repo_name
stringlengths
6
103
path
stringlengths
5
191
copies
stringlengths
1
4
size
stringlengths
4
6
content
stringlengths
986
970k
license
stringclasses
15 values
vdumoulin/fuel
fuel/transformers/sequences.py
7
4963
from fuel.transformers import Transformer class Window(Transformer): """Return pairs of source and target windows from a stream. This data stream wrapper takes as an input a data stream outputting sequences of potentially varying lengths (e.g. sentences, audio tracks, etc.). It then returns two sliding windows (source and target) over these sequences. For example, to train an n-gram model set `source_window` to n, `target_window` to 1, no offset, and `overlapping` to false. This will give chunks [1, N] and [N + 1]. To train an RNN you often want to set the source and target window to the same size and use an offset of 1 with overlap, this would give you chunks [1, N] and [2, N + 1]. Parameters ---------- offset : int The offset from the source window where the target window starts. source_window : int The size of the source window. target_window : int The size of the target window. overlapping : bool If true, the source and target windows overlap i.e. the offset of the target window is taken to be from the beginning of the source window. If false, the target window offset is taken to be from the end of the source window. data_stream : :class:`.DataStream` instance The data stream providing sequences. Each example is assumed to be an object that supports slicing. target_source : str, optional This data stream adds a new source for the target words. By default this source is 'targets'. """ def __init__(self, offset, source_window, target_window, overlapping, data_stream, target_source='targets', **kwargs): if not data_stream.produces_examples: raise ValueError('the wrapped data stream must produce examples, ' 'not batches of examples.') if len(data_stream.sources) > 1: raise ValueError('{} expects only one source' .format(self.__class__.__name__)) super(Window, self).__init__(data_stream, produces_examples=True, **kwargs) self.sources = self.sources + (target_source,) self.offset = offset self.source_window = source_window self.target_window = target_window self.overlapping = overlapping self.sentence = [] self._set_index() def _set_index(self): """Set the starting index of the source window.""" self.index = 0 # If offset is negative, target window might start before 0 self.index = -min(0, self._get_target_index()) def _get_target_index(self): """Return the index where the target window starts.""" return (self.index + self.source_window * (not self.overlapping) + self.offset) def _get_end_index(self): """Return the end of both windows.""" return max(self.index + self.source_window, self._get_target_index() + self.target_window) def get_data(self, request=None): if request is not None: raise ValueError while not self._get_end_index() <= len(self.sentence): self.sentence, = next(self.child_epoch_iterator) self._set_index() source = self.sentence[self.index:self.index + self.source_window] target = self.sentence[self._get_target_index(): self._get_target_index() + self.target_window] self.index += 1 return (source, target) class NGrams(Window): """Return n-grams from a stream. This data stream wrapper takes as an input a data stream outputting sentences. From these sentences n-grams of a fixed order (e.g. bigrams, trigrams, etc.) are extracted and returned. It also creates a ``targets`` data source. For each example, the target is the word immediately following that n-gram. It is normally used for language modeling, where we try to predict the next word from the previous *n* words. .. note:: Unlike the :class:`Window` stream, the target returned by :class:`NGrams` is a single element instead of a window. Parameters ---------- ngram_order : int The order of the n-grams to output e.g. 3 for trigrams. data_stream : :class:`.DataStream` instance The data stream providing sentences. Each example is assumed to be a list of integers. target_source : str, optional This data stream adds a new source for the target words. By default this source is 'targets'. """ def __init__(self, ngram_order, *args, **kwargs): super(NGrams, self).__init__( 0, ngram_order, 1, False, *args, **kwargs) def get_data(self, *args, **kwargs): source, target = super(NGrams, self).get_data(*args, **kwargs) return (source, target[0])
mit
yafeunteun/wikipedia-spam-classifier
revscoring/revscoring/utilities/tune.py
1
9729
""" Tunes a set of models against a training set to identify the best model/configuration. Usage: tune <params-config> <features> <label> [--observations=<path>] [--scoring=<type>] [--test-prop=<prop>] [--folds=<num>] [--report=<path>] [--label-type=<type>] [--processes=<num>] [--cv-timeout=<mins>] [--scale-features] [--verbose] [--debug] Options: <params-config> The path to a YAML configuration file containing the models and parameter values to search when tuning <features> The classpath to a feature_list to use when interpreting the feature values of the observations <label> The name of the field to be predicted --observations=<path> The path to a file containing observations to train and test against. [default: <stdin>] --scoring=<type> The type of scoring strategy to optimize for when choosing parameter sets [default: roc_auc] --folds=<num> The number of cross-validation folds to try [default: 5] --report=<path> Path to a file to write the tuning report to [default: <stdout>] --processes=<num> The number of parallel processes to start for model building [default: <cpu-count>] --cv-timeout=<mins> The number of minutes to wait for a model to cross-validate before timing out [default: <forever>] --scale-features Scales the feature values before tuning --verbose Print progress information to stderr --debug Print debug information to stderr """ import datetime import json import logging import multiprocessing import sys import time import traceback from collections import defaultdict import docopt import numpy import yamlconf from sklearn import cross_validation, grid_search, preprocessing from tabulate import tabulate from . import metrics from .. import __version__ from ..dependencies import solve from .util import Timeout, read_observations logger = logging.getLogger(__name__) def main(argv=None): args = docopt.docopt(__doc__, argv=argv) logging.basicConfig( level=logging.INFO if not args['--debug'] else logging.DEBUG, format='%(asctime)s %(levelname)s:%(name)s -- %(message)s' ) params_config = yamlconf.load(open(args['<params-config>'])) features_path = args['<features>'] features = yamlconf.import_path(features_path) if args['--observations'] == "<stdin>": observations = read_observations(sys.stdin) else: observations = read_observations(open(args['--observations'])) logger.info("Reading feature values & labels...") label_name = args['<label>'] value_labels = \ [(list(solve(features, cache=ob['cache'])), ob[label_name]) for ob in observations] # Get a sepecialized scorer if we have one scoring = metrics.SCORERS.get(args['--scoring'], args['--scoring']) folds = int(args['--folds']) if args['--report'] == "<stdout>": report = sys.stdout else: report = open(args['--report'], "w") if args['--processes'] == "<cpu-count>": processes = multiprocessing.cpu_count() else: processes = int(args['--processes']) if args['--cv-timeout'] == "<forever>": cv_timeout = None else: cv_timeout = float(args['--cv-timeout']) * 60 # Convert to seconds scale_features = args['--scale-features'] verbose = args['--verbose'] run(params_config, features_path, value_labels, scoring, folds, report, processes, cv_timeout, scale_features, verbose) def run(params_config, features_path, value_labels, scoring, folds, report, processes, cv_timeout, scale_features, verbose): if scale_features: logger.debug("Scaling features...") ss = preprocessing.StandardScaler() feature_values, labels = (list(vect) for vect in zip(*value_labels)) scaled_feature_values = ss.fit_transform(feature_values) value_labels = list(zip(scaled_feature_values, labels)) # Prepare the worker pool logger.debug("Starting up multiprocessing pool (processes={0})" .format(processes)) pool = multiprocessing.Pool(processes=processes) # Start writing the model tuning report possible_labels = set(label for _, label in value_labels) report.write("# Model tuning report\n") report.write("- Revscoring version: {0}\n".format(__version__)) report.write("- Features: {0}\n".format(features_path)) report.write("- Date: {0}\n".format(datetime.datetime.now().isoformat())) report.write("- Observations: {0}\n".format(len(value_labels))) report.write("- Labels: {0}\n".format(json.dumps(list(possible_labels)))) report.write("- Scoring: {0}\n".format(scoring)) report.write("- Folds: {0}\n".format(folds)) report.write("\n") # For each estimator and paramset, submit the job. cv_result_sets = defaultdict(lambda: []) for name, estimator, param_grid in _estimator_param_grid(params_config): logger.debug("Submitting jobs for {0}:".format(name)) for params in param_grid: logger.debug("\tsubmitting {0}..." .format(format_params(params))) result = pool.apply_async(_cross_validate, [value_labels, estimator, params], {'cv_timeout': cv_timeout, 'scoring': scoring, 'folds': folds}) cv_result_sets[name].append((params, result)) # Barrier synchronization logger.info("Running gridsearch for {0} model/params pairs ..." .format(sum(len(p_r) for p_r in cv_result_sets))) grid_scores = [] for name, param_results in cv_result_sets.items(): for params, result in param_results: scores = result.get() # This is a line that blocks grid_scores.append((name, params, scores.mean(), scores.std())) # Write the rest of the report! First, print the top 10 combinations report.write("# Top scoring configurations\n") grid_scores.sort(key=lambda gs: gs[2], reverse=True) table = tabulate( ((name, round(mean_score, 3), round(std_score, 3), format_params(params)) for name, params, mean_score, std_score in grid_scores[:10]), headers=["model", "mean(scores)", "std(scores)", "params"], tablefmt="pipe" ) report.write(table + "\n") report.write("\n") # Now print out scores for each model. report.write("# Models\n") for name, param_results in cv_result_sets.items(): report.write("## {0}\n".format(name)) param_scores = ((p, r.get()) for p, r in param_results) param_stats = [(p, s.mean(), s.std()) for p, s in param_scores] param_stats.sort(key=lambda v: v[1], reverse=True) table = tabulate( ((round(mean_score, 3), round(std_score, 3), format_params(params)) for params, mean_score, std_score in param_stats), headers=["mean(scores)", "std(scores)", "params"], tablefmt="pipe" ) report.write(table + "\n") report.write("\n") report.close() def format_params(doc): return ", ".join("{0}={1}".format(k, json.dumps(v)) for k, v in doc.items()) def _estimator_param_grid(params_config): for name, config in params_config.items(): try: EstimatorClass = yamlconf.import_module(config['class']) estimator = EstimatorClass() except Exception: logger.warn("Could not load estimator {0}" .format(config['class'])) logger.warn("Exception:\n" + traceback.format_exc()) continue if not hasattr(estimator, "fit"): logger.warn("Estimator {0} does not have a fit() method." .format(config['class'])) continue param_grid = grid_search.ParameterGrid(config['params']) yield name, estimator, param_grid def _cross_validate(value_labels, estimator, params, scoring="roc_auc", folds=5, cv_timeout=None, verbose=False): start = time.time() feature_values, labels = (list(vect) for vect in zip(*value_labels)) estimator.set_params(**params) try: logger.debug("Running cross-validation for " + "{0} with timeout of {1} seconds" .format(estimator.__class__.__name__, cv_timeout)) with Timeout(cv_timeout): scores = cross_validation.cross_val_score( estimator, feature_values, labels, scoring=scoring, cv=folds) duration = time.time() - start logger.debug("Cross-validated {0} with {1} in {2} minutes: {3} ({4})" .format(estimator.__class__.__name__, format_params(params), round(duration / 60, 3), round(scores.mean(), 3), round(scores.std(), 3))) return scores except Exception: logger.warn("Could not cross-validate estimator {0}" .format(estimator.__class__.__name__)) logger.warn("Exception:\n" + traceback.format_exc()) return numpy.array([0] * folds)
mit
lakshayg/tensorflow
tensorflow/contrib/learn/python/learn/preprocessing/categorical.py
151
4269
# Copyright 2016 The TensorFlow Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """Implements preprocessing transformers for categorical variables.""" from __future__ import absolute_import from __future__ import division from __future__ import print_function import math import numpy as np # pylint: disable=g-bad-import-order from . import categorical_vocabulary from ..learn_io.data_feeder import setup_processor_data_feeder # pylint: enable=g-bad-import-order class CategoricalProcessor(object): """Maps documents to sequences of word ids. As a common convention, Nan values are handled as unknown tokens. Both float('nan') and np.nan are accepted. """ def __init__(self, min_frequency=0, share=False, vocabularies=None): """Initializes a CategoricalProcessor instance. Args: min_frequency: Minimum frequency of categories in the vocabulary. share: Share vocabulary between variables. vocabularies: list of CategoricalVocabulary objects for each variable in the input dataset. Attributes: vocabularies_: list of CategoricalVocabulary objects. """ self.min_frequency = min_frequency self.share = share self.vocabularies_ = vocabularies def freeze(self, freeze=True): """Freeze or unfreeze all vocabularies. Args: freeze: Boolean, indicate if vocabularies should be frozen. """ for vocab in self.vocabularies_: vocab.freeze(freeze) def fit(self, x, unused_y=None): """Learn a vocabulary dictionary of all categories in `x`. Args: x: numpy matrix or iterable of lists/numpy arrays. unused_y: to match fit format signature of estimators. Returns: self """ x = setup_processor_data_feeder(x) for row in x: # Create vocabularies if not given. if self.vocabularies_ is None: # If not share, one per column, else one shared across. if not self.share: self.vocabularies_ = [ categorical_vocabulary.CategoricalVocabulary() for _ in row ] else: vocab = categorical_vocabulary.CategoricalVocabulary() self.vocabularies_ = [vocab for _ in row] for idx, value in enumerate(row): # Nans are handled as unknowns. if (isinstance(value, float) and math.isnan(value)) or value == np.nan: continue self.vocabularies_[idx].add(value) if self.min_frequency > 0: for vocab in self.vocabularies_: vocab.trim(self.min_frequency) self.freeze() return self def fit_transform(self, x, unused_y=None): """Learn the vocabulary dictionary and return indexies of categories. Args: x: numpy matrix or iterable of lists/numpy arrays. unused_y: to match fit_transform signature of estimators. Returns: x: iterable, [n_samples]. Category-id matrix. """ self.fit(x) return self.transform(x) def transform(self, x): """Transform documents to category-id matrix. Converts categories to ids give fitted vocabulary from `fit` or one provided in the constructor. Args: x: numpy matrix or iterable of lists/numpy arrays. Yields: x: iterable, [n_samples]. Category-id matrix. """ self.freeze() x = setup_processor_data_feeder(x) for row in x: output_row = [] for idx, value in enumerate(row): # Return <UNK> when it's Nan. if (isinstance(value, float) and math.isnan(value)) or value == np.nan: output_row.append(0) continue output_row.append(self.vocabularies_[idx].get(value)) yield np.array(output_row, dtype=np.int64)
apache-2.0
musically-ut/statsmodels
examples/python/tsa_dates.py
29
1169
## Dates in timeseries models from __future__ import print_function import statsmodels.api as sm import pandas as pd # ## Getting started data = sm.datasets.sunspots.load() # Right now an annual date series must be datetimes at the end of the year. dates = sm.tsa.datetools.dates_from_range('1700', length=len(data.endog)) # ## Using Pandas # # Make a pandas TimeSeries or DataFrame endog = pd.TimeSeries(data.endog, index=dates) # Instantiate the model ar_model = sm.tsa.AR(endog, freq='A') pandas_ar_res = ar_model.fit(maxlag=9, method='mle', disp=-1) # Out-of-sample prediction pred = pandas_ar_res.predict(start='2005', end='2015') print(pred) # ## Using explicit dates ar_model = sm.tsa.AR(data.endog, dates=dates, freq='A') ar_res = ar_model.fit(maxlag=9, method='mle', disp=-1) pred = ar_res.predict(start='2005', end='2015') print(pred) # This just returns a regular array, but since the model has date information attached, you can get the prediction dates in a roundabout way. print(ar_res.data.predict_dates) # Note: This attribute only exists if predict has been called. It holds the dates associated with the last call to predict.
bsd-3-clause
jpzk/evopy
evopy/examples/experiments/cv_ppv_dsesscv/plot_precisions.py
1
4256
''' This file is part of evopy. Copyright 2012 - 2013, Jendrik Poloczek evopy is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. evopy is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with evopy. If not, see <http://www.gnu.org/licenses/>. ''' from sys import path path.append("../../../..") from pickle import load from copy import deepcopy from numpy import matrix, log10, array from scipy.stats import wilcoxon from itertools import chain from pylab import errorbar from matplotlib.backends.backend_pdf import PdfPages from evopy.strategies.ori_dses_svc_repair import ORIDSESSVCR from evopy.strategies.ori_dses_svc import ORIDSESSVC from evopy.strategies.ori_dses import ORIDSES from evopy.simulators.simulator import Simulator from evopy.problems.sphere_problem_origin_r1 import SphereProblemOriginR1 from evopy.problems.sphere_problem_origin_r2 import SphereProblemOriginR2 from evopy.problems.schwefels_problem_26 import SchwefelsProblem26 from evopy.problems.tr_problem import TRProblem from evopy.metamodel.dses_svc_linear_meta_model import DSESSVCLinearMetaModel from sklearn.cross_validation import KFold from evopy.operators.scaling.scaling_standardscore import ScalingStandardscore from evopy.operators.scaling.scaling_dummy import ScalingDummy from evopy.metamodel.cv.svc_cv_sklearn_grid_linear import SVCCVSkGridLinear from evopy.operators.termination.or_combinator import ORCombinator from evopy.operators.termination.accuracy import Accuracy from evopy.operators.termination.generations import Generations from evopy.operators.termination.convergence import Convergence from evopy.helper.timeseries_aggregator import TimeseriesAggregator import matplotlib.pyplot as plt from setup import * precisionfile = file("output/precision_file.save", "r") precisions = load(precisionfile) none = lambda x : type(x) != type(None) for problem in precisions.keys(): figure_accs = plt.figure(figsize=(8,6), dpi=10, facecolor="w", edgecolor="k") plt.xlabel("Generation") plt.ylabel("Gemittelter Positiver Vorhersagewert") plt.xlim([0, 50]) plt.ylim([0.0, 1.0]) o_colors = { get_method_TR_none: "g",\ get_method_TR_nor: "k",\ get_method_TR_ssc: "#044977",\ get_method_SphereProblemR1_none: "g",\ get_method_SphereProblemR1_nor: "k",\ get_method_SphereProblemR1_ssc: "#044977",\ get_method_SphereProblemR2_none: "g",\ get_method_SphereProblemR2_nor: "k",\ get_method_SphereProblemR2_ssc: "#044977",\ get_method_Schwefel26_none: "g",\ get_method_Schwefel26_nor: "k",\ get_method_Schwefel26_ssc: "#044977"} o_markers = { get_method_TR_none: "x",\ get_method_TR_nor: "+",\ get_method_TR_ssc: ".",\ get_method_SphereProblemR1_none: "x",\ get_method_SphereProblemR1_nor: "+",\ get_method_SphereProblemR1_ssc: ".",\ get_method_SphereProblemR2_none: "x",\ get_method_SphereProblemR2_nor: "+",\ get_method_SphereProblemR2_ssc: ".",\ get_method_Schwefel26_none: "x",\ get_method_Schwefel26_nor: "+",\ get_method_Schwefel26_ssc: "."} optimizers = precisions[problem].keys() for index, optimizer in enumerate(optimizers): precisions_po = precisions[problem][optimizer] precisions_agg, errors_agg =\ TimeseriesAggregator(precisions_po).get_aggregate() generations = range(0, len(precisions_agg)) eb = errorbar(generations,\ precisions_agg,\ marker=o_markers[optimizer], color=o_colors[optimizer],\ ecolor="#CCCCCC",\ linestyle="none", yerr=errors_agg) pp = PdfPages("output/p_%s.pdf" % str(problem).split('.')[-1]) plt.savefig(pp, format='pdf') pp.close()
gpl-3.0
lakshayg/tensorflow
tensorflow/python/keras/_impl/keras/datasets/fashion_mnist.py
12
2055
# Copyright 2017 The TensorFlow Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """Fashion-MNIST dataset. """ from __future__ import absolute_import from __future__ import division from __future__ import print_function import gzip import os import numpy as np from tensorflow.python.keras._impl.keras.utils.data_utils import get_file def load_data(): """Loads the Fashion-MNIST dataset. Returns: Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`. """ dirname = os.path.join('datasets', 'fashion-mnist') base = 'http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/' files = [ 'train-labels-idx1-ubyte.gz', 'train-images-idx3-ubyte.gz', 't10k-labels-idx1-ubyte.gz', 't10k-images-idx3-ubyte.gz' ] paths = [] for given_file in files: paths.append( get_file(given_file, origin=base + given_file, cache_subdir=dirname)) with gzip.open(paths[0], 'rb') as lbpath: y_train = np.frombuffer(lbpath.read(), np.uint8, offset=8) with gzip.open(paths[1], 'rb') as imgpath: x_train = np.frombuffer( imgpath.read(), np.uint8, offset=16).reshape(len(y_train), 28, 28) with gzip.open(paths[2], 'rb') as lbpath: y_test = np.frombuffer(lbpath.read(), np.uint8, offset=8) with gzip.open(paths[3], 'rb') as imgpath: x_test = np.frombuffer( imgpath.read(), np.uint8, offset=16).reshape(len(y_test), 28, 28) return (x_train, y_train), (x_test, y_test)
apache-2.0
musically-ut/statsmodels
statsmodels/datasets/statecrime/data.py
25
3128
#! /usr/bin/env python """Statewide Crime Data""" __docformat__ = 'restructuredtext' COPYRIGHT = """Public domain.""" TITLE = """Statewide Crime Data 2009""" SOURCE = """ All data is for 2009 and was obtained from the American Statistical Abstracts except as indicated below. """ DESCRSHORT = """State crime data 2009""" DESCRLONG = DESCRSHORT #suggested notes NOTE = """:: Number of observations: 51 Number of variables: 8 Variable name definitions: state All 50 states plus DC. violent Rate of violent crimes / 100,000 population. Includes murder, forcible rape, robbery, and aggravated assault. Numbers for Illinois and Minnesota do not include forcible rapes. Footnote included with the American Statistical Abstract table reads: "The data collection methodology for the offense of forcible rape used by the Illinois and the Minnesota state Uniform Crime Reporting (UCR) Programs (with the exception of Rockford, Illinois, and Minneapolis and St. Paul, Minnesota) does not comply with national UCR guidelines. Consequently, their state figures for forcible rape and violent crime (of which forcible rape is a part) are not published in this table." murder Rate of murders / 100,000 population. hs_grad Precent of population having graduated from high school or higher. poverty % of individuals below the poverty line white Percent of population that is one race - white only. From 2009 American Community Survey single Calculated from 2009 1-year American Community Survey obtained obtained from Census. Variable is Male householder, no wife present, family household combined with Female household, no husband prsent, family household, divided by the total number of Family households. urban % of population in Urbanized Areas as of 2010 Census. Urbanized Areas are area of 50,000 or more people.""" import numpy as np from statsmodels.datasets import utils as du from os.path import dirname, abspath def load(): """ Load the statecrime data and return a Dataset class instance. Returns ------- Dataset instance: See DATASET_PROPOSAL.txt for more information. """ data = _get_data() ##### SET THE INDICES ##### #NOTE: None for exog_idx is the complement of endog_idx return du.process_recarray(data, endog_idx=2, exog_idx=[7, 4, 3, 5], dtype=float) def load_pandas(): data = _get_data() ##### SET THE INDICES ##### #NOTE: None for exog_idx is the complement of endog_idx return du.process_recarray_pandas(data, endog_idx=2, exog_idx=[7,4,3,5], dtype=float, index_idx=0) def _get_data(): filepath = dirname(abspath(__file__)) ##### EDIT THE FOLLOWING TO POINT TO DatasetName.csv ##### data = np.recfromtxt(open(filepath + '/statecrime.csv', 'rb'), delimiter=",", names=True, dtype=None) return data
bsd-3-clause
PAIR-code/recommendation-rudders
hyperbolic-rs/preprocess.py
1
11964
# Copyright 2017 The Rudders Authors. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from absl import app, flags import pickle import tensorflow as tf import numpy as np import random from tqdm import tqdm from pathlib import Path from rudders.relations import Relations from rudders.datasets import movielens, keen, amazon, amazon_relations, synopsis from rudders.config import CONFIG from rudders.utils import set_seed, sort_items_by_popularity, save_as_pickle, add_to_train_split FLAGS = flags.FLAGS flags.DEFINE_string('prep_id', default='foobar', help='Name of prep to store') flags.DEFINE_string('item', default='amazon', help='Item to process: "keen", "gem", "ml-1m", "amazon" or "synopsis"') flags.DEFINE_string('dataset_path', default='data/amazon', help='Path to raw dataset') flags.DEFINE_string('amazon_reviews', default='Musical_Instruments_5.json.gz', help='Name of the 5-core amazon reviews file') flags.DEFINE_string('amazon_meta', default='meta_Musical_Instruments.json.gz', help='Name of the 5-core amazon reviews file') flags.DEFINE_string('item_item_file', default='Musical_Instruments_th0.6_cosdistances.pickle', help='Path to the item-item distance file') flags.DEFINE_boolean('plot_graph', default=False, help='Plots the user-item graph') flags.DEFINE_boolean('shuffle', default=False, help='Whether to shuffle the interactions or not') flags.DEFINE_boolean('add_extra_relations', default=True, help='For the amazon dataset, adds extra relations') flags.DEFINE_boolean('export_splits', default=True, help='Exports (user_id, item_id) pairs of all splits') flags.DEFINE_integer('min_user_interactions', default=5, help='Keens users with less than min_user_interactions are filtered') flags.DEFINE_integer('min_item_interactions', default=2, help='Keens/gems with less than this interactions are filtered') flags.DEFINE_integer('max_item_interactions', default=150, help='Keens/gems with more than this interactions are filtered') flags.DEFINE_integer('similarity_items_per_item', default=10, help='Amount of similarity items to add per item') flags.DEFINE_integer('seed', default=42, help='Random seed') flags.DEFINE_integer('filter_most_popular', default=-1, help='Filters out most popular keens/gems. If -1 it does not filter') def plot_graph(samples): """Plot user-item graph, setting different colors for items and users.""" import networkx as nx import matplotlib.pyplot as plt graph = nx.Graph() for uid, ints in samples.items(): for iid in ints: graph.add_edge(uid, iid) color_map = ["red" if node in samples else "blue" for node in graph] fig = plt.figure() pos = nx.spring_layout(graph, iterations=100) nx.draw(graph, pos, ax=fig.add_subplot(111), node_size=20, node_color=color_map) plt.show() def map_raw_ids_to_sequential_ids(samples): """ For each unique user or item id, this function creates a mapping to a sequence of number starting in 0. This will be the index of the embeddings in the model. Items ids will be from 0 to n_items - 1. Users ids will be from n_items to n_items + n_users - 1 This condition is required to later build the distance matrix :param samples: dict of <user_id1>: [<item_id1>, <item_id2>, ...] :return: dicts of {<user_idX>: indexY} and {<item_idX>: indexW} """ uid2id, iid2id = {}, {} sorted_samples = sorted(samples.items(), key=lambda x: x[0]) # first sets items ids only for _, ints in sorted_samples: sorted_ints = sorted(ints) for iid in sorted_ints: if iid not in iid2id: iid2id[iid] = len(iid2id) # users ids come after item ids for uid, _ in sorted_samples: if uid not in uid2id: uid2id[uid] = len(uid2id) + len(iid2id) return uid2id, iid2id def create_splits(samples, relation_id, do_random=False, seed=42): """ Splits (user, item) dataset to train, dev and test. :param samples: Dict of sorted examples. :param relation_id: number that identifies the user-item interaction relation to form the triplets :param do_random: Bool whether to extract dev and test by random sampling. If False, dev, test are the last two items per user. :return: examples: Dictionary with 'train','dev','test' splits as numpy arrays containing corresponding (user_id, item_id) pairs, and 'to_skip' to a Dictionary containing filters for each user. """ train, dev, test = [], [], [] for uid, ints in samples.items(): if do_random: random.seed(seed) random.shuffle(ints) if len(ints) >= 3: test.append((uid, relation_id, ints[-1])) dev.append((uid, relation_id, ints[-2])) for iid in ints[:-2]: train.append((uid, relation_id, iid)) else: for iid in ints: train.append((uid, relation_id, iid)) return { 'samples': samples, 'train': np.array(train).astype('int64'), 'dev': np.array(dev).astype('int64'), 'test': np.array(test).astype('int64') } def load_item_item_distances(item_item_file_path): """Loads item-item distances that were precomputed with item_graph.py.""" print(f"Loading data from {item_item_file_path}") with tf.io.gfile.GFile(str(item_item_file_path), 'rb') as f: data = pickle.load(f) return data["item_item_distances"] def build_item_item_triplets(item_item_distances_dict, iid2id, top_k): """ Builds item item triples from the item-item distances :param item_item_distances_dict: dict of src_iid: [(dst_iid, distance)] :param iid2id: dict of item ids :param top_k: adds top_k items per item at most :return: """ triplets = set() for src_iid, dists in tqdm(item_item_distances_dict.items(), desc="item_item_triplets"): if src_iid not in iid2id: continue src_id = iid2id[src_iid] sorted_dists = sorted(dists, key=lambda t: t[1]) added = 0 for dst_iid, cos_dist in sorted_dists: if dst_iid not in iid2id or cos_dist > 0.3: continue dst_id = iid2id[dst_iid] if cos_dist <= 0.1: triplets.add((src_id, Relations.SEM_HIGH_SIM.value, dst_id)) elif 0.2 >= cos_dist > 0.1: triplets.add((src_id, Relations.SEM_MEDIUM_SIM.value, dst_id)) else: # 0.3 >= cos_dist > 0.2 triplets.add((src_id, Relations.SEM_LOW_SIM.value, dst_id)) added += 1 if added >= top_k: break return list(triplets) def export_splits(data, to_save_dir, prep_id): """Exports (user_id, item_id) pairs of all splits splits""" split_names = ["train", "dev", "test"] id2uid, id2iid = data["id2uid"], data["id2iid"] for split_name in split_names: split = data[split_name] if split_name == "train": split = [(uid, r, iid) for uid, r, iid in split if r == Relations.USER_ITEM.value] lines = [f"{id2uid[u_id]},{id2iid[i_id]}\n" for u_id, _, i_id in split] with open(to_save_dir / f"{prep_id}_ui_{split_name}.csv", "w") as f: f.writelines(lines) def main(_): set_seed(FLAGS.seed, set_tf_seed=True) dataset_path = Path(FLAGS.dataset_path) if FLAGS.item == "keen": samples = keen.load_user_keen_interactions(dataset_path, min_user_ints=FLAGS.min_user_interactions, min_item_ints=FLAGS.min_item_interactions, max_item_ints=FLAGS.max_item_interactions) iid2name = keen.build_iid2title(item_id_key="keen_id", item_title_key="keen_title") elif FLAGS.item == "gem": samples = keen.load_keen_gems_interactions(dataset_path, min_keen_keen_edges=2, max_keen_keen_edges=1000, min_overlapping_users=2, min_keen_ints=FLAGS.min_user_interactions, min_item_ints=FLAGS.min_item_interactions, max_item_ints=FLAGS.max_item_interactions) iid2name = keen.build_iid2title(item_id_key="gem_id", item_title_key="gem_link_title") elif FLAGS.item == "ml-1m": samples = movielens.movielens_to_dict(dataset_path) iid2name = movielens.build_movieid2title(dataset_path) elif "amazon" in FLAGS.item: samples = amazon.load_interactions(dataset_path / FLAGS.amazon_reviews) iid2name = amazon.build_itemid2name(dataset_path / FLAGS.amazon_meta) elif FLAGS.item == "synopsis": samples = synopsis.synopsis_to_dict(dataset_path) iid2name = synopsis.build_movieid2title(dataset_path) else: raise ValueError(f"Unknown item: {FLAGS.item}") if FLAGS.filter_most_popular > 0: print(f"Filtering {FLAGS.filter_most_popular} most popular items") sorted_items = sort_items_by_popularity(samples) iid_to_filter = set([iid for iid, _ in sorted_items[:FLAGS.filter_most_popular]]) samples = {uid: list(set(ints) - iid_to_filter) for uid, ints in samples.items()} samples = {uid: ints for uid, ints in samples.items() if ints} if FLAGS.plot_graph: plot_graph(samples) return uid2id, iid2id = map_raw_ids_to_sequential_ids(samples) id_samples = {} for uid, ints in samples.items(): if FLAGS.item == "keen" or FLAGS.item == "gem": ints = sorted(ints) id_samples[uid2id[uid]] = [iid2id[iid] for iid in ints] data = create_splits(id_samples, Relations.USER_ITEM.value, do_random=FLAGS.shuffle, seed=FLAGS.seed) data["iid2name"] = {iid: iid2name.get(iid, "None") for iid in iid2id} data["id2uid"] = {v: k for k, v in uid2id.items()} data["id2iid"] = {v: k for k, v in iid2id.items()} print(f"User item interaction triplets: {len(data['train'])}") n_entities = len(uid2id) + len(iid2id) # if there is an item-item graph, we preprocess it if FLAGS.item_item_file: item_item_distances_dict = load_item_item_distances(dataset_path / FLAGS.item_item_file) item_item_triplets = build_item_item_triplets(item_item_distances_dict, iid2id, FLAGS.similarity_items_per_item) add_to_train_split(data, item_item_triplets) print(f"Added item-item similarity triplets: {len(item_item_triplets)}") if "amazon" in FLAGS.item and FLAGS.add_extra_relations: print("Adding extra relations") n_entities = amazon_relations.load_relations(dataset_path / FLAGS.amazon_meta, data, iid2id, n_entities) data["n_entities"] = n_entities # creates directories to save preprocessed data print(f"Final training split: {len(data['train'])} triplets") prep_path = Path(CONFIG["string"]["prep_dir"][1]) prep_path.mkdir(parents=True, exist_ok=True) to_save_dir = prep_path / FLAGS.item to_save_dir.mkdir(parents=True, exist_ok=True) save_as_pickle(to_save_dir / f'{FLAGS.prep_id}.pickle', data) if FLAGS.export_splits: export_splits(data, to_save_dir, FLAGS.prep_id) print("Done!") if __name__ == '__main__': app.run(main)
apache-2.0
openplans/shareabouts-api
src/sa_api_v2/tasks.py
1
10690
import requests import ujson as json from celery import shared_task from celery.result import AsyncResult from django.db import transaction from django.test.client import RequestFactory from django.utils.timezone import now from itertools import chain #from social.apps.django_app.default.models import UserSocialAuth from .models import DataSnapshotRequest, DataSnapshot, DataSet, User, Place, Submission from .serializers import SimplePlaceSerializer, SimpleSubmissionSerializer, SimpleDataSetSerializer from .renderers import CSVRenderer, JSONRenderer, GeoJSONRenderer import logging log = logging.getLogger(__name__) # ========================================================= # Generating snapshots # def generate_bulk_content(dataset, submission_set_name, **flags): renderer_classes = { 'csv': CSVRenderer, 'json': GeoJSONRenderer if submission_set_name == 'places' else JSONRenderer } if submission_set_name == 'places': submissions = dataset.places.all() serializer = SimplePlaceSerializer(submissions, many=True) else: submissions = dataset.submissions.filter(set_name=submission_set_name) serializer = SimpleSubmissionSerializer(submissions, many=True) # Construct a request for the serializer context r_data = {} for flag_attr, flag_val in flags.items(): if flag_val: r_data[flag_attr] = 'true' r = RequestFactory().get('', data=r_data) r.get_dataset = lambda: dataset # Render the data in each format serializer.context['request'] = r data = serializer.data content = {} for format, renderer_class in list(renderer_classes.items()): renderer = renderer_class() content[format] = renderer.render(data) return content @shared_task def store_bulk_data(request_id): task_id = store_bulk_data.request.id log.info('Creating a snapshot request with task id %s' % (task_id,)) datarequest = DataSnapshotRequest.objects.get(pk=request_id) datarequest.guid = task_id datarequest.save() # Generate the content content = generate_bulk_content( datarequest.dataset, datarequest.submission_set, include_submissions=datarequest.include_submissions, include_private=datarequest.include_private, include_invisible=datarequest.include_invisible) # Store the information bulk_data = DataSnapshot( request=datarequest, csv=content['csv'], json=content['json']) bulk_data.save() datarequest.fulfilled_at = now() datarequest.save() return task_id @shared_task def bulk_data_status_update(uuid): """ A callback task that updates the status of a data snapshot request, whether successful or not. """ taskresult = AsyncResult(uuid) datarequest = DataSnapshotRequest.objects.get(guid=uuid) datarequest.status = taskresult.status.lower() datarequest.save() @shared_task def clone_related_dataset_data(orig_dataset_id, new_dataset_id): qs = DataSet.objects.select_related('owner')\ .filter(id__in=(orig_dataset_id, new_dataset_id))\ .prefetch_related('things', 'things__place', 'things__place__dataset', 'things__place__submitter', 'things__place__submissions', 'things__place__submissions__dataset', 'things__place__submissions__submitter', 'permissions', 'groups', 'groups__submitters', 'groups__permissions', 'keys', 'keys__permissions', 'origins', 'origins__permissions', ) datasets = list(qs) if datasets[0].id == orig_dataset_id: orig_dataset, new_dataset = datasets else: new_dataset, orig_dataset = datasets with transaction.atomic(): orig_dataset.clone_related(onto=new_dataset) # ========================================================= # Loading a dataset # def get_twitter_extra_data(user_data): return { 'id': user_data.get('provider_id'), 'profile_image_url': user_data.get('avatar_url'), 'access_token': { 'screen_name': user_data.get('username'), 'oauth_token_secret': 'abc', 'oauth_token': '123', 'user_id': user_data.get('provider_id') }, 'name': user_data.get('name') } def get_facebook_extra_data(user_data): return { 'access_token': 'abc123', 'picture': { "data": { "url": user_data.get('avatar_url'), } }, "id": user_data.get('provider_id'), "name": user_data.get('name'), } def get_or_create_user(user_data, users_map): if user_data is None: return # Check whether the user is already cached username = user_data.get('username') user = users_map.get(username) if user: return user # Create and cache the user user = User.objects.create(username=username, password='!') users_map[username] = user # Create a social auth entry for the user, if appropriate provider = user_data.get('provider_type') uid = user_data.get('provider_id') if provider and uid: UserSocialAuth.objects.create( user=user, provider=provider, uid=uid, extra_data= get_twitter_extra_data(user_data) if provider == 'twitter' else get_facebook_extra_data(user_data) ) def preload_users(data): """ Construct a mapping from usernames to users for Users that already exist in the API. """ usernames = set() def collect_username(data): submitter_data = data.get('submitter') if submitter_data: usernames.add(submitter_data.get('username')) for place_data in data.get('features', []): collect_username(place_data['properties']) for _, submissions_data in place_data['properties'].get('submission_sets', {}).items(): for submission_data in submissions_data: collect_username(submission_data) users = User.objects.filter(username__in=usernames) users_map = dict([(user.username, user) for user in users]) return users_map def list_errors(errors): errors_list = [] for key, l in list(errors.items()): if isinstance(l, list): for msg in l: errors_list.append('%s: %s' % (key, str(msg))) else: msg = l errors_list.append('%s: %s' % (key, str(msg))) return errors_list @shared_task def load_dataset_archive(dataset_id, archive_url): dataset = DataSet.objects.get(id=dataset_id) archive_response = requests.get(archive_url) if archive_response.status_code == 200: data = archive_response.json() # Preload users users_map = preload_users(data) with transaction.atomic(): # Construct the dataset from metadata metadata = data.get('metadata') if metadata: metadata.pop('id', None) metadata.pop('owner', None) serializer = SimpleDataSetSerializer(dataset, data=data.get('metadata')) assert serializer.is_valid, list_errors(serializer.errors) serializer.save() # Create a stub view object to use in serializer contexts. class Stub (object): pass view = Stub() view.request = Stub() view.request.META = {'HTTP_X_SHAREABOUTS_SILENT': 'True'} view.request.user = Stub() view.request.user.is_authenticated = lambda: False # Construct each place and submission individually for place_data in data.get('features'): place_data.pop('type', None) place_data.update(place_data.pop('properties', {})) place_data.pop('id', None) place_data.pop('dataset', None) place_data.pop('created_datetime', None) place_data.pop('updated_datetime', None) submission_sets_data = place_data.pop('submission_sets', {}) submitter_data = place_data.pop('submitter', None) serializer_context = {'view': view, 'request': view.request} serializer = SimplePlaceSerializer(data=place_data, context=serializer_context) assert serializer.is_valid(), list_errors(serializer.errors) place = Place() for attr, value in serializer.validated_data.items(): setattr(place, attr, value) place.dataset = dataset place.submitter = get_or_create_user(submitter_data, users_map) place.save(silent=True, reindex=False) for set_name, submissions_data in submission_sets_data.items(): for submission_data in submissions_data: submission_data.pop('id', None) submission_data.pop('place', None) submission_data.pop('dataset', None) submission_data.pop('attachments', None) submission_data.pop('created_datetime', None) submission_data.pop('updated_datetime', None) submitter_data = submission_data.pop('submitter', None) serializer_context = {'view': view, 'request': view.request} serializer = SimpleSubmissionSerializer(data=submission_data, context=serializer_context) assert serializer.is_valid(), list_errors(serializer.errors) submission = Submission() for attr, value in serializer.validated_data.items(): setattr(submission, attr, value) submission.set_name = set_name submission.place = place submission.dataset = dataset submission.submitter = get_or_create_user(submitter_data, users_map) submission.save(silent=True, reindex=False) dataset.reindex() # Load meta-data like permissions and such # metadata = data.get('metadata') # for permission_data in metadata.get('permissions'):
gpl-3.0
kastnerkyle/pylearn2
pylearn2/datasets/stl10.py
1
5304
""" .. todo:: WRITEME """ __authors__ = "Ian Goodfellow" __copyright__ = "Copyright 2010-2012, Universite de Montreal" __credits__ = ["Ian Goodfellow"] __license__ = "3-clause BSD" __maintainer__ = "LISA Lab" __email__ = "pylearn-dev@googlegroups" import numpy as np from pylearn2.datasets import dense_design_matrix from pylearn2.utils.serial import load from pylearn2.utils import contains_nan class STL10(dense_design_matrix.DenseDesignMatrix): """ The STL-10 dataset Adam Coates, Honglak Lee, Andrew Y. Ng An Analysis of Single Layer Networks in Unsupervised Feature Learning AISTATS, 2011 http://www.stanford.edu/~acoates//stl10/ When reporting results on this dataset, you are meant to use a somewhat unusal evaluation procedure. Use STL10(which_set='train') to load the training set. Then restrict the training set to one of the ten folds using the restrict function below. You must then train only on the data from that fold. For the test set, report the average test set performance over the ten trials obtained by training on each of the ten folds. The folds here do not define the splits you should use for cross validation. You are free to make your own split within each fold. Parameters ---------- which_set : WRITEME center : WRITEME example_range : WRITEME """ def __init__(self, which_set, center=False, example_range=None): """ .. todo:: WRITEME """ if which_set == 'train': train = load('${PYLEARN2_DATA_PATH}/stl10/stl10_matlab/train.mat') # Load the class names self.class_names = [array[0].encode('utf-8') for array in train['class_names'][0]] # Load the fold indices fold_indices = train['fold_indices'] assert fold_indices.shape == (1, 10) self.fold_indices = np.zeros((10, 1000), dtype='uint16') for i in xrange(10): indices = fold_indices[0, i] assert indices.shape == (1000, 1) assert indices.dtype == 'uint16' self.fold_indices[i, :] = indices[:, 0] # The data is stored as uint8 # If we leave it as uint8, it will cause the CAE to silently fail # since theano will treat derivatives wrt X as 0 X = np.cast['float32'](train['X']) assert X.shape == (5000, 96 * 96 * 3) if example_range is not None: X = X[example_range[0]:example_range[1], :] # this is uint8 y = train['y'][:, 0] assert y.shape == (5000,) elif which_set == 'test': test = load('${PYLEARN2_DATA_PATH}/stl10/stl10_matlab/test.mat') # Load the class names self.class_names = [array[0].encode('utf-8') for array in test['class_names'][0]] # The data is stored as uint8 # If we leave it as uint8, it will cause the CAE to silently fail # since theano will treat derivatives wrt X as 0 X = np.cast['float32'](test['X']) assert X.shape == (8000, 96 * 96 * 3) if example_range is not None: X = X[example_range[0]:example_range[1], :] # this is uint8 y = test['y'][:, 0] assert y.shape == (8000,) elif which_set == 'unlabeled': unlabeled = load('${PYLEARN2_DATA_PATH}/stl10/stl10_matlab/' 'unlabeled.mat') X = unlabeled['X'] # this file is stored in HDF format, which transposes everything assert X.shape == (96 * 96 * 3, 100000) assert X.dtype == 'uint8' if example_range is None: X = X.value else: X = X.value[:, example_range[0]:example_range[1]] X = np.cast['float32'](X.T) unlabeled.close() y = None else: raise ValueError('"' + which_set + '" is not an STL10 dataset. ' 'Recognized values are "train", "test", and ' '"unlabeled".') if center: X -= 127.5 view_converter = dense_design_matrix.DefaultViewConverter((96, 96, 3)) super(STL10, self).__init__(X=X, y=y, view_converter=view_converter) for i in xrange(self.X.shape[0]): mat = X[i:i + 1, :] topo = self.get_topological_view(mat) for j in xrange(topo.shape[3]): temp = topo[0, :, :, j].T.copy() topo[0, :, :, j] = temp mat = self.get_design_matrix(topo) X[i:i + 1, :] = mat assert not contains_nan(self.X) def restrict(dataset, fold): """ Restricts the dataset to use the specified fold (1 to 10). dataset should be the training set. """ fold_indices = dataset.fold_indices assert fold_indices.shape == (10, 1000) idxs = fold_indices[fold, :] - 1 dataset.X = dataset.X[idxs, :].copy() assert dataset.X.shape[0] == 1000 dataset.y = dataset.y[idxs, ...].copy() assert dataset.y.shape[0] == 1000 return dataset
bsd-3-clause
nicproulx/mne-python
examples/connectivity/plot_mixed_source_space_connectity.py
3
6976
""" =============================================================================== Compute mixed source space connectivity and visualize it using a circular graph =============================================================================== This example computes the all-to-all connectivity between 75 regions in a mixed source space based on dSPM inverse solutions and a FreeSurfer cortical parcellation. The connectivity is visualized using a circular graph which is ordered based on the locations of the regions. """ # Author: Annalisa Pascarella <a.pascarella@iac.cnr.it> # # License: BSD (3-clause) import os.path as op import numpy as np import mne from mne.datasets import sample from mne import setup_volume_source_space, setup_source_space from mne import make_forward_solution from mne.io import read_raw_fif from mne.minimum_norm import make_inverse_operator, apply_inverse_epochs from mne.connectivity import spectral_connectivity from mne.viz import circular_layout, plot_connectivity_circle # Set dir data_path = sample.data_path() subject = 'sample' data_dir = op.join(data_path, 'MEG', subject) subjects_dir = op.join(data_path, 'subjects') bem_dir = op.join(subjects_dir, subject, 'bem') # Set file names fname_aseg = op.join(subjects_dir, subject, 'mri', 'aseg.mgz') fname_model = op.join(bem_dir, '%s-5120-bem.fif' % subject) fname_bem = op.join(bem_dir, '%s-5120-bem-sol.fif' % subject) fname_raw = data_dir + '/sample_audvis_filt-0-40_raw.fif' fname_trans = data_dir + '/sample_audvis_raw-trans.fif' fname_cov = data_dir + '/ernoise-cov.fif' fname_event = data_dir + '/sample_audvis_filt-0-40_raw-eve.fif' # List of sub structures we are interested in. We select only the # sub structures we want to include in the source space labels_vol = ['Left-Amygdala', 'Left-Thalamus-Proper', 'Left-Cerebellum-Cortex', 'Brain-Stem', 'Right-Amygdala', 'Right-Thalamus-Proper', 'Right-Cerebellum-Cortex'] # Setup a surface-based source space src = setup_source_space(subject, fname=None, subjects_dir=subjects_dir, spacing='oct6', add_dist=False) # Setup a volume source space # set pos=7.0 for speed issue vol_src = setup_volume_source_space(subject, mri=fname_aseg, pos=7.0, bem=fname_model, volume_label=labels_vol, subjects_dir=subjects_dir) # Generate the mixed source space src += vol_src # compute the fwd matrix fwd = make_forward_solution(fname_raw, fname_trans, src, fname_bem, mindist=5.0, # ignore sources<=5mm from innerskull meg=True, eeg=False, n_jobs=1) # Load data raw = read_raw_fif(fname_raw, preload=True) noise_cov = mne.read_cov(fname_cov) events = mne.read_events(fname_event) # Add a bad channel raw.info['bads'] += ['MEG 2443'] # Pick MEG channels picks = mne.pick_types(raw.info, meg=True, eeg=False, stim=False, eog=True, exclude='bads') # Define epochs for left-auditory condition event_id, tmin, tmax = 1, -0.2, 0.5 epochs = mne.Epochs(raw, events, event_id, tmin, tmax, picks=picks, baseline=(None, 0), reject=dict(mag=4e-12, grad=4000e-13, eog=150e-6)) # Compute inverse solution and for each epoch snr = 1.0 # use smaller SNR for raw data inv_method = 'dSPM' # sLORETA, MNE, dSPM parc = 'aparc' # the parcellation to use, e.g., 'aparc' 'aparc.a2009s' lambda2 = 1.0 / snr ** 2 # Compute inverse operator inverse_operator = make_inverse_operator(raw.info, fwd, noise_cov, loose=None, depth=None, fixed=False) stcs = apply_inverse_epochs(epochs, inverse_operator, lambda2, inv_method, pick_ori=None, return_generator=True) # Get labels for FreeSurfer 'aparc' cortical parcellation with 34 labels/hemi labels_parc = mne.read_labels_from_annot(subject, parc=parc, subjects_dir=subjects_dir) # Average the source estimates within each label of the cortical parcellation # and each sub structures contained in the src space # If mode = 'mean_flip' this option is used only for the cortical label src = inverse_operator['src'] label_ts = mne.extract_label_time_course(stcs, labels_parc, src, mode='mean_flip', allow_empty=True, return_generator=False) # We compute the connectivity in the alpha band and plot it using a circular # graph layout fmin = 8. fmax = 13. sfreq = raw.info['sfreq'] # the sampling frequency con, freqs, times, n_epochs, n_tapers = spectral_connectivity( label_ts, method='pli', mode='multitaper', sfreq=sfreq, fmin=fmin, fmax=fmax, faverage=True, mt_adaptive=True, n_jobs=1) # We create a list of Label containing also the sub structures labels_aseg = mne.get_volume_labels_from_src(src, subject, subjects_dir) labels = labels_parc + labels_aseg # read colors node_colors = [label.color for label in labels] # We reorder the labels based on their location in the left hemi label_names = [label.name for label in labels] lh_labels = [name for name in label_names if name.endswith('lh')] rh_labels = [name for name in label_names if name.endswith('rh')] # Get the y-location of the label label_ypos_lh = list() for name in lh_labels: idx = label_names.index(name) ypos = np.mean(labels[idx].pos[:, 1]) label_ypos_lh.append(ypos) try: idx = label_names.index('Brain-Stem') ypos = np.mean(labels[idx].pos[:, 1]) lh_labels.append('Brain-Stem') label_ypos_lh.append(ypos) except ValueError: pass # Reorder the labels based on their location lh_labels = [label for (yp, label) in sorted(zip(label_ypos_lh, lh_labels))] # For the right hemi rh_labels = [label[:-2] + 'rh' for label in lh_labels if label != 'Brain-Stem' and label[:-2] + 'rh' in rh_labels] # Save the plot order node_order = list() node_order = lh_labels[::-1] + rh_labels node_angles = circular_layout(label_names, node_order, start_pos=90, group_boundaries=[0, len(label_names) // 2]) # Plot the graph using node colors from the FreeSurfer parcellation. We only # show the 300 strongest connections. conmat = con[:, :, 0] plot_connectivity_circle(conmat, label_names, n_lines=300, node_angles=node_angles, node_colors=node_colors, title='All-to-All Connectivity left-Auditory ' 'Condition (PLI)') # Uncomment the following line to save the figure ''' import matplotlib.pyplot as plt plt.savefig('circle.png', facecolor='black') '''
bsd-3-clause
tabhitmy/MLTF
WORKFLOW/code/python_code/sklearnTrainer.py
1
12746
# sklearnTrainer import numpy import numpy as np import copy from toolkitJ import cell2dmatlab_jsp import matplotlib as mpl from matplotlib.font_manager import FontProperties zhfont = FontProperties(fname="/usr/share/fonts/cjkuni-ukai/ukai.ttc") # 图片显示中文字体 mpl.use('Agg') import pprint from sklearn.externals.six import StringIO # import pydot import sklearn.model_selection as skmdls import sklearn.ensemble as skemb import sklearn.tree as sktree import sklearn.linear_model as sklinmdl import sklearn.discriminant_analysis as skdisa import sklearn.svm as sksvm import sklearn.naive_bayes as sknb import GVal from controlPanelSubFunc_NFDA_J import dVM from trainerSubFunc_NFDA_J import * ################################### # Classifier Subfunction ################ ################################### def adaboost(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) weakClf_list = { 27: 'decisionTree', 271: 'decisionTree' } clf = skemb.AdaBoostClassifier(sktree.DecisionTreeClassifier(max_depth=2, min_samples_split=30, min_samples_leaf=5), algorithm=dVM[2702][2], n_estimators=dVM[2700][2], learning_rate=dVM[2701][2], random_state=dVM[2703][2]) clf.fit(X_tra, y_tra) return processLearning(clf, X_tra, y_tra, X_val, y_val) ## def lda(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clf = skdisa.LinearDiscriminantAnalysis(solver=dVM[2300][2], n_components=dVM[2303][2]) clf.fit(X_tra, y_tra) return processLearning(clf, X_tra, y_tra, X_val, y_val) ## def qda(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clf = skdisa.QuadraticDiscriminantAnalysis() clf.fit(X_tra, y_tra) return processLearning(clf, X_tra, y_tra, X_val, y_val) ## def naiveBayes(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clfname_list = { 25: sknb.GaussianNB, 251: sknb.GaussianNB, 252: sknb.MultinomialNB, 253: sknb.BernoulliNB, } clf = clfname_list[classifier_num]() clf.fit(X_tra, y_tra, sample_weight=weights) return processLearning(clf, X_tra, y_tra, X_val, y_val) ## def svmKernel(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) kernelname_list = { 22: 'rbf', 221: 'rbf', 222: 'poly', 223: 'sigmoid', 224: 'precompute' } kernelname = kernelname_list[classifier_num] clf = sksvm.SVC(C=0.1, kernel=kernelname, degree=3, gamma=0.7) clf.fit(X_tra, y_tra, sample_weight=weights) return processLearning(clf, X_tra, y_tra, X_val, y_val) ## def svmLinear(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clf = sksvm.LinearSVC(penalty=dVM[2100][2], loss=dVM[2101][2], dual=dVM[2102][2], tol=dVM[2103][2], C=dVM[2104][2]) # clf = sksvm.LinearSVC() clf.fit(X_tra, y_tra, sample_weight=weights) cx = clf.coef_[0] clfc = np.around(cx, decimals=2) print('### Feature coefficient with L penalty: ' + str(clfc)) return processLearning(clf, X_tra, y_tra, X_val, y_val) ## def linearRegression(X_tra, y_tra, X_val, y_val, index_no, classifier_num): # Not Applicable at this moment # weights = Y_train_raw[:, 0] # weights[np.nonzero(weights == 0)[0]] = 1 # weights = weights / 7 # y_tra, y_val, X_val = dataRegulation(y_tra, y_val, X_val, index_no) # clf = sklinmdl.LinearRegression() # clf.fit(X_tra, y_tra, sample_weight=weights) # score = clf.score(X_tra, y_tra, sample_weight=weights) # print() # Z = clf.predict(X_val) # print(Z.shape) # TP = np.nonzero(np.logical_and(Z == 1, y_val == 1))[0] # print(TP) # print(TP.shape) # print(max(weights)) # print(min(weights)) return clf, score, FRAP ## def sgdClassifier(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clf = sklinmdl.SGDClassifier(loss='hinge', penalty='l2', alpha=0.1) clf.fit(X_tra, y_tra, sample_weight=weights) return processLearning(clf, X_tra, y_tra, X_val, y_val) ## def logiRegression(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clf = sklinmdl.LogisticRegression(penalty=dVM[3000][2], dual=dVM[3001][2], tol=dVM[3002][2], C=dVM[3003][2], random_state=dVM[3007][2], solver=dVM[3008][2], max_iter=dVM[3009][2]) clf.fit(X_tra, y_tra, sample_weight=weights) return processLearning(clf, X_tra, y_tra, X_val, y_val) ## def decisionTree(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clf = sktree.DecisionTreeClassifier(criterion=dVM[3100][2], splitter=dVM[3101][2], max_depth=dVM[3102][2], min_samples_split=dVM[3103][2], min_samples_leaf=dVM[3104][2], max_features=dVM[3106][2], random_state=dVM[3107][2]) clf.fit(X_tra, y_tra, sample_weight=weights) path = GVal.getPARA('path_PARA') with open(path['fig_path'] + 'dtclf.dot', 'w') as f: f = sktree.export_graphviz(clf, out_file=f, class_names=['0', '1']) # sktree.export_graphviz(clf, out_file=path['fig_path'] + 'tree.dot') # exit() return processLearning(clf, X_tra, y_tra, X_val, y_val) ## def randomForest(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) # http://blog.csdn.net/xuxiatian/article/details/54410086 clf = skemb.RandomForestClassifier(n_estimators=dVM[3200][2], criterion=dVM[3201][2], max_features=dVM[3202][2], max_depth=dVM[3203][2], min_samples_split=dVM[3204][2], min_samples_leaf=dVM[3205][2], min_weight_fraction_leaf=dVM[3206][2], random_state=dVM[3213][2]) # GVal.show('dVM_PARA') clf.fit(X_tra, y_tra, sample_weight=weights) # print(clf.get_params()) # print(clf) path = GVal.getPARA('path_PARA') i_tree = 0 for tree_in_forest in clf.estimators_: with open(path['fig_path'] + '/RF/tree_' + str(i_tree) + '.dot', 'w') as my_file: my_file = sktree.export_graphviz(tree_in_forest, out_file=my_file, class_names=['0', '1']) i_tree = i_tree + 1 return processLearning(clf, X_tra, y_tra, X_val, y_val) def bagging(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clf = skemb.BaggingClassifier(base_estimator=sktree.DecisionTreeClassifier(max_depth=2, min_samples_split=30, min_samples_leaf=5), n_estimators=dVM[3300][2], max_samples=dVM[3301][2], max_features=dVM[3302][2], bootstrap=dVM[3303][2], random_state=dVM[3308][2]) clf.fit(X_tra, y_tra, sample_weight=weights) return processLearning(clf, X_tra, y_tra, X_val, y_val) def voting(X_tra, y_tra, X_val, y_val, index_no, classifier_num): # classifier_list = GVal.getPARA('classifier_list_PARA') # dVM[3400] = ['estimators', [21, 23, 25, 30, 31], [21, 23, 25, 30, 31]] estims = [] for i in range(len(dVM[3400][2])): clf_temp = (classifier_list[dVM[3400][2][i]][1], classifier_list[int(str(dVM[3400][2][i])[0:2])][0](X_tra, y_tra, X_val, y_val, index_no, dVM[3400][2][i])[0]) estims.append(clf_temp) y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clf = skemb.VotingClassifier(estimators=estims, voting=dVM[3401][2]) clf.fit(X_tra, y_tra) return processLearning(clf, X_tra, y_tra, X_val, y_val) def gradboost(X_tra, y_tra, X_val, y_val, index_no, classifier_num): y_tra, X_tra, y_val, X_val, weights = dataRegulationSKL(y_tra, X_tra, y_val, X_val, index_no) clf = skemb.GradientBoostingClassifier(loss=dVM[3500][2], learning_rate=dVM[3501][2], n_estimators=dVM[3502][2], max_depth=dVM[3503][2], criterion=dVM[3504][2], min_samples_split=dVM[3505][2], min_samples_leaf=dVM[3506][2], subsample=dVM[3508][2], random_state=dVM[3515][2]) clf.fit(X_tra, y_tra, sample_weight=weights) return processLearning(clf, X_tra, y_tra, X_val, y_val) ################################### # Main ############################# ################################### def sklearnTrainer(classifier_num, X_train_raw, Y_train_raw, X_valid_raw, Y_valid_raw, path): feature_index = GVal.getPARA('feature_index_PARA') X, y, X_valid, y_valid, index_no = dataSetPreparation(feature_index, X_train_raw, Y_train_raw, X_valid_raw, Y_valid_raw) classifier_list = { 21: [svmLinear, 'Linear SVM', []], 22: [svmKernel, 'Kernel SVM (Default:rbf)'], 221: [svmKernel, 'Kernel SVM (rbf)'], 222: [svmKernel, 'Kernel SVM (poly)'], 223: [svmKernel, 'Kernel SVM (sigmoid)'], 224: [svmKernel, 'Kernel SVM (precompute)'], 23: [lda, 'LDA'], 24: [qda, 'QDA'], 25: [naiveBayes, 'Naive Bayes (Default: Gaussian)'], 251: [naiveBayes, 'Naive Bayes (Guassian)'], 252: [naiveBayes, 'Naive Bayes (Multinominal)'], 253: [naiveBayes, 'Naive Bayes (Bernoulli)'], # 26: neuralNetwork, 27: [adaboost, 'Adaboost'], 271: [adaboost, 'Adaboost(WC:DecisionTree)'], # 28: [linearRegression, 'Linear Regression'], 29: [sgdClassifier, 'SGD Classifier'], 30: [logiRegression, 'Logistic Regression'], 31: [decisionTree, 'Decision Tree'], 32: [randomForest, 'Random Forest'], 33: [bagging, 'bagging with DT'], 34: [voting, 'Voter'], 35: [gradboost, 'Gradient Tree Boosting'] } GVal.setPARA('classifier_list_cache', classifier_list) # classifier serial code: [[model], [training score], [predicting rate]] clf_cache = { 21: cell2dmatlab_jsp([1], 1, []), 22: cell2dmatlab_jsp([1], 1, []), 221: cell2dmatlab_jsp([1], 1, []), 222: cell2dmatlab_jsp([1], 1, []), 223: cell2dmatlab_jsp([1], 1, []), 224: cell2dmatlab_jsp([1], 1, []), 23: cell2dmatlab_jsp([1], 1, []), 24: cell2dmatlab_jsp([1], 1, []), 25: cell2dmatlab_jsp([1], 1, []), 251: cell2dmatlab_jsp([1], 1, []), 252: cell2dmatlab_jsp([1], 1, []), 253: cell2dmatlab_jsp([1], 1, []), 27: cell2dmatlab_jsp([1], 1, []), 271: cell2dmatlab_jsp([1], 1, []), 28: cell2dmatlab_jsp([1], 1, []), 29: cell2dmatlab_jsp([1], 1, []), 30: cell2dmatlab_jsp([1], 1, []), 31: cell2dmatlab_jsp([1], 1, []), 32: cell2dmatlab_jsp([1], 1, []), 33: cell2dmatlab_jsp([1], 1, []), 34: cell2dmatlab_jsp([1], 1, []) } print('### With model: [' + classifier_list[classifier_num][1] + ']') # Loading model to do the classification clf, score, FRAP = classifier_list[int(str(classifier_num)[0:2])][0](X, y, X_valid, y_valid, index_no, classifier_num) clf_cache[classifier_num] = clf # return clf,score,FRAP clf_info = cell2dmatlab_jsp([3], 1, []) clf_info[0] = classifier_num clf_info[1] = classifier_list[classifier_num][1] clf_info[2] = clf return clf_info, score, FRAP
mit
aranega/pyecore
experimental/m2m/transfo_example.py
2
2226
import motra # generated using # https://github.com/kolovos/datasets/blob/master/github-mde/ghmde.ecore # as input metamodel import ghmde from pyecore.ecore import * # Define a graph like metamodel in a static way eClass = EPackage('graph', nsURI='http://graph/1.0', nsPrefix='graph') @EMetaclass class Node(object): name = EAttribute(eType=EString) @EMetaclass class Graph(object): name = EAttribute(eType=EString) nodes = EReference(eType=Node, upper=-1, containment=True) # Transfo definition ghmde2graph = motra.Transformation('ghmde2graph', inputs=['ghmde_model'], outputs=['graph_model']) @ghmde2graph.main def main(ghmde_model=None, graph_model=None): print('Transforming repository to graph', graph_model) for repository in motra.objects_of_kind(ghmde_model, ghmde.File): file2node(repository) for repository in motra.objects_of_kind(ghmde_model, ghmde.Repository): repository2graph(repository, postfix='_graph') def does_not_starts_with(self, postfix): return not self.name.startswith(postfix) @ghmde2graph.mapping(when=does_not_starts_with) def repository2graph(self: ghmde.Repository, postfix: str) -> Graph: result.name = self.name + postfix for repo_file in self.files: result.nodes.append(file2node(repo_file)) @ghmde2graph.mapping def file2node(self: ghmde.File) -> Node: result.name = self.path # @transfo.main # def main(inputs, outputs): # print('in main') # print(inputs.ghmde.contents) # for o in motra.objects_of_kind(inputs.ghmde, ghmde.Repository): # test_dispatch(o) # # # @transfo.mapping(when=lambda self: self.name is not None) # def test1(self: ghmde.Repository) -> ghmde.Repository: # print('changing name', result is self, self.name) # result.name = self.name # self.name += '_toto' # # # @transfo.mapping(output_model='test2', # when=lambda self: self.name is None) # def test2(self: ghmde.Repository) -> ghmde.Repository: # result.name = 'from_empty_' + str(self) # # # @transfo.disjunct(mappings=[test1, test2]) # def test_dispatch(self: ghmde.Repository) -> ghmde.Repository: # pass
bsd-3-clause
vdumoulin/fuel
fuel/converters/cifar100.py
18
3576
import os import tarfile import h5py import numpy import six from six.moves import cPickle from fuel.converters.base import fill_hdf5_file, check_exists DISTRIBUTION_FILE = 'cifar-100-python.tar.gz' @check_exists(required_files=[DISTRIBUTION_FILE]) def convert_cifar100(directory, output_directory, output_filename='cifar100.hdf5'): """Converts the CIFAR-100 dataset to HDF5. Converts the CIFAR-100 dataset to an HDF5 dataset compatible with :class:`fuel.datasets.CIFAR100`. The converted dataset is saved as 'cifar100.hdf5'. This method assumes the existence of the following file: `cifar-100-python.tar.gz` Parameters ---------- directory : str Directory in which the required input files reside. output_directory : str Directory in which to save the converted dataset. output_filename : str, optional Name of the saved dataset. Defaults to 'cifar100.hdf5'. Returns ------- output_paths : tuple of str Single-element tuple containing the path to the converted dataset. """ output_path = os.path.join(output_directory, output_filename) h5file = h5py.File(output_path, mode="w") input_file = os.path.join(directory, 'cifar-100-python.tar.gz') tar_file = tarfile.open(input_file, 'r:gz') file = tar_file.extractfile('cifar-100-python/train') try: if six.PY3: train = cPickle.load(file, encoding='latin1') else: train = cPickle.load(file) finally: file.close() train_features = train['data'].reshape(train['data'].shape[0], 3, 32, 32) train_coarse_labels = numpy.array(train['coarse_labels'], dtype=numpy.uint8) train_fine_labels = numpy.array(train['fine_labels'], dtype=numpy.uint8) file = tar_file.extractfile('cifar-100-python/test') try: if six.PY3: test = cPickle.load(file, encoding='latin1') else: test = cPickle.load(file) finally: file.close() test_features = test['data'].reshape(test['data'].shape[0], 3, 32, 32) test_coarse_labels = numpy.array(test['coarse_labels'], dtype=numpy.uint8) test_fine_labels = numpy.array(test['fine_labels'], dtype=numpy.uint8) data = (('train', 'features', train_features), ('train', 'coarse_labels', train_coarse_labels.reshape((-1, 1))), ('train', 'fine_labels', train_fine_labels.reshape((-1, 1))), ('test', 'features', test_features), ('test', 'coarse_labels', test_coarse_labels.reshape((-1, 1))), ('test', 'fine_labels', test_fine_labels.reshape((-1, 1)))) fill_hdf5_file(h5file, data) h5file['features'].dims[0].label = 'batch' h5file['features'].dims[1].label = 'channel' h5file['features'].dims[2].label = 'height' h5file['features'].dims[3].label = 'width' h5file['coarse_labels'].dims[0].label = 'batch' h5file['coarse_labels'].dims[1].label = 'index' h5file['fine_labels'].dims[0].label = 'batch' h5file['fine_labels'].dims[1].label = 'index' h5file.flush() h5file.close() return (output_path,) def fill_subparser(subparser): """Sets up a subparser to convert the CIFAR100 dataset files. Parameters ---------- subparser : :class:`argparse.ArgumentParser` Subparser handling the `cifar100` command. """ return convert_cifar100
mit
agnusfeec/tattCBIR
lib_sistema.py
1
25313
# -*- coding: utf-8 -*- """ Created on Thu Jul 14 13:36:05 2016 @author: agnus """ #%% def monta_lista_imagens(path = '.', ext='.png'): import os imagens = {} for dirname, dirnames, filenames in os.walk(path): # print path to all filenames with extension py. for filename in filenames: fname_path = os.path.join(dirname, filename) fext = os.path.splitext(fname_path)[1] if fext == ext: #file_dat = [filename, dirname] #imagens.append(file_dat) imagens[filename]=dirname else: continue return imagens #%% def grava_db_imagens(arquivo, imagens): #arquivo = './tatt_c.db' with open(arquivo, 'wb') as db_image_file: for nome_img, caminho in imagens.items(): db_image_file.write(nome_img+ '\t' + caminho + '\n') db_image_file.close() #%% def grava_config(arquivo = './example_mem.cfg'): import ConfigParser config = ConfigParser.RawConfigParser() # When adding sections or items, add them in the reverse order of # how you want them to be displayed in the actual file. # In addition, please note that using RawConfigParser's and the raw # mode of ConfigParser's respective set functions, you can assign # non-string values to keys internally, but will receive an error # when attempting to write to a file or when you get it in non-raw # mode. SafeConfigParser does not allow such assignments to take place. config.add_section('Geral') config.set('Geral', 'Image Database', 'Tatt-C') config.set('Geral', 'Database Image Folder', '/media/sf_Projeto/dataset/tatt_dca/') config.set('Geral', 'Indexa image database', 'True') config.set('Geral', 'Database filename', './tatt_c.db') config.set('Geral', 'Image filename extension','.jpg') config.set('Geral', 'Training File', 'train1') config.set('Geral', 'Testing File', 'test1') config.add_section('Folds') config.set('Folds', 'Folds Folder', '/media/sf_Projeto/dataset/tatt_dca/folds/') config.set('Folds', 'Quantidade subsets', '3') config.set('Folds', 'Subset_1', 'gallery{1}.txt') config.set('Folds', 'Subset_2', 'probes{1}.txt') config.set('Folds', 'Subset_3', 'bg{1}.txt') config.set('Folds', 'Ground_truth', 'ground_truth.txt') config.add_section('SIFT') config.set('SIFT','SIFT Folder', '/media/sf_Projeto/dataset/tatt_dca/SIFT/') # Writing our configuration file to 'example.cfg' with open(arquivo, 'wb') as configfile: config.write(configfile) #%% def folds_construct(subsets, folds_folder): n_folds =len(subsets[0]) n_subsets = len(subsets) folds = [] for i in range(n_folds): sub = [] for j in range(n_subsets): arquivo = subsets[j][i] aux = [] with open(folds_folder+arquivo, 'r') as imagefiles: for nomef in imagefiles: if nomef[-1] == '\n' : nomef = nomef[:-1] aux.append(nomef) imagefiles.close() sub.append(aux) folds.append(sub) return folds #%% def le_config(): import ConfigParser config = ConfigParser.RawConfigParser() config.read('./example_mem.cfg') # getfloat() raises an exception if the value is not a float # getint() and getboolean() also do this for their respective types base = config.get('Geral', 'image database') indexa = config.getboolean('Geral', 'indexa image database') print base if indexa: print "indexa base" arquivo = config.get('Geral','database filename') caminho = config.get('Geral', 'database image folder') extensao = config.get('Geral', 'image filename extension') print arquivo, caminho, extensao imagens = monta_lista_imagens(caminho, extensao) grava_db_imagens(arquivo, imagens) folds_folder = config.get('Folds','folds folder') n_subsets = config.getint('Folds', 'quantidade subsets') subsets=[] for i in range(n_subsets): sub = config.get('Folds', 'subset_'+str(i+1)) ps = sub.find("{") pe = sub.find("}") ped = sub[ps+1:pe] indices = ped.split(',') aux = [] for ind in indices: aux.append(sub[:ps]+ind+'.txt') # incluir extensão variável subsets.append(aux) #print subsets #n_folds = config.getint('Folds', 'quantidade folds') n_folds =len(subsets[0]) folds = [] for i in range(n_folds): sub = [] for j in range(n_subsets): arquivo = subsets[j][i] aux = [] with open(folds_folder+arquivo, 'r') as imagefiles: for nomef in imagefiles: if nomef[-1] == '\n' : nomef = nomef[:-1] aux.append(nomef) imagefiles.close() sub.append(aux) folds.append(sub) #print folds[0] gt_filename = config.get('Folds', 'ground_truth') sift_folder = config.get('SIFT', 'sift folder') print sift_folder, folds_folder, caminho return (folds, imagens, gt_filename, sift_folder, folds_folder, caminho, subsets) #%% def sift(nomes_imagens, imagens, sift_folder): import cv2 import os from math import sqrt #ds = [] #kp = [] t = len(nomes_imagens) i=1 for filename in nomes_imagens: fname = os.path.join(sift_folder, filename[:-3]+'sift_ds') if os.path.isfile(fname) == False : print filename #file_img = os.path.join(diretorio, filename) diretorio = imagens[filename] img = cv2.imread(os.path.join(diretorio, filename)) #file_img) # Redimensiona imagem para aplicação do Fisher Vectors #img = cv2.resize(img, (256,256)) aux = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) gray = cv2.equalizeHist(aux) k = sqrt((240.0*480.0*0.5)/(gray.shape[0]*gray.shape[1])) res = cv2.resize(gray,None,fx=k, fy=k, interpolation = cv2.INTER_CUBIC) cv2.imwrite("/media/sf_Projeto/dataset/tatt_dca//img_Reduzido/"+filename,res) sift = cv2.xfeatures2d.SIFT_create() (kps, descs) = sift.detectAndCompute(res, None) #ds.append(descs) #kp.append(kps) arquivo = os.path.join(sift_folder, filename[:-3]+'sift_ds') with open(arquivo, 'wb') as sift_file: for desc in descs: sift_file.write(','.join(str(x) for x in desc)+'\n') sift_file.close() arquivo = os.path.join(sift_folder, filename[:-3]+'sift_kp') with open(arquivo, 'wb') as sift_file: for point in kps: temp = [point.pt[0], point.pt[1], point.size, point.angle, point.response, point.octave, point.class_id] sift_file.write(','.join(str(x) for x in temp)+'\n') sift_file.close() print (i*100)/t, i=i+1 #return ds #%% def sift_match(ds1, kp1, ds2, kp2): import cv2 MIN_MATCH_COUNT = 10 bf = cv2.BFMatcher() matches = bf.knnMatch(ds1,ds2, k=2) # Apply ratio test good = [] for m,n in matches: if m.distance < 0.7*n.distance: good.append(m) qm = len(good) (nr1,c) = ds1.shape (nr2,c) = ds2.shape # if qm>MIN_MATCH_COUNT: # src_pts = np.float32([ kp1[m.queryIdx].pt for m in good ]).reshape(-1,1,2) # dst_pts = np.float32([ kp2[m.trainIdx].pt for m in good ]).reshape(-1,1,2) # # M, mask = cv2.findHomography(src_pts, dst_pts, cv2.RANSAC,5.0) # if mask != None: # matchesMask = mask.ravel().tolist() # rt = np.sum(np.asarray(matchesMask)) # else: # rt = 0 # else: # #print "Not enough matches are found - %d/%d" % (len(good),MIN_MATCH_COUNT) # #matchesMask = None # rt = 0 nr = nr1 if nr2>nr: nr = nr2 rt = (100.0*qm/nr) # if qm > 0: # rt = 1.0/qm # else: # rt = 10^8 return rt #%% def gera_sift_base(folds, imagens, sift_folder): # Inicialmente gera se necessario o SIFT para as imagens de treinamento e teste # pode ser otimizado, gerando para toda a base, caso se utilize toda a base # o que pode ter um custo alto pois na base existem imagens para outros casos # de uso. n_folds = len(folds) #Poder ser implementado diferente pois as linhas abaixo apenas agregram os nomes #das imagens para que sejam gerados os sifts para cada um dos folds for i in range(n_folds): test = folds[i][1] train = folds[i][0] bg = folds[i][2] for j in range(n_folds): if j!=i : train = train + folds[j][0]+folds[j][1]+folds[j][2] print 'Gerando sift do conjunto de treinamento' #train_kp, train_ds = sift(train, imagens, sift_folder) sift(train, imagens, sift_folder) print 'Gerando sift do conjunto de teste' #test_kp, test_ds = sift(test, imagens) sift(test, imagens, sift_folder) print 'Gerando sift do conjunto de bg' #bg_kp, bg_ds = sift(bg, imagens) sift(bg, imagens, sift_folder) #%% def processa_sift(folds, imagens, sift_folder): import numpy as np import os import cv2 n_folds = len(folds) #Alterei para que inclua nas imagens da galeria i no conj. train, de forma a que as # imagens correspondentes ao probe existam na galeria (train) for i in range(n_folds): test = folds[i][1] bg = folds[i][2] train = folds[i][0]#+bg for j in range(n_folds): if j!=i : train = train + folds[j][0]+folds[j][1]+folds[j][2] n_test = len(test) n_train = len(train) dist = np.zeros((n_train), dtype=np.float) nn = n_test * n_train print 'Gerando o match entre o treinamento e o conjunto de teste' mem = True if mem==True : ds=[] ks=[] arquivo = './clist_mem_'+str(i+1)+'.txt' with open(arquivo, 'w') as clist_file: l = 0 for file_test in test: fname = os.path.join(sift_folder, file_test[:-3]+'sift_ds') ds1 = (np.loadtxt(open(fname,"r"),delimiter=",")).astype(np.uint8) #,skiprows=1) fname = os.path.join(sift_folder, file_test[:-3]+'sift_kp') kps = (np.loadtxt(open(fname,"r"),delimiter=",")).astype(np.float) #,skiprows=1) kp1=[] kp2=[] for kp in kps: kpoint = cv2.KeyPoint(float(kp[0]), float(kp[1]), float(kp[2]), float(kp[3]), float(kp[4]), int(kp[5]), int(kp[6])) kp1.append(kpoint) diretorio = imagens[file_test] img1 = cv2.imread(os.path.join(diretorio, file_test),0) #print os.path.join(diretorio, file_test) j = 0 for file_train in train: diretorio = imagens[file_train] img2 = cv2.imread(os.path.join(diretorio, file_train),0) #print os.path.join(diretorio, file_train) if (mem == True and len(ds)<len(train)): fname = os.path.join(sift_folder, file_train[:-3]+'sift_ds') ds.append ( np.asarray((np.loadtxt(open(fname,"r"),delimiter=",")).astype(np.uint8)) ) #,skiprows=1) ds2 = ds[j] fname = os.path.join(sift_folder, file_train[:-3]+'sift_kp') kps = (np.loadtxt(open(fname,"r"),delimiter=",")).astype(np.float) #,skiprows=1) aux =[] for kp in kps: kpoint = cv2.KeyPoint(float(kp[0]), float(kp[1]), float(kp[2]), float(kp[3]), float(kp[4]), int(kp[5]), int(kp[6])) aux.append(kpoint) ks.append(aux) kp2 = ks[j] elif (mem == True and len(ds)==len(train)): ds2 = ds[j] kp2 = ks[j] elif mem == False: fname = os.path.join(sift_folder, file_train[:-3]+'sift_ds') ds2 = ( (np.loadtxt(open(fname,"r"),delimiter=",")).astype(np.uint8) ) fname = os.path.join(sift_folder, file_train[:-3]+'sift_kp') kps = (np.loadtxt(open(fname,"r"),delimiter=",")).astype(np.float) #,skiprows=1) kp2 = [] for kp in kps: kpoint = cv2.KeyPoint(float(kp[0]), float(kp[1]), float(kp[2]), float(kp[3]), float(kp[4]), int(kp[5]), int(kp[6])) kp2.append(kpoint) #print ds1 #print ds2 rt = sift_match(ds1, np.asarray(kp1), ds2, np.asarray(kp2)) dist[j] = rt j = j + 1 print i,(((l*n_train)+j)*100)/nn, indice = np.argsort(dist)[::-1] k = 1 for id in indice: clist_file.write(file_test+'|'+ str(k) + '|' + train[id] + '|' + str(dist[id]) +'\n') k = k + 1 l = l + 1 clist_file.close() break #%% def ground_truth(folds_folder, gt_filename): """Reads a ground truth table from text file. Keyword arguments: folds_folder -- the path for the ground truth file gt_filename -- the file name of the ground truth file with extension Returns: gt_images -- ground truth table stored in a dictionary """ #folds_folder = '/media/sf_Projeto/dataset/tatt-c_update_v1.4/5-fold/tattoo_identification/' #gt_filename = 'ground_truth.txt' gt_imagens = {} with open(folds_folder+gt_filename, 'r') as gt_arq: for nomef in gt_arq: imgs = nomef.split('|') if imgs[1][-1] == '\n' : imgs[1] = imgs[1][:-1] #print imgs[0], imgs[1] gt_imagens[imgs[0]] = imgs[1] gt_arq.close() return gt_imagens #%% def compute_cmc(arquivo, gt_imagens): """Reads a classification list from text file and sumarize rank results for every image reference based in the ground truth dictionary. Keyword arguments: arquivo -- the filename of classification list file gt_images -- ground truth table stored in a dictionary Returns: cmc -- acummulated accuracy for each rank stored in a numpy array """ import numpy as np i = 0 acc = np.zeros(400) #arquivo = './clist_mem_'+str(i+1)+'.txt' with open(arquivo, 'r') as clist_file: for nomef in clist_file: imgs = nomef.split('|') if imgs[3][-1] == '\n' : imgs[3] = imgs[3][:-1] if gt_imagens[imgs[0]] == imgs[2] : r = int(imgs[1]) acc[r] = acc[r]+1 clist_file.close() #print cmc ft = sum(acc) #print cmc/ft cmc = np.zeros(400) for i in range(1,400): cmc[i] = cmc[i-1]+acc[i]/ft #print cmc1 return cmc #%% def plot_cmc(cmc, ni=200): import matplotlib.pyplot as plt import pylab as P import numpy as np fig = P.figure() fig.suptitle('Acumulative Match Characteristic', fontsize=18, fontweight='bold') P.ylabel('%', fontsize=16) P.xlabel('Rank', fontsize=16) P.xlim(0, ni) P.ylim(0,101) P.xticks(np.arange(0, ni, 10.0)) P.yticks(np.arange(0, 101, 5.0)) xticklabels = P.getp(P.gca(), 'xticklabels') yticklabels = P.getp(P.gca(), 'yticklabels') P.setp(yticklabels, 'color', 'k', fontsize='x-large') P.setp(xticklabels, 'color', 'k', fontsize='x-large') P.grid(True) fig.set_size_inches(19,7) #P.plot(cmc*100) P.plot(cmc*100) fig.savefig('cmc_bf_knn.png') P.show() #%%% #Author: Jacob Gildenblat, 2014 #http://jacobcv.blogspot.com.br/2014/12/fisher-vector-in-python.html #License: you may use this for whatever you like #Adaptation: Agnus A. Horta def fv_dictionary(descriptors, N): import numpy as np import cv2 em = cv2.ml.EM_create() em.setClustersNumber(N) #em = cv2.EM(N) em.trainEM(descriptors) return np.float32(em.getMeans()), \ np.float32(em.getCovs()), np.float32(em.getWeights())[0] def fv_generate_gmm(descriptors, N, dt): import numpy as np words = np.concatenate(descriptors) #np.concatenate([folder_descriptors(folder) for folder in glob.glob(input_folder + '*')]) #print("Training GMM of size", N) means, covs, weights = fv_dictionary(words, N) #Throw away gaussians with weights that are too small: th = 1.0 / N means = np.float32([m for k,m in zip(range(0, len(weights)), means) if weights[k] > th]) covs = np.float32([m for k,m in zip(range(0, len(weights)), covs) if weights[k] > th]) weights = np.float32([m for k,m in zip(range(0, len(weights)), weights) if weights[k] > th]) #print 'Means: ',means #print 'Covs: ',covs #print 'Weights: ',weights np.save("./dat/means" + dt + ".gmm", means) np.save("./dat/covs" + dt + ".gmm", covs) np.save("./dat/weights" + dt + ".gmm", weights) return means, covs, weights def fv_load_gmm(dt, folder = "./dat"): import numpy as np files = ["means" + dt + ".gmm" +".npy", "covs" + dt + ".gmm.npy", "weights" + dt + ".gmm.npy"] try: return map(lambda file: np.load(file), map(lambda s : folder + "/" + s , files)) except IOError: return (None, None, None) def fv_likelihood_moment(x, ytk, moment): import numpy as np x_moment = np.power(np.float32(x), moment) if moment > 0 else np.float32([1]) return x_moment * ytk def fv_likelihood_statistics(samples, means, covs, weights): from scipy.stats import multivariate_normal import numpy as np gaussians, s0, s1,s2 = {}, {}, {}, {} samples = zip(range(0, len(samples)), samples) #print samples g = [multivariate_normal(mean=means[k], cov=covs[k]) for k in range(0, len(weights)) ] for index, x in samples: gaussians[index] = np.array([g_k.pdf(x) for g_k in g]) for k in range(0, len(weights)): s0[k], s1[k], s2[k] = 0, 0, 0 for index, x in samples: probabilities = np.multiply(gaussians[index], weights) probabilities = probabilities / np.sum(probabilities) s0[k] = s0[k] + fv_likelihood_moment(x, probabilities[k], 0) s1[k] = s1[k] + fv_likelihood_moment(x, probabilities[k], 1) s2[k] = s2[k] + fv_likelihood_moment(x, probabilities[k], 2) return s0, s1, s2 def fv_fisher_vector_weights(s0, s1, s2, means, covs, w, T): import numpy as np return np.float32([((s0[k] - T * w[k]) / np.sqrt(w[k]) ) for k in range(0, len(w))]) def fv_fisher_vector_means(s0, s1, s2, means, sigma, w, T): import numpy as np return np.float32([(s1[k] - means[k] * s0[k]) / (np.sqrt(w[k] * sigma[k])) for k in range(0, len(w))]) def fv_fisher_vector_sigma(s0, s1, s2, means, sigma, w, T): import numpy as np return np.float32([(s2[k] - 2 * means[k]*s1[k] + (means[k]*means[k] - sigma[k]) * s0[k]) / (np.sqrt(2*w[k])*sigma[k]) for k in range(0, len(w))]) def fv_normalize(fisher_vector): import numpy as np v = np.sqrt(abs(fisher_vector)) * np.sign(fisher_vector) return v / np.sqrt(np.dot(v, v)) def fv_fisher_vector(samples, means, covs, w): import numpy as np #print 'fisher_vector(samples, means, covs, w)' s0, s1, s2 = fv_likelihood_statistics(samples, means, covs, w) T = samples.shape[0] covs = np.float32([np.diagonal(covs[k]) for k in range(0, covs.shape[0])]) a = fv_fisher_vector_weights(s0, s1, s2, means, covs, w, T) b = fv_fisher_vector_means(s0, s1, s2, means, covs, w, T) c = fv_fisher_vector_sigma(s0, s1, s2, means, covs, w, T) fv = np.concatenate([np.concatenate(a), np.concatenate(b), np.concatenate(c)]) fv = fv_normalize(fv) #print 'fv = ', fv return fv def le_descritores(sift_folder, subset, tipo=1): import os import numpy as np #n_folds = len(folds) #Alterei para que inclua nas imagens da galeria i no conj. train, de forma a que as # imagens correspondentes ao probe existam na galeria (train) # for i in range(n_folds): # train = folds[i][0] # for j in range(n_folds): # if j!=i : # train = train + folds[j][0]+folds[j][1]+folds[j][2] # # n_train = len(train) ch = 0 ds = [] id_ds = [] for image in subset: fname = os.path.join(sift_folder, image[:-3]+'sift_ds') ds1 = (np.loadtxt(open(fname,"r"),delimiter=",")).astype(np.uint8) #,skiprows=1) if tipo == 1: if ch == 0: ch = 1 ds = [] ds.append(ds1) id_ds.append(ds1.shape[0]) else: ds.append(ds1) id_ds.append(ds1.shape[0]) else: if ch == 0: ch = 1 ds = np.empty_like(ds1) ds[:] = ds1 id_ds.append(ds1.shape[0]) else: print ds.shape, ds1.shape ds = np.concatenate((ds, ds1), axis=0) id_ds.append(ds1.shape[0]) return ds, id_ds #%% def bov_histogramas_grava(arquivo, hists, dt): resultFile = open(arquivo, 'w') i = len(hists) for h in hists: line = (''.join(str(e) + ", " for e in h.tolist()))[:-2] resultFile.write(line) if i > 0: resultFile.write("\n") i = i - 1 resultFile.close() #%% def bov_codebook_gera(l_sift, nc, tipo): if tipo == 1: # http://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans.fit from sklearn.cluster import KMeans est = KMeans(n_clusters=nc, init='k-means++', n_init=10, max_iter=100, tol=0.0001, precompute_distances='auto', verbose=0, random_state=None, copy_x=True, n_jobs=4) est.fit(l_sift) labels = est.labels_ centers = est.cluster_centers_ elif tipo == 2: from sklearn.cluster import MiniBatchKMeans est = MiniBatchKMeans(n_clusters=nc, init='k-means++', max_iter=100, batch_size=3*nc, verbose=0, compute_labels=True, random_state=None, tol=0.0, max_no_improvement=10, init_size=None, n_init=3, reassignment_ratio=0.01) est.fit(l_sift) labels = est.labels_ centers = est.cluster_centers_ else: import random from scipy.cluster.vq import vq import numpy as np list_of_random_items = random.sample(np.arange(l_sift.shape[0]), nc) l_centroids = [] for i in list_of_random_items: l_centroids.append(l_sift[i]) centers = np.asarray(l_centroids) labels, _ = vq(l_sift, centers) return (centers, labels) #%% def bov_histogramas_gera(labels, id_ds, k, nomes_imagens, vis=False): from matplotlib import pyplot as plt import numpy as np #fv = np.vectorize(f) hists = [] i = 0 for j in range(len(nomes_imagens)): #ld = X[indices[j]].tolist() n = id_ds[j] sl = labels[i:i+n] hist, bins = np.histogram(sl, bins=k, range=(0, k), normed=False, weights=None, density=True) if vis == True: width = 0.7 * (bins[1] - bins[0]) center = (bins[:-1] + bins[1:]) / 2 plt.title("Histogram "+nomes_imagens[j]) plt.xlabel("Visual Word") plt.ylabel("Frequency") plt.bar(center, hist, align='center', width=width) plt.show() #print j hists.append(hist) #print hist i = i + n #j = j +1 return hists def bov_descritores_codifica(X, centers): from scipy.cluster.vq import vq labels,_ = vq(X,centers) return labels
gpl-3.0
aalmah/pylearn2
pylearn2/devtools/tests/test_format.py
24
25785
""" Unit tests for format checking """ from __future__ import print_function from nose.plugins.skip import SkipTest import os import pylearn2 from pylearn2.devtools.tests.docscrape import docstring_errors from pylearn2.devtools.list_files import list_files from pylearn2.devtools.tests.pep8.pep8 import StyleGuide whitelist_pep8 = [ "rbm_tools.py", "distributions/mnd.py", "models/sparse_autoencoder.py", "models/tests/test_dbm.py", "models/tests/test_s3c_inference.py", "models/tests/test_mnd.py", "models/tests/test_s3c_misc.py", "models/gsn.py", "models/dbm/layer.py", "models/dbm/__init__.py", "models/dbm/ising.py", "models/differentiable_sparse_coding.py", "models/local_coordinate_coding.py", "models/mnd.py", "models/s3c.py", "tests/test_monitor.py", "kmeans.py", "packaged_dependencies/theano_linear/conv2d.py", "packaged_dependencies/theano_linear/imaging.py", "packaged_dependencies/theano_linear/pyramid.py", "packaged_dependencies/theano_linear/unshared_conv/" "test_gpu_unshared_conv.py", "packaged_dependencies/theano_linear/unshared_conv/" "test_localdot.py", "packaged_dependencies/theano_linear/unshared_conv/localdot.py", "packaged_dependencies/theano_linear/unshared_conv/" "unshared_conv.py", "packaged_dependencies/theano_linear/linear.py", "packaged_dependencies/theano_linear/test_spconv.py", "packaged_dependencies/theano_linear/test_matrixmul.py", "packaged_dependencies/theano_linear/spconv.py", "expr/tests/test_coding.py", "expr/tests/test_normalize.py", "expr/tests/test_stochastic_pool.py", "expr/stochastic_pool.py", "expr/sampling.py", "expr/information_theory.py", "expr/basic.py", "gui/graph_2D.py", "sandbox/cuda_convnet/weight_acts.py", "sandbox/cuda_convnet/filter_acts.py", "sandbox/cuda_convnet/tests/test_filter_acts_strided.py", "sandbox/cuda_convnet/tests/test_probabilistic_max_pooling.py", "sandbox/cuda_convnet/tests/test_filter_acts.py", "sandbox/cuda_convnet/tests/test_weight_acts_strided.py", "sandbox/cuda_convnet/tests/test_image_acts_strided.py", "sandbox/cuda_convnet/tests/test_img_acts.py", "sandbox/cuda_convnet/tests/test_stochastic_pool.py", "sandbox/cuda_convnet/specialized_bench.py", "sandbox/cuda_convnet/response_norm.py", "sandbox/cuda_convnet/__init__.py", "sandbox/cuda_convnet/img_acts.py", "sandbox/cuda_convnet/convnet_compile.py", "sandbox/cuda_convnet/pthreads.py", "sandbox/cuda_convnet/pool.py", "sandbox/cuda_convnet/bench.py", "sandbox/cuda_convnet/stochastic_pool.py", "sandbox/cuda_convnet/probabilistic_max_pooling.py", "sandbox/tuple_var.py", "sandbox/lisa_rl/bandit/average_agent.py", "sandbox/lisa_rl/bandit/classifier_bandit.py", "sandbox/lisa_rl/bandit/classifier_agent.py", "sandbox/lisa_rl/bandit/plot_reward.py", "config/old_config.py", "utils/utlc.py", "utils/tests/test_serial.py", "utils/common_strings.py", "utils/mem.py", "dataset_get/dataset-get.py", "dataset_get/helper-scripts/make-archive.py", "dataset_get/dataset_resolver.py", "optimization/minres.py", "linear/conv2d.py", "linear/local_c01b.py", "linear/linear_transform.py", "linear/conv2d_c01b.py", "energy_functions/rbm_energy.py", "scripts/pkl_inspector.py", "scripts/show_binocular_greyscale_examples.py", "scripts/jobman/tester.py", "scripts/dbm/dbm_metrics.py", "scripts/papers/maxout/svhn_preprocessing.py", "scripts/papers/jia_huang_wkshp_11/fit_final_model.py", "scripts/papers/jia_huang_wkshp_11/evaluate.py", "scripts/papers/jia_huang_wkshp_11/extract_features.py", "scripts/papers/jia_huang_wkshp_11/assemble.py", "scripts/gpu_pkl_to_cpu_pkl.py", "scripts/gsn_example.py", "scripts/tutorials/deep_trainer/run_deep_trainer.py", "scripts/tutorials/grbm_smd/test_grbm_smd.py", "scripts/icml_2013_wrepl/multimodal/" "extract_layer_2_kmeans_features.py", "scripts/icml_2013_wrepl/multimodal/make_submission.py", "scripts/icml_2013_wrepl/multimodal/lcn.py", "scripts/icml_2013_wrepl/multimodal/extract_kmeans_features.py", "scripts/icml_2013_wrepl/emotions/emotions_dataset.py", "scripts/icml_2013_wrepl/emotions/make_submission.py", "scripts/icml_2013_wrepl/black_box/black_box_dataset.py", "scripts/icml_2013_wrepl/black_box/make_submission.py", "scripts/diff_monitor.py", "corruption.py", "sandbox/lisa_rl/bandit/gaussian_bandit.py", "utils/track_version.py", "scripts/get_version.py", "training_algorithms/tests/test_bgd.py", "training_algorithms/tests/test_default.py", "training_algorithms/default.py", "training_algorithms/training_algorithm.py", "distributions/tests/test_mnd.py", "distributions/parzen.py", "distributions/uniform_hypersphere.py", "models/setup.py", "models/independent_multiclass_logistic.py", "models/softmax_regression.py", "models/tests/test_reflection_clip.py", "models/tests/test_maxout.py", "models/tests/test_convelemwise_sigm.py", "models/dbm/sampling_procedure.py", "models/rbm.py", "models/pca.py", "tests/test_train.py", "packaged_dependencies/theano_linear/unshared_conv/gpu_unshared_conv.py", "packaged_dependencies/theano_linear/unshared_conv/test_unshared_conv.py", "packaged_dependencies/theano_linear/linearmixin.py", "packaged_dependencies/theano_linear/util.py", "packaged_dependencies/theano_linear/__init__.py", "packaged_dependencies/theano_linear/test_linear.py", "expr/tests/test_nnet.py", "expr/image.py", "expr/coding.py", "expr/normalize.py", "expr/probabilistic_max_pooling.py", "testing/tests/test.py", "testing/skip.py", "testing/prereqs.py", "testing/__init__.py", "gui/get_weights_report.py", "gui/patch_viewer.py", "sandbox/cuda_convnet/tests/test_response_norm.py", "sandbox/cuda_convnet/tests/profile_probabilistic_max_pooling.py", "sandbox/cuda_convnet/tests/test_rop_pool.py", "sandbox/cuda_convnet/tests/test_pool.py", "sandbox/cuda_convnet/tests/test_common.py", "sandbox/cuda_convnet/shared_code.py", "sandbox/cuda_convnet/code_templates.py", "sandbox/lisa_rl/bandit/agent.py", "sandbox/lisa_rl/bandit/algorithm.py", "sandbox/lisa_rl/bandit/environment.py", "sandbox/lisa_rl/__init__.py", "datasets/avicenna.py", "datasets/iris.py", "datasets/adult.py", "datasets/npy_npz.py", "datasets/control.py", "datasets/cifar100.py", "datasets/transformer_dataset.py", "termination_criteria/__init__.py", "__init__.py", "utils/logger.py", "utils/tests/test_mnist_ubyte.py", "utils/tests/test_data_specs.py", "utils/tests/test_bit_strings.py", "utils/tests/test_iteration.py", "utils/theano_graph.py", "utils/__init__.py", "utils/datasets.py", "utils/data_specs.py", "utils/insert_along_axis.py", "utils/environ.py", "utils/call_check.py", "utils/python26.py", "deprecated/classifier.py", "train.py", "classifier.py", "dataset_get/helper-scripts/make-sources.py", "pca.py", "optimization/test_linesearch.py", "optimization/test_minres.py", "optimization/test_batch_gradient_descent.py", "optimization/linear_cg.py", "optimization/test_feature_sign.py", "optimization/feature_sign.py", "optimization/test_linear_cg.py", "optimization/linesearch.py", "linear/tests/test_conv2d.py", "linear/tests/test_conv2d_c01b.py", "linear/matrixmul.py", "energy_functions/energy_function.py", "scripts/make_weights_image.py", "scripts/plot_monitor.py", "scripts/print_monitor.py", "scripts/num_parameters.py", "scripts/benchmark/time_relu.py", "scripts/jobman/experiment.py", "scripts/jobman/__init__.py", "scripts/dbm/show_negative_chains.py", "scripts/papers/maxout/compute_test_err.py", "scripts/papers/jia_huang_wkshp_11/npy2mat.py", "scripts/datasets/step_through_small_norb.py", "scripts/datasets/step_through_norb_foveated.py", "scripts/datasets/make_downsampled_stl10.py", "scripts/datasets/browse_small_norb.py", "scripts/datasets/make_mnistplus.py", "scripts/mlp/predict_csv.py", "scripts/find_gpu_fields.py", "scripts/tutorials/deep_trainer/test_deep_trainer.py", "scripts/icml_2013_wrepl/multimodal/make_wordlist.py", "base.py", "devtools/tests/test_via_pyflakes.py", "devtools/tests/test_shebangs.py", "devtools/tests/pep8/pep8.py", "devtools/tests/docscrape.py", "devtools/run_pyflakes.py", "devtools/record.py", "train_extensions/tests/test_window_flip.py", "train_extensions/__init__.py", ] whitelist_docstrings = [ 'scripts/datasets/step_through_norb_foveated.py', 'blocks.py', 'datasets/hdf5.py', 'rbm_tools.py', 'training_algorithms/tests/test_bgd.py', 'training_algorithms/tests/test_sgd.py', 'training_algorithms/tests/test_default.py', 'training_algorithms/bgd.py', 'training_algorithms/default.py', 'training_algorithms/training_algorithm.py', 'training_algorithms/__init__.py', 'training_algorithms/sgd.py', 'distributions/tests/test_mnd.py', 'distributions/multinomial.py', 'distributions/parzen.py', 'distributions/__init__.py', 'distributions/mnd.py', 'distributions/uniform_hypersphere.py', 'models/setup.py', 'models/independent_multiclass_logistic.py', 'models/softmax_regression.py', 'models/sparse_autoencoder.py', 'models/tests/test_reflection_clip.py', 'models/tests/test_dbm.py', 'models/tests/test_gsn.py', 'models/tests/test_dropout.py', 'models/tests/test_autoencoder.py', 'models/tests/test_mlp.py', 'models/tests/test_s3c_inference.py', 'models/tests/test_maxout.py', 'models/tests/test_mnd.py', 'models/tests/test_vae.py', 'models/tests/test_rbm.py', 'models/tests/test_s3c_misc.py', 'models/gsn.py', 'models/dbm/sampling_procedure.py', 'models/dbm/layer.py', 'models/dbm/__init__.py', 'models/dbm/dbm.py', 'models/dbm/ising.py', 'models/differentiable_sparse_coding.py', 'models/local_coordinate_coding.py', 'models/maxout.py', 'models/s3c.py', 'models/mnd.py', 'models/rbm.py', 'models/autoencoder.py', 'tests/test_dbm_metrics.py', 'tests/test_monitor.py', 'tests/test_train.py', 'tests/rbm/test_ais.py', 'kmeans.py', 'packaged_dependencies/__init__.py', 'packaged_dependencies/theano_linear/imaging.py', 'packaged_dependencies/theano_linear/unshared_conv/__init__.py', 'packaged_dependencies/theano_linear/unshared_conv/unshared_conv.py', 'packaged_dependencies/theano_linear/linearmixin.py', 'packaged_dependencies/theano_linear/linear.py', 'packaged_dependencies/theano_linear/test_spconv.py', 'expr/activations.py', 'expr/tests/test_probabilistic_max_pooling.py', 'expr/tests/test_preprocessing.py', 'expr/tests/test_nnet.py', 'expr/tests/test_coding.py', 'expr/tests/test_normalize.py', 'expr/tests/test_stochastic_pool.py', 'expr/preprocessing.py', 'expr/image.py', 'expr/coding.py', 'expr/__init__.py', 'expr/stochastic_pool.py', 'expr/sampling.py', 'expr/normalize.py', 'expr/probabilistic_max_pooling.py', 'expr/information_theory.py', 'expr/basic.py', 'testing/tests/test.py', 'testing/skip.py', 'testing/prereqs.py', 'testing/__init__.py', 'testing/datasets.py', 'gui/get_weights_report.py', 'gui/__init__.py', 'gui/patch_viewer.py', 'scalar.py', 'sandbox/cuda_convnet/weight_acts.py', 'sandbox/cuda_convnet/filter_acts.py', 'sandbox/cuda_convnet/tests/test_filter_acts_strided.py', 'sandbox/cuda_convnet/tests/test_probabilistic_max_pooling.py', 'sandbox/cuda_convnet/tests/test_filter_acts.py', 'sandbox/cuda_convnet/tests/test_img_acts.py', 'sandbox/cuda_convnet/tests/test_response_norm.py', 'sandbox/cuda_convnet/tests/profile_probabilistic_max_pooling.py', 'sandbox/cuda_convnet/tests/test_weight_acts.py', 'sandbox/cuda_convnet/tests/test_rop_pool.py', 'sandbox/cuda_convnet/tests/test_pool.py', 'sandbox/cuda_convnet/tests/test_common.py', 'sandbox/cuda_convnet/tests/test_stochastic_pool.py', 'sandbox/cuda_convnet/shared_code.py', 'sandbox/cuda_convnet/__init__.py', 'sandbox/cuda_convnet/img_acts.py', 'sandbox/cuda_convnet/base_acts.py', 'sandbox/cuda_convnet/pool.py', 'sandbox/cuda_convnet/stochastic_pool.py', 'sandbox/cuda_convnet/code_templates.py', 'sandbox/cuda_convnet/probabilistic_max_pooling.py', 'sandbox/tuple_var.py', 'sandbox/__init__.py', 'sandbox/lisa_rl/bandit/simulator.py', 'sandbox/lisa_rl/bandit/agent.py', 'sandbox/lisa_rl/bandit/algorithm.py', 'sandbox/lisa_rl/bandit/environment.py', 'sandbox/lisa_rl/bandit/average_agent.py', 'sandbox/lisa_rl/bandit/classifier_bandit.py', 'sandbox/lisa_rl/bandit/__init__.py', 'sandbox/lisa_rl/bandit/classifier_agent.py', 'sandbox/lisa_rl/bandit/gaussian_bandit.py', 'sandbox/lisa_rl/__init__.py', 'config/old_config.py', 'config/tests/test_yaml_parse.py', 'config/yaml_parse.py', 'space/tests/test_space.py', 'space/__init__.py', 'datasets/norb.py', 'datasets/utlc.py', 'datasets/mnistplus.py', 'datasets/cos_dataset.py', 'datasets/cifar10.py', 'datasets/svhn.py', 'datasets/tests/test_preprocessing.py', 'datasets/tests/test_mnist.py', 'datasets/tests/test_imports.py', 'datasets/tests/test_cifar10.py', 'datasets/tests/test_norb.py', 'datasets/tests/test_dense_design_matrix.py', 'datasets/tests/test_vector_spaces_dataset.py', 'datasets/tests/test_four_regions.py', 'datasets/tests/test_csv_dataset.py', 'datasets/tests/test_icml07.py', 'datasets/tests/test_utlc.py', 'datasets/preprocessing.py', 'datasets/avicenna.py', 'datasets/iris.py', 'datasets/config.py', 'datasets/dense_design_matrix.py', 'datasets/adult.py', 'datasets/tfd.py', 'datasets/icml07.py', 'datasets/filetensor.py', 'datasets/npy_npz.py', 'datasets/hepatitis.py', 'datasets/wiskott.py', 'datasets/control.py', 'datasets/exc.py', 'datasets/__init__.py', 'datasets/mnist.py', 'datasets/sparse_dataset.py', 'datasets/csv_dataset.py', 'datasets/cifar100.py', 'datasets/tl_challenge.py', 'datasets/transformer_dataset.py', 'datasets/norb_small.py', 'datasets/retina.py', 'datasets/ocr.py', 'datasets/stl10.py', 'datasets/matlab_dataset.py', 'datasets/vector_spaces_dataset.py', 'datasets/four_regions.py', 'datasets/debug.py', 'datasets/binarizer.py', 'termination_criteria/__init__.py', '__init__.py', 'utils/utlc.py', 'utils/setup.py', 'utils/compile.py', 'utils/logger.py', 'utils/general.py', 'utils/testing.py', 'utils/tests/test_mnist_ubyte.py', 'utils/tests/test_data_specs.py', 'utils/tests/test_video.py', 'utils/tests/test_bit_strings.py', 'utils/tests/test_rng.py', 'utils/tests/test_pooling.py', 'utils/tests/test_iteration.py', 'utils/tests/test_insert_along_axis.py', 'utils/tests/test_utlc.py', 'utils/tests/test_compile.py', 'utils/tests/test_key_aware.py', 'utils/key_aware.py', 'utils/video.py', 'utils/bit_strings.py', 'utils/iteration.py', 'utils/pooling.py', 'utils/theano_graph.py', 'utils/common_strings.py', 'utils/datasets.py', 'utils/data_specs.py', 'utils/shell.py', 'utils/rng.py', 'utils/insert_along_axis.py', 'utils/environ.py', 'utils/call_check.py', 'utils/mnist_ubyte.py', 'utils/track_version.py', 'utils/mem.py', 'utils/python26.py', 'utils/timing.py', 'deprecated/__init__.py', 'deprecated/classifier.py', 'train.py', 'format/tests/test_target_format.py', 'format/__init__.py', 'dataset_get/dataset-get.py', 'dataset_get/helper-scripts/make-sources.py', 'dataset_get/helper-scripts/make-archive.py', 'dataset_get/dataset_resolver.py', 'pca.py', 'monitor.py', 'optimization/batch_gradient_descent.py', 'optimization/__init__.py', 'optimization/test_batch_gradient_descent.py', 'optimization/linear_cg.py', 'optimization/minres.py', 'optimization/test_feature_sign.py', 'optimization/feature_sign.py', 'optimization/linesearch.py', 'linear/conv2d.py', 'linear/tests/test_matrixmul.py', 'linear/local_c01b.py', 'linear/matrixmul.py', 'linear/__init__.py', 'linear/linear_transform.py', 'linear/conv2d_c01b.py', 'energy_functions/tests/__init__.py', 'energy_functions/rbm_energy.py', 'energy_functions/__init__.py', 'energy_functions/energy_function.py', 'scripts/plot_monitor.py', 'scripts/print_model.py', 'scripts/tests/__init__.py', 'scripts/pkl_inspector.py', 'scripts/get_version.py', 'scripts/print_monitor.py', 'scripts/show_binocular_greyscale_examples.py', 'scripts/num_parameters.py', 'scripts/jobman/tester.py', 'scripts/jobman/experiment.py', 'scripts/jobman/__init__.py', 'scripts/dbm/__init__.py', 'scripts/dbm/dbm_metrics.py', 'scripts/papers/__init__.py', 'scripts/papers/jia_huang_wkshp_11/extract_features.py', 'scripts/print_channel_doc.py', 'scripts/gpu_pkl_to_cpu_pkl.py', 'scripts/datasets/step_through_small_norb.py', 'scripts/datasets/download_mnist.py', 'scripts/datasets/download_binarized_mnist.py', 'scripts/datasets/browse_small_norb.py', 'scripts/datasets/make_mnistplus.py', 'scripts/__init__.py', 'scripts/gsn_example.py', 'scripts/mlp/predict_csv.py', 'scripts/mlp/__init__.py', 'scripts/find_gpu_fields.py', 'scripts/tutorials/dbm_demo/train_dbm.py', 'scripts/tutorials/dbm_demo/__init__.py', 'scripts/tutorials/tests/test_dbm.py', 'scripts/tutorials/tests/test_mlp_nested.py', 'scripts/tutorials/multilayer_perceptron/tests/test_mlp.py', 'scripts/tutorials/softmax_regression/tests/test_softmaxreg.py', 'scripts/tutorials/deep_trainer/__init__.py', 'scripts/tutorials/deep_trainer/run_deep_trainer.py', 'scripts/tutorials/grbm_smd/make_dataset.py', 'scripts/tutorials/grbm_smd/__init__.py', 'scripts/tutorials/grbm_smd/test_grbm_smd.py', 'scripts/tutorials/__init__.py', 'scripts/tutorials/jobman_demo/utils.py', 'scripts/tutorials/jobman_demo/__init__.py', 'scripts/tutorials/stacked_autoencoders/tests/test_dae.py', 'scripts/icml_2013_wrepl/__init__.py', 'scripts/icml_2013_wrepl/multimodal/extract_layer_2_kmeans_features.py', 'scripts/icml_2013_wrepl/multimodal/make_submission.py', 'scripts/icml_2013_wrepl/multimodal/lcn.py', 'scripts/icml_2013_wrepl/multimodal/__init__.py', 'scripts/icml_2013_wrepl/multimodal/extract_kmeans_features.py', 'scripts/icml_2013_wrepl/emotions/emotions_dataset.py', 'scripts/icml_2013_wrepl/emotions/make_submission.py', 'scripts/icml_2013_wrepl/emotions/__init__.py', 'scripts/icml_2013_wrepl/black_box/black_box_dataset.py', 'scripts/icml_2013_wrepl/black_box/make_submission.py', 'scripts/icml_2013_wrepl/black_box/__init__.py', 'scripts/diff_monitor.py', 'base.py', 'devtools/tests/test_via_pyflakes.py', 'devtools/tests/test_shebangs.py', 'devtools/tests/__init__.py', 'devtools/tests/docscrape.py', 'devtools/run_pyflakes.py', 'devtools/__init__.py', 'devtools/record.py', 'corruption.py', 'datasets/tests/test_tl_challenge.py', 'datasets/tests/test_tfd.py', 'datasets/tests/test_npy_npz.py', 'linear/tests/test_conv2d.py', 'devtools/tests/pep8/pep8.py', 'devtools/tests/pep8/__init__.py', 'scripts/lcc_tangents/make_dataset.py', 'scripts/icml_2013_wrepl/multimodal/make_wordlist.py', 'scripts/datasets/make_stl10_whitened.py', 'scripts/datasets/make_stl10_patches_8x8.py', 'scripts/datasets/make_stl10_patches.py', 'scripts/datasets/make_cifar10_whitened.py', 'scripts/datasets/make_cifar10_gcn_whitened.py', 'scripts/datasets/make_cifar100_patches.py', 'scripts/datasets/make_cifar100_gcn_whitened.py', 'scripts/datasets/make_svhn_pytables.py', 'energy_functions/tests/test_rbm_energy.py', ] # add files which fail to run to whitelist_docstrings whitelist_docstrings.extend([ 'sandbox/rnn/models/mlp_hook.py', 'training_algorithms/tests/test_learning_rule.py', 'models/pca.py', 'datasets/tests/test_hdf5.py', 'linear/tests/test_conv2d_c01b.py', 'packaged_dependencies/theano_linear/conv2d.py', 'packaged_dependencies/theano_linear/pyramid.py', 'packaged_dependencies/theano_linear/unshared_conv/gpu_unshared_conv.py', 'packaged_dependencies/theano_linear/unshared_conv/' 'test_gpu_unshared_conv.py', 'packaged_dependencies/theano_linear/unshared_conv/test_localdot.py', 'packaged_dependencies/theano_linear/unshared_conv/test_unshared_conv.py', 'packaged_dependencies/theano_linear/unshared_conv/localdot.py', 'packaged_dependencies/theano_linear/util.py', 'packaged_dependencies/theano_linear/__init__.py', 'packaged_dependencies/theano_linear/test_matrixmul.py', 'packaged_dependencies/theano_linear/test_linear.py', 'packaged_dependencies/theano_linear/spconv.py', 'sandbox/cuda_convnet/tests/test_weight_acts_strided.py', 'sandbox/cuda_convnet/tests/test_image_acts_strided.py', 'sandbox/cuda_convnet/specialized_bench.py', 'sandbox/cuda_convnet/response_norm.py', 'sandbox/cuda_convnet/convnet_compile.py', 'sandbox/cuda_convnet/pthreads.py', 'sandbox/cuda_convnet/bench.py', 'sandbox/lisa_rl/bandit/plot_reward.py', 'sandbox/lisa_rl/bandit/simulate.py', 'config/__init__.py', 'utils/__init__.py', 'optimization/test_linesearch.py', 'optimization/test_minres.py', 'optimization/test_linear_cg.py', 'scripts/papers/maxout/svhn_preprocessing.py', 'scripts/papers/maxout/compute_test_err.py', 'scripts/papers/jia_huang_wkshp_11/fit_final_model.py', 'scripts/papers/jia_huang_wkshp_11/evaluate.py', 'scripts/papers/jia_huang_wkshp_11/npy2mat.py', 'scripts/papers/jia_huang_wkshp_11/assemble.py', 'scripts/datasets/make_cifar100_patches_8x8.py', 'scripts/datasets/make_downsampled_stl10.py', 'scripts/datasets/make_cifar100_whitened.py', 'scripts/tutorials/deep_trainer/test_deep_trainer.py', 'scripts/icml_2013_wrepl/black_box/learn_zca.py', 'train_extensions/tests/test_window_flip.py', 'train_extensions/window_flip.py', 'linear/tests/test_local_c01b.py', 'sandbox/cuda_convnet/debug.py', ]) def test_format_pep8(): """ Test if pep8 is respected. """ pep8_checker = StyleGuide() files_to_check = [] for path in list_files(".py"): rel_path = os.path.relpath(path, pylearn2.__path__[0]) if rel_path in whitelist_pep8: continue else: files_to_check.append(path) report = pep8_checker.check_files(files_to_check) if report.total_errors > 0: raise AssertionError("PEP8 Format not respected") def print_files_information_pep8(): """ Print the list of files which can be removed from the whitelist and the list of files which do not respect PEP8 formatting that aren't in the whitelist """ infracting_files = [] non_infracting_files = [] pep8_checker = StyleGuide(quiet=True) for path in list_files(".py"): number_of_infractions = pep8_checker.input_file(path) rel_path = os.path.relpath(path, pylearn2.__path__[0]) if number_of_infractions > 0: if rel_path not in whitelist_pep8: infracting_files.append(path) else: if rel_path in whitelist_pep8: non_infracting_files.append(path) print("Files that must be corrected or added to whitelist:") for file in infracting_files: print(file) print("Files that can be removed from whitelist:") for file in non_infracting_files: print(file) def test_format_docstrings(): """ Test if docstrings are well formatted. """ try: verify_format_docstrings() except SkipTest as e: import traceback traceback.print_exc(e) raise AssertionError( "Some file raised SkipTest on import, and inadvertently" " canceled the documentation testing." ) def verify_format_docstrings(): """ Implementation of `test_format_docstrings`. The implementation is factored out so it can be placed inside a guard against SkipTest. """ format_infractions = [] for path in list_files(".py"): rel_path = os.path.relpath(path, pylearn2.__path__[0]) if rel_path in whitelist_docstrings: continue try: format_infractions.extend(docstring_errors(path)) except Exception as e: format_infractions.append(["%s failed to run so format cannot " "be checked. Error message:\n %s" % (rel_path, e)]) if len(format_infractions) > 0: msg = "\n".join(':'.join(line) for line in format_infractions) raise AssertionError("Docstring format not respected:\n%s" % msg) if __name__ == "__main__": print_files_information_pep8()
bsd-3-clause
jfsantos/ift6266h14
old/test_timit_iy.py
1
2996
from timit_full import TimitFullCorpusReader import itertools import numpy as np from pylearn2.datasets import DenseDesignMatrix from pylearn2.models.mlp import * from pylearn2.costs.mlp.dropout import Dropout from pylearn2.termination_criteria import EpochCounter from pylearn2.training_algorithms.sgd import SGD from pylearn2.training_algorithms import learning_rule from pylearn2.train import Train from pylearn2.train_extensions import best_params import cPickle as pickle import theano # Gets all utterances from <spkrid>, splits them into <framelen> # frames with <overlap> overlaps. Returns the frames and correspondent # phone symbols. spkrid = 'MTCS0' class TimitPhoneData(DenseDesignMatrix): def __init__(self, spkrid, phone, framelen, overlap, start, stop): data = TimitFullCorpusReader('/home/jfsantos/data/TIMIT/') # Some list comprehension/zip magic here (but it works!) spkrfr = [data.frames(z, 160, 159) for z in data.utteranceids(spkrid=spkrid)] fr, ph = zip(*[(x[0], x[1]) for x in spkrfr]) fr = np.vstack(fr)*2**-15 ph = list(itertools.chain(*ph)) # Get all elements for which the phone is 'iy' iy_idx = [i for i,x in enumerate(ph) if x == 'iy'] fr_iy = fr[iy_idx] X = fr_iy[:,0:159] y = np.array([fr_iy[:,159]]).T # y.ndim has to be 2 super(TimitPhoneData,self).__init__(X=X[start:stop], y=y[start:stop]) train = TimitPhoneData(spkrid='FPLS0', phone='iy', framelen=160, overlap=159, start=0, stop=10000) valid = TimitPhoneData(spkrid='FPLS0', phone='iy', framelen=160, overlap=159, start=10000, stop=12000) test = TimitPhoneData(spkrid='FPLS0', phone='iy', framelen=160, overlap=159, start=12000, stop=18000) i0 = VectorSpace(159) s0 = Sigmoid(layer_name='h0', dim=500, sparse_init=15) l0 = Linear(layer_name='y', dim=1, sparse_init=15) mdl = MLP(layers=[s0, l0], nvis=159, input_space=i0) trainer = SGD(batch_size=512, learning_rate = .01, init_momentum = .5, monitoring_dataset = {'train' : train, 'valid': valid, 'test' : test}, termination_criterion = EpochCounter(max_epochs=200)) watcher = best_params.MonitorBasedSaveBest( channel_name='test_objective', save_path='nextsample_iy_FPLS0_mlp_sig_lin_watcher.pkl') experiment = Train(dataset=train, model=mdl, algorithm=trainer, extensions = [watcher]) experiment.main_loop() # Now we have the best model, let's load it and use it to generate some # samples! bestmdl = pickle.load(open('nextsample_iy_FPLS0_mlp_sig_lin_watcher.pkl')) X = theano.tensor.dmatrix('X') y = bestmdl.fprop(X) predict = theano.function([X], y) # Let's start with a all zero vector, then use the prediction to populate the next sample x0 = np.asmatrix(np.zeros((1,16000))) for k in np.arange(160,16000): frame = x0[:,k-160:k-1] x0[0,k] = predict(frame)
mit
neuroneuro15/natnetclient
build/lib/natnetclient/utils.py
1
1318
__author__ = 'ratcave' import numpy as np from sklearn.decomposition import PCA def rotate_to_var(markers): """Returns degrees to rotate about y axis so greatest marker variance points in +X direction""" # Mean-Center markers -= np.mean(markers, axis=0) # Vector in direction of greatest variance pca = PCA(n_components=1).fit(markers[:, [0, 2]]) coeff_vec = pca.components_[0] # # Flip coeff_vec in direction of max variance along the vector. # marker_var = markers[markers[:,2].argsort(), 2] # Check variance along component to determine whether to flip. # winlen = int(len(marker_var)/2+1) # Window length for moving mean (two steps, with slight overlap) # var_means = np.array([marker_var[:winlen], marker_var[-winlen:]]).mean(axis=1) # coeff_vec = coeff_vec * -1 if np.diff(var_means)[0] < 0 else coeff_vec # Rotation amount, in radians base_vec = np.array([1, 0]) # Vector in +X direction msin, mcos = np.cross(coeff_vec, base_vec), np.dot(coeff_vec, base_vec) angle = np.degrees(np.arctan2(msin, mcos)) print("Angle within function: {}".format(angle)) return angle # def get_pca_rotation(markers): # # markers_2d = markers[:, [0, 2]] # pca = PCA(n_components=1).fit(markers[:, [0, 2]]) # # coeff = pca.components_[0]
gpl-2.0
joshbohde/scikit-learn
sklearn/linear_model/tests/test_sgd.py
1
16225
import numpy as np from numpy.testing import assert_array_equal, assert_approx_equal from numpy.testing import assert_almost_equal, assert_array_almost_equal from sklearn import linear_model, datasets, metrics from sklearn import preprocessing import unittest from nose.tools import raises from nose.tools import assert_raises ## ## Test Data ## # test sample 1 X = np.array([[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]]) Y = [1, 1, 1, 2, 2, 2] T = np.array([[-1, -1], [2, 2], [3, 2]]) true_result = [1, 2, 2] # test sample 2 X2 = np.array([[-1, 1], [-0.75, 0.5], [-1.5, 1.5], [1, 1], [0.75, 0.5], [1.5, 1.5], [-1, -1], [0, -0.5], [1, -1]]) Y2 = [1, 1, 1, 2, 2, 2, 3, 3, 3] T2 = np.array([[-1.5, 0.5], [1, 2], [0, -2]]) true_result2 = [1, 2, 3] # test sample 3 X3 = np.array([[1, 1, 0, 0, 0, 0], [1, 1, 0, 0, 0, 0], [0, 0, 1, 0, 0, 0], [0, 0, 1, 0, 0, 0], [0, 0, 0, 0, 1, 1], [0, 0, 0, 0, 1, 1], [0, 0, 0, 1, 0, 0], [0, 0, 0, 1, 0, 0]]) Y3 = np.array([1, 1, 1, 1, 2, 2, 2, 2]) # test sample 4 - two more or less redundent feature groups X4 = np.array([[1, 0.9, 0.8, 0, 0, 0], [1, .84, .98, 0, 0, 0], [1, .96, .88, 0, 0, 0], [1, .91, .99, 0, 0, 0], [0, 0, 0, .89, .91, 1], [0, 0, 0, .79, .84, 1], [0, 0, 0, .91, .95, 1], [0, 0, 0, .93, 1, 1]]) Y4 = np.array([1, 1, 1, 1, 2, 2, 2, 2]) iris = datasets.load_iris() ## ## Classification Test Case ## class DenseSGDClassifierTestCase(unittest.TestCase): """Test suite for the dense representation variant of SGD""" factory = linear_model.SGDClassifier def test_sgd(self): """Check that SGD gives any results :-)""" clf = self.factory(penalty='l2', alpha=0.01, fit_intercept=True, n_iter=10, shuffle=True) clf.fit(X, Y) #assert_almost_equal(clf.coef_[0], clf.coef_[1], decimal=7) assert_array_equal(clf.predict(T), true_result) def test_sgd_penalties(self): """Check whether penalties and hyperparameters are set properly""" clf = self.factory(penalty='l2') assert clf.rho == 1.0 clf = self.factory(penalty='l1') assert clf.rho == 0.0 clf = self.factory(penalty='elasticnet', rho=0.85) assert clf.rho == 0.85 @raises(ValueError) def test_sgd_bad_penalty(self): """Check whether expected ValueError on bad penalty""" self.factory(penalty='foobar', rho=0.85) def test_sgd_losses(self): """Check whether losses and hyperparameters are set properly""" clf = self.factory(loss='hinge') assert isinstance(clf.loss_function, linear_model.Hinge) clf = self.factory(loss='log') assert isinstance(clf.loss_function, linear_model.Log) clf = self.factory(loss='modified_huber') assert isinstance(clf.loss_function, linear_model.ModifiedHuber) @raises(ValueError) def test_sgd_bad_loss(self): """Check whether expected ValueError on bad loss""" self.factory(loss="foobar") @raises(ValueError) def test_sgd_n_iter_param(self): """Test parameter validity check""" self.factory(n_iter=-10000) @raises(ValueError) def test_sgd_shuffle_param(self): """Test parameter validity check""" self.factory(shuffle="false") @raises(TypeError) def test_arument_coef(self): """Checks coef_init not allowed as model argument (only fit)""" # Provided coef_ does not match dataset. self.factory(coef_init=np.zeros((3,))).fit(X, Y) @raises(ValueError) def test_provide_coef(self): """Checks coef_init shape for the warm starts""" # Provided coef_ does not match dataset. self.factory().fit(X, Y, coef_init=np.zeros((3,))) @raises(ValueError) def test_set_intercept(self): """Checks intercept_ shape for the warm starts""" # Provided intercept_ does not match dataset. self.factory().fit(X, Y, intercept_init=np.zeros((3,))) @raises(ValueError) def test_sgd_at_least_two_labels(self): """Target must have at least two labels""" self.factory(alpha=0.01, n_iter=20).fit(X2, np.ones(9)) def test_sgd_multiclass(self): """Multi-class test case""" clf = self.factory(alpha=0.01, n_iter=20).fit(X2, Y2) assert clf.coef_.shape == (3, 2) assert clf.intercept_.shape == (3,) assert clf.decision_function([0, 0]).shape == (1, 3) pred = clf.predict(T2) assert_array_equal(pred, true_result2) def test_sgd_multiclass_with_init_coef(self): """Multi-class test case""" clf = self.factory(alpha=0.01, n_iter=20) clf.fit(X2, Y2, coef_init=np.zeros((3, 2)), intercept_init=np.zeros(3)) assert clf.coef_.shape == (3, 2) assert clf.intercept_.shape == (3,) pred = clf.predict(T2) assert_array_equal(pred, true_result2) def test_sgd_multiclass_njobs(self): """Multi-class test case with multi-core support""" clf = self.factory(alpha=0.01, n_iter=20, n_jobs=2).fit(X2, Y2) assert clf.coef_.shape == (3, 2) assert clf.intercept_.shape == (3,) assert clf.decision_function([0, 0]).shape == (1, 3) pred = clf.predict(T2) assert_array_equal(pred, true_result2) def test_set_coef_multiclass(self): """Checks coef_init and intercept_init shape for for multi-class problems""" # Provided coef_ does not match dataset clf = self.factory() assert_raises(ValueError, clf.fit, X2, Y2, coef_init=np.zeros((2, 2))) # Provided coef_ does match dataset clf = self.factory().fit(X2, Y2, coef_init=np.zeros((3, 2))) # Provided intercept_ does not match dataset clf = self.factory() assert_raises(ValueError, clf.fit, X2, Y2, intercept_init=np.zeros((1,))) # Provided intercept_ does match dataset. clf = self.factory().fit(X2, Y2, intercept_init=np.zeros((3,))) def test_sgd_proba(self): """Check SGD.predict_proba for log loss only""" # hinge loss does not allow for conditional prob estimate clf = self.factory(loss="hinge", alpha=0.01, n_iter=10).fit(X, Y) assert_raises(NotImplementedError, clf.predict_proba, [3, 2]) # log loss implements the logistic regression prob estimate clf = self.factory(loss="log", alpha=0.01, n_iter=10).fit(X, Y) p = clf.predict_proba([3, 2]) assert p > 0.5 p = clf.predict_proba([-1, -1]) assert p < 0.5 def test_sgd_l1(self): """Test L1 regularization""" n = len(X4) np.random.seed(13) idx = np.arange(n) np.random.shuffle(idx) X = X4[idx, :] Y = Y4[idx, :] clf = self.factory(penalty='l1', alpha=.2, fit_intercept=False, n_iter=2000) clf.fit(X, Y) assert_array_equal(clf.coef_[0, 1:-1], np.zeros((4,))) pred = clf.predict(X) assert_array_equal(pred, Y) def test_class_weight(self): """ Test class weights. """ X = np.array([[-1.0, -1.0], [-1.0, 0], [-.8, -1.0], [1.0, 1.0], [1.0, 0.0]]) y = [1, 1, 1, -1, -1] clf = self.factory(alpha=0.1, n_iter=1000, fit_intercept=False) clf.fit(X, y) assert_array_equal(clf.predict([[0.2, -1.0]]), np.array([1])) # we give a small weights to class 1 clf.fit(X, y, class_weight={1: 0.001}) # now the hyperplane should rotate clock-wise and # the prediction on this point should shift assert_array_equal(clf.predict([[0.2, -1.0]]), np.array([-1])) def test_equal_class_weight(self): """Test if equal class weights approx. equals no class weights. """ X = [[1, 0], [1, 0], [0, 1], [0, 1]] y = [0, 0, 1, 1] clf = self.factory(alpha=0.1, n_iter=1000) clf.fit(X, y) X = [[1, 0], [0, 1]] y = [0, 1] clf_weighted = self.factory(alpha=0.1, n_iter=1000) clf_weighted.fit(X, y, class_weight={0: 0.5, 1: 0.5}) # should be similar up to some epsilon due to learning rate schedule assert_almost_equal(clf.coef_, clf_weighted.coef_, decimal=2) @raises(ValueError) def test_wrong_class_weight_label(self): """ValueError due to not existing class label.""" clf = self.factory(alpha=0.1, n_iter=1000) clf.fit(X, Y, class_weight={0: 0.5}) @raises(ValueError) def test_wrong_class_weight_format(self): """ValueError due to wrong class_weight argument type.""" clf = self.factory(alpha=0.1, n_iter=1000) clf.fit(X, Y, class_weight=[0.5]) def test_auto_weight(self): """Test class weights for imbalanced data""" # compute reference metrics on iris dataset that is quite balanced by # default X, y = iris.data, iris.target X = preprocessing.scale(X) idx = np.arange(X.shape[0]) np.random.seed(13) np.random.shuffle(idx) X = X[idx] y = y[idx] clf = self.factory(alpha=0.0001, n_iter=1000).fit(X, y) assert_approx_equal(metrics.f1_score(y, clf.predict(X)), 0.96, 2) # make the same prediction using automated class_weight clf_auto = self.factory(alpha=0.0001, n_iter=1000).fit(X, y, class_weight="auto") assert_approx_equal(metrics.f1_score(y, clf_auto.predict(X)), 0.96, 2) # Make sure that in the balanced case it does not change anything # to use "auto" assert_array_almost_equal(clf.coef_, clf_auto.coef_, 6) # build an very very imbalanced dataset out of iris data X_0 = X[y == 0, :] y_0 = y[y == 0] X_imbalanced = np.vstack([X] + [X_0] * 10) y_imbalanced = np.concatenate([y] + [y_0] * 10) # fit a model on the imbalanced data without class weight info clf = self.factory(n_iter=1000) clf.fit(X_imbalanced, y_imbalanced) y_pred = clf.predict(X) assert metrics.f1_score(y, y_pred) < 0.96 # fit a model with auto class_weight enabled clf = self.factory(n_iter=1000) clf.fit(X_imbalanced, y_imbalanced, class_weight="auto") y_pred = clf.predict(X) assert metrics.f1_score(y, y_pred) > 0.96 def test_sample_weights(self): """ Test weights on individual samples """ X = np.array([[-1.0, -1.0], [-1.0, 0], [-.8, -1.0], [1.0, 1.0], [1.0, 0.0]]) y = [1, 1, 1, -1, -1] clf = self.factory(alpha=0.1, n_iter=1000, fit_intercept=False) clf.fit(X, y) assert_array_equal(clf.predict([[0.2, -1.0]]), np.array([1])) # we give a small weights to class 1 clf.fit(X, y, sample_weight=[0.001] * 3 + [1] * 2) # now the hyperplane should rotate clock-wise and # the prediction on this point should shift assert_array_equal(clf.predict([[0.2, -1.0]]), np.array([-1])) @raises(ValueError) def test_wrong_sample_weights(self): """Test if ValueError is raised if sample_weight has wrong shape""" clf = self.factory(alpha=0.1, n_iter=1000, fit_intercept=False) # provided sample_weight too long clf.fit(X, Y, sample_weight=range(7)) class SparseSGDClassifierTestCase(DenseSGDClassifierTestCase): """Run exactly the same tests using the sparse representation variant""" factory = linear_model.sparse.SGDClassifier ################################################################################ # Regression Test Case class DenseSGDRegressorTestCase(unittest.TestCase): """Test suite for the dense representation variant of SGD""" factory = linear_model.SGDRegressor def test_sgd(self): """Check that SGD gives any results.""" clf = self.factory(alpha=0.1, n_iter=2, fit_intercept=False) clf.fit([[0, 0], [1, 1], [2, 2]], [0, 1, 2]) assert clf.coef_[0] == clf.coef_[1] def test_sgd_penalties(self): """Check whether penalties and hyperparameters are set properly""" clf = self.factory(penalty='l2') assert clf.rho == 1.0 clf = self.factory(penalty='l1') assert clf.rho == 0.0 clf = self.factory(penalty='elasticnet', rho=0.85) assert clf.rho == 0.85 @raises(ValueError) def test_sgd_bad_penalty(self): """Check whether expected ValueError on bad penalty""" self.factory(penalty='foobar', rho=0.85) def test_sgd_losses(self): """Check whether losses and hyperparameters are set properly""" clf = self.factory(loss='squared_loss') assert isinstance(clf.loss_function, linear_model.SquaredLoss) clf = self.factory(loss='huber', p=0.5) assert isinstance(clf.loss_function, linear_model.Huber) assert clf.p == 0.5 @raises(ValueError) def test_sgd_bad_loss(self): """Check whether expected ValueError on bad loss""" self.factory(loss="foobar") def test_sgd_least_squares_fit(self): xmin, xmax = -5, 5 n_samples = 100 X = np.linspace(xmin, xmax, n_samples).reshape(n_samples, 1) # simple linear function without noise y = 0.5 * X.ravel() clf = self.factory(loss='squared_loss', alpha=0.1, n_iter=20, fit_intercept=False) clf.fit(X, y) score = clf.score(X, y) assert score > 0.99 # simple linear function with noise y = 0.5 * X.ravel() \ + np.random.randn(n_samples, 1).ravel() clf = self.factory(loss='squared_loss', alpha=0.1, n_iter=20, fit_intercept=False) clf.fit(X, y) score = clf.score(X, y) assert score > 0.5 def test_sgd_huber_fit(self): xmin, xmax = -5, 5 n_samples = 100 X = np.linspace(xmin, xmax, n_samples).reshape(n_samples, 1) # simple linear function without noise y = 0.5 * X.ravel() clf = self.factory(loss="huber", p=0.1, alpha=0.1, n_iter=20, fit_intercept=False) clf.fit(X, y) score = clf.score(X, y) assert score > 0.99 # simple linear function with noise y = 0.5 * X.ravel() \ + np.random.randn(n_samples, 1).ravel() clf = self.factory(loss="huber", p=0.1, alpha=0.1, n_iter=20, fit_intercept=False) clf.fit(X, y) score = clf.score(X, y) assert score > 0.5 def test_elasticnet_convergence(self): """Check that the SGD ouput is consistent with coordinate descent""" n_samples, n_features = 1000, 5 np.random.seed(0) X = np.random.randn(n_samples, n_features) # ground_truth linear model that generate y from X and to which the # models should converge if the regularizer would be set to 0.0 ground_truth_coef = np.random.randn(n_features) y = np.dot(X, ground_truth_coef) # XXX: alpha = 0.1 seems to cause convergence problems for alpha in [0.01, 0.001]: for rho in [0.5, 0.8, 1.0]: cd = linear_model.ElasticNet(alpha=alpha, rho=rho, fit_intercept=False) cd.fit(X, y) sgd = self.factory(penalty='elasticnet', n_iter=50, alpha=alpha, rho=rho, fit_intercept=False) sgd.fit(X, y) err_msg = ("cd and sgd did not converge to comparable " "results for alpha=%f and rho=%f" % (alpha, rho)) assert_almost_equal(cd.coef_, sgd.coef_, decimal=2, err_msg=err_msg) class SparseSGDRegressorTestCase(DenseSGDRegressorTestCase): """Run exactly the same tests using the sparse representation variant""" factory = linear_model.sparse.SGDRegressor
bsd-3-clause
ephes/scikit-learn
sklearn/semi_supervised/label_propagation.py
127
15312
# coding=utf8 """ Label propagation in the context of this module refers to a set of semisupervised classification algorithms. In the high level, these algorithms work by forming a fully-connected graph between all points given and solving for the steady-state distribution of labels at each point. These algorithms perform very well in practice. The cost of running can be very expensive, at approximately O(N^3) where N is the number of (labeled and unlabeled) points. The theory (why they perform so well) is motivated by intuitions from random walk algorithms and geometric relationships in the data. For more information see the references below. Model Features -------------- Label clamping: The algorithm tries to learn distributions of labels over the dataset. In the "Hard Clamp" mode, the true ground labels are never allowed to change. They are clamped into position. In the "Soft Clamp" mode, they are allowed some wiggle room, but some alpha of their original value will always be retained. Hard clamp is the same as soft clamping with alpha set to 1. Kernel: A function which projects a vector into some higher dimensional space. This implementation supprots RBF and KNN kernels. Using the RBF kernel generates a dense matrix of size O(N^2). KNN kernel will generate a sparse matrix of size O(k*N) which will run much faster. See the documentation for SVMs for more info on kernels. Examples -------- >>> from sklearn import datasets >>> from sklearn.semi_supervised import LabelPropagation >>> label_prop_model = LabelPropagation() >>> iris = datasets.load_iris() >>> random_unlabeled_points = np.where(np.random.random_integers(0, 1, ... size=len(iris.target))) >>> labels = np.copy(iris.target) >>> labels[random_unlabeled_points] = -1 >>> label_prop_model.fit(iris.data, labels) ... # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS LabelPropagation(...) Notes ----- References: [1] Yoshua Bengio, Olivier Delalleau, Nicolas Le Roux. In Semi-Supervised Learning (2006), pp. 193-216 [2] Olivier Delalleau, Yoshua Bengio, Nicolas Le Roux. Efficient Non-Parametric Function Induction in Semi-Supervised Learning. AISTAT 2005 """ # Authors: Clay Woolam <clay@woolam.org> # Licence: BSD from abc import ABCMeta, abstractmethod from scipy import sparse import numpy as np from ..base import BaseEstimator, ClassifierMixin from ..metrics.pairwise import rbf_kernel from ..utils.graph import graph_laplacian from ..utils.extmath import safe_sparse_dot from ..utils.validation import check_X_y, check_is_fitted from ..externals import six from ..neighbors.unsupervised import NearestNeighbors ### Helper functions def _not_converged(y_truth, y_prediction, tol=1e-3): """basic convergence check""" return np.abs(y_truth - y_prediction).sum() > tol class BaseLabelPropagation(six.with_metaclass(ABCMeta, BaseEstimator, ClassifierMixin)): """Base class for label propagation module. Parameters ---------- kernel : {'knn', 'rbf'} String identifier for kernel function to use. Only 'rbf' and 'knn' kernels are currently supported.. gamma : float Parameter for rbf kernel alpha : float Clamping factor max_iter : float Change maximum number of iterations allowed tol : float Convergence tolerance: threshold to consider the system at steady state n_neighbors : integer > 0 Parameter for knn kernel """ def __init__(self, kernel='rbf', gamma=20, n_neighbors=7, alpha=1, max_iter=30, tol=1e-3): self.max_iter = max_iter self.tol = tol # kernel parameters self.kernel = kernel self.gamma = gamma self.n_neighbors = n_neighbors # clamping factor self.alpha = alpha def _get_kernel(self, X, y=None): if self.kernel == "rbf": if y is None: return rbf_kernel(X, X, gamma=self.gamma) else: return rbf_kernel(X, y, gamma=self.gamma) elif self.kernel == "knn": if self.nn_fit is None: self.nn_fit = NearestNeighbors(self.n_neighbors).fit(X) if y is None: return self.nn_fit.kneighbors_graph(self.nn_fit._fit_X, self.n_neighbors, mode='connectivity') else: return self.nn_fit.kneighbors(y, return_distance=False) else: raise ValueError("%s is not a valid kernel. Only rbf and knn" " are supported at this time" % self.kernel) @abstractmethod def _build_graph(self): raise NotImplementedError("Graph construction must be implemented" " to fit a label propagation model.") def predict(self, X): """Performs inductive inference across the model. Parameters ---------- X : array_like, shape = [n_samples, n_features] Returns ------- y : array_like, shape = [n_samples] Predictions for input data """ probas = self.predict_proba(X) return self.classes_[np.argmax(probas, axis=1)].ravel() def predict_proba(self, X): """Predict probability for each possible outcome. Compute the probability estimates for each single sample in X and each possible outcome seen during training (categorical distribution). Parameters ---------- X : array_like, shape = [n_samples, n_features] Returns ------- probabilities : array, shape = [n_samples, n_classes] Normalized probability distributions across class labels """ check_is_fitted(self, 'X_') if sparse.isspmatrix(X): X_2d = X else: X_2d = np.atleast_2d(X) weight_matrices = self._get_kernel(self.X_, X_2d) if self.kernel == 'knn': probabilities = [] for weight_matrix in weight_matrices: ine = np.sum(self.label_distributions_[weight_matrix], axis=0) probabilities.append(ine) probabilities = np.array(probabilities) else: weight_matrices = weight_matrices.T probabilities = np.dot(weight_matrices, self.label_distributions_) normalizer = np.atleast_2d(np.sum(probabilities, axis=1)).T probabilities /= normalizer return probabilities def fit(self, X, y): """Fit a semi-supervised label propagation model based All the input data is provided matrix X (labeled and unlabeled) and corresponding label matrix y with a dedicated marker value for unlabeled samples. Parameters ---------- X : array-like, shape = [n_samples, n_features] A {n_samples by n_samples} size matrix will be created from this y : array_like, shape = [n_samples] n_labeled_samples (unlabeled points are marked as -1) All unlabeled samples will be transductively assigned labels Returns ------- self : returns an instance of self. """ X, y = check_X_y(X, y) self.X_ = X # actual graph construction (implementations should override this) graph_matrix = self._build_graph() # label construction # construct a categorical distribution for classification only classes = np.unique(y) classes = (classes[classes != -1]) self.classes_ = classes n_samples, n_classes = len(y), len(classes) y = np.asarray(y) unlabeled = y == -1 clamp_weights = np.ones((n_samples, 1)) clamp_weights[unlabeled, 0] = self.alpha # initialize distributions self.label_distributions_ = np.zeros((n_samples, n_classes)) for label in classes: self.label_distributions_[y == label, classes == label] = 1 y_static = np.copy(self.label_distributions_) if self.alpha > 0.: y_static *= 1 - self.alpha y_static[unlabeled] = 0 l_previous = np.zeros((self.X_.shape[0], n_classes)) remaining_iter = self.max_iter if sparse.isspmatrix(graph_matrix): graph_matrix = graph_matrix.tocsr() while (_not_converged(self.label_distributions_, l_previous, self.tol) and remaining_iter > 1): l_previous = self.label_distributions_ self.label_distributions_ = safe_sparse_dot( graph_matrix, self.label_distributions_) # clamp self.label_distributions_ = np.multiply( clamp_weights, self.label_distributions_) + y_static remaining_iter -= 1 normalizer = np.sum(self.label_distributions_, axis=1)[:, np.newaxis] self.label_distributions_ /= normalizer # set the transduction item transduction = self.classes_[np.argmax(self.label_distributions_, axis=1)] self.transduction_ = transduction.ravel() self.n_iter_ = self.max_iter - remaining_iter return self class LabelPropagation(BaseLabelPropagation): """Label Propagation classifier Read more in the :ref:`User Guide <label_propagation>`. Parameters ---------- kernel : {'knn', 'rbf'} String identifier for kernel function to use. Only 'rbf' and 'knn' kernels are currently supported.. gamma : float Parameter for rbf kernel n_neighbors : integer > 0 Parameter for knn kernel alpha : float Clamping factor max_iter : float Change maximum number of iterations allowed tol : float Convergence tolerance: threshold to consider the system at steady state Attributes ---------- X_ : array, shape = [n_samples, n_features] Input array. classes_ : array, shape = [n_classes] The distinct labels used in classifying instances. label_distributions_ : array, shape = [n_samples, n_classes] Categorical distribution for each item. transduction_ : array, shape = [n_samples] Label assigned to each item via the transduction. n_iter_ : int Number of iterations run. Examples -------- >>> from sklearn import datasets >>> from sklearn.semi_supervised import LabelPropagation >>> label_prop_model = LabelPropagation() >>> iris = datasets.load_iris() >>> random_unlabeled_points = np.where(np.random.random_integers(0, 1, ... size=len(iris.target))) >>> labels = np.copy(iris.target) >>> labels[random_unlabeled_points] = -1 >>> label_prop_model.fit(iris.data, labels) ... # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS LabelPropagation(...) References ---------- Xiaojin Zhu and Zoubin Ghahramani. Learning from labeled and unlabeled data with label propagation. Technical Report CMU-CALD-02-107, Carnegie Mellon University, 2002 http://pages.cs.wisc.edu/~jerryzhu/pub/CMU-CALD-02-107.pdf See Also -------- LabelSpreading : Alternate label propagation strategy more robust to noise """ def _build_graph(self): """Matrix representing a fully connected graph between each sample This basic implementation creates a non-stochastic affinity matrix, so class distributions will exceed 1 (normalization may be desired). """ if self.kernel == 'knn': self.nn_fit = None affinity_matrix = self._get_kernel(self.X_) normalizer = affinity_matrix.sum(axis=0) if sparse.isspmatrix(affinity_matrix): affinity_matrix.data /= np.diag(np.array(normalizer)) else: affinity_matrix /= normalizer[:, np.newaxis] return affinity_matrix class LabelSpreading(BaseLabelPropagation): """LabelSpreading model for semi-supervised learning This model is similar to the basic Label Propgation algorithm, but uses affinity matrix based on the normalized graph Laplacian and soft clamping across the labels. Read more in the :ref:`User Guide <label_propagation>`. Parameters ---------- kernel : {'knn', 'rbf'} String identifier for kernel function to use. Only 'rbf' and 'knn' kernels are currently supported. gamma : float parameter for rbf kernel n_neighbors : integer > 0 parameter for knn kernel alpha : float clamping factor max_iter : float maximum number of iterations allowed tol : float Convergence tolerance: threshold to consider the system at steady state Attributes ---------- X_ : array, shape = [n_samples, n_features] Input array. classes_ : array, shape = [n_classes] The distinct labels used in classifying instances. label_distributions_ : array, shape = [n_samples, n_classes] Categorical distribution for each item. transduction_ : array, shape = [n_samples] Label assigned to each item via the transduction. n_iter_ : int Number of iterations run. Examples -------- >>> from sklearn import datasets >>> from sklearn.semi_supervised import LabelSpreading >>> label_prop_model = LabelSpreading() >>> iris = datasets.load_iris() >>> random_unlabeled_points = np.where(np.random.random_integers(0, 1, ... size=len(iris.target))) >>> labels = np.copy(iris.target) >>> labels[random_unlabeled_points] = -1 >>> label_prop_model.fit(iris.data, labels) ... # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS LabelSpreading(...) References ---------- Dengyong Zhou, Olivier Bousquet, Thomas Navin Lal, Jason Weston, Bernhard Schoelkopf. Learning with local and global consistency (2004) http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.115.3219 See Also -------- LabelPropagation : Unregularized graph based semi-supervised learning """ def __init__(self, kernel='rbf', gamma=20, n_neighbors=7, alpha=0.2, max_iter=30, tol=1e-3): # this one has different base parameters super(LabelSpreading, self).__init__(kernel=kernel, gamma=gamma, n_neighbors=n_neighbors, alpha=alpha, max_iter=max_iter, tol=tol) def _build_graph(self): """Graph matrix for Label Spreading computes the graph laplacian""" # compute affinity matrix (or gram matrix) if self.kernel == 'knn': self.nn_fit = None n_samples = self.X_.shape[0] affinity_matrix = self._get_kernel(self.X_) laplacian = graph_laplacian(affinity_matrix, normed=True) laplacian = -laplacian if sparse.isspmatrix(laplacian): diag_mask = (laplacian.row == laplacian.col) laplacian.data[diag_mask] = 0.0 else: laplacian.flat[::n_samples + 1] = 0.0 # set diag to 0.0 return laplacian
bsd-3-clause
google-research/scenic
scenic/projects/robust_segvit/datasets/cityscapes_variants.py
1
11019
"""Data generators for the Cityscapes dataset variants. Supported datasets, set by dataset_configs.dataset_name in the config file: cityscapes_corrupted: https://arxiv.org/pdf/1907.07484.pdf fishyscapes: https://link.springer.com/article/10.1007/s11263-021-01511-6 Implementation details: cityscapes_c: https://github.com/ekellbuch/cityscapes-c """ import functools from typing import Optional from absl import logging from flax import jax_utils import jax.numpy as jnp from scenic.dataset_lib import cityscapes_dataset from scenic.dataset_lib import dataset_utils from scenic.dataset_lib import datasets import tensorflow as tf import tensorflow_datasets as tfds CITYSCAPES_C_CORRUPTIONS = [ 'gaussian_noise', ] FISHYSCAPES_CORRUPTIONS = [ 'Static', ] CITYSCAPES_C_SEVERITIES = range(1, 6) DATASET_INFO = { 'cityscapes': { 'tfds_name': 'cityscapes', 'split': 'validation', 'num_of_examples': 500, }, 'cityscapes_corrupted': { 'tfds_name': 'internal', 'split': 'validation', 'num_of_examples': 500, }, 'fishycapes': { 'tfds_name': 'internal', 'split': 'validation', 'num_of_examples': 30, }, } # Adds cityscapes_c for severity in CITYSCAPES_C_SEVERITIES: for corruption in CITYSCAPES_C_CORRUPTIONS: temp_dataset_name = f'cityscapes_corrupted/semantic_segmentation_{corruption}_{severity}' DATASET_INFO[temp_dataset_name] = { 'tfds_name': temp_dataset_name, 'split': 'validation', 'num_of_examples': 500, } # Adds fishyscapes for corruption in FISHYSCAPES_CORRUPTIONS: temp_dataset_name = f'fishyscapes/{corruption}' DATASET_INFO[temp_dataset_name] = { 'tfds_name': temp_dataset_name, 'split': 'validation', 'num_of_examples': 30, } cityscapes_meta_data = { 'num_classes': len([c.id for c in cityscapes_dataset.CLASSES if not c.ignore_in_eval]), 'class_names': cityscapes_dataset.get_class_names(), 'class_colors': cityscapes_dataset.get_class_colors(), 'class_proportions': cityscapes_dataset.get_class_proportions(), } fishyscapes_meta_data = { 'num_classes': 2, 'class_names': ['ind', 'ood'], 'class_colors': [(0, 0, 1), (1, 0, 0)], } def normalize(image, dtype=tf.float32): """Normalizes the value of pixels in the given image. Args: image: `Tensor` representing an image binary of arbitrary size. dtype: Tensorflow data type, Data type of the image. Returns: A normalized image `Tensor`. """ image = tf.cast(image, dtype=dtype) if dtype not in [tf.int32, tf.int64, tf.uint32, tf.uint64]: image /= tf.constant(255.0, shape=[1, 1, 1], dtype=dtype) return image def preprocess_example_fishyscapes(example, train, dtype=tf.float32, resize=None, include_mask=True): """Preprocesses the given image. Args: example: dict; Example coming from TFDS. train: bool; Whether to apply training-specific preprocessing or not. dtype: Tensorflow data type; Data type of the image. resize: sequence; [H, W] to which image and labels should be resized. include_mask: include batch_mask to ignore specific classes. Returns: An example dict as required by the model. """ image = normalize(example['image_left'], dtype) mask = example['mask'] # Resize test images (train images are cropped/resized during augmentation): if not train: if resize is not None: image = tf.image.resize(image, resize, 'bilinear') mask = tf.image.resize(mask, resize, 'nearest') image = tf.cast(image, dtype) mask = tf.cast(mask, dtype) mask = tf.squeeze(mask, axis=2) outputs = {'inputs': image, 'label': mask} if include_mask: # Fishyscapes mask has values 0,1, 255, background pixels are set as 255. # create batch_mask array and set background pixels to 0 and # pixels that should be included during eval to 1 batch_mask = tf.ones_like(mask, dtype) batch_mask = tf.cast(batch_mask*(1-tf.cast(mask == 255, dtype)), dtype) # update the mask array to be 0 or by setting cls 255 to cls 0. mask = tf.cast(mask*(1-tf.cast(mask == 255, dtype)), dtype) outputs = {'inputs': image, 'label': mask, 'batch_mask': batch_mask} return outputs preprocess_examples = { 'cityscapes': cityscapes_dataset.preprocess_example, 'fishyscapes': preprocess_example_fishyscapes, } def cityscapes_load_split( dataset_name, batch_size, train=False, dtype=tf.float32, shuffle_buffer_size=10, shuffle_seed=None, data_augmentations=None, preprocess_ex_eval=None, cache=True, data_dir: Optional[str] = None, ): """Creates a split from the Cityscapes dataset using TensorFlow Datasets. For the training set, we drop the last partial batch. This is fine to do because we additionally shuffle the data randomly each epoch, thus the trainer will see all data in expectation. For the validation set, we pad the final batch to the desired batch size. Args: dataset_name: string; Dataset name defined in DATASET_INFO. batch_size: int; The batch size returned by the data pipeline. train: bool; Whether to load the train or evaluation split. dtype: TF data type; Data type of the image. shuffle_buffer_size: int; Buffer size for the TFDS prefetch. shuffle_seed: The seed to use when shuffling the train split. data_augmentations: list(str); Types of data augmentation applied on preprocess_ex_eval: preprocessing function. Default None. cache: bool; Whether to cache dataset in memory. data_dir: directory with data. Returns: A `tf.data.Dataset`. """ assert not train, 'Only evaluation is supported.' assert dataset_name in DATASET_INFO del data_augmentations cityscapes_variant_info = DATASET_INFO.get(dataset_name, {}) split = cityscapes_variant_info['split'] # only supports validation # Load the preprocessing function if 'cityscapes' in cityscapes_variant_info.get('tfds_name'): if dataset_name == 'cityscapes': builder = tfds.builder(dataset_name, dtype=dtype) elif 'cityscapes_corrupted' in dataset_name: if data_dir is None: # pylint: disable=line-too-long data_dir = 'gs://ub-ekb/tensorflow_datasets/cityscapes_corrupted/tfrecords/v.0.0' # pylint: disable=line-too-long # pylint: enable=line-too-long builder = tfds.builder(dataset_name, data_dir=data_dir) elif 'fishyscapes' in cityscapes_variant_info.get('tfds_name'): if data_dir is None: data_dir = 'gs://ub-ekb/tensorflow_datasets/fishyscapes/tfrecords/v.0.0' builder = tfds.builder(dataset_name, data_dir=data_dir) else: raise NotImplementedError(f'{dataset_name} not available') ds, ds_info = dataset_utils.load_split_from_tfds_builder( builder=builder, batch_size=batch_size, split=split, preprocess_example=preprocess_ex_eval, shuffle_buffer_size=shuffle_buffer_size, shuffle_seed=shuffle_seed, cache=cache) return ds, ds_info def _check_dataset_exists(dataset_configs): assert 'dataset_name' in dataset_configs, ('Must specify dataset_name in ' 'dataset_configs.') dataset_name = dataset_configs['dataset_name'] assert dataset_configs[ 'dataset_name'] in DATASET_INFO, f'{dataset_name} is not supported.' return dataset_name @datasets.add_dataset('cityscapes_variants') def get_dataset(*, batch_size, eval_batch_size, num_shards, dtype_str='float32', shuffle_seed=0, prefetch_buffer_size=2, rng=None, dataset_configs=None, dataset_service_address: Optional[str] = None): """Returns generators for the Cityscapes validation, and test set. Args: batch_size: int; Determines the train batch size. eval_batch_size: int; Determines the evaluation batch size. num_shards: int; Number of shards --> batch shape: [num_shards, bs, ...]. dtype_str: Data type of the image (e.g. 'float32'). shuffle_seed: int; Seed for shuffling the training data. prefetch_buffer_size: int; Buffer size for the TFDS prefetch. rng: JAX rng key, which can be used for augmentation, shuffling, etc. dataset_configs: dict; Dataset specific configurations. dataset_service_address: If set, will distribute the training dataset using the given tf.data service at the given address. Returns: A dataset_utils.Dataset() which includes a train_iter, a valid_iter, a test_iter, and a dict of meta_data. """ del batch_size del shuffle_seed, rng del dataset_service_address dtype = getattr(tf, dtype_str) dataset_configs = dataset_configs or {} dataset_name = _check_dataset_exists(dataset_configs) cityscapes_variant_info = DATASET_INFO.get(dataset_name) target_size = dataset_configs.get('target_size', None) if 'cityscapes' in dataset_name: preprocess_example = preprocess_examples['cityscapes'] elif 'fishyscapes' in dataset_name: preprocess_example = preprocess_examples['fishyscapes'] preprocess_ex_eval = functools.partial( preprocess_example, train=False, dtype=dtype, resize=target_size) logging.info('Loading validation split of the %s dataset.', dataset_name) eval_ds, _ = cityscapes_load_split( dataset_name=dataset_name, batch_size=eval_batch_size, train=False, dtype=dtype, preprocess_ex_eval=preprocess_ex_eval) maybe_pad_batches_eval = functools.partial( dataset_utils.maybe_pad_batch, train=False, batch_size=eval_batch_size, pixel_level=True) shard_batches = functools.partial(dataset_utils.shard, n_devices=num_shards) exclude_classes = functools.partial( cityscapes_dataset.exclude_bad_classes, new_labels=cityscapes_dataset.get_post_exclusion_labels()) eval_iter = iter(eval_ds) eval_iter = map(dataset_utils.tf_to_numpy, eval_iter) eval_iter = map(maybe_pad_batches_eval, eval_iter) if 'cityscapes' in dataset_name: eval_iter = map(exclude_classes, eval_iter) eval_iter = map(shard_batches, eval_iter) eval_iter = jax_utils.prefetch_to_device(eval_iter, prefetch_buffer_size) if target_size is None: input_shape = (-1, 1024, 2048, 3) else: input_shape = (-1,) + tuple(target_size) + (3,) meta_data = { 'input_shape': input_shape, 'num_train_examples': 0, 'num_eval_examples': cityscapes_variant_info['num_of_examples'], 'input_dtype': getattr(jnp, dtype_str), 'target_is_onehot': False, } if 'cityscapes' in dataset_name: meta_data.update(cityscapes_meta_data) elif 'fishyscapes' in dataset_name: meta_data.update(fishyscapes_meta_data) return dataset_utils.Dataset(None, eval_iter, None, meta_data)
apache-2.0
starimpact/fast-rcnn
tools/train_net.py
23
3134
#!/usr/bin/env python # -------------------------------------------------------- # Fast R-CNN # Copyright (c) 2015 Microsoft # Licensed under The MIT License [see LICENSE for details] # Written by Ross Girshick # -------------------------------------------------------- """Train a Fast R-CNN network on a region of interest database.""" import _init_paths from fast_rcnn.train import get_training_roidb, train_net from fast_rcnn.config import cfg, cfg_from_file, cfg_from_list, get_output_dir from datasets.factory import get_imdb import caffe import argparse import pprint import numpy as np import sys def parse_args(): """ Parse input arguments """ parser = argparse.ArgumentParser(description='Train a Fast R-CNN network') parser.add_argument('--gpu', dest='gpu_id', help='GPU device id to use [0]', default=0, type=int) parser.add_argument('--solver', dest='solver', help='solver prototxt', default=None, type=str) parser.add_argument('--iters', dest='max_iters', help='number of iterations to train', default=40000, type=int) parser.add_argument('--weights', dest='pretrained_model', help='initialize with pretrained model weights', default=None, type=str) parser.add_argument('--cfg', dest='cfg_file', help='optional config file', default=None, type=str) parser.add_argument('--imdb', dest='imdb_name', help='dataset to train on', default='voc_2007_trainval', type=str) parser.add_argument('--rand', dest='randomize', help='randomize (do not use a fixed seed)', action='store_true') parser.add_argument('--set', dest='set_cfgs', help='set config keys', default=None, nargs=argparse.REMAINDER) if len(sys.argv) == 1: parser.print_help() sys.exit(1) args = parser.parse_args() return args if __name__ == '__main__': args = parse_args() print('Called with args:') print(args) if args.cfg_file is not None: cfg_from_file(args.cfg_file) if args.set_cfgs is not None: cfg_from_list(args.set_cfgs) print('Using config:') pprint.pprint(cfg) if not args.randomize: # fix the random seeds (numpy and caffe) for reproducibility np.random.seed(cfg.RNG_SEED) caffe.set_random_seed(cfg.RNG_SEED) # set up caffe caffe.set_mode_gpu() if args.gpu_id is not None: caffe.set_device(args.gpu_id) imdb = get_imdb(args.imdb_name) print 'Loaded dataset `{:s}` for training'.format(imdb.name) roidb = get_training_roidb(imdb) output_dir = get_output_dir(imdb, None) print 'Output will be saved to `{:s}`'.format(output_dir) train_net(args.solver, roidb, output_dir, pretrained_model=args.pretrained_model, max_iters=args.max_iters)
mit
YzPaul3/h2o-3
py2/testdir_single_jvm/test_GBM_basic.py
20
4883
import unittest, sys, time sys.path.extend(['.','..','../..','py']) import h2o2 as h2o import h2o_cmd, h2o_import as h2i from h2o_test import dump_json, verboseprint, OutputObj from tabulate import tabulate class Basic(unittest.TestCase): def tearDown(self): h2o.check_sandbox_for_errors() @classmethod def setUpClass(cls): h2o.init(1, java_heap_GB=12) @classmethod def tearDownClass(cls): h2o.tear_down_cloud() def test_GBM_basic(self): bucket = 'home-0xdiag-datasets' importFolderPath = 'standard' trainFilename = 'covtype.shuffled.90pct.data' train_key = 'covtype.train.hex' model_key = 'GBMModelKey' timeoutSecs = 1800 csvPathname = importFolderPath + "/" + trainFilename # FIX! do I need to force enum for classification? what if I do regression after this? columnTypeDict = {54: 'Enum'} parseResult = h2i.import_parse(bucket=bucket, path=csvPathname, columnTypeDict=columnTypeDict, schema='local', chunk_size=4194304, hex_key=train_key, timeoutSecs=timeoutSecs) pA = h2o_cmd.ParseObj(parseResult) iA = h2o_cmd.InspectObj(pA.parse_key) parse_key = pA.parse_key numRows = iA.numRows numCols = iA.numCols labelList = iA.labelList labelListUsed = list(labelList) numColsUsed = numCols # run through a couple of parameter sets parameters = [] parameters.append({ 'response_column': 'C55', 'ntrees': 2, 'max_depth': 10, 'min_rows': 3, 'nbins': 40, 'learn_rate': 0.2, # 'loss': 'multinomial', # FIX! doesn't like it? # 'loss': 'Bernoulli', # FIX..no variable importance for GBM yet? # 'variable_importance': False, # 'seed': }) parameters.append({ 'response_column': 'C55', 'loss': 'multinomial', # This does nothing! intent is solely based on type of response col 'ntrees': 1, 'max_depth': 20, 'min_rows': 3, 'nbins': 40, 'learn_rate': 0.2, }) model_key = 'covtype_gbm.hex' for p in parameters: bmResult = h2o.n0.build_model( algo='gbm', model_id=model_key, training_frame=train_key, validation_frame=train_key, parameters=p, timeoutSecs=60) bm = OutputObj(bmResult, 'bm') modelResult = h2o.n0.models(key=model_key) model = OutputObj(modelResult['models'][0]['output'], 'model') cmmResult = h2o.n0.compute_model_metrics(model=model_key, frame=parse_key, timeoutSecs=60) # cmm = OutputObj(cmmResult, 'cmm') # print "\nLook!, can use dot notation: cmm.cm.confusion_matrix", cmm.cm.confusion_matrix, "\n" vis = OutputObj(model.variable_importances, 'vis') # just the first 10 visDataChopped = [v[0:9] for v in vis.data] names = visDataChopped[0] relativeImportance = visDataChopped[1] print "names:", names print "relativeImportance:", relativeImportance scaledImportance = visDataChopped[2] percentage = visDataChopped[3] print "\nvis\n", tabulate(visDataChopped[1:], headers=names) # print "\nrelativeImportance (10)\n", tabulate(relativeImportance, headers=names) # print "\nscaledImportance (10)\n", tabulate(scaledImportance, headers=names) # print "\npercentage (10)\n", tabulate(percentage, headers=names) print "will say Regression or Classification. no Multinomial?" print "model.model_category", model.model_category assert model.model_category=='Multinomial', model.model_category # FIX! # print "FIX! why is mse 0 and mse_train Nan?" # print "model.mse:", model.mse # print "model.mse_train:", model.mse_train if 1==0: print "" for i,c in enumerate(cmm.cm): print "\ncmms.cm[%s]" % i, tabulate(c) print "" mmResult = h2o.n0.model_metrics(model=model_key, frame=parse_key, timeoutSecs=60) mmResultShort = mmResult['model_metrics'][0] del mmResultShort['frame'] # too much! mm = OutputObj(mmResultShort, 'mm') prResult = h2o.n0.predict(model=model_key, frame=parse_key, timeoutSecs=60) pr = OutputObj(prResult['model_metrics'][0]['predictions'], 'pr') # too slow! # h2o_cmd.runStoreView() if __name__ == '__main__': h2o.unit_main()
apache-2.0
ephes/scikit-learn
sklearn/ensemble/partial_dependence.py
249
15097
"""Partial dependence plots for tree ensembles. """ # Authors: Peter Prettenhofer # License: BSD 3 clause from itertools import count import numbers import numpy as np from scipy.stats.mstats import mquantiles from ..utils.extmath import cartesian from ..externals.joblib import Parallel, delayed from ..externals import six from ..externals.six.moves import map, range, zip from ..utils import check_array from ..tree._tree import DTYPE from ._gradient_boosting import _partial_dependence_tree from .gradient_boosting import BaseGradientBoosting def _grid_from_X(X, percentiles=(0.05, 0.95), grid_resolution=100): """Generate a grid of points based on the ``percentiles of ``X``. The grid is generated by placing ``grid_resolution`` equally spaced points between the ``percentiles`` of each column of ``X``. Parameters ---------- X : ndarray The data percentiles : tuple of floats The percentiles which are used to construct the extreme values of the grid axes. grid_resolution : int The number of equally spaced points that are placed on the grid. Returns ------- grid : ndarray All data points on the grid; ``grid.shape[1] == X.shape[1]`` and ``grid.shape[0] == grid_resolution * X.shape[1]``. axes : seq of ndarray The axes with which the grid has been created. """ if len(percentiles) != 2: raise ValueError('percentile must be tuple of len 2') if not all(0. <= x <= 1. for x in percentiles): raise ValueError('percentile values must be in [0, 1]') axes = [] for col in range(X.shape[1]): uniques = np.unique(X[:, col]) if uniques.shape[0] < grid_resolution: # feature has low resolution use unique vals axis = uniques else: emp_percentiles = mquantiles(X, prob=percentiles, axis=0) # create axis based on percentiles and grid resolution axis = np.linspace(emp_percentiles[0, col], emp_percentiles[1, col], num=grid_resolution, endpoint=True) axes.append(axis) return cartesian(axes), axes def partial_dependence(gbrt, target_variables, grid=None, X=None, percentiles=(0.05, 0.95), grid_resolution=100): """Partial dependence of ``target_variables``. Partial dependence plots show the dependence between the joint values of the ``target_variables`` and the function represented by the ``gbrt``. Read more in the :ref:`User Guide <partial_dependence>`. Parameters ---------- gbrt : BaseGradientBoosting A fitted gradient boosting model. target_variables : array-like, dtype=int The target features for which the partial dependecy should be computed (size should be smaller than 3 for visual renderings). grid : array-like, shape=(n_points, len(target_variables)) The grid of ``target_variables`` values for which the partial dependecy should be evaluated (either ``grid`` or ``X`` must be specified). X : array-like, shape=(n_samples, n_features) The data on which ``gbrt`` was trained. It is used to generate a ``grid`` for the ``target_variables``. The ``grid`` comprises ``grid_resolution`` equally spaced points between the two ``percentiles``. percentiles : (low, high), default=(0.05, 0.95) The lower and upper percentile used create the extreme values for the ``grid``. Only if ``X`` is not None. grid_resolution : int, default=100 The number of equally spaced points on the ``grid``. Returns ------- pdp : array, shape=(n_classes, n_points) The partial dependence function evaluated on the ``grid``. For regression and binary classification ``n_classes==1``. axes : seq of ndarray or None The axes with which the grid has been created or None if the grid has been given. Examples -------- >>> samples = [[0, 0, 2], [1, 0, 0]] >>> labels = [0, 1] >>> from sklearn.ensemble import GradientBoostingClassifier >>> gb = GradientBoostingClassifier(random_state=0).fit(samples, labels) >>> kwargs = dict(X=samples, percentiles=(0, 1), grid_resolution=2) >>> partial_dependence(gb, [0], **kwargs) # doctest: +SKIP (array([[-4.52..., 4.52...]]), [array([ 0., 1.])]) """ if not isinstance(gbrt, BaseGradientBoosting): raise ValueError('gbrt has to be an instance of BaseGradientBoosting') if gbrt.estimators_.shape[0] == 0: raise ValueError('Call %s.fit before partial_dependence' % gbrt.__class__.__name__) if (grid is None and X is None) or (grid is not None and X is not None): raise ValueError('Either grid or X must be specified') target_variables = np.asarray(target_variables, dtype=np.int32, order='C').ravel() if any([not (0 <= fx < gbrt.n_features) for fx in target_variables]): raise ValueError('target_variables must be in [0, %d]' % (gbrt.n_features - 1)) if X is not None: X = check_array(X, dtype=DTYPE, order='C') grid, axes = _grid_from_X(X[:, target_variables], percentiles, grid_resolution) else: assert grid is not None # dont return axes if grid is given axes = None # grid must be 2d if grid.ndim == 1: grid = grid[:, np.newaxis] if grid.ndim != 2: raise ValueError('grid must be 2d but is %dd' % grid.ndim) grid = np.asarray(grid, dtype=DTYPE, order='C') assert grid.shape[1] == target_variables.shape[0] n_trees_per_stage = gbrt.estimators_.shape[1] n_estimators = gbrt.estimators_.shape[0] pdp = np.zeros((n_trees_per_stage, grid.shape[0],), dtype=np.float64, order='C') for stage in range(n_estimators): for k in range(n_trees_per_stage): tree = gbrt.estimators_[stage, k].tree_ _partial_dependence_tree(tree, grid, target_variables, gbrt.learning_rate, pdp[k]) return pdp, axes def plot_partial_dependence(gbrt, X, features, feature_names=None, label=None, n_cols=3, grid_resolution=100, percentiles=(0.05, 0.95), n_jobs=1, verbose=0, ax=None, line_kw=None, contour_kw=None, **fig_kw): """Partial dependence plots for ``features``. The ``len(features)`` plots are arranged in a grid with ``n_cols`` columns. Two-way partial dependence plots are plotted as contour plots. Read more in the :ref:`User Guide <partial_dependence>`. Parameters ---------- gbrt : BaseGradientBoosting A fitted gradient boosting model. X : array-like, shape=(n_samples, n_features) The data on which ``gbrt`` was trained. features : seq of tuples or ints If seq[i] is an int or a tuple with one int value, a one-way PDP is created; if seq[i] is a tuple of two ints, a two-way PDP is created. feature_names : seq of str Name of each feature; feature_names[i] holds the name of the feature with index i. label : object The class label for which the PDPs should be computed. Only if gbrt is a multi-class model. Must be in ``gbrt.classes_``. n_cols : int The number of columns in the grid plot (default: 3). percentiles : (low, high), default=(0.05, 0.95) The lower and upper percentile used to create the extreme values for the PDP axes. grid_resolution : int, default=100 The number of equally spaced points on the axes. n_jobs : int The number of CPUs to use to compute the PDs. -1 means 'all CPUs'. Defaults to 1. verbose : int Verbose output during PD computations. Defaults to 0. ax : Matplotlib axis object, default None An axis object onto which the plots will be drawn. line_kw : dict Dict with keywords passed to the ``pylab.plot`` call. For one-way partial dependence plots. contour_kw : dict Dict with keywords passed to the ``pylab.plot`` call. For two-way partial dependence plots. fig_kw : dict Dict with keywords passed to the figure() call. Note that all keywords not recognized above will be automatically included here. Returns ------- fig : figure The Matplotlib Figure object. axs : seq of Axis objects A seq of Axis objects, one for each subplot. Examples -------- >>> from sklearn.datasets import make_friedman1 >>> from sklearn.ensemble import GradientBoostingRegressor >>> X, y = make_friedman1() >>> clf = GradientBoostingRegressor(n_estimators=10).fit(X, y) >>> fig, axs = plot_partial_dependence(clf, X, [0, (0, 1)]) #doctest: +SKIP ... """ import matplotlib.pyplot as plt from matplotlib import transforms from matplotlib.ticker import MaxNLocator from matplotlib.ticker import ScalarFormatter if not isinstance(gbrt, BaseGradientBoosting): raise ValueError('gbrt has to be an instance of BaseGradientBoosting') if gbrt.estimators_.shape[0] == 0: raise ValueError('Call %s.fit before partial_dependence' % gbrt.__class__.__name__) # set label_idx for multi-class GBRT if hasattr(gbrt, 'classes_') and np.size(gbrt.classes_) > 2: if label is None: raise ValueError('label is not given for multi-class PDP') label_idx = np.searchsorted(gbrt.classes_, label) if gbrt.classes_[label_idx] != label: raise ValueError('label %s not in ``gbrt.classes_``' % str(label)) else: # regression and binary classification label_idx = 0 X = check_array(X, dtype=DTYPE, order='C') if gbrt.n_features != X.shape[1]: raise ValueError('X.shape[1] does not match gbrt.n_features') if line_kw is None: line_kw = {'color': 'green'} if contour_kw is None: contour_kw = {} # convert feature_names to list if feature_names is None: # if not feature_names use fx indices as name feature_names = [str(i) for i in range(gbrt.n_features)] elif isinstance(feature_names, np.ndarray): feature_names = feature_names.tolist() def convert_feature(fx): if isinstance(fx, six.string_types): try: fx = feature_names.index(fx) except ValueError: raise ValueError('Feature %s not in feature_names' % fx) return fx # convert features into a seq of int tuples tmp_features = [] for fxs in features: if isinstance(fxs, (numbers.Integral,) + six.string_types): fxs = (fxs,) try: fxs = np.array([convert_feature(fx) for fx in fxs], dtype=np.int32) except TypeError: raise ValueError('features must be either int, str, or tuple ' 'of int/str') if not (1 <= np.size(fxs) <= 2): raise ValueError('target features must be either one or two') tmp_features.append(fxs) features = tmp_features names = [] try: for fxs in features: l = [] # explicit loop so "i" is bound for exception below for i in fxs: l.append(feature_names[i]) names.append(l) except IndexError: raise ValueError('features[i] must be in [0, n_features) ' 'but was %d' % i) # compute PD functions pd_result = Parallel(n_jobs=n_jobs, verbose=verbose)( delayed(partial_dependence)(gbrt, fxs, X=X, grid_resolution=grid_resolution, percentiles=percentiles) for fxs in features) # get global min and max values of PD grouped by plot type pdp_lim = {} for pdp, axes in pd_result: min_pd, max_pd = pdp[label_idx].min(), pdp[label_idx].max() n_fx = len(axes) old_min_pd, old_max_pd = pdp_lim.get(n_fx, (min_pd, max_pd)) min_pd = min(min_pd, old_min_pd) max_pd = max(max_pd, old_max_pd) pdp_lim[n_fx] = (min_pd, max_pd) # create contour levels for two-way plots if 2 in pdp_lim: Z_level = np.linspace(*pdp_lim[2], num=8) if ax is None: fig = plt.figure(**fig_kw) else: fig = ax.get_figure() fig.clear() n_cols = min(n_cols, len(features)) n_rows = int(np.ceil(len(features) / float(n_cols))) axs = [] for i, fx, name, (pdp, axes) in zip(count(), features, names, pd_result): ax = fig.add_subplot(n_rows, n_cols, i + 1) if len(axes) == 1: ax.plot(axes[0], pdp[label_idx].ravel(), **line_kw) else: # make contour plot assert len(axes) == 2 XX, YY = np.meshgrid(axes[0], axes[1]) Z = pdp[label_idx].reshape(list(map(np.size, axes))).T CS = ax.contour(XX, YY, Z, levels=Z_level, linewidths=0.5, colors='k') ax.contourf(XX, YY, Z, levels=Z_level, vmax=Z_level[-1], vmin=Z_level[0], alpha=0.75, **contour_kw) ax.clabel(CS, fmt='%2.2f', colors='k', fontsize=10, inline=True) # plot data deciles + axes labels deciles = mquantiles(X[:, fx[0]], prob=np.arange(0.1, 1.0, 0.1)) trans = transforms.blended_transform_factory(ax.transData, ax.transAxes) ylim = ax.get_ylim() ax.vlines(deciles, [0], 0.05, transform=trans, color='k') ax.set_xlabel(name[0]) ax.set_ylim(ylim) # prevent x-axis ticks from overlapping ax.xaxis.set_major_locator(MaxNLocator(nbins=6, prune='lower')) tick_formatter = ScalarFormatter() tick_formatter.set_powerlimits((-3, 4)) ax.xaxis.set_major_formatter(tick_formatter) if len(axes) > 1: # two-way PDP - y-axis deciles + labels deciles = mquantiles(X[:, fx[1]], prob=np.arange(0.1, 1.0, 0.1)) trans = transforms.blended_transform_factory(ax.transAxes, ax.transData) xlim = ax.get_xlim() ax.hlines(deciles, [0], 0.05, transform=trans, color='k') ax.set_ylabel(name[1]) # hline erases xlim ax.set_xlim(xlim) else: ax.set_ylabel('Partial dependence') if len(axes) == 1: ax.set_ylim(pdp_lim[1]) axs.append(ax) fig.subplots_adjust(bottom=0.15, top=0.7, left=0.1, right=0.95, wspace=0.4, hspace=0.3) return fig, axs
bsd-3-clause
gibiansky/tensorflow
tensorflow/contrib/learn/python/learn/datasets/load_csv_test.py
31
1334
# Copyright 2016 The TensorFlow Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== from __future__ import absolute_import from __future__ import division from __future__ import print_function import tensorflow as tf from tensorflow.contrib.learn.python.learn import datasets class LoadCsvTest(tf.test.TestCase): """Test load csv functions.""" def testIris(self): iris = datasets.load_iris() self.assertTupleEqual(iris.data.shape, (150, 4)) self.assertTupleEqual(iris.target.shape, (150,)) def testBoston(self): boston = datasets.load_boston() self.assertTupleEqual(boston.data.shape, (506, 13)) self.assertTupleEqual(boston.target.shape, (506,)) if __name__ == "__main__": tf.test.main()
apache-2.0
RafaelCosman/pybrain
pybrain/rl/learners/valuebased/nfq.py
31
1994
from scipy import r_ from pybrain.rl.learners.valuebased.valuebased import ValueBasedLearner from pybrain.datasets import SupervisedDataSet from pybrain.supervised.trainers.rprop import RPropMinusTrainer from pybrain.supervised.trainers import BackpropTrainer from pybrain.utilities import one_to_n class NFQ(ValueBasedLearner): """ Neuro-fitted Q-learning""" def __init__(self, maxEpochs=20): ValueBasedLearner.__init__(self) self.gamma = 0.9 self.maxEpochs = maxEpochs def learn(self): # convert reinforcement dataset to NFQ supervised dataset supervised = SupervisedDataSet(self.module.network.indim, 1) for seq in self.dataset: lastexperience = None for state, action, reward in seq: if not lastexperience: # delay each experience in sequence by one lastexperience = (state, action, reward) continue # use experience from last timestep to do Q update (state_, action_, reward_) = lastexperience Q = self.module.getValue(state_, action_[0]) inp = r_[state_, one_to_n(action_[0], self.module.numActions)] tgt = Q + 0.5*(reward_ + self.gamma * max(self.module.getActionValues(state)) - Q) supervised.addSample(inp, tgt) # update last experience with current one lastexperience = (state, action, reward) # train module with backprop/rprop on dataset trainer = RPropMinusTrainer(self.module.network, dataset=supervised, batchlearning=True, verbose=False) trainer.trainUntilConvergence(maxEpochs=self.maxEpochs) # alternative: backprop, was not as stable as rprop # trainer = BackpropTrainer(self.module.network, dataset=supervised, learningrate=0.005, batchlearning=True, verbose=True) # trainer.trainUntilConvergence(maxEpochs=self.maxEpochs)
bsd-3-clause
Cito/sqlalchemy
lib/sqlalchemy/orm/query.py
2
131160
# orm/query.py # Copyright (C) 2005-2014 the SQLAlchemy authors and contributors <see AUTHORS file> # # This module is part of SQLAlchemy and is released under # the MIT License: http://www.opensource.org/licenses/mit-license.php """The Query class and support. Defines the :class:`.Query` class, the central construct used by the ORM to construct database queries. The :class:`.Query` class should not be confused with the :class:`.Select` class, which defines database SELECT operations at the SQL (non-ORM) level. ``Query`` differs from ``Select`` in that it returns ORM-mapped objects and interacts with an ORM session, whereas the ``Select`` construct interacts directly with the database to return iterable result sets. """ from itertools import chain from . import ( attributes, interfaces, object_mapper, persistence, exc as orm_exc, loading ) from .base import _entity_descriptor, _is_aliased_class, \ _is_mapped_class, _orm_columns, _generative from .path_registry import PathRegistry from .util import ( AliasedClass, ORMAdapter, join as orm_join, with_parent, aliased ) from .. import sql, util, log, exc as sa_exc, inspect, inspection from ..sql.expression import _interpret_as_from from ..sql import ( util as sql_util, expression, visitors ) from ..sql.base import ColumnCollection from . import properties __all__ = ['Query', 'QueryContext', 'aliased'] _path_registry = PathRegistry.root @inspection._self_inspects @log.class_logger class Query(object): """ORM-level SQL construction object. :class:`.Query` is the source of all SELECT statements generated by the ORM, both those formulated by end-user query operations as well as by high level internal operations such as related collection loading. It features a generative interface whereby successive calls return a new :class:`.Query` object, a copy of the former with additional criteria and options associated with it. :class:`.Query` objects are normally initially generated using the :meth:`~.Session.query` method of :class:`.Session`. For a full walkthrough of :class:`.Query` usage, see the :ref:`ormtutorial_toplevel`. """ _enable_eagerloads = True _enable_assertions = True _with_labels = False _criterion = None _yield_per = None _order_by = False _group_by = False _having = None _distinct = False _prefixes = None _offset = None _limit = None _for_update_arg = None _statement = None _correlate = frozenset() _populate_existing = False _invoke_all_eagers = True _version_check = False _autoflush = True _only_load_props = None _refresh_state = None _from_obj = () _join_entities = () _select_from_entity = None _mapper_adapter_map = {} _filter_aliases = None _from_obj_alias = None _joinpath = _joinpoint = util.immutabledict() _execution_options = util.immutabledict() _params = util.immutabledict() _attributes = util.immutabledict() _with_options = () _with_hints = () _enable_single_crit = True _current_path = _path_registry def __init__(self, entities, session=None): self.session = session self._polymorphic_adapters = {} self._set_entities(entities) def _set_entities(self, entities, entity_wrapper=None): if entity_wrapper is None: entity_wrapper = _QueryEntity self._entities = [] self._primary_entity = None for ent in util.to_list(entities): entity_wrapper(self, ent) self._set_entity_selectables(self._entities) def _set_entity_selectables(self, entities): self._mapper_adapter_map = d = self._mapper_adapter_map.copy() for ent in entities: for entity in ent.entities: if entity not in d: ext_info = inspect(entity) if not ext_info.is_aliased_class and \ ext_info.mapper.with_polymorphic: if ext_info.mapper.mapped_table not in \ self._polymorphic_adapters: self._mapper_loads_polymorphically_with( ext_info.mapper, sql_util.ColumnAdapter( ext_info.selectable, ext_info.mapper._equivalent_columns ) ) aliased_adapter = None elif ext_info.is_aliased_class: aliased_adapter = sql_util.ColumnAdapter( ext_info.selectable, ext_info.mapper._equivalent_columns ) else: aliased_adapter = None d[entity] = ( ext_info, aliased_adapter ) ent.setup_entity(*d[entity]) def _mapper_loads_polymorphically_with(self, mapper, adapter): for m2 in mapper._with_polymorphic_mappers or [mapper]: self._polymorphic_adapters[m2] = adapter for m in m2.iterate_to_root(): self._polymorphic_adapters[m.local_table] = adapter def _set_select_from(self, obj, set_base_alias): fa = [] select_from_alias = None for from_obj in obj: info = inspect(from_obj) if hasattr(info, 'mapper') and \ (info.is_mapper or info.is_aliased_class): if set_base_alias: raise sa_exc.ArgumentError( "A selectable (FromClause) instance is " "expected when the base alias is being set.") fa.append(info.selectable) elif not info.is_selectable: raise sa_exc.ArgumentError( "argument is not a mapped class, mapper, " "aliased(), or FromClause instance.") else: if isinstance(from_obj, expression.SelectBase): from_obj = from_obj.alias() if set_base_alias: select_from_alias = from_obj fa.append(from_obj) self._from_obj = tuple(fa) if set_base_alias and \ len(self._from_obj) == 1 and \ isinstance(select_from_alias, expression.Alias): equivs = self.__all_equivs() self._from_obj_alias = sql_util.ColumnAdapter( self._from_obj[0], equivs) def _reset_polymorphic_adapter(self, mapper): for m2 in mapper._with_polymorphic_mappers: self._polymorphic_adapters.pop(m2, None) for m in m2.iterate_to_root(): self._polymorphic_adapters.pop(m.local_table, None) def _adapt_polymorphic_element(self, element): if "parententity" in element._annotations: search = element._annotations['parententity'] alias = self._polymorphic_adapters.get(search, None) if alias: return alias.adapt_clause(element) if isinstance(element, expression.FromClause): search = element elif hasattr(element, 'table'): search = element.table else: return None alias = self._polymorphic_adapters.get(search, None) if alias: return alias.adapt_clause(element) def _adapt_col_list(self, cols): return [ self._adapt_clause( expression._literal_as_text(o), True, True) for o in cols ] @_generative() def _adapt_all_clauses(self): self._orm_only_adapt = False def _adapt_clause(self, clause, as_filter, orm_only): """Adapt incoming clauses to transformations which have been applied within this query.""" adapters = [] # do we adapt all expression elements or only those # tagged as 'ORM' constructs ? orm_only = getattr(self, '_orm_only_adapt', orm_only) if as_filter and self._filter_aliases: for fa in self._filter_aliases._visitor_iterator: adapters.append( ( orm_only, fa.replace ) ) if self._from_obj_alias: # for the "from obj" alias, apply extra rule to the # 'ORM only' check, if this query were generated from a # subquery of itself, i.e. _from_selectable(), apply adaption # to all SQL constructs. adapters.append( ( getattr(self, '_orm_only_from_obj_alias', orm_only), self._from_obj_alias.replace ) ) if self._polymorphic_adapters: adapters.append( ( orm_only, self._adapt_polymorphic_element ) ) if not adapters: return clause def replace(elem): for _orm_only, adapter in adapters: # if 'orm only', look for ORM annotations # in the element before adapting. if not _orm_only or \ '_orm_adapt' in elem._annotations or \ "parententity" in elem._annotations: e = adapter(elem) if e is not None: return e return visitors.replacement_traverse( clause, {}, replace ) def _entity_zero(self): return self._entities[0] def _mapper_zero(self): return self._select_from_entity or \ self._entity_zero().entity_zero @property def _mapper_entities(self): for ent in self._entities: if isinstance(ent, _MapperEntity): yield ent def _joinpoint_zero(self): return self._joinpoint.get( '_joinpoint_entity', self._mapper_zero() ) def _mapper_zero_or_none(self): if self._primary_entity: return self._primary_entity.mapper else: return None def _only_mapper_zero(self, rationale=None): if len(self._entities) > 1: raise sa_exc.InvalidRequestError( rationale or "This operation requires a Query " "against a single mapper." ) return self._mapper_zero() def _only_full_mapper_zero(self, methname): if self._entities != [self._primary_entity]: raise sa_exc.InvalidRequestError( "%s() can only be used against " "a single mapped class." % methname) return self._primary_entity.entity_zero def _only_entity_zero(self, rationale=None): if len(self._entities) > 1: raise sa_exc.InvalidRequestError( rationale or "This operation requires a Query " "against a single mapper." ) return self._entity_zero() def __all_equivs(self): equivs = {} for ent in self._mapper_entities: equivs.update(ent.mapper._equivalent_columns) return equivs def _get_condition(self): return self._no_criterion_condition("get", order_by=False, distinct=False) def _get_existing_condition(self): self._no_criterion_assertion("get", order_by=False, distinct=False) def _no_criterion_assertion(self, meth, order_by=True, distinct=True): if not self._enable_assertions: return if self._criterion is not None or \ self._statement is not None or self._from_obj or \ self._limit is not None or self._offset is not None or \ self._group_by or (order_by and self._order_by) or \ (distinct and self._distinct): raise sa_exc.InvalidRequestError( "Query.%s() being called on a " "Query with existing criterion. " % meth) def _no_criterion_condition(self, meth, order_by=True, distinct=True): self._no_criterion_assertion(meth, order_by, distinct) self._from_obj = () self._statement = self._criterion = None self._order_by = self._group_by = self._distinct = False def _no_clauseelement_condition(self, meth): if not self._enable_assertions: return if self._order_by: raise sa_exc.InvalidRequestError( "Query.%s() being called on a " "Query with existing criterion. " % meth) self._no_criterion_condition(meth) def _no_statement_condition(self, meth): if not self._enable_assertions: return if self._statement is not None: raise sa_exc.InvalidRequestError( ("Query.%s() being called on a Query with an existing full " "statement - can't apply criterion.") % meth) def _no_limit_offset(self, meth): if not self._enable_assertions: return if self._limit is not None or self._offset is not None: raise sa_exc.InvalidRequestError( "Query.%s() being called on a Query which already has LIMIT " "or OFFSET applied. To modify the row-limited results of a " " Query, call from_self() first. " "Otherwise, call %s() before limit() or offset() " "are applied." % (meth, meth) ) def _no_select_modifiers(self, meth): if not self._enable_assertions: return for attr, methname, notset in ( ('_limit', 'limit()', None), ('_offset', 'offset()', None), ('_order_by', 'order_by()', False), ('_group_by', 'group_by()', False), ('_distinct', 'distinct()', False), ): if getattr(self, attr) is not notset: raise sa_exc.InvalidRequestError( "Can't call Query.%s() when %s has been called" % (meth, methname) ) def _get_options(self, populate_existing=None, version_check=None, only_load_props=None, refresh_state=None): if populate_existing: self._populate_existing = populate_existing if version_check: self._version_check = version_check if refresh_state: self._refresh_state = refresh_state if only_load_props: self._only_load_props = set(only_load_props) return self def _clone(self): cls = self.__class__ q = cls.__new__(cls) q.__dict__ = self.__dict__.copy() return q @property def statement(self): """The full SELECT statement represented by this Query. The statement by default will not have disambiguating labels applied to the construct unless with_labels(True) is called first. """ stmt = self._compile_context(labels=self._with_labels).\ statement if self._params: stmt = stmt.params(self._params) # TODO: there's no tests covering effects of # the annotation not being there return stmt._annotate({'no_replacement_traverse': True}) def subquery(self, name=None, with_labels=False, reduce_columns=False): """return the full SELECT statement represented by this :class:`.Query`, embedded within an :class:`.Alias`. Eager JOIN generation within the query is disabled. :param name: string name to be assigned as the alias; this is passed through to :meth:`.FromClause.alias`. If ``None``, a name will be deterministically generated at compile time. :param with_labels: if True, :meth:`.with_labels` will be called on the :class:`.Query` first to apply table-qualified labels to all columns. :param reduce_columns: if True, :meth:`.Select.reduce_columns` will be called on the resulting :func:`.select` construct, to remove same-named columns where one also refers to the other via foreign key or WHERE clause equivalence. .. versionchanged:: 0.8 the ``with_labels`` and ``reduce_columns`` keyword arguments were added. """ q = self.enable_eagerloads(False) if with_labels: q = q.with_labels() q = q.statement if reduce_columns: q = q.reduce_columns() return q.alias(name=name) def cte(self, name=None, recursive=False): """Return the full SELECT statement represented by this :class:`.Query` represented as a common table expression (CTE). .. versionadded:: 0.7.6 Parameters and usage are the same as those of the :meth:`.SelectBase.cte` method; see that method for further details. Here is the `Postgresql WITH RECURSIVE example <http://www.postgresql.org/docs/8.4/static/queries-with.html>`_. Note that, in this example, the ``included_parts`` cte and the ``incl_alias`` alias of it are Core selectables, which means the columns are accessed via the ``.c.`` attribute. The ``parts_alias`` object is an :func:`.orm.aliased` instance of the ``Part`` entity, so column-mapped attributes are available directly:: from sqlalchemy.orm import aliased class Part(Base): __tablename__ = 'part' part = Column(String, primary_key=True) sub_part = Column(String, primary_key=True) quantity = Column(Integer) included_parts = session.query( Part.sub_part, Part.part, Part.quantity).\\ filter(Part.part=="our part").\\ cte(name="included_parts", recursive=True) incl_alias = aliased(included_parts, name="pr") parts_alias = aliased(Part, name="p") included_parts = included_parts.union_all( session.query( parts_alias.part, parts_alias.sub_part, parts_alias.quantity).\\ filter(parts_alias.part==incl_alias.c.sub_part) ) q = session.query( included_parts.c.sub_part, func.sum(included_parts.c.quantity). label('total_quantity') ).\\ group_by(included_parts.c.sub_part) .. seealso:: :meth:`.SelectBase.cte` """ return self.enable_eagerloads(False).\ statement.cte(name=name, recursive=recursive) def label(self, name): """Return the full SELECT statement represented by this :class:`.Query`, converted to a scalar subquery with a label of the given name. Analogous to :meth:`sqlalchemy.sql.expression.SelectBase.label`. .. versionadded:: 0.6.5 """ return self.enable_eagerloads(False).statement.label(name) def as_scalar(self): """Return the full SELECT statement represented by this :class:`.Query`, converted to a scalar subquery. Analogous to :meth:`sqlalchemy.sql.expression.SelectBase.as_scalar`. .. versionadded:: 0.6.5 """ return self.enable_eagerloads(False).statement.as_scalar() @property def selectable(self): """Return the :class:`.Select` object emitted by this :class:`.Query`. Used for :func:`.inspect` compatibility, this is equivalent to:: query.enable_eagerloads(False).with_labels().statement """ return self.__clause_element__() def __clause_element__(self): return self.enable_eagerloads(False).with_labels().statement @_generative() def enable_eagerloads(self, value): """Control whether or not eager joins and subqueries are rendered. When set to False, the returned Query will not render eager joins regardless of :func:`~sqlalchemy.orm.joinedload`, :func:`~sqlalchemy.orm.subqueryload` options or mapper-level ``lazy='joined'``/``lazy='subquery'`` configurations. This is used primarily when nesting the Query's statement into a subquery or other selectable. """ self._enable_eagerloads = value @_generative() def with_labels(self): """Apply column labels to the return value of Query.statement. Indicates that this Query's `statement` accessor should return a SELECT statement that applies labels to all columns in the form <tablename>_<columnname>; this is commonly used to disambiguate columns from multiple tables which have the same name. When the `Query` actually issues SQL to load rows, it always uses column labeling. """ self._with_labels = True @_generative() def enable_assertions(self, value): """Control whether assertions are generated. When set to False, the returned Query will not assert its state before certain operations, including that LIMIT/OFFSET has not been applied when filter() is called, no criterion exists when get() is called, and no "from_statement()" exists when filter()/order_by()/group_by() etc. is called. This more permissive mode is used by custom Query subclasses to specify criterion or other modifiers outside of the usual usage patterns. Care should be taken to ensure that the usage pattern is even possible. A statement applied by from_statement() will override any criterion set by filter() or order_by(), for example. """ self._enable_assertions = value @property def whereclause(self): """A readonly attribute which returns the current WHERE criterion for this Query. This returned value is a SQL expression construct, or ``None`` if no criterion has been established. """ return self._criterion @_generative() def _with_current_path(self, path): """indicate that this query applies to objects loaded within a certain path. Used by deferred loaders (see strategies.py) which transfer query options from an originating query to a newly generated query intended for the deferred load. """ self._current_path = path @_generative(_no_clauseelement_condition) def with_polymorphic(self, cls_or_mappers, selectable=None, polymorphic_on=None): """Load columns for inheriting classes. :meth:`.Query.with_polymorphic` applies transformations to the "main" mapped class represented by this :class:`.Query`. The "main" mapped class here means the :class:`.Query` object's first argument is a full class, i.e. ``session.query(SomeClass)``. These transformations allow additional tables to be present in the FROM clause so that columns for a joined-inheritance subclass are available in the query, both for the purposes of load-time efficiency as well as the ability to use these columns at query time. See the documentation section :ref:`with_polymorphic` for details on how this method is used. .. versionchanged:: 0.8 A new and more flexible function :func:`.orm.with_polymorphic` supersedes :meth:`.Query.with_polymorphic`, as it can apply the equivalent functionality to any set of columns or classes in the :class:`.Query`, not just the "zero mapper". See that function for a description of arguments. """ if not self._primary_entity: raise sa_exc.InvalidRequestError( "No primary mapper set up for this Query.") entity = self._entities[0]._clone() self._entities = [entity] + self._entities[1:] entity.set_with_polymorphic(self, cls_or_mappers, selectable=selectable, polymorphic_on=polymorphic_on) @_generative() def yield_per(self, count): """Yield only ``count`` rows at a time. WARNING: use this method with caution; if the same instance is present in more than one batch of rows, end-user changes to attributes will be overwritten. In particular, it's usually impossible to use this setting with eagerly loaded collections (i.e. any lazy='joined' or 'subquery') since those collections will be cleared for a new load when encountered in a subsequent result batch. In the case of 'subquery' loading, the full result for all rows is fetched which generally defeats the purpose of :meth:`~sqlalchemy.orm.query.Query.yield_per`. Also note that while :meth:`~sqlalchemy.orm.query.Query.yield_per` will set the ``stream_results`` execution option to True, currently this is only understood by :mod:`~sqlalchemy.dialects.postgresql.psycopg2` dialect which will stream results using server side cursors instead of pre-buffer all rows for this query. Other DBAPIs pre-buffer all rows before making them available. """ self._yield_per = count self._execution_options = self._execution_options.union( {"stream_results": True}) def get(self, ident): """Return an instance based on the given primary key identifier, or ``None`` if not found. E.g.:: my_user = session.query(User).get(5) some_object = session.query(VersionedFoo).get((5, 10)) :meth:`~.Query.get` is special in that it provides direct access to the identity map of the owning :class:`.Session`. If the given primary key identifier is present in the local identity map, the object is returned directly from this collection and no SQL is emitted, unless the object has been marked fully expired. If not present, a SELECT is performed in order to locate the object. :meth:`~.Query.get` also will perform a check if the object is present in the identity map and marked as expired - a SELECT is emitted to refresh the object as well as to ensure that the row is still present. If not, :class:`~sqlalchemy.orm.exc.ObjectDeletedError` is raised. :meth:`~.Query.get` is only used to return a single mapped instance, not multiple instances or individual column constructs, and strictly on a single primary key value. The originating :class:`.Query` must be constructed in this way, i.e. against a single mapped entity, with no additional filtering criterion. Loading options via :meth:`~.Query.options` may be applied however, and will be used if the object is not yet locally present. A lazy-loading, many-to-one attribute configured by :func:`.relationship`, using a simple foreign-key-to-primary-key criterion, will also use an operation equivalent to :meth:`~.Query.get` in order to retrieve the target value from the local identity map before querying the database. See :doc:`/orm/loading` for further details on relationship loading. :param ident: A scalar or tuple value representing the primary key. For a composite primary key, the order of identifiers corresponds in most cases to that of the mapped :class:`.Table` object's primary key columns. For a :func:`.mapper` that was given the ``primary key`` argument during construction, the order of identifiers corresponds to the elements present in this collection. :return: The object instance, or ``None``. """ # convert composite types to individual args if hasattr(ident, '__composite_values__'): ident = ident.__composite_values__() ident = util.to_list(ident) mapper = self._only_full_mapper_zero("get") if len(ident) != len(mapper.primary_key): raise sa_exc.InvalidRequestError( "Incorrect number of values in identifier to formulate " "primary key for query.get(); primary key columns are %s" % ','.join("'%s'" % c for c in mapper.primary_key)) key = mapper.identity_key_from_primary_key(ident) if not self._populate_existing and \ not mapper.always_refresh and \ self._for_update_arg is None: instance = loading.get_from_identity( self.session, key, attributes.PASSIVE_OFF) if instance is not None: self._get_existing_condition() # reject calls for id in identity map but class # mismatch. if not issubclass(instance.__class__, mapper.class_): return None return instance return loading.load_on_ident(self, key) @_generative() def correlate(self, *args): """Return a :class:`.Query` construct which will correlate the given FROM clauses to that of an enclosing :class:`.Query` or :func:`~.expression.select`. The method here accepts mapped classes, :func:`.aliased` constructs, and :func:`.mapper` constructs as arguments, which are resolved into expression constructs, in addition to appropriate expression constructs. The correlation arguments are ultimately passed to :meth:`.Select.correlate` after coercion to expression constructs. The correlation arguments take effect in such cases as when :meth:`.Query.from_self` is used, or when a subquery as returned by :meth:`.Query.subquery` is embedded in another :func:`~.expression.select` construct. """ self._correlate = self._correlate.union( _interpret_as_from(s) if s is not None else None for s in args) @_generative() def autoflush(self, setting): """Return a Query with a specific 'autoflush' setting. Note that a Session with autoflush=False will not autoflush, even if this flag is set to True at the Query level. Therefore this flag is usually used only to disable autoflush for a specific Query. """ self._autoflush = setting @_generative() def populate_existing(self): """Return a :class:`.Query` that will expire and refresh all instances as they are loaded, or reused from the current :class:`.Session`. :meth:`.populate_existing` does not improve behavior when the ORM is used normally - the :class:`.Session` object's usual behavior of maintaining a transaction and expiring all attributes after rollback or commit handles object state automatically. This method is not intended for general use. """ self._populate_existing = True @_generative() def _with_invoke_all_eagers(self, value): """Set the 'invoke all eagers' flag which causes joined- and subquery loaders to traverse into already-loaded related objects and collections. Default is that of :attr:`.Query._invoke_all_eagers`. """ self._invoke_all_eagers = value def with_parent(self, instance, property=None): """Add filtering criterion that relates the given instance to a child object or collection, using its attribute state as well as an established :func:`.relationship()` configuration. The method uses the :func:`.with_parent` function to generate the clause, the result of which is passed to :meth:`.Query.filter`. Parameters are the same as :func:`.with_parent`, with the exception that the given property can be None, in which case a search is performed against this :class:`.Query` object's target mapper. """ if property is None: mapper = object_mapper(instance) for prop in mapper.iterate_properties: if isinstance(prop, properties.RelationshipProperty) and \ prop.mapper is self._mapper_zero(): property = prop break else: raise sa_exc.InvalidRequestError( "Could not locate a property which relates instances " "of class '%s' to instances of class '%s'" % ( self._mapper_zero().class_.__name__, instance.__class__.__name__) ) return self.filter(with_parent(instance, property)) @_generative() def add_entity(self, entity, alias=None): """add a mapped entity to the list of result columns to be returned.""" if alias is not None: entity = aliased(entity, alias) self._entities = list(self._entities) m = _MapperEntity(self, entity) self._set_entity_selectables([m]) @_generative() def with_session(self, session): """Return a :class:`.Query` that will use the given :class:`.Session`. """ self.session = session def from_self(self, *entities): """return a Query that selects from this Query's SELECT statement. \*entities - optional list of entities which will replace those being selected. """ fromclause = self.with_labels().enable_eagerloads(False).\ _enable_single_crit(False).\ statement.correlate(None) q = self._from_selectable(fromclause) if entities: q._set_entities(entities) return q @_generative() def _enable_single_crit(self, val): self._enable_single_crit = val @_generative() def _from_selectable(self, fromclause): for attr in ( '_statement', '_criterion', '_order_by', '_group_by', '_limit', '_offset', '_joinpath', '_joinpoint', '_distinct', '_having', '_prefixes', ): self.__dict__.pop(attr, None) self._set_select_from([fromclause], True) # this enables clause adaptation for non-ORM # expressions. self._orm_only_from_obj_alias = False old_entities = self._entities self._entities = [] for e in old_entities: e.adapt_to_selectable(self, self._from_obj[0]) def values(self, *columns): """Return an iterator yielding result tuples corresponding to the given list of columns""" if not columns: return iter(()) q = self._clone() q._set_entities(columns, entity_wrapper=_ColumnEntity) if not q._yield_per: q._yield_per = 10 return iter(q) _values = values def value(self, column): """Return a scalar result corresponding to the given column expression.""" try: return next(self.values(column))[0] except StopIteration: return None @_generative() def with_entities(self, *entities): """Return a new :class:`.Query` replacing the SELECT list with the given entities. e.g.:: # Users, filtered on some arbitrary criterion # and then ordered by related email address q = session.query(User).\\ join(User.address).\\ filter(User.name.like('%ed%')).\\ order_by(Address.email) # given *only* User.id==5, Address.email, and 'q', what # would the *next* User in the result be ? subq = q.with_entities(Address.email).\\ order_by(None).\\ filter(User.id==5).\\ subquery() q = q.join((subq, subq.c.email < Address.email)).\\ limit(1) .. versionadded:: 0.6.5 """ self._set_entities(entities) @_generative() def add_columns(self, *column): """Add one or more column expressions to the list of result columns to be returned.""" self._entities = list(self._entities) l = len(self._entities) for c in column: _ColumnEntity(self, c) # _ColumnEntity may add many entities if the # given arg is a FROM clause self._set_entity_selectables(self._entities[l:]) @util.pending_deprecation("0.7", ":meth:`.add_column` is superseded by :meth:`.add_columns`", False) def add_column(self, column): """Add a column expression to the list of result columns to be returned. Pending deprecation: :meth:`.add_column` will be superseded by :meth:`.add_columns`. """ return self.add_columns(column) def options(self, *args): """Return a new Query object, applying the given list of mapper options. Most supplied options regard changing how column- and relationship-mapped attributes are loaded. See the sections :ref:`deferred` and :doc:`/orm/loading` for reference documentation. """ return self._options(False, *args) def _conditional_options(self, *args): return self._options(True, *args) @_generative() def _options(self, conditional, *args): # most MapperOptions write to the '_attributes' dictionary, # so copy that as well self._attributes = self._attributes.copy() opts = tuple(util.flatten_iterator(args)) self._with_options = self._with_options + opts if conditional: for opt in opts: opt.process_query_conditionally(self) else: for opt in opts: opt.process_query(self) def with_transformation(self, fn): """Return a new :class:`.Query` object transformed by the given function. E.g.:: def filter_something(criterion): def transform(q): return q.filter(criterion) return transform q = q.with_transformation(filter_something(x==5)) This allows ad-hoc recipes to be created for :class:`.Query` objects. See the example at :ref:`hybrid_transformers`. .. versionadded:: 0.7.4 """ return fn(self) @_generative() def with_hint(self, selectable, text, dialect_name='*'): """Add an indexing hint for the given entity or selectable to this :class:`.Query`. Functionality is passed straight through to :meth:`~sqlalchemy.sql.expression.Select.with_hint`, with the addition that ``selectable`` can be a :class:`.Table`, :class:`.Alias`, or ORM entity / mapped class /etc. """ selectable = inspect(selectable).selectable self._with_hints += ((selectable, text, dialect_name),) @_generative() def execution_options(self, **kwargs): """ Set non-SQL options which take effect during execution. The options are the same as those accepted by :meth:`.Connection.execution_options`. Note that the ``stream_results`` execution option is enabled automatically if the :meth:`~sqlalchemy.orm.query.Query.yield_per()` method is used. """ self._execution_options = self._execution_options.union(kwargs) @_generative() def with_lockmode(self, mode): """Return a new :class:`.Query` object with the specified "locking mode", which essentially refers to the ``FOR UPDATE`` clause. .. deprecated:: 0.9.0 superseded by :meth:`.Query.with_for_update`. :param mode: a string representing the desired locking mode. Valid values are: * ``None`` - translates to no lockmode * ``'update'`` - translates to ``FOR UPDATE`` (standard SQL, supported by most dialects) * ``'update_nowait'`` - translates to ``FOR UPDATE NOWAIT`` (supported by Oracle, PostgreSQL 8.1 upwards) * ``'read'`` - translates to ``LOCK IN SHARE MODE`` (for MySQL), and ``FOR SHARE`` (for PostgreSQL) .. seealso:: :meth:`.Query.with_for_update` - improved API for specifying the ``FOR UPDATE`` clause. """ self._for_update_arg = LockmodeArg.parse_legacy_query(mode) @_generative() def with_for_update(self, read=False, nowait=False, of=None): """return a new :class:`.Query` with the specified options for the ``FOR UPDATE`` clause. The behavior of this method is identical to that of :meth:`.SelectBase.with_for_update`. When called with no arguments, the resulting ``SELECT`` statement will have a ``FOR UPDATE`` clause appended. When additional arguments are specified, backend-specific options such as ``FOR UPDATE NOWAIT`` or ``LOCK IN SHARE MODE`` can take effect. E.g.:: q = sess.query(User).with_for_update(nowait=True, of=User) The above query on a Postgresql backend will render like:: SELECT users.id AS users_id FROM users FOR UPDATE OF users NOWAIT .. versionadded:: 0.9.0 :meth:`.Query.with_for_update` supersedes the :meth:`.Query.with_lockmode` method. .. seealso:: :meth:`.GenerativeSelect.with_for_update` - Core level method with full argument and behavioral description. """ self._for_update_arg = LockmodeArg(read=read, nowait=nowait, of=of) @_generative() def params(self, *args, **kwargs): """add values for bind parameters which may have been specified in filter(). parameters may be specified using \**kwargs, or optionally a single dictionary as the first positional argument. The reason for both is that \**kwargs is convenient, however some parameter dictionaries contain unicode keys in which case \**kwargs cannot be used. """ if len(args) == 1: kwargs.update(args[0]) elif len(args) > 0: raise sa_exc.ArgumentError( "params() takes zero or one positional argument, " "which is a dictionary.") self._params = self._params.copy() self._params.update(kwargs) @_generative(_no_statement_condition, _no_limit_offset) def filter(self, *criterion): """apply the given filtering criterion to a copy of this :class:`.Query`, using SQL expressions. e.g.:: session.query(MyClass).filter(MyClass.name == 'some name') Multiple criteria are joined together by AND:: session.query(MyClass).\\ filter(MyClass.name == 'some name', MyClass.id > 5) The criterion is any SQL expression object applicable to the WHERE clause of a select. String expressions are coerced into SQL expression constructs via the :func:`.text` construct. .. versionchanged:: 0.7.5 Multiple criteria joined by AND. .. seealso:: :meth:`.Query.filter_by` - filter on keyword expressions. """ for criterion in list(criterion): criterion = expression._literal_as_text(criterion) criterion = self._adapt_clause(criterion, True, True) if self._criterion is not None: self._criterion = self._criterion & criterion else: self._criterion = criterion def filter_by(self, **kwargs): """apply the given filtering criterion to a copy of this :class:`.Query`, using keyword expressions. e.g.:: session.query(MyClass).filter_by(name = 'some name') Multiple criteria are joined together by AND:: session.query(MyClass).\\ filter_by(name = 'some name', id = 5) The keyword expressions are extracted from the primary entity of the query, or the last entity that was the target of a call to :meth:`.Query.join`. .. seealso:: :meth:`.Query.filter` - filter on SQL expressions. """ clauses = [_entity_descriptor(self._joinpoint_zero(), key) == value for key, value in kwargs.items()] return self.filter(sql.and_(*clauses)) @_generative(_no_statement_condition, _no_limit_offset) def order_by(self, *criterion): """apply one or more ORDER BY criterion to the query and return the newly resulting ``Query`` All existing ORDER BY settings can be suppressed by passing ``None`` - this will suppress any ORDER BY configured on mappers as well. Alternatively, an existing ORDER BY setting on the Query object can be entirely cancelled by passing ``False`` as the value - use this before calling methods where an ORDER BY is invalid. """ if len(criterion) == 1: if criterion[0] is False: if '_order_by' in self.__dict__: del self._order_by return if criterion[0] is None: self._order_by = None return criterion = self._adapt_col_list(criterion) if self._order_by is False or self._order_by is None: self._order_by = criterion else: self._order_by = self._order_by + criterion @_generative(_no_statement_condition, _no_limit_offset) def group_by(self, *criterion): """apply one or more GROUP BY criterion to the query and return the newly resulting :class:`.Query`""" criterion = list(chain(*[_orm_columns(c) for c in criterion])) criterion = self._adapt_col_list(criterion) if self._group_by is False: self._group_by = criterion else: self._group_by = self._group_by + criterion @_generative(_no_statement_condition, _no_limit_offset) def having(self, criterion): """apply a HAVING criterion to the query and return the newly resulting :class:`.Query`. :meth:`~.Query.having` is used in conjunction with :meth:`~.Query.group_by`. HAVING criterion makes it possible to use filters on aggregate functions like COUNT, SUM, AVG, MAX, and MIN, eg.:: q = session.query(User.id).\\ join(User.addresses).\\ group_by(User.id).\\ having(func.count(Address.id) > 2) """ if isinstance(criterion, util.string_types): criterion = sql.text(criterion) if criterion is not None and \ not isinstance(criterion, sql.ClauseElement): raise sa_exc.ArgumentError( "having() argument must be of type " "sqlalchemy.sql.ClauseElement or string") criterion = self._adapt_clause(criterion, True, True) if self._having is not None: self._having = self._having & criterion else: self._having = criterion def union(self, *q): """Produce a UNION of this Query against one or more queries. e.g.:: q1 = sess.query(SomeClass).filter(SomeClass.foo=='bar') q2 = sess.query(SomeClass).filter(SomeClass.bar=='foo') q3 = q1.union(q2) The method accepts multiple Query objects so as to control the level of nesting. A series of ``union()`` calls such as:: x.union(y).union(z).all() will nest on each ``union()``, and produces:: SELECT * FROM (SELECT * FROM (SELECT * FROM X UNION SELECT * FROM y) UNION SELECT * FROM Z) Whereas:: x.union(y, z).all() produces:: SELECT * FROM (SELECT * FROM X UNION SELECT * FROM y UNION SELECT * FROM Z) Note that many database backends do not allow ORDER BY to be rendered on a query called within UNION, EXCEPT, etc. To disable all ORDER BY clauses including those configured on mappers, issue ``query.order_by(None)`` - the resulting :class:`.Query` object will not render ORDER BY within its SELECT statement. """ return self._from_selectable( expression.union(*([self] + list(q)))) def union_all(self, *q): """Produce a UNION ALL of this Query against one or more queries. Works the same way as :meth:`~sqlalchemy.orm.query.Query.union`. See that method for usage examples. """ return self._from_selectable( expression.union_all(*([self] + list(q))) ) def intersect(self, *q): """Produce an INTERSECT of this Query against one or more queries. Works the same way as :meth:`~sqlalchemy.orm.query.Query.union`. See that method for usage examples. """ return self._from_selectable( expression.intersect(*([self] + list(q))) ) def intersect_all(self, *q): """Produce an INTERSECT ALL of this Query against one or more queries. Works the same way as :meth:`~sqlalchemy.orm.query.Query.union`. See that method for usage examples. """ return self._from_selectable( expression.intersect_all(*([self] + list(q))) ) def except_(self, *q): """Produce an EXCEPT of this Query against one or more queries. Works the same way as :meth:`~sqlalchemy.orm.query.Query.union`. See that method for usage examples. """ return self._from_selectable( expression.except_(*([self] + list(q))) ) def except_all(self, *q): """Produce an EXCEPT ALL of this Query against one or more queries. Works the same way as :meth:`~sqlalchemy.orm.query.Query.union`. See that method for usage examples. """ return self._from_selectable( expression.except_all(*([self] + list(q))) ) def join(self, *props, **kwargs): """Create a SQL JOIN against this :class:`.Query` object's criterion and apply generatively, returning the newly resulting :class:`.Query`. **Simple Relationship Joins** Consider a mapping between two classes ``User`` and ``Address``, with a relationship ``User.addresses`` representing a collection of ``Address`` objects associated with each ``User``. The most common usage of :meth:`~.Query.join` is to create a JOIN along this relationship, using the ``User.addresses`` attribute as an indicator for how this should occur:: q = session.query(User).join(User.addresses) Where above, the call to :meth:`~.Query.join` along ``User.addresses`` will result in SQL equivalent to:: SELECT user.* FROM user JOIN address ON user.id = address.user_id In the above example we refer to ``User.addresses`` as passed to :meth:`~.Query.join` as the *on clause*, that is, it indicates how the "ON" portion of the JOIN should be constructed. For a single-entity query such as the one above (i.e. we start by selecting only from ``User`` and nothing else), the relationship can also be specified by its string name:: q = session.query(User).join("addresses") :meth:`~.Query.join` can also accommodate multiple "on clause" arguments to produce a chain of joins, such as below where a join across four related entities is constructed:: q = session.query(User).join("orders", "items", "keywords") The above would be shorthand for three separate calls to :meth:`~.Query.join`, each using an explicit attribute to indicate the source entity:: q = session.query(User).\\ join(User.orders).\\ join(Order.items).\\ join(Item.keywords) **Joins to a Target Entity or Selectable** A second form of :meth:`~.Query.join` allows any mapped entity or core selectable construct as a target. In this usage, :meth:`~.Query.join` will attempt to create a JOIN along the natural foreign key relationship between two entities:: q = session.query(User).join(Address) The above calling form of :meth:`~.Query.join` will raise an error if either there are no foreign keys between the two entities, or if there are multiple foreign key linkages between them. In the above calling form, :meth:`~.Query.join` is called upon to create the "on clause" automatically for us. The target can be any mapped entity or selectable, such as a :class:`.Table`:: q = session.query(User).join(addresses_table) **Joins to a Target with an ON Clause** The third calling form allows both the target entity as well as the ON clause to be passed explicitly. Suppose for example we wanted to join to ``Address`` twice, using an alias the second time. We use :func:`~sqlalchemy.orm.aliased` to create a distinct alias of ``Address``, and join to it using the ``target, onclause`` form, so that the alias can be specified explicitly as the target along with the relationship to instruct how the ON clause should proceed:: a_alias = aliased(Address) q = session.query(User).\\ join(User.addresses).\\ join(a_alias, User.addresses).\\ filter(Address.email_address=='ed@foo.com').\\ filter(a_alias.email_address=='ed@bar.com') Where above, the generated SQL would be similar to:: SELECT user.* FROM user JOIN address ON user.id = address.user_id JOIN address AS address_1 ON user.id=address_1.user_id WHERE address.email_address = :email_address_1 AND address_1.email_address = :email_address_2 The two-argument calling form of :meth:`~.Query.join` also allows us to construct arbitrary joins with SQL-oriented "on clause" expressions, not relying upon configured relationships at all. Any SQL expression can be passed as the ON clause when using the two-argument form, which should refer to the target entity in some way as well as an applicable source entity:: q = session.query(User).join(Address, User.id==Address.user_id) .. versionchanged:: 0.7 In SQLAlchemy 0.6 and earlier, the two argument form of :meth:`~.Query.join` requires the usage of a tuple: ``query(User).join((Address, User.id==Address.user_id))``\ . This calling form is accepted in 0.7 and further, though is not necessary unless multiple join conditions are passed to a single :meth:`~.Query.join` call, which itself is also not generally necessary as it is now equivalent to multiple calls (this wasn't always the case). **Advanced Join Targeting and Adaption** There is a lot of flexibility in what the "target" can be when using :meth:`~.Query.join`. As noted previously, it also accepts :class:`.Table` constructs and other selectables such as :func:`.alias` and :func:`.select` constructs, with either the one or two-argument forms:: addresses_q = select([Address.user_id]).\\ where(Address.email_address.endswith("@bar.com")).\\ alias() q = session.query(User).\\ join(addresses_q, addresses_q.c.user_id==User.id) :meth:`~.Query.join` also features the ability to *adapt* a :meth:`~sqlalchemy.orm.relationship` -driven ON clause to the target selectable. Below we construct a JOIN from ``User`` to a subquery against ``Address``, allowing the relationship denoted by ``User.addresses`` to *adapt* itself to the altered target:: address_subq = session.query(Address).\\ filter(Address.email_address == 'ed@foo.com').\\ subquery() q = session.query(User).join(address_subq, User.addresses) Producing SQL similar to:: SELECT user.* FROM user JOIN ( SELECT address.id AS id, address.user_id AS user_id, address.email_address AS email_address FROM address WHERE address.email_address = :email_address_1 ) AS anon_1 ON user.id = anon_1.user_id The above form allows one to fall back onto an explicit ON clause at any time:: q = session.query(User).\\ join(address_subq, User.id==address_subq.c.user_id) **Controlling what to Join From** While :meth:`~.Query.join` exclusively deals with the "right" side of the JOIN, we can also control the "left" side, in those cases where it's needed, using :meth:`~.Query.select_from`. Below we construct a query against ``Address`` but can still make usage of ``User.addresses`` as our ON clause by instructing the :class:`.Query` to select first from the ``User`` entity:: q = session.query(Address).select_from(User).\\ join(User.addresses).\\ filter(User.name == 'ed') Which will produce SQL similar to:: SELECT address.* FROM user JOIN address ON user.id=address.user_id WHERE user.name = :name_1 **Constructing Aliases Anonymously** :meth:`~.Query.join` can construct anonymous aliases using the ``aliased=True`` flag. This feature is useful when a query is being joined algorithmically, such as when querying self-referentially to an arbitrary depth:: q = session.query(Node).\\ join("children", "children", aliased=True) When ``aliased=True`` is used, the actual "alias" construct is not explicitly available. To work with it, methods such as :meth:`.Query.filter` will adapt the incoming entity to the last join point:: q = session.query(Node).\\ join("children", "children", aliased=True).\\ filter(Node.name == 'grandchild 1') When using automatic aliasing, the ``from_joinpoint=True`` argument can allow a multi-node join to be broken into multiple calls to :meth:`~.Query.join`, so that each path along the way can be further filtered:: q = session.query(Node).\\ join("children", aliased=True).\\ filter(Node.name='child 1').\\ join("children", aliased=True, from_joinpoint=True).\\ filter(Node.name == 'grandchild 1') The filtering aliases above can then be reset back to the original ``Node`` entity using :meth:`~.Query.reset_joinpoint`:: q = session.query(Node).\\ join("children", "children", aliased=True).\\ filter(Node.name == 'grandchild 1').\\ reset_joinpoint().\\ filter(Node.name == 'parent 1) For an example of ``aliased=True``, see the distribution example :ref:`examples_xmlpersistence` which illustrates an XPath-like query system using algorithmic joins. :param \*props: A collection of one or more join conditions, each consisting of a relationship-bound attribute or string relationship name representing an "on clause", or a single target entity, or a tuple in the form of ``(target, onclause)``. A special two-argument calling form of the form ``target, onclause`` is also accepted. :param aliased=False: If True, indicate that the JOIN target should be anonymously aliased. Subsequent calls to :meth:`~.Query.filter` and similar will adapt the incoming criterion to the target alias, until :meth:`~.Query.reset_joinpoint` is called. :param from_joinpoint=False: When using ``aliased=True``, a setting of True here will cause the join to be from the most recent joined target, rather than starting back from the original FROM clauses of the query. .. seealso:: :ref:`ormtutorial_joins` in the ORM tutorial. :ref:`inheritance_toplevel` for details on how :meth:`~.Query.join` is used for inheritance relationships. :func:`.orm.join` - a standalone ORM-level join function, used internally by :meth:`.Query.join`, which in previous SQLAlchemy versions was the primary ORM-level joining interface. """ aliased, from_joinpoint = kwargs.pop('aliased', False),\ kwargs.pop('from_joinpoint', False) if kwargs: raise TypeError("unknown arguments: %s" % ','.join(kwargs.keys)) return self._join(props, outerjoin=False, create_aliases=aliased, from_joinpoint=from_joinpoint) def outerjoin(self, *props, **kwargs): """Create a left outer join against this ``Query`` object's criterion and apply generatively, returning the newly resulting ``Query``. Usage is the same as the ``join()`` method. """ aliased, from_joinpoint = kwargs.pop('aliased', False), \ kwargs.pop('from_joinpoint', False) if kwargs: raise TypeError("unknown arguments: %s" % ','.join(kwargs)) return self._join(props, outerjoin=True, create_aliases=aliased, from_joinpoint=from_joinpoint) def _update_joinpoint(self, jp): self._joinpoint = jp # copy backwards to the root of the _joinpath # dict, so that no existing dict in the path is mutated while 'prev' in jp: f, prev = jp['prev'] prev = prev.copy() prev[f] = jp jp['prev'] = (f, prev) jp = prev self._joinpath = jp @_generative(_no_statement_condition, _no_limit_offset) def _join(self, keys, outerjoin, create_aliases, from_joinpoint): """consumes arguments from join() or outerjoin(), places them into a consistent format with which to form the actual JOIN constructs. """ if not from_joinpoint: self._reset_joinpoint() if len(keys) == 2 and \ isinstance(keys[0], (expression.FromClause, type, AliasedClass)) and \ isinstance(keys[1], (str, expression.ClauseElement, interfaces.PropComparator)): # detect 2-arg form of join and # convert to a tuple. keys = (keys,) for arg1 in util.to_list(keys): if isinstance(arg1, tuple): # "tuple" form of join, multiple # tuples are accepted as well. The simpler # "2-arg" form is preferred. May deprecate # the "tuple" usage. arg1, arg2 = arg1 else: arg2 = None # determine onclause/right_entity. there # is a little bit of legacy behavior still at work here # which means they might be in either order. may possibly # lock this down to (right_entity, onclause) in 0.6. if isinstance(arg1, (interfaces.PropComparator, util.string_types)): right_entity, onclause = arg2, arg1 else: right_entity, onclause = arg1, arg2 left_entity = prop = None if isinstance(onclause, util.string_types): left_entity = self._joinpoint_zero() descriptor = _entity_descriptor(left_entity, onclause) onclause = descriptor # check for q.join(Class.propname, from_joinpoint=True) # and Class is that of the current joinpoint elif from_joinpoint and \ isinstance(onclause, interfaces.PropComparator): left_entity = onclause._parententity info = inspect(self._joinpoint_zero()) left_mapper, left_selectable, left_is_aliased = \ getattr(info, 'mapper', None), \ info.selectable, \ getattr(info, 'is_aliased_class', None) if left_mapper is left_entity: left_entity = self._joinpoint_zero() descriptor = _entity_descriptor(left_entity, onclause.key) onclause = descriptor if isinstance(onclause, interfaces.PropComparator): if right_entity is None: right_entity = onclause.property.mapper of_type = getattr(onclause, '_of_type', None) if of_type: right_entity = of_type else: right_entity = onclause.property.mapper left_entity = onclause._parententity prop = onclause.property if not isinstance(onclause, attributes.QueryableAttribute): onclause = prop if not create_aliases: # check for this path already present. # don't render in that case. edge = (left_entity, right_entity, prop.key) if edge in self._joinpoint: # The child's prev reference might be stale -- # it could point to a parent older than the # current joinpoint. If this is the case, # then we need to update it and then fix the # tree's spine with _update_joinpoint. Copy # and then mutate the child, which might be # shared by a different query object. jp = self._joinpoint[edge].copy() jp['prev'] = (edge, self._joinpoint) self._update_joinpoint(jp) continue elif onclause is not None and right_entity is None: # TODO: no coverage here raise NotImplementedError("query.join(a==b) not supported.") self._join_left_to_right( left_entity, right_entity, onclause, outerjoin, create_aliases, prop) def _join_left_to_right(self, left, right, onclause, outerjoin, create_aliases, prop): """append a JOIN to the query's from clause.""" self._polymorphic_adapters = self._polymorphic_adapters.copy() if left is None: if self._from_obj: left = self._from_obj[0] elif self._entities: left = self._entities[0].entity_zero_or_selectable if left is right and \ not create_aliases: raise sa_exc.InvalidRequestError( "Can't construct a join from %s to %s, they " "are the same entity" % (left, right)) l_info = inspect(left) r_info = inspect(right) overlap = False if not create_aliases: right_mapper = getattr(r_info, "mapper", None) # if the target is a joined inheritance mapping, # be more liberal about auto-aliasing. if right_mapper and ( right_mapper.with_polymorphic or isinstance(right_mapper.mapped_table, expression.Join) ): for from_obj in self._from_obj or [l_info.selectable]: if sql_util.selectables_overlap(l_info.selectable, from_obj) and \ sql_util.selectables_overlap(from_obj, r_info.selectable): overlap = True break elif sql_util.selectables_overlap(l_info.selectable, r_info.selectable): overlap = True if overlap and l_info.selectable is r_info.selectable: raise sa_exc.InvalidRequestError( "Can't join table/selectable '%s' to itself" % l_info.selectable) right, onclause = self._prepare_right_side( r_info, right, onclause, create_aliases, prop, overlap) # if joining on a MapperProperty path, # track the path to prevent redundant joins if not create_aliases and prop: self._update_joinpoint({ '_joinpoint_entity': right, 'prev': ((left, right, prop.key), self._joinpoint) }) else: self._joinpoint = {'_joinpoint_entity': right} self._join_to_left(l_info, left, right, onclause, outerjoin) def _prepare_right_side(self, r_info, right, onclause, create_aliases, prop, overlap): info = r_info right_mapper, right_selectable, right_is_aliased = \ getattr(info, 'mapper', None), \ info.selectable, \ getattr(info, 'is_aliased_class', False) if right_mapper: self._join_entities += (info, ) if right_mapper and prop and \ not right_mapper.common_parent(prop.mapper): raise sa_exc.InvalidRequestError( "Join target %s does not correspond to " "the right side of join condition %s" % (right, onclause) ) if not right_mapper and prop: right_mapper = prop.mapper need_adapter = False if right_mapper and right is right_selectable: if not right_selectable.is_derived_from( right_mapper.mapped_table): raise sa_exc.InvalidRequestError( "Selectable '%s' is not derived from '%s'" % (right_selectable.description, right_mapper.mapped_table.description)) if isinstance(right_selectable, expression.SelectBase): # TODO: this isn't even covered now! right_selectable = right_selectable.alias() need_adapter = True right = aliased(right_mapper, right_selectable) aliased_entity = right_mapper and \ not right_is_aliased and \ ( right_mapper.with_polymorphic and isinstance( right_mapper._with_polymorphic_selectable, expression.Alias) or overlap # test for overlap: # orm/inheritance/relationships.py # SelfReferentialM2MTest ) if not need_adapter and (create_aliases or aliased_entity): right = aliased(right, flat=True) need_adapter = True # if an alias() of the right side was generated here, # apply an adapter to all subsequent filter() calls # until reset_joinpoint() is called. if need_adapter: self._filter_aliases = ORMAdapter(right, equivalents=right_mapper and right_mapper._equivalent_columns or {}, chain_to=self._filter_aliases) # if the onclause is a ClauseElement, adapt it with any # adapters that are in place right now if isinstance(onclause, expression.ClauseElement): onclause = self._adapt_clause(onclause, True, True) # if an alias() on the right side was generated, # which is intended to wrap a the right side in a subquery, # ensure that columns retrieved from this target in the result # set are also adapted. if aliased_entity and not create_aliases: self._mapper_loads_polymorphically_with( right_mapper, ORMAdapter( right, equivalents=right_mapper._equivalent_columns ) ) return right, onclause def _join_to_left(self, l_info, left, right, onclause, outerjoin): info = l_info left_mapper = getattr(info, 'mapper', None) left_selectable = info.selectable if self._from_obj: replace_clause_index, clause = sql_util.find_join_source( self._from_obj, left_selectable) if clause is not None: try: clause = orm_join(clause, right, onclause, isouter=outerjoin) except sa_exc.ArgumentError as ae: raise sa_exc.InvalidRequestError( "Could not find a FROM clause to join from. " "Tried joining to %s, but got: %s" % (right, ae)) self._from_obj = \ self._from_obj[:replace_clause_index] + \ (clause, ) + \ self._from_obj[replace_clause_index + 1:] return if left_mapper: for ent in self._entities: if ent.corresponds_to(left): clause = ent.selectable break else: clause = left else: clause = left_selectable assert clause is not None try: clause = orm_join(clause, right, onclause, isouter=outerjoin) except sa_exc.ArgumentError as ae: raise sa_exc.InvalidRequestError( "Could not find a FROM clause to join from. " "Tried joining to %s, but got: %s" % (right, ae)) self._from_obj = self._from_obj + (clause,) def _reset_joinpoint(self): self._joinpoint = self._joinpath self._filter_aliases = None @_generative(_no_statement_condition) def reset_joinpoint(self): """Return a new :class:`.Query`, where the "join point" has been reset back to the base FROM entities of the query. This method is usually used in conjunction with the ``aliased=True`` feature of the :meth:`~.Query.join` method. See the example in :meth:`~.Query.join` for how this is used. """ self._reset_joinpoint() @_generative(_no_clauseelement_condition) def select_from(self, *from_obj): """Set the FROM clause of this :class:`.Query` explicitly. :meth:`.Query.select_from` is often used in conjunction with :meth:`.Query.join` in order to control which entity is selected from on the "left" side of the join. The entity or selectable object here effectively replaces the "left edge" of any calls to :meth:`~.Query.join`, when no joinpoint is otherwise established - usually, the default "join point" is the leftmost entity in the :class:`~.Query` object's list of entities to be selected. A typical example:: q = session.query(Address).select_from(User).\\ join(User.addresses).\\ filter(User.name == 'ed') Which produces SQL equivalent to:: SELECT address.* FROM user JOIN address ON user.id=address.user_id WHERE user.name = :name_1 :param \*from_obj: collection of one or more entities to apply to the FROM clause. Entities can be mapped classes, :class:`.AliasedClass` objects, :class:`.Mapper` objects as well as core :class:`.FromClause` elements like subqueries. .. versionchanged:: 0.9 This method no longer applies the given FROM object to be the selectable from which matching entities select from; the :meth:`.select_entity_from` method now accomplishes this. See that method for a description of this behavior. .. seealso:: :meth:`~.Query.join` :meth:`.Query.select_entity_from` """ self._set_select_from(from_obj, False) @_generative(_no_clauseelement_condition) def select_entity_from(self, from_obj): """Set the FROM clause of this :class:`.Query` to a core selectable, applying it as a replacement FROM clause for corresponding mapped entities. This method is similar to the :meth:`.Query.select_from` method, in that it sets the FROM clause of the query. However, where :meth:`.Query.select_from` only affects what is placed in the FROM, this method also applies the given selectable to replace the FROM which the selected entities would normally select from. The given ``from_obj`` must be an instance of a :class:`.FromClause`, e.g. a :func:`.select` or :class:`.Alias` construct. An example would be a :class:`.Query` that selects ``User`` entities, but uses :meth:`.Query.select_entity_from` to have the entities selected from a :func:`.select` construct instead of the base ``user`` table:: select_stmt = select([User]).where(User.id == 7) q = session.query(User).\\ select_entity_from(select_stmt).\\ filter(User.name == 'ed') The query generated will select ``User`` entities directly from the given :func:`.select` construct, and will be:: SELECT anon_1.id AS anon_1_id, anon_1.name AS anon_1_name FROM (SELECT "user".id AS id, "user".name AS name FROM "user" WHERE "user".id = :id_1) AS anon_1 WHERE anon_1.name = :name_1 Notice above that even the WHERE criterion was "adapted" such that the ``anon_1`` subquery effectively replaces all references to the ``user`` table, except for the one that it refers to internally. Compare this to :meth:`.Query.select_from`, which as of version 0.9, does not affect existing entities. The statement below:: q = session.query(User).\\ select_from(select_stmt).\\ filter(User.name == 'ed') Produces SQL where both the ``user`` table as well as the ``select_stmt`` construct are present as separate elements in the FROM clause. No "adaptation" of the ``user`` table is applied:: SELECT "user".id AS user_id, "user".name AS user_name FROM "user", (SELECT "user".id AS id, "user".name AS name FROM "user" WHERE "user".id = :id_1) AS anon_1 WHERE "user".name = :name_1 :meth:`.Query.select_entity_from` maintains an older behavior of :meth:`.Query.select_from`. In modern usage, similar results can also be achieved using :func:`.aliased`:: select_stmt = select([User]).where(User.id == 7) user_from_select = aliased(User, select_stmt.alias()) q = session.query(user_from_select) :param from_obj: a :class:`.FromClause` object that will replace the FROM clause of this :class:`.Query`. .. seealso:: :meth:`.Query.select_from` .. versionadded:: 0.8 :meth:`.Query.select_entity_from` was added to specify the specific behavior of entity replacement, however the :meth:`.Query.select_from` maintains this behavior as well until 0.9. """ self._set_select_from([from_obj], True) def __getitem__(self, item): if isinstance(item, slice): start, stop, step = util.decode_slice(item) if isinstance(stop, int) and \ isinstance(start, int) and \ stop - start <= 0: return [] # perhaps we should execute a count() here so that we # can still use LIMIT/OFFSET ? elif (isinstance(start, int) and start < 0) \ or (isinstance(stop, int) and stop < 0): return list(self)[item] res = self.slice(start, stop) if step is not None: return list(res)[None:None:item.step] else: return list(res) else: if item == -1: return list(self)[-1] else: return list(self[item:item + 1])[0] @_generative(_no_statement_condition) def slice(self, start, stop): """apply LIMIT/OFFSET to the ``Query`` based on a " "range and return the newly resulting ``Query``.""" if start is not None and stop is not None: self._offset = (self._offset or 0) + start self._limit = stop - start elif start is None and stop is not None: self._limit = stop elif start is not None and stop is None: self._offset = (self._offset or 0) + start if self._offset == 0: self._offset = None @_generative(_no_statement_condition) def limit(self, limit): """Apply a ``LIMIT`` to the query and return the newly resulting ``Query``. """ self._limit = limit @_generative(_no_statement_condition) def offset(self, offset): """Apply an ``OFFSET`` to the query and return the newly resulting ``Query``. """ self._offset = offset @_generative(_no_statement_condition) def distinct(self, *criterion): """Apply a ``DISTINCT`` to the query and return the newly resulting ``Query``. :param \*expr: optional column expressions. When present, the Postgresql dialect will render a ``DISTINCT ON (<expressions>>)`` construct. """ if not criterion: self._distinct = True else: criterion = self._adapt_col_list(criterion) if isinstance(self._distinct, list): self._distinct += criterion else: self._distinct = criterion @_generative() def prefix_with(self, *prefixes): """Apply the prefixes to the query and return the newly resulting ``Query``. :param \*prefixes: optional prefixes, typically strings, not using any commas. In particular is useful for MySQL keywords. e.g.:: query = sess.query(User.name).\\ prefix_with('HIGH_PRIORITY').\\ prefix_with('SQL_SMALL_RESULT', 'ALL') Would render:: SELECT HIGH_PRIORITY SQL_SMALL_RESULT ALL users.name AS users_name FROM users .. versionadded:: 0.7.7 """ if self._prefixes: self._prefixes += prefixes else: self._prefixes = prefixes def all(self): """Return the results represented by this ``Query`` as a list. This results in an execution of the underlying query. """ return list(self) @_generative(_no_clauseelement_condition) def from_statement(self, statement): """Execute the given SELECT statement and return results. This method bypasses all internal statement compilation, and the statement is executed without modification. The statement argument is either a string, a ``select()`` construct, or a ``text()`` construct, and should return the set of columns appropriate to the entity class represented by this ``Query``. """ if isinstance(statement, util.string_types): statement = sql.text(statement) if not isinstance(statement, (expression.TextClause, expression.SelectBase)): raise sa_exc.ArgumentError( "from_statement accepts text(), select(), " "and union() objects only.") self._statement = statement def first(self): """Return the first result of this ``Query`` or None if the result doesn't contain any row. first() applies a limit of one within the generated SQL, so that only one primary entity row is generated on the server side (note this may consist of multiple result rows if join-loaded collections are present). Calling ``first()`` results in an execution of the underlying query. """ if self._statement is not None: ret = list(self)[0:1] else: ret = list(self[0:1]) if len(ret) > 0: return ret[0] else: return None def one(self): """Return exactly one result or raise an exception. Raises ``sqlalchemy.orm.exc.NoResultFound`` if the query selects no rows. Raises ``sqlalchemy.orm.exc.MultipleResultsFound`` if multiple object identities are returned, or if multiple rows are returned for a query that does not return object identities. Note that an entity query, that is, one which selects one or more mapped classes as opposed to individual column attributes, may ultimately represent many rows but only one row of unique entity or entities - this is a successful result for one(). Calling ``one()`` results in an execution of the underlying query. .. versionchanged:: 0.6 ``one()`` fully fetches all results instead of applying any kind of limit, so that the "unique"-ing of entities does not conceal multiple object identities. """ ret = list(self) l = len(ret) if l == 1: return ret[0] elif l == 0: raise orm_exc.NoResultFound("No row was found for one()") else: raise orm_exc.MultipleResultsFound( "Multiple rows were found for one()") def scalar(self): """Return the first element of the first result or None if no rows present. If multiple rows are returned, raises MultipleResultsFound. >>> session.query(Item).scalar() <Item> >>> session.query(Item.id).scalar() 1 >>> session.query(Item.id).filter(Item.id < 0).scalar() None >>> session.query(Item.id, Item.name).scalar() 1 >>> session.query(func.count(Parent.id)).scalar() 20 This results in an execution of the underlying query. """ try: ret = self.one() if not isinstance(ret, tuple): return ret return ret[0] except orm_exc.NoResultFound: return None def __iter__(self): context = self._compile_context() context.statement.use_labels = True if self._autoflush and not self._populate_existing: self.session._autoflush() return self._execute_and_instances(context) def _connection_from_session(self, **kw): conn = self.session.connection( **kw) if self._execution_options: conn = conn.execution_options(**self._execution_options) return conn def _execute_and_instances(self, querycontext): conn = self._connection_from_session( mapper=self._mapper_zero_or_none(), clause=querycontext.statement, close_with_result=True) result = conn.execute(querycontext.statement, self._params) return loading.instances(self, result, querycontext) @property def column_descriptions(self): """Return metadata about the columns which would be returned by this :class:`.Query`. Format is a list of dictionaries:: user_alias = aliased(User, name='user2') q = sess.query(User, User.id, user_alias) # this expression: q.column_descriptions # would return: [ { 'name':'User', 'type':User, 'aliased':False, 'expr':User, }, { 'name':'id', 'type':Integer(), 'aliased':False, 'expr':User.id, }, { 'name':'user2', 'type':User, 'aliased':True, 'expr':user_alias } ] """ return [ { 'name': ent._label_name, 'type': ent.type, 'aliased': getattr(ent, 'is_aliased_class', False), 'expr': ent.expr } for ent in self._entities ] def instances(self, cursor, __context=None): """Given a ResultProxy cursor as returned by connection.execute(), return an ORM result as an iterator. e.g.:: result = engine.execute("select * from users") for u in session.query(User).instances(result): print u """ context = __context if context is None: context = QueryContext(self) return loading.instances(self, cursor, context) def merge_result(self, iterator, load=True): """Merge a result into this :class:`.Query` object's Session. Given an iterator returned by a :class:`.Query` of the same structure as this one, return an identical iterator of results, with all mapped instances merged into the session using :meth:`.Session.merge`. This is an optimized method which will merge all mapped instances, preserving the structure of the result rows and unmapped columns with less method overhead than that of calling :meth:`.Session.merge` explicitly for each value. The structure of the results is determined based on the column list of this :class:`.Query` - if these do not correspond, unchecked errors will occur. The 'load' argument is the same as that of :meth:`.Session.merge`. For an example of how :meth:`~.Query.merge_result` is used, see the source code for the example :ref:`examples_caching`, where :meth:`~.Query.merge_result` is used to efficiently restore state from a cache back into a target :class:`.Session`. """ return loading.merge_result(self, iterator, load) @property def _select_args(self): return { 'limit': self._limit, 'offset': self._offset, 'distinct': self._distinct, 'prefixes': self._prefixes, 'group_by': self._group_by or None, 'having': self._having } @property def _should_nest_selectable(self): kwargs = self._select_args return (kwargs.get('limit') is not None or kwargs.get('offset') is not None or kwargs.get('distinct', False)) def exists(self): """A convenience method that turns a query into an EXISTS subquery of the form EXISTS (SELECT 1 FROM ... WHERE ...). e.g.:: q = session.query(User).filter(User.name == 'fred') session.query(q.exists()) Producing SQL similar to:: SELECT EXISTS ( SELECT 1 FROM users WHERE users.name = :name_1 ) AS anon_1 .. versionadded:: 0.8.1 """ return sql.exists(self.with_labels().statement.with_only_columns(['1'])) def count(self): """Return a count of rows this Query would return. This generates the SQL for this Query as follows:: SELECT count(1) AS count_1 FROM ( SELECT <rest of query follows...> ) AS anon_1 .. versionchanged:: 0.7 The above scheme is newly refined as of 0.7b3. For fine grained control over specific columns to count, to skip the usage of a subquery or otherwise control of the FROM clause, or to use other aggregate functions, use :attr:`~sqlalchemy.sql.expression.func` expressions in conjunction with :meth:`~.Session.query`, i.e.:: from sqlalchemy import func # count User records, without # using a subquery. session.query(func.count(User.id)) # return count of user "id" grouped # by "name" session.query(func.count(User.id)).\\ group_by(User.name) from sqlalchemy import distinct # count distinct "name" values session.query(func.count(distinct(User.name))) """ col = sql.func.count(sql.literal_column('*')) return self.from_self(col).scalar() def delete(self, synchronize_session='evaluate'): """Perform a bulk delete query. Deletes rows matched by this query from the database. :param synchronize_session: chooses the strategy for the removal of matched objects from the session. Valid values are: ``False`` - don't synchronize the session. This option is the most efficient and is reliable once the session is expired, which typically occurs after a commit(), or explicitly using expire_all(). Before the expiration, objects may still remain in the session which were in fact deleted which can lead to confusing results if they are accessed via get() or already loaded collections. ``'fetch'`` - performs a select query before the delete to find objects that are matched by the delete query and need to be removed from the session. Matched objects are removed from the session. ``'evaluate'`` - Evaluate the query's criteria in Python straight on the objects in the session. If evaluation of the criteria isn't implemented, an error is raised. In that case you probably want to use the 'fetch' strategy as a fallback. The expression evaluator currently doesn't account for differing string collations between the database and Python. :return: the count of rows matched as returned by the database's "row count" feature. This method has several key caveats: * The method does **not** offer in-Python cascading of relationships - it is assumed that ON DELETE CASCADE/SET NULL/etc. is configured for any foreign key references which require it, otherwise the database may emit an integrity violation if foreign key references are being enforced. After the DELETE, dependent objects in the :class:`.Session` which were impacted by an ON DELETE may not contain the current state, or may have been deleted. This issue is resolved once the :class:`.Session` is expired, which normally occurs upon :meth:`.Session.commit` or can be forced by using :meth:`.Session.expire_all`. Accessing an expired object whose row has been deleted will invoke a SELECT to locate the row; when the row is not found, an :class:`~sqlalchemy.orm.exc.ObjectDeletedError` is raised. * The :meth:`.MapperEvents.before_delete` and :meth:`.MapperEvents.after_delete` events are **not** invoked from this method. Instead, the :meth:`.SessionEvents.after_bulk_delete` method is provided to act upon a mass DELETE of entity rows. .. seealso:: :meth:`.Query.update` :ref:`inserts_and_updates` - Core SQL tutorial """ #TODO: cascades need handling. delete_op = persistence.BulkDelete.factory( self, synchronize_session) delete_op.exec_() return delete_op.rowcount def update(self, values, synchronize_session='evaluate'): """Perform a bulk update query. Updates rows matched by this query in the database. :param values: a dictionary with attributes names as keys and literal values or sql expressions as values. :param synchronize_session: chooses the strategy to update the attributes on objects in the session. Valid values are: ``False`` - don't synchronize the session. This option is the most efficient and is reliable once the session is expired, which typically occurs after a commit(), or explicitly using expire_all(). Before the expiration, updated objects may still remain in the session with stale values on their attributes, which can lead to confusing results. ``'fetch'`` - performs a select query before the update to find objects that are matched by the update query. The updated attributes are expired on matched objects. ``'evaluate'`` - Evaluate the Query's criteria in Python straight on the objects in the session. If evaluation of the criteria isn't implemented, an exception is raised. The expression evaluator currently doesn't account for differing string collations between the database and Python. :return: the count of rows matched as returned by the database's "row count" feature. This method has several key caveats: * The method does **not** offer in-Python cascading of relationships - it is assumed that ON UPDATE CASCADE is configured for any foreign key references which require it, otherwise the database may emit an integrity violation if foreign key references are being enforced. After the UPDATE, dependent objects in the :class:`.Session` which were impacted by an ON UPDATE CASCADE may not contain the current state; this issue is resolved once the :class:`.Session` is expired, which normally occurs upon :meth:`.Session.commit` or can be forced by using :meth:`.Session.expire_all`. * As of 0.8, this method will support multiple table updates, as detailed in :ref:`multi_table_updates`, and this behavior does extend to support updates of joined-inheritance and other multiple table mappings. However, the **join condition of an inheritance mapper is currently not automatically rendered**. Care must be taken in any multiple-table update to explicitly include the joining condition between those tables, even in mappings where this is normally automatic. E.g. if a class ``Engineer`` subclasses ``Employee``, an UPDATE of the ``Engineer`` local table using criteria against the ``Employee`` local table might look like:: session.query(Engineer).\\ filter(Engineer.id == Employee.id).\\ filter(Employee.name == 'dilbert').\\ update({"engineer_type": "programmer"}) * The :meth:`.MapperEvents.before_update` and :meth:`.MapperEvents.after_update` events are **not** invoked from this method. Instead, the :meth:`.SessionEvents.after_bulk_update` method is provided to act upon a mass UPDATE of entity rows. .. seealso:: :meth:`.Query.delete` :ref:`inserts_and_updates` - Core SQL tutorial """ #TODO: value keys need to be mapped to corresponding sql cols and # instr.attr.s to string keys #TODO: updates of manytoone relationships need to be converted to # fk assignments #TODO: cascades need handling. update_op = persistence.BulkUpdate.factory( self, synchronize_session, values) update_op.exec_() return update_op.rowcount def _compile_context(self, labels=True): context = QueryContext(self) if context.statement is not None: return context context.labels = labels context._for_update_arg = self._for_update_arg for entity in self._entities: entity.setup_context(self, context) for rec in context.create_eager_joins: strategy = rec[0] strategy(*rec[1:]) if context.from_clause: # "load from explicit FROMs" mode, # i.e. when select_from() or join() is used context.froms = list(context.from_clause) else: # "load from discrete FROMs" mode, # i.e. when each _MappedEntity has its own FROM context.froms = context.froms if self._enable_single_crit: self._adjust_for_single_inheritance(context) if not context.primary_columns: if self._only_load_props: raise sa_exc.InvalidRequestError( "No column-based properties specified for " "refresh operation. Use session.expire() " "to reload collections and related items.") else: raise sa_exc.InvalidRequestError( "Query contains no columns with which to " "SELECT from.") if context.multi_row_eager_loaders and self._should_nest_selectable: context.statement = self._compound_eager_statement(context) else: context.statement = self._simple_statement(context) return context def _compound_eager_statement(self, context): # for eager joins present and LIMIT/OFFSET/DISTINCT, # wrap the query inside a select, # then append eager joins onto that if context.order_by: order_by_col_expr = list( chain(*[ sql_util.unwrap_order_by(o) for o in context.order_by ]) ) else: context.order_by = None order_by_col_expr = [] inner = sql.select( context.primary_columns + order_by_col_expr, context.whereclause, from_obj=context.froms, use_labels=context.labels, # TODO: this order_by is only needed if # LIMIT/OFFSET is present in self._select_args, # else the application on the outside is enough order_by=context.order_by, **self._select_args ) for hint in self._with_hints: inner = inner.with_hint(*hint) if self._correlate: inner = inner.correlate(*self._correlate) inner = inner.alias() equivs = self.__all_equivs() context.adapter = sql_util.ColumnAdapter(inner, equivs) statement = sql.select( [inner] + context.secondary_columns, use_labels=context.labels) statement._for_update_arg = context._for_update_arg from_clause = inner for eager_join in context.eager_joins.values(): # EagerLoader places a 'stop_on' attribute on the join, # giving us a marker as to where the "splice point" of # the join should be from_clause = sql_util.splice_joins( from_clause, eager_join, eager_join.stop_on) statement.append_from(from_clause) if context.order_by: statement.append_order_by( *context.adapter.copy_and_process( context.order_by ) ) statement.append_order_by(*context.eager_order_by) return statement def _simple_statement(self, context): if not context.order_by: context.order_by = None if self._distinct and context.order_by: order_by_col_expr = list( chain(*[ sql_util.unwrap_order_by(o) for o in context.order_by ]) ) context.primary_columns += order_by_col_expr context.froms += tuple(context.eager_joins.values()) statement = sql.select( context.primary_columns + context.secondary_columns, context.whereclause, from_obj=context.froms, use_labels=context.labels, order_by=context.order_by, **self._select_args ) statement._for_update_arg = context._for_update_arg for hint in self._with_hints: statement = statement.with_hint(*hint) if self._correlate: statement = statement.correlate(*self._correlate) if context.eager_order_by: statement.append_order_by(*context.eager_order_by) return statement def _adjust_for_single_inheritance(self, context): """Apply single-table-inheritance filtering. For all distinct single-table-inheritance mappers represented in the columns clause of this query, add criterion to the WHERE clause of the given QueryContext such that only the appropriate subtypes are selected from the total results. """ for (ext_info, adapter) in self._mapper_adapter_map.values(): if ext_info in self._join_entities: continue single_crit = ext_info.mapper._single_table_criterion if single_crit is not None: if adapter: single_crit = adapter.traverse(single_crit) single_crit = self._adapt_clause(single_crit, False, False) context.whereclause = sql.and_( sql.True_._ifnone(context.whereclause), single_crit) def __str__(self): return str(self._compile_context().statement) from ..sql.selectable import ForUpdateArg class LockmodeArg(ForUpdateArg): @classmethod def parse_legacy_query(self, mode): if mode in (None, False): return None if mode == "read": read = True nowait = False elif mode == "update": read = nowait = False elif mode == "update_nowait": nowait = True read = False else: raise sa_exc.ArgumentError( "Unknown with_lockmode argument: %r" % mode) return LockmodeArg(read=read, nowait=nowait) class _QueryEntity(object): """represent an entity column returned within a Query result.""" def __new__(cls, *args, **kwargs): if cls is _QueryEntity: entity = args[1] if not isinstance(entity, util.string_types) and \ _is_mapped_class(entity): cls = _MapperEntity elif isinstance(entity, Bundle): cls = _BundleEntity else: cls = _ColumnEntity return object.__new__(cls) def _clone(self): q = self.__class__.__new__(self.__class__) q.__dict__ = self.__dict__.copy() return q class _MapperEntity(_QueryEntity): """mapper/class/AliasedClass entity""" def __init__(self, query, entity): if not query._primary_entity: query._primary_entity = self query._entities.append(self) self.entities = [entity] self.expr = entity supports_single_entity = True def setup_entity(self, ext_info, aliased_adapter): self.mapper = ext_info.mapper self.aliased_adapter = aliased_adapter self.selectable = ext_info.selectable self.is_aliased_class = ext_info.is_aliased_class self._with_polymorphic = ext_info.with_polymorphic_mappers self._polymorphic_discriminator = \ ext_info.polymorphic_on self.entity_zero = ext_info if ext_info.is_aliased_class: self._label_name = self.entity_zero.name else: self._label_name = self.mapper.class_.__name__ self.path = self.entity_zero._path_registry self.custom_rows = bool(self.mapper.dispatch.append_result) def set_with_polymorphic(self, query, cls_or_mappers, selectable, polymorphic_on): """Receive an update from a call to query.with_polymorphic(). Note the newer style of using a free standing with_polymporphic() construct doesn't make use of this method. """ if self.is_aliased_class: # TODO: invalidrequest ? raise NotImplementedError( "Can't use with_polymorphic() against " "an Aliased object" ) if cls_or_mappers is None: query._reset_polymorphic_adapter(self.mapper) return mappers, from_obj = self.mapper._with_polymorphic_args( cls_or_mappers, selectable) self._with_polymorphic = mappers self._polymorphic_discriminator = polymorphic_on self.selectable = from_obj query._mapper_loads_polymorphically_with(self.mapper, sql_util.ColumnAdapter(from_obj, self.mapper._equivalent_columns)) filter_fn = id @property def type(self): return self.mapper.class_ @property def entity_zero_or_selectable(self): return self.entity_zero def corresponds_to(self, entity): if entity.is_aliased_class: if self.is_aliased_class: if entity._base_alias is self.entity_zero._base_alias: return True return False elif self.is_aliased_class: if self.entity_zero._use_mapper_path: return entity in self._with_polymorphic else: return entity is self.entity_zero return entity.common_parent(self.entity_zero) def adapt_to_selectable(self, query, sel): query._entities.append(self) def _get_entity_clauses(self, query, context): adapter = None if not self.is_aliased_class: if query._polymorphic_adapters: adapter = query._polymorphic_adapters.get(self.mapper, None) else: adapter = self.aliased_adapter if adapter: if query._from_obj_alias: ret = adapter.wrap(query._from_obj_alias) else: ret = adapter else: ret = query._from_obj_alias return ret def row_processor(self, query, context, custom_rows): adapter = self._get_entity_clauses(query, context) if context.adapter and adapter: adapter = adapter.wrap(context.adapter) elif not adapter: adapter = context.adapter # polymorphic mappers which have concrete tables in # their hierarchy usually # require row aliasing unconditionally. if not adapter and self.mapper._requires_row_aliasing: adapter = sql_util.ColumnAdapter( self.selectable, self.mapper._equivalent_columns) if query._primary_entity is self: _instance = loading.instance_processor( self.mapper, context, self.path, adapter, only_load_props=query._only_load_props, refresh_state=context.refresh_state, polymorphic_discriminator=self._polymorphic_discriminator ) else: _instance = loading.instance_processor( self.mapper, context, self.path, adapter, polymorphic_discriminator=self._polymorphic_discriminator ) return _instance, self._label_name def setup_context(self, query, context): adapter = self._get_entity_clauses(query, context) #if self._adapted_selectable is None: context.froms += (self.selectable,) if context.order_by is False and self.mapper.order_by: context.order_by = self.mapper.order_by # apply adaptation to the mapper's order_by if needed. if adapter: context.order_by = adapter.adapt_list( util.to_list( context.order_by ) ) if self._with_polymorphic: poly_properties = self.mapper._iterate_polymorphic_properties( self._with_polymorphic) else: poly_properties = self.mapper._polymorphic_properties for value in poly_properties: if query._only_load_props and \ value.key not in query._only_load_props: continue value.setup( context, self, self.path, adapter, only_load_props=query._only_load_props, column_collection=context.primary_columns ) if self._polymorphic_discriminator is not None and \ self._polymorphic_discriminator \ is not self.mapper.polymorphic_on: if adapter: pd = adapter.columns[self._polymorphic_discriminator] else: pd = self._polymorphic_discriminator context.primary_columns.append(pd) def __str__(self): return str(self.mapper) @inspection._self_inspects class Bundle(object): """A grouping of SQL expressions that are returned by a :class:`.Query` under one namespace. The :class:`.Bundle` essentially allows nesting of the tuple-based results returned by a column-oriented :class:`.Query` object. It also is extensible via simple subclassing, where the primary capability to override is that of how the set of expressions should be returned, allowing post-processing as well as custom return types, without involving ORM identity-mapped classes. .. versionadded:: 0.9.0 .. seealso:: :ref:`bundles` """ single_entity = False """If True, queries for a single Bundle will be returned as a single entity, rather than an element within a keyed tuple.""" def __init__(self, name, *exprs, **kw): """Construct a new :class:`.Bundle`. e.g.:: bn = Bundle("mybundle", MyClass.x, MyClass.y) for row in session.query(bn).filter(bn.c.x == 5).filter(bn.c.y == 4): print(row.mybundle.x, row.mybundle.y) :param name: name of the bundle. :param \*exprs: columns or SQL expressions comprising the bundle. :param single_entity=False: if True, rows for this :class:`.Bundle` can be returned as a "single entity" outside of any enclosing tuple in the same manner as a mapped entity. """ self.name = self._label = name self.exprs = exprs self.c = self.columns = ColumnCollection() self.columns.update((getattr(col, "key", col._label), col) for col in exprs) self.single_entity = kw.pop('single_entity', self.single_entity) columns = None """A namespace of SQL expressions referred to by this :class:`.Bundle`. e.g.:: bn = Bundle("mybundle", MyClass.x, MyClass.y) q = sess.query(bn).filter(bn.c.x == 5) Nesting of bundles is also supported:: b1 = Bundle("b1", Bundle('b2', MyClass.a, MyClass.b), Bundle('b3', MyClass.x, MyClass.y) ) q = sess.query(b1).filter(b1.c.b2.c.a == 5).filter(b1.c.b3.c.y == 9) .. seealso:: :attr:`.Bundle.c` """ c = None """An alias for :attr:`.Bundle.columns`.""" def _clone(self): cloned = self.__class__.__new__(self.__class__) cloned.__dict__.update(self.__dict__) return cloned def __clause_element__(self): return expression.ClauseList(group=False, *self.c) @property def clauses(self): return self.__clause_element__().clauses def label(self, name): """Provide a copy of this :class:`.Bundle` passing a new label.""" cloned = self._clone() cloned.name = name return cloned def create_row_processor(self, query, procs, labels): """Produce the "row processing" function for this :class:`.Bundle`. May be overridden by subclasses. .. seealso:: :ref:`bundles` - includes an example of subclassing. """ def proc(row, result): return util.KeyedTuple([proc(row, None) for proc in procs], labels) return proc class _BundleEntity(_QueryEntity): def __init__(self, query, bundle, setup_entities=True): query._entities.append(self) self.bundle = self.expr = bundle self.type = type(bundle) self._label_name = bundle.name self._entities = [] if setup_entities: for expr in bundle.exprs: if isinstance(expr, Bundle): _BundleEntity(self, expr) else: _ColumnEntity(self, expr, namespace=self) self.entities = () self.filter_fn = lambda item: item self.supports_single_entity = self.bundle.single_entity custom_rows = False @property def entity_zero(self): for ent in self._entities: ezero = ent.entity_zero if ezero is not None: return ezero else: return None def corresponds_to(self, entity): # TODO: this seems to have no effect for # _ColumnEntity either return False @property def entity_zero_or_selectable(self): for ent in self._entities: ezero = ent.entity_zero_or_selectable if ezero is not None: return ezero else: return None def adapt_to_selectable(self, query, sel): c = _BundleEntity(query, self.bundle, setup_entities=False) #c._label_name = self._label_name #c.entity_zero = self.entity_zero #c.entities = self.entities for ent in self._entities: ent.adapt_to_selectable(c, sel) def setup_entity(self, ext_info, aliased_adapter): for ent in self._entities: ent.setup_entity(ext_info, aliased_adapter) def setup_context(self, query, context): for ent in self._entities: ent.setup_context(query, context) def row_processor(self, query, context, custom_rows): procs, labels = zip( *[ent.row_processor(query, context, custom_rows) for ent in self._entities] ) proc = self.bundle.create_row_processor(query, procs, labels) return proc, self._label_name class _ColumnEntity(_QueryEntity): """Column/expression based entity.""" def __init__(self, query, column, namespace=None): self.expr = column self.namespace = namespace if isinstance(column, util.string_types): column = sql.literal_column(column) self._label_name = column.name elif isinstance(column, ( attributes.QueryableAttribute, interfaces.PropComparator )): self._label_name = column.key column = column._query_clause_element() else: self._label_name = getattr(column, 'key', None) if not isinstance(column, expression.ColumnElement) and \ hasattr(column, '_select_iterable'): for c in column._select_iterable: if c is column: break _ColumnEntity(query, c, namespace=column) if c is not column: return elif isinstance(column, Bundle): _BundleEntity(query, column) return if not isinstance(column, sql.ColumnElement): raise sa_exc.InvalidRequestError( "SQL expression, column, or mapped entity " "expected - got '%r'" % (column, ) ) self.type = type_ = column.type if type_.hashable: self.filter_fn = lambda item: item else: counter = util.counter() self.filter_fn = lambda item: counter() # If the Column is unnamed, give it a # label() so that mutable column expressions # can be located in the result even # if the expression's identity has been changed # due to adaption. if not column._label and not getattr(column, 'is_literal', False): column = column.label(self._label_name) query._entities.append(self) self.column = column self.froms = set() # look for ORM entities represented within the # given expression. Try to count only entities # for columns whose FROM object is in the actual list # of FROMs for the overall expression - this helps # subqueries which were built from ORM constructs from # leaking out their entities into the main select construct self.actual_froms = actual_froms = set(column._from_objects) self.entities = util.OrderedSet( elem._annotations['parententity'] for elem in visitors.iterate(column, {}) if 'parententity' in elem._annotations and actual_froms.intersection(elem._from_objects) ) if self.entities: self.entity_zero = list(self.entities)[0] elif self.namespace is not None: self.entity_zero = self.namespace else: self.entity_zero = None supports_single_entity = False custom_rows = False @property def entity_zero_or_selectable(self): if self.entity_zero is not None: return self.entity_zero elif self.actual_froms: return list(self.actual_froms)[0] else: return None def adapt_to_selectable(self, query, sel): c = _ColumnEntity(query, sel.corresponding_column(self.column)) c._label_name = self._label_name c.entity_zero = self.entity_zero c.entities = self.entities def setup_entity(self, ext_info, aliased_adapter): if 'selectable' not in self.__dict__: self.selectable = ext_info.selectable self.froms.add(ext_info.selectable) def corresponds_to(self, entity): # TODO: just returning False here, # no tests fail if self.entity_zero is None: return False elif _is_aliased_class(entity): # TODO: polymorphic subclasses ? return entity is self.entity_zero else: return not _is_aliased_class(self.entity_zero) and \ entity.common_parent(self.entity_zero) def _resolve_expr_against_query_aliases(self, query, expr, context): return query._adapt_clause(expr, False, True) def row_processor(self, query, context, custom_rows): column = self._resolve_expr_against_query_aliases( query, self.column, context) if context.adapter: column = context.adapter.columns[column] def proc(row, result): return row[column] return proc, self._label_name def setup_context(self, query, context): column = self._resolve_expr_against_query_aliases( query, self.column, context) context.froms += tuple(self.froms) context.primary_columns.append(column) def __str__(self): return str(self.column) class QueryContext(object): multi_row_eager_loaders = False adapter = None froms = () for_update = None def __init__(self, query): if query._statement is not None: if isinstance(query._statement, expression.SelectBase) and \ not query._statement._textual and \ not query._statement.use_labels: self.statement = query._statement.apply_labels() else: self.statement = query._statement else: self.statement = None self.from_clause = query._from_obj self.whereclause = query._criterion self.order_by = query._order_by self.query = query self.session = query.session self.populate_existing = query._populate_existing self.invoke_all_eagers = query._invoke_all_eagers self.version_check = query._version_check self.refresh_state = query._refresh_state self.primary_columns = [] self.secondary_columns = [] self.eager_order_by = [] self.eager_joins = {} self.create_eager_joins = [] self.propagate_options = set(o for o in query._with_options if o.propagate_to_loaders) self.attributes = query._attributes.copy() class AliasOption(interfaces.MapperOption): def __init__(self, alias): """Return a :class:`.MapperOption` that will indicate to the :class:`.Query` that the main table has been aliased. This is a seldom-used option to suit the very rare case that :func:`.contains_eager` is being used in conjunction with a user-defined SELECT statement that aliases the parent table. E.g.:: # define an aliased UNION called 'ulist' ulist = users.select(users.c.user_id==7).\\ union(users.select(users.c.user_id>7)).\\ alias('ulist') # add on an eager load of "addresses" statement = ulist.outerjoin(addresses).\\ select().apply_labels() # create query, indicating "ulist" will be an # alias for the main table, "addresses" # property should be eager loaded query = session.query(User).options( contains_alias(ulist), contains_eager(User.addresses)) # then get results via the statement results = query.from_statement(statement).all() :param alias: is the string name of an alias, or a :class:`~.sql.expression.Alias` object representing the alias. """ self.alias = alias def process_query(self, query): if isinstance(self.alias, util.string_types): alias = query._mapper_zero().mapped_table.alias(self.alias) else: alias = self.alias query._from_obj_alias = sql_util.ColumnAdapter(alias)
mit
irit-melodi/attelo
attelo/metrics/classification_structured.py
3
5266
"""Classification metrics for structured outputs. """ from collections import Counter from itertools import chain, izip import numpy as np def _unique_labels(y): """Set of unique labels in y""" return set(y_ij[1] for y_ij in chain.from_iterable(y_i for y_i in y)) def unique_labels(*ys): """Extract an ordered array of unique labels. Parameters ---------- elt_type: string Type of each element, determines how to find the label See also -------- This is the structured version of `sklearn.utils.multiclass.unique_labels` """ ys_labels = set(chain.from_iterable(_unique_labels(y) for y in ys)) # TODO check the set of labels contains a unique (e.g. string) type # of values return np.array(sorted(ys_labels)) def precision_recall_fscore_support(y_true, y_pred, labels=None, average=None): """Compute precision, recall, F-measure and support for each class. The support is the number of occurrences of each class in ``y_true``. This is essentially a structured version of sklearn.metrics.classification.precision_recall_fscore_support . It should apply equally well to lists of constituency tree spans and lists of dependency edges. Parameters ---------- y_true: list of iterable Ground truth target structures, encoded in a sparse format (e.g. list of edges or span descriptions). y_pred: list of iterable Estimated target structures, encoded in a sparse format (e.g. list of edges or span descriptions). labels: list, optional The set of labels to include, and their order if ``average is None``. average: string, [None (default), 'binary', 'micro', 'macro'] If ``None``, the scores for each class are returned. Otherwise, this determines the type of averaging performed on the data: ``'binary'``: Only report results for the positive class. This is applicable only if targets are binary. ``'micro'``: Calculate metrics globally by counting the total true positives, false negatives and false positives. ``'macro'``: Calculate metrics for each label, and find their unweighted mean. This does not take label imbalance into account. Returns ------- precision: float (if average is not None) or array of float, shape=\ [n_unique_labels] recall: float (if average is not None) or array of float, shape=\ [n_unique_labels] fscore: float (if average is not None) or array of float, shape=\ [n_unique_labels] support: int (if average is not None) or array of int, shape=\ [n_unique_labels] The number of occurrences of each label in ``ctree_true``. """ average_options = frozenset([None, 'micro', 'macro']) if average not in average_options: raise ValueError('average has to be one of' + str(average_options)) # TMP if average == 'macro': raise NotImplementedError('average currently has to be micro or None') # end TMP # gather an ordered list of unique labels from y_true and y_pred present_labels = unique_labels(y_true, y_pred) if labels is None: labels = present_labels # n_labels = None else: # EXPERIMENTAL labels = [lbl for lbl in labels if lbl in present_labels] # n_labels = len(labels) # FIXME complete/fix this # raise ValueError('Parameter `labels` is currently unsupported') # end EXPERIMENTAL # compute tp_sum, pred_sum, true_sum # true positives for each tree tp = [set(yi_true) & set(yi_pred) for yi_true, yi_pred in izip(y_true, y_pred)] # TODO find a nicer and faster design that resembles sklearn's, e.g. # use np.bincount instead of collections.Counter tp_sum = Counter(y_ij[1] for y_ij in chain.from_iterable(tp)) true_sum = Counter(y_ij[1] for y_ij in chain.from_iterable(y_true)) pred_sum = Counter(y_ij[1] for y_ij in chain.from_iterable(y_pred)) # transform to np arrays of floats tp_sum = np.array([float(tp_sum[lbl]) for lbl in labels]) true_sum = np.array([float(true_sum[lbl]) for lbl in labels]) pred_sum = np.array([float(pred_sum[lbl]) for lbl in labels]) # TODO rewrite to compute by summing over scores broken down by label if average == 'micro': tp_sum = np.array([tp_sum.sum()]) true_sum = np.array([true_sum.sum()]) pred_sum = np.array([pred_sum.sum()]) # finally compute the desired statistics # when the div denominator is 0, assign 0.0 (instead of np.inf) precision = tp_sum / pred_sum precision[pred_sum == 0] = 0.0 recall = tp_sum / true_sum recall[true_sum == 0] = 0.0 f_score = 2 * (precision * recall) / (precision + recall) f_score[precision + recall == 0] = 0.0 if average is not None: precision = np.average(precision) recall = np.average(recall) f_score = np.average(f_score) true_sum = np.average(true_sum) # != sklearn: we keep the support return precision, recall, f_score, true_sum
gpl-3.0
roxyboy/scikit-learn
sklearn/tree/tests/test_export.py
130
9950
""" Testing for export functions of decision trees (sklearn.tree.export). """ from re import finditer from numpy.testing import assert_equal from nose.tools import assert_raises from sklearn.tree import DecisionTreeClassifier, DecisionTreeRegressor from sklearn.ensemble import GradientBoostingClassifier from sklearn.tree import export_graphviz from sklearn.externals.six import StringIO from sklearn.utils.testing import assert_in # toy sample X = [[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]] y = [-1, -1, -1, 1, 1, 1] y2 = [[-1, 1], [-1, 2], [-1, 3], [1, 1], [1, 2], [1, 3]] w = [1, 1, 1, .5, .5, .5] def test_graphviz_toy(): # Check correctness of export_graphviz clf = DecisionTreeClassifier(max_depth=3, min_samples_split=1, criterion="gini", random_state=2) clf.fit(X, y) # Test export code out = StringIO() export_graphviz(clf, out_file=out) contents1 = out.getvalue() contents2 = 'digraph Tree {\n' \ 'node [shape=box] ;\n' \ '0 [label="X[0] <= 0.0\\ngini = 0.5\\nsamples = 6\\n' \ 'value = [3, 3]"] ;\n' \ '1 [label="gini = 0.0\\nsamples = 3\\nvalue = [3, 0]"] ;\n' \ '0 -> 1 [labeldistance=2.5, labelangle=45, ' \ 'headlabel="True"] ;\n' \ '2 [label="gini = 0.0\\nsamples = 3\\nvalue = [0, 3]"] ;\n' \ '0 -> 2 [labeldistance=2.5, labelangle=-45, ' \ 'headlabel="False"] ;\n' \ '}' assert_equal(contents1, contents2) # Test with feature_names out = StringIO() export_graphviz(clf, out_file=out, feature_names=["feature0", "feature1"]) contents1 = out.getvalue() contents2 = 'digraph Tree {\n' \ 'node [shape=box] ;\n' \ '0 [label="feature0 <= 0.0\\ngini = 0.5\\nsamples = 6\\n' \ 'value = [3, 3]"] ;\n' \ '1 [label="gini = 0.0\\nsamples = 3\\nvalue = [3, 0]"] ;\n' \ '0 -> 1 [labeldistance=2.5, labelangle=45, ' \ 'headlabel="True"] ;\n' \ '2 [label="gini = 0.0\\nsamples = 3\\nvalue = [0, 3]"] ;\n' \ '0 -> 2 [labeldistance=2.5, labelangle=-45, ' \ 'headlabel="False"] ;\n' \ '}' assert_equal(contents1, contents2) # Test with class_names out = StringIO() export_graphviz(clf, out_file=out, class_names=["yes", "no"]) contents1 = out.getvalue() contents2 = 'digraph Tree {\n' \ 'node [shape=box] ;\n' \ '0 [label="X[0] <= 0.0\\ngini = 0.5\\nsamples = 6\\n' \ 'value = [3, 3]\\nclass = yes"] ;\n' \ '1 [label="gini = 0.0\\nsamples = 3\\nvalue = [3, 0]\\n' \ 'class = yes"] ;\n' \ '0 -> 1 [labeldistance=2.5, labelangle=45, ' \ 'headlabel="True"] ;\n' \ '2 [label="gini = 0.0\\nsamples = 3\\nvalue = [0, 3]\\n' \ 'class = no"] ;\n' \ '0 -> 2 [labeldistance=2.5, labelangle=-45, ' \ 'headlabel="False"] ;\n' \ '}' assert_equal(contents1, contents2) # Test plot_options out = StringIO() export_graphviz(clf, out_file=out, filled=True, impurity=False, proportion=True, special_characters=True, rounded=True) contents1 = out.getvalue() contents2 = 'digraph Tree {\n' \ 'node [shape=box, style="filled, rounded", color="black", ' \ 'fontname=helvetica] ;\n' \ 'edge [fontname=helvetica] ;\n' \ '0 [label=<X<SUB>0</SUB> &le; 0.0<br/>samples = 100.0%<br/>' \ 'value = [0.5, 0.5]>, fillcolor="#e5813900"] ;\n' \ '1 [label=<samples = 50.0%<br/>value = [1.0, 0.0]>, ' \ 'fillcolor="#e58139ff"] ;\n' \ '0 -> 1 [labeldistance=2.5, labelangle=45, ' \ 'headlabel="True"] ;\n' \ '2 [label=<samples = 50.0%<br/>value = [0.0, 1.0]>, ' \ 'fillcolor="#399de5ff"] ;\n' \ '0 -> 2 [labeldistance=2.5, labelangle=-45, ' \ 'headlabel="False"] ;\n' \ '}' assert_equal(contents1, contents2) # Test max_depth out = StringIO() export_graphviz(clf, out_file=out, max_depth=0, class_names=True) contents1 = out.getvalue() contents2 = 'digraph Tree {\n' \ 'node [shape=box] ;\n' \ '0 [label="X[0] <= 0.0\\ngini = 0.5\\nsamples = 6\\n' \ 'value = [3, 3]\\nclass = y[0]"] ;\n' \ '1 [label="(...)"] ;\n' \ '0 -> 1 ;\n' \ '2 [label="(...)"] ;\n' \ '0 -> 2 ;\n' \ '}' assert_equal(contents1, contents2) # Test max_depth with plot_options out = StringIO() export_graphviz(clf, out_file=out, max_depth=0, filled=True, node_ids=True) contents1 = out.getvalue() contents2 = 'digraph Tree {\n' \ 'node [shape=box, style="filled", color="black"] ;\n' \ '0 [label="node #0\\nX[0] <= 0.0\\ngini = 0.5\\n' \ 'samples = 6\\nvalue = [3, 3]", fillcolor="#e5813900"] ;\n' \ '1 [label="(...)", fillcolor="#C0C0C0"] ;\n' \ '0 -> 1 ;\n' \ '2 [label="(...)", fillcolor="#C0C0C0"] ;\n' \ '0 -> 2 ;\n' \ '}' assert_equal(contents1, contents2) # Test multi-output with weighted samples clf = DecisionTreeClassifier(max_depth=2, min_samples_split=1, criterion="gini", random_state=2) clf = clf.fit(X, y2, sample_weight=w) out = StringIO() export_graphviz(clf, out_file=out, filled=True, impurity=False) contents1 = out.getvalue() contents2 = 'digraph Tree {\n' \ 'node [shape=box, style="filled", color="black"] ;\n' \ '0 [label="X[0] <= 0.0\\nsamples = 6\\n' \ 'value = [[3.0, 1.5, 0.0]\\n' \ '[1.5, 1.5, 1.5]]", fillcolor="#e5813900"] ;\n' \ '1 [label="X[1] <= -1.5\\nsamples = 3\\n' \ 'value = [[3, 0, 0]\\n[1, 1, 1]]", ' \ 'fillcolor="#e5813965"] ;\n' \ '0 -> 1 [labeldistance=2.5, labelangle=45, ' \ 'headlabel="True"] ;\n' \ '2 [label="samples = 1\\nvalue = [[1, 0, 0]\\n' \ '[0, 0, 1]]", fillcolor="#e58139ff"] ;\n' \ '1 -> 2 ;\n' \ '3 [label="samples = 2\\nvalue = [[2, 0, 0]\\n' \ '[1, 1, 0]]", fillcolor="#e581398c"] ;\n' \ '1 -> 3 ;\n' \ '4 [label="X[0] <= 1.5\\nsamples = 3\\n' \ 'value = [[0.0, 1.5, 0.0]\\n[0.5, 0.5, 0.5]]", ' \ 'fillcolor="#e5813965"] ;\n' \ '0 -> 4 [labeldistance=2.5, labelangle=-45, ' \ 'headlabel="False"] ;\n' \ '5 [label="samples = 2\\nvalue = [[0.0, 1.0, 0.0]\\n' \ '[0.5, 0.5, 0.0]]", fillcolor="#e581398c"] ;\n' \ '4 -> 5 ;\n' \ '6 [label="samples = 1\\nvalue = [[0.0, 0.5, 0.0]\\n' \ '[0.0, 0.0, 0.5]]", fillcolor="#e58139ff"] ;\n' \ '4 -> 6 ;\n' \ '}' assert_equal(contents1, contents2) # Test regression output with plot_options clf = DecisionTreeRegressor(max_depth=3, min_samples_split=1, criterion="mse", random_state=2) clf.fit(X, y) out = StringIO() export_graphviz(clf, out_file=out, filled=True, leaves_parallel=True, rotate=True, rounded=True) contents1 = out.getvalue() contents2 = 'digraph Tree {\n' \ 'node [shape=box, style="filled, rounded", color="black", ' \ 'fontname=helvetica] ;\n' \ 'graph [ranksep=equally, splines=polyline] ;\n' \ 'edge [fontname=helvetica] ;\n' \ 'rankdir=LR ;\n' \ '0 [label="X[0] <= 0.0\\nmse = 1.0\\nsamples = 6\\n' \ 'value = 0.0", fillcolor="#e581397f"] ;\n' \ '1 [label="mse = 0.0\\nsamples = 3\\nvalue = -1.0", ' \ 'fillcolor="#e5813900"] ;\n' \ '0 -> 1 [labeldistance=2.5, labelangle=-45, ' \ 'headlabel="True"] ;\n' \ '2 [label="mse = 0.0\\nsamples = 3\\nvalue = 1.0", ' \ 'fillcolor="#e58139ff"] ;\n' \ '0 -> 2 [labeldistance=2.5, labelangle=45, ' \ 'headlabel="False"] ;\n' \ '{rank=same ; 0} ;\n' \ '{rank=same ; 1; 2} ;\n' \ '}' assert_equal(contents1, contents2) def test_graphviz_errors(): # Check for errors of export_graphviz clf = DecisionTreeClassifier(max_depth=3, min_samples_split=1) clf.fit(X, y) # Check feature_names error out = StringIO() assert_raises(IndexError, export_graphviz, clf, out, feature_names=[]) # Check class_names error out = StringIO() assert_raises(IndexError, export_graphviz, clf, out, class_names=[]) def test_friedman_mse_in_graphviz(): clf = DecisionTreeRegressor(criterion="friedman_mse", random_state=0) clf.fit(X, y) dot_data = StringIO() export_graphviz(clf, out_file=dot_data) clf = GradientBoostingClassifier(n_estimators=2, random_state=0) clf.fit(X, y) for estimator in clf.estimators_: export_graphviz(estimator[0], out_file=dot_data) for finding in finditer("\[.*?samples.*?\]", dot_data.getvalue()): assert_in("friedman_mse", finding.group())
bsd-3-clause
roxyboy/scikit-learn
setup.py
142
7364
#! /usr/bin/env python # # Copyright (C) 2007-2009 Cournapeau David <cournape@gmail.com> # 2010 Fabian Pedregosa <fabian.pedregosa@inria.fr> # License: 3-clause BSD descr = """A set of python modules for machine learning and data mining""" import sys import os import shutil from distutils.command.clean import clean as Clean if sys.version_info[0] < 3: import __builtin__ as builtins else: import builtins # This is a bit (!) hackish: we are setting a global variable so that the main # sklearn __init__ can detect if it is being loaded by the setup routine, to # avoid attempting to load components that aren't built yet: # the numpy distutils extensions that are used by scikit-learn to recursively # build the compiled extensions in sub-packages is based on the Python import # machinery. builtins.__SKLEARN_SETUP__ = True DISTNAME = 'scikit-learn' DESCRIPTION = 'A set of python modules for machine learning and data mining' with open('README.rst') as f: LONG_DESCRIPTION = f.read() MAINTAINER = 'Andreas Mueller' MAINTAINER_EMAIL = 'amueller@ais.uni-bonn.de' URL = 'http://scikit-learn.org' LICENSE = 'new BSD' DOWNLOAD_URL = 'http://sourceforge.net/projects/scikit-learn/files/' # We can actually import a restricted version of sklearn that # does not need the compiled code import sklearn VERSION = sklearn.__version__ # Optional setuptools features # We need to import setuptools early, if we want setuptools features, # as it monkey-patches the 'setup' function # For some commands, use setuptools SETUPTOOLS_COMMANDS = set([ 'develop', 'release', 'bdist_egg', 'bdist_rpm', 'bdist_wininst', 'install_egg_info', 'build_sphinx', 'egg_info', 'easy_install', 'upload', 'bdist_wheel', '--single-version-externally-managed', ]) if SETUPTOOLS_COMMANDS.intersection(sys.argv): import setuptools extra_setuptools_args = dict( zip_safe=False, # the package can run out of an .egg file include_package_data=True, ) else: extra_setuptools_args = dict() # Custom clean command to remove build artifacts class CleanCommand(Clean): description = "Remove build artifacts from the source tree" def run(self): Clean.run(self) if os.path.exists('build'): shutil.rmtree('build') for dirpath, dirnames, filenames in os.walk('sklearn'): for filename in filenames: if (filename.endswith('.so') or filename.endswith('.pyd') or filename.endswith('.dll') or filename.endswith('.pyc')): os.unlink(os.path.join(dirpath, filename)) for dirname in dirnames: if dirname == '__pycache__': shutil.rmtree(os.path.join(dirpath, dirname)) cmdclass = {'clean': CleanCommand} # Optional wheelhouse-uploader features # To automate release of binary packages for scikit-learn we need a tool # to download the packages generated by travis and appveyor workers (with # version number matching the current release) and upload them all at once # to PyPI at release time. # The URL of the artifact repositories are configured in the setup.cfg file. WHEELHOUSE_UPLOADER_COMMANDS = set(['fetch_artifacts', 'upload_all']) if WHEELHOUSE_UPLOADER_COMMANDS.intersection(sys.argv): import wheelhouse_uploader.cmd cmdclass.update(vars(wheelhouse_uploader.cmd)) def configuration(parent_package='', top_path=None): if os.path.exists('MANIFEST'): os.remove('MANIFEST') from numpy.distutils.misc_util import Configuration config = Configuration(None, parent_package, top_path) # Avoid non-useful msg: # "Ignoring attempt to set 'name' (from ... " config.set_options(ignore_setup_xxx_py=True, assume_default_configuration=True, delegate_options_to_subpackages=True, quiet=True) config.add_subpackage('sklearn') return config def is_scipy_installed(): try: import scipy except ImportError: return False return True def is_numpy_installed(): try: import numpy except ImportError: return False return True def setup_package(): metadata = dict(name=DISTNAME, maintainer=MAINTAINER, maintainer_email=MAINTAINER_EMAIL, description=DESCRIPTION, license=LICENSE, url=URL, version=VERSION, download_url=DOWNLOAD_URL, long_description=LONG_DESCRIPTION, classifiers=['Intended Audience :: Science/Research', 'Intended Audience :: Developers', 'License :: OSI Approved', 'Programming Language :: C', 'Programming Language :: Python', 'Topic :: Software Development', 'Topic :: Scientific/Engineering', 'Operating System :: Microsoft :: Windows', 'Operating System :: POSIX', 'Operating System :: Unix', 'Operating System :: MacOS', 'Programming Language :: Python :: 2', 'Programming Language :: Python :: 2.6', 'Programming Language :: Python :: 2.7', 'Programming Language :: Python :: 3', 'Programming Language :: Python :: 3.3', 'Programming Language :: Python :: 3.4', ], cmdclass=cmdclass, **extra_setuptools_args) if (len(sys.argv) >= 2 and ('--help' in sys.argv[1:] or sys.argv[1] in ('--help-commands', 'egg_info', '--version', 'clean'))): # For these actions, NumPy is not required. # # They are required to succeed without Numpy for example when # pip is used to install Scikit-learn when Numpy is not yet present in # the system. try: from setuptools import setup except ImportError: from distutils.core import setup metadata['version'] = VERSION else: if is_numpy_installed() is False: raise ImportError("Numerical Python (NumPy) is not installed.\n" "scikit-learn requires NumPy.\n" "Installation instructions are available on scikit-learn website: " "http://scikit-learn.org/stable/install.html\n") if is_scipy_installed() is False: raise ImportError("Scientific Python (SciPy) is not installed.\n" "scikit-learn requires SciPy.\n" "Installation instructions are available on scikit-learn website: " "http://scikit-learn.org/stable/install.html\n") from numpy.distutils.core import setup metadata['configuration'] = configuration setup(**metadata) if __name__ == "__main__": setup_package()
bsd-3-clause
cpausmit/IntelROCCS
Monitor/dynamoToPickle.py
2
2945
#!/usr/bin/env python import os, sys import re, glob, time, json, pprint import cPickle as pickle import multiprocessing as mp from Dataset import Dataset,Request import dynamoDB import requestParse import config #=================================================================================================== # M A I N #=================================================================================================== if __name__=="__main__": sw = Stopwatch() print 'Loading interesting datasets' # load all interesting datasets cursor = dynamoDB.getDbCursor() table_dump = dynamoDB.getDatasetsAndProps(cursor) datasets = {} for name,nfiles,size,dt in table_dump: if not re.match(config.dataset_pattern,name): continue ds = Dataset(name) ds.nFiles = int(nfiles) ds.sizeGB = size/(10.**9) ds.cTime = time.mktime(dt.timetuple()) datasets[name] = ds sw.report() print 'Considering %i relevant datasets'%(len(datasets)) print 'Importing transfer history' # import phedex history pool = mp.Pool(processes=10) all_transfers = pool.map(requestParse.parseTransfer, glob.glob(config.requests_dir+'/requests_transfer_*.json')) for reqs in all_transfers: for d in reqs: if not d in datasets: continue for t,s in reqs[d]: if not re.match(config.site_pattern,s): continue datasets[d].addTransfer(s,t) sw.report() print 'Importing deletion history' all_deletions = pool.map(requestParse.parseDeletion, glob.glob(config.requests_dir+'/requests_delete_*.json')) for reqs in all_deletions: for d in reqs: if not d in datasets: continue for t,s in reqs[d]: if not re.match(config.site_pattern,s): continue datasets[d].addDeletion(s,t) sw.report() print 'Sorting history' # organize the history for name,d in datasets.iteritems(): d.sortRequests() sw.report() print 'Importing CRAB+xrootd accesses' # import access history all_accesses = dynamoDB.getDatasetsAccesses(cursor) for name,node,dt,n in all_accesses: if name not in datasets: continue if not re.match(config.site_pattern,node): continue timestamp = time.mktime(dt.timetuple()) datasets[name].addAccesses(node,n,timestamp) sw.report() print 'Exporting to pickle' i=0 for k in datasets: if i == 10: break print k print datasets[k] i+= 1 pickleJar = open('monitorCache_'+ config.name+'.pkl',"wb") pickle.dump(datasets,pickleJar,2) # put the pickle in a jar pickleJar.close() # close the jar sw.report() sys.exit(0)
mit
cpausmit/IntelROCCS
Detox/python/phedexApi.py
3
17588
#!/usr/bin/python #--------------------------------------------------------------------------------------------------- # # This script provide an API to PhEDEX communications one can delete or subscribe datasets using # methods defined in this class. # # This script uses auxilary tool for logging purposes in case the request runs into the error. # #--------------------------------------------------------------------------------------------------- __author__ = '*Bjorn Barrefors, $Maxim Goncharov, $Christoph Paus' __organization__ = '*Holland Computing Center - University of Nebraska-Lincoln, $MIT' __email__ = 'bbarrefo@cse.unl.edu, maxi@mit.edu, paus@mit.edu' import sys import os import re import urllib import urllib2 import httplib import time import datetime try: import json except ImportError: import simplejson as json from cmsDataLogger import cmsDataLogger #################################################################################################### # # P h E D E x A P I # #################################################################################################### class phedexApi: """ _phedexApi_ Interface to submit queries to the PhEDEx API For specifications of calls see https://cmsweb.cern.ch/phedex/datasvc/doc Class variables: phedexBase -- Base URL to the PhEDEx web API logger -- Used to print log and error messages to log file """ # Useful variables # phedexBase = "https://cmsweb.cern.ch/phedex/datasvc/" # phedexInstance = "prod" or "dev # dataType = "json" or "xml" # site = "T2_US_Nebraska" # dataset = "/BTau/GowdyTest10-Run2010Av3/RAW" # group = 'local' or 'AnalysisOps' def __init__(self, logPath=''): """ __init__ Set up class constants """ statusDirectory = os.environ['DETOX_DB'] + '/' + os.environ['DETOX_STATUS'] self.logger = cmsDataLogger(statusDirectory+'/') self.phedexBase = "https://cmsweb.cern.ch/phedex/datasvc/" ############################################################################ # # # P h E D E x C A L L # # # ############################################################################ def phedexCall(self, url, values): """ _phedexCall_ Make http post call to PhEDEx API. Function only gaurantees that something is returned, the caller need to check the response for correctness. Keyword arguments: url -- URL to make API call values -- Arguments to pass to the call Return values: 1 -- Status, 0 = everything went well, 1 = something went wrong 2 -- IF status == 0 : HTTP response ELSE : Error message """ name = "phedexCall" data = urllib.urlencode(values) opener = urllib2.build_opener(HTTPSGridAuthHandler()) request = urllib2.Request(url, data) try: response = opener.open(request) except urllib2.HTTPError, e: self.logger.error(name, e.read()) self.logger.error(name, "URL: %s" % (str(url),)) self.logger.error(name, "VALUES: %s" % (str(values),)) #print (name, "VALUES: %s" % (str(values),)) return 1, " ERROR - urllib2.HTTPError" except urllib2.URLError, e: self.logger.error(name, e.args) self.logger.error(name, "URL: %s" % (str(url),)) self.logger.error(name, "VALUES: %s" % (str(values),)) #print (name, "VALUES: %s" % (str(values))) return 1, " ERROR - urllib2.URLError" return 0, response ############################################################################ # # # D A T A # # # ############################################################################ def data(self, dataset='', block='', fileName='', level='block', createSince='', format='json', instance='prod'): """ _data_ PhEDEx data call At least one of the arguments dataset, block, file have to be passed No checking is made for xml data Even if JSON data is returned no gaurantees are made for the structure of it Keyword arguments: dataset -- Name of dataset to look up block -- Name of block to look up file -- Name of file to look up block -- Only return data for this block fileName -- Data for file fileName returned level -- Which granularity of dataset information to show createSince -- Files/blocks/datasets created since this date/time format -- Which format to return data as, XML or JSON instance -- Which instance of PhEDEx to query, dev or prod Return values: check -- 0 if all went well, 1 if error occured data -- json structure if json format, xml structure if xml format """ name = "data" if not (dataset or block or fileName): self.logger.error(name, "Need to pass at least one of dataset/block/fileName") return 1, "Error" values = { 'dataset' : dataset, 'block' : block, 'file' : fileName, 'level' : level, 'create_since' : createSince } dataURL = urllib.basejoin(self.phedexBase, "%s/%s/data" % (format, instance)) check, response = self.phedexCall(dataURL, values) if check: # An error occurred self.logger.error(name, "Data call failed") return 1, "Error" if format == "json": try: data = json.load(response) except ValueError, e: # This usually means that PhEDEx didn't like the URL self.logger.error(name, "In call to url %s : %s" % (dataURL, str(e))) return 1, " ERROR - ValueError" if not data: self.logger.error(name, "No json data available") return 1, " ERROR - no data" else: data = response.read() return 0, data ############################################################################ # # # P A R S E # # # ############################################################################ def parse(self, data, xml): """ _parse_ Take data output from PhEDEx and parse it into xml syntax corresponding to subscribe and delete calls. """ for k, v in data.iteritems(): k = k.replace("_", "-") if type(v) is list: xml = "%s>" % (xml,) for v1 in v: xml = "%s<%s" % (xml, k) xml = self.parse(v1, xml) if (k == "file"): xml = "%s/>" % (xml,) else: xml = "%s</%s>" % (xml, k) else: if k == "lfn": k = "name" elif k == "size": k = "bytes" if (k == "name" or k == "is-open" or k == "is-transient" or \ k == "bytes" or k== "checksum"): xml = '%s %s="%s"' % (xml, k, v) return xml ############################################################################ # # # X M L D A T A # # # ############################################################################ def xmlData(self, datasets=[], instance='prod', level='file'): """ _xmlData_ Get json data from PhEDEx for all datasets and convert it to a xml structure complient with the PhEDEx delete/subscribe call. Keyword arguments: datasets -- List of dataset names instance -- The instance on which the datasets resides, prod/dev Return values: error -- 1 if an error occurred, 0 if everything went as expected xml -- The converted data now represented as an xml structure """ name = "xmlData" # @CHANGED: Function now takes a list of datasets instead of only one if not datasets: self.logger.error(name, "Need to pass at least one of dataset") return 1, "Error" xml = '<data version="2">' xml = '%s<%s name="https://cmsweb.cern.ch/dbs/%s/global/DBSReader">'\ % (xml, 'dbs', instance) for dataset in datasets: if '#' in dataset: parts = dataset.split('#') check, response = self.data(block=dataset, level=level, instance=instance) else: check, response = self.data(dataset=dataset, level=level, instance=instance) if check: return 1, "Error" data = response.get('phedex').get('dbs') if not data: return 1, "Error" xml = "%s<%s" % (xml, 'dataset') data = data[0].get('dataset') xml = self.parse(data[0], xml) xml = "%s</%s>" % (xml, 'dataset') xml = "%s</%s>" % (xml, 'dbs') xml_data = "%s</data>" % (xml,) #print xml_data return 0, xml_data ############################################################################ # # # S U B S C R I B E # # # ############################################################################ def subscribe(self, node='', data='', level='dataset', priority='low', move='n', static='n', custodial='n', group='local', timeStart='', requestOnly='n', noMail='n', comments='', format='json', instance='prod'): """ _subscribe_ Set up subscription call to PhEDEx API. """ name = "subscribe" if not (node and data): self.logger.error(name, "Need to pass both node and data") return 1, "Error" values = { 'node' : node, 'data' : data, 'level' : level, 'priority' : priority, 'move' : move, 'static' : static, 'custodial' : custodial, 'group' : group, 'time_start' : timeStart, 'request_only' : requestOnly, 'no_mail' : noMail, 'comments' : comments } subscriptionURL = urllib.basejoin(self.phedexBase, "%s/%s/subscribe" % (format, instance)) check, response = self.phedexCall(subscriptionURL, values) if check: # An error occurred self.logger.error(name, "Subscription call failed") return 1, "Error" return 0, response ############################################################################ # # # D E L E T E # # # ############################################################################ def delete(self, node='', data='', level='dataset', rmSubscriptions='y', comments='', format='json', instance='prod'): name = "delete" if not (node and data): self.logger.error(name, "Need to pass both node and data") return 1, "Error" values = { 'node' : node, 'data' : data, 'level' : level, 'rm_subscriptions' : rmSubscriptions, 'comments' : comments } deleteURL = urllib.basejoin(self.phedexBase, "%s/%s/delete" % (format, instance)) check, response = self.phedexCall(deleteURL, values) if check: self.logger.error(name, "Delete call failed") return 1, "ERROR - self.phedexCall with response: " + response return 0, response ############################################################################ # # # U P D A T E R E Q U E S T # # # ############################################################################ def updateRequest(self, decision, request, node, comments='',format='json', instance='prod'): name = "update" values = {'decision':decision, 'request':request, 'node':node, 'comments':comments} url = urllib.basejoin(self.phedexBase, "%s/%s/updaterequest" % (format, instance)) check, response = self.phedexCall(url, values) if check: self.logger.error(name, "Update call failed") return 1, "ERROR - self.phedexCall with response: " + response return 0, response def changeGroup(self, node, dataset, group, comments='',format='json', instance='prod', level='dataset'): name = "changegroup" values = {'node':node, 'dataset':dataset, 'group':group} url = urllib.basejoin(self.phedexBase, "%s/%s/updatesubscription" % (format, instance)) check, response = self.phedexCall(url, values) if check: self.logger.error(name, "Change group call failed") print response print check return 1, "ERROR - self.phedexCall with response: " + response return 0, response def getDelRequests(self,node, request_since=1428200998, complete='y',format='json', instance = 'prod'): name = 'deletions' values = {'node':node, 'request_since':request_since, 'complete':complete} url = urllib.basejoin(self.phedexBase, "%s/%s/deletions" % (format, instance)) check, response = self.phedexCall(url, values) if check: self.logger.error(name, "Get deletions for site") print response print check return 1, "ERROR - self.phedexCall with response: " + response return 0, response #################################################################################################### # # H T T P S G R I D A U T H H A N D L E R # #################################################################################################### class HTTPSGridAuthHandler(urllib2.HTTPSHandler): """ _HTTPSGridAuthHandler_ Get proxy to acces PhEDEx API Needed for subscribe and delete calls Class variables: key -- User key to CERN with access to PhEDEx cert -- User certificate connected to key """ def __init__(self): urllib2.HTTPSHandler.__init__(self) self.key = self.getProxy() self.cert = self.key def https_open(self, req): return self.do_open(self.getConnection, req) def getProxy(self): proxy = os.environ['DETOX_X509UP'] return proxy def getConnection(self, host, timeout=300): return httplib.HTTPSConnection(host, key_file=self.key, cert_file=self.cert) #################################################################################################### # # M A I N # #################################################################################################### if __name__ == '__main__': """ __main__ For testing purpose only """ # Example for deletion #phedexApi = phedexApi(logPath='./') #check, data = phedexApi.xmlData(datasets=['/MET/Run2012A-22Jan2013-v1/AOD'], instance='prod') #if check: # sys.exit(1) #print data # #check, response = \ # phedexApi.delete(node='T2_US_MIT', data=data, instance='prod', # comments='Just a test by Christoph Paus for Maxim Goncharov.') #if check: # print "This is the response from phedexApi.delete: " + response # sys.exit(1) # #print response.read() # Example for subscription #phedexApi = phedexApi(logPath='./') #check, data = phedexApi.xmlData( # datasets=['/Muplus_Pt1_PositiveEta-gun/Muon2023Upg14-DES23_62_V1-v1/GEN-SIM'], instance='prod') #if check: # sys.exit(1) #print data # #check, response = \ # phedexApi.subscribe(node='T2_US_MIT', data=data, instance='prod', group='AnalysisOps', # comments='Just a test by Christoph Paus for Maxim Goncharov.') #if check: # print "This is the response from phedexApi.delete: " + response # sys.exit(1) # #print response.read() #sys.exit(0)
mit
ephes/scikit-learn
sklearn/learning_curve.py
109
13467
"""Utilities to evaluate models with respect to a variable """ # Author: Alexander Fabisch <afabisch@informatik.uni-bremen.de> # # License: BSD 3 clause import warnings import numpy as np from .base import is_classifier, clone from .cross_validation import check_cv from .externals.joblib import Parallel, delayed from .cross_validation import _safe_split, _score, _fit_and_score from .metrics.scorer import check_scoring from .utils import indexable from .utils.fixes import astype __all__ = ['learning_curve', 'validation_curve'] def learning_curve(estimator, X, y, train_sizes=np.linspace(0.1, 1.0, 5), cv=None, scoring=None, exploit_incremental_learning=False, n_jobs=1, pre_dispatch="all", verbose=0): """Learning curve. Determines cross-validated training and test scores for different training set sizes. A cross-validation generator splits the whole dataset k times in training and test data. Subsets of the training set with varying sizes will be used to train the estimator and a score for each training subset size and the test set will be computed. Afterwards, the scores will be averaged over all k runs for each training subset size. Read more in the :ref:`User Guide <learning_curves>`. Parameters ---------- estimator : object type that implements the "fit" and "predict" methods An object of that type which is cloned for each validation. X : array-like, shape (n_samples, n_features) Training vector, where n_samples is the number of samples and n_features is the number of features. y : array-like, shape (n_samples) or (n_samples, n_features), optional Target relative to X for classification or regression; None for unsupervised learning. train_sizes : array-like, shape (n_ticks,), dtype float or int Relative or absolute numbers of training examples that will be used to generate the learning curve. If the dtype is float, it is regarded as a fraction of the maximum size of the training set (that is determined by the selected validation method), i.e. it has to be within (0, 1]. Otherwise it is interpreted as absolute sizes of the training sets. Note that for classification the number of samples usually have to be big enough to contain at least one sample from each class. (default: np.linspace(0.1, 1.0, 5)) cv : integer, cross-validation generator, optional If an integer is passed, it is the number of folds (defaults to 3). Specific cross-validation objects can be passed, see sklearn.cross_validation module for the list of possible objects scoring : string, callable or None, optional, default: None A string (see model evaluation documentation) or a scorer callable object / function with signature ``scorer(estimator, X, y)``. exploit_incremental_learning : boolean, optional, default: False If the estimator supports incremental learning, this will be used to speed up fitting for different training set sizes. n_jobs : integer, optional Number of jobs to run in parallel (default 1). pre_dispatch : integer or string, optional Number of predispatched jobs for parallel execution (default is all). The option can reduce the allocated memory. The string can be an expression like '2*n_jobs'. verbose : integer, optional Controls the verbosity: the higher, the more messages. Returns ------- train_sizes_abs : array, shape = (n_unique_ticks,), dtype int Numbers of training examples that has been used to generate the learning curve. Note that the number of ticks might be less than n_ticks because duplicate entries will be removed. train_scores : array, shape (n_ticks, n_cv_folds) Scores on training sets. test_scores : array, shape (n_ticks, n_cv_folds) Scores on test set. Notes ----- See :ref:`examples/model_selection/plot_learning_curve.py <example_model_selection_plot_learning_curve.py>` """ if exploit_incremental_learning and not hasattr(estimator, "partial_fit"): raise ValueError("An estimator must support the partial_fit interface " "to exploit incremental learning") X, y = indexable(X, y) # Make a list since we will be iterating multiple times over the folds cv = list(check_cv(cv, X, y, classifier=is_classifier(estimator))) scorer = check_scoring(estimator, scoring=scoring) # HACK as long as boolean indices are allowed in cv generators if cv[0][0].dtype == bool: new_cv = [] for i in range(len(cv)): new_cv.append((np.nonzero(cv[i][0])[0], np.nonzero(cv[i][1])[0])) cv = new_cv n_max_training_samples = len(cv[0][0]) # Because the lengths of folds can be significantly different, it is # not guaranteed that we use all of the available training data when we # use the first 'n_max_training_samples' samples. train_sizes_abs = _translate_train_sizes(train_sizes, n_max_training_samples) n_unique_ticks = train_sizes_abs.shape[0] if verbose > 0: print("[learning_curve] Training set sizes: " + str(train_sizes_abs)) parallel = Parallel(n_jobs=n_jobs, pre_dispatch=pre_dispatch, verbose=verbose) if exploit_incremental_learning: classes = np.unique(y) if is_classifier(estimator) else None out = parallel(delayed(_incremental_fit_estimator)( clone(estimator), X, y, classes, train, test, train_sizes_abs, scorer, verbose) for train, test in cv) else: out = parallel(delayed(_fit_and_score)( clone(estimator), X, y, scorer, train[:n_train_samples], test, verbose, parameters=None, fit_params=None, return_train_score=True) for train, test in cv for n_train_samples in train_sizes_abs) out = np.array(out)[:, :2] n_cv_folds = out.shape[0] // n_unique_ticks out = out.reshape(n_cv_folds, n_unique_ticks, 2) out = np.asarray(out).transpose((2, 1, 0)) return train_sizes_abs, out[0], out[1] def _translate_train_sizes(train_sizes, n_max_training_samples): """Determine absolute sizes of training subsets and validate 'train_sizes'. Examples: _translate_train_sizes([0.5, 1.0], 10) -> [5, 10] _translate_train_sizes([5, 10], 10) -> [5, 10] Parameters ---------- train_sizes : array-like, shape (n_ticks,), dtype float or int Numbers of training examples that will be used to generate the learning curve. If the dtype is float, it is regarded as a fraction of 'n_max_training_samples', i.e. it has to be within (0, 1]. n_max_training_samples : int Maximum number of training samples (upper bound of 'train_sizes'). Returns ------- train_sizes_abs : array, shape (n_unique_ticks,), dtype int Numbers of training examples that will be used to generate the learning curve. Note that the number of ticks might be less than n_ticks because duplicate entries will be removed. """ train_sizes_abs = np.asarray(train_sizes) n_ticks = train_sizes_abs.shape[0] n_min_required_samples = np.min(train_sizes_abs) n_max_required_samples = np.max(train_sizes_abs) if np.issubdtype(train_sizes_abs.dtype, np.float): if n_min_required_samples <= 0.0 or n_max_required_samples > 1.0: raise ValueError("train_sizes has been interpreted as fractions " "of the maximum number of training samples and " "must be within (0, 1], but is within [%f, %f]." % (n_min_required_samples, n_max_required_samples)) train_sizes_abs = astype(train_sizes_abs * n_max_training_samples, dtype=np.int, copy=False) train_sizes_abs = np.clip(train_sizes_abs, 1, n_max_training_samples) else: if (n_min_required_samples <= 0 or n_max_required_samples > n_max_training_samples): raise ValueError("train_sizes has been interpreted as absolute " "numbers of training samples and must be within " "(0, %d], but is within [%d, %d]." % (n_max_training_samples, n_min_required_samples, n_max_required_samples)) train_sizes_abs = np.unique(train_sizes_abs) if n_ticks > train_sizes_abs.shape[0]: warnings.warn("Removed duplicate entries from 'train_sizes'. Number " "of ticks will be less than than the size of " "'train_sizes' %d instead of %d)." % (train_sizes_abs.shape[0], n_ticks), RuntimeWarning) return train_sizes_abs def _incremental_fit_estimator(estimator, X, y, classes, train, test, train_sizes, scorer, verbose): """Train estimator on training subsets incrementally and compute scores.""" train_scores, test_scores = [], [] partitions = zip(train_sizes, np.split(train, train_sizes)[:-1]) for n_train_samples, partial_train in partitions: train_subset = train[:n_train_samples] X_train, y_train = _safe_split(estimator, X, y, train_subset) X_partial_train, y_partial_train = _safe_split(estimator, X, y, partial_train) X_test, y_test = _safe_split(estimator, X, y, test, train_subset) if y_partial_train is None: estimator.partial_fit(X_partial_train, classes=classes) else: estimator.partial_fit(X_partial_train, y_partial_train, classes=classes) train_scores.append(_score(estimator, X_train, y_train, scorer)) test_scores.append(_score(estimator, X_test, y_test, scorer)) return np.array((train_scores, test_scores)).T def validation_curve(estimator, X, y, param_name, param_range, cv=None, scoring=None, n_jobs=1, pre_dispatch="all", verbose=0): """Validation curve. Determine training and test scores for varying parameter values. Compute scores for an estimator with different values of a specified parameter. This is similar to grid search with one parameter. However, this will also compute training scores and is merely a utility for plotting the results. Read more in the :ref:`User Guide <validation_curve>`. Parameters ---------- estimator : object type that implements the "fit" and "predict" methods An object of that type which is cloned for each validation. X : array-like, shape (n_samples, n_features) Training vector, where n_samples is the number of samples and n_features is the number of features. y : array-like, shape (n_samples) or (n_samples, n_features), optional Target relative to X for classification or regression; None for unsupervised learning. param_name : string Name of the parameter that will be varied. param_range : array-like, shape (n_values,) The values of the parameter that will be evaluated. cv : integer, cross-validation generator, optional If an integer is passed, it is the number of folds (defaults to 3). Specific cross-validation objects can be passed, see sklearn.cross_validation module for the list of possible objects scoring : string, callable or None, optional, default: None A string (see model evaluation documentation) or a scorer callable object / function with signature ``scorer(estimator, X, y)``. n_jobs : integer, optional Number of jobs to run in parallel (default 1). pre_dispatch : integer or string, optional Number of predispatched jobs for parallel execution (default is all). The option can reduce the allocated memory. The string can be an expression like '2*n_jobs'. verbose : integer, optional Controls the verbosity: the higher, the more messages. Returns ------- train_scores : array, shape (n_ticks, n_cv_folds) Scores on training sets. test_scores : array, shape (n_ticks, n_cv_folds) Scores on test set. Notes ----- See :ref:`examples/model_selection/plot_validation_curve.py <example_model_selection_plot_validation_curve.py>` """ X, y = indexable(X, y) cv = check_cv(cv, X, y, classifier=is_classifier(estimator)) scorer = check_scoring(estimator, scoring=scoring) parallel = Parallel(n_jobs=n_jobs, pre_dispatch=pre_dispatch, verbose=verbose) out = parallel(delayed(_fit_and_score)( estimator, X, y, scorer, train, test, verbose, parameters={param_name: v}, fit_params=None, return_train_score=True) for train, test in cv for v in param_range) out = np.asarray(out)[:, :2] n_params = len(param_range) n_cv_folds = out.shape[0] // n_params out = out.reshape(n_cv_folds, n_params, 2).transpose((2, 1, 0)) return out[0], out[1]
bsd-3-clause
vijayaganesh/Kanjoos-HackGT
django-webapp/kanjoos/myapp/src/runner/dataset.py
1
3274
import cv2 import os import glob from sklearn.utils import shuffle import numpy as np def load_train(train_path, image_size, classes): images = [] labels = [] img_names = [] cls = [] print('Going to read training images') for fields in classes: index = classes.index(fields) print('Now going to read {} files (Index: {})'.format(fields, index)) path = os.path.join(train_path, fields, '*g') files = glob.glob(path) for fl in files: image = cv2.imread(fl) image = cv2.resize(image, (image_size, image_size),0,0, cv2.INTER_LINEAR) image = image.astype(np.float32) image = np.multiply(image, 1.0 / 255.0) images.append(image) label = np.zeros(len(classes)) label[index] = 1.0 labels.append(label) flbase = os.path.basename(fl) img_names.append(flbase) cls.append(fields) images = np.array(images) labels = np.array(labels) img_names = np.array(img_names) cls = np.array(cls) return images, labels, img_names, cls class DataSet(object): def __init__(self, images, labels, img_names, cls): self._num_examples = images.shape[0] self._images = images self._labels = labels self._img_names = img_names self._cls = cls self._epochs_done = 0 self._index_in_epoch = 0 @property def images(self): return self._images @property def labels(self): return self._labels @property def img_names(self): return self._img_names @property def cls(self): return self._cls @property def num_examples(self): return self._num_examples @property def epochs_done(self): return self._epochs_done def next_batch(self, batch_size): """Return the next `batch_size` examples from this data set.""" start = self._index_in_epoch self._index_in_epoch += batch_size if self._index_in_epoch > self._num_examples: # After each epoch we update this self._epochs_done += 1 start = 0 self._index_in_epoch = batch_size assert batch_size <= self._num_examples end = self._index_in_epoch return self._images[start:end], self._labels[start:end], self._img_names[start:end], self._cls[start:end] def read_train_sets(train_path, image_size, classes, validation_size): class DataSets(object): pass data_sets = DataSets() images, labels, img_names, cls = load_train(train_path, image_size, classes) images, labels, img_names, cls = shuffle(images, labels, img_names, cls) if isinstance(validation_size, float): validation_size = int(validation_size * images.shape[0]) validation_images = images[:validation_size] validation_labels = labels[:validation_size] validation_img_names = img_names[:validation_size] validation_cls = cls[:validation_size] train_images = images[validation_size:] train_labels = labels[validation_size:] train_img_names = img_names[validation_size:] train_cls = cls[validation_size:] data_sets.train = DataSet(train_images, train_labels, train_img_names, train_cls) data_sets.valid = DataSet(validation_images, validation_labels, validation_img_names, validation_cls) return data_sets
mit
bromjiri/Presto
predictor/predictor_sklearn.py
1
10590
import settings import pandas as pd import numpy as np import os from datetime import timedelta import predictor.predictor_statistic as stat import random import pickle from sklearn.feature_extraction import DictVectorizer from sklearn.model_selection import train_test_split from sklearn.linear_model import LogisticRegression from sklearn import metrics class Stock: def __init__(self, subject): input_file = settings.PREDICTOR_STOCK + "/" + subject + ".csv" self.stock_df = pd.read_csv(input_file, sep=',', index_col='Date') def create_dict(self, from_date, to_date): self.stock_ser = self.stock_df['Diff'].loc[from_date:to_date] # binning self.stock_ser = self.stock_ser.apply(binning_none) self.stock_dict = self.stock_ser.dropna().astype(int).to_dict() def get_dict(self): return self.stock_dict def get_stock_dates(self): return self.stock_ser.index.values class Sent: def __init__(self, subject, source): input_file = settings.PREDICTOR_SENTIMENT + "/" + source + "/" + source + "-sent-" + subject + ".csv" self.sent_df = pd.read_csv(input_file, sep=',', index_col='Date') def get_weekend(self, col_name, stock_dates): weekend_df = np.round(self.sent_df, 2) aggreg = 0 days = 1 for idx, row in weekend_df.iterrows(): value = row[col_name] date = pd.to_datetime(idx) date_plus = date + timedelta(days=1) if str(date_plus.date()) not in stock_dates: # print("weekend") value += aggreg aggreg = value days += 1 else: total = value + aggreg mean = total / days aggreg = 0 days = 1 weekend_df.set_value(idx, col_name, mean) # print(date.date(), row[col_name], value) return np.round(weekend_df[col_name].diff().loc[stock_dates], 2) def create_dict(self, precision, method, from_date, to_date, stock_dates, binning): sentiment_col = "Sent" + precision sent_ser = self.sent_df[sentiment_col] if method == "Natural": sent_ser = sent_ser.diff().loc[from_date:to_date] elif method == "Friday": sent_ser = sent_ser.loc[stock_dates].diff() elif method == "Sunday": sent_ser = sent_ser.diff().loc[stock_dates] elif method == "Weekend": sent_ser = self.get_weekend(sentiment_col, stock_dates) # binning std_dev1 = sent_ser.std() / 4 std_dev2 = sent_ser.std() if binning == 'none': sent_ser_new = sent_ser.apply(binning_none) elif binning == 'low': sent_ser_new = sent_ser.apply(binning_low, args=(std_dev1,)) else: sent_ser_new = sent_ser.apply(binning_high, args=(std_dev1, std_dev2,)) # print(pd.concat([sent_ser, sent_ser_new], axis=1)) self.sent_dict = sent_ser_new.dropna().astype(int).to_dict() self.key_list = sorted(self.sent_dict.keys()) def get_dict(self): return self.sent_dict def get_features(self, key): index = self.key_list.index(key) features = dict() features['d1'] = self.sent_dict[self.key_list[index-3]] features['d2'] = self.sent_dict[self.key_list[index-2]] features['d3'] = self.sent_dict[self.key_list[index-1]] return features def binning_none(row): if row > 0: return 4 elif row < 0: return 0 else: return row def binning_low(row, std_dev1): if row > std_dev1: return 4 elif row < std_dev1 and row > -std_dev1: return 2 elif row < -std_dev1: return 0 else: return row def binning_high(row, std_dev1, std_dev2): if row > std_dev2: return 4 elif row < std_dev2 and row > std_dev1: return 3 elif row < std_dev1 and row > -std_dev1: return 2 elif row < -std_dev1 and row > -std_dev2: return 1 elif row < -std_dev2: return 0 else: return row def run_one(source, subject, precision, method, from_date, to_date, binning, filename_nltk, filename_skl, filename_lr): # stock dataframe stock = Stock(subject) stock.create_dict(from_date, to_date) stock_dict = stock.get_dict() # print(sorted(stock_dict.items())) indexes = ["djia", "snp", "nasdaq"] # if subject in indexes: # subject = "the" # sentiment dataframe sent = Sent(subject, source) sent.create_dict(precision, method, from_date, to_date, stock.get_stock_dates(), binning) # print(sorted(sent.get_dict().items())) # features features_list = list() for key in sorted(stock_dict)[3:]: features = sent.get_features(key) features_list.append([features, stock_dict[key]]) # print([key, sorted(features.items()), stock_dict[key]]) features_list_pos = list() features_list_neg = list() for feature in features_list: if feature[1] == 0: features_list_neg.append(feature) else: features_list_pos.append(feature) statistic = stat.Statistic(source, subject, precision, method, binning) # print(len(features_list), len(features_list_pos), len(features_list_neg)) max_half = min(len(features_list_pos), len(features_list_neg)) train_border = int(max_half * 4 / 5) # print(train_border, max_half) # exit() if pickle_switch: cycles = 1 else: cycles = 50 for x in range(0, cycles): random.shuffle(features_list_pos) random.shuffle(features_list_neg) trainfeats = features_list_pos[:train_border] + features_list_neg[:train_border] testfeats = features_list_pos[train_border:max_half] + features_list_neg[train_border:max_half] # print(len(trainfeats), len(testfeats)) # print(trainfeats) X = list() y = list() feats = trainfeats + testfeats for feat in feats: X.append(feat[0]) y.append(feat[1]) # vectorize input v = DictVectorizer(sparse=False) X_train, X_test, y_train, y_test = train_test_split(X, y) X_train_vect = v.fit_transform(X_train) X_test_vect = v.transform(X_test) # instantiate model logreg = LogisticRegression() # fit model logreg.fit(X_train_vect, y_train) # pickle if pickle_switch: os.makedirs('pickled', exist_ok=True) logreg_f = open("pickled/tesla_logreg.pickle", "wb") pickle.dump(logreg, logreg_f) logreg_f.close() vector_f = open("pickled/tesla_vector.pickle", "wb") pickle.dump(v, vector_f) vector_f.close() # make class predictions for the testing set y_pred_class = logreg.predict(X_test_vect) ################# # get the outputs lr_output = dict() lr_output['accuracy'] = metrics.accuracy_score(y_test, y_pred_class) * 100 # print(lr_output['accuracy']) confusion = metrics.confusion_matrix(y_test, y_pred_class) # print('True:', y_test) # print('False:', y_pred_class) # print(confusion) TP = confusion[1, 1] TN = confusion[0, 0] FP = confusion[0, 1] FN = confusion[1, 0] # exit() lr_output['pos_prec'] = TP / float(TP + FP) * 100 lr_output['neg_prec'] = TN / float(TN + FN) * 100 lr_output['pos_rec'] = TP / float(TP + FN) * 100 lr_output['neg_rec'] = TN / float(TN + FP) * 100 # nlt_output, skl_output = cls.train(trainfeats, testfeats, nlt=nltk_run, skl=sklearn_run) # print(nlt_output['most1']) # if nltk_run: # statistic.add_nltk(nlt_output) # if sklearn_run: # statistic.add_skl(skl_output) if lr_run: statistic.add_lr(lr_output) if nltk_run: statistic.mean_nltk(cycles) statistic.print_nltk() # statistic.write_nltk(filename_nltk) if sklearn_run: statistic.mean_skl(cycles) statistic.print_skl() statistic.print_stddev() # statistic.write_skl(filename_skl) if lr_run: statistic.mean_lr(cycles) statistic.print_lr() # statistic.write_lr(filename_lr) nltk_run = False sklearn_run = False lr_run = True pickle_switch = False from_date = '2016-11-01' to_date = '2017-08-31' source = "stwits" # binnings = ['none', 'low', 'high'] binnings = ['none'] # subjects = ["coca-cola", "mcdonalds", "microsoft", "netflix", "nike", "samsung", "tesla", "djia", "snp", "nasdaq"] subjects = ["tesla"] # precisions = ["0.6", "0.8", "1.0"] precisions = ["0.6"] # methods = ["Friday", "Natural", "Weekend"] methods = ["Friday"] for subject in subjects: folder = settings.PREDICTOR_PREDICTION + '/' + source + '/' + subject + '/' os.makedirs(folder, exist_ok=True) filename_nltk = folder + source + '-prediction-' + subject + "-nltk.csv" filename_skl = folder + source + '-prediction-' + subject + "-skl.csv" filename_lr = folder + source + '-prediction-' + subject + "-lr.csv" # if nltk_run: # open(filename_nltk, 'w').close() # # if sklearn_run: # open(filename_skl, 'w').close() if lr_run: open(filename_lr, 'w').close() for method in methods: # if nltk_run: # f = open(filename_nltk, 'a') # f.write(source + ", " + subject + ", " + method + ", NLTK\n") # f.write("precision, binning, accuracy, pos_prec, neg_prec, pos_rec, neg_rec, d1, d2, d3\n") # f.close() # # if sklearn_run: # f = open(filename_skl, 'a') # f.write(source + ", " + subject + ", " + method + ", SKL\n") # f.write("precision, binning, mnb, bnb, lr, lsvc, nsvc, voted\n") # f.close() # if lr_run: # f = open(filename_lr, 'a') # f.write(source + ", " + subject + ", " + method + ", LR\n") # f.write("precision, binning, accuracy, pos_prec, neg_prec, pos_rec, neg_rec\n") # f.close() for precision in precisions: for binning in binnings: # print(source, subject, precision, method) run_one(source, subject, precision, method, from_date, to_date, binning, filename_nltk, filename_skl, filename_lr)
mit
lensacom/sparkit-learn
splearn/base.py
2
1701
# -*- coding: utf-8 -*- from sklearn.base import BaseEstimator, ClassifierMixin, TransformerMixin from sklearn.metrics import accuracy_score class SparkBroadcasterMixin(object): # TODO: consider caching in case of streaming def broadcast(self, func, context): bcvars = {name: context.broadcast(getattr(self, name)) for name in self.__transient__} def func_wrapper(*args, **kwargs): for k, v in bcvars.items(): setattr(func.__self__, k, v.value) return func(*args, **kwargs) return func_wrapper class SparkBaseEstimator(BaseEstimator): pass class SparkClassifierMixin(ClassifierMixin): """Mixin class for all classifiers in sparkit-learn.""" def score(self, Z): X, y, w = Z[:, 'X'], Z[:, 'y'], None if 'w' in Z.columns: w = Z[:, 'w'] return accuracy_score(y.toarray(), self.predict(X).toarray(), sample_weight=w) class SparkTransformerMixin(TransformerMixin): """Mixin class for all transformers in sparkit-learn.""" def fit_transform(self, Z, **fit_params): """Fit to data, then transform it. Fits transformer to Z with optional parameters fit_params and returns a transformed version of Z. Parameters ---------- Z : ArrayRDD or DictRDD Training set. Returns ------- Z_new : ArrayRDD or DictRDD Transformed array. """ # non-optimized default implementation; override when a better # method is possible return self.fit(Z, **fit_params).transform(Z)
apache-2.0
3quarterstack/simple_blog
djangoappengine/db/base.py
18
12838
import datetime import decimal import logging import os import shutil from django.db.utils import DatabaseError from google.appengine.api.datastore import Delete, Query from google.appengine.api.datastore_errors import BadArgumentError, \ BadValueError from google.appengine.api.datastore_types import Blob, Key, Text, \ ValidateInteger from google.appengine.api.namespace_manager import set_namespace from google.appengine.ext.db.metadata import get_kinds, get_namespaces from djangotoolbox.db.base import ( NonrelDatabaseClient, NonrelDatabaseFeatures, NonrelDatabaseIntrospection, NonrelDatabaseOperations, NonrelDatabaseValidation, NonrelDatabaseWrapper) from djangotoolbox.db.utils import decimal_to_string from ..boot import DATA_ROOT from ..utils import appid, on_production_server from .creation import DatabaseCreation from .stubs import stub_manager DATASTORE_PATHS = { 'datastore_path': os.path.join(DATA_ROOT, 'datastore'), 'blobstore_path': os.path.join(DATA_ROOT, 'blobstore'), #'rdbms_sqlite_path': os.path.join(DATA_ROOT, 'rdbms'), 'prospective_search_path': os.path.join(DATA_ROOT, 'prospective-search'), } def key_from_path(db_table, value): """ Workaround for GAE choosing not to validate integer ids when creating keys. TODO: Should be removed if it gets fixed. """ if isinstance(value, (int, long)): ValidateInteger(value, 'id') return Key.from_path(db_table, value) def get_datastore_paths(options): paths = {} for key, path in DATASTORE_PATHS.items(): paths[key] = options.get(key, path) return paths def destroy_datastore(paths): """Destroys the appengine datastore at the specified paths.""" for path in paths.values(): if not path: continue try: if os.path.isdir(path): shutil.rmtree(path) else: os.remove(path) except OSError, error: if error.errno != 2: logging.error("Failed to clear datastore: %s" % error) class DatabaseFeatures(NonrelDatabaseFeatures): # GAE only allow strictly positive integers (and strings) to be # used as key values. allows_primary_key_0 = False # Anything that results in a something different than a positive # integer or a string cannot be directly used as a key on GAE. # Note that DecimalField values are encoded as strings, so can be # used as keys. # With some encoding, we could allow most fields to be used as a # primary key, but for now only mark what can and what cannot be # safely used. supports_primary_key_on = \ NonrelDatabaseFeatures.supports_primary_key_on - set(( 'FloatField', 'DateField', 'DateTimeField', 'TimeField', 'BooleanField', 'NullBooleanField', 'TextField', 'XMLField')) class DatabaseOperations(NonrelDatabaseOperations): compiler_module = __name__.rsplit('.', 1)[0] + '.compiler' # Date used to store times as datetimes. # TODO: Use just date()? DEFAULT_DATE = datetime.date(1970, 1, 1) # Time used to store dates as datetimes. DEFAULT_TIME = datetime.time() def sql_flush(self, style, tables, sequences, allow_cascade=False): self.connection.flush() return [] def value_to_db_auto(self, value): """ New keys generated by the GAE datastore hold longs. """ if value is None: return None return long(value) def value_for_db(self, value, field, lookup=None): """ We'll simulate `startswith` lookups with two inequalities: property >= value and property <= value + u'\ufffd', and need to "double" the value before passing it through the actual datastore conversions. """ super_value_for_db = super(DatabaseOperations, self).value_for_db if lookup == 'startswith': return [super_value_for_db(value, field, lookup), super_value_for_db(value + u'\ufffd', field, lookup)] return super_value_for_db(value, field, lookup) def _value_for_db(self, value, field, field_kind, db_type, lookup): """ GAE database may store a restricted set of Python types, for some cases it has its own types like Key, Text or Blob. TODO: Consider moving empty list handling here (from insert). """ # Store Nones as Nones to handle nullable fields, even keys. if value is None: return None # Parent can handle iterable fields and Django wrappers. value = super(DatabaseOperations, self)._value_for_db( value, field, field_kind, db_type, lookup) # Convert decimals to strings preserving order. if field_kind == 'DecimalField': value = decimal_to_string( value, field.max_digits, field.decimal_places) # Create GAE db.Keys from Django keys. # We use model's table name as key kind (the table of the model # of the instance that the key identifies, for ForeignKeys and # other relations). if db_type == 'key': # value = self._value_for_db_key(value, field_kind) try: value = key_from_path(field.model._meta.db_table, value) except (BadArgumentError, BadValueError,): raise DatabaseError("Only strings and positive integers " "may be used as keys on GAE.") # Store all strings as unicode, use db.Text for longer content. elif db_type == 'string' or db_type == 'text': if isinstance(value, str): value = value.decode('utf-8') if db_type == 'text': value = Text(value) # Store all date / time values as datetimes, by using some # default time or date. elif db_type == 'date': value = datetime.datetime.combine(value, self.DEFAULT_TIME) elif db_type == 'time': value = datetime.datetime.combine(self.DEFAULT_DATE, value) # Store BlobField, DictField and EmbeddedModelField values as Blobs. elif db_type == 'bytes': value = Blob(value) return value def _value_from_db(self, value, field, field_kind, db_type): """ Undoes conversions done in value_for_db. """ # We could have stored None for a null field. if value is None: return None # All keys were converted to the Key class. if db_type == 'key': assert isinstance(value, Key), \ "GAE db.Key expected! Try changing to old storage, " \ "dumping data, changing to new storage and reloading." assert value.parent() is None, "Parents are not yet supported!" value = value.id_or_name() # value = self._value_from_db_key(value, field_kind) # Always retrieve strings as unicode (old datasets may # contain non-unicode strings). elif db_type == 'string' or db_type == 'text': if isinstance(value, str): value = value.decode('utf-8') else: value = unicode(value) # Dates and times are stored as datetimes, drop the added part. elif db_type == 'date': value = value.date() elif db_type == 'time': value = value.time() # Convert GAE Blobs to plain strings for Django. elif db_type == 'bytes': value = str(value) # Revert the decimal-to-string encoding. if field_kind == 'DecimalField': value = decimal.Decimal(value) return super(DatabaseOperations, self)._value_from_db( value, field, field_kind, db_type) # def _value_for_db_key(self, value, field_kind): # """ # Converts values to be used as entity keys to strings, # trying (but not fully succeeding) to preserve comparisons. # """ # # Bools as positive integers. # if field_kind == 'BooleanField': # value = int(value) + 1 # # Encode floats as strings. # elif field_kind == 'FloatField': # value = self.value_to_db_decimal( # decimal.Decimal(value), None, None) # # Integers as strings (string keys sort after int keys, so # # all need to be encoded to preserve comparisons). # elif field_kind in ('IntegerField', 'BigIntegerField', # 'PositiveIntegerField', 'PositiveSmallIntegerField', # 'SmallIntegerField'): # value = self.value_to_db_decimal( # decimal.Decimal(value), None, 0) # return value # def value_from_db_key(self, value, field_kind): # """ # Decodes value previously encoded in a key. # """ # if field_kind == 'BooleanField': # value = bool(value - 1) # elif field_kind == 'FloatField': # value = float(value) # elif field_kind in ('IntegerField', 'BigIntegerField', # 'PositiveIntegerField', 'PositiveSmallIntegerField', # 'SmallIntegerField'): # value = int(value) # return value class DatabaseClient(NonrelDatabaseClient): pass class DatabaseValidation(NonrelDatabaseValidation): pass class DatabaseIntrospection(NonrelDatabaseIntrospection): def table_names(self, cursor=None): """ Returns a list of names of all tables that exist in the database. """ return [kind.key().name() for kind in Query(kind='__kind__').Run()] class DatabaseWrapper(NonrelDatabaseWrapper): def __init__(self, *args, **kwds): super(DatabaseWrapper, self).__init__(*args, **kwds) self.features = DatabaseFeatures(self) self.ops = DatabaseOperations(self) self.client = DatabaseClient(self) self.creation = DatabaseCreation(self) self.validation = DatabaseValidation(self) self.introspection = DatabaseIntrospection(self) options = self.settings_dict self.remote_app_id = options.get('REMOTE_APP_ID', appid) self.domain = options.get('DOMAIN', 'appspot.com') self.remote_api_path = options.get('REMOTE_API_PATH', None) self.secure_remote_api = options.get('SECURE_REMOTE_API', True) remote = options.get('REMOTE', False) if on_production_server: remote = False if remote: stub_manager.setup_remote_stubs(self) else: stub_manager.setup_stubs(self) def flush(self): """ Helper function to remove the current datastore and re-open the stubs. """ if stub_manager.active_stubs == 'remote': import random import string code = ''.join([random.choice(string.ascii_letters) for x in range(4)]) print "\n\n!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!" print "!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!" print "Warning! You're about to delete the *production* datastore!" print "Only models defined in your INSTALLED_APPS can be removed!" print "If you want to clear the whole datastore you have to use " \ "the datastore viewer in the dashboard. Also, in order to " \ "delete all unneeded indexes you have to run appcfg.py " \ "vacuum_indexes." print "In order to proceed you have to enter the following code:" print code response = raw_input("Repeat: ") if code == response: print "Deleting..." delete_all_entities() print "Datastore flushed! Please check your dashboard's " \ "datastore viewer for any remaining entities and " \ "remove all unneeded indexes with appcfg.py " \ "vacuum_indexes." else: print "Aborting." exit() elif stub_manager.active_stubs == 'test': stub_manager.deactivate_test_stubs() stub_manager.activate_test_stubs(self) else: destroy_datastore(get_datastore_paths(self.settings_dict)) stub_manager.setup_local_stubs(self) def delete_all_entities(): for namespace in get_namespaces(): set_namespace(namespace) for kind in get_kinds(): if kind.startswith('__'): continue while True: data = Query(kind=kind, keys_only=True).Get(200) if not data: break Delete(data)
mit
alasdairtran/mclearn
projects/jakub/learning_curve/uncertainty_curve.py
2
1191
import json import sys import numpy as np import sklearn.gaussian_process # Import splitter sys.path.insert(1, '..') import splitter TRAINING_SAMPLES_NUM = 1000000 TESTING_SAMPLES_NUM = 1000 MAX_GP = 3000 STEP = 100 ALPHA = .002 LENGTH_SCALE = 1 def perform_gp(train_X, train_y, test_X): kernel = sklearn.gaussian_process.kernels.RBF( length_scale=LENGTH_SCALE) gp = sklearn.gaussian_process.GaussianProcessRegressor( kernel=kernel, alpha=ALPHA, copy_X_train=False) gp.fit(train_X, train_y) _, sigmas = gp.predict(test_X, return_std=True) return np.mean(sigmas) def main(path_in, path_out): data = splitter.load(path_in) (train_X, train_y), (test_X, test_y) \ = splitter.split(data, TRAINING_SAMPLES_NUM, TESTING_SAMPLES_NUM) gp_x = list(range(STEP, MAX_GP+1, STEP)) gp_y = [] for i, n in enumerate(gp_x): print('Starting GP', i + 1) gp_y.append(perform_gp(train_X[:n], train_y[:n], test_X)) with open(path_out, 'w') as f: json.dump({ 'gp_x': gp_x, 'gp_y': gp_y, }, f) if __name__ == '__main__': main(*sys.argv[1:3])
bsd-3-clause
Orpine/py-R-FCN
lib/datasets/pascal_voc.py
11
14217
# -------------------------------------------------------- # Fast R-CNN # Copyright (c) 2015 Microsoft # Licensed under The MIT License [see LICENSE for details] # Written by Ross Girshick # -------------------------------------------------------- import os from datasets.imdb import imdb import datasets.ds_utils as ds_utils import xml.etree.ElementTree as ET import numpy as np import scipy.sparse import scipy.io as sio import utils.cython_bbox import cPickle import subprocess import uuid from voc_eval import voc_eval from fast_rcnn.config import cfg class pascal_voc(imdb): def __init__(self, image_set, year, devkit_path=None): imdb.__init__(self, 'voc_' + year + '_' + image_set) self._year = year self._image_set = image_set self._devkit_path = self._get_default_path() if devkit_path is None \ else devkit_path self._data_path = os.path.join(self._devkit_path, 'VOC' + self._year) self._classes = ('__background__', # always index 0 'aeroplane', 'bicycle', 'bird', 'boat', 'bottle', 'bus', 'car', 'cat', 'chair', 'cow', 'diningtable', 'dog', 'horse', 'motorbike', 'person', 'pottedplant', 'sheep', 'sofa', 'train', 'tvmonitor') self._class_to_ind = dict(zip(self.classes, xrange(self.num_classes))) self._image_ext = '.jpg' self._image_index = self._load_image_set_index() # Default to roidb handler self._roidb_handler = self.selective_search_roidb self._salt = str(uuid.uuid4()) self._comp_id = 'comp4' # PASCAL specific config options self.config = {'cleanup' : True, 'use_salt' : True, 'use_diff' : False, 'matlab_eval' : False, 'rpn_file' : None, 'min_size' : 2} assert os.path.exists(self._devkit_path), \ 'VOCdevkit path does not exist: {}'.format(self._devkit_path) assert os.path.exists(self._data_path), \ 'Path does not exist: {}'.format(self._data_path) def image_path_at(self, i): """ Return the absolute path to image i in the image sequence. """ return self.image_path_from_index(self._image_index[i]) def image_path_from_index(self, index): """ Construct an image path from the image's "index" identifier. """ image_path = os.path.join(self._data_path, 'JPEGImages', index + self._image_ext) assert os.path.exists(image_path), \ 'Path does not exist: {}'.format(image_path) return image_path def _load_image_set_index(self): """ Load the indexes listed in this dataset's image set file. """ # Example path to image set file: # self._devkit_path + /VOCdevkit2007/VOC2007/ImageSets/Main/val.txt image_set_file = os.path.join(self._data_path, 'ImageSets', 'Main', self._image_set + '.txt') assert os.path.exists(image_set_file), \ 'Path does not exist: {}'.format(image_set_file) with open(image_set_file) as f: image_index = [x.strip() for x in f.readlines()] return image_index def _get_default_path(self): """ Return the default path where PASCAL VOC is expected to be installed. """ return os.path.join(cfg.DATA_DIR, 'VOCdevkit' + self._year) def gt_roidb(self): """ Return the database of ground-truth regions of interest. This function loads/saves from/to a cache file to speed up future calls. """ cache_file = os.path.join(self.cache_path, self.name + '_gt_roidb.pkl') if os.path.exists(cache_file): with open(cache_file, 'rb') as fid: roidb = cPickle.load(fid) print '{} gt roidb loaded from {}'.format(self.name, cache_file) return roidb gt_roidb = [self._load_pascal_annotation(index) for index in self.image_index] with open(cache_file, 'wb') as fid: cPickle.dump(gt_roidb, fid, cPickle.HIGHEST_PROTOCOL) print 'wrote gt roidb to {}'.format(cache_file) return gt_roidb def selective_search_roidb(self): """ Return the database of selective search regions of interest. Ground-truth ROIs are also included. This function loads/saves from/to a cache file to speed up future calls. """ cache_file = os.path.join(self.cache_path, self.name + '_selective_search_roidb.pkl') if os.path.exists(cache_file): with open(cache_file, 'rb') as fid: roidb = cPickle.load(fid) print '{} ss roidb loaded from {}'.format(self.name, cache_file) return roidb if int(self._year) == 2007 or self._image_set != 'test': gt_roidb = self.gt_roidb() ss_roidb = self._load_selective_search_roidb(gt_roidb) roidb = imdb.merge_roidbs(gt_roidb, ss_roidb) else: roidb = self._load_selective_search_roidb(None) with open(cache_file, 'wb') as fid: cPickle.dump(roidb, fid, cPickle.HIGHEST_PROTOCOL) print 'wrote ss roidb to {}'.format(cache_file) return roidb def rpn_roidb(self): if int(self._year) == 2007 or self._image_set != 'test': gt_roidb = self.gt_roidb() rpn_roidb = self._load_rpn_roidb(gt_roidb) roidb = imdb.merge_roidbs(gt_roidb, rpn_roidb) else: roidb = self._load_rpn_roidb(None) return roidb def _load_rpn_roidb(self, gt_roidb): filename = self.config['rpn_file'] print 'loading {}'.format(filename) assert os.path.exists(filename), \ 'rpn data not found at: {}'.format(filename) with open(filename, 'rb') as f: box_list = cPickle.load(f) return self.create_roidb_from_box_list(box_list, gt_roidb) def _load_selective_search_roidb(self, gt_roidb): filename = os.path.abspath(os.path.join(cfg.DATA_DIR, 'selective_search_data', self.name + '.mat')) assert os.path.exists(filename), \ 'Selective search data not found at: {}'.format(filename) raw_data = sio.loadmat(filename)['boxes'].ravel() box_list = [] for i in xrange(raw_data.shape[0]): boxes = raw_data[i][:, (1, 0, 3, 2)] - 1 keep = ds_utils.unique_boxes(boxes) boxes = boxes[keep, :] keep = ds_utils.filter_small_boxes(boxes, self.config['min_size']) boxes = boxes[keep, :] box_list.append(boxes) return self.create_roidb_from_box_list(box_list, gt_roidb) def _load_pascal_annotation(self, index): """ Load image and bounding boxes info from XML file in the PASCAL VOC format. """ filename = os.path.join(self._data_path, 'Annotations', index + '.xml') tree = ET.parse(filename) objs = tree.findall('object') if not self.config['use_diff']: # Exclude the samples labeled as difficult non_diff_objs = [ obj for obj in objs if int(obj.find('difficult').text) == 0] # if len(non_diff_objs) != len(objs): # print 'Removed {} difficult objects'.format( # len(objs) - len(non_diff_objs)) objs = non_diff_objs num_objs = len(objs) boxes = np.zeros((num_objs, 4), dtype=np.uint16) gt_classes = np.zeros((num_objs), dtype=np.int32) overlaps = np.zeros((num_objs, self.num_classes), dtype=np.float32) # "Seg" area for pascal is just the box area seg_areas = np.zeros((num_objs), dtype=np.float32) # Load object bounding boxes into a data frame. for ix, obj in enumerate(objs): bbox = obj.find('bndbox') # Make pixel indexes 0-based x1 = float(bbox.find('xmin').text) - 1 y1 = float(bbox.find('ymin').text) - 1 x2 = float(bbox.find('xmax').text) - 1 y2 = float(bbox.find('ymax').text) - 1 cls = self._class_to_ind[obj.find('name').text.lower().strip()] boxes[ix, :] = [x1, y1, x2, y2] gt_classes[ix] = cls overlaps[ix, cls] = 1.0 seg_areas[ix] = (x2 - x1 + 1) * (y2 - y1 + 1) overlaps = scipy.sparse.csr_matrix(overlaps) return {'boxes' : boxes, 'gt_classes': gt_classes, 'gt_overlaps' : overlaps, 'flipped' : False, 'seg_areas' : seg_areas} def _get_comp_id(self): comp_id = (self._comp_id + '_' + self._salt if self.config['use_salt'] else self._comp_id) return comp_id def _get_voc_results_file_template(self): # VOCdevkit/results/VOC2007/Main/<comp_id>_det_test_aeroplane.txt filename = self._get_comp_id() + '_det_' + self._image_set + '_{:s}.txt' path = os.path.join( self._devkit_path, 'results', 'VOC' + self._year, 'Main', filename) return path def _write_voc_results_file(self, all_boxes): for cls_ind, cls in enumerate(self.classes): if cls == '__background__': continue print 'Writing {} VOC results file'.format(cls) filename = self._get_voc_results_file_template().format(cls) with open(filename, 'wt') as f: for im_ind, index in enumerate(self.image_index): dets = all_boxes[cls_ind][im_ind] if dets == []: continue # the VOCdevkit expects 1-based indices for k in xrange(dets.shape[0]): f.write('{:s} {:.3f} {:.1f} {:.1f} {:.1f} {:.1f}\n'. format(index, dets[k, -1], dets[k, 0] + 1, dets[k, 1] + 1, dets[k, 2] + 1, dets[k, 3] + 1)) def _do_python_eval(self, output_dir = 'output'): annopath = os.path.join( self._devkit_path, 'VOC' + self._year, 'Annotations', '{:s}.xml') imagesetfile = os.path.join( self._devkit_path, 'VOC' + self._year, 'ImageSets', 'Main', self._image_set + '.txt') cachedir = os.path.join(self._devkit_path, 'annotations_cache') aps = [] # The PASCAL VOC metric changed in 2010 use_07_metric = True if int(self._year) < 2010 else False print 'VOC07 metric? ' + ('Yes' if use_07_metric else 'No') if not os.path.isdir(output_dir): os.mkdir(output_dir) for i, cls in enumerate(self._classes): if cls == '__background__': continue filename = self._get_voc_results_file_template().format(cls) rec, prec, ap = voc_eval( filename, annopath, imagesetfile, cls, cachedir, ovthresh=0.5, use_07_metric=use_07_metric) aps += [ap] print('AP for {} = {:.4f}'.format(cls, ap)) with open(os.path.join(output_dir, cls + '_pr.pkl'), 'w') as f: cPickle.dump({'rec': rec, 'prec': prec, 'ap': ap}, f) print('Mean AP = {:.4f}'.format(np.mean(aps))) print('~~~~~~~~') print('Results:') for ap in aps: print('{:.3f}'.format(ap)) print('{:.3f}'.format(np.mean(aps))) print('~~~~~~~~') print('') print('--------------------------------------------------------------') print('Results computed with the **unofficial** Python eval code.') print('Results should be very close to the official MATLAB eval code.') print('Recompute with `./tools/reval.py --matlab ...` for your paper.') print('-- Thanks, The Management') print('--------------------------------------------------------------') def _do_matlab_eval(self, output_dir='output'): print '-----------------------------------------------------' print 'Computing results with the official MATLAB eval code.' print '-----------------------------------------------------' path = os.path.join(cfg.ROOT_DIR, 'lib', 'datasets', 'VOCdevkit-matlab-wrapper') cmd = 'cd {} && '.format(path) cmd += '{:s} -nodisplay -nodesktop '.format(cfg.MATLAB) cmd += '-r "dbstop if error; ' cmd += 'voc_eval(\'{:s}\',\'{:s}\',\'{:s}\',\'{:s}\'); quit;"' \ .format(self._devkit_path, self._get_comp_id(), self._image_set, output_dir) print('Running:\n{}'.format(cmd)) status = subprocess.call(cmd, shell=True) def evaluate_detections(self, all_boxes, output_dir): self._write_voc_results_file(all_boxes) self._do_python_eval(output_dir) if self.config['matlab_eval']: self._do_matlab_eval(output_dir) if self.config['cleanup']: for cls in self._classes: if cls == '__background__': continue filename = self._get_voc_results_file_template().format(cls) os.remove(filename) def competition_mode(self, on): if on: self.config['use_salt'] = False self.config['cleanup'] = False else: self.config['use_salt'] = True self.config['cleanup'] = True if __name__ == '__main__': from datasets.pascal_voc import pascal_voc d = pascal_voc('trainval', '2007') res = d.roidb from IPython import embed; embed()
mit
roxyboy/scikit-learn
sklearn/utils/__init__.py
131
14185
""" The :mod:`sklearn.utils` module includes various utilities. """ from collections import Sequence import numpy as np from scipy.sparse import issparse import warnings from .murmurhash import murmurhash3_32 from .validation import (as_float_array, assert_all_finite, check_random_state, column_or_1d, check_array, check_consistent_length, check_X_y, indexable, check_symmetric, DataConversionWarning) from .class_weight import compute_class_weight, compute_sample_weight from ..externals.joblib import cpu_count __all__ = ["murmurhash3_32", "as_float_array", "assert_all_finite", "check_array", "check_random_state", "compute_class_weight", "compute_sample_weight", "column_or_1d", "safe_indexing", "check_consistent_length", "check_X_y", 'indexable', "check_symmetric"] class deprecated(object): """Decorator to mark a function or class as deprecated. Issue a warning when the function is called/the class is instantiated and adds a warning to the docstring. The optional extra argument will be appended to the deprecation message and the docstring. Note: to use this with the default value for extra, put in an empty of parentheses: >>> from sklearn.utils import deprecated >>> deprecated() # doctest: +ELLIPSIS <sklearn.utils.deprecated object at ...> >>> @deprecated() ... def some_function(): pass """ # Adapted from http://wiki.python.org/moin/PythonDecoratorLibrary, # but with many changes. def __init__(self, extra=''): """ Parameters ---------- extra: string to be added to the deprecation messages """ self.extra = extra def __call__(self, obj): if isinstance(obj, type): return self._decorate_class(obj) else: return self._decorate_fun(obj) def _decorate_class(self, cls): msg = "Class %s is deprecated" % cls.__name__ if self.extra: msg += "; %s" % self.extra # FIXME: we should probably reset __new__ for full generality init = cls.__init__ def wrapped(*args, **kwargs): warnings.warn(msg, category=DeprecationWarning) return init(*args, **kwargs) cls.__init__ = wrapped wrapped.__name__ = '__init__' wrapped.__doc__ = self._update_doc(init.__doc__) wrapped.deprecated_original = init return cls def _decorate_fun(self, fun): """Decorate function fun""" msg = "Function %s is deprecated" % fun.__name__ if self.extra: msg += "; %s" % self.extra def wrapped(*args, **kwargs): warnings.warn(msg, category=DeprecationWarning) return fun(*args, **kwargs) wrapped.__name__ = fun.__name__ wrapped.__dict__ = fun.__dict__ wrapped.__doc__ = self._update_doc(fun.__doc__) return wrapped def _update_doc(self, olddoc): newdoc = "DEPRECATED" if self.extra: newdoc = "%s: %s" % (newdoc, self.extra) if olddoc: newdoc = "%s\n\n%s" % (newdoc, olddoc) return newdoc def safe_mask(X, mask): """Return a mask which is safe to use on X. Parameters ---------- X : {array-like, sparse matrix} Data on which to apply mask. mask: array Mask to be used on X. Returns ------- mask """ mask = np.asarray(mask) if np.issubdtype(mask.dtype, np.int): return mask if hasattr(X, "toarray"): ind = np.arange(mask.shape[0]) mask = ind[mask] return mask def safe_indexing(X, indices): """Return items or rows from X using indices. Allows simple indexing of lists or arrays. Parameters ---------- X : array-like, sparse-matrix, list. Data from which to sample rows or items. indices : array-like, list Indices according to which X will be subsampled. """ if hasattr(X, "iloc"): # Pandas Dataframes and Series try: return X.iloc[indices] except ValueError: # Cython typed memoryviews internally used in pandas do not support # readonly buffers. warnings.warn("Copying input dataframe for slicing.", DataConversionWarning) return X.copy().iloc[indices] elif hasattr(X, "shape"): if hasattr(X, 'take') and (hasattr(indices, 'dtype') and indices.dtype.kind == 'i'): # This is often substantially faster than X[indices] return X.take(indices, axis=0) else: return X[indices] else: return [X[idx] for idx in indices] def resample(*arrays, **options): """Resample arrays or sparse matrices in a consistent way The default strategy implements one step of the bootstrapping procedure. Parameters ---------- *arrays : sequence of indexable data-structures Indexable data-structures can be arrays, lists, dataframes or scipy sparse matrices with consistent first dimension. replace : boolean, True by default Implements resampling with replacement. If False, this will implement (sliced) random permutations. n_samples : int, None by default Number of samples to generate. If left to None this is automatically set to the first dimension of the arrays. random_state : int or RandomState instance Control the shuffling for reproducible behavior. Returns ------- resampled_arrays : sequence of indexable data-structures Sequence of resampled views of the collections. The original arrays are not impacted. Examples -------- It is possible to mix sparse and dense arrays in the same run:: >>> X = np.array([[1., 0.], [2., 1.], [0., 0.]]) >>> y = np.array([0, 1, 2]) >>> from scipy.sparse import coo_matrix >>> X_sparse = coo_matrix(X) >>> from sklearn.utils import resample >>> X, X_sparse, y = resample(X, X_sparse, y, random_state=0) >>> X array([[ 1., 0.], [ 2., 1.], [ 1., 0.]]) >>> X_sparse # doctest: +ELLIPSIS +NORMALIZE_WHITESPACE <3x2 sparse matrix of type '<... 'numpy.float64'>' with 4 stored elements in Compressed Sparse Row format> >>> X_sparse.toarray() array([[ 1., 0.], [ 2., 1.], [ 1., 0.]]) >>> y array([0, 1, 0]) >>> resample(y, n_samples=2, random_state=0) array([0, 1]) See also -------- :func:`sklearn.utils.shuffle` """ random_state = check_random_state(options.pop('random_state', None)) replace = options.pop('replace', True) max_n_samples = options.pop('n_samples', None) if options: raise ValueError("Unexpected kw arguments: %r" % options.keys()) if len(arrays) == 0: return None first = arrays[0] n_samples = first.shape[0] if hasattr(first, 'shape') else len(first) if max_n_samples is None: max_n_samples = n_samples if max_n_samples > n_samples: raise ValueError("Cannot sample %d out of arrays with dim %d" % ( max_n_samples, n_samples)) check_consistent_length(*arrays) if replace: indices = random_state.randint(0, n_samples, size=(max_n_samples,)) else: indices = np.arange(n_samples) random_state.shuffle(indices) indices = indices[:max_n_samples] # convert sparse matrices to CSR for row-based indexing arrays = [a.tocsr() if issparse(a) else a for a in arrays] resampled_arrays = [safe_indexing(a, indices) for a in arrays] if len(resampled_arrays) == 1: # syntactic sugar for the unit argument case return resampled_arrays[0] else: return resampled_arrays def shuffle(*arrays, **options): """Shuffle arrays or sparse matrices in a consistent way This is a convenience alias to ``resample(*arrays, replace=False)`` to do random permutations of the collections. Parameters ---------- *arrays : sequence of indexable data-structures Indexable data-structures can be arrays, lists, dataframes or scipy sparse matrices with consistent first dimension. random_state : int or RandomState instance Control the shuffling for reproducible behavior. n_samples : int, None by default Number of samples to generate. If left to None this is automatically set to the first dimension of the arrays. Returns ------- shuffled_arrays : sequence of indexable data-structures Sequence of shuffled views of the collections. The original arrays are not impacted. Examples -------- It is possible to mix sparse and dense arrays in the same run:: >>> X = np.array([[1., 0.], [2., 1.], [0., 0.]]) >>> y = np.array([0, 1, 2]) >>> from scipy.sparse import coo_matrix >>> X_sparse = coo_matrix(X) >>> from sklearn.utils import shuffle >>> X, X_sparse, y = shuffle(X, X_sparse, y, random_state=0) >>> X array([[ 0., 0.], [ 2., 1.], [ 1., 0.]]) >>> X_sparse # doctest: +ELLIPSIS +NORMALIZE_WHITESPACE <3x2 sparse matrix of type '<... 'numpy.float64'>' with 3 stored elements in Compressed Sparse Row format> >>> X_sparse.toarray() array([[ 0., 0.], [ 2., 1.], [ 1., 0.]]) >>> y array([2, 1, 0]) >>> shuffle(y, n_samples=2, random_state=0) array([0, 1]) See also -------- :func:`sklearn.utils.resample` """ options['replace'] = False return resample(*arrays, **options) def safe_sqr(X, copy=True): """Element wise squaring of array-likes and sparse matrices. Parameters ---------- X : array like, matrix, sparse matrix copy : boolean, optional, default True Whether to create a copy of X and operate on it or to perform inplace computation (default behaviour). Returns ------- X ** 2 : element wise square """ X = check_array(X, accept_sparse=['csr', 'csc', 'coo']) if issparse(X): if copy: X = X.copy() X.data **= 2 else: if copy: X = X ** 2 else: X **= 2 return X def gen_batches(n, batch_size): """Generator to create slices containing batch_size elements, from 0 to n. The last slice may contain less than batch_size elements, when batch_size does not divide n. Examples -------- >>> from sklearn.utils import gen_batches >>> list(gen_batches(7, 3)) [slice(0, 3, None), slice(3, 6, None), slice(6, 7, None)] >>> list(gen_batches(6, 3)) [slice(0, 3, None), slice(3, 6, None)] >>> list(gen_batches(2, 3)) [slice(0, 2, None)] """ start = 0 for _ in range(int(n // batch_size)): end = start + batch_size yield slice(start, end) start = end if start < n: yield slice(start, n) def gen_even_slices(n, n_packs, n_samples=None): """Generator to create n_packs slices going up to n. Pass n_samples when the slices are to be used for sparse matrix indexing; slicing off-the-end raises an exception, while it works for NumPy arrays. Examples -------- >>> from sklearn.utils import gen_even_slices >>> list(gen_even_slices(10, 1)) [slice(0, 10, None)] >>> list(gen_even_slices(10, 10)) #doctest: +ELLIPSIS [slice(0, 1, None), slice(1, 2, None), ..., slice(9, 10, None)] >>> list(gen_even_slices(10, 5)) #doctest: +ELLIPSIS [slice(0, 2, None), slice(2, 4, None), ..., slice(8, 10, None)] >>> list(gen_even_slices(10, 3)) [slice(0, 4, None), slice(4, 7, None), slice(7, 10, None)] """ start = 0 if n_packs < 1: raise ValueError("gen_even_slices got n_packs=%s, must be >=1" % n_packs) for pack_num in range(n_packs): this_n = n // n_packs if pack_num < n % n_packs: this_n += 1 if this_n > 0: end = start + this_n if n_samples is not None: end = min(n_samples, end) yield slice(start, end, None) start = end def _get_n_jobs(n_jobs): """Get number of jobs for the computation. This function reimplements the logic of joblib to determine the actual number of jobs depending on the cpu count. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Parameters ---------- n_jobs : int Number of jobs stated in joblib convention. Returns ------- n_jobs : int The actual number of jobs as positive integer. Examples -------- >>> from sklearn.utils import _get_n_jobs >>> _get_n_jobs(4) 4 >>> jobs = _get_n_jobs(-2) >>> assert jobs == max(cpu_count() - 1, 1) >>> _get_n_jobs(0) Traceback (most recent call last): ... ValueError: Parameter n_jobs == 0 has no meaning. """ if n_jobs < 0: return max(cpu_count() + 1 + n_jobs, 1) elif n_jobs == 0: raise ValueError('Parameter n_jobs == 0 has no meaning.') else: return n_jobs def tosequence(x): """Cast iterable x to a Sequence, avoiding a copy if possible.""" if isinstance(x, np.ndarray): return np.asarray(x) elif isinstance(x, Sequence): return x else: return list(x) class ConvergenceWarning(UserWarning): """Custom warning to capture convergence problems""" class DataDimensionalityWarning(UserWarning): """Custom warning to notify potential issues with data dimensionality"""
bsd-3-clause
ephes/scikit-learn
examples/svm/plot_svm_margin.py
315
2328
#!/usr/bin/python # -*- coding: utf-8 -*- """ ========================================================= SVM Margins Example ========================================================= The plots below illustrate the effect the parameter `C` has on the separation line. A large value of `C` basically tells our model that we do not have that much faith in our data's distribution, and will only consider points close to line of separation. A small value of `C` includes more/all the observations, allowing the margins to be calculated using all the data in the area. """ print(__doc__) # Code source: Gaël Varoquaux # Modified for documentation by Jaques Grobler # License: BSD 3 clause import numpy as np import matplotlib.pyplot as plt from sklearn import svm # we create 40 separable points np.random.seed(0) X = np.r_[np.random.randn(20, 2) - [2, 2], np.random.randn(20, 2) + [2, 2]] Y = [0] * 20 + [1] * 20 # figure number fignum = 1 # fit the model for name, penalty in (('unreg', 1), ('reg', 0.05)): clf = svm.SVC(kernel='linear', C=penalty) clf.fit(X, Y) # get the separating hyperplane w = clf.coef_[0] a = -w[0] / w[1] xx = np.linspace(-5, 5) yy = a * xx - (clf.intercept_[0]) / w[1] # plot the parallels to the separating hyperplane that pass through the # support vectors margin = 1 / np.sqrt(np.sum(clf.coef_ ** 2)) yy_down = yy + a * margin yy_up = yy - a * margin # plot the line, the points, and the nearest vectors to the plane plt.figure(fignum, figsize=(4, 3)) plt.clf() plt.plot(xx, yy, 'k-') plt.plot(xx, yy_down, 'k--') plt.plot(xx, yy_up, 'k--') plt.scatter(clf.support_vectors_[:, 0], clf.support_vectors_[:, 1], s=80, facecolors='none', zorder=10) plt.scatter(X[:, 0], X[:, 1], c=Y, zorder=10, cmap=plt.cm.Paired) plt.axis('tight') x_min = -4.8 x_max = 4.2 y_min = -6 y_max = 6 XX, YY = np.mgrid[x_min:x_max:200j, y_min:y_max:200j] Z = clf.predict(np.c_[XX.ravel(), YY.ravel()]) # Put the result into a color plot Z = Z.reshape(XX.shape) plt.figure(fignum, figsize=(4, 3)) plt.pcolormesh(XX, YY, Z, cmap=plt.cm.Paired) plt.xlim(x_min, x_max) plt.ylim(y_min, y_max) plt.xticks(()) plt.yticks(()) fignum = fignum + 1 plt.show()
bsd-3-clause
roxyboy/scikit-learn
sklearn/svm/base.py
155
36018
from __future__ import print_function import numpy as np import scipy.sparse as sp import warnings from abc import ABCMeta, abstractmethod from . import libsvm, liblinear from . import libsvm_sparse from ..base import BaseEstimator, ClassifierMixin, ChangedBehaviorWarning from ..preprocessing import LabelEncoder from ..multiclass import _ovr_decision_function from ..utils import check_array, check_random_state, column_or_1d from ..utils import ConvergenceWarning, compute_class_weight, deprecated from ..utils.extmath import safe_sparse_dot from ..utils.validation import check_is_fitted, NotFittedError from ..externals import six LIBSVM_IMPL = ['c_svc', 'nu_svc', 'one_class', 'epsilon_svr', 'nu_svr'] def _one_vs_one_coef(dual_coef, n_support, support_vectors): """Generate primal coefficients from dual coefficients for the one-vs-one multi class LibSVM in the case of a linear kernel.""" # get 1vs1 weights for all n*(n-1) classifiers. # this is somewhat messy. # shape of dual_coef_ is nSV * (n_classes -1) # see docs for details n_class = dual_coef.shape[0] + 1 # XXX we could do preallocation of coef but # would have to take care in the sparse case coef = [] sv_locs = np.cumsum(np.hstack([[0], n_support])) for class1 in range(n_class): # SVs for class1: sv1 = support_vectors[sv_locs[class1]:sv_locs[class1 + 1], :] for class2 in range(class1 + 1, n_class): # SVs for class1: sv2 = support_vectors[sv_locs[class2]:sv_locs[class2 + 1], :] # dual coef for class1 SVs: alpha1 = dual_coef[class2 - 1, sv_locs[class1]:sv_locs[class1 + 1]] # dual coef for class2 SVs: alpha2 = dual_coef[class1, sv_locs[class2]:sv_locs[class2 + 1]] # build weight for class1 vs class2 coef.append(safe_sparse_dot(alpha1, sv1) + safe_sparse_dot(alpha2, sv2)) return coef class BaseLibSVM(six.with_metaclass(ABCMeta, BaseEstimator)): """Base class for estimators that use libsvm as backing library This implements support vector machine classification and regression. Parameter documentation is in the derived `SVC` class. """ # The order of these must match the integer values in LibSVM. # XXX These are actually the same in the dense case. Need to factor # this out. _sparse_kernels = ["linear", "poly", "rbf", "sigmoid", "precomputed"] @abstractmethod def __init__(self, impl, kernel, degree, gamma, coef0, tol, C, nu, epsilon, shrinking, probability, cache_size, class_weight, verbose, max_iter, random_state): if impl not in LIBSVM_IMPL: # pragma: no cover raise ValueError("impl should be one of %s, %s was given" % ( LIBSVM_IMPL, impl)) # FIXME Remove gamma=0.0 support in 0.18 if gamma == 0: msg = ("gamma=%s has been deprecated in favor of " "gamma='%s' as of 0.17. Backward compatibility" " for gamma=%s will be removed in %s") invalid_gamma = 0.0 warnings.warn(msg % (invalid_gamma, "auto", invalid_gamma, "0.18"), DeprecationWarning) self._impl = impl self.kernel = kernel self.degree = degree self.gamma = gamma self.coef0 = coef0 self.tol = tol self.C = C self.nu = nu self.epsilon = epsilon self.shrinking = shrinking self.probability = probability self.cache_size = cache_size self.class_weight = class_weight self.verbose = verbose self.max_iter = max_iter self.random_state = random_state @property def _pairwise(self): # Used by cross_val_score. kernel = self.kernel return kernel == "precomputed" or callable(kernel) def fit(self, X, y, sample_weight=None): """Fit the SVM model according to the given training data. Parameters ---------- X : {array-like, sparse matrix}, shape (n_samples, n_features) Training vectors, where n_samples is the number of samples and n_features is the number of features. For kernel="precomputed", the expected shape of X is (n_samples, n_samples). y : array-like, shape (n_samples,) Target values (class labels in classification, real numbers in regression) sample_weight : array-like, shape (n_samples,) Per-sample weights. Rescale C per sample. Higher weights force the classifier to put more emphasis on these points. Returns ------- self : object Returns self. Notes ------ If X and y are not C-ordered and contiguous arrays of np.float64 and X is not a scipy.sparse.csr_matrix, X and/or y may be copied. If X is a dense array, then the other methods will not support sparse matrices as input. """ rnd = check_random_state(self.random_state) sparse = sp.isspmatrix(X) if sparse and self.kernel == "precomputed": raise TypeError("Sparse precomputed kernels are not supported.") self._sparse = sparse and not callable(self.kernel) X = check_array(X, accept_sparse='csr', dtype=np.float64, order='C') y = self._validate_targets(y) sample_weight = np.asarray([] if sample_weight is None else sample_weight, dtype=np.float64) solver_type = LIBSVM_IMPL.index(self._impl) # input validation if solver_type != 2 and X.shape[0] != y.shape[0]: raise ValueError("X and y have incompatible shapes.\n" + "X has %s samples, but y has %s." % (X.shape[0], y.shape[0])) if self.kernel == "precomputed" and X.shape[0] != X.shape[1]: raise ValueError("X.shape[0] should be equal to X.shape[1]") if sample_weight.shape[0] > 0 and sample_weight.shape[0] != X.shape[0]: raise ValueError("sample_weight and X have incompatible shapes: " "%r vs %r\n" "Note: Sparse matrices cannot be indexed w/" "boolean masks (use `indices=True` in CV)." % (sample_weight.shape, X.shape)) # FIXME remove (self.gamma == 0) in 0.18 if (self.kernel in ['poly', 'rbf']) and ((self.gamma == 0) or (self.gamma == 'auto')): # if custom gamma is not provided ... self._gamma = 1.0 / X.shape[1] elif self.gamma == 'auto': self._gamma = 0.0 else: self._gamma = self.gamma kernel = self.kernel if callable(kernel): kernel = 'precomputed' fit = self._sparse_fit if self._sparse else self._dense_fit if self.verbose: # pragma: no cover print('[LibSVM]', end='') seed = rnd.randint(np.iinfo('i').max) fit(X, y, sample_weight, solver_type, kernel, random_seed=seed) # see comment on the other call to np.iinfo in this file self.shape_fit_ = X.shape # In binary case, we need to flip the sign of coef, intercept and # decision function. Use self._intercept_ and self._dual_coef_ internally. self._intercept_ = self.intercept_.copy() self._dual_coef_ = self.dual_coef_ if self._impl in ['c_svc', 'nu_svc'] and len(self.classes_) == 2: self.intercept_ *= -1 self.dual_coef_ = -self.dual_coef_ return self def _validate_targets(self, y): """Validation of y and class_weight. Default implementation for SVR and one-class; overridden in BaseSVC. """ # XXX this is ugly. # Regression models should not have a class_weight_ attribute. self.class_weight_ = np.empty(0) return column_or_1d(y, warn=True).astype(np.float64) def _warn_from_fit_status(self): assert self.fit_status_ in (0, 1) if self.fit_status_ == 1: warnings.warn('Solver terminated early (max_iter=%i).' ' Consider pre-processing your data with' ' StandardScaler or MinMaxScaler.' % self.max_iter, ConvergenceWarning) def _dense_fit(self, X, y, sample_weight, solver_type, kernel, random_seed): if callable(self.kernel): # you must store a reference to X to compute the kernel in predict # TODO: add keyword copy to copy on demand self.__Xfit = X X = self._compute_kernel(X) if X.shape[0] != X.shape[1]: raise ValueError("X.shape[0] should be equal to X.shape[1]") libsvm.set_verbosity_wrap(self.verbose) # we don't pass **self.get_params() to allow subclasses to # add other parameters to __init__ self.support_, self.support_vectors_, self.n_support_, \ self.dual_coef_, self.intercept_, self.probA_, \ self.probB_, self.fit_status_ = libsvm.fit( X, y, svm_type=solver_type, sample_weight=sample_weight, class_weight=self.class_weight_, kernel=kernel, C=self.C, nu=self.nu, probability=self.probability, degree=self.degree, shrinking=self.shrinking, tol=self.tol, cache_size=self.cache_size, coef0=self.coef0, gamma=self._gamma, epsilon=self.epsilon, max_iter=self.max_iter, random_seed=random_seed) self._warn_from_fit_status() def _sparse_fit(self, X, y, sample_weight, solver_type, kernel, random_seed): X.data = np.asarray(X.data, dtype=np.float64, order='C') X.sort_indices() kernel_type = self._sparse_kernels.index(kernel) libsvm_sparse.set_verbosity_wrap(self.verbose) self.support_, self.support_vectors_, dual_coef_data, \ self.intercept_, self.n_support_, \ self.probA_, self.probB_, self.fit_status_ = \ libsvm_sparse.libsvm_sparse_train( X.shape[1], X.data, X.indices, X.indptr, y, solver_type, kernel_type, self.degree, self._gamma, self.coef0, self.tol, self.C, self.class_weight_, sample_weight, self.nu, self.cache_size, self.epsilon, int(self.shrinking), int(self.probability), self.max_iter, random_seed) self._warn_from_fit_status() if hasattr(self, "classes_"): n_class = len(self.classes_) - 1 else: # regression n_class = 1 n_SV = self.support_vectors_.shape[0] dual_coef_indices = np.tile(np.arange(n_SV), n_class) dual_coef_indptr = np.arange(0, dual_coef_indices.size + 1, dual_coef_indices.size / n_class) self.dual_coef_ = sp.csr_matrix( (dual_coef_data, dual_coef_indices, dual_coef_indptr), (n_class, n_SV)) def predict(self, X): """Perform regression on samples in X. For an one-class model, +1 or -1 is returned. Parameters ---------- X : {array-like, sparse matrix}, shape (n_samples, n_features) For kernel="precomputed", the expected shape of X is (n_samples_test, n_samples_train). Returns ------- y_pred : array, shape (n_samples,) """ X = self._validate_for_predict(X) predict = self._sparse_predict if self._sparse else self._dense_predict return predict(X) def _dense_predict(self, X): n_samples, n_features = X.shape X = self._compute_kernel(X) if X.ndim == 1: X = check_array(X, order='C') kernel = self.kernel if callable(self.kernel): kernel = 'precomputed' if X.shape[1] != self.shape_fit_[0]: raise ValueError("X.shape[1] = %d should be equal to %d, " "the number of samples at training time" % (X.shape[1], self.shape_fit_[0])) svm_type = LIBSVM_IMPL.index(self._impl) return libsvm.predict( X, self.support_, self.support_vectors_, self.n_support_, self._dual_coef_, self._intercept_, self.probA_, self.probB_, svm_type=svm_type, kernel=kernel, degree=self.degree, coef0=self.coef0, gamma=self._gamma, cache_size=self.cache_size) def _sparse_predict(self, X): # Precondition: X is a csr_matrix of dtype np.float64. kernel = self.kernel if callable(kernel): kernel = 'precomputed' kernel_type = self._sparse_kernels.index(kernel) C = 0.0 # C is not useful here return libsvm_sparse.libsvm_sparse_predict( X.data, X.indices, X.indptr, self.support_vectors_.data, self.support_vectors_.indices, self.support_vectors_.indptr, self._dual_coef_.data, self._intercept_, LIBSVM_IMPL.index(self._impl), kernel_type, self.degree, self._gamma, self.coef0, self.tol, C, self.class_weight_, self.nu, self.epsilon, self.shrinking, self.probability, self.n_support_, self.probA_, self.probB_) def _compute_kernel(self, X): """Return the data transformed by a callable kernel""" if callable(self.kernel): # in the case of precomputed kernel given as a function, we # have to compute explicitly the kernel matrix kernel = self.kernel(X, self.__Xfit) if sp.issparse(kernel): kernel = kernel.toarray() X = np.asarray(kernel, dtype=np.float64, order='C') return X @deprecated(" and will be removed in 0.19") def decision_function(self, X): """Distance of the samples X to the separating hyperplane. Parameters ---------- X : array-like, shape (n_samples, n_features) For kernel="precomputed", the expected shape of X is [n_samples_test, n_samples_train]. Returns ------- X : array-like, shape (n_samples, n_class * (n_class-1) / 2) Returns the decision function of the sample for each class in the model. """ return self._decision_function(X) def _decision_function(self, X): """Distance of the samples X to the separating hyperplane. Parameters ---------- X : array-like, shape (n_samples, n_features) Returns ------- X : array-like, shape (n_samples, n_class * (n_class-1) / 2) Returns the decision function of the sample for each class in the model. """ # NOTE: _validate_for_predict contains check for is_fitted # hence must be placed before any other attributes are used. X = self._validate_for_predict(X) X = self._compute_kernel(X) if self._sparse: dec_func = self._sparse_decision_function(X) else: dec_func = self._dense_decision_function(X) # In binary case, we need to flip the sign of coef, intercept and # decision function. if self._impl in ['c_svc', 'nu_svc'] and len(self.classes_) == 2: return -dec_func.ravel() return dec_func def _dense_decision_function(self, X): X = check_array(X, dtype=np.float64, order="C") kernel = self.kernel if callable(kernel): kernel = 'precomputed' return libsvm.decision_function( X, self.support_, self.support_vectors_, self.n_support_, self._dual_coef_, self._intercept_, self.probA_, self.probB_, svm_type=LIBSVM_IMPL.index(self._impl), kernel=kernel, degree=self.degree, cache_size=self.cache_size, coef0=self.coef0, gamma=self._gamma) def _sparse_decision_function(self, X): X.data = np.asarray(X.data, dtype=np.float64, order='C') kernel = self.kernel if hasattr(kernel, '__call__'): kernel = 'precomputed' kernel_type = self._sparse_kernels.index(kernel) return libsvm_sparse.libsvm_sparse_decision_function( X.data, X.indices, X.indptr, self.support_vectors_.data, self.support_vectors_.indices, self.support_vectors_.indptr, self._dual_coef_.data, self._intercept_, LIBSVM_IMPL.index(self._impl), kernel_type, self.degree, self._gamma, self.coef0, self.tol, self.C, self.class_weight_, self.nu, self.epsilon, self.shrinking, self.probability, self.n_support_, self.probA_, self.probB_) def _validate_for_predict(self, X): check_is_fitted(self, 'support_') X = check_array(X, accept_sparse='csr', dtype=np.float64, order="C") if self._sparse and not sp.isspmatrix(X): X = sp.csr_matrix(X) if self._sparse: X.sort_indices() if sp.issparse(X) and not self._sparse and not callable(self.kernel): raise ValueError( "cannot use sparse input in %r trained on dense data" % type(self).__name__) n_samples, n_features = X.shape if self.kernel == "precomputed": if X.shape[1] != self.shape_fit_[0]: raise ValueError("X.shape[1] = %d should be equal to %d, " "the number of samples at training time" % (X.shape[1], self.shape_fit_[0])) elif n_features != self.shape_fit_[1]: raise ValueError("X.shape[1] = %d should be equal to %d, " "the number of features at training time" % (n_features, self.shape_fit_[1])) return X @property def coef_(self): if self.kernel != 'linear': raise ValueError('coef_ is only available when using a ' 'linear kernel') coef = self._get_coef() # coef_ being a read-only property, it's better to mark the value as # immutable to avoid hiding potential bugs for the unsuspecting user. if sp.issparse(coef): # sparse matrix do not have global flags coef.data.flags.writeable = False else: # regular dense array coef.flags.writeable = False return coef def _get_coef(self): return safe_sparse_dot(self._dual_coef_, self.support_vectors_) class BaseSVC(six.with_metaclass(ABCMeta, BaseLibSVM, ClassifierMixin)): """ABC for LibSVM-based classifiers.""" @abstractmethod def __init__(self, impl, kernel, degree, gamma, coef0, tol, C, nu, shrinking, probability, cache_size, class_weight, verbose, max_iter, decision_function_shape, random_state): self.decision_function_shape = decision_function_shape super(BaseSVC, self).__init__( impl=impl, kernel=kernel, degree=degree, gamma=gamma, coef0=coef0, tol=tol, C=C, nu=nu, epsilon=0., shrinking=shrinking, probability=probability, cache_size=cache_size, class_weight=class_weight, verbose=verbose, max_iter=max_iter, random_state=random_state) def _validate_targets(self, y): y_ = column_or_1d(y, warn=True) cls, y = np.unique(y_, return_inverse=True) self.class_weight_ = compute_class_weight(self.class_weight, cls, y_) if len(cls) < 2: raise ValueError( "The number of classes has to be greater than one; got %d" % len(cls)) self.classes_ = cls return np.asarray(y, dtype=np.float64, order='C') def decision_function(self, X): """Distance of the samples X to the separating hyperplane. Parameters ---------- X : array-like, shape (n_samples, n_features) Returns ------- X : array-like, shape (n_samples, n_classes * (n_classes-1) / 2) Returns the decision function of the sample for each class in the model. If decision_function_shape='ovr', the shape is (n_samples, n_classes) """ dec = self._decision_function(X) if self.decision_function_shape is None and len(self.classes_) > 2: warnings.warn("The decision_function_shape default value will " "change from 'ovo' to 'ovr' in 0.18. This will change " "the shape of the decision function returned by " "SVC.", ChangedBehaviorWarning) if self.decision_function_shape == 'ovr': return _ovr_decision_function(dec < 0, dec, len(self.classes_)) return dec def predict(self, X): """Perform classification on samples in X. For an one-class model, +1 or -1 is returned. Parameters ---------- X : {array-like, sparse matrix}, shape (n_samples, n_features) For kernel="precomputed", the expected shape of X is [n_samples_test, n_samples_train] Returns ------- y_pred : array, shape (n_samples,) Class labels for samples in X. """ y = super(BaseSVC, self).predict(X) return self.classes_.take(np.asarray(y, dtype=np.intp)) # Hacky way of getting predict_proba to raise an AttributeError when # probability=False using properties. Do not use this in new code; when # probabilities are not available depending on a setting, introduce two # estimators. def _check_proba(self): if not self.probability: raise AttributeError("predict_proba is not available when " " probability=False") if self._impl not in ('c_svc', 'nu_svc'): raise AttributeError("predict_proba only implemented for SVC" " and NuSVC") @property def predict_proba(self): """Compute probabilities of possible outcomes for samples in X. The model need to have probability information computed at training time: fit with attribute `probability` set to True. Parameters ---------- X : array-like, shape (n_samples, n_features) For kernel="precomputed", the expected shape of X is [n_samples_test, n_samples_train] Returns ------- T : array-like, shape (n_samples, n_classes) Returns the probability of the sample for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute `classes_`. Notes ----- The probability model is created using cross validation, so the results can be slightly different than those obtained by predict. Also, it will produce meaningless results on very small datasets. """ self._check_proba() return self._predict_proba def _predict_proba(self, X): X = self._validate_for_predict(X) if self.probA_.size == 0 or self.probB_.size == 0: raise NotFittedError("predict_proba is not available when fitted " "with probability=False") pred_proba = (self._sparse_predict_proba if self._sparse else self._dense_predict_proba) return pred_proba(X) @property def predict_log_proba(self): """Compute log probabilities of possible outcomes for samples in X. The model need to have probability information computed at training time: fit with attribute `probability` set to True. Parameters ---------- X : array-like, shape (n_samples, n_features) For kernel="precomputed", the expected shape of X is [n_samples_test, n_samples_train] Returns ------- T : array-like, shape (n_samples, n_classes) Returns the log-probabilities of the sample for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute `classes_`. Notes ----- The probability model is created using cross validation, so the results can be slightly different than those obtained by predict. Also, it will produce meaningless results on very small datasets. """ self._check_proba() return self._predict_log_proba def _predict_log_proba(self, X): return np.log(self.predict_proba(X)) def _dense_predict_proba(self, X): X = self._compute_kernel(X) kernel = self.kernel if callable(kernel): kernel = 'precomputed' svm_type = LIBSVM_IMPL.index(self._impl) pprob = libsvm.predict_proba( X, self.support_, self.support_vectors_, self.n_support_, self._dual_coef_, self._intercept_, self.probA_, self.probB_, svm_type=svm_type, kernel=kernel, degree=self.degree, cache_size=self.cache_size, coef0=self.coef0, gamma=self._gamma) return pprob def _sparse_predict_proba(self, X): X.data = np.asarray(X.data, dtype=np.float64, order='C') kernel = self.kernel if callable(kernel): kernel = 'precomputed' kernel_type = self._sparse_kernels.index(kernel) return libsvm_sparse.libsvm_sparse_predict_proba( X.data, X.indices, X.indptr, self.support_vectors_.data, self.support_vectors_.indices, self.support_vectors_.indptr, self._dual_coef_.data, self._intercept_, LIBSVM_IMPL.index(self._impl), kernel_type, self.degree, self._gamma, self.coef0, self.tol, self.C, self.class_weight_, self.nu, self.epsilon, self.shrinking, self.probability, self.n_support_, self.probA_, self.probB_) def _get_coef(self): if self.dual_coef_.shape[0] == 1: # binary classifier coef = safe_sparse_dot(self.dual_coef_, self.support_vectors_) else: # 1vs1 classifier coef = _one_vs_one_coef(self.dual_coef_, self.n_support_, self.support_vectors_) if sp.issparse(coef[0]): coef = sp.vstack(coef).tocsr() else: coef = np.vstack(coef) return coef def _get_liblinear_solver_type(multi_class, penalty, loss, dual): """Find the liblinear magic number for the solver. This number depends on the values of the following attributes: - multi_class - penalty - loss - dual The same number is also internally used by LibLinear to determine which solver to use. """ # nested dicts containing level 1: available loss functions, # level2: available penalties for the given loss functin, # level3: wether the dual solver is available for the specified # combination of loss function and penalty _solver_type_dict = { 'logistic_regression': { 'l1': {False: 6}, 'l2': {False: 0, True: 7}}, 'hinge': { 'l2': {True: 3}}, 'squared_hinge': { 'l1': {False: 5}, 'l2': {False: 2, True: 1}}, 'epsilon_insensitive': { 'l2': {True: 13}}, 'squared_epsilon_insensitive': { 'l2': {False: 11, True: 12}}, 'crammer_singer': 4 } if multi_class == 'crammer_singer': return _solver_type_dict[multi_class] elif multi_class != 'ovr': raise ValueError("`multi_class` must be one of `ovr`, " "`crammer_singer`, got %r" % multi_class) # FIXME loss.lower() --> loss in 0.18 _solver_pen = _solver_type_dict.get(loss.lower(), None) if _solver_pen is None: error_string = ("loss='%s' is not supported" % loss) else: # FIME penalty.lower() --> penalty in 0.18 _solver_dual = _solver_pen.get(penalty.lower(), None) if _solver_dual is None: error_string = ("The combination of penalty='%s' " "and loss='%s' is not supported" % (penalty, loss)) else: solver_num = _solver_dual.get(dual, None) if solver_num is None: error_string = ("The combination of penalty='%s' and " "loss='%s' are not supported when dual=%s" % (penalty, loss, dual)) else: return solver_num raise ValueError('Unsupported set of arguments: %s, ' 'Parameters: penalty=%r, loss=%r, dual=%r' % (error_string, penalty, loss, dual)) def _fit_liblinear(X, y, C, fit_intercept, intercept_scaling, class_weight, penalty, dual, verbose, max_iter, tol, random_state=None, multi_class='ovr', loss='logistic_regression', epsilon=0.1): """Used by Logistic Regression (and CV) and LinearSVC. Preprocessing is done in this function before supplying it to liblinear. Parameters ---------- X : {array-like, sparse matrix}, shape (n_samples, n_features) Training vector, where n_samples in the number of samples and n_features is the number of features. y : array-like, shape (n_samples,) Target vector relative to X C : float Inverse of cross-validation parameter. Lower the C, the more the penalization. fit_intercept : bool Whether or not to fit the intercept, that is to add a intercept term to the decision function. intercept_scaling : float LibLinear internally penalizes the intercept and this term is subject to regularization just like the other terms of the feature vector. In order to avoid this, one should increase the intercept_scaling. such that the feature vector becomes [x, intercept_scaling]. class_weight : {dict, 'balanced'}, optional Weights associated with classes in the form ``{class_label: weight}``. If not given, all classes are supposed to have weight one. For multi-output problems, a list of dicts can be provided in the same order as the columns of y. The "balanced" mode uses the values of y to automatically adjust weights inversely proportional to class frequencies in the input data as ``n_samples / (n_classes * np.bincount(y))`` penalty : str, {'l1', 'l2'} The norm of the penalty used in regularization. dual : bool Dual or primal formulation, verbose : int Set verbose to any positive number for verbosity. max_iter : int Number of iterations. tol : float Stopping condition. random_state : int seed, RandomState instance, or None (default) The seed of the pseudo random number generator to use when shuffling the data. multi_class : str, {'ovr', 'crammer_singer'} `ovr` trains n_classes one-vs-rest classifiers, while `crammer_singer` optimizes a joint objective over all classes. While `crammer_singer` is interesting from an theoretical perspective as it is consistent it is seldom used in practice and rarely leads to better accuracy and is more expensive to compute. If `crammer_singer` is chosen, the options loss, penalty and dual will be ignored. loss : str, {'logistic_regression', 'hinge', 'squared_hinge', 'epsilon_insensitive', 'squared_epsilon_insensitive} The loss function used to fit the model. epsilon : float, optional (default=0.1) Epsilon parameter in the epsilon-insensitive loss function. Note that the value of this parameter depends on the scale of the target variable y. If unsure, set epsilon=0. Returns ------- coef_ : ndarray, shape (n_features, n_features + 1) The coefficent vector got by minimizing the objective function. intercept_ : float The intercept term added to the vector. n_iter_ : int Maximum number of iterations run across all classes. """ # FIXME Remove case insensitivity in 0.18 --------------------- loss_l, penalty_l = loss.lower(), penalty.lower() msg = ("loss='%s' has been deprecated in favor of " "loss='%s' as of 0.16. Backward compatibility" " for the uppercase notation will be removed in %s") if (not loss.islower()) and loss_l not in ('l1', 'l2'): warnings.warn(msg % (loss, loss_l, "0.18"), DeprecationWarning) if not penalty.islower(): warnings.warn(msg.replace("loss", "penalty") % (penalty, penalty_l, "0.18"), DeprecationWarning) # ------------------------------------------------------------- # FIXME loss_l --> loss in 0.18 if loss_l not in ['epsilon_insensitive', 'squared_epsilon_insensitive']: enc = LabelEncoder() y_ind = enc.fit_transform(y) classes_ = enc.classes_ if len(classes_) < 2: raise ValueError("This solver needs samples of at least 2 classes" " in the data, but the data contains only one" " class: %r" % classes_[0]) class_weight_ = compute_class_weight(class_weight, classes_, y) else: class_weight_ = np.empty(0, dtype=np.float) y_ind = y liblinear.set_verbosity_wrap(verbose) rnd = check_random_state(random_state) if verbose: print('[LibLinear]', end='') # LinearSVC breaks when intercept_scaling is <= 0 bias = -1.0 if fit_intercept: if intercept_scaling <= 0: raise ValueError("Intercept scaling is %r but needs to be greater than 0." " To disable fitting an intercept," " set fit_intercept=False." % intercept_scaling) else: bias = intercept_scaling libsvm.set_verbosity_wrap(verbose) libsvm_sparse.set_verbosity_wrap(verbose) liblinear.set_verbosity_wrap(verbose) # LibLinear wants targets as doubles, even for classification y_ind = np.asarray(y_ind, dtype=np.float64).ravel() solver_type = _get_liblinear_solver_type(multi_class, penalty, loss, dual) raw_coef_, n_iter_ = liblinear.train_wrap( X, y_ind, sp.isspmatrix(X), solver_type, tol, bias, C, class_weight_, max_iter, rnd.randint(np.iinfo('i').max), epsilon) # Regarding rnd.randint(..) in the above signature: # seed for srand in range [0..INT_MAX); due to limitations in Numpy # on 32-bit platforms, we can't get to the UINT_MAX limit that # srand supports n_iter_ = max(n_iter_) if n_iter_ >= max_iter and verbose > 0: warnings.warn("Liblinear failed to converge, increase " "the number of iterations.", ConvergenceWarning) if fit_intercept: coef_ = raw_coef_[:, :-1] intercept_ = intercept_scaling * raw_coef_[:, -1] else: coef_ = raw_coef_ intercept_ = 0. return coef_, intercept_, n_iter_
bsd-3-clause
edx/edx-platform
openedx/features/content_type_gating/block_transformers.py
4
4525
""" Content Type Gate Transformer implementation. Limits access for certain users to certain types of content. """ from django.conf import settings from lms.djangoapps.course_blocks.transformers.user_partitions import UserPartitionTransformer from openedx.core.djangoapps.content.block_structure.transformer import BlockStructureTransformer from openedx.features.content_type_gating.helpers import CONTENT_GATING_PARTITION_ID from openedx.features.content_type_gating.models import ContentTypeGatingConfig class ContentTypeGateTransformer(BlockStructureTransformer): """ A transformer that adds a partition condition for all graded content so that the content is only visible to verified users. This transformer requires that the UserPartitionTransformer also be included in your transformer list. """ WRITE_VERSION = 1 READ_VERSION = 1 @classmethod def name(cls): """ Unique identifier for the transformer's class; same identifier used in setup.py. """ return "content_type_gate" @classmethod def collect(cls, block_structure): """ Collects any information that's necessary to execute this transformer's transform method. """ block_structure.request_xblock_fields('group_access', 'graded', 'has_score', 'weight') def _set_contains_gated_content_on_parents(self, block_structure, block_key): """ This will recursively set a field on all the parents of a block if one of the problems inside of it is content gated. `contains_gated_content` can then be used to indicate something in the blocks subtree is gated. """ if block_structure.get_xblock_field(block_key, 'contains_gated_content'): return block_structure.override_xblock_field(block_key, 'contains_gated_content', True) for parent_block_key in block_structure.get_parents(block_key): self._set_contains_gated_content_on_parents(block_structure, parent_block_key) @staticmethod def _get_block_group_access(block_structure, block_key): """ Gets the current group_access value for a block, supporting inheritance when possible. In order to support inheritance, UserPartitionTransformer must also be used. """ # See user_partitions.py for the code that sets this field. merged_access = block_structure.get_transformer_block_field( block_key, UserPartitionTransformer, 'merged_group_access', None ) if merged_access: # merged_access holds a dictionary of sets, but group_access is a dictionary of lists, so we convert here # (sets seem like a better format for this, but existing code already expects lists) current_access = {p: list(g) for (p, g) in merged_access.get_allowed_groups().items()} else: # This fallback code has a bug if UserPartitionTranformer is not being used -- it does not consider # inheritance from parent blocks. This is why our class docstring recommends UserPartitionTranformer. current_access = block_structure.get_xblock_field(block_key, 'group_access') return current_access or {} def transform(self, usage_info, block_structure): if not ContentTypeGatingConfig.enabled_for_enrollment( user=usage_info.user, course_key=usage_info.course_key, ): return for block_key in block_structure.topological_traversal(): graded = block_structure.get_xblock_field(block_key, 'graded') has_score = block_structure.get_xblock_field(block_key, 'has_score') weight_not_zero = block_structure.get_xblock_field(block_key, 'weight') != 0 problem_eligible_for_content_gating = graded and has_score and weight_not_zero if problem_eligible_for_content_gating: current_access = self._get_block_group_access(block_structure, block_key) current_access.setdefault( CONTENT_GATING_PARTITION_ID, [settings.CONTENT_TYPE_GATE_GROUP_IDS['full_access']] ) block_structure.override_xblock_field(block_key, 'group_access', current_access) if current_access[CONTENT_GATING_PARTITION_ID] == [settings.CONTENT_TYPE_GATE_GROUP_IDS['full_access']]: self._set_contains_gated_content_on_parents(block_structure, block_key)
agpl-3.0
ephes/scikit-learn
sklearn/grid_search.py
102
36232
""" The :mod:`sklearn.grid_search` includes utilities to fine-tune the parameters of an estimator. """ from __future__ import print_function # Author: Alexandre Gramfort <alexandre.gramfort@inria.fr>, # Gael Varoquaux <gael.varoquaux@normalesup.org> # Andreas Mueller <amueller@ais.uni-bonn.de> # Olivier Grisel <olivier.grisel@ensta.org> # License: BSD 3 clause from abc import ABCMeta, abstractmethod from collections import Mapping, namedtuple, Sized from functools import partial, reduce from itertools import product import operator import warnings import numpy as np from .base import BaseEstimator, is_classifier, clone from .base import MetaEstimatorMixin, ChangedBehaviorWarning from .cross_validation import check_cv from .cross_validation import _fit_and_score from .externals.joblib import Parallel, delayed from .externals import six from .utils import check_random_state from .utils.random import sample_without_replacement from .utils.validation import _num_samples, indexable from .utils.metaestimators import if_delegate_has_method from .metrics.scorer import check_scoring __all__ = ['GridSearchCV', 'ParameterGrid', 'fit_grid_point', 'ParameterSampler', 'RandomizedSearchCV'] class ParameterGrid(object): """Grid of parameters with a discrete number of values for each. Can be used to iterate over parameter value combinations with the Python built-in function iter. Read more in the :ref:`User Guide <grid_search>`. Parameters ---------- param_grid : dict of string to sequence, or sequence of such The parameter grid to explore, as a dictionary mapping estimator parameters to sequences of allowed values. An empty dict signifies default parameters. A sequence of dicts signifies a sequence of grids to search, and is useful to avoid exploring parameter combinations that make no sense or have no effect. See the examples below. Examples -------- >>> from sklearn.grid_search import ParameterGrid >>> param_grid = {'a': [1, 2], 'b': [True, False]} >>> list(ParameterGrid(param_grid)) == ( ... [{'a': 1, 'b': True}, {'a': 1, 'b': False}, ... {'a': 2, 'b': True}, {'a': 2, 'b': False}]) True >>> grid = [{'kernel': ['linear']}, {'kernel': ['rbf'], 'gamma': [1, 10]}] >>> list(ParameterGrid(grid)) == [{'kernel': 'linear'}, ... {'kernel': 'rbf', 'gamma': 1}, ... {'kernel': 'rbf', 'gamma': 10}] True >>> ParameterGrid(grid)[1] == {'kernel': 'rbf', 'gamma': 1} True See also -------- :class:`GridSearchCV`: uses ``ParameterGrid`` to perform a full parallelized parameter search. """ def __init__(self, param_grid): if isinstance(param_grid, Mapping): # wrap dictionary in a singleton list to support either dict # or list of dicts param_grid = [param_grid] self.param_grid = param_grid def __iter__(self): """Iterate over the points in the grid. Returns ------- params : iterator over dict of string to any Yields dictionaries mapping each estimator parameter to one of its allowed values. """ for p in self.param_grid: # Always sort the keys of a dictionary, for reproducibility items = sorted(p.items()) if not items: yield {} else: keys, values = zip(*items) for v in product(*values): params = dict(zip(keys, v)) yield params def __len__(self): """Number of points on the grid.""" # Product function that can handle iterables (np.product can't). product = partial(reduce, operator.mul) return sum(product(len(v) for v in p.values()) if p else 1 for p in self.param_grid) def __getitem__(self, ind): """Get the parameters that would be ``ind``th in iteration Parameters ---------- ind : int The iteration index Returns ------- params : dict of string to any Equal to list(self)[ind] """ # This is used to make discrete sampling without replacement memory # efficient. for sub_grid in self.param_grid: # XXX: could memoize information used here if not sub_grid: if ind == 0: return {} else: ind -= 1 continue # Reverse so most frequent cycling parameter comes first keys, values_lists = zip(*sorted(sub_grid.items())[::-1]) sizes = [len(v_list) for v_list in values_lists] total = np.product(sizes) if ind >= total: # Try the next grid ind -= total else: out = {} for key, v_list, n in zip(keys, values_lists, sizes): ind, offset = divmod(ind, n) out[key] = v_list[offset] return out raise IndexError('ParameterGrid index out of range') class ParameterSampler(object): """Generator on parameters sampled from given distributions. Non-deterministic iterable over random candidate combinations for hyper- parameter search. If all parameters are presented as a list, sampling without replacement is performed. If at least one parameter is given as a distribution, sampling with replacement is used. It is highly recommended to use continuous distributions for continuous parameters. Note that as of SciPy 0.12, the ``scipy.stats.distributions`` do not accept a custom RNG instance and always use the singleton RNG from ``numpy.random``. Hence setting ``random_state`` will not guarantee a deterministic iteration whenever ``scipy.stats`` distributions are used to define the parameter search space. Read more in the :ref:`User Guide <grid_search>`. Parameters ---------- param_distributions : dict Dictionary where the keys are parameters and values are distributions from which a parameter is to be sampled. Distributions either have to provide a ``rvs`` function to sample from them, or can be given as a list of values, where a uniform distribution is assumed. n_iter : integer Number of parameter settings that are produced. random_state : int or RandomState Pseudo random number generator state used for random uniform sampling from lists of possible values instead of scipy.stats distributions. Returns ------- params : dict of string to any **Yields** dictionaries mapping each estimator parameter to as sampled value. Examples -------- >>> from sklearn.grid_search import ParameterSampler >>> from scipy.stats.distributions import expon >>> import numpy as np >>> np.random.seed(0) >>> param_grid = {'a':[1, 2], 'b': expon()} >>> param_list = list(ParameterSampler(param_grid, n_iter=4)) >>> rounded_list = [dict((k, round(v, 6)) for (k, v) in d.items()) ... for d in param_list] >>> rounded_list == [{'b': 0.89856, 'a': 1}, ... {'b': 0.923223, 'a': 1}, ... {'b': 1.878964, 'a': 2}, ... {'b': 1.038159, 'a': 2}] True """ def __init__(self, param_distributions, n_iter, random_state=None): self.param_distributions = param_distributions self.n_iter = n_iter self.random_state = random_state def __iter__(self): # check if all distributions are given as lists # in this case we want to sample without replacement all_lists = np.all([not hasattr(v, "rvs") for v in self.param_distributions.values()]) rnd = check_random_state(self.random_state) if all_lists: # look up sampled parameter settings in parameter grid param_grid = ParameterGrid(self.param_distributions) grid_size = len(param_grid) if grid_size < self.n_iter: raise ValueError( "The total space of parameters %d is smaller " "than n_iter=%d." % (grid_size, self.n_iter) + " For exhaustive searches, use GridSearchCV.") for i in sample_without_replacement(grid_size, self.n_iter, random_state=rnd): yield param_grid[i] else: # Always sort the keys of a dictionary, for reproducibility items = sorted(self.param_distributions.items()) for _ in six.moves.range(self.n_iter): params = dict() for k, v in items: if hasattr(v, "rvs"): params[k] = v.rvs() else: params[k] = v[rnd.randint(len(v))] yield params def __len__(self): """Number of points that will be sampled.""" return self.n_iter def fit_grid_point(X, y, estimator, parameters, train, test, scorer, verbose, error_score='raise', **fit_params): """Run fit on one set of parameters. Parameters ---------- X : array-like, sparse matrix or list Input data. y : array-like or None Targets for input data. estimator : estimator object This estimator will be cloned and then fitted. parameters : dict Parameters to be set on estimator for this grid point. train : ndarray, dtype int or bool Boolean mask or indices for training set. test : ndarray, dtype int or bool Boolean mask or indices for test set. scorer : callable or None. If provided must be a scorer callable object / function with signature ``scorer(estimator, X, y)``. verbose : int Verbosity level. **fit_params : kwargs Additional parameter passed to the fit function of the estimator. error_score : 'raise' (default) or numeric Value to assign to the score if an error occurs in estimator fitting. If set to 'raise', the error is raised. If a numeric value is given, FitFailedWarning is raised. This parameter does not affect the refit step, which will always raise the error. Returns ------- score : float Score of this parameter setting on given training / test split. parameters : dict The parameters that have been evaluated. n_samples_test : int Number of test samples in this split. """ score, n_samples_test, _ = _fit_and_score(estimator, X, y, scorer, train, test, verbose, parameters, fit_params, error_score) return score, parameters, n_samples_test def _check_param_grid(param_grid): if hasattr(param_grid, 'items'): param_grid = [param_grid] for p in param_grid: for v in p.values(): if isinstance(v, np.ndarray) and v.ndim > 1: raise ValueError("Parameter array should be one-dimensional.") check = [isinstance(v, k) for k in (list, tuple, np.ndarray)] if True not in check: raise ValueError("Parameter values should be a list.") if len(v) == 0: raise ValueError("Parameter values should be a non-empty " "list.") class _CVScoreTuple (namedtuple('_CVScoreTuple', ('parameters', 'mean_validation_score', 'cv_validation_scores'))): # A raw namedtuple is very memory efficient as it packs the attributes # in a struct to get rid of the __dict__ of attributes in particular it # does not copy the string for the keys on each instance. # By deriving a namedtuple class just to introduce the __repr__ method we # would also reintroduce the __dict__ on the instance. By telling the # Python interpreter that this subclass uses static __slots__ instead of # dynamic attributes. Furthermore we don't need any additional slot in the # subclass so we set __slots__ to the empty tuple. __slots__ = () def __repr__(self): """Simple custom repr to summarize the main info""" return "mean: {0:.5f}, std: {1:.5f}, params: {2}".format( self.mean_validation_score, np.std(self.cv_validation_scores), self.parameters) class BaseSearchCV(six.with_metaclass(ABCMeta, BaseEstimator, MetaEstimatorMixin)): """Base class for hyper parameter search with cross-validation.""" @abstractmethod def __init__(self, estimator, scoring=None, fit_params=None, n_jobs=1, iid=True, refit=True, cv=None, verbose=0, pre_dispatch='2*n_jobs', error_score='raise'): self.scoring = scoring self.estimator = estimator self.n_jobs = n_jobs self.fit_params = fit_params if fit_params is not None else {} self.iid = iid self.refit = refit self.cv = cv self.verbose = verbose self.pre_dispatch = pre_dispatch self.error_score = error_score @property def _estimator_type(self): return self.estimator._estimator_type def score(self, X, y=None): """Returns the score on the given data, if the estimator has been refit This uses the score defined by ``scoring`` where provided, and the ``best_estimator_.score`` method otherwise. Parameters ---------- X : array-like, shape = [n_samples, n_features] Input data, where n_samples is the number of samples and n_features is the number of features. y : array-like, shape = [n_samples] or [n_samples, n_output], optional Target relative to X for classification or regression; None for unsupervised learning. Returns ------- score : float Notes ----- * The long-standing behavior of this method changed in version 0.16. * It no longer uses the metric provided by ``estimator.score`` if the ``scoring`` parameter was set when fitting. """ if self.scorer_ is None: raise ValueError("No score function explicitly defined, " "and the estimator doesn't provide one %s" % self.best_estimator_) if self.scoring is not None and hasattr(self.best_estimator_, 'score'): warnings.warn("The long-standing behavior to use the estimator's " "score function in {0}.score has changed. The " "scoring parameter is now used." "".format(self.__class__.__name__), ChangedBehaviorWarning) return self.scorer_(self.best_estimator_, X, y) @if_delegate_has_method(delegate='estimator') def predict(self, X): """Call predict on the estimator with the best found parameters. Only available if ``refit=True`` and the underlying estimator supports ``predict``. Parameters ----------- X : indexable, length n_samples Must fulfill the input assumptions of the underlying estimator. """ return self.best_estimator_.predict(X) @if_delegate_has_method(delegate='estimator') def predict_proba(self, X): """Call predict_proba on the estimator with the best found parameters. Only available if ``refit=True`` and the underlying estimator supports ``predict_proba``. Parameters ----------- X : indexable, length n_samples Must fulfill the input assumptions of the underlying estimator. """ return self.best_estimator_.predict_proba(X) @if_delegate_has_method(delegate='estimator') def predict_log_proba(self, X): """Call predict_log_proba on the estimator with the best found parameters. Only available if ``refit=True`` and the underlying estimator supports ``predict_log_proba``. Parameters ----------- X : indexable, length n_samples Must fulfill the input assumptions of the underlying estimator. """ return self.best_estimator_.predict_log_proba(X) @if_delegate_has_method(delegate='estimator') def decision_function(self, X): """Call decision_function on the estimator with the best found parameters. Only available if ``refit=True`` and the underlying estimator supports ``decision_function``. Parameters ----------- X : indexable, length n_samples Must fulfill the input assumptions of the underlying estimator. """ return self.best_estimator_.decision_function(X) @if_delegate_has_method(delegate='estimator') def transform(self, X): """Call transform on the estimator with the best found parameters. Only available if the underlying estimator supports ``transform`` and ``refit=True``. Parameters ----------- X : indexable, length n_samples Must fulfill the input assumptions of the underlying estimator. """ return self.best_estimator_.transform(X) @if_delegate_has_method(delegate='estimator') def inverse_transform(self, Xt): """Call inverse_transform on the estimator with the best found parameters. Only available if the underlying estimator implements ``inverse_transform`` and ``refit=True``. Parameters ----------- Xt : indexable, length n_samples Must fulfill the input assumptions of the underlying estimator. """ return self.best_estimator_.transform(Xt) def _fit(self, X, y, parameter_iterable): """Actual fitting, performing the search over parameters.""" estimator = self.estimator cv = self.cv self.scorer_ = check_scoring(self.estimator, scoring=self.scoring) n_samples = _num_samples(X) X, y = indexable(X, y) if y is not None: if len(y) != n_samples: raise ValueError('Target variable (y) has a different number ' 'of samples (%i) than data (X: %i samples)' % (len(y), n_samples)) cv = check_cv(cv, X, y, classifier=is_classifier(estimator)) if self.verbose > 0: if isinstance(parameter_iterable, Sized): n_candidates = len(parameter_iterable) print("Fitting {0} folds for each of {1} candidates, totalling" " {2} fits".format(len(cv), n_candidates, n_candidates * len(cv))) base_estimator = clone(self.estimator) pre_dispatch = self.pre_dispatch out = Parallel( n_jobs=self.n_jobs, verbose=self.verbose, pre_dispatch=pre_dispatch )( delayed(_fit_and_score)(clone(base_estimator), X, y, self.scorer_, train, test, self.verbose, parameters, self.fit_params, return_parameters=True, error_score=self.error_score) for parameters in parameter_iterable for train, test in cv) # Out is a list of triplet: score, estimator, n_test_samples n_fits = len(out) n_folds = len(cv) scores = list() grid_scores = list() for grid_start in range(0, n_fits, n_folds): n_test_samples = 0 score = 0 all_scores = [] for this_score, this_n_test_samples, _, parameters in \ out[grid_start:grid_start + n_folds]: all_scores.append(this_score) if self.iid: this_score *= this_n_test_samples n_test_samples += this_n_test_samples score += this_score if self.iid: score /= float(n_test_samples) else: score /= float(n_folds) scores.append((score, parameters)) # TODO: shall we also store the test_fold_sizes? grid_scores.append(_CVScoreTuple( parameters, score, np.array(all_scores))) # Store the computed scores self.grid_scores_ = grid_scores # Find the best parameters by comparing on the mean validation score: # note that `sorted` is deterministic in the way it breaks ties best = sorted(grid_scores, key=lambda x: x.mean_validation_score, reverse=True)[0] self.best_params_ = best.parameters self.best_score_ = best.mean_validation_score if self.refit: # fit the best estimator using the entire dataset # clone first to work around broken estimators best_estimator = clone(base_estimator).set_params( **best.parameters) if y is not None: best_estimator.fit(X, y, **self.fit_params) else: best_estimator.fit(X, **self.fit_params) self.best_estimator_ = best_estimator return self class GridSearchCV(BaseSearchCV): """Exhaustive search over specified parameter values for an estimator. Important members are fit, predict. GridSearchCV implements a "fit" method and a "predict" method like any classifier except that the parameters of the classifier used to predict is optimized by cross-validation. Read more in the :ref:`User Guide <grid_search>`. Parameters ---------- estimator : object type that implements the "fit" and "predict" methods A object of that type is instantiated for each grid point. param_grid : dict or list of dictionaries Dictionary with parameters names (string) as keys and lists of parameter settings to try as values, or a list of such dictionaries, in which case the grids spanned by each dictionary in the list are explored. This enables searching over any sequence of parameter settings. scoring : string, callable or None, optional, default: None A string (see model evaluation documentation) or a scorer callable object / function with signature ``scorer(estimator, X, y)``. fit_params : dict, optional Parameters to pass to the fit method. n_jobs : int, default 1 Number of jobs to run in parallel. pre_dispatch : int, or string, optional Controls the number of jobs that get dispatched during parallel execution. Reducing this number can be useful to avoid an explosion of memory consumption when more jobs get dispatched than CPUs can process. This parameter can be: - None, in which case all the jobs are immediately created and spawned. Use this for lightweight and fast-running jobs, to avoid delays due to on-demand spawning of the jobs - An int, giving the exact number of total jobs that are spawned - A string, giving an expression as a function of n_jobs, as in '2*n_jobs' iid : boolean, default=True If True, the data is assumed to be identically distributed across the folds, and the loss minimized is the total loss per sample, and not the mean loss across the folds. cv : integer or cross-validation generator, default=3 If an integer is passed, it is the number of folds. Specific cross-validation objects can be passed, see sklearn.cross_validation module for the list of possible objects refit : boolean, default=True Refit the best estimator with the entire dataset. If "False", it is impossible to make predictions using this GridSearchCV instance after fitting. verbose : integer Controls the verbosity: the higher, the more messages. error_score : 'raise' (default) or numeric Value to assign to the score if an error occurs in estimator fitting. If set to 'raise', the error is raised. If a numeric value is given, FitFailedWarning is raised. This parameter does not affect the refit step, which will always raise the error. Examples -------- >>> from sklearn import svm, grid_search, datasets >>> iris = datasets.load_iris() >>> parameters = {'kernel':('linear', 'rbf'), 'C':[1, 10]} >>> svr = svm.SVC() >>> clf = grid_search.GridSearchCV(svr, parameters) >>> clf.fit(iris.data, iris.target) ... # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS GridSearchCV(cv=None, error_score=..., estimator=SVC(C=1.0, cache_size=..., class_weight=..., coef0=..., decision_function_shape=None, degree=..., gamma=..., kernel='rbf', max_iter=-1, probability=False, random_state=None, shrinking=True, tol=..., verbose=False), fit_params={}, iid=..., n_jobs=1, param_grid=..., pre_dispatch=..., refit=..., scoring=..., verbose=...) Attributes ---------- grid_scores_ : list of named tuples Contains scores for all parameter combinations in param_grid. Each entry corresponds to one parameter setting. Each named tuple has the attributes: * ``parameters``, a dict of parameter settings * ``mean_validation_score``, the mean score over the cross-validation folds * ``cv_validation_scores``, the list of scores for each fold best_estimator_ : estimator Estimator that was chosen by the search, i.e. estimator which gave highest score (or smallest loss if specified) on the left out data. Not available if refit=False. best_score_ : float Score of best_estimator on the left out data. best_params_ : dict Parameter setting that gave the best results on the hold out data. scorer_ : function Scorer function used on the held out data to choose the best parameters for the model. Notes ------ The parameters selected are those that maximize the score of the left out data, unless an explicit score is passed in which case it is used instead. If `n_jobs` was set to a value higher than one, the data is copied for each point in the grid (and not `n_jobs` times). This is done for efficiency reasons if individual jobs take very little time, but may raise errors if the dataset is large and not enough memory is available. A workaround in this case is to set `pre_dispatch`. Then, the memory is copied only `pre_dispatch` many times. A reasonable value for `pre_dispatch` is `2 * n_jobs`. See Also --------- :class:`ParameterGrid`: generates all the combinations of a an hyperparameter grid. :func:`sklearn.cross_validation.train_test_split`: utility function to split the data into a development set usable for fitting a GridSearchCV instance and an evaluation set for its final evaluation. :func:`sklearn.metrics.make_scorer`: Make a scorer from a performance metric or loss function. """ def __init__(self, estimator, param_grid, scoring=None, fit_params=None, n_jobs=1, iid=True, refit=True, cv=None, verbose=0, pre_dispatch='2*n_jobs', error_score='raise'): super(GridSearchCV, self).__init__( estimator, scoring, fit_params, n_jobs, iid, refit, cv, verbose, pre_dispatch, error_score) self.param_grid = param_grid _check_param_grid(param_grid) def fit(self, X, y=None): """Run fit with all sets of parameters. Parameters ---------- X : array-like, shape = [n_samples, n_features] Training vector, where n_samples is the number of samples and n_features is the number of features. y : array-like, shape = [n_samples] or [n_samples, n_output], optional Target relative to X for classification or regression; None for unsupervised learning. """ return self._fit(X, y, ParameterGrid(self.param_grid)) class RandomizedSearchCV(BaseSearchCV): """Randomized search on hyper parameters. RandomizedSearchCV implements a "fit" method and a "predict" method like any classifier except that the parameters of the classifier used to predict is optimized by cross-validation. In contrast to GridSearchCV, not all parameter values are tried out, but rather a fixed number of parameter settings is sampled from the specified distributions. The number of parameter settings that are tried is given by n_iter. If all parameters are presented as a list, sampling without replacement is performed. If at least one parameter is given as a distribution, sampling with replacement is used. It is highly recommended to use continuous distributions for continuous parameters. Read more in the :ref:`User Guide <randomized_parameter_search>`. Parameters ---------- estimator : object type that implements the "fit" and "predict" methods A object of that type is instantiated for each parameter setting. param_distributions : dict Dictionary with parameters names (string) as keys and distributions or lists of parameters to try. Distributions must provide a ``rvs`` method for sampling (such as those from scipy.stats.distributions). If a list is given, it is sampled uniformly. n_iter : int, default=10 Number of parameter settings that are sampled. n_iter trades off runtime vs quality of the solution. scoring : string, callable or None, optional, default: None A string (see model evaluation documentation) or a scorer callable object / function with signature ``scorer(estimator, X, y)``. fit_params : dict, optional Parameters to pass to the fit method. n_jobs : int, default=1 Number of jobs to run in parallel. pre_dispatch : int, or string, optional Controls the number of jobs that get dispatched during parallel execution. Reducing this number can be useful to avoid an explosion of memory consumption when more jobs get dispatched than CPUs can process. This parameter can be: - None, in which case all the jobs are immediately created and spawned. Use this for lightweight and fast-running jobs, to avoid delays due to on-demand spawning of the jobs - An int, giving the exact number of total jobs that are spawned - A string, giving an expression as a function of n_jobs, as in '2*n_jobs' iid : boolean, default=True If True, the data is assumed to be identically distributed across the folds, and the loss minimized is the total loss per sample, and not the mean loss across the folds. cv : integer or cross-validation generator, optional If an integer is passed, it is the number of folds (default 3). Specific cross-validation objects can be passed, see sklearn.cross_validation module for the list of possible objects refit : boolean, default=True Refit the best estimator with the entire dataset. If "False", it is impossible to make predictions using this RandomizedSearchCV instance after fitting. verbose : integer Controls the verbosity: the higher, the more messages. random_state : int or RandomState Pseudo random number generator state used for random uniform sampling from lists of possible values instead of scipy.stats distributions. error_score : 'raise' (default) or numeric Value to assign to the score if an error occurs in estimator fitting. If set to 'raise', the error is raised. If a numeric value is given, FitFailedWarning is raised. This parameter does not affect the refit step, which will always raise the error. Attributes ---------- grid_scores_ : list of named tuples Contains scores for all parameter combinations in param_grid. Each entry corresponds to one parameter setting. Each named tuple has the attributes: * ``parameters``, a dict of parameter settings * ``mean_validation_score``, the mean score over the cross-validation folds * ``cv_validation_scores``, the list of scores for each fold best_estimator_ : estimator Estimator that was chosen by the search, i.e. estimator which gave highest score (or smallest loss if specified) on the left out data. Not available if refit=False. best_score_ : float Score of best_estimator on the left out data. best_params_ : dict Parameter setting that gave the best results on the hold out data. Notes ----- The parameters selected are those that maximize the score of the held-out data, according to the scoring parameter. If `n_jobs` was set to a value higher than one, the data is copied for each parameter setting(and not `n_jobs` times). This is done for efficiency reasons if individual jobs take very little time, but may raise errors if the dataset is large and not enough memory is available. A workaround in this case is to set `pre_dispatch`. Then, the memory is copied only `pre_dispatch` many times. A reasonable value for `pre_dispatch` is `2 * n_jobs`. See Also -------- :class:`GridSearchCV`: Does exhaustive search over a grid of parameters. :class:`ParameterSampler`: A generator over parameter settins, constructed from param_distributions. """ def __init__(self, estimator, param_distributions, n_iter=10, scoring=None, fit_params=None, n_jobs=1, iid=True, refit=True, cv=None, verbose=0, pre_dispatch='2*n_jobs', random_state=None, error_score='raise'): self.param_distributions = param_distributions self.n_iter = n_iter self.random_state = random_state super(RandomizedSearchCV, self).__init__( estimator=estimator, scoring=scoring, fit_params=fit_params, n_jobs=n_jobs, iid=iid, refit=refit, cv=cv, verbose=verbose, pre_dispatch=pre_dispatch, error_score=error_score) def fit(self, X, y=None): """Run fit on the estimator with randomly drawn parameters. Parameters ---------- X : array-like, shape = [n_samples, n_features] Training vector, where n_samples in the number of samples and n_features is the number of features. y : array-like, shape = [n_samples] or [n_samples, n_output], optional Target relative to X for classification or regression; None for unsupervised learning. """ sampled_params = ParameterSampler(self.param_distributions, self.n_iter, random_state=self.random_state) return self._fit(X, y, sampled_params)
bsd-3-clause
luo66/scikit-learn
sklearn/ensemble/gradient_boosting.py
50
67625
"""Gradient Boosted Regression Trees This module contains methods for fitting gradient boosted regression trees for both classification and regression. The module structure is the following: - The ``BaseGradientBoosting`` base class implements a common ``fit`` method for all the estimators in the module. Regression and classification only differ in the concrete ``LossFunction`` used. - ``GradientBoostingClassifier`` implements gradient boosting for classification problems. - ``GradientBoostingRegressor`` implements gradient boosting for regression problems. """ # Authors: Peter Prettenhofer, Scott White, Gilles Louppe, Emanuele Olivetti, # Arnaud Joly, Jacob Schreiber # License: BSD 3 clause from __future__ import print_function from __future__ import division from abc import ABCMeta, abstractmethod from time import time import numbers import numpy as np from scipy import stats from .base import BaseEnsemble from ..base import BaseEstimator from ..base import ClassifierMixin from ..base import RegressorMixin from ..utils import check_random_state, check_array, check_X_y, column_or_1d from ..utils import check_consistent_length, deprecated from ..utils.extmath import logsumexp from ..utils.fixes import expit, bincount from ..utils.stats import _weighted_percentile from ..utils.validation import check_is_fitted, NotFittedError from ..externals import six from ..feature_selection.from_model import _LearntSelectorMixin from ..tree.tree import DecisionTreeRegressor from ..tree._tree import DTYPE, TREE_LEAF from ..tree._splitter import PresortBestSplitter from ..tree._criterion import FriedmanMSE from ._gradient_boosting import predict_stages from ._gradient_boosting import predict_stage from ._gradient_boosting import _random_sample_mask class QuantileEstimator(BaseEstimator): """An estimator predicting the alpha-quantile of the training targets.""" def __init__(self, alpha=0.9): if not 0 < alpha < 1.0: raise ValueError("`alpha` must be in (0, 1.0) but was %r" % alpha) self.alpha = alpha def fit(self, X, y, sample_weight=None): if sample_weight is None: self.quantile = stats.scoreatpercentile(y, self.alpha * 100.0) else: self.quantile = _weighted_percentile(y, sample_weight, self.alpha * 100.0) def predict(self, X): check_is_fitted(self, 'quantile') y = np.empty((X.shape[0], 1), dtype=np.float64) y.fill(self.quantile) return y class MeanEstimator(BaseEstimator): """An estimator predicting the mean of the training targets.""" def fit(self, X, y, sample_weight=None): if sample_weight is None: self.mean = np.mean(y) else: self.mean = np.average(y, weights=sample_weight) def predict(self, X): check_is_fitted(self, 'mean') y = np.empty((X.shape[0], 1), dtype=np.float64) y.fill(self.mean) return y class LogOddsEstimator(BaseEstimator): """An estimator predicting the log odds ratio.""" scale = 1.0 def fit(self, X, y, sample_weight=None): # pre-cond: pos, neg are encoded as 1, 0 if sample_weight is None: pos = np.sum(y) neg = y.shape[0] - pos else: pos = np.sum(sample_weight * y) neg = np.sum(sample_weight * (1 - y)) if neg == 0 or pos == 0: raise ValueError('y contains non binary labels.') self.prior = self.scale * np.log(pos / neg) def predict(self, X): check_is_fitted(self, 'prior') y = np.empty((X.shape[0], 1), dtype=np.float64) y.fill(self.prior) return y class ScaledLogOddsEstimator(LogOddsEstimator): """Log odds ratio scaled by 0.5 -- for exponential loss. """ scale = 0.5 class PriorProbabilityEstimator(BaseEstimator): """An estimator predicting the probability of each class in the training data. """ def fit(self, X, y, sample_weight=None): if sample_weight is None: sample_weight = np.ones_like(y, dtype=np.float64) class_counts = bincount(y, weights=sample_weight) self.priors = class_counts / class_counts.sum() def predict(self, X): check_is_fitted(self, 'priors') y = np.empty((X.shape[0], self.priors.shape[0]), dtype=np.float64) y[:] = self.priors return y class ZeroEstimator(BaseEstimator): """An estimator that simply predicts zero. """ def fit(self, X, y, sample_weight=None): if np.issubdtype(y.dtype, int): # classification self.n_classes = np.unique(y).shape[0] if self.n_classes == 2: self.n_classes = 1 else: # regression self.n_classes = 1 def predict(self, X): check_is_fitted(self, 'n_classes') y = np.empty((X.shape[0], self.n_classes), dtype=np.float64) y.fill(0.0) return y class LossFunction(six.with_metaclass(ABCMeta, object)): """Abstract base class for various loss functions. Attributes ---------- K : int The number of regression trees to be induced; 1 for regression and binary classification; ``n_classes`` for multi-class classification. """ is_multi_class = False def __init__(self, n_classes): self.K = n_classes def init_estimator(self): """Default ``init`` estimator for loss function. """ raise NotImplementedError() @abstractmethod def __call__(self, y, pred, sample_weight=None): """Compute the loss of prediction ``pred`` and ``y``. """ @abstractmethod def negative_gradient(self, y, y_pred, **kargs): """Compute the negative gradient. Parameters --------- y : np.ndarray, shape=(n,) The target labels. y_pred : np.ndarray, shape=(n,): The predictions. """ def update_terminal_regions(self, tree, X, y, residual, y_pred, sample_weight, sample_mask, learning_rate=1.0, k=0): """Update the terminal regions (=leaves) of the given tree and updates the current predictions of the model. Traverses tree and invokes template method `_update_terminal_region`. Parameters ---------- tree : tree.Tree The tree object. X : ndarray, shape=(n, m) The data array. y : ndarray, shape=(n,) The target labels. residual : ndarray, shape=(n,) The residuals (usually the negative gradient). y_pred : ndarray, shape=(n,) The predictions. sample_weight : ndarray, shape=(n,) The weight of each sample. sample_mask : ndarray, shape=(n,) The sample mask to be used. learning_rate : float, default=0.1 learning rate shrinks the contribution of each tree by ``learning_rate``. k : int, default 0 The index of the estimator being updated. """ # compute leaf for each sample in ``X``. terminal_regions = tree.apply(X) # mask all which are not in sample mask. masked_terminal_regions = terminal_regions.copy() masked_terminal_regions[~sample_mask] = -1 # update each leaf (= perform line search) for leaf in np.where(tree.children_left == TREE_LEAF)[0]: self._update_terminal_region(tree, masked_terminal_regions, leaf, X, y, residual, y_pred[:, k], sample_weight) # update predictions (both in-bag and out-of-bag) y_pred[:, k] += (learning_rate * tree.value[:, 0, 0].take(terminal_regions, axis=0)) @abstractmethod def _update_terminal_region(self, tree, terminal_regions, leaf, X, y, residual, pred, sample_weight): """Template method for updating terminal regions (=leaves). """ class RegressionLossFunction(six.with_metaclass(ABCMeta, LossFunction)): """Base class for regression loss functions. """ def __init__(self, n_classes): if n_classes != 1: raise ValueError("``n_classes`` must be 1 for regression but " "was %r" % n_classes) super(RegressionLossFunction, self).__init__(n_classes) class LeastSquaresError(RegressionLossFunction): """Loss function for least squares (LS) estimation. Terminal regions need not to be updated for least squares. """ def init_estimator(self): return MeanEstimator() def __call__(self, y, pred, sample_weight=None): if sample_weight is None: return np.mean((y - pred.ravel()) ** 2.0) else: return (1.0 / sample_weight.sum() * np.sum(sample_weight * ((y - pred.ravel()) ** 2.0))) def negative_gradient(self, y, pred, **kargs): return y - pred.ravel() def update_terminal_regions(self, tree, X, y, residual, y_pred, sample_weight, sample_mask, learning_rate=1.0, k=0): """Least squares does not need to update terminal regions. But it has to update the predictions. """ # update predictions y_pred[:, k] += learning_rate * tree.predict(X).ravel() def _update_terminal_region(self, tree, terminal_regions, leaf, X, y, residual, pred, sample_weight): pass class LeastAbsoluteError(RegressionLossFunction): """Loss function for least absolute deviation (LAD) regression. """ def init_estimator(self): return QuantileEstimator(alpha=0.5) def __call__(self, y, pred, sample_weight=None): if sample_weight is None: return np.abs(y - pred.ravel()).mean() else: return (1.0 / sample_weight.sum() * np.sum(sample_weight * np.abs(y - pred.ravel()))) def negative_gradient(self, y, pred, **kargs): """1.0 if y - pred > 0.0 else -1.0""" pred = pred.ravel() return 2.0 * (y - pred > 0.0) - 1.0 def _update_terminal_region(self, tree, terminal_regions, leaf, X, y, residual, pred, sample_weight): """LAD updates terminal regions to median estimates. """ terminal_region = np.where(terminal_regions == leaf)[0] sample_weight = sample_weight.take(terminal_region, axis=0) diff = y.take(terminal_region, axis=0) - pred.take(terminal_region, axis=0) tree.value[leaf, 0, 0] = _weighted_percentile(diff, sample_weight, percentile=50) class HuberLossFunction(RegressionLossFunction): """Huber loss function for robust regression. M-Regression proposed in Friedman 2001. References ---------- J. Friedman, Greedy Function Approximation: A Gradient Boosting Machine, The Annals of Statistics, Vol. 29, No. 5, 2001. """ def __init__(self, n_classes, alpha=0.9): super(HuberLossFunction, self).__init__(n_classes) self.alpha = alpha self.gamma = None def init_estimator(self): return QuantileEstimator(alpha=0.5) def __call__(self, y, pred, sample_weight=None): pred = pred.ravel() diff = y - pred gamma = self.gamma if gamma is None: if sample_weight is None: gamma = stats.scoreatpercentile(np.abs(diff), self.alpha * 100) else: gamma = _weighted_percentile(np.abs(diff), sample_weight, self.alpha * 100) gamma_mask = np.abs(diff) <= gamma if sample_weight is None: sq_loss = np.sum(0.5 * diff[gamma_mask] ** 2.0) lin_loss = np.sum(gamma * (np.abs(diff[~gamma_mask]) - gamma / 2.0)) loss = (sq_loss + lin_loss) / y.shape[0] else: sq_loss = np.sum(0.5 * sample_weight[gamma_mask] * diff[gamma_mask] ** 2.0) lin_loss = np.sum(gamma * sample_weight[~gamma_mask] * (np.abs(diff[~gamma_mask]) - gamma / 2.0)) loss = (sq_loss + lin_loss) / sample_weight.sum() return loss def negative_gradient(self, y, pred, sample_weight=None, **kargs): pred = pred.ravel() diff = y - pred if sample_weight is None: gamma = stats.scoreatpercentile(np.abs(diff), self.alpha * 100) else: gamma = _weighted_percentile(np.abs(diff), sample_weight, self.alpha * 100) gamma_mask = np.abs(diff) <= gamma residual = np.zeros((y.shape[0],), dtype=np.float64) residual[gamma_mask] = diff[gamma_mask] residual[~gamma_mask] = gamma * np.sign(diff[~gamma_mask]) self.gamma = gamma return residual def _update_terminal_region(self, tree, terminal_regions, leaf, X, y, residual, pred, sample_weight): terminal_region = np.where(terminal_regions == leaf)[0] sample_weight = sample_weight.take(terminal_region, axis=0) gamma = self.gamma diff = (y.take(terminal_region, axis=0) - pred.take(terminal_region, axis=0)) median = _weighted_percentile(diff, sample_weight, percentile=50) diff_minus_median = diff - median tree.value[leaf, 0] = median + np.mean( np.sign(diff_minus_median) * np.minimum(np.abs(diff_minus_median), gamma)) class QuantileLossFunction(RegressionLossFunction): """Loss function for quantile regression. Quantile regression allows to estimate the percentiles of the conditional distribution of the target. """ def __init__(self, n_classes, alpha=0.9): super(QuantileLossFunction, self).__init__(n_classes) assert 0 < alpha < 1.0 self.alpha = alpha self.percentile = alpha * 100.0 def init_estimator(self): return QuantileEstimator(self.alpha) def __call__(self, y, pred, sample_weight=None): pred = pred.ravel() diff = y - pred alpha = self.alpha mask = y > pred if sample_weight is None: loss = (alpha * diff[mask].sum() + (1.0 - alpha) * diff[~mask].sum()) / y.shape[0] else: loss = ((alpha * np.sum(sample_weight[mask] * diff[mask]) + (1.0 - alpha) * np.sum(sample_weight[~mask] * diff[~mask])) / sample_weight.sum()) return loss def negative_gradient(self, y, pred, **kargs): alpha = self.alpha pred = pred.ravel() mask = y > pred return (alpha * mask) - ((1.0 - alpha) * ~mask) def _update_terminal_region(self, tree, terminal_regions, leaf, X, y, residual, pred, sample_weight): terminal_region = np.where(terminal_regions == leaf)[0] diff = (y.take(terminal_region, axis=0) - pred.take(terminal_region, axis=0)) sample_weight = sample_weight.take(terminal_region, axis=0) val = _weighted_percentile(diff, sample_weight, self.percentile) tree.value[leaf, 0] = val class ClassificationLossFunction(six.with_metaclass(ABCMeta, LossFunction)): """Base class for classification loss functions. """ def _score_to_proba(self, score): """Template method to convert scores to probabilities. the does not support probabilites raises AttributeError. """ raise TypeError('%s does not support predict_proba' % type(self).__name__) @abstractmethod def _score_to_decision(self, score): """Template method to convert scores to decisions. Returns int arrays. """ class BinomialDeviance(ClassificationLossFunction): """Binomial deviance loss function for binary classification. Binary classification is a special case; here, we only need to fit one tree instead of ``n_classes`` trees. """ def __init__(self, n_classes): if n_classes != 2: raise ValueError("{0:s} requires 2 classes.".format( self.__class__.__name__)) # we only need to fit one tree for binary clf. super(BinomialDeviance, self).__init__(1) def init_estimator(self): return LogOddsEstimator() def __call__(self, y, pred, sample_weight=None): """Compute the deviance (= 2 * negative log-likelihood). """ # logaddexp(0, v) == log(1.0 + exp(v)) pred = pred.ravel() if sample_weight is None: return -2.0 * np.mean((y * pred) - np.logaddexp(0.0, pred)) else: return (-2.0 / sample_weight.sum() * np.sum(sample_weight * ((y * pred) - np.logaddexp(0.0, pred)))) def negative_gradient(self, y, pred, **kargs): """Compute the residual (= negative gradient). """ return y - expit(pred.ravel()) def _update_terminal_region(self, tree, terminal_regions, leaf, X, y, residual, pred, sample_weight): """Make a single Newton-Raphson step. our node estimate is given by: sum(w * (y - prob)) / sum(w * prob * (1 - prob)) we take advantage that: y - prob = residual """ terminal_region = np.where(terminal_regions == leaf)[0] residual = residual.take(terminal_region, axis=0) y = y.take(terminal_region, axis=0) sample_weight = sample_weight.take(terminal_region, axis=0) numerator = np.sum(sample_weight * residual) denominator = np.sum(sample_weight * (y - residual) * (1 - y + residual)) if denominator == 0.0: tree.value[leaf, 0, 0] = 0.0 else: tree.value[leaf, 0, 0] = numerator / denominator def _score_to_proba(self, score): proba = np.ones((score.shape[0], 2), dtype=np.float64) proba[:, 1] = expit(score.ravel()) proba[:, 0] -= proba[:, 1] return proba def _score_to_decision(self, score): proba = self._score_to_proba(score) return np.argmax(proba, axis=1) class MultinomialDeviance(ClassificationLossFunction): """Multinomial deviance loss function for multi-class classification. For multi-class classification we need to fit ``n_classes`` trees at each stage. """ is_multi_class = True def __init__(self, n_classes): if n_classes < 3: raise ValueError("{0:s} requires more than 2 classes.".format( self.__class__.__name__)) super(MultinomialDeviance, self).__init__(n_classes) def init_estimator(self): return PriorProbabilityEstimator() def __call__(self, y, pred, sample_weight=None): # create one-hot label encoding Y = np.zeros((y.shape[0], self.K), dtype=np.float64) for k in range(self.K): Y[:, k] = y == k if sample_weight is None: return np.sum(-1 * (Y * pred).sum(axis=1) + logsumexp(pred, axis=1)) else: return np.sum(-1 * sample_weight * (Y * pred).sum(axis=1) + logsumexp(pred, axis=1)) def negative_gradient(self, y, pred, k=0, **kwargs): """Compute negative gradient for the ``k``-th class. """ return y - np.nan_to_num(np.exp(pred[:, k] - logsumexp(pred, axis=1))) def _update_terminal_region(self, tree, terminal_regions, leaf, X, y, residual, pred, sample_weight): """Make a single Newton-Raphson step. """ terminal_region = np.where(terminal_regions == leaf)[0] residual = residual.take(terminal_region, axis=0) y = y.take(terminal_region, axis=0) sample_weight = sample_weight.take(terminal_region, axis=0) numerator = np.sum(sample_weight * residual) numerator *= (self.K - 1) / self.K denominator = np.sum(sample_weight * (y - residual) * (1.0 - y + residual)) if denominator == 0.0: tree.value[leaf, 0, 0] = 0.0 else: tree.value[leaf, 0, 0] = numerator / denominator def _score_to_proba(self, score): return np.nan_to_num( np.exp(score - (logsumexp(score, axis=1)[:, np.newaxis]))) def _score_to_decision(self, score): proba = self._score_to_proba(score) return np.argmax(proba, axis=1) class ExponentialLoss(ClassificationLossFunction): """Exponential loss function for binary classification. Same loss as AdaBoost. References ---------- Greg Ridgeway, Generalized Boosted Models: A guide to the gbm package, 2007 """ def __init__(self, n_classes): if n_classes != 2: raise ValueError("{0:s} requires 2 classes.".format( self.__class__.__name__)) # we only need to fit one tree for binary clf. super(ExponentialLoss, self).__init__(1) def init_estimator(self): return ScaledLogOddsEstimator() def __call__(self, y, pred, sample_weight=None): pred = pred.ravel() if sample_weight is None: return np.mean(np.exp(-(2. * y - 1.) * pred)) else: return (1.0 / sample_weight.sum() * np.sum(sample_weight * np.exp(-(2 * y - 1) * pred))) def negative_gradient(self, y, pred, **kargs): y_ = -(2. * y - 1.) return y_ * np.exp(y_ * pred.ravel()) def _update_terminal_region(self, tree, terminal_regions, leaf, X, y, residual, pred, sample_weight): terminal_region = np.where(terminal_regions == leaf)[0] pred = pred.take(terminal_region, axis=0) y = y.take(terminal_region, axis=0) sample_weight = sample_weight.take(terminal_region, axis=0) y_ = 2. * y - 1. numerator = np.sum(y_ * sample_weight * np.exp(-y_ * pred)) denominator = np.sum(sample_weight * np.exp(-y_ * pred)) if denominator == 0.0: tree.value[leaf, 0, 0] = 0.0 else: tree.value[leaf, 0, 0] = numerator / denominator def _score_to_proba(self, score): proba = np.ones((score.shape[0], 2), dtype=np.float64) proba[:, 1] = expit(2.0 * score.ravel()) proba[:, 0] -= proba[:, 1] return proba def _score_to_decision(self, score): return (score.ravel() >= 0.0).astype(np.int) LOSS_FUNCTIONS = {'ls': LeastSquaresError, 'lad': LeastAbsoluteError, 'huber': HuberLossFunction, 'quantile': QuantileLossFunction, 'deviance': None, # for both, multinomial and binomial 'exponential': ExponentialLoss, } INIT_ESTIMATORS = {'zero': ZeroEstimator} class VerboseReporter(object): """Reports verbose output to stdout. If ``verbose==1`` output is printed once in a while (when iteration mod verbose_mod is zero).; if larger than 1 then output is printed for each update. """ def __init__(self, verbose): self.verbose = verbose def init(self, est, begin_at_stage=0): # header fields and line format str header_fields = ['Iter', 'Train Loss'] verbose_fmt = ['{iter:>10d}', '{train_score:>16.4f}'] # do oob? if est.subsample < 1: header_fields.append('OOB Improve') verbose_fmt.append('{oob_impr:>16.4f}') header_fields.append('Remaining Time') verbose_fmt.append('{remaining_time:>16s}') # print the header line print(('%10s ' + '%16s ' * (len(header_fields) - 1)) % tuple(header_fields)) self.verbose_fmt = ' '.join(verbose_fmt) # plot verbose info each time i % verbose_mod == 0 self.verbose_mod = 1 self.start_time = time() self.begin_at_stage = begin_at_stage def update(self, j, est): """Update reporter with new iteration. """ do_oob = est.subsample < 1 # we need to take into account if we fit additional estimators. i = j - self.begin_at_stage # iteration relative to the start iter if (i + 1) % self.verbose_mod == 0: oob_impr = est.oob_improvement_[j] if do_oob else 0 remaining_time = ((est.n_estimators - (j + 1)) * (time() - self.start_time) / float(i + 1)) if remaining_time > 60: remaining_time = '{0:.2f}m'.format(remaining_time / 60.0) else: remaining_time = '{0:.2f}s'.format(remaining_time) print(self.verbose_fmt.format(iter=j + 1, train_score=est.train_score_[j], oob_impr=oob_impr, remaining_time=remaining_time)) if self.verbose == 1 and ((i + 1) // (self.verbose_mod * 10) > 0): # adjust verbose frequency (powers of 10) self.verbose_mod *= 10 class BaseGradientBoosting(six.with_metaclass(ABCMeta, BaseEnsemble, _LearntSelectorMixin)): """Abstract base class for Gradient Boosting. """ @abstractmethod def __init__(self, loss, learning_rate, n_estimators, min_samples_split, min_samples_leaf, min_weight_fraction_leaf, max_depth, init, subsample, max_features, random_state, alpha=0.9, verbose=0, max_leaf_nodes=None, warm_start=False): self.n_estimators = n_estimators self.learning_rate = learning_rate self.loss = loss self.min_samples_split = min_samples_split self.min_samples_leaf = min_samples_leaf self.min_weight_fraction_leaf = min_weight_fraction_leaf self.subsample = subsample self.max_features = max_features self.max_depth = max_depth self.init = init self.random_state = random_state self.alpha = alpha self.verbose = verbose self.max_leaf_nodes = max_leaf_nodes self.warm_start = warm_start self.estimators_ = np.empty((0, 0), dtype=np.object) def _fit_stage(self, i, X, y, y_pred, sample_weight, sample_mask, criterion, splitter, random_state): """Fit another stage of ``n_classes_`` trees to the boosting model. """ assert sample_mask.dtype == np.bool loss = self.loss_ original_y = y for k in range(loss.K): if loss.is_multi_class: y = np.array(original_y == k, dtype=np.float64) residual = loss.negative_gradient(y, y_pred, k=k, sample_weight=sample_weight) # induce regression tree on residuals tree = DecisionTreeRegressor( criterion=criterion, splitter=splitter, max_depth=self.max_depth, min_samples_split=self.min_samples_split, min_samples_leaf=self.min_samples_leaf, min_weight_fraction_leaf=self.min_weight_fraction_leaf, max_features=self.max_features, max_leaf_nodes=self.max_leaf_nodes, random_state=random_state) if self.subsample < 1.0: # no inplace multiplication! sample_weight = sample_weight * sample_mask.astype(np.float64) tree.fit(X, residual, sample_weight=sample_weight, check_input=False) # update tree leaves loss.update_terminal_regions(tree.tree_, X, y, residual, y_pred, sample_weight, sample_mask, self.learning_rate, k=k) # add tree to ensemble self.estimators_[i, k] = tree return y_pred def _check_params(self): """Check validity of parameters and raise ValueError if not valid. """ if self.n_estimators <= 0: raise ValueError("n_estimators must be greater than 0 but " "was %r" % self.n_estimators) if self.learning_rate <= 0.0: raise ValueError("learning_rate must be greater than 0 but " "was %r" % self.learning_rate) if (self.loss not in self._SUPPORTED_LOSS or self.loss not in LOSS_FUNCTIONS): raise ValueError("Loss '{0:s}' not supported. ".format(self.loss)) if self.loss == 'deviance': loss_class = (MultinomialDeviance if len(self.classes_) > 2 else BinomialDeviance) else: loss_class = LOSS_FUNCTIONS[self.loss] if self.loss in ('huber', 'quantile'): self.loss_ = loss_class(self.n_classes_, self.alpha) else: self.loss_ = loss_class(self.n_classes_) if not (0.0 < self.subsample <= 1.0): raise ValueError("subsample must be in (0,1] but " "was %r" % self.subsample) if self.init is not None: if isinstance(self.init, six.string_types): if self.init not in INIT_ESTIMATORS: raise ValueError('init="%s" is not supported' % self.init) else: if (not hasattr(self.init, 'fit') or not hasattr(self.init, 'predict')): raise ValueError("init=%r must be valid BaseEstimator " "and support both fit and " "predict" % self.init) if not (0.0 < self.alpha < 1.0): raise ValueError("alpha must be in (0.0, 1.0) but " "was %r" % self.alpha) if isinstance(self.max_features, six.string_types): if self.max_features == "auto": # if is_classification if self.n_classes_ > 1: max_features = max(1, int(np.sqrt(self.n_features))) else: # is regression max_features = self.n_features elif self.max_features == "sqrt": max_features = max(1, int(np.sqrt(self.n_features))) elif self.max_features == "log2": max_features = max(1, int(np.log2(self.n_features))) else: raise ValueError("Invalid value for max_features: %r. " "Allowed string values are 'auto', 'sqrt' " "or 'log2'." % self.max_features) elif self.max_features is None: max_features = self.n_features elif isinstance(self.max_features, (numbers.Integral, np.integer)): max_features = self.max_features else: # float if 0. < self.max_features <= 1.: max_features = max(int(self.max_features * self.n_features), 1) else: raise ValueError("max_features must be in (0, n_features]") self.max_features_ = max_features def _init_state(self): """Initialize model state and allocate model state data structures. """ if self.init is None: self.init_ = self.loss_.init_estimator() elif isinstance(self.init, six.string_types): self.init_ = INIT_ESTIMATORS[self.init]() else: self.init_ = self.init self.estimators_ = np.empty((self.n_estimators, self.loss_.K), dtype=np.object) self.train_score_ = np.zeros((self.n_estimators,), dtype=np.float64) # do oob? if self.subsample < 1.0: self.oob_improvement_ = np.zeros((self.n_estimators), dtype=np.float64) def _clear_state(self): """Clear the state of the gradient boosting model. """ if hasattr(self, 'estimators_'): self.estimators_ = np.empty((0, 0), dtype=np.object) if hasattr(self, 'train_score_'): del self.train_score_ if hasattr(self, 'oob_improvement_'): del self.oob_improvement_ if hasattr(self, 'init_'): del self.init_ def _resize_state(self): """Add additional ``n_estimators`` entries to all attributes. """ # self.n_estimators is the number of additional est to fit total_n_estimators = self.n_estimators if total_n_estimators < self.estimators_.shape[0]: raise ValueError('resize with smaller n_estimators %d < %d' % (total_n_estimators, self.estimators_[0])) self.estimators_.resize((total_n_estimators, self.loss_.K)) self.train_score_.resize(total_n_estimators) if (self.subsample < 1 or hasattr(self, 'oob_improvement_')): # if do oob resize arrays or create new if not available if hasattr(self, 'oob_improvement_'): self.oob_improvement_.resize(total_n_estimators) else: self.oob_improvement_ = np.zeros((total_n_estimators,), dtype=np.float64) def _is_initialized(self): return len(getattr(self, 'estimators_', [])) > 0 def _check_initialized(self): """Check that the estimator is initialized, raising an error if not.""" if self.estimators_ is None or len(self.estimators_) == 0: raise NotFittedError("Estimator not fitted, call `fit`" " before making predictions`.") def fit(self, X, y, sample_weight=None, monitor=None): """Fit the gradient boosting model. Parameters ---------- X : array-like, shape = [n_samples, n_features] Training vectors, where n_samples is the number of samples and n_features is the number of features. y : array-like, shape = [n_samples] Target values (integers in classification, real numbers in regression) For classification, labels must correspond to classes. sample_weight : array-like, shape = [n_samples] or None Sample weights. If None, then samples are equally weighted. Splits that would create child nodes with net zero or negative weight are ignored while searching for a split in each node. In the case of classification, splits are also ignored if they would result in any single class carrying a negative weight in either child node. monitor : callable, optional The monitor is called after each iteration with the current iteration, a reference to the estimator and the local variables of ``_fit_stages`` as keyword arguments ``callable(i, self, locals())``. If the callable returns ``True`` the fitting procedure is stopped. The monitor can be used for various things such as computing held-out estimates, early stopping, model introspect, and snapshoting. Returns ------- self : object Returns self. """ # if not warmstart - clear the estimator state if not self.warm_start: self._clear_state() # Check input X, y = check_X_y(X, y, dtype=DTYPE) n_samples, self.n_features = X.shape if sample_weight is None: sample_weight = np.ones(n_samples, dtype=np.float32) else: sample_weight = column_or_1d(sample_weight, warn=True) check_consistent_length(X, y, sample_weight) y = self._validate_y(y) random_state = check_random_state(self.random_state) self._check_params() if not self._is_initialized(): # init state self._init_state() # fit initial model - FIXME make sample_weight optional self.init_.fit(X, y, sample_weight) # init predictions y_pred = self.init_.predict(X) begin_at_stage = 0 else: # add more estimators to fitted model # invariant: warm_start = True if self.n_estimators < self.estimators_.shape[0]: raise ValueError('n_estimators=%d must be larger or equal to ' 'estimators_.shape[0]=%d when ' 'warm_start==True' % (self.n_estimators, self.estimators_.shape[0])) begin_at_stage = self.estimators_.shape[0] y_pred = self._decision_function(X) self._resize_state() # fit the boosting stages n_stages = self._fit_stages(X, y, y_pred, sample_weight, random_state, begin_at_stage, monitor) # change shape of arrays after fit (early-stopping or additional ests) if n_stages != self.estimators_.shape[0]: self.estimators_ = self.estimators_[:n_stages] self.train_score_ = self.train_score_[:n_stages] if hasattr(self, 'oob_improvement_'): self.oob_improvement_ = self.oob_improvement_[:n_stages] return self def _fit_stages(self, X, y, y_pred, sample_weight, random_state, begin_at_stage=0, monitor=None): """Iteratively fits the stages. For each stage it computes the progress (OOB, train score) and delegates to ``_fit_stage``. Returns the number of stages fit; might differ from ``n_estimators`` due to early stopping. """ n_samples = X.shape[0] do_oob = self.subsample < 1.0 sample_mask = np.ones((n_samples, ), dtype=np.bool) n_inbag = max(1, int(self.subsample * n_samples)) loss_ = self.loss_ # Set min_weight_leaf from min_weight_fraction_leaf if self.min_weight_fraction_leaf != 0. and sample_weight is not None: min_weight_leaf = (self.min_weight_fraction_leaf * np.sum(sample_weight)) else: min_weight_leaf = 0. # init criterion and splitter criterion = FriedmanMSE(1) splitter = PresortBestSplitter(criterion, self.max_features_, self.min_samples_leaf, min_weight_leaf, random_state) if self.verbose: verbose_reporter = VerboseReporter(self.verbose) verbose_reporter.init(self, begin_at_stage) # perform boosting iterations i = begin_at_stage for i in range(begin_at_stage, self.n_estimators): # subsampling if do_oob: sample_mask = _random_sample_mask(n_samples, n_inbag, random_state) # OOB score before adding this stage old_oob_score = loss_(y[~sample_mask], y_pred[~sample_mask], sample_weight[~sample_mask]) # fit next stage of trees y_pred = self._fit_stage(i, X, y, y_pred, sample_weight, sample_mask, criterion, splitter, random_state) # track deviance (= loss) if do_oob: self.train_score_[i] = loss_(y[sample_mask], y_pred[sample_mask], sample_weight[sample_mask]) self.oob_improvement_[i] = ( old_oob_score - loss_(y[~sample_mask], y_pred[~sample_mask], sample_weight[~sample_mask])) else: # no need to fancy index w/ no subsampling self.train_score_[i] = loss_(y, y_pred, sample_weight) if self.verbose > 0: verbose_reporter.update(i, self) if monitor is not None: early_stopping = monitor(i, self, locals()) if early_stopping: break return i + 1 def _make_estimator(self, append=True): # we don't need _make_estimator raise NotImplementedError() def _init_decision_function(self, X): """Check input and compute prediction of ``init``. """ self._check_initialized() if X.shape[1] != self.n_features: raise ValueError("X.shape[1] should be {0:d}, not {1:d}.".format( self.n_features, X.shape[1])) score = self.init_.predict(X).astype(np.float64) return score def _decision_function(self, X): # for use in inner loop, not raveling the output in single-class case, # not doing input validation. score = self._init_decision_function(X) predict_stages(self.estimators_, X, self.learning_rate, score) return score @deprecated(" and will be removed in 0.19") def decision_function(self, X): """Compute the decision function of ``X``. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- score : array, shape = [n_samples, n_classes] or [n_samples] The decision function of the input samples. The order of the classes corresponds to that in the attribute `classes_`. Regression and binary classification produce an array of shape [n_samples]. """ X = check_array(X, dtype=DTYPE, order="C") score = self._decision_function(X) if score.shape[1] == 1: return score.ravel() return score def _staged_decision_function(self, X): """Compute decision function of ``X`` for each iteration. This method allows monitoring (i.e. determine error on testing set) after each stage. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- score : generator of array, shape = [n_samples, k] The decision function of the input samples. The order of the classes corresponds to that in the attribute `classes_`. Regression and binary classification are special cases with ``k == 1``, otherwise ``k==n_classes``. """ X = check_array(X, dtype=DTYPE, order="C") score = self._init_decision_function(X) for i in range(self.estimators_.shape[0]): predict_stage(self.estimators_, i, X, self.learning_rate, score) yield score.copy() @deprecated(" and will be removed in 0.19") def staged_decision_function(self, X): """Compute decision function of ``X`` for each iteration. This method allows monitoring (i.e. determine error on testing set) after each stage. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- score : generator of array, shape = [n_samples, k] The decision function of the input samples. The order of the classes corresponds to that in the attribute `classes_`. Regression and binary classification are special cases with ``k == 1``, otherwise ``k==n_classes``. """ for dec in self._staged_decision_function(X): # no yield from in Python2.X yield dec @property def feature_importances_(self): """Return the feature importances (the higher, the more important the feature). Returns ------- feature_importances_ : array, shape = [n_features] """ self._check_initialized() total_sum = np.zeros((self.n_features, ), dtype=np.float64) for stage in self.estimators_: stage_sum = sum(tree.feature_importances_ for tree in stage) / len(stage) total_sum += stage_sum importances = total_sum / len(self.estimators_) return importances def _validate_y(self, y): self.n_classes_ = 1 if y.dtype.kind == 'O': y = y.astype(np.float64) # Default implementation return y def apply(self, X): """Apply trees in the ensemble to X, return leaf indices. Parameters ---------- X : array-like or sparse matrix, shape = [n_samples, n_features] The input samples. Internally, it will be converted to ``dtype=np.float32`` and if a sparse matrix is provided to a sparse ``csr_matrix``. Returns ------- X_leaves : array_like, shape = [n_samples, n_estimators, n_classes] For each datapoint x in X and for each tree in the ensemble, return the index of the leaf x ends up in in each estimator. In the case of binary classification n_classes is 1. """ self._check_initialized() X = self.estimators_[0, 0]._validate_X_predict(X, check_input=True) # n_classes will be equal to 1 in the binary classification or the # regression case. n_estimators, n_classes = self.estimators_.shape leaves = np.zeros((X.shape[0], n_estimators, n_classes)) for i in range(n_estimators): for j in range(n_classes): estimator = self.estimators_[i, j] leaves[:, i, j] = estimator.apply(X, check_input=False) return leaves class GradientBoostingClassifier(BaseGradientBoosting, ClassifierMixin): """Gradient Boosting for classification. GB builds an additive model in a forward stage-wise fashion; it allows for the optimization of arbitrary differentiable loss functions. In each stage ``n_classes_`` regression trees are fit on the negative gradient of the binomial or multinomial deviance loss function. Binary classification is a special case where only a single regression tree is induced. Read more in the :ref:`User Guide <gradient_boosting>`. Parameters ---------- loss : {'deviance', 'exponential'}, optional (default='deviance') loss function to be optimized. 'deviance' refers to deviance (= logistic regression) for classification with probabilistic outputs. For loss 'exponential' gradient boosting recovers the AdaBoost algorithm. learning_rate : float, optional (default=0.1) learning rate shrinks the contribution of each tree by `learning_rate`. There is a trade-off between learning_rate and n_estimators. n_estimators : int (default=100) The number of boosting stages to perform. Gradient boosting is fairly robust to over-fitting so a large number usually results in better performance. max_depth : integer, optional (default=3) maximum depth of the individual regression estimators. The maximum depth limits the number of nodes in the tree. Tune this parameter for best performance; the best value depends on the interaction of the input variables. Ignored if ``max_leaf_nodes`` is not None. min_samples_split : integer, optional (default=2) The minimum number of samples required to split an internal node. min_samples_leaf : integer, optional (default=1) The minimum number of samples required to be at a leaf node. min_weight_fraction_leaf : float, optional (default=0.) The minimum weighted fraction of the input samples required to be at a leaf node. subsample : float, optional (default=1.0) The fraction of samples to be used for fitting the individual base learners. If smaller than 1.0 this results in Stochastic Gradient Boosting. `subsample` interacts with the parameter `n_estimators`. Choosing `subsample < 1.0` leads to a reduction of variance and an increase in bias. max_features : int, float, string or None, optional (default=None) The number of features to consider when looking for the best split: - If int, then consider `max_features` features at each split. - If float, then `max_features` is a percentage and `int(max_features * n_features)` features are considered at each split. - If "auto", then `max_features=sqrt(n_features)`. - If "sqrt", then `max_features=sqrt(n_features)`. - If "log2", then `max_features=log2(n_features)`. - If None, then `max_features=n_features`. Choosing `max_features < n_features` leads to a reduction of variance and an increase in bias. Note: the search for a split does not stop until at least one valid partition of the node samples is found, even if it requires to effectively inspect more than ``max_features`` features. max_leaf_nodes : int or None, optional (default=None) Grow trees with ``max_leaf_nodes`` in best-first fashion. Best nodes are defined as relative reduction in impurity. If None then unlimited number of leaf nodes. If not None then ``max_depth`` will be ignored. init : BaseEstimator, None, optional (default=None) An estimator object that is used to compute the initial predictions. ``init`` has to provide ``fit`` and ``predict``. If None it uses ``loss.init_estimator``. verbose : int, default: 0 Enable verbose output. If 1 then it prints progress and performance once in a while (the more trees the lower the frequency). If greater than 1 then it prints progress and performance for every tree. warm_start : bool, default: False When set to ``True``, reuse the solution of the previous call to fit and add more estimators to the ensemble, otherwise, just erase the previous solution. random_state : int, RandomState instance or None, optional (default=None) If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. Attributes ---------- feature_importances_ : array, shape = [n_features] The feature importances (the higher, the more important the feature). oob_improvement_ : array, shape = [n_estimators] The improvement in loss (= deviance) on the out-of-bag samples relative to the previous iteration. ``oob_improvement_[0]`` is the improvement in loss of the first stage over the ``init`` estimator. train_score_ : array, shape = [n_estimators] The i-th score ``train_score_[i]`` is the deviance (= loss) of the model at iteration ``i`` on the in-bag sample. If ``subsample == 1`` this is the deviance on the training data. loss_ : LossFunction The concrete ``LossFunction`` object. init : BaseEstimator The estimator that provides the initial predictions. Set via the ``init`` argument or ``loss.init_estimator``. estimators_ : ndarray of DecisionTreeRegressor, shape = [n_estimators, ``loss_.K``] The collection of fitted sub-estimators. ``loss_.K`` is 1 for binary classification, otherwise n_classes. See also -------- sklearn.tree.DecisionTreeClassifier, RandomForestClassifier AdaBoostClassifier References ---------- J. Friedman, Greedy Function Approximation: A Gradient Boosting Machine, The Annals of Statistics, Vol. 29, No. 5, 2001. J. Friedman, Stochastic Gradient Boosting, 1999 T. Hastie, R. Tibshirani and J. Friedman. Elements of Statistical Learning Ed. 2, Springer, 2009. """ _SUPPORTED_LOSS = ('deviance', 'exponential') def __init__(self, loss='deviance', learning_rate=0.1, n_estimators=100, subsample=1.0, min_samples_split=2, min_samples_leaf=1, min_weight_fraction_leaf=0., max_depth=3, init=None, random_state=None, max_features=None, verbose=0, max_leaf_nodes=None, warm_start=False): super(GradientBoostingClassifier, self).__init__( loss=loss, learning_rate=learning_rate, n_estimators=n_estimators, min_samples_split=min_samples_split, min_samples_leaf=min_samples_leaf, min_weight_fraction_leaf=min_weight_fraction_leaf, max_depth=max_depth, init=init, subsample=subsample, max_features=max_features, random_state=random_state, verbose=verbose, max_leaf_nodes=max_leaf_nodes, warm_start=warm_start) def _validate_y(self, y): self.classes_, y = np.unique(y, return_inverse=True) self.n_classes_ = len(self.classes_) return y def decision_function(self, X): """Compute the decision function of ``X``. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- score : array, shape = [n_samples, n_classes] or [n_samples] The decision function of the input samples. The order of the classes corresponds to that in the attribute `classes_`. Regression and binary classification produce an array of shape [n_samples]. """ X = check_array(X, dtype=DTYPE, order="C") score = self._decision_function(X) if score.shape[1] == 1: return score.ravel() return score def staged_decision_function(self, X): """Compute decision function of ``X`` for each iteration. This method allows monitoring (i.e. determine error on testing set) after each stage. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- score : generator of array, shape = [n_samples, k] The decision function of the input samples. The order of the classes corresponds to that in the attribute `classes_`. Regression and binary classification are special cases with ``k == 1``, otherwise ``k==n_classes``. """ for dec in self._staged_decision_function(X): # no yield from in Python2.X yield dec def predict(self, X): """Predict class for X. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- y: array of shape = ["n_samples] The predicted values. """ score = self.decision_function(X) decisions = self.loss_._score_to_decision(score) return self.classes_.take(decisions, axis=0) def staged_predict(self, X): """Predict class at each stage for X. This method allows monitoring (i.e. determine error on testing set) after each stage. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- y : generator of array of shape = [n_samples] The predicted value of the input samples. """ for score in self._staged_decision_function(X): decisions = self.loss_._score_to_decision(score) yield self.classes_.take(decisions, axis=0) def predict_proba(self, X): """Predict class probabilities for X. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Raises ------ AttributeError If the ``loss`` does not support probabilities. Returns ------- p : array of shape = [n_samples] The class probabilities of the input samples. The order of the classes corresponds to that in the attribute `classes_`. """ score = self.decision_function(X) try: return self.loss_._score_to_proba(score) except NotFittedError: raise except AttributeError: raise AttributeError('loss=%r does not support predict_proba' % self.loss) def predict_log_proba(self, X): """Predict class log-probabilities for X. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Raises ------ AttributeError If the ``loss`` does not support probabilities. Returns ------- p : array of shape = [n_samples] The class log-probabilities of the input samples. The order of the classes corresponds to that in the attribute `classes_`. """ proba = self.predict_proba(X) return np.log(proba) def staged_predict_proba(self, X): """Predict class probabilities at each stage for X. This method allows monitoring (i.e. determine error on testing set) after each stage. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- y : generator of array of shape = [n_samples] The predicted value of the input samples. """ try: for score in self._staged_decision_function(X): yield self.loss_._score_to_proba(score) except NotFittedError: raise except AttributeError: raise AttributeError('loss=%r does not support predict_proba' % self.loss) class GradientBoostingRegressor(BaseGradientBoosting, RegressorMixin): """Gradient Boosting for regression. GB builds an additive model in a forward stage-wise fashion; it allows for the optimization of arbitrary differentiable loss functions. In each stage a regression tree is fit on the negative gradient of the given loss function. Read more in the :ref:`User Guide <gradient_boosting>`. Parameters ---------- loss : {'ls', 'lad', 'huber', 'quantile'}, optional (default='ls') loss function to be optimized. 'ls' refers to least squares regression. 'lad' (least absolute deviation) is a highly robust loss function solely based on order information of the input variables. 'huber' is a combination of the two. 'quantile' allows quantile regression (use `alpha` to specify the quantile). learning_rate : float, optional (default=0.1) learning rate shrinks the contribution of each tree by `learning_rate`. There is a trade-off between learning_rate and n_estimators. n_estimators : int (default=100) The number of boosting stages to perform. Gradient boosting is fairly robust to over-fitting so a large number usually results in better performance. max_depth : integer, optional (default=3) maximum depth of the individual regression estimators. The maximum depth limits the number of nodes in the tree. Tune this parameter for best performance; the best value depends on the interaction of the input variables. Ignored if ``max_leaf_nodes`` is not None. min_samples_split : integer, optional (default=2) The minimum number of samples required to split an internal node. min_samples_leaf : integer, optional (default=1) The minimum number of samples required to be at a leaf node. min_weight_fraction_leaf : float, optional (default=0.) The minimum weighted fraction of the input samples required to be at a leaf node. subsample : float, optional (default=1.0) The fraction of samples to be used for fitting the individual base learners. If smaller than 1.0 this results in Stochastic Gradient Boosting. `subsample` interacts with the parameter `n_estimators`. Choosing `subsample < 1.0` leads to a reduction of variance and an increase in bias. max_features : int, float, string or None, optional (default=None) The number of features to consider when looking for the best split: - If int, then consider `max_features` features at each split. - If float, then `max_features` is a percentage and `int(max_features * n_features)` features are considered at each split. - If "auto", then `max_features=n_features`. - If "sqrt", then `max_features=sqrt(n_features)`. - If "log2", then `max_features=log2(n_features)`. - If None, then `max_features=n_features`. Choosing `max_features < n_features` leads to a reduction of variance and an increase in bias. Note: the search for a split does not stop until at least one valid partition of the node samples is found, even if it requires to effectively inspect more than ``max_features`` features. max_leaf_nodes : int or None, optional (default=None) Grow trees with ``max_leaf_nodes`` in best-first fashion. Best nodes are defined as relative reduction in impurity. If None then unlimited number of leaf nodes. alpha : float (default=0.9) The alpha-quantile of the huber loss function and the quantile loss function. Only if ``loss='huber'`` or ``loss='quantile'``. init : BaseEstimator, None, optional (default=None) An estimator object that is used to compute the initial predictions. ``init`` has to provide ``fit`` and ``predict``. If None it uses ``loss.init_estimator``. verbose : int, default: 0 Enable verbose output. If 1 then it prints progress and performance once in a while (the more trees the lower the frequency). If greater than 1 then it prints progress and performance for every tree. warm_start : bool, default: False When set to ``True``, reuse the solution of the previous call to fit and add more estimators to the ensemble, otherwise, just erase the previous solution. random_state : int, RandomState instance or None, optional (default=None) If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. Attributes ---------- feature_importances_ : array, shape = [n_features] The feature importances (the higher, the more important the feature). oob_improvement_ : array, shape = [n_estimators] The improvement in loss (= deviance) on the out-of-bag samples relative to the previous iteration. ``oob_improvement_[0]`` is the improvement in loss of the first stage over the ``init`` estimator. train_score_ : array, shape = [n_estimators] The i-th score ``train_score_[i]`` is the deviance (= loss) of the model at iteration ``i`` on the in-bag sample. If ``subsample == 1`` this is the deviance on the training data. loss_ : LossFunction The concrete ``LossFunction`` object. `init` : BaseEstimator The estimator that provides the initial predictions. Set via the ``init`` argument or ``loss.init_estimator``. estimators_ : ndarray of DecisionTreeRegressor, shape = [n_estimators, 1] The collection of fitted sub-estimators. See also -------- DecisionTreeRegressor, RandomForestRegressor References ---------- J. Friedman, Greedy Function Approximation: A Gradient Boosting Machine, The Annals of Statistics, Vol. 29, No. 5, 2001. J. Friedman, Stochastic Gradient Boosting, 1999 T. Hastie, R. Tibshirani and J. Friedman. Elements of Statistical Learning Ed. 2, Springer, 2009. """ _SUPPORTED_LOSS = ('ls', 'lad', 'huber', 'quantile') def __init__(self, loss='ls', learning_rate=0.1, n_estimators=100, subsample=1.0, min_samples_split=2, min_samples_leaf=1, min_weight_fraction_leaf=0., max_depth=3, init=None, random_state=None, max_features=None, alpha=0.9, verbose=0, max_leaf_nodes=None, warm_start=False): super(GradientBoostingRegressor, self).__init__( loss=loss, learning_rate=learning_rate, n_estimators=n_estimators, min_samples_split=min_samples_split, min_samples_leaf=min_samples_leaf, min_weight_fraction_leaf=min_weight_fraction_leaf, max_depth=max_depth, init=init, subsample=subsample, max_features=max_features, random_state=random_state, alpha=alpha, verbose=verbose, max_leaf_nodes=max_leaf_nodes, warm_start=warm_start) def predict(self, X): """Predict regression target for X. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- y : array of shape = [n_samples] The predicted values. """ X = check_array(X, dtype=DTYPE, order="C") return self._decision_function(X).ravel() def staged_predict(self, X): """Predict regression target at each stage for X. This method allows monitoring (i.e. determine error on testing set) after each stage. Parameters ---------- X : array-like of shape = [n_samples, n_features] The input samples. Returns ------- y : generator of array of shape = [n_samples] The predicted value of the input samples. """ for y in self._staged_decision_function(X): yield y.ravel() def apply(self, X): """Apply trees in the ensemble to X, return leaf indices. Parameters ---------- X : array-like or sparse matrix, shape = [n_samples, n_features] The input samples. Internally, it will be converted to ``dtype=np.float32`` and if a sparse matrix is provided to a sparse ``csr_matrix``. Returns ------- X_leaves : array_like, shape = [n_samples, n_estimators] For each datapoint x in X and for each tree in the ensemble, return the index of the leaf x ends up in in each estimator. """ leaves = super(GradientBoostingRegressor, self).apply(X) leaves = leaves.reshape(X.shape[0], self.estimators_.shape[0]) return leaves
bsd-3-clause
luo66/scikit-learn
sklearn/utils/__init__.py
79
14202
""" The :mod:`sklearn.utils` module includes various utilities. """ from collections import Sequence import numpy as np from scipy.sparse import issparse import warnings from .murmurhash import murmurhash3_32 from .validation import (as_float_array, assert_all_finite, check_random_state, column_or_1d, check_array, check_consistent_length, check_X_y, indexable, check_symmetric, DataConversionWarning) from .class_weight import compute_class_weight, compute_sample_weight from ..externals.joblib import cpu_count __all__ = ["murmurhash3_32", "as_float_array", "assert_all_finite", "check_array", "check_random_state", "compute_class_weight", "compute_sample_weight", "column_or_1d", "safe_indexing", "check_consistent_length", "check_X_y", 'indexable', "check_symmetric"] class deprecated(object): """Decorator to mark a function or class as deprecated. Issue a warning when the function is called/the class is instantiated and adds a warning to the docstring. The optional extra argument will be appended to the deprecation message and the docstring. Note: to use this with the default value for extra, put in an empty of parentheses: >>> from sklearn.utils import deprecated >>> deprecated() # doctest: +ELLIPSIS <sklearn.utils.deprecated object at ...> >>> @deprecated() ... def some_function(): pass """ # Adapted from http://wiki.python.org/moin/PythonDecoratorLibrary, # but with many changes. def __init__(self, extra=''): """ Parameters ---------- extra: string to be added to the deprecation messages """ self.extra = extra def __call__(self, obj): if isinstance(obj, type): return self._decorate_class(obj) else: return self._decorate_fun(obj) def _decorate_class(self, cls): msg = "Class %s is deprecated" % cls.__name__ if self.extra: msg += "; %s" % self.extra # FIXME: we should probably reset __new__ for full generality init = cls.__init__ def wrapped(*args, **kwargs): warnings.warn(msg, category=DeprecationWarning) return init(*args, **kwargs) cls.__init__ = wrapped wrapped.__name__ = '__init__' wrapped.__doc__ = self._update_doc(init.__doc__) wrapped.deprecated_original = init return cls def _decorate_fun(self, fun): """Decorate function fun""" msg = "Function %s is deprecated" % fun.__name__ if self.extra: msg += "; %s" % self.extra def wrapped(*args, **kwargs): warnings.warn(msg, category=DeprecationWarning) return fun(*args, **kwargs) wrapped.__name__ = fun.__name__ wrapped.__dict__ = fun.__dict__ wrapped.__doc__ = self._update_doc(fun.__doc__) return wrapped def _update_doc(self, olddoc): newdoc = "DEPRECATED" if self.extra: newdoc = "%s: %s" % (newdoc, self.extra) if olddoc: newdoc = "%s\n\n%s" % (newdoc, olddoc) return newdoc def safe_mask(X, mask): """Return a mask which is safe to use on X. Parameters ---------- X : {array-like, sparse matrix} Data on which to apply mask. mask: array Mask to be used on X. Returns ------- mask """ mask = np.asarray(mask) if np.issubdtype(mask.dtype, np.int): return mask if hasattr(X, "toarray"): ind = np.arange(mask.shape[0]) mask = ind[mask] return mask def safe_indexing(X, indices): """Return items or rows from X using indices. Allows simple indexing of lists or arrays. Parameters ---------- X : array-like, sparse-matrix, list. Data from which to sample rows or items. indices : array-like, list Indices according to which X will be subsampled. """ if hasattr(X, "iloc"): # Pandas Dataframes and Series try: return X.iloc[indices] except ValueError: # Cython typed memoryviews internally used in pandas do not support # readonly buffers. warnings.warn("Copying input dataframe for slicing.", DataConversionWarning) return X.copy().iloc[indices] elif hasattr(X, "shape"): if hasattr(X, 'take') and (hasattr(indices, 'dtype') and indices.dtype.kind == 'i'): # This is often substantially faster than X[indices] return X.take(indices, axis=0) else: return X[indices] else: return [X[idx] for idx in indices] def resample(*arrays, **options): """Resample arrays or sparse matrices in a consistent way The default strategy implements one step of the bootstrapping procedure. Parameters ---------- *arrays : sequence of indexable data-structures Indexable data-structures can be arrays, lists, dataframes or scipy sparse matrices with consistent first dimension. replace : boolean, True by default Implements resampling with replacement. If False, this will implement (sliced) random permutations. n_samples : int, None by default Number of samples to generate. If left to None this is automatically set to the first dimension of the arrays. random_state : int or RandomState instance Control the shuffling for reproducible behavior. Returns ------- resampled_arrays : sequence of indexable data-structures Sequence of resampled views of the collections. The original arrays are not impacted. Examples -------- It is possible to mix sparse and dense arrays in the same run:: >>> X = np.array([[1., 0.], [2., 1.], [0., 0.]]) >>> y = np.array([0, 1, 2]) >>> from scipy.sparse import coo_matrix >>> X_sparse = coo_matrix(X) >>> from sklearn.utils import resample >>> X, X_sparse, y = resample(X, X_sparse, y, random_state=0) >>> X array([[ 1., 0.], [ 2., 1.], [ 1., 0.]]) >>> X_sparse # doctest: +ELLIPSIS +NORMALIZE_WHITESPACE <3x2 sparse matrix of type '<... 'numpy.float64'>' with 4 stored elements in Compressed Sparse Row format> >>> X_sparse.toarray() array([[ 1., 0.], [ 2., 1.], [ 1., 0.]]) >>> y array([0, 1, 0]) >>> resample(y, n_samples=2, random_state=0) array([0, 1]) See also -------- :func:`sklearn.utils.shuffle` """ random_state = check_random_state(options.pop('random_state', None)) replace = options.pop('replace', True) max_n_samples = options.pop('n_samples', None) if options: raise ValueError("Unexpected kw arguments: %r" % options.keys()) if len(arrays) == 0: return None first = arrays[0] n_samples = first.shape[0] if hasattr(first, 'shape') else len(first) if max_n_samples is None: max_n_samples = n_samples if max_n_samples > n_samples: raise ValueError("Cannot sample %d out of arrays with dim %d" % ( max_n_samples, n_samples)) check_consistent_length(*arrays) if replace: indices = random_state.randint(0, n_samples, size=(max_n_samples,)) else: indices = np.arange(n_samples) random_state.shuffle(indices) indices = indices[:max_n_samples] # convert sparse matrices to CSR for row-based indexing arrays = [a.tocsr() if issparse(a) else a for a in arrays] resampled_arrays = [safe_indexing(a, indices) for a in arrays] if len(resampled_arrays) == 1: # syntactic sugar for the unit argument case return resampled_arrays[0] else: return resampled_arrays def shuffle(*arrays, **options): """Shuffle arrays or sparse matrices in a consistent way This is a convenience alias to ``resample(*arrays, replace=False)`` to do random permutations of the collections. Parameters ---------- *arrays : sequence of indexable data-structures Indexable data-structures can be arrays, lists, dataframes or scipy sparse matrices with consistent first dimension. random_state : int or RandomState instance Control the shuffling for reproducible behavior. n_samples : int, None by default Number of samples to generate. If left to None this is automatically set to the first dimension of the arrays. Returns ------- shuffled_arrays : sequence of indexable data-structures Sequence of shuffled views of the collections. The original arrays are not impacted. Examples -------- It is possible to mix sparse and dense arrays in the same run:: >>> X = np.array([[1., 0.], [2., 1.], [0., 0.]]) >>> y = np.array([0, 1, 2]) >>> from scipy.sparse import coo_matrix >>> X_sparse = coo_matrix(X) >>> from sklearn.utils import shuffle >>> X, X_sparse, y = shuffle(X, X_sparse, y, random_state=0) >>> X array([[ 0., 0.], [ 2., 1.], [ 1., 0.]]) >>> X_sparse # doctest: +ELLIPSIS +NORMALIZE_WHITESPACE <3x2 sparse matrix of type '<... 'numpy.float64'>' with 3 stored elements in Compressed Sparse Row format> >>> X_sparse.toarray() array([[ 0., 0.], [ 2., 1.], [ 1., 0.]]) >>> y array([2, 1, 0]) >>> shuffle(y, n_samples=2, random_state=0) array([0, 1]) See also -------- :func:`sklearn.utils.resample` """ options['replace'] = False return resample(*arrays, **options) def safe_sqr(X, copy=True): """Element wise squaring of array-likes and sparse matrices. Parameters ---------- X : array like, matrix, sparse matrix copy : boolean, optional, default True Whether to create a copy of X and operate on it or to perform inplace computation (default behaviour). Returns ------- X ** 2 : element wise square """ X = check_array(X, accept_sparse=['csr', 'csc', 'coo'], ensure_2d=False) if issparse(X): if copy: X = X.copy() X.data **= 2 else: if copy: X = X ** 2 else: X **= 2 return X def gen_batches(n, batch_size): """Generator to create slices containing batch_size elements, from 0 to n. The last slice may contain less than batch_size elements, when batch_size does not divide n. Examples -------- >>> from sklearn.utils import gen_batches >>> list(gen_batches(7, 3)) [slice(0, 3, None), slice(3, 6, None), slice(6, 7, None)] >>> list(gen_batches(6, 3)) [slice(0, 3, None), slice(3, 6, None)] >>> list(gen_batches(2, 3)) [slice(0, 2, None)] """ start = 0 for _ in range(int(n // batch_size)): end = start + batch_size yield slice(start, end) start = end if start < n: yield slice(start, n) def gen_even_slices(n, n_packs, n_samples=None): """Generator to create n_packs slices going up to n. Pass n_samples when the slices are to be used for sparse matrix indexing; slicing off-the-end raises an exception, while it works for NumPy arrays. Examples -------- >>> from sklearn.utils import gen_even_slices >>> list(gen_even_slices(10, 1)) [slice(0, 10, None)] >>> list(gen_even_slices(10, 10)) #doctest: +ELLIPSIS [slice(0, 1, None), slice(1, 2, None), ..., slice(9, 10, None)] >>> list(gen_even_slices(10, 5)) #doctest: +ELLIPSIS [slice(0, 2, None), slice(2, 4, None), ..., slice(8, 10, None)] >>> list(gen_even_slices(10, 3)) [slice(0, 4, None), slice(4, 7, None), slice(7, 10, None)] """ start = 0 if n_packs < 1: raise ValueError("gen_even_slices got n_packs=%s, must be >=1" % n_packs) for pack_num in range(n_packs): this_n = n // n_packs if pack_num < n % n_packs: this_n += 1 if this_n > 0: end = start + this_n if n_samples is not None: end = min(n_samples, end) yield slice(start, end, None) start = end def _get_n_jobs(n_jobs): """Get number of jobs for the computation. This function reimplements the logic of joblib to determine the actual number of jobs depending on the cpu count. If -1 all CPUs are used. If 1 is given, no parallel computing code is used at all, which is useful for debugging. For n_jobs below -1, (n_cpus + 1 + n_jobs) are used. Thus for n_jobs = -2, all CPUs but one are used. Parameters ---------- n_jobs : int Number of jobs stated in joblib convention. Returns ------- n_jobs : int The actual number of jobs as positive integer. Examples -------- >>> from sklearn.utils import _get_n_jobs >>> _get_n_jobs(4) 4 >>> jobs = _get_n_jobs(-2) >>> assert jobs == max(cpu_count() - 1, 1) >>> _get_n_jobs(0) Traceback (most recent call last): ... ValueError: Parameter n_jobs == 0 has no meaning. """ if n_jobs < 0: return max(cpu_count() + 1 + n_jobs, 1) elif n_jobs == 0: raise ValueError('Parameter n_jobs == 0 has no meaning.') else: return n_jobs def tosequence(x): """Cast iterable x to a Sequence, avoiding a copy if possible.""" if isinstance(x, np.ndarray): return np.asarray(x) elif isinstance(x, Sequence): return x else: return list(x) class ConvergenceWarning(UserWarning): """Custom warning to capture convergence problems""" class DataDimensionalityWarning(UserWarning): """Custom warning to notify potential issues with data dimensionality"""
bsd-3-clause
switowski/invenio
invenio/modules/indexer/tokenizers/BibIndexFiletypeTokenizer.py
12
2411
# -*- coding: utf-8 -*- # # This file is part of Invenio. # Copyright (C) 2010, 2011, 2012, 2013, 2014, 2015 CERN. # # Invenio is free software; you can redistribute it and/or # modify it under the terms of the GNU General Public License as # published by the Free Software Foundation; either version 2 of the # License, or (at your option) any later version. # # Invenio is distributed in the hope that it will be useful, but # WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU # General Public License for more details. # # You should have received a copy of the GNU General Public License # along with Invenio; if not, write to the Free Software Foundation, Inc., # 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA. """BibIndexFiletypeTokenizer: 'tokenizes' for file extensions. Tokenizer is adapted to work with recjson and its get_record function. """ from invenio.modules.indexer.tokenizers.BibIndexRecJsonTokenizer import BibIndexRecJsonTokenizer class BibIndexFiletypeTokenizer(BibIndexRecJsonTokenizer): """ Tokenizes for file extensions. Tokenizer is adapted to work with recjson and its get_record function. It accepts as an input a record created by a get_record function: from invenio.modules.records.api import get_record record16 = get_record(16) tokenizer = BibIndexFiletypeTokenizer() new_words = tokenizer.tokenize(record16) """ def __init__(self, stemming_language = None, remove_stopwords = False, remove_html_markup = False, remove_latex_markup = False): pass def tokenize(self, record): """'record' is a recjson record. Function uses derived field 'filetypes' from the record. @param urls: recjson record """ values = [] try: if 'filetypes' in record: values = record['filetypes'] except KeyError: pass except TypeError: return [] return values def tokenize_for_words(self, record): return self.tokenize(record) def tokenize_for_pairs(self, record): return self.tokenize(record) def tokenize_for_phrases(self, record): return self.tokenize(record) def get_tokenizing_function(self, wordtable_type): return self.tokenize
gpl-2.0
MohammedWasim/scikit-learn
examples/cluster/plot_affinity_propagation.py
346
2304
""" ================================================= Demo of affinity propagation clustering algorithm ================================================= Reference: Brendan J. Frey and Delbert Dueck, "Clustering by Passing Messages Between Data Points", Science Feb. 2007 """ print(__doc__) from sklearn.cluster import AffinityPropagation from sklearn import metrics from sklearn.datasets.samples_generator import make_blobs ############################################################################## # Generate sample data centers = [[1, 1], [-1, -1], [1, -1]] X, labels_true = make_blobs(n_samples=300, centers=centers, cluster_std=0.5, random_state=0) ############################################################################## # Compute Affinity Propagation af = AffinityPropagation(preference=-50).fit(X) cluster_centers_indices = af.cluster_centers_indices_ labels = af.labels_ n_clusters_ = len(cluster_centers_indices) print('Estimated number of clusters: %d' % n_clusters_) print("Homogeneity: %0.3f" % metrics.homogeneity_score(labels_true, labels)) print("Completeness: %0.3f" % metrics.completeness_score(labels_true, labels)) print("V-measure: %0.3f" % metrics.v_measure_score(labels_true, labels)) print("Adjusted Rand Index: %0.3f" % metrics.adjusted_rand_score(labels_true, labels)) print("Adjusted Mutual Information: %0.3f" % metrics.adjusted_mutual_info_score(labels_true, labels)) print("Silhouette Coefficient: %0.3f" % metrics.silhouette_score(X, labels, metric='sqeuclidean')) ############################################################################## # Plot result import matplotlib.pyplot as plt from itertools import cycle plt.close('all') plt.figure(1) plt.clf() colors = cycle('bgrcmykbgrcmykbgrcmykbgrcmyk') for k, col in zip(range(n_clusters_), colors): class_members = labels == k cluster_center = X[cluster_centers_indices[k]] plt.plot(X[class_members, 0], X[class_members, 1], col + '.') plt.plot(cluster_center[0], cluster_center[1], 'o', markerfacecolor=col, markeredgecolor='k', markersize=14) for x in X[class_members]: plt.plot([cluster_center[0], x[0]], [cluster_center[1], x[1]], col) plt.title('Estimated number of clusters: %d' % n_clusters_) plt.show()
bsd-3-clause
glouppe/scikit-learn
examples/cluster/plot_affinity_propagation.py
346
2304
""" ================================================= Demo of affinity propagation clustering algorithm ================================================= Reference: Brendan J. Frey and Delbert Dueck, "Clustering by Passing Messages Between Data Points", Science Feb. 2007 """ print(__doc__) from sklearn.cluster import AffinityPropagation from sklearn import metrics from sklearn.datasets.samples_generator import make_blobs ############################################################################## # Generate sample data centers = [[1, 1], [-1, -1], [1, -1]] X, labels_true = make_blobs(n_samples=300, centers=centers, cluster_std=0.5, random_state=0) ############################################################################## # Compute Affinity Propagation af = AffinityPropagation(preference=-50).fit(X) cluster_centers_indices = af.cluster_centers_indices_ labels = af.labels_ n_clusters_ = len(cluster_centers_indices) print('Estimated number of clusters: %d' % n_clusters_) print("Homogeneity: %0.3f" % metrics.homogeneity_score(labels_true, labels)) print("Completeness: %0.3f" % metrics.completeness_score(labels_true, labels)) print("V-measure: %0.3f" % metrics.v_measure_score(labels_true, labels)) print("Adjusted Rand Index: %0.3f" % metrics.adjusted_rand_score(labels_true, labels)) print("Adjusted Mutual Information: %0.3f" % metrics.adjusted_mutual_info_score(labels_true, labels)) print("Silhouette Coefficient: %0.3f" % metrics.silhouette_score(X, labels, metric='sqeuclidean')) ############################################################################## # Plot result import matplotlib.pyplot as plt from itertools import cycle plt.close('all') plt.figure(1) plt.clf() colors = cycle('bgrcmykbgrcmykbgrcmykbgrcmyk') for k, col in zip(range(n_clusters_), colors): class_members = labels == k cluster_center = X[cluster_centers_indices[k]] plt.plot(X[class_members, 0], X[class_members, 1], col + '.') plt.plot(cluster_center[0], cluster_center[1], 'o', markerfacecolor=col, markeredgecolor='k', markersize=14) for x in X[class_members]: plt.plot([cluster_center[0], x[0]], [cluster_center[1], x[1]], col) plt.title('Estimated number of clusters: %d' % n_clusters_) plt.show()
bsd-3-clause
mileistone/test
brambox/boxes/annotations/kitti.py
1
4525
# # Copyright EAVISE # Author: Tanguy Ophoff # """ KITTI ----- """ from .annotation import * __all__ = ["KittiAnnotation", "KittiParser"] class KittiAnnotation(Annotation): """ KITI image annotation """ def serialize(self): """ generate a KITTI annotation string """ truncated = 1.0 if self.lost else self.truncated_fraction class_label = self.class_label if self.class_label != '' else '?' if self.occluded_fraction >= 0.5: occluded = 2 elif self.occluded_fraction > 0.0: occluded = 1 else: occluded = 0 return f'{class_label} {truncated:.2f} {occluded} -10 {self.x_top_left:.2f} {self.y_top_left:.2f} {self.x_top_left+self.width:.2f} {self.y_top_left+self.height:.2f} -1 -1 -1 -1000 -1000 -1000 -10' def deserialize(self, string): """ parse a KITTI annotation string """ elements = string.split() self.class_label = elements[0] if elements[0] != '?' else '' self.truncated_fraction = max(float(elements[1]), 0.0) self.x_top_left = float(elements[4]) self.y_top_left = float(elements[5]) self.width = float(elements[6]) - self.x_top_left self.height = float(elements[7]) - self.y_top_left if elements[2] == '1': self.occluded_fraction = 0.25 elif elements[2] == '2': self.occluded_fraction = 0.5 else: self.occluded_fraction = 0.0 class KittiParser(Parser): """ This parser can read and write kitti_ annotation files. |br| Some of the values of this dataset are not present in the brambox annotation objects and are thus not used. When serializing this format, these values will be set to their default value, as per specification. ================== ================ =========== Name Number of Values Description ================== ================ =========== class_label 1 Annotation class_label. In the official dataset this can be one of: |br| 'Car', 'Van', 'Truck', 'Pedestrian', 'Person_sitting', 'Cyclist', 'Tram', 'Misc' or 'DontCare' truncated_fraction 1 Float in range [0-1] indicating whether object is truncated occluded_state 1 Integer (0,1,2,3) indicating occlusion state: |br| 0=fully visible, 1=partly occluded, 2=largely occluded, 3=unknown alpha 1 *[Not used in brambox]* Observation angle of the object bbox 4 2D bounding box of the image, expressed in pixel coordinates dimensions 3 *[Not used in brambox]* 3D object dimensions location 3 *[Not used in brambox]* 3D object location rotation_y 1 *[Not used in brambox]* Rotation around Y-axis in camera coordinates ================== ================ =========== Example: >>> image_000.txt <class_label> <truncated_fraction> <occluded_state> -10 <bbox_left> <bbox_top> <bbox_right> <bbox_bottom> -1 -1 -1 -1000 -1000 -1000 -10 <class_label> <truncated_fraction> <occluded_state> -10 <bbox_left> <bbox_top> <bbox_right> <bbox_bottom> -1 -1 -1 -1000 -1000 -1000 -10 >>> image_001.txt <class_label> <truncated_fraction> <occluded_state> -10 <bbox_left> <bbox_top> <bbox_right> <bbox_bottom> -1 -1 -1 -1000 -1000 -1000 -10 <class_label> <truncated_fraction> <occluded_state> -10 <bbox_left> <bbox_top> <bbox_right> <bbox_bottom> -1 -1 -1 -1000 -1000 -1000 -10 <class_label> <truncated_fraction> <occluded_state> -10 <bbox_left> <bbox_top> <bbox_right> <bbox_bottom> -1 -1 -1 -1000 -1000 -1000 -10 Note: This parser will convert the ``occluded_state`` to an ``occluded_fraction``. |br| Partly occluded (1) will be converted to a fraction of 0.25 and largely occluded (2) to 0.5. The other states will be converted to a fraction of 0. |br| When serializing, all fractions bigger or equal to 0.5 will be converted to largely occluded (2), fractions between 0.5 and 0 to partly occluded (1) and fractions of 0 will be converted to fully visible (0). .. _kitti: https://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=2d """ parser_type = ParserType.MULTI_FILE box_type = KittiAnnotation
mit
LohithBlaze/scikit-learn
sklearn/datasets/tests/test_samples_generator.py
35
15016
from __future__ import division from collections import defaultdict from functools import partial import numpy as np from sklearn.externals.six.moves import zip from sklearn.utils.testing import assert_equal from sklearn.utils.testing import assert_array_equal from sklearn.utils.testing import assert_almost_equal from sklearn.utils.testing import assert_array_almost_equal from sklearn.utils.testing import assert_true from sklearn.utils.testing import assert_less from sklearn.utils.testing import assert_raises from sklearn.datasets import make_classification from sklearn.datasets import make_multilabel_classification from sklearn.datasets import make_hastie_10_2 from sklearn.datasets import make_regression from sklearn.datasets import make_blobs from sklearn.datasets import make_friedman1 from sklearn.datasets import make_friedman2 from sklearn.datasets import make_friedman3 from sklearn.datasets import make_low_rank_matrix from sklearn.datasets import make_sparse_coded_signal from sklearn.datasets import make_sparse_uncorrelated from sklearn.datasets import make_spd_matrix from sklearn.datasets import make_swiss_roll from sklearn.datasets import make_s_curve from sklearn.datasets import make_biclusters from sklearn.datasets import make_checkerboard from sklearn.utils.validation import assert_all_finite def test_make_classification(): X, y = make_classification(n_samples=100, n_features=20, n_informative=5, n_redundant=1, n_repeated=1, n_classes=3, n_clusters_per_class=1, hypercube=False, shift=None, scale=None, weights=[0.1, 0.25], random_state=0) assert_equal(X.shape, (100, 20), "X shape mismatch") assert_equal(y.shape, (100,), "y shape mismatch") assert_equal(np.unique(y).shape, (3,), "Unexpected number of classes") assert_equal(sum(y == 0), 10, "Unexpected number of samples in class #0") assert_equal(sum(y == 1), 25, "Unexpected number of samples in class #1") assert_equal(sum(y == 2), 65, "Unexpected number of samples in class #2") def test_make_classification_informative_features(): """Test the construction of informative features in make_classification Also tests `n_clusters_per_class`, `n_classes`, `hypercube` and fully-specified `weights`. """ # Create very separate clusters; check that vertices are unique and # correspond to classes class_sep = 1e6 make = partial(make_classification, class_sep=class_sep, n_redundant=0, n_repeated=0, flip_y=0, shift=0, scale=1, shuffle=False) for n_informative, weights, n_clusters_per_class in [(2, [1], 1), (2, [1/3] * 3, 1), (2, [1/4] * 4, 1), (2, [1/2] * 2, 2), (2, [3/4, 1/4], 2), (10, [1/3] * 3, 10) ]: n_classes = len(weights) n_clusters = n_classes * n_clusters_per_class n_samples = n_clusters * 50 for hypercube in (False, True): X, y = make(n_samples=n_samples, n_classes=n_classes, weights=weights, n_features=n_informative, n_informative=n_informative, n_clusters_per_class=n_clusters_per_class, hypercube=hypercube, random_state=0) assert_equal(X.shape, (n_samples, n_informative)) assert_equal(y.shape, (n_samples,)) # Cluster by sign, viewed as strings to allow uniquing signs = np.sign(X) signs = signs.view(dtype='|S{0}'.format(signs.strides[0])) unique_signs, cluster_index = np.unique(signs, return_inverse=True) assert_equal(len(unique_signs), n_clusters, "Wrong number of clusters, or not in distinct " "quadrants") clusters_by_class = defaultdict(set) for cluster, cls in zip(cluster_index, y): clusters_by_class[cls].add(cluster) for clusters in clusters_by_class.values(): assert_equal(len(clusters), n_clusters_per_class, "Wrong number of clusters per class") assert_equal(len(clusters_by_class), n_classes, "Wrong number of classes") assert_array_almost_equal(np.bincount(y) / len(y) // weights, [1] * n_classes, err_msg="Wrong number of samples " "per class") # Ensure on vertices of hypercube for cluster in range(len(unique_signs)): centroid = X[cluster_index == cluster].mean(axis=0) if hypercube: assert_array_almost_equal(np.abs(centroid), [class_sep] * n_informative, decimal=0, err_msg="Clusters are not " "centered on hypercube " "vertices") else: assert_raises(AssertionError, assert_array_almost_equal, np.abs(centroid), [class_sep] * n_informative, decimal=0, err_msg="Clusters should not be cenetered " "on hypercube vertices") assert_raises(ValueError, make, n_features=2, n_informative=2, n_classes=5, n_clusters_per_class=1) assert_raises(ValueError, make, n_features=2, n_informative=2, n_classes=3, n_clusters_per_class=2) def test_make_multilabel_classification_return_sequences(): for allow_unlabeled, min_length in zip((True, False), (0, 1)): X, Y = make_multilabel_classification(n_samples=100, n_features=20, n_classes=3, random_state=0, return_indicator=False, allow_unlabeled=allow_unlabeled) assert_equal(X.shape, (100, 20), "X shape mismatch") if not allow_unlabeled: assert_equal(max([max(y) for y in Y]), 2) assert_equal(min([len(y) for y in Y]), min_length) assert_true(max([len(y) for y in Y]) <= 3) def test_make_multilabel_classification_return_indicator(): for allow_unlabeled, min_length in zip((True, False), (0, 1)): X, Y = make_multilabel_classification(n_samples=25, n_features=20, n_classes=3, random_state=0, allow_unlabeled=allow_unlabeled) assert_equal(X.shape, (25, 20), "X shape mismatch") assert_equal(Y.shape, (25, 3), "Y shape mismatch") assert_true(np.all(np.sum(Y, axis=0) > min_length)) # Also test return_distributions X2, Y2, p_c, p_w_c = make_multilabel_classification( n_samples=25, n_features=20, n_classes=3, random_state=0, allow_unlabeled=allow_unlabeled, return_distributions=True) assert_array_equal(X, X2) assert_array_equal(Y, Y2) assert_equal(p_c.shape, (3,)) assert_almost_equal(p_c.sum(), 1) assert_equal(p_w_c.shape, (20, 3)) assert_almost_equal(p_w_c.sum(axis=0), [1] * 3) def test_make_hastie_10_2(): X, y = make_hastie_10_2(n_samples=100, random_state=0) assert_equal(X.shape, (100, 10), "X shape mismatch") assert_equal(y.shape, (100,), "y shape mismatch") assert_equal(np.unique(y).shape, (2,), "Unexpected number of classes") def test_make_regression(): X, y, c = make_regression(n_samples=100, n_features=10, n_informative=3, effective_rank=5, coef=True, bias=0.0, noise=1.0, random_state=0) assert_equal(X.shape, (100, 10), "X shape mismatch") assert_equal(y.shape, (100,), "y shape mismatch") assert_equal(c.shape, (10,), "coef shape mismatch") assert_equal(sum(c != 0.0), 3, "Unexpected number of informative features") # Test that y ~= np.dot(X, c) + bias + N(0, 1.0). assert_almost_equal(np.std(y - np.dot(X, c)), 1.0, decimal=1) # Test with small number of features. X, y = make_regression(n_samples=100, n_features=1) # n_informative=3 assert_equal(X.shape, (100, 1)) def test_make_regression_multitarget(): X, y, c = make_regression(n_samples=100, n_features=10, n_informative=3, n_targets=3, coef=True, noise=1., random_state=0) assert_equal(X.shape, (100, 10), "X shape mismatch") assert_equal(y.shape, (100, 3), "y shape mismatch") assert_equal(c.shape, (10, 3), "coef shape mismatch") assert_array_equal(sum(c != 0.0), 3, "Unexpected number of informative features") # Test that y ~= np.dot(X, c) + bias + N(0, 1.0) assert_almost_equal(np.std(y - np.dot(X, c)), 1.0, decimal=1) def test_make_blobs(): cluster_stds = np.array([0.05, 0.2, 0.4]) cluster_centers = np.array([[0.0, 0.0], [1.0, 1.0], [0.0, 1.0]]) X, y = make_blobs(random_state=0, n_samples=50, n_features=2, centers=cluster_centers, cluster_std=cluster_stds) assert_equal(X.shape, (50, 2), "X shape mismatch") assert_equal(y.shape, (50,), "y shape mismatch") assert_equal(np.unique(y).shape, (3,), "Unexpected number of blobs") for i, (ctr, std) in enumerate(zip(cluster_centers, cluster_stds)): assert_almost_equal((X[y == i] - ctr).std(), std, 1, "Unexpected std") def test_make_friedman1(): X, y = make_friedman1(n_samples=5, n_features=10, noise=0.0, random_state=0) assert_equal(X.shape, (5, 10), "X shape mismatch") assert_equal(y.shape, (5,), "y shape mismatch") assert_array_almost_equal(y, 10 * np.sin(np.pi * X[:, 0] * X[:, 1]) + 20 * (X[:, 2] - 0.5) ** 2 + 10 * X[:, 3] + 5 * X[:, 4]) def test_make_friedman2(): X, y = make_friedman2(n_samples=5, noise=0.0, random_state=0) assert_equal(X.shape, (5, 4), "X shape mismatch") assert_equal(y.shape, (5,), "y shape mismatch") assert_array_almost_equal(y, (X[:, 0] ** 2 + (X[:, 1] * X[:, 2] - 1 / (X[:, 1] * X[:, 3])) ** 2) ** 0.5) def test_make_friedman3(): X, y = make_friedman3(n_samples=5, noise=0.0, random_state=0) assert_equal(X.shape, (5, 4), "X shape mismatch") assert_equal(y.shape, (5,), "y shape mismatch") assert_array_almost_equal(y, np.arctan((X[:, 1] * X[:, 2] - 1 / (X[:, 1] * X[:, 3])) / X[:, 0])) def test_make_low_rank_matrix(): X = make_low_rank_matrix(n_samples=50, n_features=25, effective_rank=5, tail_strength=0.01, random_state=0) assert_equal(X.shape, (50, 25), "X shape mismatch") from numpy.linalg import svd u, s, v = svd(X) assert_less(sum(s) - 5, 0.1, "X rank is not approximately 5") def test_make_sparse_coded_signal(): Y, D, X = make_sparse_coded_signal(n_samples=5, n_components=8, n_features=10, n_nonzero_coefs=3, random_state=0) assert_equal(Y.shape, (10, 5), "Y shape mismatch") assert_equal(D.shape, (10, 8), "D shape mismatch") assert_equal(X.shape, (8, 5), "X shape mismatch") for col in X.T: assert_equal(len(np.flatnonzero(col)), 3, 'Non-zero coefs mismatch') assert_array_almost_equal(np.dot(D, X), Y) assert_array_almost_equal(np.sqrt((D ** 2).sum(axis=0)), np.ones(D.shape[1])) def test_make_sparse_uncorrelated(): X, y = make_sparse_uncorrelated(n_samples=5, n_features=10, random_state=0) assert_equal(X.shape, (5, 10), "X shape mismatch") assert_equal(y.shape, (5,), "y shape mismatch") def test_make_spd_matrix(): X = make_spd_matrix(n_dim=5, random_state=0) assert_equal(X.shape, (5, 5), "X shape mismatch") assert_array_almost_equal(X, X.T) from numpy.linalg import eig eigenvalues, _ = eig(X) assert_array_equal(eigenvalues > 0, np.array([True] * 5), "X is not positive-definite") def test_make_swiss_roll(): X, t = make_swiss_roll(n_samples=5, noise=0.0, random_state=0) assert_equal(X.shape, (5, 3), "X shape mismatch") assert_equal(t.shape, (5,), "t shape mismatch") assert_array_almost_equal(X[:, 0], t * np.cos(t)) assert_array_almost_equal(X[:, 2], t * np.sin(t)) def test_make_s_curve(): X, t = make_s_curve(n_samples=5, noise=0.0, random_state=0) assert_equal(X.shape, (5, 3), "X shape mismatch") assert_equal(t.shape, (5,), "t shape mismatch") assert_array_almost_equal(X[:, 0], np.sin(t)) assert_array_almost_equal(X[:, 2], np.sign(t) * (np.cos(t) - 1)) def test_make_biclusters(): X, rows, cols = make_biclusters( shape=(100, 100), n_clusters=4, shuffle=True, random_state=0) assert_equal(X.shape, (100, 100), "X shape mismatch") assert_equal(rows.shape, (4, 100), "rows shape mismatch") assert_equal(cols.shape, (4, 100,), "columns shape mismatch") assert_all_finite(X) assert_all_finite(rows) assert_all_finite(cols) X2, _, _ = make_biclusters(shape=(100, 100), n_clusters=4, shuffle=True, random_state=0) assert_array_almost_equal(X, X2) def test_make_checkerboard(): X, rows, cols = make_checkerboard( shape=(100, 100), n_clusters=(20, 5), shuffle=True, random_state=0) assert_equal(X.shape, (100, 100), "X shape mismatch") assert_equal(rows.shape, (100, 100), "rows shape mismatch") assert_equal(cols.shape, (100, 100,), "columns shape mismatch") X, rows, cols = make_checkerboard( shape=(100, 100), n_clusters=2, shuffle=True, random_state=0) assert_all_finite(X) assert_all_finite(rows) assert_all_finite(cols) X1, _, _ = make_checkerboard(shape=(100, 100), n_clusters=2, shuffle=True, random_state=0) X2, _, _ = make_checkerboard(shape=(100, 100), n_clusters=2, shuffle=True, random_state=0) assert_array_equal(X1, X2)
bsd-3-clause
mlflow/mlflow
mlflow/recipes/steps/ingest/__init__.py
1
10689
import abc import logging import os from pathlib import Path from mlflow.exceptions import MlflowException from mlflow.recipes.artifacts import DataframeArtifact from mlflow.recipes.cards import BaseCard from mlflow.recipes.step import BaseStep from mlflow.recipes.step import StepClass from mlflow.recipes.utils.step import get_pandas_data_profiles from mlflow.protos.databricks_pb2 import INVALID_PARAMETER_VALUE from mlflow.utils.file_utils import read_parquet_as_pandas_df from mlflow.recipes.steps.ingest.datasets import ( ParquetDataset, DeltaTableDataset, SparkSqlDataset, CustomDataset, ) from typing import Dict, Any import pandas as pd _logger = logging.getLogger(__name__) class BaseIngestStep(BaseStep, metaclass=abc.ABCMeta): _DATASET_FORMAT_SPARK_TABLE = "spark_table" _DATASET_FORMAT_DELTA = "delta" _DATASET_FORMAT_PARQUET = "parquet" _DATASET_PROFILE_OUTPUT_NAME = "dataset_profile.html" _STEP_CARD_OUTPUT_NAME = "card.pkl" _SUPPORTED_DATASETS = [ ParquetDataset, DeltaTableDataset, SparkSqlDataset, # NB: The custom dataset is deliberately listed last as a catch-all for any # format not matched by the datasets above. When mapping a format to a dataset, # datasets are explored in the listed order CustomDataset, ] def _validate_and_apply_step_config(self): dataset_format = self.step_config.get("using") if not dataset_format: raise MlflowException( message=( "Dataset format must be specified via the `using` key within the `ingest`" " section of recipe.yaml" ), error_code=INVALID_PARAMETER_VALUE, ) if self.step_class() == StepClass.TRAINING: self.target_col = self.step_config.get("target_col") if self.target_col is None: raise MlflowException( "Missing target_col config in recipe config.", error_code=INVALID_PARAMETER_VALUE, ) if ( "positive_class" not in self.step_config and self.step_config["recipe"] == "classification/v1" ): raise MlflowException( "`positive_class` must be specified for classification/v1 recipes.", error_code=INVALID_PARAMETER_VALUE, ) self.positive_class = self.step_config.get("positive_class") for dataset_class in BaseIngestStep._SUPPORTED_DATASETS: if dataset_class.handles_format(dataset_format): self.dataset = dataset_class.from_config( dataset_config=self.step_config, recipe_root=self.recipe_root, ) break else: raise MlflowException( message=f"Unrecognized dataset format: {dataset_format}", error_code=INVALID_PARAMETER_VALUE, ) self.skip_data_profiling = self.step_config.get("skip_data_profiling", False) def _run(self, output_directory: str) -> BaseCard: dataset_dst_path = os.path.abspath(os.path.join(output_directory, self.dataset_output_name)) self.dataset.resolve_to_parquet( dst_path=dataset_dst_path, ) _logger.debug("Successfully stored data in parquet format at '%s'", dataset_dst_path) ingested_df = read_parquet_as_pandas_df(data_parquet_path=dataset_dst_path) if self.step_class() == StepClass.TRAINING: if self.target_col not in ingested_df.columns: raise MlflowException( f"Target column '{self.target_col}' not found in ingested dataset.", error_code=INVALID_PARAMETER_VALUE, ) if self.positive_class is not None: cardinality = ingested_df[self.target_col].nunique() if cardinality != 2: raise MlflowException( f"Target column '{self.target_col}' must have a cardinality of 2," f"found '{cardinality}'.", error_code=INVALID_PARAMETER_VALUE, ) ingested_dataset_profile = None if not self.skip_data_profiling: _logger.debug("Profiling ingested dataset") ingested_dataset_profile = get_pandas_data_profiles( [["Profile of Ingested Dataset", ingested_df]] ) dataset_profile_path = Path( str(os.path.join(output_directory, BaseIngestStep._DATASET_PROFILE_OUTPUT_NAME)) ) dataset_profile_path.write_text(ingested_dataset_profile, encoding="utf-8") _logger.debug(f"Wrote dataset profile to '{dataset_profile_path}'") schema = pd.io.json.build_table_schema(ingested_df, index=False) step_card = self._build_step_card( ingested_dataset_profile=ingested_dataset_profile, ingested_rows=len(ingested_df), schema=schema, data_preview=ingested_df.head(), dataset_src_location=getattr(self.dataset, "location", None), dataset_sql=getattr(self.dataset, "sql", None), ) return step_card def _build_step_card( self, ingested_dataset_profile: str, ingested_rows: int, schema: Dict, data_preview: pd.DataFrame = None, dataset_src_location: str = None, dataset_sql: str = None, ) -> BaseCard: """ Constructs a step card instance corresponding to the current ingest step state. :param ingested_dataset_path: The local filesystem path to the ingested parquet dataset file. :param dataset_src_location: The source location of the dataset (e.g. '/tmp/myfile.parquet', 's3://mybucket/mypath', ...), if the dataset is a location-based dataset. Either ``dataset_src_location`` or ``dataset_sql`` must be specified. :param dataset_sql: The Spark SQL query string that defines the dataset (e.g. 'SELECT * FROM my_spark_table'), if the dataset is a Spark SQL dataset. Either ``dataset_src_location`` or ``dataset_sql`` must be specified. :return: An BaseCard instance corresponding to the current ingest step state. """ if dataset_src_location is None and dataset_sql is None: raise MlflowException( message=( "Failed to build step card because neither a dataset location nor a" " dataset Spark SQL query were specified" ), error_code=INVALID_PARAMETER_VALUE, ) card = BaseCard(self.recipe_name, self.name) if not self.skip_data_profiling: ( # Tab #1 -- Ingested dataset profile. card.add_tab("Data Profile", "{{PROFILE}}").add_pandas_profile( "PROFILE", ingested_dataset_profile ) ) # Tab #2 -- Ingested dataset schema. schema_html = BaseCard.render_table(schema["fields"]) card.add_tab("Data Schema", "{{SCHEMA}}").add_html("SCHEMA", schema_html) if data_preview is not None: # Tab #3 -- Ingested dataset preview. card.add_tab("Data Preview", "{{DATA_PREVIEW}}").add_html( "DATA_PREVIEW", BaseCard.render_table(data_preview) ) ( # Tab #4 -- Step run summary. card.add_tab( "Run Summary", "{{ INGESTED_ROWS }}" + "{{ DATA_SOURCE }}" + "{{ EXE_DURATION }}" + "{{ LAST_UPDATE_TIME }}", ) .add_markdown( name="INGESTED_ROWS", markdown=f"**Number of rows ingested:** `{ingested_rows}`", ) .add_markdown( name="DATA_SOURCE", markdown=( f"**Dataset source location:** `{dataset_src_location}`" if dataset_src_location is not None else f"**Dataset SQL:** `{dataset_sql}`" ), ) ) return card class IngestStep(BaseIngestStep): _DATASET_OUTPUT_NAME = "dataset.parquet" def __init__(self, step_config: Dict[str, Any], recipe_root: str): super().__init__(step_config, recipe_root) self.dataset_output_name = IngestStep._DATASET_OUTPUT_NAME @classmethod def from_recipe_config(cls, recipe_config: Dict[str, Any], recipe_root: str): ingest_config = recipe_config.get("steps", {}).get("ingest", {}) target_config = {"target_col": recipe_config.get("target_col")} if "positive_class" in recipe_config: target_config["positive_class"] = recipe_config.get("positive_class") return cls( step_config={ **ingest_config, **target_config, **{"recipe": recipe_config.get("recipe")}, }, recipe_root=recipe_root, ) @property def name(self) -> str: return "ingest" def get_artifacts(self): return [ DataframeArtifact( "ingested_data", self.recipe_root, self.name, IngestStep._DATASET_OUTPUT_NAME ) ] def step_class(self): return StepClass.TRAINING class IngestScoringStep(BaseIngestStep): _DATASET_OUTPUT_NAME = "scoring-dataset.parquet" def __init__(self, step_config: Dict[str, Any], recipe_root: str): super().__init__(step_config, recipe_root) self.dataset_output_name = IngestScoringStep._DATASET_OUTPUT_NAME @classmethod def from_recipe_config(cls, recipe_config: Dict[str, Any], recipe_root: str): step_config = recipe_config.get("steps", {}).get("ingest_scoring", {}) return cls( step_config=step_config, recipe_root=recipe_root, ) @property def name(self) -> str: return "ingest_scoring" def get_artifacts(self): return [ DataframeArtifact( "ingested_scoring_data", self.recipe_root, self.name, IngestScoringStep._DATASET_OUTPUT_NAME, ) ] def step_class(self): return StepClass.PREDICTION
apache-2.0
jzt5132/scikit-learn
sklearn/cross_validation.py
47
67782
""" The :mod:`sklearn.cross_validation` module includes utilities for cross- validation and performance evaluation. """ # Author: Alexandre Gramfort <alexandre.gramfort@inria.fr>, # Gael Varoquaux <gael.varoquaux@normalesup.org>, # Olivier Grisel <olivier.grisel@ensta.org> # License: BSD 3 clause from __future__ import print_function from __future__ import division import warnings from itertools import chain, combinations from math import ceil, floor, factorial import numbers import time from abc import ABCMeta, abstractmethod import numpy as np import scipy.sparse as sp from .base import is_classifier, clone from .utils import indexable, check_random_state, safe_indexing from .utils.validation import (_is_arraylike, _num_samples, check_array, column_or_1d) from .utils.multiclass import type_of_target from .externals.joblib import Parallel, delayed, logger from .externals.six import with_metaclass from .externals.six.moves import zip from .metrics.scorer import check_scoring from .utils.fixes import bincount __all__ = ['KFold', 'LabelKFold', 'LeaveOneLabelOut', 'LeaveOneOut', 'LeavePLabelOut', 'LeavePOut', 'ShuffleSplit', 'StratifiedKFold', 'StratifiedShuffleSplit', 'PredefinedSplit', 'LabelShuffleSplit', 'check_cv', 'cross_val_score', 'cross_val_predict', 'permutation_test_score', 'train_test_split'] class _PartitionIterator(with_metaclass(ABCMeta)): """Base class for CV iterators where train_mask = ~test_mask Implementations must define `_iter_test_masks` or `_iter_test_indices`. Parameters ---------- n : int Total number of elements in dataset. """ def __init__(self, n): if abs(n - int(n)) >= np.finfo('f').eps: raise ValueError("n must be an integer") self.n = int(n) def __iter__(self): ind = np.arange(self.n) for test_index in self._iter_test_masks(): train_index = np.logical_not(test_index) train_index = ind[train_index] test_index = ind[test_index] yield train_index, test_index # Since subclasses must implement either _iter_test_masks or # _iter_test_indices, neither can be abstract. def _iter_test_masks(self): """Generates boolean masks corresponding to test sets. By default, delegates to _iter_test_indices() """ for test_index in self._iter_test_indices(): test_mask = self._empty_mask() test_mask[test_index] = True yield test_mask def _iter_test_indices(self): """Generates integer indices corresponding to test sets.""" raise NotImplementedError def _empty_mask(self): return np.zeros(self.n, dtype=np.bool) class LeaveOneOut(_PartitionIterator): """Leave-One-Out cross validation iterator. Provides train/test indices to split data in train test sets. Each sample is used once as a test set (singleton) while the remaining samples form the training set. Note: ``LeaveOneOut(n)`` is equivalent to ``KFold(n, n_folds=n)`` and ``LeavePOut(n, p=1)``. Due to the high number of test sets (which is the same as the number of samples) this cross validation method can be very costly. For large datasets one should favor KFold, StratifiedKFold or ShuffleSplit. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- n : int Total number of elements in dataset. Examples -------- >>> from sklearn import cross_validation >>> X = np.array([[1, 2], [3, 4]]) >>> y = np.array([1, 2]) >>> loo = cross_validation.LeaveOneOut(2) >>> len(loo) 2 >>> print(loo) sklearn.cross_validation.LeaveOneOut(n=2) >>> for train_index, test_index in loo: ... print("TRAIN:", train_index, "TEST:", test_index) ... X_train, X_test = X[train_index], X[test_index] ... y_train, y_test = y[train_index], y[test_index] ... print(X_train, X_test, y_train, y_test) TRAIN: [1] TEST: [0] [[3 4]] [[1 2]] [2] [1] TRAIN: [0] TEST: [1] [[1 2]] [[3 4]] [1] [2] See also -------- LeaveOneLabelOut for splitting the data according to explicit, domain-specific stratification of the dataset. """ def _iter_test_indices(self): return range(self.n) def __repr__(self): return '%s.%s(n=%i)' % ( self.__class__.__module__, self.__class__.__name__, self.n, ) def __len__(self): return self.n class LeavePOut(_PartitionIterator): """Leave-P-Out cross validation iterator Provides train/test indices to split data in train test sets. This results in testing on all distinct samples of size p, while the remaining n - p samples form the training set in each iteration. Note: ``LeavePOut(n, p)`` is NOT equivalent to ``KFold(n, n_folds=n // p)`` which creates non-overlapping test sets. Due to the high number of iterations which grows combinatorically with the number of samples this cross validation method can be very costly. For large datasets one should favor KFold, StratifiedKFold or ShuffleSplit. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- n : int Total number of elements in dataset. p : int Size of the test sets. Examples -------- >>> from sklearn import cross_validation >>> X = np.array([[1, 2], [3, 4], [5, 6], [7, 8]]) >>> y = np.array([1, 2, 3, 4]) >>> lpo = cross_validation.LeavePOut(4, 2) >>> len(lpo) 6 >>> print(lpo) sklearn.cross_validation.LeavePOut(n=4, p=2) >>> for train_index, test_index in lpo: ... print("TRAIN:", train_index, "TEST:", test_index) ... X_train, X_test = X[train_index], X[test_index] ... y_train, y_test = y[train_index], y[test_index] TRAIN: [2 3] TEST: [0 1] TRAIN: [1 3] TEST: [0 2] TRAIN: [1 2] TEST: [0 3] TRAIN: [0 3] TEST: [1 2] TRAIN: [0 2] TEST: [1 3] TRAIN: [0 1] TEST: [2 3] """ def __init__(self, n, p): super(LeavePOut, self).__init__(n) self.p = p def _iter_test_indices(self): for comb in combinations(range(self.n), self.p): yield np.array(comb) def __repr__(self): return '%s.%s(n=%i, p=%i)' % ( self.__class__.__module__, self.__class__.__name__, self.n, self.p, ) def __len__(self): return int(factorial(self.n) / factorial(self.n - self.p) / factorial(self.p)) class _BaseKFold(with_metaclass(ABCMeta, _PartitionIterator)): """Base class to validate KFold approaches""" @abstractmethod def __init__(self, n, n_folds, shuffle, random_state): super(_BaseKFold, self).__init__(n) if abs(n_folds - int(n_folds)) >= np.finfo('f').eps: raise ValueError("n_folds must be an integer") self.n_folds = n_folds = int(n_folds) if n_folds <= 1: raise ValueError( "k-fold cross validation requires at least one" " train / test split by setting n_folds=2 or more," " got n_folds={0}.".format(n_folds)) if n_folds > self.n: raise ValueError( ("Cannot have number of folds n_folds={0} greater" " than the number of samples: {1}.").format(n_folds, n)) if not isinstance(shuffle, bool): raise TypeError("shuffle must be True or False;" " got {0}".format(shuffle)) self.shuffle = shuffle self.random_state = random_state class KFold(_BaseKFold): """K-Folds cross validation iterator. Provides train/test indices to split data in train test sets. Split dataset into k consecutive folds (without shuffling by default). Each fold is then used a validation set once while the k - 1 remaining fold form the training set. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- n : int Total number of elements. n_folds : int, default=3 Number of folds. Must be at least 2. shuffle : boolean, optional Whether to shuffle the data before splitting into batches. random_state : None, int or RandomState When shuffle=True, pseudo-random number generator state used for shuffling. If None, use default numpy RNG for shuffling. Examples -------- >>> from sklearn.cross_validation import KFold >>> X = np.array([[1, 2], [3, 4], [1, 2], [3, 4]]) >>> y = np.array([1, 2, 3, 4]) >>> kf = KFold(4, n_folds=2) >>> len(kf) 2 >>> print(kf) # doctest: +NORMALIZE_WHITESPACE sklearn.cross_validation.KFold(n=4, n_folds=2, shuffle=False, random_state=None) >>> for train_index, test_index in kf: ... print("TRAIN:", train_index, "TEST:", test_index) ... X_train, X_test = X[train_index], X[test_index] ... y_train, y_test = y[train_index], y[test_index] TRAIN: [2 3] TEST: [0 1] TRAIN: [0 1] TEST: [2 3] Notes ----- The first n % n_folds folds have size n // n_folds + 1, other folds have size n // n_folds. See also -------- StratifiedKFold: take label information into account to avoid building folds with imbalanced class distributions (for binary or multiclass classification tasks). LabelKFold: K-fold iterator variant with non-overlapping labels. """ def __init__(self, n, n_folds=3, shuffle=False, random_state=None): super(KFold, self).__init__(n, n_folds, shuffle, random_state) self.idxs = np.arange(n) if shuffle: rng = check_random_state(self.random_state) rng.shuffle(self.idxs) def _iter_test_indices(self): n = self.n n_folds = self.n_folds fold_sizes = (n // n_folds) * np.ones(n_folds, dtype=np.int) fold_sizes[:n % n_folds] += 1 current = 0 for fold_size in fold_sizes: start, stop = current, current + fold_size yield self.idxs[start:stop] current = stop def __repr__(self): return '%s.%s(n=%i, n_folds=%i, shuffle=%s, random_state=%s)' % ( self.__class__.__module__, self.__class__.__name__, self.n, self.n_folds, self.shuffle, self.random_state, ) def __len__(self): return self.n_folds class LabelKFold(_BaseKFold): """K-fold iterator variant with non-overlapping labels. The same label will not appear in two different folds (the number of distinct labels has to be at least equal to the number of folds). The folds are approximately balanced in the sense that the number of distinct labels is approximately the same in each fold. Parameters ---------- labels : array-like with shape (n_samples, ) Contains a label for each sample. The folds are built so that the same label does not appear in two different folds. n_folds : int, default=3 Number of folds. Must be at least 2. shuffle : boolean, optional Whether to shuffle the data before splitting into batches. random_state : None, int or RandomState When shuffle=True, pseudo-random number generator state used for shuffling. If None, use default numpy RNG for shuffling. Examples -------- >>> from sklearn.cross_validation import LabelKFold >>> X = np.array([[1, 2], [3, 4], [5, 6], [7, 8]]) >>> y = np.array([1, 2, 3, 4]) >>> labels = np.array([0, 0, 2, 2]) >>> label_kfold = LabelKFold(labels, n_folds=2) >>> len(label_kfold) 2 >>> print(label_kfold) sklearn.cross_validation.LabelKFold(n_labels=4, n_folds=2) >>> for train_index, test_index in label_kfold: ... print("TRAIN:", train_index, "TEST:", test_index) ... X_train, X_test = X[train_index], X[test_index] ... y_train, y_test = y[train_index], y[test_index] ... print(X_train, X_test, y_train, y_test) ... TRAIN: [0 1] TEST: [2 3] [[1 2] [3 4]] [[5 6] [7 8]] [1 2] [3 4] TRAIN: [2 3] TEST: [0 1] [[5 6] [7 8]] [[1 2] [3 4]] [3 4] [1 2] See also -------- LeaveOneLabelOut for splitting the data according to explicit, domain-specific stratification of the dataset. """ def __init__(self, labels, n_folds=3, shuffle=False, random_state=None): super(LabelKFold, self).__init__(len(labels), n_folds, shuffle, random_state) unique_labels, labels = np.unique(labels, return_inverse=True) n_labels = len(unique_labels) if n_folds > n_labels: raise ValueError( ("Cannot have number of folds n_folds={0} greater" " than the number of labels: {1}.").format(n_folds, n_labels)) # Weight labels by their number of occurences n_samples_per_label = np.bincount(labels) # Distribute the most frequent labels first indices = np.argsort(n_samples_per_label)[::-1] n_samples_per_label = n_samples_per_label[indices] # Total weight of each fold n_samples_per_fold = np.zeros(n_folds) # Mapping from label index to fold index label_to_fold = np.zeros(len(unique_labels)) # Distribute samples by adding the largest weight to the lightest fold for label_index, weight in enumerate(n_samples_per_label): lightest_fold = np.argmin(n_samples_per_fold) n_samples_per_fold[lightest_fold] += weight label_to_fold[indices[label_index]] = lightest_fold self.idxs = label_to_fold[labels] if shuffle: rng = check_random_state(self.random_state) rng.shuffle(self.idxs) def _iter_test_indices(self): for i in range(self.n_folds): yield (self.idxs == i) def __repr__(self): return '{0}.{1}(n_labels={2}, n_folds={3})'.format( self.__class__.__module__, self.__class__.__name__, self.n, self.n_folds, ) def __len__(self): return self.n_folds class StratifiedKFold(_BaseKFold): """Stratified K-Folds cross validation iterator Provides train/test indices to split data in train test sets. This cross-validation object is a variation of KFold that returns stratified folds. The folds are made by preserving the percentage of samples for each class. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- y : array-like, [n_samples] Samples to split in K folds. n_folds : int, default=3 Number of folds. Must be at least 2. shuffle : boolean, optional Whether to shuffle each stratification of the data before splitting into batches. random_state : None, int or RandomState When shuffle=True, pseudo-random number generator state used for shuffling. If None, use default numpy RNG for shuffling. Examples -------- >>> from sklearn.cross_validation import StratifiedKFold >>> X = np.array([[1, 2], [3, 4], [1, 2], [3, 4]]) >>> y = np.array([0, 0, 1, 1]) >>> skf = StratifiedKFold(y, n_folds=2) >>> len(skf) 2 >>> print(skf) # doctest: +NORMALIZE_WHITESPACE sklearn.cross_validation.StratifiedKFold(labels=[0 0 1 1], n_folds=2, shuffle=False, random_state=None) >>> for train_index, test_index in skf: ... print("TRAIN:", train_index, "TEST:", test_index) ... X_train, X_test = X[train_index], X[test_index] ... y_train, y_test = y[train_index], y[test_index] TRAIN: [1 3] TEST: [0 2] TRAIN: [0 2] TEST: [1 3] Notes ----- All the folds have size trunc(n_samples / n_folds), the last one has the complementary. See also -------- LabelKFold: K-fold iterator variant with non-overlapping labels. """ def __init__(self, y, n_folds=3, shuffle=False, random_state=None): super(StratifiedKFold, self).__init__( len(y), n_folds, shuffle, random_state) y = np.asarray(y) n_samples = y.shape[0] unique_labels, y_inversed = np.unique(y, return_inverse=True) label_counts = bincount(y_inversed) min_labels = np.min(label_counts) if self.n_folds > min_labels: warnings.warn(("The least populated class in y has only %d" " members, which is too few. The minimum" " number of labels for any class cannot" " be less than n_folds=%d." % (min_labels, self.n_folds)), Warning) # don't want to use the same seed in each label's shuffle if self.shuffle: rng = check_random_state(self.random_state) else: rng = self.random_state # pre-assign each sample to a test fold index using individual KFold # splitting strategies for each label so as to respect the # balance of labels per_label_cvs = [ KFold(max(c, self.n_folds), self.n_folds, shuffle=self.shuffle, random_state=rng) for c in label_counts] test_folds = np.zeros(n_samples, dtype=np.int) for test_fold_idx, per_label_splits in enumerate(zip(*per_label_cvs)): for label, (_, test_split) in zip(unique_labels, per_label_splits): label_test_folds = test_folds[y == label] # the test split can be too big because we used # KFold(max(c, self.n_folds), self.n_folds) instead of # KFold(c, self.n_folds) to make it possible to not crash even # if the data is not 100% stratifiable for all the labels # (we use a warning instead of raising an exception) # If this is the case, let's trim it: test_split = test_split[test_split < len(label_test_folds)] label_test_folds[test_split] = test_fold_idx test_folds[y == label] = label_test_folds self.test_folds = test_folds self.y = y def _iter_test_masks(self): for i in range(self.n_folds): yield self.test_folds == i def __repr__(self): return '%s.%s(labels=%s, n_folds=%i, shuffle=%s, random_state=%s)' % ( self.__class__.__module__, self.__class__.__name__, self.y, self.n_folds, self.shuffle, self.random_state, ) def __len__(self): return self.n_folds class LeaveOneLabelOut(_PartitionIterator): """Leave-One-Label_Out cross-validation iterator Provides train/test indices to split data according to a third-party provided label. This label information can be used to encode arbitrary domain specific stratifications of the samples as integers. For instance the labels could be the year of collection of the samples and thus allow for cross-validation against time-based splits. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- labels : array-like of int with shape (n_samples,) Arbitrary domain-specific stratification of the data to be used to draw the splits. Examples -------- >>> from sklearn import cross_validation >>> X = np.array([[1, 2], [3, 4], [5, 6], [7, 8]]) >>> y = np.array([1, 2, 1, 2]) >>> labels = np.array([1, 1, 2, 2]) >>> lol = cross_validation.LeaveOneLabelOut(labels) >>> len(lol) 2 >>> print(lol) sklearn.cross_validation.LeaveOneLabelOut(labels=[1 1 2 2]) >>> for train_index, test_index in lol: ... print("TRAIN:", train_index, "TEST:", test_index) ... X_train, X_test = X[train_index], X[test_index] ... y_train, y_test = y[train_index], y[test_index] ... print(X_train, X_test, y_train, y_test) TRAIN: [2 3] TEST: [0 1] [[5 6] [7 8]] [[1 2] [3 4]] [1 2] [1 2] TRAIN: [0 1] TEST: [2 3] [[1 2] [3 4]] [[5 6] [7 8]] [1 2] [1 2] See also -------- LabelKFold: K-fold iterator variant with non-overlapping labels. """ def __init__(self, labels): super(LeaveOneLabelOut, self).__init__(len(labels)) # We make a copy of labels to avoid side-effects during iteration self.labels = np.array(labels, copy=True) self.unique_labels = np.unique(labels) self.n_unique_labels = len(self.unique_labels) def _iter_test_masks(self): for i in self.unique_labels: yield self.labels == i def __repr__(self): return '%s.%s(labels=%s)' % ( self.__class__.__module__, self.__class__.__name__, self.labels, ) def __len__(self): return self.n_unique_labels class LeavePLabelOut(_PartitionIterator): """Leave-P-Label_Out cross-validation iterator Provides train/test indices to split data according to a third-party provided label. This label information can be used to encode arbitrary domain specific stratifications of the samples as integers. For instance the labels could be the year of collection of the samples and thus allow for cross-validation against time-based splits. The difference between LeavePLabelOut and LeaveOneLabelOut is that the former builds the test sets with all the samples assigned to ``p`` different values of the labels while the latter uses samples all assigned the same labels. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- labels : array-like of int with shape (n_samples,) Arbitrary domain-specific stratification of the data to be used to draw the splits. p : int Number of samples to leave out in the test split. Examples -------- >>> from sklearn import cross_validation >>> X = np.array([[1, 2], [3, 4], [5, 6]]) >>> y = np.array([1, 2, 1]) >>> labels = np.array([1, 2, 3]) >>> lpl = cross_validation.LeavePLabelOut(labels, p=2) >>> len(lpl) 3 >>> print(lpl) sklearn.cross_validation.LeavePLabelOut(labels=[1 2 3], p=2) >>> for train_index, test_index in lpl: ... print("TRAIN:", train_index, "TEST:", test_index) ... X_train, X_test = X[train_index], X[test_index] ... y_train, y_test = y[train_index], y[test_index] ... print(X_train, X_test, y_train, y_test) TRAIN: [2] TEST: [0 1] [[5 6]] [[1 2] [3 4]] [1] [1 2] TRAIN: [1] TEST: [0 2] [[3 4]] [[1 2] [5 6]] [2] [1 1] TRAIN: [0] TEST: [1 2] [[1 2]] [[3 4] [5 6]] [1] [2 1] See also -------- LabelKFold: K-fold iterator variant with non-overlapping labels. """ def __init__(self, labels, p): # We make a copy of labels to avoid side-effects during iteration super(LeavePLabelOut, self).__init__(len(labels)) self.labels = np.array(labels, copy=True) self.unique_labels = np.unique(labels) self.n_unique_labels = len(self.unique_labels) self.p = p def _iter_test_masks(self): comb = combinations(range(self.n_unique_labels), self.p) for idx in comb: test_index = self._empty_mask() idx = np.array(idx) for l in self.unique_labels[idx]: test_index[self.labels == l] = True yield test_index def __repr__(self): return '%s.%s(labels=%s, p=%s)' % ( self.__class__.__module__, self.__class__.__name__, self.labels, self.p, ) def __len__(self): return int(factorial(self.n_unique_labels) / factorial(self.n_unique_labels - self.p) / factorial(self.p)) class BaseShuffleSplit(with_metaclass(ABCMeta)): """Base class for ShuffleSplit and StratifiedShuffleSplit""" def __init__(self, n, n_iter=10, test_size=0.1, train_size=None, random_state=None): self.n = n self.n_iter = n_iter self.test_size = test_size self.train_size = train_size self.random_state = random_state self.n_train, self.n_test = _validate_shuffle_split(n, test_size, train_size) def __iter__(self): for train, test in self._iter_indices(): yield train, test return @abstractmethod def _iter_indices(self): """Generate (train, test) indices""" class ShuffleSplit(BaseShuffleSplit): """Random permutation cross-validation iterator. Yields indices to split data into training and test sets. Note: contrary to other cross-validation strategies, random splits do not guarantee that all folds will be different, although this is still very likely for sizeable datasets. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- n : int Total number of elements in the dataset. n_iter : int (default 10) Number of re-shuffling & splitting iterations. test_size : float (default 0.1), int, or None If float, should be between 0.0 and 1.0 and represent the proportion of the dataset to include in the test split. If int, represents the absolute number of test samples. If None, the value is automatically set to the complement of the train size. train_size : float, int, or None (default is None) If float, should be between 0.0 and 1.0 and represent the proportion of the dataset to include in the train split. If int, represents the absolute number of train samples. If None, the value is automatically set to the complement of the test size. random_state : int or RandomState Pseudo-random number generator state used for random sampling. Examples -------- >>> from sklearn import cross_validation >>> rs = cross_validation.ShuffleSplit(4, n_iter=3, ... test_size=.25, random_state=0) >>> len(rs) 3 >>> print(rs) ... # doctest: +ELLIPSIS ShuffleSplit(4, n_iter=3, test_size=0.25, ...) >>> for train_index, test_index in rs: ... print("TRAIN:", train_index, "TEST:", test_index) ... TRAIN: [3 1 0] TEST: [2] TRAIN: [2 1 3] TEST: [0] TRAIN: [0 2 1] TEST: [3] >>> rs = cross_validation.ShuffleSplit(4, n_iter=3, ... train_size=0.5, test_size=.25, random_state=0) >>> for train_index, test_index in rs: ... print("TRAIN:", train_index, "TEST:", test_index) ... TRAIN: [3 1] TEST: [2] TRAIN: [2 1] TEST: [0] TRAIN: [0 2] TEST: [3] """ def _iter_indices(self): rng = check_random_state(self.random_state) for i in range(self.n_iter): # random partition permutation = rng.permutation(self.n) ind_test = permutation[:self.n_test] ind_train = permutation[self.n_test:self.n_test + self.n_train] yield ind_train, ind_test def __repr__(self): return ('%s(%d, n_iter=%d, test_size=%s, ' 'random_state=%s)' % ( self.__class__.__name__, self.n, self.n_iter, str(self.test_size), self.random_state, )) def __len__(self): return self.n_iter def _validate_shuffle_split(n, test_size, train_size): if test_size is None and train_size is None: raise ValueError( 'test_size and train_size can not both be None') if test_size is not None: if np.asarray(test_size).dtype.kind == 'f': if test_size >= 1.: raise ValueError( 'test_size=%f should be smaller ' 'than 1.0 or be an integer' % test_size) elif np.asarray(test_size).dtype.kind == 'i': if test_size >= n: raise ValueError( 'test_size=%d should be smaller ' 'than the number of samples %d' % (test_size, n)) else: raise ValueError("Invalid value for test_size: %r" % test_size) if train_size is not None: if np.asarray(train_size).dtype.kind == 'f': if train_size >= 1.: raise ValueError("train_size=%f should be smaller " "than 1.0 or be an integer" % train_size) elif np.asarray(test_size).dtype.kind == 'f' and \ train_size + test_size > 1.: raise ValueError('The sum of test_size and train_size = %f, ' 'should be smaller than 1.0. Reduce ' 'test_size and/or train_size.' % (train_size + test_size)) elif np.asarray(train_size).dtype.kind == 'i': if train_size >= n: raise ValueError("train_size=%d should be smaller " "than the number of samples %d" % (train_size, n)) else: raise ValueError("Invalid value for train_size: %r" % train_size) if np.asarray(test_size).dtype.kind == 'f': n_test = ceil(test_size * n) elif np.asarray(test_size).dtype.kind == 'i': n_test = float(test_size) if train_size is None: n_train = n - n_test else: if np.asarray(train_size).dtype.kind == 'f': n_train = floor(train_size * n) else: n_train = float(train_size) if test_size is None: n_test = n - n_train if n_train + n_test > n: raise ValueError('The sum of train_size and test_size = %d, ' 'should be smaller than the number of ' 'samples %d. Reduce test_size and/or ' 'train_size.' % (n_train + n_test, n)) return int(n_train), int(n_test) class StratifiedShuffleSplit(BaseShuffleSplit): """Stratified ShuffleSplit cross validation iterator Provides train/test indices to split data in train test sets. This cross-validation object is a merge of StratifiedKFold and ShuffleSplit, which returns stratified randomized folds. The folds are made by preserving the percentage of samples for each class. Note: like the ShuffleSplit strategy, stratified random splits do not guarantee that all folds will be different, although this is still very likely for sizeable datasets. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- y : array, [n_samples] Labels of samples. n_iter : int (default 10) Number of re-shuffling & splitting iterations. test_size : float (default 0.1), int, or None If float, should be between 0.0 and 1.0 and represent the proportion of the dataset to include in the test split. If int, represents the absolute number of test samples. If None, the value is automatically set to the complement of the train size. train_size : float, int, or None (default is None) If float, should be between 0.0 and 1.0 and represent the proportion of the dataset to include in the train split. If int, represents the absolute number of train samples. If None, the value is automatically set to the complement of the test size. random_state : int or RandomState Pseudo-random number generator state used for random sampling. Examples -------- >>> from sklearn.cross_validation import StratifiedShuffleSplit >>> X = np.array([[1, 2], [3, 4], [1, 2], [3, 4]]) >>> y = np.array([0, 0, 1, 1]) >>> sss = StratifiedShuffleSplit(y, 3, test_size=0.5, random_state=0) >>> len(sss) 3 >>> print(sss) # doctest: +ELLIPSIS StratifiedShuffleSplit(labels=[0 0 1 1], n_iter=3, ...) >>> for train_index, test_index in sss: ... print("TRAIN:", train_index, "TEST:", test_index) ... X_train, X_test = X[train_index], X[test_index] ... y_train, y_test = y[train_index], y[test_index] TRAIN: [1 2] TEST: [3 0] TRAIN: [0 2] TEST: [1 3] TRAIN: [0 2] TEST: [3 1] """ def __init__(self, y, n_iter=10, test_size=0.1, train_size=None, random_state=None): super(StratifiedShuffleSplit, self).__init__( len(y), n_iter, test_size, train_size, random_state) self.y = np.array(y) self.classes, self.y_indices = np.unique(y, return_inverse=True) n_cls = self.classes.shape[0] if np.min(bincount(self.y_indices)) < 2: raise ValueError("The least populated class in y has only 1" " member, which is too few. The minimum" " number of labels for any class cannot" " be less than 2.") if self.n_train < n_cls: raise ValueError('The train_size = %d should be greater or ' 'equal to the number of classes = %d' % (self.n_train, n_cls)) if self.n_test < n_cls: raise ValueError('The test_size = %d should be greater or ' 'equal to the number of classes = %d' % (self.n_test, n_cls)) def _iter_indices(self): rng = check_random_state(self.random_state) cls_count = bincount(self.y_indices) p_i = cls_count / float(self.n) n_i = np.round(self.n_train * p_i).astype(int) t_i = np.minimum(cls_count - n_i, np.round(self.n_test * p_i).astype(int)) for n in range(self.n_iter): train = [] test = [] for i, cls in enumerate(self.classes): permutation = rng.permutation(cls_count[i]) cls_i = np.where((self.y == cls))[0][permutation] train.extend(cls_i[:n_i[i]]) test.extend(cls_i[n_i[i]:n_i[i] + t_i[i]]) # Because of rounding issues (as n_train and n_test are not # dividers of the number of elements per class), we may end # up here with less samples in train and test than asked for. if len(train) < self.n_train or len(test) < self.n_test: # We complete by affecting randomly the missing indexes missing_idx = np.where(bincount(train + test, minlength=len(self.y)) == 0, )[0] missing_idx = rng.permutation(missing_idx) train.extend(missing_idx[:(self.n_train - len(train))]) test.extend(missing_idx[-(self.n_test - len(test)):]) train = rng.permutation(train) test = rng.permutation(test) yield train, test def __repr__(self): return ('%s(labels=%s, n_iter=%d, test_size=%s, ' 'random_state=%s)' % ( self.__class__.__name__, self.y, self.n_iter, str(self.test_size), self.random_state, )) def __len__(self): return self.n_iter class PredefinedSplit(_PartitionIterator): """Predefined split cross validation iterator Splits the data into training/test set folds according to a predefined scheme. Each sample can be assigned to at most one test set fold, as specified by the user through the ``test_fold`` parameter. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- test_fold : "array-like, shape (n_samples,) test_fold[i] gives the test set fold of sample i. A value of -1 indicates that the corresponding sample is not part of any test set folds, but will instead always be put into the training fold. Examples -------- >>> from sklearn.cross_validation import PredefinedSplit >>> X = np.array([[1, 2], [3, 4], [1, 2], [3, 4]]) >>> y = np.array([0, 0, 1, 1]) >>> ps = PredefinedSplit(test_fold=[0, 1, -1, 1]) >>> len(ps) 2 >>> print(ps) # doctest: +NORMALIZE_WHITESPACE +ELLIPSIS sklearn.cross_validation.PredefinedSplit(test_fold=[ 0 1 -1 1]) >>> for train_index, test_index in ps: ... print("TRAIN:", train_index, "TEST:", test_index) ... X_train, X_test = X[train_index], X[test_index] ... y_train, y_test = y[train_index], y[test_index] TRAIN: [1 2 3] TEST: [0] TRAIN: [0 2] TEST: [1 3] """ def __init__(self, test_fold): super(PredefinedSplit, self).__init__(len(test_fold)) self.test_fold = np.array(test_fold, dtype=np.int) self.test_fold = column_or_1d(self.test_fold) self.unique_folds = np.unique(self.test_fold) self.unique_folds = self.unique_folds[self.unique_folds != -1] def _iter_test_indices(self): for f in self.unique_folds: yield np.where(self.test_fold == f)[0] def __repr__(self): return '%s.%s(test_fold=%s)' % ( self.__class__.__module__, self.__class__.__name__, self.test_fold) def __len__(self): return len(self.unique_folds) class LabelShuffleSplit(ShuffleSplit): '''Shuffle-Labels-Out cross-validation iterator Provides randomized train/test indices to split data according to a third-party provided label. This label information can be used to encode arbitrary domain specific stratifications of the samples as integers. For instance the labels could be the year of collection of the samples and thus allow for cross-validation against time-based splits. The difference between LeavePLabelOut and LabelShuffleSplit is that the former generates splits using all subsets of size ``p`` unique labels, whereas LabelShuffleSplit generates a user-determined number of random test splits, each with a user-determined fraction of unique labels. For example, a less computationally intensive alternative to ``LeavePLabelOut(labels, p=10)`` would be ``LabelShuffleSplit(labels, test_size=10, n_iter=100)``. Note: The parameters ``test_size`` and ``train_size`` refer to labels, and not to samples, as in ShuffleSplit. Parameters ---------- labels : array, [n_samples] Labels of samples n_iter : int (default 5) Number of re-shuffling & splitting iterations. test_size : float (default 0.2), int, or None If float, should be between 0.0 and 1.0 and represent the proportion of the labels to include in the test split. If int, represents the absolute number of test labels. If None, the value is automatically set to the complement of the train size. train_size : float, int, or None (default is None) If float, should be between 0.0 and 1.0 and represent the proportion of the labels to include in the train split. If int, represents the absolute number of train labels. If None, the value is automatically set to the complement of the test size. random_state : int or RandomState Pseudo-random number generator state used for random sampling. ''' def __init__(self, labels, n_iter=5, test_size=0.2, train_size=None, random_state=None): classes, label_indices = np.unique(labels, return_inverse=True) super(LabelShuffleSplit, self).__init__( len(classes), n_iter=n_iter, test_size=test_size, train_size=train_size, random_state=random_state) self.labels = labels self.classes = classes self.label_indices = label_indices def __repr__(self): return ('%s(labels=%s, n_iter=%d, test_size=%s, ' 'random_state=%s)' % ( self.__class__.__name__, self.labels, self.n_iter, str(self.test_size), self.random_state, )) def __len__(self): return self.n_iter def _iter_indices(self): for label_train, label_test in super(LabelShuffleSplit, self)._iter_indices(): # these are the indices of classes in the partition # invert them into data indices train = np.flatnonzero(np.in1d(self.label_indices, label_train)) test = np.flatnonzero(np.in1d(self.label_indices, label_test)) yield train, test ############################################################################## def _index_param_value(X, v, indices): """Private helper function for parameter value indexing.""" if not _is_arraylike(v) or _num_samples(v) != _num_samples(X): # pass through: skip indexing return v if sp.issparse(v): v = v.tocsr() return safe_indexing(v, indices) def cross_val_predict(estimator, X, y=None, cv=None, n_jobs=1, verbose=0, fit_params=None, pre_dispatch='2*n_jobs'): """Generate cross-validated estimates for each input data point Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- estimator : estimator object implementing 'fit' and 'predict' The object to use to fit the data. X : array-like The data to fit. Can be, for example a list, or an array at least 2d. y : array-like, optional, default: None The target variable to try to predict in the case of supervised learning. cv : int, cross-validation generator or an iterable, optional Determines the cross-validation splitting strategy. Possible inputs for cv are: - None, to use the default 3-fold cross-validation, - integer, to specify the number of folds. - An object to be used as a cross-validation generator. - An iterable yielding train/test splits. For integer/None inputs, if ``y`` is binary or multiclass, :class:`StratifiedKFold` used. If the estimator is a classifier or if ``y`` is neither binary nor multiclass, :class:`KFold` is used. Refer :ref:`User Guide <cross_validation>` for the various cross-validation strategies that can be used here. n_jobs : integer, optional The number of CPUs to use to do the computation. -1 means 'all CPUs'. verbose : integer, optional The verbosity level. fit_params : dict, optional Parameters to pass to the fit method of the estimator. pre_dispatch : int, or string, optional Controls the number of jobs that get dispatched during parallel execution. Reducing this number can be useful to avoid an explosion of memory consumption when more jobs get dispatched than CPUs can process. This parameter can be: - None, in which case all the jobs are immediately created and spawned. Use this for lightweight and fast-running jobs, to avoid delays due to on-demand spawning of the jobs - An int, giving the exact number of total jobs that are spawned - A string, giving an expression as a function of n_jobs, as in '2*n_jobs' Returns ------- preds : ndarray This is the result of calling 'predict' """ X, y = indexable(X, y) cv = check_cv(cv, X, y, classifier=is_classifier(estimator)) # We clone the estimator to make sure that all the folds are # independent, and that it is pickle-able. parallel = Parallel(n_jobs=n_jobs, verbose=verbose, pre_dispatch=pre_dispatch) preds_blocks = parallel(delayed(_fit_and_predict)(clone(estimator), X, y, train, test, verbose, fit_params) for train, test in cv) preds = [p for p, _ in preds_blocks] locs = np.concatenate([loc for _, loc in preds_blocks]) if not _check_is_partition(locs, _num_samples(X)): raise ValueError('cross_val_predict only works for partitions') inv_locs = np.empty(len(locs), dtype=int) inv_locs[locs] = np.arange(len(locs)) # Check for sparse predictions if sp.issparse(preds[0]): preds = sp.vstack(preds, format=preds[0].format) else: preds = np.concatenate(preds) return preds[inv_locs] def _fit_and_predict(estimator, X, y, train, test, verbose, fit_params): """Fit estimator and predict values for a given dataset split. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- estimator : estimator object implementing 'fit' and 'predict' The object to use to fit the data. X : array-like of shape at least 2D The data to fit. y : array-like, optional, default: None The target variable to try to predict in the case of supervised learning. train : array-like, shape (n_train_samples,) Indices of training samples. test : array-like, shape (n_test_samples,) Indices of test samples. verbose : integer The verbosity level. fit_params : dict or None Parameters that will be passed to ``estimator.fit``. Returns ------- preds : sequence Result of calling 'estimator.predict' test : array-like This is the value of the test parameter """ # Adjust length of sample weights fit_params = fit_params if fit_params is not None else {} fit_params = dict([(k, _index_param_value(X, v, train)) for k, v in fit_params.items()]) X_train, y_train = _safe_split(estimator, X, y, train) X_test, _ = _safe_split(estimator, X, y, test, train) if y_train is None: estimator.fit(X_train, **fit_params) else: estimator.fit(X_train, y_train, **fit_params) preds = estimator.predict(X_test) return preds, test def _check_is_partition(locs, n): """Check whether locs is a reordering of the array np.arange(n) Parameters ---------- locs : ndarray integer array to test n : int number of expected elements Returns ------- is_partition : bool True iff sorted(locs) is range(n) """ if len(locs) != n: return False hit = np.zeros(n, bool) hit[locs] = True if not np.all(hit): return False return True def cross_val_score(estimator, X, y=None, scoring=None, cv=None, n_jobs=1, verbose=0, fit_params=None, pre_dispatch='2*n_jobs'): """Evaluate a score by cross-validation Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- estimator : estimator object implementing 'fit' The object to use to fit the data. X : array-like The data to fit. Can be, for example a list, or an array at least 2d. y : array-like, optional, default: None The target variable to try to predict in the case of supervised learning. scoring : string, callable or None, optional, default: None A string (see model evaluation documentation) or a scorer callable object / function with signature ``scorer(estimator, X, y)``. cv : int, cross-validation generator or an iterable, optional Determines the cross-validation splitting strategy. Possible inputs for cv are: - None, to use the default 3-fold cross-validation, - integer, to specify the number of folds. - An object to be used as a cross-validation generator. - An iterable yielding train/test splits. For integer/None inputs, if ``y`` is binary or multiclass, :class:`StratifiedKFold` used. If the estimator is a classifier or if ``y`` is neither binary nor multiclass, :class:`KFold` is used. Refer :ref:`User Guide <cross_validation>` for the various cross-validation strategies that can be used here. n_jobs : integer, optional The number of CPUs to use to do the computation. -1 means 'all CPUs'. verbose : integer, optional The verbosity level. fit_params : dict, optional Parameters to pass to the fit method of the estimator. pre_dispatch : int, or string, optional Controls the number of jobs that get dispatched during parallel execution. Reducing this number can be useful to avoid an explosion of memory consumption when more jobs get dispatched than CPUs can process. This parameter can be: - None, in which case all the jobs are immediately created and spawned. Use this for lightweight and fast-running jobs, to avoid delays due to on-demand spawning of the jobs - An int, giving the exact number of total jobs that are spawned - A string, giving an expression as a function of n_jobs, as in '2*n_jobs' Returns ------- scores : array of float, shape=(len(list(cv)),) Array of scores of the estimator for each run of the cross validation. """ X, y = indexable(X, y) cv = check_cv(cv, X, y, classifier=is_classifier(estimator)) scorer = check_scoring(estimator, scoring=scoring) # We clone the estimator to make sure that all the folds are # independent, and that it is pickle-able. parallel = Parallel(n_jobs=n_jobs, verbose=verbose, pre_dispatch=pre_dispatch) scores = parallel(delayed(_fit_and_score)(clone(estimator), X, y, scorer, train, test, verbose, None, fit_params) for train, test in cv) return np.array(scores)[:, 0] class FitFailedWarning(RuntimeWarning): pass def _fit_and_score(estimator, X, y, scorer, train, test, verbose, parameters, fit_params, return_train_score=False, return_parameters=False, error_score='raise'): """Fit estimator and compute scores for a given dataset split. Parameters ---------- estimator : estimator object implementing 'fit' The object to use to fit the data. X : array-like of shape at least 2D The data to fit. y : array-like, optional, default: None The target variable to try to predict in the case of supervised learning. scorer : callable A scorer callable object / function with signature ``scorer(estimator, X, y)``. train : array-like, shape (n_train_samples,) Indices of training samples. test : array-like, shape (n_test_samples,) Indices of test samples. verbose : integer The verbosity level. error_score : 'raise' (default) or numeric Value to assign to the score if an error occurs in estimator fitting. If set to 'raise', the error is raised. If a numeric value is given, FitFailedWarning is raised. This parameter does not affect the refit step, which will always raise the error. parameters : dict or None Parameters to be set on the estimator. fit_params : dict or None Parameters that will be passed to ``estimator.fit``. return_train_score : boolean, optional, default: False Compute and return score on training set. return_parameters : boolean, optional, default: False Return parameters that has been used for the estimator. Returns ------- train_score : float, optional Score on training set, returned only if `return_train_score` is `True`. test_score : float Score on test set. n_test_samples : int Number of test samples. scoring_time : float Time spent for fitting and scoring in seconds. parameters : dict or None, optional The parameters that have been evaluated. """ if verbose > 1: if parameters is None: msg = "no parameters to be set" else: msg = '%s' % (', '.join('%s=%s' % (k, v) for k, v in parameters.items())) print("[CV] %s %s" % (msg, (64 - len(msg)) * '.')) # Adjust length of sample weights fit_params = fit_params if fit_params is not None else {} fit_params = dict([(k, _index_param_value(X, v, train)) for k, v in fit_params.items()]) if parameters is not None: estimator.set_params(**parameters) start_time = time.time() X_train, y_train = _safe_split(estimator, X, y, train) X_test, y_test = _safe_split(estimator, X, y, test, train) try: if y_train is None: estimator.fit(X_train, **fit_params) else: estimator.fit(X_train, y_train, **fit_params) except Exception as e: if error_score == 'raise': raise elif isinstance(error_score, numbers.Number): test_score = error_score if return_train_score: train_score = error_score warnings.warn("Classifier fit failed. The score on this train-test" " partition for these parameters will be set to %f. " "Details: \n%r" % (error_score, e), FitFailedWarning) else: raise ValueError("error_score must be the string 'raise' or a" " numeric value. (Hint: if using 'raise', please" " make sure that it has been spelled correctly.)" ) else: test_score = _score(estimator, X_test, y_test, scorer) if return_train_score: train_score = _score(estimator, X_train, y_train, scorer) scoring_time = time.time() - start_time if verbose > 2: msg += ", score=%f" % test_score if verbose > 1: end_msg = "%s -%s" % (msg, logger.short_format_time(scoring_time)) print("[CV] %s %s" % ((64 - len(end_msg)) * '.', end_msg)) ret = [train_score] if return_train_score else [] ret.extend([test_score, _num_samples(X_test), scoring_time]) if return_parameters: ret.append(parameters) return ret def _safe_split(estimator, X, y, indices, train_indices=None): """Create subset of dataset and properly handle kernels.""" if hasattr(estimator, 'kernel') and callable(estimator.kernel): # cannot compute the kernel values with custom function raise ValueError("Cannot use a custom kernel function. " "Precompute the kernel matrix instead.") if not hasattr(X, "shape"): if getattr(estimator, "_pairwise", False): raise ValueError("Precomputed kernels or affinity matrices have " "to be passed as arrays or sparse matrices.") X_subset = [X[idx] for idx in indices] else: if getattr(estimator, "_pairwise", False): # X is a precomputed square kernel matrix if X.shape[0] != X.shape[1]: raise ValueError("X should be a square kernel matrix") if train_indices is None: X_subset = X[np.ix_(indices, indices)] else: X_subset = X[np.ix_(indices, train_indices)] else: X_subset = safe_indexing(X, indices) if y is not None: y_subset = safe_indexing(y, indices) else: y_subset = None return X_subset, y_subset def _score(estimator, X_test, y_test, scorer): """Compute the score of an estimator on a given test set.""" if y_test is None: score = scorer(estimator, X_test) else: score = scorer(estimator, X_test, y_test) if not isinstance(score, numbers.Number): raise ValueError("scoring must return a number, got %s (%s) instead." % (str(score), type(score))) return score def _permutation_test_score(estimator, X, y, cv, scorer): """Auxiliary function for permutation_test_score""" avg_score = [] for train, test in cv: estimator.fit(X[train], y[train]) avg_score.append(scorer(estimator, X[test], y[test])) return np.mean(avg_score) def _shuffle(y, labels, random_state): """Return a shuffled copy of y eventually shuffle among same labels.""" if labels is None: ind = random_state.permutation(len(y)) else: ind = np.arange(len(labels)) for label in np.unique(labels): this_mask = (labels == label) ind[this_mask] = random_state.permutation(ind[this_mask]) return y[ind] def check_cv(cv, X=None, y=None, classifier=False): """Input checker utility for building a CV in a user friendly way. Parameters ---------- cv : int, cross-validation generator or an iterable, optional Determines the cross-validation splitting strategy. Possible inputs for cv are: - None, to use the default 3-fold cross-validation, - integer, to specify the number of folds. - An object to be used as a cross-validation generator. - An iterable yielding train/test splits. For integer/None inputs, if ``y`` is binary or multiclass, :class:`StratifiedKFold` used. If the estimator is a classifier or if ``y`` is neither binary nor multiclass, :class:`KFold` is used. Refer :ref:`User Guide <cross_validation>` for the various cross-validation strategies that can be used here. X : array-like The data the cross-val object will be applied on. y : array-like The target variable for a supervised learning problem. classifier : boolean optional Whether the task is a classification task, in which case stratified KFold will be used. Returns ------- checked_cv: a cross-validation generator instance. The return value is guaranteed to be a cv generator instance, whatever the input type. """ is_sparse = sp.issparse(X) if cv is None: cv = 3 if isinstance(cv, numbers.Integral): if classifier: if type_of_target(y) in ['binary', 'multiclass']: cv = StratifiedKFold(y, cv) else: cv = KFold(_num_samples(y), cv) else: if not is_sparse: n_samples = len(X) else: n_samples = X.shape[0] cv = KFold(n_samples, cv) return cv def permutation_test_score(estimator, X, y, cv=None, n_permutations=100, n_jobs=1, labels=None, random_state=0, verbose=0, scoring=None): """Evaluate the significance of a cross-validated score with permutations Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- estimator : estimator object implementing 'fit' The object to use to fit the data. X : array-like of shape at least 2D The data to fit. y : array-like The target variable to try to predict in the case of supervised learning. scoring : string, callable or None, optional, default: None A string (see model evaluation documentation) or a scorer callable object / function with signature ``scorer(estimator, X, y)``. cv : int, cross-validation generator or an iterable, optional Determines the cross-validation splitting strategy. Possible inputs for cv are: - None, to use the default 3-fold cross-validation, - integer, to specify the number of folds. - An object to be used as a cross-validation generator. - An iterable yielding train/test splits. For integer/None inputs, if ``y`` is binary or multiclass, :class:`StratifiedKFold` used. If the estimator is a classifier or if ``y`` is neither binary nor multiclass, :class:`KFold` is used. Refer :ref:`User Guide <cross_validation>` for the various cross-validation strategies that can be used here. n_permutations : integer, optional Number of times to permute ``y``. n_jobs : integer, optional The number of CPUs to use to do the computation. -1 means 'all CPUs'. labels : array-like of shape [n_samples] (optional) Labels constrain the permutation among groups of samples with a same label. random_state : RandomState or an int seed (0 by default) A random number generator instance to define the state of the random permutations generator. verbose : integer, optional The verbosity level. Returns ------- score : float The true score without permuting targets. permutation_scores : array, shape (n_permutations,) The scores obtained for each permutations. pvalue : float The returned value equals p-value if `scoring` returns bigger numbers for better scores (e.g., accuracy_score). If `scoring` is rather a loss function (i.e. when lower is better such as with `mean_squared_error`) then this is actually the complement of the p-value: 1 - p-value. Notes ----- This function implements Test 1 in: Ojala and Garriga. Permutation Tests for Studying Classifier Performance. The Journal of Machine Learning Research (2010) vol. 11 """ X, y = indexable(X, y) cv = check_cv(cv, X, y, classifier=is_classifier(estimator)) scorer = check_scoring(estimator, scoring=scoring) random_state = check_random_state(random_state) # We clone the estimator to make sure that all the folds are # independent, and that it is pickle-able. score = _permutation_test_score(clone(estimator), X, y, cv, scorer) permutation_scores = Parallel(n_jobs=n_jobs, verbose=verbose)( delayed(_permutation_test_score)( clone(estimator), X, _shuffle(y, labels, random_state), cv, scorer) for _ in range(n_permutations)) permutation_scores = np.array(permutation_scores) pvalue = (np.sum(permutation_scores >= score) + 1.0) / (n_permutations + 1) return score, permutation_scores, pvalue permutation_test_score.__test__ = False # to avoid a pb with nosetests def train_test_split(*arrays, **options): """Split arrays or matrices into random train and test subsets Quick utility that wraps input validation and ``next(iter(ShuffleSplit(n_samples)))`` and application to input data into a single call for splitting (and optionally subsampling) data in a oneliner. Read more in the :ref:`User Guide <cross_validation>`. Parameters ---------- *arrays : sequence of arrays or scipy.sparse matrices with same shape[0] Python lists or tuples occurring in arrays are converted to 1D numpy arrays. test_size : float, int, or None (default is None) If float, should be between 0.0 and 1.0 and represent the proportion of the dataset to include in the test split. If int, represents the absolute number of test samples. If None, the value is automatically set to the complement of the train size. If train size is also None, test size is set to 0.25. train_size : float, int, or None (default is None) If float, should be between 0.0 and 1.0 and represent the proportion of the dataset to include in the train split. If int, represents the absolute number of train samples. If None, the value is automatically set to the complement of the test size. random_state : int or RandomState Pseudo-random number generator state used for random sampling. stratify : array-like or None (default is None) If not None, data is split in a stratified fashion, using this as the labels array. Returns ------- splitting : list of arrays, length=2 * len(arrays) List containing train-test split of input array. Examples -------- >>> import numpy as np >>> from sklearn.cross_validation import train_test_split >>> X, y = np.arange(10).reshape((5, 2)), range(5) >>> X array([[0, 1], [2, 3], [4, 5], [6, 7], [8, 9]]) >>> list(y) [0, 1, 2, 3, 4] >>> X_train, X_test, y_train, y_test = train_test_split( ... X, y, test_size=0.33, random_state=42) ... >>> X_train array([[4, 5], [0, 1], [6, 7]]) >>> y_train [2, 0, 3] >>> X_test array([[2, 3], [8, 9]]) >>> y_test [1, 4] """ n_arrays = len(arrays) if n_arrays == 0: raise ValueError("At least one array required as input") test_size = options.pop('test_size', None) train_size = options.pop('train_size', None) random_state = options.pop('random_state', None) dtype = options.pop('dtype', None) if dtype is not None: warnings.warn("dtype option is ignored and will be removed in 0.18.", DeprecationWarning) allow_nd = options.pop('allow_nd', None) allow_lists = options.pop('allow_lists', None) stratify = options.pop('stratify', None) if allow_lists is not None: warnings.warn("The allow_lists option is deprecated and will be " "assumed True in 0.18 and removed.", DeprecationWarning) if options: raise TypeError("Invalid parameters passed: %s" % str(options)) if allow_nd is not None: warnings.warn("The allow_nd option is deprecated and will be " "assumed True in 0.18 and removed.", DeprecationWarning) if allow_lists is False or allow_nd is False: arrays = [check_array(x, 'csr', allow_nd=allow_nd, force_all_finite=False, ensure_2d=False) if x is not None else x for x in arrays] if test_size is None and train_size is None: test_size = 0.25 arrays = indexable(*arrays) if stratify is not None: cv = StratifiedShuffleSplit(stratify, test_size=test_size, train_size=train_size, random_state=random_state) else: n_samples = _num_samples(arrays[0]) cv = ShuffleSplit(n_samples, test_size=test_size, train_size=train_size, random_state=random_state) train, test = next(iter(cv)) return list(chain.from_iterable((safe_indexing(a, train), safe_indexing(a, test)) for a in arrays)) train_test_split.__test__ = False # to avoid a pb with nosetests
bsd-3-clause
marionleborgne/nupic.research
projects/capybara/supervised_baseline/v1_no_sequences/plot_results.py
9
3714
# ---------------------------------------------------------------------- # Numenta Platform for Intelligent Computing (NuPIC) # Copyright (C) 2016, Numenta, Inc. Unless you have an agreement # with Numenta, Inc., for a separate license for this software code, the # following terms and conditions apply: # # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU Affero Public License version 3 as # published by the Free Software Foundation. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. # See the GNU Affero Public License for more details. # # You should have received a copy of the GNU Affero Public License # along with this program. If not, see http://www.gnu.org/licenses. # # http://numenta.org/licenses/ # ---------------------------------------------------------------------- import argparse import os import pandas as pd from sklearn.metrics import (classification_report, confusion_matrix, accuracy_score) from baseline_utils import predictions_vote from plot_utils import (plot_confusion_matrix, plot_train_history, plot_classification_report, plot_predictions) if __name__ == '__main__': # Path to CSV files (training history and predictions) parser = argparse.ArgumentParser() parser.add_argument('--vote_window', '-v', dest='vote_window', type=int, default=11) parser.add_argument('--input_dir', '-i', dest='input_dir', type=str, default='results') parser.add_argument('--output_dir', '-o', dest='output_dir',type=str, default='plots') options = parser.parse_args() vote_window = options.vote_window input_dir = options.input_dir output_dir = options.output_dir if not os.path.exists(output_dir): os.makedirs(output_dir) train_history_path = os.path.join(input_dir, 'train_history.csv') predictions_path = os.path.join(input_dir, 'predictions.csv') # Training history df = pd.read_csv(train_history_path) epochs = range(len(df.epoch.values)) acc = df.acc.values loss = df.loss.values output_file = os.path.join(output_dir, 'train_history.html') plot_train_history(epochs, acc, loss, output_file) print 'Plot saved:', output_file # Predictions df = pd.read_csv(predictions_path) t = df.t.values X_values = df.scalar_value.values y_true = df.y_true.values y_pred = df.y_pred.values if vote_window > 0: y_pred = predictions_vote(y_pred, vote_window) # Accuracy acc = accuracy_score(y_true, y_pred) print 'Accuracy on test set:', acc label_list = sorted(df.y_true.unique()) # Plot normalized confusion matrix cnf_matrix = confusion_matrix(y_true, y_pred) output_file = os.path.join(output_dir, 'confusion_matrix.png') _ = plot_confusion_matrix(cnf_matrix, output_file, classes=label_list, normalize=True, title='Confusion matrix (accuracy=%.2f)' % acc) print 'Plot saved:', output_file # Classification report (F1 score, etc.) clf_report = classification_report(y_true, y_pred) output_file = os.path.join(output_dir, 'classification_report.png') plot_classification_report(clf_report, output_file) print 'Plot saved:', output_file # Plot predictions output_file = os.path.join(output_dir, 'predictions.html') title = 'Predictions (accuracy=%s)' % acc plot_predictions(t, X_values, y_true, y_pred, output_file, title) print 'Plot saved:', output_file
agpl-3.0
tomsilver/nupic
tests/integration/nupic/opf/opf_description_template_test/experiments/gym/base.py
1
15721
# ---------------------------------------------------------------------- # Numenta Platform for Intelligent Computing (NuPIC) # Copyright (C) 2014, Numenta, Inc. Unless you have an agreement # with Numenta, Inc., for a separate license for this software code, the # following terms and conditions apply: # # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License version 3 as # published by the Free Software Foundation. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. # See the GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program. If not, see http://www.gnu.org/licenses. # # http://numenta.org/licenses/ # ---------------------------------------------------------------------- """ Template file used by the OPF Experiment Generator to generate the actual description.py file by replacing $XXXXXXXX tokens with desired values. """ from nupic.frameworks.opf.expdescriptionapi import ExperimentDescriptionAPI from nupic.frameworks.opf.expdescriptionhelpers import ( updateConfigFromSubConfig, applyValueGettersToContainer, DeferredDictLookup) from nupic.frameworks.opf.clamodelcallbacks import * from nupic.frameworks.opf.metrics import MetricSpec from nupic.frameworks.opf.opfutils import (InferenceType, InferenceElement) from nupic.support import aggregationDivide from nupic.frameworks.opf.opftaskdriver import ( IterationPhaseSpecLearnOnly, IterationPhaseSpecInferOnly, IterationPhaseSpecLearnAndInfer) # Model Configuration Dictionary: # # Define the model parameters and adjust for any modifications if imported # from a sub-experiment. # # These fields might be modified by a sub-experiment; this dict is passed # between the sub-experiment and base experiment # # # NOTE: Use of DEFERRED VALUE-GETTERs: dictionary fields and list elements # within the config dictionary may be assigned futures derived from the # ValueGetterBase class, such as DeferredDictLookup. # This facility is particularly handy for enabling substitution of values in # the config dictionary from other values in the config dictionary, which is # needed by permutation.py-based experiments. These values will be resolved # during the call to applyValueGettersToContainer(), # which we call after the base experiment's config dictionary is updated from # the sub-experiment. See ValueGetterBase and # DeferredDictLookup for more details about value-getters. # # For each custom encoder parameter to be exposed to the sub-experiment/ # permutation overrides, define a variable in this section, using key names # beginning with a single underscore character to avoid collisions with # pre-defined keys (e.g., _dsEncoderFieldName2_N). # # Example: # config = dict( # _dsEncoderFieldName2_N = 70, # _dsEncoderFieldName2_W = 5, # dsEncoderSchema = [ # base=dict( # fieldname='Name2', type='ScalarEncoder', # name='Name2', minval=0, maxval=270, clipInput=True, # n=DeferredDictLookup('_dsEncoderFieldName2_N'), # w=DeferredDictLookup('_dsEncoderFieldName2_W')), # ], # ) # updateConfigFromSubConfig(config) # applyValueGettersToContainer(config) config = { # Type of model that the rest of these parameters apply to. 'model': "CLA", # Version that specifies the format of the config. 'version': 1, # Intermediate variables used to compute fields in modelParams and also # referenced from the control section. 'aggregationInfo': { 'fields': [ ('numericFieldNameA', 'mean'), ('numericFieldNameB', 'sum'), ('categoryFieldNameC', 'first')], 'hours': 0}, 'predictAheadTime': None, # Model parameter dictionary. 'modelParams': { # The type of inference that this model will perform 'inferenceType': 'TemporalNextStep', 'sensorParams': { # Sensor diagnostic output verbosity control; # if > 0: sensor region will print out on screen what it's sensing # at each step 0: silent; >=1: some info; >=2: more info; # >=3: even more info (see compute() in py/regions/RecordSensor.py) 'verbosity' : 0, # Example: # dsEncoderSchema = [ # DeferredDictLookup('__field_name_encoder'), # ], # # (value generated from DS_ENCODER_SCHEMA) 'encoders': { 'timestamp': dict(fieldname='timestamp', type='DateEncoder',timeOfDay=(5,5)), 'attendeeCount': dict(fieldname='attendeeCount', type='ScalarEncoder', name='attendeeCount', minval=0, maxval=270, clipInput=True, w=5, resolution=10, forced=True), 'consumption': dict(fieldname='consumption',type='ScalarEncoder', name='consumption', minval=0,maxval=115, clipInput=True, w=5, resolution=5, forced=True), }, # A dictionary specifying the period for automatically-generated # resets from a RecordSensor; # # None = disable automatically-generated resets (also disabled if # all of the specified values evaluate to 0). # Valid keys is the desired combination of the following: # days, hours, minutes, seconds, milliseconds, microseconds, weeks # # Example for 1.5 days: sensorAutoReset = dict(days=1,hours=12), # # (value generated from SENSOR_AUTO_RESET) 'sensorAutoReset' : None, }, 'spEnable': True, 'spParams': { # SP diagnostic output verbosity control; # 0: silent; >=1: some info; >=2: more info; 'spVerbosity' : 0, 'globalInhibition': 1, # Number of cell columns in the cortical region (same number for # SP and TP) # (see also tpNCellsPerCol) 'columnCount': 2048, 'inputWidth': 0, # SP inhibition control (absolute value); # Maximum number of active columns in the SP region's output (when # there are more, the weaker ones are suppressed) 'numActiveColumnsPerInhArea': 20, 'seed': 1956, # potentialPct # What percent of the columns's receptive field is available # for potential synapses. At initialization time, we will # choose potentialPct * (2*potentialRadius+1)^2 'potentialPct': 0.5, # The default connected threshold. Any synapse whose # permanence value is above the connected threshold is # a "connected synapse", meaning it can contribute to the # cell's firing. Typical value is 0.10. Cells whose activity # level before inhibition falls below minDutyCycleBeforeInh # will have their own internal synPermConnectedCell # threshold set below this default value. # (This concept applies to both SP and TP and so 'cells' # is correct here as opposed to 'columns') 'synPermConnected': 0.1, 'synPermActiveInc': 0.1, 'synPermInactiveDec': 0.01, }, # Controls whether TP is enabled or disabled; # TP is necessary for making temporal predictions, such as predicting # the next inputs. Without TP, the model is only capable of # reconstructing missing sensor inputs (via SP). 'tpEnable' : True, 'tpParams': { # TP diagnostic output verbosity control; # 0: silent; [1..6]: increasing levels of verbosity # (see verbosity in nupic/trunk/py/nupic/research/TP.py and TP10X*.py) 'verbosity': 0, # Number of cell columns in the cortical region (same number for # SP and TP) # (see also tpNCellsPerCol) 'columnCount': 2048, # The number of cells (i.e., states), allocated per column. 'cellsPerColumn': 8, 'inputWidth': 2048, 'seed': 1960, # Temporal Pooler implementation selector (see _getTPClass in # CLARegion.py). 'temporalImp': 'cpp', # New Synapse formation count # NOTE: If None, use spNumActivePerInhArea # # TODO: need better explanation 'newSynapseCount': 15, # Maximum number of synapses per segment # > 0 for fixed-size CLA # -1 for non-fixed-size CLA # # TODO: for Ron: once the appropriate value is placed in TP # constructor, see if we should eliminate this parameter from # description.py. 'maxSynapsesPerSegment': 32, # Maximum number of segments per cell # > 0 for fixed-size CLA # -1 for non-fixed-size CLA # # TODO: for Ron: once the appropriate value is placed in TP # constructor, see if we should eliminate this parameter from # description.py. 'maxSegmentsPerCell': 128, # Initial Permanence # TODO: need better explanation 'initialPerm': 0.21, # Permanence Increment 'permanenceInc': 0.1, # Permanence Decrement # If set to None, will automatically default to tpPermanenceInc # value. 'permanenceDec' : 0.1, 'globalDecay': 0.0, 'maxAge': 0, # Minimum number of active synapses for a segment to be considered # during search for the best-matching segments. # None=use default # Replaces: tpMinThreshold 'minThreshold': 12, # Segment activation threshold. # A segment is active if it has >= tpSegmentActivationThreshold # connected synapses that are active due to infActiveState # None=use default # Replaces: tpActivationThreshold 'activationThreshold': float("nan"), 'outputType': 'normal', # "Pay Attention Mode" length. This tells the TP how many new # elements to append to the end of a learned sequence at a time. # Smaller values are better for datasets with short sequences, # higher values are better for datasets with long sequences. 'pamLength': 1, }, 'clParams': { 'regionName' : 'CLAClassifierRegion', # Classifier diagnostic output verbosity control; # 0: silent; [1..6]: increasing levels of verbosity 'clVerbosity' : 0, # This controls how fast the classifier learns/forgets. Higher values # make it adapt faster and forget older patterns faster. 'alpha': 0.001, # This is set after the call to updateConfigFromSubConfig and is # computed from the aggregationInfo and predictAheadTime. 'steps': '1', }, 'trainSPNetOnlyIfRequested': False, }, } # end of config dictionary # Adjust base config dictionary for any modifications if imported from a # sub-experiment updateConfigFromSubConfig(config) # Compute predictionSteps based on the predictAheadTime and the aggregation # period, which may be permuted over. if config['predictAheadTime'] is not None: predictionSteps = int(round(aggregationDivide( config['predictAheadTime'], config['aggregationInfo']))) assert (predictionSteps >= 1) config['modelParams']['clParams']['steps'] = str(predictionSteps) # Adjust config by applying ValueGetterBase-derived # futures. NOTE: this MUST be called after updateConfigFromSubConfig() in order # to support value-getter-based substitutions from the sub-experiment (if any) applyValueGettersToContainer(config) ################################################################################ control = dict( environment = 'opfExperiment', # [optional] A sequence of one or more tasks that describe what to do with the # model. Each task consists of a task label, an input spec., iteration count, # and a task-control spec per opfTaskSchema.json # # NOTE: The tasks are intended for OPF clients that make use of OPFTaskDriver. # Clients that interact with OPFExperiment directly do not make use of # the tasks specification. # tasks = [ { # Task label; this label string may be used for diagnostic logging and for # constructing filenames or directory pathnames for task-specific files, etc. 'taskLabel' : "OnlineLearning", # Input stream specification per py/nupicengine/cluster/database/StreamDef.json. # 'dataset' : { 'info': 'test_NoProviders', 'version': 1, 'streams': [ { 'columns': ['*'], 'info': 'my gym.csv dataset', 'source': 'file://extra/gym/gym.csv', 'first_record': 0, 'last_record': 4000 } ], # TODO: Aggregation is not supported yet by run_opf_experiment.py #'aggregation' : config['aggregationInfo'] }, # Iteration count: maximum number of iterations. Each iteration corresponds # to one record from the (possibly aggregated) dataset. The task is # terminated when either number of iterations reaches iterationCount or # all records in the (possibly aggregated) database have been processed, # whichever occurs first. # # iterationCount of -1 = iterate over the entire dataset 'iterationCount' : -1, # Task Control parameters for OPFTaskDriver (per opfTaskControlSchema.json) 'taskControl' : { # Iteration cycle list consisting of opftaskdriver.IterationPhaseSpecXXXXX # instances. 'iterationCycle' : [ #IterationPhaseSpecLearnOnly(1000), IterationPhaseSpecLearnAndInfer(1000, dict(predictedField="consumption")), #IterationPhaseSpecInferOnly(10), ], 'metrics' :[ MetricSpec(metric='rmse', field="consumption", inferenceElement=InferenceElement.prediction), ], # Callbacks for experimentation/research (optional) 'callbacks' : { # Callbacks to be called at the beginning of a task, before model iterations. # Signature: callback(<reference to OPFExperiment>); returns nothing 'setup' : [], # Callbacks to be called after every learning/inference iteration # Signature: callback(<reference to OPFExperiment>); returns nothing 'postIter' : [], # Callbacks to be called when the experiment task is finished # Signature: callback(<reference to OPFExperiment>); returns nothing 'finish' : [] } } # End of taskControl }, # End of task ] ) ################################################################################ descriptionInterface = ExperimentDescriptionAPI(modelConfig=config, control=control)
gpl-3.0
mlperf/training_results_v0.5
v0.5.0/google/cloud_v2.512/resnet-tpuv2-512/code/resnet/model/models/official/mnist/mnist_eager.py
5
7794
# Copyright 2018 The TensorFlow Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """MNIST model training with TensorFlow eager execution. See: https://research.googleblog.com/2017/10/eager-execution-imperative-define-by.html This program demonstrates training of the convolutional neural network model defined in mnist.py with eager execution enabled. If you are not interested in eager execution, you should ignore this file. """ from __future__ import absolute_import from __future__ import division from __future__ import print_function import os import time # pylint: disable=g-bad-import-order from absl import app as absl_app from absl import flags import tensorflow as tf # pylint: enable=g-bad-import-order from official.mnist import dataset as mnist_dataset from official.mnist import mnist from official.utils.flags import core as flags_core from official.utils.misc import model_helpers tfe = tf.contrib.eager def loss(logits, labels): return tf.reduce_mean( tf.nn.sparse_softmax_cross_entropy_with_logits( logits=logits, labels=labels)) def compute_accuracy(logits, labels): predictions = tf.argmax(logits, axis=1, output_type=tf.int64) labels = tf.cast(labels, tf.int64) batch_size = int(logits.shape[0]) return tf.reduce_sum( tf.cast(tf.equal(predictions, labels), dtype=tf.float32)) / batch_size def train(model, optimizer, dataset, step_counter, log_interval=None): """Trains model on `dataset` using `optimizer`.""" start = time.time() for (batch, (images, labels)) in enumerate(dataset): with tf.contrib.summary.record_summaries_every_n_global_steps( 10, global_step=step_counter): # Record the operations used to compute the loss given the input, # so that the gradient of the loss with respect to the variables # can be computed. with tf.GradientTape() as tape: logits = model(images, training=True) loss_value = loss(logits, labels) tf.contrib.summary.scalar('loss', loss_value) tf.contrib.summary.scalar('accuracy', compute_accuracy(logits, labels)) grads = tape.gradient(loss_value, model.variables) optimizer.apply_gradients( zip(grads, model.variables), global_step=step_counter) if log_interval and batch % log_interval == 0: rate = log_interval / (time.time() - start) print('Step #%d\tLoss: %.6f (%d steps/sec)' % (batch, loss_value, rate)) start = time.time() def test(model, dataset): """Perform an evaluation of `model` on the examples from `dataset`.""" avg_loss = tfe.metrics.Mean('loss', dtype=tf.float32) accuracy = tfe.metrics.Accuracy('accuracy', dtype=tf.float32) for (images, labels) in dataset: logits = model(images, training=False) avg_loss(loss(logits, labels)) accuracy( tf.argmax(logits, axis=1, output_type=tf.int64), tf.cast(labels, tf.int64)) print('Test set: Average loss: %.4f, Accuracy: %4f%%\n' % (avg_loss.result(), 100 * accuracy.result())) with tf.contrib.summary.always_record_summaries(): tf.contrib.summary.scalar('loss', avg_loss.result()) tf.contrib.summary.scalar('accuracy', accuracy.result()) def run_mnist_eager(flags_obj): """Run MNIST training and eval loop in eager mode. Args: flags_obj: An object containing parsed flag values. """ tf.enable_eager_execution() model_helpers.apply_clean(flags.FLAGS) # Automatically determine device and data_format (device, data_format) = ('/gpu:0', 'channels_first') if flags_obj.no_gpu or not tf.test.is_gpu_available(): (device, data_format) = ('/cpu:0', 'channels_last') # If data_format is defined in FLAGS, overwrite automatically set value. if flags_obj.data_format is not None: data_format = flags_obj.data_format print('Using device %s, and data format %s.' % (device, data_format)) # Load the datasets train_ds = mnist_dataset.train(flags_obj.data_dir).shuffle(60000).batch( flags_obj.batch_size) test_ds = mnist_dataset.test(flags_obj.data_dir).batch( flags_obj.batch_size) # Create the model and optimizer model = mnist.create_model(data_format) optimizer = tf.train.MomentumOptimizer(flags_obj.lr, flags_obj.momentum) # Create file writers for writing TensorBoard summaries. if flags_obj.output_dir: # Create directories to which summaries will be written # tensorboard --logdir=<output_dir> # can then be used to see the recorded summaries. train_dir = os.path.join(flags_obj.output_dir, 'train') test_dir = os.path.join(flags_obj.output_dir, 'eval') tf.gfile.MakeDirs(flags_obj.output_dir) else: train_dir = None test_dir = None summary_writer = tf.contrib.summary.create_file_writer( train_dir, flush_millis=10000) test_summary_writer = tf.contrib.summary.create_file_writer( test_dir, flush_millis=10000, name='test') # Create and restore checkpoint (if one exists on the path) checkpoint_prefix = os.path.join(flags_obj.model_dir, 'ckpt') step_counter = tf.train.get_or_create_global_step() checkpoint = tf.train.Checkpoint( model=model, optimizer=optimizer, step_counter=step_counter) # Restore variables on creation if a checkpoint exists. checkpoint.restore(tf.train.latest_checkpoint(flags_obj.model_dir)) # Train and evaluate for a set number of epochs. with tf.device(device): for _ in range(flags_obj.train_epochs): start = time.time() with summary_writer.as_default(): train(model, optimizer, train_ds, step_counter, flags_obj.log_interval) end = time.time() print('\nTrain time for epoch #%d (%d total steps): %f' % (checkpoint.save_counter.numpy() + 1, step_counter.numpy(), end - start)) with test_summary_writer.as_default(): test(model, test_ds) checkpoint.save(checkpoint_prefix) def define_mnist_eager_flags(): """Defined flags and defaults for MNIST in eager mode.""" flags_core.define_base_eager() flags_core.define_image() flags.adopt_module_key_flags(flags_core) flags.DEFINE_integer( name='log_interval', short_name='li', default=10, help=flags_core.help_wrap('batches between logging training status')) flags.DEFINE_string( name='output_dir', short_name='od', default=None, help=flags_core.help_wrap('Directory to write TensorBoard summaries')) flags.DEFINE_float(name='learning_rate', short_name='lr', default=0.01, help=flags_core.help_wrap('Learning rate.')) flags.DEFINE_float(name='momentum', short_name='m', default=0.5, help=flags_core.help_wrap('SGD momentum.')) flags.DEFINE_bool(name='no_gpu', short_name='nogpu', default=False, help=flags_core.help_wrap( 'disables GPU usage even if a GPU is available')) flags_core.set_defaults( data_dir='/tmp/tensorflow/mnist/input_data', model_dir='/tmp/tensorflow/mnist/checkpoints/', batch_size=100, train_epochs=10, ) def main(_): run_mnist_eager(flags.FLAGS) if __name__ == '__main__': define_mnist_eager_flags() absl_app.run(main=main)
apache-2.0
eadgarchen/tensorflow
tensorflow/contrib/factorization/python/ops/gmm_test.py
40
9763
# Copyright 2016 The TensorFlow Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """Tests for ops.gmm.""" from __future__ import absolute_import from __future__ import division from __future__ import print_function import numpy as np from six.moves import xrange # pylint: disable=redefined-builtin from tensorflow.contrib.factorization.python.ops import gmm as gmm_lib from tensorflow.contrib.learn.python.learn.estimators import kmeans from tensorflow.contrib.learn.python.learn.estimators import run_config from tensorflow.python.framework import constant_op from tensorflow.python.framework import dtypes from tensorflow.python.framework import random_seed as random_seed_lib from tensorflow.python.ops import array_ops from tensorflow.python.ops import data_flow_ops from tensorflow.python.ops import random_ops from tensorflow.python.platform import flags from tensorflow.python.platform import test from tensorflow.python.training import queue_runner FLAGS = flags.FLAGS class GMMTest(test.TestCase): def input_fn(self, batch_size=None, points=None): batch_size = batch_size or self.batch_size points = points if points is not None else self.points num_points = points.shape[0] def _fn(): x = constant_op.constant(points) if batch_size == num_points: return x, None indices = random_ops.random_uniform(constant_op.constant([batch_size]), minval=0, maxval=num_points-1, dtype=dtypes.int32, seed=10) return array_ops.gather(x, indices), None return _fn def setUp(self): np.random.seed(3) random_seed_lib.set_random_seed(2) self.num_centers = 2 self.num_dims = 2 self.num_points = 4000 self.batch_size = self.num_points self.true_centers = self.make_random_centers(self.num_centers, self.num_dims) self.points, self.assignments, self.scores = self.make_random_points( self.true_centers, self.num_points) self.true_score = np.add.reduce(self.scores) # Use initial means from kmeans (just like scikit-learn does). clusterer = kmeans.KMeansClustering(num_clusters=self.num_centers) clusterer.fit(input_fn=lambda: (constant_op.constant(self.points), None), steps=30) self.initial_means = clusterer.clusters() @staticmethod def make_random_centers(num_centers, num_dims): return np.round( np.random.rand(num_centers, num_dims).astype(np.float32) * 500) @staticmethod def make_random_points(centers, num_points): num_centers, num_dims = centers.shape assignments = np.random.choice(num_centers, num_points) offsets = np.round( np.random.randn(num_points, num_dims).astype(np.float32) * 20) points = centers[assignments] + offsets means = [ np.mean( points[assignments == center], axis=0) for center in xrange(num_centers) ] covs = [ np.cov(points[assignments == center].T) for center in xrange(num_centers) ] scores = [] for r in xrange(num_points): scores.append( np.sqrt( np.dot( np.dot(points[r, :] - means[assignments[r]], np.linalg.inv(covs[assignments[r]])), points[r, :] - means[assignments[r]]))) return (points, assignments, scores) def test_weights(self): """Tests the shape of the weights.""" gmm = gmm_lib.GMM(self.num_centers, initial_clusters=self.initial_means, random_seed=4, config=run_config.RunConfig(tf_random_seed=2)) gmm.fit(input_fn=self.input_fn(), steps=0) weights = gmm.weights() self.assertAllEqual(list(weights.shape), [self.num_centers]) def test_clusters(self): """Tests the shape of the clusters.""" gmm = gmm_lib.GMM(self.num_centers, initial_clusters=self.initial_means, random_seed=4, config=run_config.RunConfig(tf_random_seed=2)) gmm.fit(input_fn=self.input_fn(), steps=0) clusters = gmm.clusters() self.assertAllEqual(list(clusters.shape), [self.num_centers, self.num_dims]) def test_fit(self): gmm = gmm_lib.GMM(self.num_centers, initial_clusters='random', random_seed=4, config=run_config.RunConfig(tf_random_seed=2)) gmm.fit(input_fn=self.input_fn(), steps=1) score1 = gmm.score(input_fn=self.input_fn(batch_size=self.num_points), steps=1) gmm.fit(input_fn=self.input_fn(), steps=10) score2 = gmm.score(input_fn=self.input_fn(batch_size=self.num_points), steps=1) self.assertGreater(score1, score2) self.assertNear(self.true_score, score2, self.true_score * 0.15) def test_infer(self): gmm = gmm_lib.GMM(self.num_centers, initial_clusters=self.initial_means, random_seed=4, config=run_config.RunConfig(tf_random_seed=2)) gmm.fit(input_fn=self.input_fn(), steps=60) clusters = gmm.clusters() # Make a small test set num_points = 40 points, true_assignments, true_offsets = ( self.make_random_points(clusters, num_points)) assignments = [] for item in gmm.predict_assignments( input_fn=self.input_fn(points=points, batch_size=num_points)): assignments.append(item) assignments = np.ravel(assignments) self.assertAllEqual(true_assignments, assignments) # Test score score = gmm.score(input_fn=self.input_fn(points=points, batch_size=num_points), steps=1) self.assertNear(score, np.sum(true_offsets), 4.05) def _compare_with_sklearn(self, cov_type): # sklearn version. iterations = 40 np.random.seed(5) sklearn_assignments = np.asarray([0, 0, 1, 0, 0, 0, 1, 0, 0, 1]) sklearn_means = np.asarray([[144.83417719, 254.20130341], [274.38754816, 353.16074346]]) sklearn_covs = np.asarray([[[395.0081194, -4.50389512], [-4.50389512, 408.27543989]], [[385.17484203, -31.27834935], [-31.27834935, 391.74249925]]]) # skflow version. gmm = gmm_lib.GMM(self.num_centers, initial_clusters=self.initial_means, covariance_type=cov_type, config=run_config.RunConfig(tf_random_seed=2)) gmm.fit(input_fn=self.input_fn(), steps=iterations) points = self.points[:10, :] skflow_assignments = [] for item in gmm.predict_assignments( input_fn=self.input_fn(points=points, batch_size=10)): skflow_assignments.append(item) self.assertAllClose(sklearn_assignments, np.ravel(skflow_assignments).astype(int)) self.assertAllClose(sklearn_means, gmm.clusters()) if cov_type == 'full': self.assertAllClose(sklearn_covs, gmm.covariances(), rtol=0.01) else: for d in [0, 1]: self.assertAllClose( np.diag(sklearn_covs[d]), gmm.covariances()[d, :], rtol=0.01) def test_compare_full(self): self._compare_with_sklearn('full') def test_compare_diag(self): self._compare_with_sklearn('diag') def test_random_input_large(self): # sklearn version. iterations = 5 # that should be enough to know whether this diverges np.random.seed(5) num_classes = 20 x = np.array([[np.random.random() for _ in range(100)] for _ in range(num_classes)], dtype=np.float32) # skflow version. gmm = gmm_lib.GMM(num_classes, covariance_type='full', config=run_config.RunConfig(tf_random_seed=2)) def get_input_fn(x): def input_fn(): return constant_op.constant(x.astype(np.float32)), None return input_fn gmm.fit(input_fn=get_input_fn(x), steps=iterations) self.assertFalse(np.isnan(gmm.clusters()).any()) class GMMTestQueues(test.TestCase): def input_fn(self): def _fn(): queue = data_flow_ops.FIFOQueue(capacity=10, dtypes=dtypes.float32, shapes=[10, 3]) enqueue_op = queue.enqueue(array_ops.zeros([10, 3], dtype=dtypes.float32)) queue_runner.add_queue_runner(queue_runner.QueueRunner(queue, [enqueue_op])) return queue.dequeue(), None return _fn # This test makes sure that there are no deadlocks when using a QueueRunner. # Note that since cluster initialization is dependendent on inputs, if input # is generated using a QueueRunner, one has to make sure that these runners # are started before the initialization. def test_queues(self): gmm = gmm_lib.GMM(2, covariance_type='diag') gmm.fit(input_fn=self.input_fn(), steps=1) if __name__ == '__main__': test.main()
apache-2.0
ASethi77/StateOfTheMedia
src/model/tune_generic_hyperparams.py
1
2503
# adding this to suppress sklearn DeprecationWarnings... from mpl_toolkits.mplot3d import Axes3D from model.linear_regression_model import LinearRegressionModel from model.MLPRegressionModel import MLPRegressionModel def warn(*args, **kwargs): pass import warnings warnings.warn = warn import time from sklearn.model_selection import train_test_split from util.config import Config, RegressionModels import matplotlib.pyplot as plt from matplotlib import cm from model.overall_runner import corpora_to_day_features, \ init_corpora, combine_day_ranges, match_features_to_labels current_milli_time = lambda: int(round(time.time() * 1000)) if __name__ == '__main__': PREDICT_DELAY_RANGE = range(1, 15) DAY_RANGE_RANGE = range(10, 30) plot_x = [] plot_y = [] plot_z = [] approval_ratings, political_article_corpora = init_corpora() for delay in PREDICT_DELAY_RANGE: plot_x.append([]) plot_y.append([]) plot_z.append([]) for day_range in DAY_RANGE_RANGE: plot_x[delay - 1].append(delay) plot_y[delay - 1].append(day_range) Config.POLL_DELAY = delay Config.DAY_RANGE = day_range features_by_day = corpora_to_day_features(political_article_corpora) features_by_range = combine_day_ranges(features_by_day, approval_ratings) X, Y = match_features_to_labels(features_by_range, approval_ratings) X_train_and_val, X_test, Y_train_and_val, Y_test = \ train_test_split(X, Y, test_size=Config.TRAINING_PARTITION, random_state=2) X_train, X_val, Y_train, Y_val = \ train_test_split(X_train_and_val, Y_train_and_val, test_size=0.125, random_state=2) # setup model and configurations if Config.REGRESSION_MODEL == RegressionModels.LINEAR_REGRESSION: model = LinearRegressionModel([X_train, Y_train]) elif Config.REGRESSION_MODEL == RegressionModels.MLP: model = MLPRegressionModel([X_train, Y_train]) print(model) model.train() mse = model.evaluate(X_val, Y_val) print("MSE is {}".format(mse)) plot_z[delay - 1].append(mse) fig = plt.figure() ax = fig.add_subplot(111, projection='3d') surf = ax.plot_surface(plot_x, plot_y, plot_z, cmap=cm.coolwarm, antialiased=True, rstride=2, cstride=2) plt.show()
apache-2.0
luo66/scikit-learn
sklearn/covariance/tests/test_robust_covariance.py
212
3359
# Author: Alexandre Gramfort <alexandre.gramfort@inria.fr> # Gael Varoquaux <gael.varoquaux@normalesup.org> # Virgile Fritsch <virgile.fritsch@inria.fr> # # License: BSD 3 clause import numpy as np from sklearn.utils.testing import assert_almost_equal from sklearn.utils.testing import assert_array_almost_equal from sklearn.utils.testing import assert_raises from sklearn.utils.validation import NotFittedError from sklearn import datasets from sklearn.covariance import empirical_covariance, MinCovDet, \ EllipticEnvelope X = datasets.load_iris().data X_1d = X[:, 0] n_samples, n_features = X.shape def test_mcd(): # Tests the FastMCD algorithm implementation # Small data set # test without outliers (random independent normal data) launch_mcd_on_dataset(100, 5, 0, 0.01, 0.1, 80) # test with a contaminated data set (medium contamination) launch_mcd_on_dataset(100, 5, 20, 0.01, 0.01, 70) # test with a contaminated data set (strong contamination) launch_mcd_on_dataset(100, 5, 40, 0.1, 0.1, 50) # Medium data set launch_mcd_on_dataset(1000, 5, 450, 0.1, 0.1, 540) # Large data set launch_mcd_on_dataset(1700, 5, 800, 0.1, 0.1, 870) # 1D data set launch_mcd_on_dataset(500, 1, 100, 0.001, 0.001, 350) def launch_mcd_on_dataset(n_samples, n_features, n_outliers, tol_loc, tol_cov, tol_support): rand_gen = np.random.RandomState(0) data = rand_gen.randn(n_samples, n_features) # add some outliers outliers_index = rand_gen.permutation(n_samples)[:n_outliers] outliers_offset = 10. * \ (rand_gen.randint(2, size=(n_outliers, n_features)) - 0.5) data[outliers_index] += outliers_offset inliers_mask = np.ones(n_samples).astype(bool) inliers_mask[outliers_index] = False pure_data = data[inliers_mask] # compute MCD by fitting an object mcd_fit = MinCovDet(random_state=rand_gen).fit(data) T = mcd_fit.location_ S = mcd_fit.covariance_ H = mcd_fit.support_ # compare with the estimates learnt from the inliers error_location = np.mean((pure_data.mean(0) - T) ** 2) assert(error_location < tol_loc) error_cov = np.mean((empirical_covariance(pure_data) - S) ** 2) assert(error_cov < tol_cov) assert(np.sum(H) >= tol_support) assert_array_almost_equal(mcd_fit.mahalanobis(data), mcd_fit.dist_) def test_mcd_issue1127(): # Check that the code does not break with X.shape = (3, 1) # (i.e. n_support = n_samples) rnd = np.random.RandomState(0) X = rnd.normal(size=(3, 1)) mcd = MinCovDet() mcd.fit(X) def test_outlier_detection(): rnd = np.random.RandomState(0) X = rnd.randn(100, 10) clf = EllipticEnvelope(contamination=0.1) assert_raises(NotFittedError, clf.predict, X) assert_raises(NotFittedError, clf.decision_function, X) clf.fit(X) y_pred = clf.predict(X) decision = clf.decision_function(X, raw_values=True) decision_transformed = clf.decision_function(X, raw_values=False) assert_array_almost_equal( decision, clf.mahalanobis(X)) assert_array_almost_equal(clf.mahalanobis(X), clf.dist_) assert_almost_equal(clf.score(X, np.ones(100)), (100 - y_pred[y_pred == -1].size) / 100.) assert(sum(y_pred == -1) == sum(decision_transformed < 0))
bsd-3-clause
mlflow/mlflow
mlflow/tracking/metric_value_conversion_utils.py
1
2248
import sys from mlflow.exceptions import MlflowException, INVALID_PARAMETER_VALUE def _is_module_imported(module_name: str) -> bool: return module_name in sys.modules def _try_get_item(x): try: return x.item() except Exception as e: raise MlflowException( f"Failed to convert metric value to float: {e}", error_code=INVALID_PARAMETER_VALUE, ) def _converter_requires(module_name: str): """Wrapper function that checks if specified `module_name` is already imported before invoking wrapped function.""" def decorator(func): def wrapper(x): if not _is_module_imported(module_name): return x return func(x) return wrapper return decorator def convert_metric_value_to_float_if_possible(x) -> float: if x is None or type(x) == float: return x converter_fns_to_try = [ convert_metric_value_to_float_if_ndarray, convert_metric_value_to_float_if_tensorflow_tensor, convert_metric_value_to_float_if_torch_tensor, ] for converter_fn in converter_fns_to_try: possible_float = converter_fn(x) if type(possible_float) == float: return possible_float try: return float(x) except ValueError: return x # let backend handle conversion if possible @_converter_requires("numpy") def convert_metric_value_to_float_if_ndarray(x): import numpy as np if isinstance(x, np.ndarray): return float(_try_get_item(x)) return x @_converter_requires("torch") def convert_metric_value_to_float_if_torch_tensor(x): import torch if isinstance(x, torch.Tensor): extracted_tensor_val = x.detach().cpu() return float(_try_get_item(extracted_tensor_val)) return x @_converter_requires("tensorflow") def convert_metric_value_to_float_if_tensorflow_tensor(x): import tensorflow as tf if isinstance(x, tf.Tensor): try: return float(x) except Exception as e: raise MlflowException( f"Failed to convert metric value to float: {repr(e)}", error_code=INVALID_PARAMETER_VALUE, ) return x
apache-2.0
lilleswing/deepchem
examples/factors/FACTORS_tf_singletask.py
6
3341
""" Script that trains Tensorflow Singletask models on FACTORS dataset. """ from __future__ import print_function from __future__ import division from __future__ import unicode_literals import os import numpy as np import tempfile import shutil import deepchem as dc from FACTORS_datasets import load_factors ###Load data### shard_size = 2000 num_trials = 2 print("About to load FACTORS data.") FACTORS_tasks, datasets, transformers = load_factors(shard_size=shard_size) train_dataset, valid_dataset, test_dataset = datasets print("Number of compounds in train set") print(len(train_dataset)) print("Number of compounds in validation set") print(len(valid_dataset)) print("Number of compounds in test set") print(len(test_dataset)) metric = dc.metrics.Metric(dc.metrics.pearson_r2_score, task_averager=np.mean) ###Create model### n_layers = 3 nb_epoch = 125 n_features = train_dataset.get_data_shape()[0] def task_model_builder(m_dir): return dc.models.TensorflowMultitaskRegressor( n_tasks=1, n_features=n_features, logdir=m_dir, layer_sizes=[1000] * n_layers, dropouts=[.25] * n_layers, weight_init_stddevs=[.02] * n_layers, bias_init_consts=[1.] * n_layers, learning_rate=.0003, penalty=.0001, penalty_type="l2", optimizer="adam", batch_size=100) all_results = [] for trial in range(num_trials): print("Starting trial %d" % trial) model = dc.models.SingletaskToMultitask(FACTORS_tasks, task_model_builder) print("Fitting Model") model.fit(train_dataset, nb_epoch=nb_epoch) print("Evaluating models") train_score, train_task_scores = model.evaluate( train_dataset, [metric], transformers, per_task_metrics=True) valid_score, valid_task_scores = model.evaluate( valid_dataset, [metric], transformers, per_task_metrics=True) test_score, test_task_scores = model.evaluate( test_dataset, [metric], transformers, per_task_metrics=True) all_results.append((train_score, train_task_scores, valid_score, valid_task_scores, test_score, test_task_scores)) print("----------------------------------------------------------------") print("Scores for trial %d" % trial) print("----------------------------------------------------------------") print("train_task_scores") print(train_task_scores) print("Mean Train score") print(train_score) print("valid_task_scores") print(valid_task_scores) print("Mean Validation score") print(valid_score) print("test_task_scores") print(test_task_scores) print("Mean Test score") print(test_score) print("####################################################################") for trial in range(num_trials): (train_score, train_task_scores, valid_score, valid_task_scores, test_score, test_task_scores) = all_results[trial] print("----------------------------------------------------------------") print("Scores for trial %d" % trial) print("----------------------------------------------------------------") print("train_task_scores") print(train_task_scores) print("Mean Train score") print(train_score) print("valid_task_scores") print(valid_task_scores) print("Mean Validation score") print(valid_score) print("test_task_scores") print(test_task_scores) print("Mean Test score") print(test_score)
mit
LohithBlaze/scikit-learn
examples/feature_selection/plot_feature_selection.py
248
2827
""" =============================== Univariate Feature Selection =============================== An example showing univariate feature selection. Noisy (non informative) features are added to the iris data and univariate feature selection is applied. For each feature, we plot the p-values for the univariate feature selection and the corresponding weights of an SVM. We can see that univariate feature selection selects the informative features and that these have larger SVM weights. In the total set of features, only the 4 first ones are significant. We can see that they have the highest score with univariate feature selection. The SVM assigns a large weight to one of these features, but also Selects many of the non-informative features. Applying univariate feature selection before the SVM increases the SVM weight attributed to the significant features, and will thus improve classification. """ print(__doc__) import numpy as np import matplotlib.pyplot as plt from sklearn import datasets, svm from sklearn.feature_selection import SelectPercentile, f_classif ############################################################################### # import some data to play with # The iris dataset iris = datasets.load_iris() # Some noisy data not correlated E = np.random.uniform(0, 0.1, size=(len(iris.data), 20)) # Add the noisy data to the informative features X = np.hstack((iris.data, E)) y = iris.target ############################################################################### plt.figure(1) plt.clf() X_indices = np.arange(X.shape[-1]) ############################################################################### # Univariate feature selection with F-test for feature scoring # We use the default selection function: the 10% most significant features selector = SelectPercentile(f_classif, percentile=10) selector.fit(X, y) scores = -np.log10(selector.pvalues_) scores /= scores.max() plt.bar(X_indices - .45, scores, width=.2, label=r'Univariate score ($-Log(p_{value})$)', color='g') ############################################################################### # Compare to the weights of an SVM clf = svm.SVC(kernel='linear') clf.fit(X, y) svm_weights = (clf.coef_ ** 2).sum(axis=0) svm_weights /= svm_weights.max() plt.bar(X_indices - .25, svm_weights, width=.2, label='SVM weight', color='r') clf_selected = svm.SVC(kernel='linear') clf_selected.fit(selector.transform(X), y) svm_weights_selected = (clf_selected.coef_ ** 2).sum(axis=0) svm_weights_selected /= svm_weights_selected.max() plt.bar(X_indices[selector.get_support()] - .05, svm_weights_selected, width=.2, label='SVM weights after selection', color='b') plt.title("Comparing feature selection") plt.xlabel('Feature number') plt.yticks(()) plt.axis('tight') plt.legend(loc='upper right') plt.show()
bsd-3-clause
fx2003/tensorflow-study
TensorFlow实战/models/inception/inception/data/build_imagenet_data.py
12
26205
# Copyright 2016 Google Inc. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """Converts ImageNet data to TFRecords file format with Example protos. The raw ImageNet data set is expected to reside in JPEG files located in the following directory structure. data_dir/n01440764/ILSVRC2012_val_00000293.JPEG data_dir/n01440764/ILSVRC2012_val_00000543.JPEG ... where 'n01440764' is the unique synset label associated with these images. The training data set consists of 1000 sub-directories (i.e. labels) each containing 1200 JPEG images for a total of 1.2M JPEG images. The evaluation data set consists of 1000 sub-directories (i.e. labels) each containing 50 JPEG images for a total of 50K JPEG images. This TensorFlow script converts the training and evaluation data into a sharded data set consisting of 1024 and 128 TFRecord files, respectively. train_directory/train-00000-of-01024 train_directory/train-00001-of-01024 ... train_directory/train-01023-of-01024 and validation_directory/validation-00000-of-00128 validation_directory/validation-00001-of-00128 ... validation_directory/validation-00127-of-00128 Each validation TFRecord file contains ~390 records. Each training TFREcord file contains ~1250 records. Each record within the TFRecord file is a serialized Example proto. The Example proto contains the following fields: image/encoded: string containing JPEG encoded image in RGB colorspace image/height: integer, image height in pixels image/width: integer, image width in pixels image/colorspace: string, specifying the colorspace, always 'RGB' image/channels: integer, specifying the number of channels, always 3 image/format: string, specifying the format, always 'JPEG' image/filename: string containing the basename of the image file e.g. 'n01440764_10026.JPEG' or 'ILSVRC2012_val_00000293.JPEG' image/class/label: integer specifying the index in a classification layer. The label ranges from [1, 1000] where 0 is not used. image/class/synset: string specifying the unique ID of the label, e.g. 'n01440764' image/class/text: string specifying the human-readable version of the label e.g. 'red fox, Vulpes vulpes' image/object/bbox/xmin: list of integers specifying the 0+ human annotated bounding boxes image/object/bbox/xmax: list of integers specifying the 0+ human annotated bounding boxes image/object/bbox/ymin: list of integers specifying the 0+ human annotated bounding boxes image/object/bbox/ymax: list of integers specifying the 0+ human annotated bounding boxes image/object/bbox/label: integer specifying the index in a classification layer. The label ranges from [1, 1000] where 0 is not used. Note this is always identical to the image label. Note that the length of xmin is identical to the length of xmax, ymin and ymax for each example. Running this script using 16 threads may take around ~2.5 hours on an HP Z420. """ from __future__ import absolute_import from __future__ import division from __future__ import print_function from datetime import datetime import os import random import sys import threading import numpy as np import tensorflow as tf tf.app.flags.DEFINE_string('train_directory', '/tmp/', 'Training data directory') tf.app.flags.DEFINE_string('validation_directory', '/tmp/', 'Validation data directory') tf.app.flags.DEFINE_string('output_directory', '/tmp/', 'Output data directory') tf.app.flags.DEFINE_integer('train_shards', 1024, 'Number of shards in training TFRecord files.') tf.app.flags.DEFINE_integer('validation_shards', 128, 'Number of shards in validation TFRecord files.') tf.app.flags.DEFINE_integer('num_threads', 8, 'Number of threads to preprocess the images.') # The labels file contains a list of valid labels are held in this file. # Assumes that the file contains entries as such: # n01440764 # n01443537 # n01484850 # where each line corresponds to a label expressed as a synset. We map # each synset contained in the file to an integer (based on the alphabetical # ordering). See below for details. tf.app.flags.DEFINE_string('labels_file', 'imagenet_lsvrc_2015_synsets.txt', 'Labels file') # This file containing mapping from synset to human-readable label. # Assumes each line of the file looks like: # # n02119247 black fox # n02119359 silver fox # n02119477 red fox, Vulpes fulva # # where each line corresponds to a unique mapping. Note that each line is # formatted as <synset>\t<human readable label>. tf.app.flags.DEFINE_string('imagenet_metadata_file', 'imagenet_metadata.txt', 'ImageNet metadata file') # This file is the output of process_bounding_box.py # Assumes each line of the file looks like: # # n00007846_64193.JPEG,0.0060,0.2620,0.7545,0.9940 # # where each line corresponds to one bounding box annotation associated # with an image. Each line can be parsed as: # # <JPEG file name>, <xmin>, <ymin>, <xmax>, <ymax> # # Note that there might exist mulitple bounding box annotations associated # with an image file. tf.app.flags.DEFINE_string('bounding_box_file', './imagenet_2012_bounding_boxes.csv', 'Bounding box file') FLAGS = tf.app.flags.FLAGS def _int64_feature(value): """Wrapper for inserting int64 features into Example proto.""" if not isinstance(value, list): value = [value] return tf.train.Feature(int64_list=tf.train.Int64List(value=value)) def _float_feature(value): """Wrapper for inserting float features into Example proto.""" if not isinstance(value, list): value = [value] return tf.train.Feature(float_list=tf.train.FloatList(value=value)) def _bytes_feature(value): """Wrapper for inserting bytes features into Example proto.""" return tf.train.Feature(bytes_list=tf.train.BytesList(value=[value])) def _convert_to_example(filename, image_buffer, label, synset, human, bbox, height, width): """Build an Example proto for an example. Args: filename: string, path to an image file, e.g., '/path/to/example.JPG' image_buffer: string, JPEG encoding of RGB image label: integer, identifier for the ground truth for the network synset: string, unique WordNet ID specifying the label, e.g., 'n02323233' human: string, human-readable label, e.g., 'red fox, Vulpes vulpes' bbox: list of bounding boxes; each box is a list of integers specifying [xmin, ymin, xmax, ymax]. All boxes are assumed to belong to the same label as the image label. height: integer, image height in pixels width: integer, image width in pixels Returns: Example proto """ xmin = [] ymin = [] xmax = [] ymax = [] for b in bbox: assert len(b) == 4 # pylint: disable=expression-not-assigned [l.append(point) for l, point in zip([xmin, ymin, xmax, ymax], b)] # pylint: enable=expression-not-assigned colorspace = 'RGB' channels = 3 image_format = 'JPEG' example = tf.train.Example(features=tf.train.Features(feature={ 'image/height': _int64_feature(height), 'image/width': _int64_feature(width), 'image/colorspace': _bytes_feature(colorspace), 'image/channels': _int64_feature(channels), 'image/class/label': _int64_feature(label), 'image/class/synset': _bytes_feature(synset), 'image/class/text': _bytes_feature(human), 'image/object/bbox/xmin': _float_feature(xmin), 'image/object/bbox/xmax': _float_feature(xmax), 'image/object/bbox/ymin': _float_feature(ymin), 'image/object/bbox/ymax': _float_feature(ymax), 'image/object/bbox/label': _int64_feature([label] * len(xmin)), 'image/format': _bytes_feature(image_format), 'image/filename': _bytes_feature(os.path.basename(filename)), 'image/encoded': _bytes_feature(image_buffer)})) return example class ImageCoder(object): """Helper class that provides TensorFlow image coding utilities.""" def __init__(self): # Create a single Session to run all image coding calls. self._sess = tf.Session() # Initializes function that converts PNG to JPEG data. self._png_data = tf.placeholder(dtype=tf.string) image = tf.image.decode_png(self._png_data, channels=3) self._png_to_jpeg = tf.image.encode_jpeg(image, format='rgb', quality=100) # Initializes function that converts CMYK JPEG data to RGB JPEG data. self._cmyk_data = tf.placeholder(dtype=tf.string) image = tf.image.decode_jpeg(self._cmyk_data, channels=0) self._cmyk_to_rgb = tf.image.encode_jpeg(image, format='rgb', quality=100) # Initializes function that decodes RGB JPEG data. self._decode_jpeg_data = tf.placeholder(dtype=tf.string) self._decode_jpeg = tf.image.decode_jpeg(self._decode_jpeg_data, channels=3) def png_to_jpeg(self, image_data): return self._sess.run(self._png_to_jpeg, feed_dict={self._png_data: image_data}) def cmyk_to_rgb(self, image_data): return self._sess.run(self._cmyk_to_rgb, feed_dict={self._cmyk_data: image_data}) def decode_jpeg(self, image_data): image = self._sess.run(self._decode_jpeg, feed_dict={self._decode_jpeg_data: image_data}) assert len(image.shape) == 3 assert image.shape[2] == 3 return image def _is_png(filename): """Determine if a file contains a PNG format image. Args: filename: string, path of the image file. Returns: boolean indicating if the image is a PNG. """ # File list from: # https://groups.google.com/forum/embed/?place=forum/torch7#!topic/torch7/fOSTXHIESSU return 'n02105855_2933.JPEG' in filename def _is_cmyk(filename): """Determine if file contains a CMYK JPEG format image. Args: filename: string, path of the image file. Returns: boolean indicating if the image is a JPEG encoded with CMYK color space. """ # File list from: # https://github.com/cytsai/ilsvrc-cmyk-image-list blacklist = ['n01739381_1309.JPEG', 'n02077923_14822.JPEG', 'n02447366_23489.JPEG', 'n02492035_15739.JPEG', 'n02747177_10752.JPEG', 'n03018349_4028.JPEG', 'n03062245_4620.JPEG', 'n03347037_9675.JPEG', 'n03467068_12171.JPEG', 'n03529860_11437.JPEG', 'n03544143_17228.JPEG', 'n03633091_5218.JPEG', 'n03710637_5125.JPEG', 'n03961711_5286.JPEG', 'n04033995_2932.JPEG', 'n04258138_17003.JPEG', 'n04264628_27969.JPEG', 'n04336792_7448.JPEG', 'n04371774_5854.JPEG', 'n04596742_4225.JPEG', 'n07583066_647.JPEG', 'n13037406_4650.JPEG'] return filename.split('/')[-1] in blacklist def _process_image(filename, coder): """Process a single image file. Args: filename: string, path to an image file e.g., '/path/to/example.JPG'. coder: instance of ImageCoder to provide TensorFlow image coding utils. Returns: image_buffer: string, JPEG encoding of RGB image. height: integer, image height in pixels. width: integer, image width in pixels. """ # Read the image file. with tf.gfile.FastGFile(filename, 'r') as f: image_data = f.read() # Clean the dirty data. if _is_png(filename): # 1 image is a PNG. print('Converting PNG to JPEG for %s' % filename) image_data = coder.png_to_jpeg(image_data) elif _is_cmyk(filename): # 22 JPEG images are in CMYK colorspace. print('Converting CMYK to RGB for %s' % filename) image_data = coder.cmyk_to_rgb(image_data) # Decode the RGB JPEG. image = coder.decode_jpeg(image_data) # Check that image converted to RGB assert len(image.shape) == 3 height = image.shape[0] width = image.shape[1] assert image.shape[2] == 3 return image_data, height, width def _process_image_files_batch(coder, thread_index, ranges, name, filenames, synsets, labels, humans, bboxes, num_shards): """Processes and saves list of images as TFRecord in 1 thread. Args: coder: instance of ImageCoder to provide TensorFlow image coding utils. thread_index: integer, unique batch to run index is within [0, len(ranges)). ranges: list of pairs of integers specifying ranges of each batches to analyze in parallel. name: string, unique identifier specifying the data set filenames: list of strings; each string is a path to an image file synsets: list of strings; each string is a unique WordNet ID labels: list of integer; each integer identifies the ground truth humans: list of strings; each string is a human-readable label bboxes: list of bounding boxes for each image. Note that each entry in this list might contain from 0+ entries corresponding to the number of bounding box annotations for the image. num_shards: integer number of shards for this data set. """ # Each thread produces N shards where N = int(num_shards / num_threads). # For instance, if num_shards = 128, and the num_threads = 2, then the first # thread would produce shards [0, 64). num_threads = len(ranges) assert not num_shards % num_threads num_shards_per_batch = int(num_shards / num_threads) shard_ranges = np.linspace(ranges[thread_index][0], ranges[thread_index][1], num_shards_per_batch + 1).astype(int) num_files_in_thread = ranges[thread_index][1] - ranges[thread_index][0] counter = 0 for s in range(num_shards_per_batch): # Generate a sharded version of the file name, e.g. 'train-00002-of-00010' shard = thread_index * num_shards_per_batch + s output_filename = '%s-%.5d-of-%.5d' % (name, shard, num_shards) output_file = os.path.join(FLAGS.output_directory, output_filename) writer = tf.python_io.TFRecordWriter(output_file) shard_counter = 0 files_in_shard = np.arange(shard_ranges[s], shard_ranges[s + 1], dtype=int) for i in files_in_shard: filename = filenames[i] label = labels[i] synset = synsets[i] human = humans[i] bbox = bboxes[i] image_buffer, height, width = _process_image(filename, coder) example = _convert_to_example(filename, image_buffer, label, synset, human, bbox, height, width) writer.write(example.SerializeToString()) shard_counter += 1 counter += 1 if not counter % 1000: print('%s [thread %d]: Processed %d of %d images in thread batch.' % (datetime.now(), thread_index, counter, num_files_in_thread)) sys.stdout.flush() writer.close() print('%s [thread %d]: Wrote %d images to %s' % (datetime.now(), thread_index, shard_counter, output_file)) sys.stdout.flush() shard_counter = 0 print('%s [thread %d]: Wrote %d images to %d shards.' % (datetime.now(), thread_index, counter, num_files_in_thread)) sys.stdout.flush() def _process_image_files(name, filenames, synsets, labels, humans, bboxes, num_shards): """Process and save list of images as TFRecord of Example protos. Args: name: string, unique identifier specifying the data set filenames: list of strings; each string is a path to an image file synsets: list of strings; each string is a unique WordNet ID labels: list of integer; each integer identifies the ground truth humans: list of strings; each string is a human-readable label bboxes: list of bounding boxes for each image. Note that each entry in this list might contain from 0+ entries corresponding to the number of bounding box annotations for the image. num_shards: integer number of shards for this data set. """ assert len(filenames) == len(synsets) assert len(filenames) == len(labels) assert len(filenames) == len(humans) assert len(filenames) == len(bboxes) # Break all images into batches with a [ranges[i][0], ranges[i][1]]. spacing = np.linspace(0, len(filenames), FLAGS.num_threads + 1).astype(np.int) ranges = [] threads = [] for i in range(len(spacing) - 1): ranges.append([spacing[i], spacing[i + 1]]) # Launch a thread for each batch. print('Launching %d threads for spacings: %s' % (FLAGS.num_threads, ranges)) sys.stdout.flush() # Create a mechanism for monitoring when all threads are finished. coord = tf.train.Coordinator() # Create a generic TensorFlow-based utility for converting all image codings. coder = ImageCoder() threads = [] for thread_index in range(len(ranges)): args = (coder, thread_index, ranges, name, filenames, synsets, labels, humans, bboxes, num_shards) t = threading.Thread(target=_process_image_files_batch, args=args) t.start() threads.append(t) # Wait for all the threads to terminate. coord.join(threads) print('%s: Finished writing all %d images in data set.' % (datetime.now(), len(filenames))) sys.stdout.flush() def _find_image_files(data_dir, labels_file): """Build a list of all images files and labels in the data set. Args: data_dir: string, path to the root directory of images. Assumes that the ImageNet data set resides in JPEG files located in the following directory structure. data_dir/n01440764/ILSVRC2012_val_00000293.JPEG data_dir/n01440764/ILSVRC2012_val_00000543.JPEG where 'n01440764' is the unique synset label associated with these images. labels_file: string, path to the labels file. The list of valid labels are held in this file. Assumes that the file contains entries as such: n01440764 n01443537 n01484850 where each line corresponds to a label expressed as a synset. We map each synset contained in the file to an integer (based on the alphabetical ordering) starting with the integer 1 corresponding to the synset contained in the first line. The reason we start the integer labels at 1 is to reserve label 0 as an unused background class. Returns: filenames: list of strings; each string is a path to an image file. synsets: list of strings; each string is a unique WordNet ID. labels: list of integer; each integer identifies the ground truth. """ print('Determining list of input files and labels from %s.' % data_dir) challenge_synsets = [l.strip() for l in tf.gfile.FastGFile(labels_file, 'r').readlines()] labels = [] filenames = [] synsets = [] # Leave label index 0 empty as a background class. label_index = 1 # Construct the list of JPEG files and labels. for synset in challenge_synsets: jpeg_file_path = '%s/%s/*.JPEG' % (data_dir, synset) matching_files = tf.gfile.Glob(jpeg_file_path) labels.extend([label_index] * len(matching_files)) synsets.extend([synset] * len(matching_files)) filenames.extend(matching_files) if not label_index % 100: print('Finished finding files in %d of %d classes.' % ( label_index, len(challenge_synsets))) label_index += 1 # Shuffle the ordering of all image files in order to guarantee # random ordering of the images with respect to label in the # saved TFRecord files. Make the randomization repeatable. shuffled_index = list(range(len(filenames))) random.seed(12345) random.shuffle(shuffled_index) filenames = [filenames[i] for i in shuffled_index] synsets = [synsets[i] for i in shuffled_index] labels = [labels[i] for i in shuffled_index] print('Found %d JPEG files across %d labels inside %s.' % (len(filenames), len(challenge_synsets), data_dir)) return filenames, synsets, labels def _find_human_readable_labels(synsets, synset_to_human): """Build a list of human-readable labels. Args: synsets: list of strings; each string is a unique WordNet ID. synset_to_human: dict of synset to human labels, e.g., 'n02119022' --> 'red fox, Vulpes vulpes' Returns: List of human-readable strings corresponding to each synset. """ humans = [] for s in synsets: assert s in synset_to_human, ('Failed to find: %s' % s) humans.append(synset_to_human[s]) return humans def _find_image_bounding_boxes(filenames, image_to_bboxes): """Find the bounding boxes for a given image file. Args: filenames: list of strings; each string is a path to an image file. image_to_bboxes: dictionary mapping image file names to a list of bounding boxes. This list contains 0+ bounding boxes. Returns: List of bounding boxes for each image. Note that each entry in this list might contain from 0+ entries corresponding to the number of bounding box annotations for the image. """ num_image_bbox = 0 bboxes = [] for f in filenames: basename = os.path.basename(f) if basename in image_to_bboxes: bboxes.append(image_to_bboxes[basename]) num_image_bbox += 1 else: bboxes.append([]) print('Found %d images with bboxes out of %d images' % ( num_image_bbox, len(filenames))) return bboxes def _process_dataset(name, directory, num_shards, synset_to_human, image_to_bboxes): """Process a complete data set and save it as a TFRecord. Args: name: string, unique identifier specifying the data set. directory: string, root path to the data set. num_shards: integer number of shards for this data set. synset_to_human: dict of synset to human labels, e.g., 'n02119022' --> 'red fox, Vulpes vulpes' image_to_bboxes: dictionary mapping image file names to a list of bounding boxes. This list contains 0+ bounding boxes. """ filenames, synsets, labels = _find_image_files(directory, FLAGS.labels_file) humans = _find_human_readable_labels(synsets, synset_to_human) bboxes = _find_image_bounding_boxes(filenames, image_to_bboxes) _process_image_files(name, filenames, synsets, labels, humans, bboxes, num_shards) def _build_synset_lookup(imagenet_metadata_file): """Build lookup for synset to human-readable label. Args: imagenet_metadata_file: string, path to file containing mapping from synset to human-readable label. Assumes each line of the file looks like: n02119247 black fox n02119359 silver fox n02119477 red fox, Vulpes fulva where each line corresponds to a unique mapping. Note that each line is formatted as <synset>\t<human readable label>. Returns: Dictionary of synset to human labels, such as: 'n02119022' --> 'red fox, Vulpes vulpes' """ lines = tf.gfile.FastGFile(imagenet_metadata_file, 'r').readlines() synset_to_human = {} for l in lines: if l: parts = l.strip().split('\t') assert len(parts) == 2 synset = parts[0] human = parts[1] synset_to_human[synset] = human return synset_to_human def _build_bounding_box_lookup(bounding_box_file): """Build a lookup from image file to bounding boxes. Args: bounding_box_file: string, path to file with bounding boxes annotations. Assumes each line of the file looks like: n00007846_64193.JPEG,0.0060,0.2620,0.7545,0.9940 where each line corresponds to one bounding box annotation associated with an image. Each line can be parsed as: <JPEG file name>, <xmin>, <ymin>, <xmax>, <ymax> Note that there might exist mulitple bounding box annotations associated with an image file. This file is the output of process_bounding_boxes.py. Returns: Dictionary mapping image file names to a list of bounding boxes. This list contains 0+ bounding boxes. """ lines = tf.gfile.FastGFile(bounding_box_file, 'r').readlines() images_to_bboxes = {} num_bbox = 0 num_image = 0 for l in lines: if l: parts = l.split(',') assert len(parts) == 5, ('Failed to parse: %s' % l) filename = parts[0] xmin = float(parts[1]) ymin = float(parts[2]) xmax = float(parts[3]) ymax = float(parts[4]) box = [xmin, ymin, xmax, ymax] if filename not in images_to_bboxes: images_to_bboxes[filename] = [] num_image += 1 images_to_bboxes[filename].append(box) num_bbox += 1 print('Successfully read %d bounding boxes ' 'across %d images.' % (num_bbox, num_image)) return images_to_bboxes def main(unused_argv): assert not FLAGS.train_shards % FLAGS.num_threads, ( 'Please make the FLAGS.num_threads commensurate with FLAGS.train_shards') assert not FLAGS.validation_shards % FLAGS.num_threads, ( 'Please make the FLAGS.num_threads commensurate with ' 'FLAGS.validation_shards') print('Saving results to %s' % FLAGS.output_directory) # Build a map from synset to human-readable label. synset_to_human = _build_synset_lookup(FLAGS.imagenet_metadata_file) image_to_bboxes = _build_bounding_box_lookup(FLAGS.bounding_box_file) # Run it! _process_dataset('validation', FLAGS.validation_directory, FLAGS.validation_shards, synset_to_human, image_to_bboxes) _process_dataset('train', FLAGS.train_directory, FLAGS.train_shards, synset_to_human, image_to_bboxes) if __name__ == '__main__': tf.app.run()
mit
sgenoud/scikit-learn
examples/svm/plot_custom_kernel.py
3
1522
""" ====================== SVM with custom kernel ====================== Simple usage of Support Vector Machines to classify a sample. It will plot the decision surface and the support vectors. """ print __doc__ import numpy as np import pylab as pl from sklearn import svm, datasets # import some data to play with iris = datasets.load_iris() X = iris.data[:, :2] # we only take the first two features. We could # avoid this ugly slicing by using a two-dim dataset Y = iris.target def my_kernel(x, y): """ We create a custom kernel: (2 0) k(x, y) = x ( ) y.T (0 1) """ M = np.array([[2, 0], [0, 1.0]]) return np.dot(np.dot(x, M), y.T) h = .02 # step size in the mesh # we create an instance of SVM and fit out data. clf = svm.SVC(kernel=my_kernel) clf.fit(X, Y) # Plot the decision boundary. For that, we will asign a color to each # point in the mesh [x_min, m_max]x[y_min, y_max]. x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1 y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1 xx, yy = np.meshgrid(np.arange(x_min, x_max, h), np.arange(y_min, y_max, h)) Z = clf.predict(np.c_[xx.ravel(), yy.ravel()]) # Put the result into a color plot Z = Z.reshape(xx.shape) pl.pcolormesh(xx, yy, Z, cmap=pl.cm.Paired) # Plot also the training points pl.scatter(X[:, 0], X[:, 1], c=Y, cmap=pl.cm.Paired) pl.title('3-Class classification using Support Vector Machine with custom' ' kernel') pl.axis('tight') pl.show()
bsd-3-clause
glouppe/scikit-learn
sklearn/decomposition/tests/test_nmf.py
25
8544
import numpy as np from scipy import linalg from sklearn.decomposition import (NMF, ProjectedGradientNMF, non_negative_factorization) from sklearn.decomposition import nmf # For testing internals from scipy.sparse import csc_matrix from sklearn.utils.testing import assert_true from sklearn.utils.testing import assert_false from sklearn.utils.testing import assert_raise_message from sklearn.utils.testing import assert_array_almost_equal from sklearn.utils.testing import assert_almost_equal from sklearn.utils.testing import assert_greater from sklearn.utils.testing import assert_less from sklearn.utils.testing import ignore_warnings from sklearn.base import clone random_state = np.random.mtrand.RandomState(0) def test_initialize_nn_output(): # Test that initialization does not return negative values data = np.abs(random_state.randn(10, 10)) for init in ('random', 'nndsvd', 'nndsvda', 'nndsvdar'): W, H = nmf._initialize_nmf(data, 10, init=init, random_state=0) assert_false((W < 0).any() or (H < 0).any()) @ignore_warnings def test_parameter_checking(): A = np.ones((2, 2)) name = 'spam' msg = "Invalid solver parameter: got 'spam' instead of one of" assert_raise_message(ValueError, msg, NMF(solver=name).fit, A) msg = "Invalid init parameter: got 'spam' instead of one of" assert_raise_message(ValueError, msg, NMF(init=name).fit, A) msg = "Invalid sparseness parameter: got 'spam' instead of one of" assert_raise_message(ValueError, msg, NMF(sparseness=name).fit, A) msg = "Negative values in data passed to" assert_raise_message(ValueError, msg, NMF().fit, -A) assert_raise_message(ValueError, msg, nmf._initialize_nmf, -A, 2, 'nndsvd') clf = NMF(2, tol=0.1).fit(A) assert_raise_message(ValueError, msg, clf.transform, -A) def test_initialize_close(): # Test NNDSVD error # Test that _initialize_nmf error is less than the standard deviation of # the entries in the matrix. A = np.abs(random_state.randn(10, 10)) W, H = nmf._initialize_nmf(A, 10, init='nndsvd') error = linalg.norm(np.dot(W, H) - A) sdev = linalg.norm(A - A.mean()) assert_true(error <= sdev) def test_initialize_variants(): # Test NNDSVD variants correctness # Test that the variants 'nndsvda' and 'nndsvdar' differ from basic # 'nndsvd' only where the basic version has zeros. data = np.abs(random_state.randn(10, 10)) W0, H0 = nmf._initialize_nmf(data, 10, init='nndsvd') Wa, Ha = nmf._initialize_nmf(data, 10, init='nndsvda') War, Har = nmf._initialize_nmf(data, 10, init='nndsvdar', random_state=0) for ref, evl in ((W0, Wa), (W0, War), (H0, Ha), (H0, Har)): assert_true(np.allclose(evl[ref != 0], ref[ref != 0])) @ignore_warnings def test_nmf_fit_nn_output(): # Test that the decomposition does not contain negative values A = np.c_[5 * np.ones(5) - np.arange(1, 6), 5 * np.ones(5) + np.arange(1, 6)] for solver in ('pg', 'cd'): for init in (None, 'nndsvd', 'nndsvda', 'nndsvdar'): model = NMF(n_components=2, solver=solver, init=init, random_state=0) transf = model.fit_transform(A) assert_false((model.components_ < 0).any() or (transf < 0).any()) @ignore_warnings def test_nmf_fit_close(): # Test that the fit is not too far away for solver in ('pg', 'cd'): pnmf = NMF(5, solver=solver, init='nndsvd', random_state=0) X = np.abs(random_state.randn(6, 5)) assert_less(pnmf.fit(X).reconstruction_err_, 0.05) def test_nls_nn_output(): # Test that NLS solver doesn't return negative values A = np.arange(1, 5).reshape(1, -1) Ap, _, _ = nmf._nls_subproblem(np.dot(A.T, -A), A.T, A, 0.001, 100) assert_false((Ap < 0).any()) def test_nls_close(): # Test that the NLS results should be close A = np.arange(1, 5).reshape(1, -1) Ap, _, _ = nmf._nls_subproblem(np.dot(A.T, A), A.T, np.zeros_like(A), 0.001, 100) assert_true((np.abs(Ap - A) < 0.01).all()) @ignore_warnings def test_nmf_transform(): # Test that NMF.transform returns close values A = np.abs(random_state.randn(6, 5)) for solver in ('pg', 'cd'): m = NMF(solver=solver, n_components=4, init='nndsvd', random_state=0) ft = m.fit_transform(A) t = m.transform(A) assert_array_almost_equal(ft, t, decimal=2) @ignore_warnings def test_n_components_greater_n_features(): # Smoke test for the case of more components than features. A = np.abs(random_state.randn(30, 10)) NMF(n_components=15, random_state=0, tol=1e-2).fit(A) @ignore_warnings def test_projgrad_nmf_sparseness(): # Test sparseness # Test that sparsity constraints actually increase sparseness in the # part where they are applied. tol = 1e-2 A = np.abs(random_state.randn(10, 10)) m = ProjectedGradientNMF(n_components=5, random_state=0, tol=tol).fit(A) data_sp = ProjectedGradientNMF(n_components=5, sparseness='data', random_state=0, tol=tol).fit(A).data_sparseness_ comp_sp = ProjectedGradientNMF(n_components=5, sparseness='components', random_state=0, tol=tol).fit(A).comp_sparseness_ assert_greater(data_sp, m.data_sparseness_) assert_greater(comp_sp, m.comp_sparseness_) @ignore_warnings def test_sparse_input(): # Test that sparse matrices are accepted as input from scipy.sparse import csc_matrix A = np.abs(random_state.randn(10, 10)) A[:, 2 * np.arange(5)] = 0 A_sparse = csc_matrix(A) for solver in ('pg', 'cd'): est1 = NMF(solver=solver, n_components=5, init='random', random_state=0, tol=1e-2) est2 = clone(est1) W1 = est1.fit_transform(A) W2 = est2.fit_transform(A_sparse) H1 = est1.components_ H2 = est2.components_ assert_array_almost_equal(W1, W2) assert_array_almost_equal(H1, H2) @ignore_warnings def test_sparse_transform(): # Test that transform works on sparse data. Issue #2124 A = np.abs(random_state.randn(3, 2)) A[A > 1.0] = 0 A = csc_matrix(A) for solver in ('pg', 'cd'): model = NMF(solver=solver, random_state=0, tol=1e-4, n_components=2) A_fit_tr = model.fit_transform(A) A_tr = model.transform(A) assert_array_almost_equal(A_fit_tr, A_tr, decimal=1) @ignore_warnings def test_non_negative_factorization_consistency(): # Test that the function is called in the same way, either directly # or through the NMF class A = np.abs(random_state.randn(10, 10)) A[:, 2 * np.arange(5)] = 0 for solver in ('pg', 'cd'): W_nmf, H, _ = non_negative_factorization( A, solver=solver, random_state=1, tol=1e-2) W_nmf_2, _, _ = non_negative_factorization( A, H=H, update_H=False, solver=solver, random_state=1, tol=1e-2) model_class = NMF(solver=solver, random_state=1, tol=1e-2) W_cls = model_class.fit_transform(A) W_cls_2 = model_class.transform(A) assert_array_almost_equal(W_nmf, W_cls, decimal=10) assert_array_almost_equal(W_nmf_2, W_cls_2, decimal=10) @ignore_warnings def test_non_negative_factorization_checking(): A = np.ones((2, 2)) # Test parameters checking is public function nnmf = non_negative_factorization msg = "Number of components must be positive; got (n_components='2')" assert_raise_message(ValueError, msg, nnmf, A, A, A, '2') msg = "Negative values in data passed to NMF (input H)" assert_raise_message(ValueError, msg, nnmf, A, A, -A, 2, 'custom') msg = "Negative values in data passed to NMF (input W)" assert_raise_message(ValueError, msg, nnmf, A, -A, A, 2, 'custom') msg = "Array passed to NMF (input H) is full of zeros" assert_raise_message(ValueError, msg, nnmf, A, A, 0 * A, 2, 'custom') def test_safe_compute_error(): A = np.abs(random_state.randn(10, 10)) A[:, 2 * np.arange(5)] = 0 A_sparse = csc_matrix(A) W, H = nmf._initialize_nmf(A, 5, init='random', random_state=0) error = nmf._safe_compute_error(A, W, H) error_sparse = nmf._safe_compute_error(A_sparse, W, H) assert_almost_equal(error, error_sparse)
bsd-3-clause
elkingtonmcb/h2o-2
py/testdir_single_jvm/test_rf_histo_fail_fvec.py
9
1394
import unittest, random, sys, time sys.path.extend(['.','..','../..','py']) import h2o, h2o_cmd, h2o_rf, h2o_import as h2i paramDict = { 'destination_key': 'model_keyA', 'ntrees': 13, 'response': 'C55', 'mtries': 3, 'source': u'covtype.hex', 'seed': '1231231', 'importance': 0, 'balance_classes': 0, } class Basic(unittest.TestCase): def tearDown(self): h2o.check_sandbox_for_errors() @classmethod def setUpClass(cls): global SEED SEED = h2o.setup_random_seed() h2o.init(java_heap_GB=10) @classmethod def tearDownClass(cls): h2o.tear_down_cloud() def test_rf_histo_fail_fvec(self): csvPathname = 'standard/covtype.data' for trial in range(3): kwargs = paramDict.copy() timeoutSecs = 180 start = time.time() parseResult = h2i.import_parse(bucket='home-0xdiag-datasets', path=csvPathname, schema='put') rfSeed = random.randint(0, sys.maxint) kwargs.update({'seed': rfSeed}) h2o_cmd.runRF(parseResult=parseResult, timeoutSecs=timeoutSecs, retryDelaySecs=1, **kwargs) elapsed = time.time()-start print "Trial #", trial, "completed in", elapsed, "seconds.", "%d pct. of timeout" % ((elapsed*100)/timeoutSecs) if __name__ == '__main__': h2o.unit_main()
apache-2.0
MohammedWasim/scikit-learn
sklearn/metrics/cluster/supervised.py
206
27395
"""Utilities to evaluate the clustering performance of models Functions named as *_score return a scalar value to maximize: the higher the better. """ # Authors: Olivier Grisel <olivier.grisel@ensta.org> # Wei LI <kuantkid@gmail.com> # Diego Molla <dmolla-aliod@gmail.com> # License: BSD 3 clause from math import log from scipy.misc import comb from scipy.sparse import coo_matrix import numpy as np from .expected_mutual_info_fast import expected_mutual_information from ...utils.fixes import bincount def comb2(n): # the exact version is faster for k == 2: use it by default globally in # this module instead of the float approximate variant return comb(n, 2, exact=1) def check_clusterings(labels_true, labels_pred): """Check that the two clusterings matching 1D integer arrays""" labels_true = np.asarray(labels_true) labels_pred = np.asarray(labels_pred) # input checks if labels_true.ndim != 1: raise ValueError( "labels_true must be 1D: shape is %r" % (labels_true.shape,)) if labels_pred.ndim != 1: raise ValueError( "labels_pred must be 1D: shape is %r" % (labels_pred.shape,)) if labels_true.shape != labels_pred.shape: raise ValueError( "labels_true and labels_pred must have same size, got %d and %d" % (labels_true.shape[0], labels_pred.shape[0])) return labels_true, labels_pred def contingency_matrix(labels_true, labels_pred, eps=None): """Build a contengency matrix describing the relationship between labels. Parameters ---------- labels_true : int array, shape = [n_samples] Ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] Cluster labels to evaluate eps: None or float If a float, that value is added to all values in the contingency matrix. This helps to stop NaN propagation. If ``None``, nothing is adjusted. Returns ------- contingency: array, shape=[n_classes_true, n_classes_pred] Matrix :math:`C` such that :math:`C_{i, j}` is the number of samples in true class :math:`i` and in predicted class :math:`j`. If ``eps is None``, the dtype of this array will be integer. If ``eps`` is given, the dtype will be float. """ classes, class_idx = np.unique(labels_true, return_inverse=True) clusters, cluster_idx = np.unique(labels_pred, return_inverse=True) n_classes = classes.shape[0] n_clusters = clusters.shape[0] # Using coo_matrix to accelerate simple histogram calculation, # i.e. bins are consecutive integers # Currently, coo_matrix is faster than histogram2d for simple cases contingency = coo_matrix((np.ones(class_idx.shape[0]), (class_idx, cluster_idx)), shape=(n_classes, n_clusters), dtype=np.int).toarray() if eps is not None: # don't use += as contingency is integer contingency = contingency + eps return contingency # clustering measures def adjusted_rand_score(labels_true, labels_pred): """Rand index adjusted for chance The Rand Index computes a similarity measure between two clusterings by considering all pairs of samples and counting pairs that are assigned in the same or different clusters in the predicted and true clusterings. The raw RI score is then "adjusted for chance" into the ARI score using the following scheme:: ARI = (RI - Expected_RI) / (max(RI) - Expected_RI) The adjusted Rand index is thus ensured to have a value close to 0.0 for random labeling independently of the number of clusters and samples and exactly 1.0 when the clusterings are identical (up to a permutation). ARI is a symmetric measure:: adjusted_rand_score(a, b) == adjusted_rand_score(b, a) Read more in the :ref:`User Guide <adjusted_rand_score>`. Parameters ---------- labels_true : int array, shape = [n_samples] Ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] Cluster labels to evaluate Returns ------- ari : float Similarity score between -1.0 and 1.0. Random labelings have an ARI close to 0.0. 1.0 stands for perfect match. Examples -------- Perfectly maching labelings have a score of 1 even >>> from sklearn.metrics.cluster import adjusted_rand_score >>> adjusted_rand_score([0, 0, 1, 1], [0, 0, 1, 1]) 1.0 >>> adjusted_rand_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 Labelings that assign all classes members to the same clusters are complete be not always pure, hence penalized:: >>> adjusted_rand_score([0, 0, 1, 2], [0, 0, 1, 1]) # doctest: +ELLIPSIS 0.57... ARI is symmetric, so labelings that have pure clusters with members coming from the same classes but unnecessary splits are penalized:: >>> adjusted_rand_score([0, 0, 1, 1], [0, 0, 1, 2]) # doctest: +ELLIPSIS 0.57... If classes members are completely split across different clusters, the assignment is totally incomplete, hence the ARI is very low:: >>> adjusted_rand_score([0, 0, 0, 0], [0, 1, 2, 3]) 0.0 References ---------- .. [Hubert1985] `L. Hubert and P. Arabie, Comparing Partitions, Journal of Classification 1985` http://www.springerlink.com/content/x64124718341j1j0/ .. [wk] http://en.wikipedia.org/wiki/Rand_index#Adjusted_Rand_index See also -------- adjusted_mutual_info_score: Adjusted Mutual Information """ labels_true, labels_pred = check_clusterings(labels_true, labels_pred) n_samples = labels_true.shape[0] classes = np.unique(labels_true) clusters = np.unique(labels_pred) # Special limit cases: no clustering since the data is not split; # or trivial clustering where each document is assigned a unique cluster. # These are perfect matches hence return 1.0. if (classes.shape[0] == clusters.shape[0] == 1 or classes.shape[0] == clusters.shape[0] == 0 or classes.shape[0] == clusters.shape[0] == len(labels_true)): return 1.0 contingency = contingency_matrix(labels_true, labels_pred) # Compute the ARI using the contingency data sum_comb_c = sum(comb2(n_c) for n_c in contingency.sum(axis=1)) sum_comb_k = sum(comb2(n_k) for n_k in contingency.sum(axis=0)) sum_comb = sum(comb2(n_ij) for n_ij in contingency.flatten()) prod_comb = (sum_comb_c * sum_comb_k) / float(comb(n_samples, 2)) mean_comb = (sum_comb_k + sum_comb_c) / 2. return ((sum_comb - prod_comb) / (mean_comb - prod_comb)) def homogeneity_completeness_v_measure(labels_true, labels_pred): """Compute the homogeneity and completeness and V-Measure scores at once Those metrics are based on normalized conditional entropy measures of the clustering labeling to evaluate given the knowledge of a Ground Truth class labels of the same samples. A clustering result satisfies homogeneity if all of its clusters contain only data points which are members of a single class. A clustering result satisfies completeness if all the data points that are members of a given class are elements of the same cluster. Both scores have positive values between 0.0 and 1.0, larger values being desirable. Those 3 metrics are independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score values in any way. V-Measure is furthermore symmetric: swapping ``labels_true`` and ``label_pred`` will give the same score. This does not hold for homogeneity and completeness. Read more in the :ref:`User Guide <homogeneity_completeness>`. Parameters ---------- labels_true : int array, shape = [n_samples] ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] cluster labels to evaluate Returns ------- homogeneity: float score between 0.0 and 1.0. 1.0 stands for perfectly homogeneous labeling completeness: float score between 0.0 and 1.0. 1.0 stands for perfectly complete labeling v_measure: float harmonic mean of the first two See also -------- homogeneity_score completeness_score v_measure_score """ labels_true, labels_pred = check_clusterings(labels_true, labels_pred) if len(labels_true) == 0: return 1.0, 1.0, 1.0 entropy_C = entropy(labels_true) entropy_K = entropy(labels_pred) MI = mutual_info_score(labels_true, labels_pred) homogeneity = MI / (entropy_C) if entropy_C else 1.0 completeness = MI / (entropy_K) if entropy_K else 1.0 if homogeneity + completeness == 0.0: v_measure_score = 0.0 else: v_measure_score = (2.0 * homogeneity * completeness / (homogeneity + completeness)) return homogeneity, completeness, v_measure_score def homogeneity_score(labels_true, labels_pred): """Homogeneity metric of a cluster labeling given a ground truth A clustering result satisfies homogeneity if all of its clusters contain only data points which are members of a single class. This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is not symmetric: switching ``label_true`` with ``label_pred`` will return the :func:`completeness_score` which will be different in general. Read more in the :ref:`User Guide <homogeneity_completeness>`. Parameters ---------- labels_true : int array, shape = [n_samples] ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] cluster labels to evaluate Returns ------- homogeneity: float score between 0.0 and 1.0. 1.0 stands for perfectly homogeneous labeling References ---------- .. [1] `Andrew Rosenberg and Julia Hirschberg, 2007. V-Measure: A conditional entropy-based external cluster evaluation measure <http://aclweb.org/anthology/D/D07/D07-1043.pdf>`_ See also -------- completeness_score v_measure_score Examples -------- Perfect labelings are homogeneous:: >>> from sklearn.metrics.cluster import homogeneity_score >>> homogeneity_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 Non-perfect labelings that further split classes into more clusters can be perfectly homogeneous:: >>> print("%.6f" % homogeneity_score([0, 0, 1, 1], [0, 0, 1, 2])) ... # doctest: +ELLIPSIS 1.0... >>> print("%.6f" % homogeneity_score([0, 0, 1, 1], [0, 1, 2, 3])) ... # doctest: +ELLIPSIS 1.0... Clusters that include samples from different classes do not make for an homogeneous labeling:: >>> print("%.6f" % homogeneity_score([0, 0, 1, 1], [0, 1, 0, 1])) ... # doctest: +ELLIPSIS 0.0... >>> print("%.6f" % homogeneity_score([0, 0, 1, 1], [0, 0, 0, 0])) ... # doctest: +ELLIPSIS 0.0... """ return homogeneity_completeness_v_measure(labels_true, labels_pred)[0] def completeness_score(labels_true, labels_pred): """Completeness metric of a cluster labeling given a ground truth A clustering result satisfies completeness if all the data points that are members of a given class are elements of the same cluster. This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is not symmetric: switching ``label_true`` with ``label_pred`` will return the :func:`homogeneity_score` which will be different in general. Read more in the :ref:`User Guide <homogeneity_completeness>`. Parameters ---------- labels_true : int array, shape = [n_samples] ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] cluster labels to evaluate Returns ------- completeness: float score between 0.0 and 1.0. 1.0 stands for perfectly complete labeling References ---------- .. [1] `Andrew Rosenberg and Julia Hirschberg, 2007. V-Measure: A conditional entropy-based external cluster evaluation measure <http://aclweb.org/anthology/D/D07/D07-1043.pdf>`_ See also -------- homogeneity_score v_measure_score Examples -------- Perfect labelings are complete:: >>> from sklearn.metrics.cluster import completeness_score >>> completeness_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 Non-perfect labelings that assign all classes members to the same clusters are still complete:: >>> print(completeness_score([0, 0, 1, 1], [0, 0, 0, 0])) 1.0 >>> print(completeness_score([0, 1, 2, 3], [0, 0, 1, 1])) 1.0 If classes members are split across different clusters, the assignment cannot be complete:: >>> print(completeness_score([0, 0, 1, 1], [0, 1, 0, 1])) 0.0 >>> print(completeness_score([0, 0, 0, 0], [0, 1, 2, 3])) 0.0 """ return homogeneity_completeness_v_measure(labels_true, labels_pred)[1] def v_measure_score(labels_true, labels_pred): """V-measure cluster labeling given a ground truth. This score is identical to :func:`normalized_mutual_info_score`. The V-measure is the harmonic mean between homogeneity and completeness:: v = 2 * (homogeneity * completeness) / (homogeneity + completeness) This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is furthermore symmetric: switching ``label_true`` with ``label_pred`` will return the same score value. This can be useful to measure the agreement of two independent label assignments strategies on the same dataset when the real ground truth is not known. Read more in the :ref:`User Guide <homogeneity_completeness>`. Parameters ---------- labels_true : int array, shape = [n_samples] ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] cluster labels to evaluate Returns ------- v_measure: float score between 0.0 and 1.0. 1.0 stands for perfectly complete labeling References ---------- .. [1] `Andrew Rosenberg and Julia Hirschberg, 2007. V-Measure: A conditional entropy-based external cluster evaluation measure <http://aclweb.org/anthology/D/D07/D07-1043.pdf>`_ See also -------- homogeneity_score completeness_score Examples -------- Perfect labelings are both homogeneous and complete, hence have score 1.0:: >>> from sklearn.metrics.cluster import v_measure_score >>> v_measure_score([0, 0, 1, 1], [0, 0, 1, 1]) 1.0 >>> v_measure_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 Labelings that assign all classes members to the same clusters are complete be not homogeneous, hence penalized:: >>> print("%.6f" % v_measure_score([0, 0, 1, 2], [0, 0, 1, 1])) ... # doctest: +ELLIPSIS 0.8... >>> print("%.6f" % v_measure_score([0, 1, 2, 3], [0, 0, 1, 1])) ... # doctest: +ELLIPSIS 0.66... Labelings that have pure clusters with members coming from the same classes are homogeneous but un-necessary splits harms completeness and thus penalize V-measure as well:: >>> print("%.6f" % v_measure_score([0, 0, 1, 1], [0, 0, 1, 2])) ... # doctest: +ELLIPSIS 0.8... >>> print("%.6f" % v_measure_score([0, 0, 1, 1], [0, 1, 2, 3])) ... # doctest: +ELLIPSIS 0.66... If classes members are completely split across different clusters, the assignment is totally incomplete, hence the V-Measure is null:: >>> print("%.6f" % v_measure_score([0, 0, 0, 0], [0, 1, 2, 3])) ... # doctest: +ELLIPSIS 0.0... Clusters that include samples from totally different classes totally destroy the homogeneity of the labeling, hence:: >>> print("%.6f" % v_measure_score([0, 0, 1, 1], [0, 0, 0, 0])) ... # doctest: +ELLIPSIS 0.0... """ return homogeneity_completeness_v_measure(labels_true, labels_pred)[2] def mutual_info_score(labels_true, labels_pred, contingency=None): """Mutual Information between two clusterings The Mutual Information is a measure of the similarity between two labels of the same data. Where :math:`P(i)` is the probability of a random sample occurring in cluster :math:`U_i` and :math:`P'(j)` is the probability of a random sample occurring in cluster :math:`V_j`, the Mutual Information between clusterings :math:`U` and :math:`V` is given as: .. math:: MI(U,V)=\sum_{i=1}^R \sum_{j=1}^C P(i,j)\log\\frac{P(i,j)}{P(i)P'(j)} This is equal to the Kullback-Leibler divergence of the joint distribution with the product distribution of the marginals. This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is furthermore symmetric: switching ``label_true`` with ``label_pred`` will return the same score value. This can be useful to measure the agreement of two independent label assignments strategies on the same dataset when the real ground truth is not known. Read more in the :ref:`User Guide <mutual_info_score>`. Parameters ---------- labels_true : int array, shape = [n_samples] A clustering of the data into disjoint subsets. labels_pred : array, shape = [n_samples] A clustering of the data into disjoint subsets. contingency: None or array, shape = [n_classes_true, n_classes_pred] A contingency matrix given by the :func:`contingency_matrix` function. If value is ``None``, it will be computed, otherwise the given value is used, with ``labels_true`` and ``labels_pred`` ignored. Returns ------- mi: float Mutual information, a non-negative value See also -------- adjusted_mutual_info_score: Adjusted against chance Mutual Information normalized_mutual_info_score: Normalized Mutual Information """ if contingency is None: labels_true, labels_pred = check_clusterings(labels_true, labels_pred) contingency = contingency_matrix(labels_true, labels_pred) contingency = np.array(contingency, dtype='float') contingency_sum = np.sum(contingency) pi = np.sum(contingency, axis=1) pj = np.sum(contingency, axis=0) outer = np.outer(pi, pj) nnz = contingency != 0.0 # normalized contingency contingency_nm = contingency[nnz] log_contingency_nm = np.log(contingency_nm) contingency_nm /= contingency_sum # log(a / b) should be calculated as log(a) - log(b) for # possible loss of precision log_outer = -np.log(outer[nnz]) + log(pi.sum()) + log(pj.sum()) mi = (contingency_nm * (log_contingency_nm - log(contingency_sum)) + contingency_nm * log_outer) return mi.sum() def adjusted_mutual_info_score(labels_true, labels_pred): """Adjusted Mutual Information between two clusterings Adjusted Mutual Information (AMI) is an adjustment of the Mutual Information (MI) score to account for chance. It accounts for the fact that the MI is generally higher for two clusterings with a larger number of clusters, regardless of whether there is actually more information shared. For two clusterings :math:`U` and :math:`V`, the AMI is given as:: AMI(U, V) = [MI(U, V) - E(MI(U, V))] / [max(H(U), H(V)) - E(MI(U, V))] This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is furthermore symmetric: switching ``label_true`` with ``label_pred`` will return the same score value. This can be useful to measure the agreement of two independent label assignments strategies on the same dataset when the real ground truth is not known. Be mindful that this function is an order of magnitude slower than other metrics, such as the Adjusted Rand Index. Read more in the :ref:`User Guide <mutual_info_score>`. Parameters ---------- labels_true : int array, shape = [n_samples] A clustering of the data into disjoint subsets. labels_pred : array, shape = [n_samples] A clustering of the data into disjoint subsets. Returns ------- ami: float(upperlimited by 1.0) The AMI returns a value of 1 when the two partitions are identical (ie perfectly matched). Random partitions (independent labellings) have an expected AMI around 0 on average hence can be negative. See also -------- adjusted_rand_score: Adjusted Rand Index mutual_information_score: Mutual Information (not adjusted for chance) Examples -------- Perfect labelings are both homogeneous and complete, hence have score 1.0:: >>> from sklearn.metrics.cluster import adjusted_mutual_info_score >>> adjusted_mutual_info_score([0, 0, 1, 1], [0, 0, 1, 1]) 1.0 >>> adjusted_mutual_info_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 If classes members are completely split across different clusters, the assignment is totally in-complete, hence the AMI is null:: >>> adjusted_mutual_info_score([0, 0, 0, 0], [0, 1, 2, 3]) 0.0 References ---------- .. [1] `Vinh, Epps, and Bailey, (2010). Information Theoretic Measures for Clusterings Comparison: Variants, Properties, Normalization and Correction for Chance, JMLR <http://jmlr.csail.mit.edu/papers/volume11/vinh10a/vinh10a.pdf>`_ .. [2] `Wikipedia entry for the Adjusted Mutual Information <http://en.wikipedia.org/wiki/Adjusted_Mutual_Information>`_ """ labels_true, labels_pred = check_clusterings(labels_true, labels_pred) n_samples = labels_true.shape[0] classes = np.unique(labels_true) clusters = np.unique(labels_pred) # Special limit cases: no clustering since the data is not split. # This is a perfect match hence return 1.0. if (classes.shape[0] == clusters.shape[0] == 1 or classes.shape[0] == clusters.shape[0] == 0): return 1.0 contingency = contingency_matrix(labels_true, labels_pred) contingency = np.array(contingency, dtype='float') # Calculate the MI for the two clusterings mi = mutual_info_score(labels_true, labels_pred, contingency=contingency) # Calculate the expected value for the mutual information emi = expected_mutual_information(contingency, n_samples) # Calculate entropy for each labeling h_true, h_pred = entropy(labels_true), entropy(labels_pred) ami = (mi - emi) / (max(h_true, h_pred) - emi) return ami def normalized_mutual_info_score(labels_true, labels_pred): """Normalized Mutual Information between two clusterings Normalized Mutual Information (NMI) is an normalization of the Mutual Information (MI) score to scale the results between 0 (no mutual information) and 1 (perfect correlation). In this function, mutual information is normalized by ``sqrt(H(labels_true) * H(labels_pred))`` This measure is not adjusted for chance. Therefore :func:`adjusted_mustual_info_score` might be preferred. This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is furthermore symmetric: switching ``label_true`` with ``label_pred`` will return the same score value. This can be useful to measure the agreement of two independent label assignments strategies on the same dataset when the real ground truth is not known. Read more in the :ref:`User Guide <mutual_info_score>`. Parameters ---------- labels_true : int array, shape = [n_samples] A clustering of the data into disjoint subsets. labels_pred : array, shape = [n_samples] A clustering of the data into disjoint subsets. Returns ------- nmi: float score between 0.0 and 1.0. 1.0 stands for perfectly complete labeling See also -------- adjusted_rand_score: Adjusted Rand Index adjusted_mutual_info_score: Adjusted Mutual Information (adjusted against chance) Examples -------- Perfect labelings are both homogeneous and complete, hence have score 1.0:: >>> from sklearn.metrics.cluster import normalized_mutual_info_score >>> normalized_mutual_info_score([0, 0, 1, 1], [0, 0, 1, 1]) 1.0 >>> normalized_mutual_info_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 If classes members are completely split across different clusters, the assignment is totally in-complete, hence the NMI is null:: >>> normalized_mutual_info_score([0, 0, 0, 0], [0, 1, 2, 3]) 0.0 """ labels_true, labels_pred = check_clusterings(labels_true, labels_pred) classes = np.unique(labels_true) clusters = np.unique(labels_pred) # Special limit cases: no clustering since the data is not split. # This is a perfect match hence return 1.0. if (classes.shape[0] == clusters.shape[0] == 1 or classes.shape[0] == clusters.shape[0] == 0): return 1.0 contingency = contingency_matrix(labels_true, labels_pred) contingency = np.array(contingency, dtype='float') # Calculate the MI for the two clusterings mi = mutual_info_score(labels_true, labels_pred, contingency=contingency) # Calculate the expected value for the mutual information # Calculate entropy for each labeling h_true, h_pred = entropy(labels_true), entropy(labels_pred) nmi = mi / max(np.sqrt(h_true * h_pred), 1e-10) return nmi def entropy(labels): """Calculates the entropy for a labeling.""" if len(labels) == 0: return 1.0 label_idx = np.unique(labels, return_inverse=True)[1] pi = bincount(label_idx).astype(np.float) pi = pi[pi > 0] pi_sum = np.sum(pi) # log(a / b) should be calculated as log(a) - log(b) for # possible loss of precision return -np.sum((pi / pi_sum) * (np.log(pi) - log(pi_sum)))
bsd-3-clause
lilleswing/deepchem
contrib/torch/torch_multitask_classification.py
8
4402
# -*- coding: utf-8 -*- """ Created on Mon Mar 13 22:31:24 2017 @author: Zhenqin Wu """ import torch import numpy as np from deepchem.metrics import from_one_hot from torch_model import TorchMultitaskModel class TorchMultitaskClassification(TorchMultitaskModel): def __init__(self, n_tasks, n_features, n_classes=2, **kwargs): """Constructs the computational graph. This function constructs the computational graph for the model. It relies subclassed methods (build/cost) to construct specific graphs. Parameters ---------- n_tasks: int Number of tasks n_features: int Number of features. n_classes: int Number of classes if this is for classification. """ # Save hyperparameters self.n_tasks = n_tasks self.n_features = n_features self.n_classes = n_classes super(TorchMultitaskClassification, self).__init__(**kwargs) def build(self): """Constructs the graph architecture as specified in its config. This method creates the following Placeholders: mol_features: Molecule descriptor (e.g. fingerprint) tensor with shape batch_size x n_features. """ layer_sizes = self.layer_sizes weight_init_stddevs = self.weight_init_stddevs bias_init_consts = self.bias_init_consts dropouts = self.dropouts lengths_set = { len(layer_sizes), len(weight_init_stddevs), len(bias_init_consts), len(dropouts), } assert len(lengths_set) == 1, 'All layer params must have same length.' n_layers = lengths_set.pop() assert n_layers > 0, 'Must have some layers defined.' prev_layer_size = self.n_features self.W_list = [] self.b_list = [] for i in range(n_layers): W_init = np.random.normal(0, weight_init_stddevs[i], (prev_layer_size, layer_sizes[i])) W_init = torch.cuda.FloatTensor(W_init) self.W_list.append(torch.autograd.Variable(W_init, requires_grad=True)) b_init = np.full((layer_sizes[i],), bias_init_consts[i]) b_init = torch.cuda.FloatTensor(b_init) self.b_list.append(torch.autograd.Variable(b_init, requires_grad=True)) prev_layer_size = layer_sizes[i] self.task_W_list = [] self.task_b_list = [] for i in range(self.n_tasks): W_init = np.random.normal(0, weight_init_stddevs[-1], (prev_layer_size, self.n_classes)) W_init = torch.cuda.FloatTensor(W_init) self.task_W_list.append( torch.autograd.Variable(W_init, requires_grad=True)) b_init = np.full((self.n_classes,), bias_init_consts[-1]) b_init = torch.cuda.FloatTensor(b_init) self.task_b_list.append( torch.autograd.Variable(b_init, requires_grad=True)) self.trainables = self.W_list + self.b_list + self.task_W_list + self.task_b_list self.regularizaed_variables = self.W_list + self.task_W_list def forward(self, X, training=False): for i, W in enumerate(self.W_list): X = X.mm(W) X += self.b_list[i].unsqueeze(0).expand_as(X) X = torch.nn.ReLU()(X) if training: X = torch.nn.Dropout(p=self.dropouts[i])(X) outputs = [] for i, W in enumerate(self.task_W_list): output = X.mm(W) output += self.task_b_list[i].unsqueeze(0).expand_as(output) if not training: output = torch.nn.functional.softmax(output) outputs.append(output) return outputs def cost(self, logit, label, weight): loss = [] for i in range(logit.size()[0]): loss.append( torch.nn.functional.cross_entropy(logit[i, :], label[i].long()).mul( weight[i])) loss = torch.cat(loss).mean() return loss def predict_on_batch(self, X_batch): X_batch = torch.autograd.Variable(torch.cuda.FloatTensor(X_batch)) outputs = self.forward(X_batch, training=False) y_pred_batch = torch.stack(outputs, 1).data.cpu().numpy()[:] y_pred_batch = from_one_hot(y_pred_batch, 2) return y_pred_batch def predict_proba_on_batch(self, X_batch): X_batch = torch.autograd.Variable(torch.cuda.FloatTensor(X_batch)) outputs = self.forward(X_batch, training=False) y_pred_batch = torch.stack(outputs, 1).data.cpu().numpy()[:] return y_pred_batch
mit
luo66/scikit-learn
sklearn/metrics/cluster/supervised.py
206
27395
"""Utilities to evaluate the clustering performance of models Functions named as *_score return a scalar value to maximize: the higher the better. """ # Authors: Olivier Grisel <olivier.grisel@ensta.org> # Wei LI <kuantkid@gmail.com> # Diego Molla <dmolla-aliod@gmail.com> # License: BSD 3 clause from math import log from scipy.misc import comb from scipy.sparse import coo_matrix import numpy as np from .expected_mutual_info_fast import expected_mutual_information from ...utils.fixes import bincount def comb2(n): # the exact version is faster for k == 2: use it by default globally in # this module instead of the float approximate variant return comb(n, 2, exact=1) def check_clusterings(labels_true, labels_pred): """Check that the two clusterings matching 1D integer arrays""" labels_true = np.asarray(labels_true) labels_pred = np.asarray(labels_pred) # input checks if labels_true.ndim != 1: raise ValueError( "labels_true must be 1D: shape is %r" % (labels_true.shape,)) if labels_pred.ndim != 1: raise ValueError( "labels_pred must be 1D: shape is %r" % (labels_pred.shape,)) if labels_true.shape != labels_pred.shape: raise ValueError( "labels_true and labels_pred must have same size, got %d and %d" % (labels_true.shape[0], labels_pred.shape[0])) return labels_true, labels_pred def contingency_matrix(labels_true, labels_pred, eps=None): """Build a contengency matrix describing the relationship between labels. Parameters ---------- labels_true : int array, shape = [n_samples] Ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] Cluster labels to evaluate eps: None or float If a float, that value is added to all values in the contingency matrix. This helps to stop NaN propagation. If ``None``, nothing is adjusted. Returns ------- contingency: array, shape=[n_classes_true, n_classes_pred] Matrix :math:`C` such that :math:`C_{i, j}` is the number of samples in true class :math:`i` and in predicted class :math:`j`. If ``eps is None``, the dtype of this array will be integer. If ``eps`` is given, the dtype will be float. """ classes, class_idx = np.unique(labels_true, return_inverse=True) clusters, cluster_idx = np.unique(labels_pred, return_inverse=True) n_classes = classes.shape[0] n_clusters = clusters.shape[0] # Using coo_matrix to accelerate simple histogram calculation, # i.e. bins are consecutive integers # Currently, coo_matrix is faster than histogram2d for simple cases contingency = coo_matrix((np.ones(class_idx.shape[0]), (class_idx, cluster_idx)), shape=(n_classes, n_clusters), dtype=np.int).toarray() if eps is not None: # don't use += as contingency is integer contingency = contingency + eps return contingency # clustering measures def adjusted_rand_score(labels_true, labels_pred): """Rand index adjusted for chance The Rand Index computes a similarity measure between two clusterings by considering all pairs of samples and counting pairs that are assigned in the same or different clusters in the predicted and true clusterings. The raw RI score is then "adjusted for chance" into the ARI score using the following scheme:: ARI = (RI - Expected_RI) / (max(RI) - Expected_RI) The adjusted Rand index is thus ensured to have a value close to 0.0 for random labeling independently of the number of clusters and samples and exactly 1.0 when the clusterings are identical (up to a permutation). ARI is a symmetric measure:: adjusted_rand_score(a, b) == adjusted_rand_score(b, a) Read more in the :ref:`User Guide <adjusted_rand_score>`. Parameters ---------- labels_true : int array, shape = [n_samples] Ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] Cluster labels to evaluate Returns ------- ari : float Similarity score between -1.0 and 1.0. Random labelings have an ARI close to 0.0. 1.0 stands for perfect match. Examples -------- Perfectly maching labelings have a score of 1 even >>> from sklearn.metrics.cluster import adjusted_rand_score >>> adjusted_rand_score([0, 0, 1, 1], [0, 0, 1, 1]) 1.0 >>> adjusted_rand_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 Labelings that assign all classes members to the same clusters are complete be not always pure, hence penalized:: >>> adjusted_rand_score([0, 0, 1, 2], [0, 0, 1, 1]) # doctest: +ELLIPSIS 0.57... ARI is symmetric, so labelings that have pure clusters with members coming from the same classes but unnecessary splits are penalized:: >>> adjusted_rand_score([0, 0, 1, 1], [0, 0, 1, 2]) # doctest: +ELLIPSIS 0.57... If classes members are completely split across different clusters, the assignment is totally incomplete, hence the ARI is very low:: >>> adjusted_rand_score([0, 0, 0, 0], [0, 1, 2, 3]) 0.0 References ---------- .. [Hubert1985] `L. Hubert and P. Arabie, Comparing Partitions, Journal of Classification 1985` http://www.springerlink.com/content/x64124718341j1j0/ .. [wk] http://en.wikipedia.org/wiki/Rand_index#Adjusted_Rand_index See also -------- adjusted_mutual_info_score: Adjusted Mutual Information """ labels_true, labels_pred = check_clusterings(labels_true, labels_pred) n_samples = labels_true.shape[0] classes = np.unique(labels_true) clusters = np.unique(labels_pred) # Special limit cases: no clustering since the data is not split; # or trivial clustering where each document is assigned a unique cluster. # These are perfect matches hence return 1.0. if (classes.shape[0] == clusters.shape[0] == 1 or classes.shape[0] == clusters.shape[0] == 0 or classes.shape[0] == clusters.shape[0] == len(labels_true)): return 1.0 contingency = contingency_matrix(labels_true, labels_pred) # Compute the ARI using the contingency data sum_comb_c = sum(comb2(n_c) for n_c in contingency.sum(axis=1)) sum_comb_k = sum(comb2(n_k) for n_k in contingency.sum(axis=0)) sum_comb = sum(comb2(n_ij) for n_ij in contingency.flatten()) prod_comb = (sum_comb_c * sum_comb_k) / float(comb(n_samples, 2)) mean_comb = (sum_comb_k + sum_comb_c) / 2. return ((sum_comb - prod_comb) / (mean_comb - prod_comb)) def homogeneity_completeness_v_measure(labels_true, labels_pred): """Compute the homogeneity and completeness and V-Measure scores at once Those metrics are based on normalized conditional entropy measures of the clustering labeling to evaluate given the knowledge of a Ground Truth class labels of the same samples. A clustering result satisfies homogeneity if all of its clusters contain only data points which are members of a single class. A clustering result satisfies completeness if all the data points that are members of a given class are elements of the same cluster. Both scores have positive values between 0.0 and 1.0, larger values being desirable. Those 3 metrics are independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score values in any way. V-Measure is furthermore symmetric: swapping ``labels_true`` and ``label_pred`` will give the same score. This does not hold for homogeneity and completeness. Read more in the :ref:`User Guide <homogeneity_completeness>`. Parameters ---------- labels_true : int array, shape = [n_samples] ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] cluster labels to evaluate Returns ------- homogeneity: float score between 0.0 and 1.0. 1.0 stands for perfectly homogeneous labeling completeness: float score between 0.0 and 1.0. 1.0 stands for perfectly complete labeling v_measure: float harmonic mean of the first two See also -------- homogeneity_score completeness_score v_measure_score """ labels_true, labels_pred = check_clusterings(labels_true, labels_pred) if len(labels_true) == 0: return 1.0, 1.0, 1.0 entropy_C = entropy(labels_true) entropy_K = entropy(labels_pred) MI = mutual_info_score(labels_true, labels_pred) homogeneity = MI / (entropy_C) if entropy_C else 1.0 completeness = MI / (entropy_K) if entropy_K else 1.0 if homogeneity + completeness == 0.0: v_measure_score = 0.0 else: v_measure_score = (2.0 * homogeneity * completeness / (homogeneity + completeness)) return homogeneity, completeness, v_measure_score def homogeneity_score(labels_true, labels_pred): """Homogeneity metric of a cluster labeling given a ground truth A clustering result satisfies homogeneity if all of its clusters contain only data points which are members of a single class. This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is not symmetric: switching ``label_true`` with ``label_pred`` will return the :func:`completeness_score` which will be different in general. Read more in the :ref:`User Guide <homogeneity_completeness>`. Parameters ---------- labels_true : int array, shape = [n_samples] ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] cluster labels to evaluate Returns ------- homogeneity: float score between 0.0 and 1.0. 1.0 stands for perfectly homogeneous labeling References ---------- .. [1] `Andrew Rosenberg and Julia Hirschberg, 2007. V-Measure: A conditional entropy-based external cluster evaluation measure <http://aclweb.org/anthology/D/D07/D07-1043.pdf>`_ See also -------- completeness_score v_measure_score Examples -------- Perfect labelings are homogeneous:: >>> from sklearn.metrics.cluster import homogeneity_score >>> homogeneity_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 Non-perfect labelings that further split classes into more clusters can be perfectly homogeneous:: >>> print("%.6f" % homogeneity_score([0, 0, 1, 1], [0, 0, 1, 2])) ... # doctest: +ELLIPSIS 1.0... >>> print("%.6f" % homogeneity_score([0, 0, 1, 1], [0, 1, 2, 3])) ... # doctest: +ELLIPSIS 1.0... Clusters that include samples from different classes do not make for an homogeneous labeling:: >>> print("%.6f" % homogeneity_score([0, 0, 1, 1], [0, 1, 0, 1])) ... # doctest: +ELLIPSIS 0.0... >>> print("%.6f" % homogeneity_score([0, 0, 1, 1], [0, 0, 0, 0])) ... # doctest: +ELLIPSIS 0.0... """ return homogeneity_completeness_v_measure(labels_true, labels_pred)[0] def completeness_score(labels_true, labels_pred): """Completeness metric of a cluster labeling given a ground truth A clustering result satisfies completeness if all the data points that are members of a given class are elements of the same cluster. This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is not symmetric: switching ``label_true`` with ``label_pred`` will return the :func:`homogeneity_score` which will be different in general. Read more in the :ref:`User Guide <homogeneity_completeness>`. Parameters ---------- labels_true : int array, shape = [n_samples] ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] cluster labels to evaluate Returns ------- completeness: float score between 0.0 and 1.0. 1.0 stands for perfectly complete labeling References ---------- .. [1] `Andrew Rosenberg and Julia Hirschberg, 2007. V-Measure: A conditional entropy-based external cluster evaluation measure <http://aclweb.org/anthology/D/D07/D07-1043.pdf>`_ See also -------- homogeneity_score v_measure_score Examples -------- Perfect labelings are complete:: >>> from sklearn.metrics.cluster import completeness_score >>> completeness_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 Non-perfect labelings that assign all classes members to the same clusters are still complete:: >>> print(completeness_score([0, 0, 1, 1], [0, 0, 0, 0])) 1.0 >>> print(completeness_score([0, 1, 2, 3], [0, 0, 1, 1])) 1.0 If classes members are split across different clusters, the assignment cannot be complete:: >>> print(completeness_score([0, 0, 1, 1], [0, 1, 0, 1])) 0.0 >>> print(completeness_score([0, 0, 0, 0], [0, 1, 2, 3])) 0.0 """ return homogeneity_completeness_v_measure(labels_true, labels_pred)[1] def v_measure_score(labels_true, labels_pred): """V-measure cluster labeling given a ground truth. This score is identical to :func:`normalized_mutual_info_score`. The V-measure is the harmonic mean between homogeneity and completeness:: v = 2 * (homogeneity * completeness) / (homogeneity + completeness) This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is furthermore symmetric: switching ``label_true`` with ``label_pred`` will return the same score value. This can be useful to measure the agreement of two independent label assignments strategies on the same dataset when the real ground truth is not known. Read more in the :ref:`User Guide <homogeneity_completeness>`. Parameters ---------- labels_true : int array, shape = [n_samples] ground truth class labels to be used as a reference labels_pred : array, shape = [n_samples] cluster labels to evaluate Returns ------- v_measure: float score between 0.0 and 1.0. 1.0 stands for perfectly complete labeling References ---------- .. [1] `Andrew Rosenberg and Julia Hirschberg, 2007. V-Measure: A conditional entropy-based external cluster evaluation measure <http://aclweb.org/anthology/D/D07/D07-1043.pdf>`_ See also -------- homogeneity_score completeness_score Examples -------- Perfect labelings are both homogeneous and complete, hence have score 1.0:: >>> from sklearn.metrics.cluster import v_measure_score >>> v_measure_score([0, 0, 1, 1], [0, 0, 1, 1]) 1.0 >>> v_measure_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 Labelings that assign all classes members to the same clusters are complete be not homogeneous, hence penalized:: >>> print("%.6f" % v_measure_score([0, 0, 1, 2], [0, 0, 1, 1])) ... # doctest: +ELLIPSIS 0.8... >>> print("%.6f" % v_measure_score([0, 1, 2, 3], [0, 0, 1, 1])) ... # doctest: +ELLIPSIS 0.66... Labelings that have pure clusters with members coming from the same classes are homogeneous but un-necessary splits harms completeness and thus penalize V-measure as well:: >>> print("%.6f" % v_measure_score([0, 0, 1, 1], [0, 0, 1, 2])) ... # doctest: +ELLIPSIS 0.8... >>> print("%.6f" % v_measure_score([0, 0, 1, 1], [0, 1, 2, 3])) ... # doctest: +ELLIPSIS 0.66... If classes members are completely split across different clusters, the assignment is totally incomplete, hence the V-Measure is null:: >>> print("%.6f" % v_measure_score([0, 0, 0, 0], [0, 1, 2, 3])) ... # doctest: +ELLIPSIS 0.0... Clusters that include samples from totally different classes totally destroy the homogeneity of the labeling, hence:: >>> print("%.6f" % v_measure_score([0, 0, 1, 1], [0, 0, 0, 0])) ... # doctest: +ELLIPSIS 0.0... """ return homogeneity_completeness_v_measure(labels_true, labels_pred)[2] def mutual_info_score(labels_true, labels_pred, contingency=None): """Mutual Information between two clusterings The Mutual Information is a measure of the similarity between two labels of the same data. Where :math:`P(i)` is the probability of a random sample occurring in cluster :math:`U_i` and :math:`P'(j)` is the probability of a random sample occurring in cluster :math:`V_j`, the Mutual Information between clusterings :math:`U` and :math:`V` is given as: .. math:: MI(U,V)=\sum_{i=1}^R \sum_{j=1}^C P(i,j)\log\\frac{P(i,j)}{P(i)P'(j)} This is equal to the Kullback-Leibler divergence of the joint distribution with the product distribution of the marginals. This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is furthermore symmetric: switching ``label_true`` with ``label_pred`` will return the same score value. This can be useful to measure the agreement of two independent label assignments strategies on the same dataset when the real ground truth is not known. Read more in the :ref:`User Guide <mutual_info_score>`. Parameters ---------- labels_true : int array, shape = [n_samples] A clustering of the data into disjoint subsets. labels_pred : array, shape = [n_samples] A clustering of the data into disjoint subsets. contingency: None or array, shape = [n_classes_true, n_classes_pred] A contingency matrix given by the :func:`contingency_matrix` function. If value is ``None``, it will be computed, otherwise the given value is used, with ``labels_true`` and ``labels_pred`` ignored. Returns ------- mi: float Mutual information, a non-negative value See also -------- adjusted_mutual_info_score: Adjusted against chance Mutual Information normalized_mutual_info_score: Normalized Mutual Information """ if contingency is None: labels_true, labels_pred = check_clusterings(labels_true, labels_pred) contingency = contingency_matrix(labels_true, labels_pred) contingency = np.array(contingency, dtype='float') contingency_sum = np.sum(contingency) pi = np.sum(contingency, axis=1) pj = np.sum(contingency, axis=0) outer = np.outer(pi, pj) nnz = contingency != 0.0 # normalized contingency contingency_nm = contingency[nnz] log_contingency_nm = np.log(contingency_nm) contingency_nm /= contingency_sum # log(a / b) should be calculated as log(a) - log(b) for # possible loss of precision log_outer = -np.log(outer[nnz]) + log(pi.sum()) + log(pj.sum()) mi = (contingency_nm * (log_contingency_nm - log(contingency_sum)) + contingency_nm * log_outer) return mi.sum() def adjusted_mutual_info_score(labels_true, labels_pred): """Adjusted Mutual Information between two clusterings Adjusted Mutual Information (AMI) is an adjustment of the Mutual Information (MI) score to account for chance. It accounts for the fact that the MI is generally higher for two clusterings with a larger number of clusters, regardless of whether there is actually more information shared. For two clusterings :math:`U` and :math:`V`, the AMI is given as:: AMI(U, V) = [MI(U, V) - E(MI(U, V))] / [max(H(U), H(V)) - E(MI(U, V))] This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is furthermore symmetric: switching ``label_true`` with ``label_pred`` will return the same score value. This can be useful to measure the agreement of two independent label assignments strategies on the same dataset when the real ground truth is not known. Be mindful that this function is an order of magnitude slower than other metrics, such as the Adjusted Rand Index. Read more in the :ref:`User Guide <mutual_info_score>`. Parameters ---------- labels_true : int array, shape = [n_samples] A clustering of the data into disjoint subsets. labels_pred : array, shape = [n_samples] A clustering of the data into disjoint subsets. Returns ------- ami: float(upperlimited by 1.0) The AMI returns a value of 1 when the two partitions are identical (ie perfectly matched). Random partitions (independent labellings) have an expected AMI around 0 on average hence can be negative. See also -------- adjusted_rand_score: Adjusted Rand Index mutual_information_score: Mutual Information (not adjusted for chance) Examples -------- Perfect labelings are both homogeneous and complete, hence have score 1.0:: >>> from sklearn.metrics.cluster import adjusted_mutual_info_score >>> adjusted_mutual_info_score([0, 0, 1, 1], [0, 0, 1, 1]) 1.0 >>> adjusted_mutual_info_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 If classes members are completely split across different clusters, the assignment is totally in-complete, hence the AMI is null:: >>> adjusted_mutual_info_score([0, 0, 0, 0], [0, 1, 2, 3]) 0.0 References ---------- .. [1] `Vinh, Epps, and Bailey, (2010). Information Theoretic Measures for Clusterings Comparison: Variants, Properties, Normalization and Correction for Chance, JMLR <http://jmlr.csail.mit.edu/papers/volume11/vinh10a/vinh10a.pdf>`_ .. [2] `Wikipedia entry for the Adjusted Mutual Information <http://en.wikipedia.org/wiki/Adjusted_Mutual_Information>`_ """ labels_true, labels_pred = check_clusterings(labels_true, labels_pred) n_samples = labels_true.shape[0] classes = np.unique(labels_true) clusters = np.unique(labels_pred) # Special limit cases: no clustering since the data is not split. # This is a perfect match hence return 1.0. if (classes.shape[0] == clusters.shape[0] == 1 or classes.shape[0] == clusters.shape[0] == 0): return 1.0 contingency = contingency_matrix(labels_true, labels_pred) contingency = np.array(contingency, dtype='float') # Calculate the MI for the two clusterings mi = mutual_info_score(labels_true, labels_pred, contingency=contingency) # Calculate the expected value for the mutual information emi = expected_mutual_information(contingency, n_samples) # Calculate entropy for each labeling h_true, h_pred = entropy(labels_true), entropy(labels_pred) ami = (mi - emi) / (max(h_true, h_pred) - emi) return ami def normalized_mutual_info_score(labels_true, labels_pred): """Normalized Mutual Information between two clusterings Normalized Mutual Information (NMI) is an normalization of the Mutual Information (MI) score to scale the results between 0 (no mutual information) and 1 (perfect correlation). In this function, mutual information is normalized by ``sqrt(H(labels_true) * H(labels_pred))`` This measure is not adjusted for chance. Therefore :func:`adjusted_mustual_info_score` might be preferred. This metric is independent of the absolute values of the labels: a permutation of the class or cluster label values won't change the score value in any way. This metric is furthermore symmetric: switching ``label_true`` with ``label_pred`` will return the same score value. This can be useful to measure the agreement of two independent label assignments strategies on the same dataset when the real ground truth is not known. Read more in the :ref:`User Guide <mutual_info_score>`. Parameters ---------- labels_true : int array, shape = [n_samples] A clustering of the data into disjoint subsets. labels_pred : array, shape = [n_samples] A clustering of the data into disjoint subsets. Returns ------- nmi: float score between 0.0 and 1.0. 1.0 stands for perfectly complete labeling See also -------- adjusted_rand_score: Adjusted Rand Index adjusted_mutual_info_score: Adjusted Mutual Information (adjusted against chance) Examples -------- Perfect labelings are both homogeneous and complete, hence have score 1.0:: >>> from sklearn.metrics.cluster import normalized_mutual_info_score >>> normalized_mutual_info_score([0, 0, 1, 1], [0, 0, 1, 1]) 1.0 >>> normalized_mutual_info_score([0, 0, 1, 1], [1, 1, 0, 0]) 1.0 If classes members are completely split across different clusters, the assignment is totally in-complete, hence the NMI is null:: >>> normalized_mutual_info_score([0, 0, 0, 0], [0, 1, 2, 3]) 0.0 """ labels_true, labels_pred = check_clusterings(labels_true, labels_pred) classes = np.unique(labels_true) clusters = np.unique(labels_pred) # Special limit cases: no clustering since the data is not split. # This is a perfect match hence return 1.0. if (classes.shape[0] == clusters.shape[0] == 1 or classes.shape[0] == clusters.shape[0] == 0): return 1.0 contingency = contingency_matrix(labels_true, labels_pred) contingency = np.array(contingency, dtype='float') # Calculate the MI for the two clusterings mi = mutual_info_score(labels_true, labels_pred, contingency=contingency) # Calculate the expected value for the mutual information # Calculate entropy for each labeling h_true, h_pred = entropy(labels_true), entropy(labels_pred) nmi = mi / max(np.sqrt(h_true * h_pred), 1e-10) return nmi def entropy(labels): """Calculates the entropy for a labeling.""" if len(labels) == 0: return 1.0 label_idx = np.unique(labels, return_inverse=True)[1] pi = bincount(label_idx).astype(np.float) pi = pi[pi > 0] pi_sum = np.sum(pi) # log(a / b) should be calculated as log(a) - log(b) for # possible loss of precision return -np.sum((pi / pi_sum) * (np.log(pi) - log(pi_sum)))
bsd-3-clause
sgenoud/scikit-learn
sklearn/covariance/tests/test_robust_covariance.py
1
2616
# Author: Alexandre Gramfort <alexandre.gramfort@inria.fr> # Gael Varoquaux <gael.varoquaux@normalesup.org> # Virgile Fritsch <virgile.fritsch@inria.fr> # # License: BSD Style. from numpy.testing import assert_almost_equal, assert_array_almost_equal import numpy as np from sklearn import datasets from sklearn.covariance import empirical_covariance, MinCovDet, \ EllipticEnvelope X = datasets.load_iris().data X_1d = X[:, 0] n_samples, n_features = X.shape def test_mcd(): """Tests the FastMCD algorithm implementation """ ### Small data set # test without outliers (random independent normal data) launch_mcd_on_dataset(100, 5, 0, 0.01, 0.1, 80) # test with a contaminated data set (medium contamination) launch_mcd_on_dataset(100, 5, 20, 0.01, 0.01, 70) # test with a contaminated data set (strong contamination) launch_mcd_on_dataset(100, 5, 40, 0.1, 0.1, 50) ### Medium data set launch_mcd_on_dataset(1000, 5, 450, 1e-3, 1e-3, 540) ### Large data set launch_mcd_on_dataset(1700, 5, 800, 1e-3, 1e-3, 870) ### 1D data set launch_mcd_on_dataset(500, 1, 100, 0.001, 0.001, 350) def launch_mcd_on_dataset( n_samples, n_features, n_outliers, tol_loc, tol_cov, tol_support): rand_gen = np.random.RandomState(0) data = rand_gen.randn(n_samples, n_features) # add some outliers outliers_index = rand_gen.permutation(n_samples)[:n_outliers] outliers_offset = 10. * \ (rand_gen.randint(2, size=(n_outliers, n_features)) - 0.5) data[outliers_index] += outliers_offset inliers_mask = np.ones(n_samples).astype(bool) inliers_mask[outliers_index] = False pure_data = data[inliers_mask] # compute MCD by fitting an object mcd_fit = MinCovDet(random_state=rand_gen).fit(data) T = mcd_fit.location_ S = mcd_fit.covariance_ H = mcd_fit.support_ # compare with the estimates learnt from the inliers error_location = np.mean((pure_data.mean(0) - T) ** 2) assert(error_location < tol_loc) error_cov = np.mean((empirical_covariance(pure_data) - S) ** 2) assert(error_cov < tol_cov) assert(np.sum(H) >= tol_support) def test_outlier_detection(): rnd = np.random.RandomState(0) X = rnd.randn(100, 10) clf = EllipticEnvelope(contamination=0.1) clf.fit(X) y_pred = clf.predict(X) assert_array_almost_equal( clf.decision_function(X, raw_mahalanobis=True), clf.mahalanobis(X - clf.location_)) assert_almost_equal(clf.score(X, np.ones(100)), (100 - y_pred[y_pred == -1].size) / 100.)
bsd-3-clause
Diyago/Machine-Learning-scripts
time series regression/autocorelation, mov avg etc/doubleExponentialSmoothing.py
1
2945
# Load modules from __future__ import print_function import os import pandas as pd import numpy as np from matplotlib import pyplot as plt # Read dataset into a pandas.DataFrame beer_df = pd.read_csv( "datasets/quarterly-beer-production-in-aus-March 1956-June 1994.csv" ) # Display shape of the dataset print("Shape of the dataframe:", beer_df.shape) beer_df.head() # Rename the 2nd column beer_df.rename( columns={ "Quarterly beer production in Australia: megalitres. March 1956 ? June 1994": "Beer_Prod" }, inplace=True, ) # Remove missing values missing = (pd.isnull(beer_df["Quarter"])) | (pd.isnull(beer_df["Beer_Prod"])) print("Number of rows with at least one missing values:", missing.sum()) beer_df = beer_df.loc[~missing, :] print("Shape after removing missing values:", beer_df.shape) # Function for Sigle exponential smoothing def double_exp_smoothing(x, alpha, beta): yhat = [x[0]] # first value is same as series for t in range(1, len(x)): if t == 1: F, T = x[0], x[1] - x[0] F_n_1, F = F, alpha * x[t] + (1 - alpha) * (F + T) T = beta * (F - F_n_1) + (1 - beta) * T yhat.append(F + T) return yhat beer_df["DEF"] = double_exp_smoothing(beer_df["Beer_Prod"], 0.4, 0.7) ### Plot Single Exponential Smoothing forecasted value fig = plt.figure(figsize=(5.5, 5.5)) ax = fig.add_subplot(2, 1, 1) beer_df["Beer_Prod"].plot(ax=ax) ax.set_title("Beer Production") ax = fig.add_subplot(2, 1, 2) beer_df["DEF"].plot(ax=ax, color="r") ax.set_title("Double Smoothing Forecast") plt.savefig("plots/ch2/B07887_03_14.png", format="png", dpi=300) # Single vs Double Forecast value # Function for Sigle exponential smoothing def single_exp_smoothing(x, alpha): F = [x[0]] # first value is same as series for t in range(1, len(x)): F.append(alpha * x[t] + (1 - alpha) * F[t - 1]) return F beer_df["Single_Exponential_Forecast"] = single_exp_smoothing(beer_df["Beer_Prod"], 0.4) ### Plot Single Exponential Smoothing forecasted value f, axarr = plt.subplots(2, sharex=True) f.set_size_inches(5.5, 5.5) beer_df["Beer_Prod"].iloc[:153].plot(color="b", linestyle="-", ax=axarr[0]) beer_df["DEF"].iloc[:153].plot(color="r", linestyle="--", ax=axarr[0]) axarr[0].set_title("Actual Vs Double Smoothing Forecasting") beer_df["Beer_Prod"].iloc[:153].plot(color="b", linestyle="-", ax=axarr[1]) beer_df["Single_Exponential_Forecast"].iloc[:153].plot( color="r", linestyle="--", ax=axarr[1] ) axarr[1].set_title("Actual Vs Single Smoothing Forecasting") # Plot single and double exponential smoothing fig = plt.figure(figsize=(5.5, 5.5)) ax = fig.add_subplot(2, 1, 1) beer_df["Single_Exponential_Forecast"].plot(ax=ax) ax.set_title("Single Exponential Smoothing") ax = fig.add_subplot(2, 1, 2) beer_df["DEF"].plot(ax=ax, color="r") ax.set_title("Double Smoothing Forecast") plt.savefig("plots/ch2/B07887_03_14.png", format="png", dpi=300)
apache-2.0
LohithBlaze/scikit-learn
sklearn/linear_model/tests/test_bayes.py
296
1770
# Author: Alexandre Gramfort <alexandre.gramfort@inria.fr> # Fabian Pedregosa <fabian.pedregosa@inria.fr> # # License: BSD 3 clause import numpy as np from sklearn.utils.testing import assert_array_equal from sklearn.utils.testing import SkipTest from sklearn.linear_model.bayes import BayesianRidge, ARDRegression from sklearn import datasets from sklearn.utils.testing import assert_array_almost_equal def test_bayesian_on_diabetes(): # Test BayesianRidge on diabetes raise SkipTest("XFailed Test") diabetes = datasets.load_diabetes() X, y = diabetes.data, diabetes.target clf = BayesianRidge(compute_score=True) # Test with more samples than features clf.fit(X, y) # Test that scores are increasing at each iteration assert_array_equal(np.diff(clf.scores_) > 0, True) # Test with more features than samples X = X[:5, :] y = y[:5] clf.fit(X, y) # Test that scores are increasing at each iteration assert_array_equal(np.diff(clf.scores_) > 0, True) def test_toy_bayesian_ridge_object(): # Test BayesianRidge on toy X = np.array([[1], [2], [6], [8], [10]]) Y = np.array([1, 2, 6, 8, 10]) clf = BayesianRidge(compute_score=True) clf.fit(X, Y) # Check that the model could approximately learn the identity function test = [[1], [3], [4]] assert_array_almost_equal(clf.predict(test), [1, 3, 4], 2) def test_toy_ard_object(): # Test BayesianRegression ARD classifier X = np.array([[1], [2], [3]]) Y = np.array([1, 2, 3]) clf = ARDRegression(compute_score=True) clf.fit(X, Y) # Check that the model could approximately learn the identity function test = [[1], [3], [4]] assert_array_almost_equal(clf.predict(test), [1, 3, 4], 2)
bsd-3-clause
LohithBlaze/scikit-learn
sklearn/ensemble/tests/test_weight_boosting.py
35
16763
"""Testing for the boost module (sklearn.ensemble.boost).""" import numpy as np from sklearn.utils.testing import assert_array_equal, assert_array_less from sklearn.utils.testing import assert_array_almost_equal from sklearn.utils.testing import assert_equal, assert_true from sklearn.utils.testing import assert_raises, assert_raises_regexp from sklearn.cross_validation import train_test_split from sklearn.grid_search import GridSearchCV from sklearn.ensemble import AdaBoostClassifier from sklearn.ensemble import AdaBoostRegressor from sklearn.ensemble import weight_boosting from scipy.sparse import csc_matrix from scipy.sparse import csr_matrix from scipy.sparse import coo_matrix from scipy.sparse import dok_matrix from scipy.sparse import lil_matrix from sklearn.svm import SVC, SVR from sklearn.tree import DecisionTreeClassifier, DecisionTreeRegressor from sklearn.utils import shuffle from sklearn import datasets # Common random state rng = np.random.RandomState(0) # Toy sample X = [[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]] y_class = ["foo", "foo", "foo", 1, 1, 1] # test string class labels y_regr = [-1, -1, -1, 1, 1, 1] T = [[-1, -1], [2, 2], [3, 2]] y_t_class = ["foo", 1, 1] y_t_regr = [-1, 1, 1] # Load the iris dataset and randomly permute it iris = datasets.load_iris() perm = rng.permutation(iris.target.size) iris.data, iris.target = shuffle(iris.data, iris.target, random_state=rng) # Load the boston dataset and randomly permute it boston = datasets.load_boston() boston.data, boston.target = shuffle(boston.data, boston.target, random_state=rng) def test_samme_proba(): # Test the `_samme_proba` helper function. # Define some example (bad) `predict_proba` output. probs = np.array([[1, 1e-6, 0], [0.19, 0.6, 0.2], [-999, 0.51, 0.5], [1e-6, 1, 1e-9]]) probs /= np.abs(probs.sum(axis=1))[:, np.newaxis] # _samme_proba calls estimator.predict_proba. # Make a mock object so I can control what gets returned. class MockEstimator(object): def predict_proba(self, X): assert_array_equal(X.shape, probs.shape) return probs mock = MockEstimator() samme_proba = weight_boosting._samme_proba(mock, 3, np.ones_like(probs)) assert_array_equal(samme_proba.shape, probs.shape) assert_true(np.isfinite(samme_proba).all()) # Make sure that the correct elements come out as smallest -- # `_samme_proba` should preserve the ordering in each example. assert_array_equal(np.argmin(samme_proba, axis=1), [2, 0, 0, 2]) assert_array_equal(np.argmax(samme_proba, axis=1), [0, 1, 1, 1]) def test_classification_toy(): # Check classification on a toy dataset. for alg in ['SAMME', 'SAMME.R']: clf = AdaBoostClassifier(algorithm=alg, random_state=0) clf.fit(X, y_class) assert_array_equal(clf.predict(T), y_t_class) assert_array_equal(np.unique(np.asarray(y_t_class)), clf.classes_) assert_equal(clf.predict_proba(T).shape, (len(T), 2)) assert_equal(clf.decision_function(T).shape, (len(T),)) def test_regression_toy(): # Check classification on a toy dataset. clf = AdaBoostRegressor(random_state=0) clf.fit(X, y_regr) assert_array_equal(clf.predict(T), y_t_regr) def test_iris(): # Check consistency on dataset iris. classes = np.unique(iris.target) clf_samme = prob_samme = None for alg in ['SAMME', 'SAMME.R']: clf = AdaBoostClassifier(algorithm=alg) clf.fit(iris.data, iris.target) assert_array_equal(classes, clf.classes_) proba = clf.predict_proba(iris.data) if alg == "SAMME": clf_samme = clf prob_samme = proba assert_equal(proba.shape[1], len(classes)) assert_equal(clf.decision_function(iris.data).shape[1], len(classes)) score = clf.score(iris.data, iris.target) assert score > 0.9, "Failed with algorithm %s and score = %f" % \ (alg, score) # Somewhat hacky regression test: prior to # ae7adc880d624615a34bafdb1d75ef67051b8200, # predict_proba returned SAMME.R values for SAMME. clf_samme.algorithm = "SAMME.R" assert_array_less(0, np.abs(clf_samme.predict_proba(iris.data) - prob_samme)) def test_boston(): # Check consistency on dataset boston house prices. clf = AdaBoostRegressor(random_state=0) clf.fit(boston.data, boston.target) score = clf.score(boston.data, boston.target) assert score > 0.85 def test_staged_predict(): # Check staged predictions. rng = np.random.RandomState(0) iris_weights = rng.randint(10, size=iris.target.shape) boston_weights = rng.randint(10, size=boston.target.shape) # AdaBoost classification for alg in ['SAMME', 'SAMME.R']: clf = AdaBoostClassifier(algorithm=alg, n_estimators=10) clf.fit(iris.data, iris.target, sample_weight=iris_weights) predictions = clf.predict(iris.data) staged_predictions = [p for p in clf.staged_predict(iris.data)] proba = clf.predict_proba(iris.data) staged_probas = [p for p in clf.staged_predict_proba(iris.data)] score = clf.score(iris.data, iris.target, sample_weight=iris_weights) staged_scores = [ s for s in clf.staged_score( iris.data, iris.target, sample_weight=iris_weights)] assert_equal(len(staged_predictions), 10) assert_array_almost_equal(predictions, staged_predictions[-1]) assert_equal(len(staged_probas), 10) assert_array_almost_equal(proba, staged_probas[-1]) assert_equal(len(staged_scores), 10) assert_array_almost_equal(score, staged_scores[-1]) # AdaBoost regression clf = AdaBoostRegressor(n_estimators=10, random_state=0) clf.fit(boston.data, boston.target, sample_weight=boston_weights) predictions = clf.predict(boston.data) staged_predictions = [p for p in clf.staged_predict(boston.data)] score = clf.score(boston.data, boston.target, sample_weight=boston_weights) staged_scores = [ s for s in clf.staged_score( boston.data, boston.target, sample_weight=boston_weights)] assert_equal(len(staged_predictions), 10) assert_array_almost_equal(predictions, staged_predictions[-1]) assert_equal(len(staged_scores), 10) assert_array_almost_equal(score, staged_scores[-1]) def test_gridsearch(): # Check that base trees can be grid-searched. # AdaBoost classification boost = AdaBoostClassifier(base_estimator=DecisionTreeClassifier()) parameters = {'n_estimators': (1, 2), 'base_estimator__max_depth': (1, 2), 'algorithm': ('SAMME', 'SAMME.R')} clf = GridSearchCV(boost, parameters) clf.fit(iris.data, iris.target) # AdaBoost regression boost = AdaBoostRegressor(base_estimator=DecisionTreeRegressor(), random_state=0) parameters = {'n_estimators': (1, 2), 'base_estimator__max_depth': (1, 2)} clf = GridSearchCV(boost, parameters) clf.fit(boston.data, boston.target) def test_pickle(): # Check pickability. import pickle # Adaboost classifier for alg in ['SAMME', 'SAMME.R']: obj = AdaBoostClassifier(algorithm=alg) obj.fit(iris.data, iris.target) score = obj.score(iris.data, iris.target) s = pickle.dumps(obj) obj2 = pickle.loads(s) assert_equal(type(obj2), obj.__class__) score2 = obj2.score(iris.data, iris.target) assert_equal(score, score2) # Adaboost regressor obj = AdaBoostRegressor(random_state=0) obj.fit(boston.data, boston.target) score = obj.score(boston.data, boston.target) s = pickle.dumps(obj) obj2 = pickle.loads(s) assert_equal(type(obj2), obj.__class__) score2 = obj2.score(boston.data, boston.target) assert_equal(score, score2) def test_importances(): # Check variable importances. X, y = datasets.make_classification(n_samples=2000, n_features=10, n_informative=3, n_redundant=0, n_repeated=0, shuffle=False, random_state=1) for alg in ['SAMME', 'SAMME.R']: clf = AdaBoostClassifier(algorithm=alg) clf.fit(X, y) importances = clf.feature_importances_ assert_equal(importances.shape[0], 10) assert_equal((importances[:3, np.newaxis] >= importances[3:]).all(), True) def test_error(): # Test that it gives proper exception on deficient input. assert_raises(ValueError, AdaBoostClassifier(learning_rate=-1).fit, X, y_class) assert_raises(ValueError, AdaBoostClassifier(algorithm="foo").fit, X, y_class) assert_raises(ValueError, AdaBoostClassifier().fit, X, y_class, sample_weight=np.asarray([-1])) def test_base_estimator(): # Test different base estimators. from sklearn.ensemble import RandomForestClassifier from sklearn.svm import SVC # XXX doesn't work with y_class because RF doesn't support classes_ # Shouldn't AdaBoost run a LabelBinarizer? clf = AdaBoostClassifier(RandomForestClassifier()) clf.fit(X, y_regr) clf = AdaBoostClassifier(SVC(), algorithm="SAMME") clf.fit(X, y_class) from sklearn.ensemble import RandomForestRegressor from sklearn.svm import SVR clf = AdaBoostRegressor(RandomForestRegressor(), random_state=0) clf.fit(X, y_regr) clf = AdaBoostRegressor(SVR(), random_state=0) clf.fit(X, y_regr) # Check that an empty discrete ensemble fails in fit, not predict. X_fail = [[1, 1], [1, 1], [1, 1], [1, 1]] y_fail = ["foo", "bar", 1, 2] clf = AdaBoostClassifier(SVC(), algorithm="SAMME") assert_raises_regexp(ValueError, "worse than random", clf.fit, X_fail, y_fail) def test_sample_weight_missing(): from sklearn.linear_model import LinearRegression from sklearn.cluster import KMeans clf = AdaBoostClassifier(LinearRegression(), algorithm="SAMME") assert_raises(ValueError, clf.fit, X, y_regr) clf = AdaBoostRegressor(LinearRegression()) assert_raises(ValueError, clf.fit, X, y_regr) clf = AdaBoostClassifier(KMeans(), algorithm="SAMME") assert_raises(ValueError, clf.fit, X, y_regr) clf = AdaBoostRegressor(KMeans()) assert_raises(ValueError, clf.fit, X, y_regr) def test_sparse_classification(): # Check classification with sparse input. class CustomSVC(SVC): """SVC variant that records the nature of the training set.""" def fit(self, X, y, sample_weight=None): """Modification on fit caries data type for later verification.""" super(CustomSVC, self).fit(X, y, sample_weight=sample_weight) self.data_type_ = type(X) return self X, y = datasets.make_multilabel_classification(n_classes=1, n_samples=15, n_features=5, random_state=42) # Flatten y to a 1d array y = np.ravel(y) X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0) for sparse_format in [csc_matrix, csr_matrix, lil_matrix, coo_matrix, dok_matrix]: X_train_sparse = sparse_format(X_train) X_test_sparse = sparse_format(X_test) # Trained on sparse format sparse_classifier = AdaBoostClassifier( base_estimator=CustomSVC(probability=True), random_state=1, algorithm="SAMME" ).fit(X_train_sparse, y_train) # Trained on dense format dense_classifier = AdaBoostClassifier( base_estimator=CustomSVC(probability=True), random_state=1, algorithm="SAMME" ).fit(X_train, y_train) # predict sparse_results = sparse_classifier.predict(X_test_sparse) dense_results = dense_classifier.predict(X_test) assert_array_equal(sparse_results, dense_results) # decision_function sparse_results = sparse_classifier.decision_function(X_test_sparse) dense_results = dense_classifier.decision_function(X_test) assert_array_equal(sparse_results, dense_results) # predict_log_proba sparse_results = sparse_classifier.predict_log_proba(X_test_sparse) dense_results = dense_classifier.predict_log_proba(X_test) assert_array_equal(sparse_results, dense_results) # predict_proba sparse_results = sparse_classifier.predict_proba(X_test_sparse) dense_results = dense_classifier.predict_proba(X_test) assert_array_equal(sparse_results, dense_results) # score sparse_results = sparse_classifier.score(X_test_sparse, y_test) dense_results = dense_classifier.score(X_test, y_test) assert_array_equal(sparse_results, dense_results) # staged_decision_function sparse_results = sparse_classifier.staged_decision_function( X_test_sparse) dense_results = dense_classifier.staged_decision_function(X_test) for sprase_res, dense_res in zip(sparse_results, dense_results): assert_array_equal(sprase_res, dense_res) # staged_predict sparse_results = sparse_classifier.staged_predict(X_test_sparse) dense_results = dense_classifier.staged_predict(X_test) for sprase_res, dense_res in zip(sparse_results, dense_results): assert_array_equal(sprase_res, dense_res) # staged_predict_proba sparse_results = sparse_classifier.staged_predict_proba(X_test_sparse) dense_results = dense_classifier.staged_predict_proba(X_test) for sprase_res, dense_res in zip(sparse_results, dense_results): assert_array_equal(sprase_res, dense_res) # staged_score sparse_results = sparse_classifier.staged_score(X_test_sparse, y_test) dense_results = dense_classifier.staged_score(X_test, y_test) for sprase_res, dense_res in zip(sparse_results, dense_results): assert_array_equal(sprase_res, dense_res) # Verify sparsity of data is maintained during training types = [i.data_type_ for i in sparse_classifier.estimators_] assert all([(t == csc_matrix or t == csr_matrix) for t in types]) def test_sparse_regression(): # Check regression with sparse input. class CustomSVR(SVR): """SVR variant that records the nature of the training set.""" def fit(self, X, y, sample_weight=None): """Modification on fit caries data type for later verification.""" super(CustomSVR, self).fit(X, y, sample_weight=sample_weight) self.data_type_ = type(X) return self X, y = datasets.make_regression(n_samples=15, n_features=50, n_targets=1, random_state=42) X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0) for sparse_format in [csc_matrix, csr_matrix, lil_matrix, coo_matrix, dok_matrix]: X_train_sparse = sparse_format(X_train) X_test_sparse = sparse_format(X_test) # Trained on sparse format sparse_classifier = AdaBoostRegressor( base_estimator=CustomSVR(), random_state=1 ).fit(X_train_sparse, y_train) # Trained on dense format dense_classifier = dense_results = AdaBoostRegressor( base_estimator=CustomSVR(), random_state=1 ).fit(X_train, y_train) # predict sparse_results = sparse_classifier.predict(X_test_sparse) dense_results = dense_classifier.predict(X_test) assert_array_equal(sparse_results, dense_results) # staged_predict sparse_results = sparse_classifier.staged_predict(X_test_sparse) dense_results = dense_classifier.staged_predict(X_test) for sprase_res, dense_res in zip(sparse_results, dense_results): assert_array_equal(sprase_res, dense_res) types = [i.data_type_ for i in sparse_classifier.estimators_] assert all([(t == csc_matrix or t == csr_matrix) for t in types])
bsd-3-clause
manipopopo/tensorflow
tensorflow/contrib/learn/python/learn/estimators/logistic_regressor_test.py
44
4901
# Copyright 2016 The TensorFlow Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """Tests for LogisticRegressor.""" from __future__ import absolute_import from __future__ import division from __future__ import print_function import numpy as np from tensorflow.contrib import layers from tensorflow.python.training import training_util from tensorflow.contrib.layers.python.layers import optimizers from tensorflow.contrib.learn.python.learn.datasets import base from tensorflow.contrib.learn.python.learn.estimators import logistic_regressor from tensorflow.contrib.learn.python.learn.estimators import metric_key from tensorflow.python.framework import constant_op from tensorflow.python.framework import dtypes from tensorflow.python.ops import array_ops from tensorflow.python.ops import init_ops from tensorflow.python.ops import math_ops from tensorflow.python.ops.losses import losses from tensorflow.python.platform import test def _iris_data_input_fn(): # Converts iris data to a logistic regression problem. iris = base.load_iris() ids = np.where((iris.target == 0) | (iris.target == 1)) features = constant_op.constant(iris.data[ids], dtype=dtypes.float32) labels = constant_op.constant(iris.target[ids], dtype=dtypes.float32) labels = array_ops.reshape(labels, labels.get_shape().concatenate(1)) return features, labels def _logistic_regression_model_fn(features, labels, mode): _ = mode logits = layers.linear( features, 1, weights_initializer=init_ops.zeros_initializer(), # Intentionally uses really awful initial values so that # AUC/precision/recall/etc will change meaningfully even on a toy dataset. biases_initializer=init_ops.constant_initializer(-10.0)) predictions = math_ops.sigmoid(logits) loss = losses.sigmoid_cross_entropy(labels, logits) train_op = optimizers.optimize_loss( loss, training_util.get_global_step(), optimizer='Adagrad', learning_rate=0.1) return predictions, loss, train_op class LogisticRegressorTest(test.TestCase): def test_fit_and_evaluate_metrics(self): """Tests basic fit and evaluate, and checks the evaluation metrics.""" regressor = logistic_regressor.LogisticRegressor( model_fn=_logistic_regression_model_fn) # Get some (intentionally horrible) baseline metrics. regressor.fit(input_fn=_iris_data_input_fn, steps=1) eval_metrics = regressor.evaluate(input_fn=_iris_data_input_fn, steps=1) self.assertNear( 0.0, eval_metrics[metric_key.MetricKey.PREDICTION_MEAN], err=1e-3) self.assertNear( 0.5, eval_metrics[metric_key.MetricKey.LABEL_MEAN], err=1e-6) self.assertNear( 0.5, eval_metrics[metric_key.MetricKey.ACCURACY_BASELINE], err=1e-6) self.assertNear(0.5, eval_metrics[metric_key.MetricKey.AUC], err=1e-6) self.assertNear( 0.5, eval_metrics[metric_key.MetricKey.ACCURACY_MEAN % 0.5], err=1e-6) self.assertNear( 0.0, eval_metrics[metric_key.MetricKey.PRECISION_MEAN % 0.5], err=1e-6) self.assertNear( 0.0, eval_metrics[metric_key.MetricKey.RECALL_MEAN % 0.5], err=1e-6) # Train for more steps and check the metrics again. regressor.fit(input_fn=_iris_data_input_fn, steps=100) eval_metrics = regressor.evaluate(input_fn=_iris_data_input_fn, steps=1) # Mean prediction moves from ~0.0 to ~0.5 as we stop predicting all 0's. self.assertNear( 0.5, eval_metrics[metric_key.MetricKey.PREDICTION_MEAN], err=1e-2) # Label mean and baseline both remain the same at 0.5. self.assertNear( 0.5, eval_metrics[metric_key.MetricKey.LABEL_MEAN], err=1e-6) self.assertNear( 0.5, eval_metrics[metric_key.MetricKey.ACCURACY_BASELINE], err=1e-6) # AUC improves from 0.5 to 1.0. self.assertNear(1.0, eval_metrics[metric_key.MetricKey.AUC], err=1e-6) # Accuracy improves from 0.5 to >0.9. self.assertTrue( eval_metrics[metric_key.MetricKey.ACCURACY_MEAN % 0.5] > 0.9) # Precision improves from 0.0 to 1.0. self.assertNear( 1.0, eval_metrics[metric_key.MetricKey.PRECISION_MEAN % 0.5], err=1e-6) # Recall improves from 0.0 to >0.9. self.assertTrue(eval_metrics[metric_key.MetricKey.RECALL_MEAN % 0.5] > 0.9) if __name__ == '__main__': test.main()
apache-2.0
codeworldprodigy/lab2
lib/jinja2/visitor.py
1402
3316
# -*- coding: utf-8 -*- """ jinja2.visitor ~~~~~~~~~~~~~~ This module implements a visitor for the nodes. :copyright: (c) 2010 by the Jinja Team. :license: BSD. """ from jinja2.nodes import Node class NodeVisitor(object): """Walks the abstract syntax tree and call visitor functions for every node found. The visitor functions may return values which will be forwarded by the `visit` method. Per default the visitor functions for the nodes are ``'visit_'`` + class name of the node. So a `TryFinally` node visit function would be `visit_TryFinally`. This behavior can be changed by overriding the `get_visitor` function. If no visitor function exists for a node (return value `None`) the `generic_visit` visitor is used instead. """ def get_visitor(self, node): """Return the visitor function for this node or `None` if no visitor exists for this node. In that case the generic visit function is used instead. """ method = 'visit_' + node.__class__.__name__ return getattr(self, method, None) def visit(self, node, *args, **kwargs): """Visit a node.""" f = self.get_visitor(node) if f is not None: return f(node, *args, **kwargs) return self.generic_visit(node, *args, **kwargs) def generic_visit(self, node, *args, **kwargs): """Called if no explicit visitor function exists for a node.""" for node in node.iter_child_nodes(): self.visit(node, *args, **kwargs) class NodeTransformer(NodeVisitor): """Walks the abstract syntax tree and allows modifications of nodes. The `NodeTransformer` will walk the AST and use the return value of the visitor functions to replace or remove the old node. If the return value of the visitor function is `None` the node will be removed from the previous location otherwise it's replaced with the return value. The return value may be the original node in which case no replacement takes place. """ def generic_visit(self, node, *args, **kwargs): for field, old_value in node.iter_fields(): if isinstance(old_value, list): new_values = [] for value in old_value: if isinstance(value, Node): value = self.visit(value, *args, **kwargs) if value is None: continue elif not isinstance(value, Node): new_values.extend(value) continue new_values.append(value) old_value[:] = new_values elif isinstance(old_value, Node): new_node = self.visit(old_value, *args, **kwargs) if new_node is None: delattr(node, field) else: setattr(node, field, new_node) return node def visit_list(self, node, *args, **kwargs): """As transformers may return lists in some places this method can be used to enforce a list as return value. """ rv = self.visit(node, *args, **kwargs) if not isinstance(rv, list): rv = [rv] return rv
apache-2.0
manipopopo/tensorflow
tensorflow/contrib/learn/python/learn/__init__.py
40
2715
# Copyright 2016 The TensorFlow Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """High level API for learning with TensorFlow (deprecated). This module and all its submodules are deprecated. See [contrib/learn/README.md](https://www.tensorflow.org/code/tensorflow/contrib/learn/README.md) for migration instructions. """ from __future__ import absolute_import from __future__ import division from __future__ import print_function # pylint: disable=wildcard-import from tensorflow.contrib.learn.python.learn import basic_session_run_hooks from tensorflow.contrib.learn.python.learn import datasets from tensorflow.contrib.learn.python.learn import estimators from tensorflow.contrib.learn.python.learn import graph_actions from tensorflow.contrib.learn.python.learn import learn_io as io from tensorflow.contrib.learn.python.learn import models from tensorflow.contrib.learn.python.learn import monitors from tensorflow.contrib.learn.python.learn import ops from tensorflow.contrib.learn.python.learn import preprocessing from tensorflow.contrib.learn.python.learn import utils from tensorflow.contrib.learn.python.learn.estimators import * from tensorflow.contrib.learn.python.learn.evaluable import Evaluable from tensorflow.contrib.learn.python.learn.experiment import Experiment from tensorflow.contrib.learn.python.learn.export_strategy import ExportStrategy from tensorflow.contrib.learn.python.learn.graph_actions import evaluate from tensorflow.contrib.learn.python.learn.graph_actions import infer from tensorflow.contrib.learn.python.learn.graph_actions import run_feeds from tensorflow.contrib.learn.python.learn.graph_actions import run_n from tensorflow.contrib.learn.python.learn.graph_actions import train from tensorflow.contrib.learn.python.learn.learn_io import * from tensorflow.contrib.learn.python.learn.metric_spec import MetricSpec from tensorflow.contrib.learn.python.learn.monitors import NanLossDuringTrainingError from tensorflow.contrib.learn.python.learn.trainable import Trainable from tensorflow.contrib.learn.python.learn.utils import * # pylint: enable=wildcard-import
apache-2.0
mileistone/test
vedanet/data/transform/_preprocess.py
1
21012
# # Image and annotations preprocessing for lightnet networks # The image transformations work with both Pillow and OpenCV images # The annotation transformations work with brambox.annotations.Annotation objects # Copyright EAVISE # # modified by mileistone import random import collections import logging as log import torch import numpy as np from PIL import Image, ImageOps import brambox.boxes as bbb from .util import BaseTransform, BaseMultiTransform try: import cv2 except ImportError: log.warn('OpenCV is not installed and cannot be used') cv2 = None __all__ = ['Letterbox', 'RandomCrop', 'RandomCropLetterbox', 'RandomFlip', 'HSVShift', 'BramboxToTensor'] class Letterbox(BaseMultiTransform): """ Transform images and annotations to the right network dimensions. Args: dimension (tuple, optional): Default size for the letterboxing, expressed as a (width, height) tuple; Default **None** dataset (lightnet.data.Dataset, optional): Dataset that uses this transform; Default **None** Note: Create 1 Letterbox object and use it for both image and annotation transforms. This object will save data from the image transform and use that on the annotation transform. """ def __init__(self, dimension=None, dataset=None): super().__init__(dimension=dimension, dataset=dataset) if self.dimension is None and self.dataset is None: raise ValueError('This transform either requires a dimension or a dataset to infer the dimension') self.pad = None self.scale = None self.fill_color = 127 def __call__(self, data): if data is None: return None elif isinstance(data, collections.Sequence): return self._tf_anno(data) elif isinstance(data, Image.Image): return self._tf_pil(data) elif isinstance(data, np.ndarray): return self._tf_cv(data) else: log.error(f'Letterbox only works with <brambox annotation lists>, <PIL images> or <OpenCV images> [{type(data)}]') return data def _tf_pil(self, img): """ Letterbox an image to fit in the network """ if self.dataset is not None: net_w, net_h = self.dataset.input_dim else: net_w, net_h = self.dimension im_w, im_h = img.size if im_w == net_w and im_h == net_h: self.scale = None self.pad = None return img # Rescaling if im_w / net_w >= im_h / net_h: self.scale = net_w / im_w else: self.scale = net_h / im_h if self.scale != 1: resample_mode = Image.NEAREST #Image.BILINEAR if self.scale > 1 else Image.ANTIALIAS img = img.resize((int(self.scale*im_w), int(self.scale*im_h)), resample_mode) im_w, im_h = img.size if im_w == net_w and im_h == net_h: self.pad = None return img # Padding img_np = np.array(img) channels = img_np.shape[2] if len(img_np.shape) > 2 else 1 pad_w = (net_w - im_w) / 2 pad_h = (net_h - im_h) / 2 self.pad = (int(pad_w), int(pad_h), int(pad_w+.5), int(pad_h+.5)) img = ImageOps.expand(img, border=self.pad, fill=(self.fill_color,)*channels) return img def _tf_cv(self, img): """ Letterbox and image to fit in the network """ if self.dataset is not None: net_w, net_h = self.dataset.input_dim else: net_w, net_h = self.dimension im_h, im_w = img.shape[:2] if im_w == net_w and im_h == net_h: self.scale = None self.pad = None return img # Rescaling if im_w / net_w >= im_h / net_h: self.scale = net_w / im_w else: self.scale = net_h / im_h if self.scale != 1: img = cv2.resize(img, None, fx=self.scale, fy=self.scale, interpolation=cv2.INTER_CUBIC) im_h, im_w = img.shape[:2] if im_w == net_w and im_h == net_h: self.pad = None return img # Padding channels = img.shape[2] if len(img.shape) > 2 else 1 pad_w = (net_w - im_w) / 2 pad_h = (net_h - im_h) / 2 self.pad = (int(pad_w), int(pad_h), int(pad_w+.5), int(pad_h+.5)) img = cv2.copyMakeBorder(img, self.pad[1], self.pad[3], self.pad[0], self.pad[2], cv2.BORDER_CONSTANT, value=(self.fill_color,)*channels) return img def _tf_anno(self, annos): """ Change coordinates of an annotation, according to the previous letterboxing """ for anno in annos: if self.scale is not None: anno.x_top_left *= self.scale anno.y_top_left *= self.scale anno.width *= self.scale anno.height *= self.scale if self.pad is not None: anno.x_top_left += self.pad[0] anno.y_top_left += self.pad[1] return annos class RandomCrop(BaseMultiTransform): """ Take random crop from the image. Args: jitter (Number [0-1]): Indicates how much of the image we can crop crop_anno(Boolean, optional): Whether we crop the annotations inside the image crop; Default **False** intersection_threshold(number or list, optional): Argument passed on to :class:`brambox.boxes.util.modifiers.CropModifier` Note: Create 1 RandomCrop object and use it for both image and annotation transforms. This object will save data from the image transform and use that on the annotation transform. """ def __init__(self, jitter, crop_anno=False, intersection_threshold=0.001, fill_color=127): super().__init__(jitter=jitter, crop_anno=crop_anno, fill_color=fill_color) self.crop_modifier = bbb.CropModifier(float('Inf'), intersection_threshold) def __call__(self, data): if data is None: return None elif isinstance(data, collections.Sequence): return self._tf_anno(data) elif isinstance(data, Image.Image): return self._tf_pil(data) elif isinstance(data, np.ndarray): return self._tf_cv(data) else: log.error(f'RandomCrop only works with <brambox annotation lists>, <PIL images> or <OpenCV images> [{type(data)}]') return data def _tf_pil(self, img): """ Take random crop from image """ im_w, im_h = img.size crop = self._get_crop(im_w, im_h) crop_w = crop[2] - crop[0] crop_h = crop[3] - crop[1] img_np = np.array(img) channels = img_np.shape[2] if len(img_np.shape) > 2 else 1 img = img.crop((max(0, crop[0]), max(0, crop[1]), min(im_w, crop[2]-1), min(im_h, crop[3]-1))) img_crop = Image.new(img.mode, (crop_w, crop_h), color=(self.fill_color,)*channels) img_crop.paste(img, (max(0, -crop[0]), max(0, -crop[1]))) return img_crop def _tf_cv(self, img): """ Take random crop from image """ im_h, im_w = img.shape[:2] crop = self._get_crop(im_w, im_h) crop_w = crop[2] - crop[0] crop_h = crop[3] - crop[1] img_crop = np.ones((crop_h, crop_w) + img.shape[2:], dtype=img.dtype) * self.fill_color src_x1 = max(0, crop[0]) src_x2 = min(crop[2], im_w) src_y1 = max(0, crop[1]) src_y2 = min(crop[3], im_h) dst_x1 = max(0, -crop[0]) dst_x2 = crop_w - max(0, crop[2]-im_w) dst_y1 = max(0, -crop[1]) dst_y2 = crop_h - max(0, crop[3]-im_h) img_crop[dst_y1:dst_y2, dst_x1:dst_x2] = img[src_y1:src_y2, src_x1:src_x2] return img_crop def _get_crop(self, im_w, im_h): dw, dh = int(im_w*self.jitter), int(im_h*self.jitter) crop_left = random.randint(-dw, dw) crop_right = random.randint(-dw, dw) crop_top = random.randint(-dh, dh) crop_bottom = random.randint(-dh, dh) crop = (crop_left, crop_top, im_w-crop_right, im_h-crop_bottom) self.crop_modifier.area = crop return crop def _tf_anno(self, annos): """ Change coordinates of an annotation, according to the previous crop """ if self.crop_anno: bbb.modify(annos, [self.crop_modifier]) else: crop = self.crop_modifier.area for i in range(len(annos)-1, -1, -1): anno = annos[i] x1 = max(crop[0], anno.x_top_left) x2 = min(crop[2], anno.x_top_left+anno.width) y1 = max(crop[1], anno.y_top_left) y2 = min(crop[3], anno.y_top_left+anno.height) w = x2-x1 h = y2-y1 if self.crop_modifier.inter_area: ratio = ((w * h) / (anno.width * anno.height)) < self.crop_modifier.inter_thresh else: ratio = (w / anno.width) < self.crop_modifier.inter_thresh[0] or (h / anno.height) < self.crop_modifier.inter_thresh[1] if w <= 0 or h <= 0 or ratio: del annos[i] continue annos[i].x_top_left -= crop[0] annos[i].y_top_left -= crop[1] return annos class RandomCropLetterbox(BaseMultiTransform): """ Take random crop from the image. Args: jitter (Number [0-1]): Indicates how much of the image we can crop crop_anno(Boolean, optional): Whether we crop the annotations inside the image crop; Default **False** intersection_threshold(number or list, optional): Argument passed on to :class:`brambox.boxes.util.modifiers.CropModifier` Note: Create 1 RandomCrop object and use it for both image and annotation transforms. This object will save data from the image transform and use that on the annotation transform. """ def __init__(self, dataset, jitter, fill_color=127): super().__init__(dataset=dataset, jitter=jitter, fill_color=fill_color) self.crop_info = None self.output_w = None self.output_h = None def __call__(self, data): if data is None: return None elif isinstance(data, collections.Sequence): return self._tf_anno(data) elif isinstance(data, Image.Image): return self._tf_pil(data) else: log.error(f'RandomCrop only works with <brambox annotation lists>, <PIL images> or <OpenCV images> [{type(data)}]') return data def _tf_pil(self, img): """ Take random crop from image """ self.output_w, self.output_h = self.dataset.input_dim #print('output shape: %d, %d' % (self.output_w, self.output_h)) orig_w, orig_h = img.size img_np = np.array(img) channels = img_np.shape[2] if len(img_np.shape) > 2 else 1 dw = int(self.jitter * orig_w) dh = int(self.jitter * orig_h) new_ar = float(orig_w + random.randint(-dw, dw)) / (orig_h + random.randint(-dh, dh)) scale = random.random()*(2-0.25) + 0.25 if new_ar < 1: nh = int(scale * orig_h) nw = int(nh * new_ar) else: nw = int(scale * orig_w) nh = int(nw / new_ar) if self.output_w > nw: dx = random.randint(0, self.output_w - nw) else: dx = random.randint(self.output_w - nw, 0) if self.output_h > nh: dy = random.randint(0, self.output_h - nh) else: dy = random.randint(self.output_h - nh, 0) nxmin = max(0, -dx) nymin = max(0, -dy) nxmax = min(nw, -dx + self.output_w - 1) nymax = min(nh, -dy + self.output_h - 1) sx, sy = float(orig_w)/nw, float(orig_h)/nh orig_xmin = int(nxmin * sx) orig_ymin = int(nymin * sy) orig_xmax = int(nxmax * sx) orig_ymax = int(nymax * sy) orig_crop = img.crop((orig_xmin, orig_ymin, orig_xmax, orig_ymax)) orig_crop_resize = orig_crop.resize((nxmax - nxmin, nymax - nymin)) output_img = Image.new(img.mode, (self.output_w, self.output_h), color=(self.fill_color,)*channels) output_img.paste(orig_crop_resize, (0, 0)) self.crop_info = [sx, sy, nxmin, nymin, nxmax, nymax] return output_img def _tf_anno(self, annos): """ Change coordinates of an annotation, according to the previous crop """ sx, sy, crop_xmin, crop_ymin, crop_xmax, crop_ymax = self.crop_info for i in range(len(annos)-1, -1, -1): anno = annos[i] x1 = max(crop_xmin, int(anno.x_top_left/sx)) x2 = min(crop_xmax, int((anno.x_top_left+anno.width)/sx)) y1 = max(crop_ymin, int(anno.y_top_left/sy)) y2 = min(crop_ymax, int((anno.y_top_left+anno.height)/sy)) w = x2-x1 h = y2-y1 if w <= 2 or h <= 2: # or w*h/(anno.width*anno.height/sx/sy) <= 0.5: del annos[i] continue annos[i].x_top_left = x1 - crop_xmin annos[i].y_top_left = y1 -crop_ymin annos[i].width = w annos[i].height = h return annos class RandomFlip(BaseMultiTransform): """ Randomly flip image. Args: threshold (Number [0-1]): Chance of flipping the image Note: Create 1 RandomFlip object and use it for both image and annotation transforms. This object will save data from the image transform and use that on the annotation transform. """ def __init__(self, threshold): self.threshold = threshold self.flip = False self.im_w = None def __call__(self, data): if data is None: return None elif isinstance(data, collections.Sequence): return [self._tf_anno(anno) for anno in data] elif isinstance(data, Image.Image): return self._tf_pil(data) elif isinstance(data, np.ndarray): return self._tf_cv(data) else: log.error(f'RandomFlip only works with <brambox annotation lists>, <PIL images> or <OpenCV images> [{type(data)}]') return data def _tf_pil(self, img): """ Randomly flip image """ self._get_flip() self.im_w = img.size[0] if self.flip: img = img.transpose(Image.FLIP_LEFT_RIGHT) return img def _tf_cv(self, img): """ Randomly flip image """ self._get_flip() self.im_w = img.shape[1] if self.flip: img = cv2.flip(img, 1) return img def _get_flip(self): self.flip = random.random() < self.threshold def _tf_anno(self, anno): """ Change coordinates of an annotation, according to the previous flip """ if self.flip and self.im_w is not None: anno.x_top_left = self.im_w - anno.x_top_left - anno.width return anno class HSVShift(BaseTransform): """ Perform random HSV shift on the RGB data. Args: hue (Number): Random number between -hue,hue is used to shift the hue saturation (Number): Random number between 1,saturation is used to shift the saturation; 50% chance to get 1/dSaturation in stead of dSaturation value (Number): Random number between 1,value is used to shift the value; 50% chance to get 1/dValue in stead of dValue Warning: If you use OpenCV as your image processing library, make sure the image is RGB before using this transform. By default OpenCV uses BGR, so you must use `cvtColor`_ function to transform it to RGB. .. _cvtColor: https://docs.opencv.org/master/d7/d1b/group__imgproc__misc.html#ga397ae87e1288a81d2363b61574eb8cab """ def __init__(self, hue, saturation, value): super().__init__(hue=hue, saturation=saturation, value=value) @classmethod def apply(cls, data, hue, saturation, value): dh = random.uniform(-hue, hue) ds = random.uniform(1, saturation) if random.random() < 0.5: ds = 1/ds dv = random.uniform(1, value) if random.random() < 0.5: dv = 1/dv if data is None: return None elif isinstance(data, Image.Image): return cls._tf_pil(data, dh, ds, dv) elif isinstance(data, np.ndarray): return cls._tf_cv(data, dh, ds, dv) else: log.error(f'HSVShift only works with <PIL images> or <OpenCV images> [{type(data)}]') return data @staticmethod def _tf_pil(img, dh, ds, dv): """ Random hsv shift """ img = img.convert('HSV') channels = list(img.split()) def change_hue(x): x += int(dh * 255) if x > 255: x -= 255 elif x < 0: x += 0 return x channels[0] = channels[0].point(change_hue) channels[1] = channels[1].point(lambda i: min(255, max(0, int(i*ds)))) channels[2] = channels[2].point(lambda i: min(255, max(0, int(i*dv)))) img = Image.merge(img.mode, tuple(channels)) img = img.convert('RGB') return img @staticmethod def _tf_cv(img, dh, ds, dv): """ Random hsv shift """ img = img.astype(np.float32) / 255.0 img = cv2.cvtColor(img, cv2.COLOR_RGB2HSV) def wrap_hue(x): x[x >= 360.0] -= 360.0 x[x < 0.0] += 360.0 return x img[:, :, 0] = wrap_hue(hsv[:, :, 0] + (360.0 * dh)) img[:, :, 1] = np.clip(ds * img[:, :, 1], 0.0, 1.0) img[:, :, 2] = np.clip(dv * img[:, :, 2], 0.0, 1.0) img = cv2.cvtColor(img, cv2.COLOR_HSV2RGB) img = (img * 255).astype(np.uint8) return img class BramboxToTensor(BaseTransform): """ Converts a list of brambox annotation objects to a tensor. Args: dimension (tuple, optional): Default size of the transformed images, expressed as a (width, height) tuple; Default **None** dataset (lightnet.data.Dataset, optional): Dataset that uses this transform; Default **None** max_anno (Number, optional): Maximum number of annotations in the list; Default **50** class_label_map (list, optional): class label map to convert class names to an index; Default **None** Return: torch.Tensor: tensor of dimension [max_anno, 5] containing [class_idx,center_x,center_y,width,height] for every detection Warning: If no class_label_map is given, this function will first try to convert the class_label to an integer. If that fails, it is simply given the number 0. """ def __init__(self, dimension=None, dataset=None, max_anno=50, class_label_map=None): super().__init__(dimension=dimension, dataset=dataset, max_anno=max_anno, class_label_map=class_label_map) if self.dimension is None and self.dataset is None: raise ValueError('This transform either requires a dimension or a dataset to infer the dimension') if self.class_label_map is None: log.warn('No class_label_map given. If the class_labels are not integers, they will be set to zero.') def __call__(self, data): if self.dataset is not None: dim = self.dataset.input_dim else: dim = self.dimension return self.apply(data, dim, self.max_anno, self.class_label_map) @classmethod def apply(cls, data, dimension, max_anno=None, class_label_map=None): if not isinstance(data, collections.Sequence): raise TypeError(f'BramboxToTensor only works with <brambox annotation list> [{type(data)}]') anno_np = np.array([cls._tf_anno(anno, dimension, class_label_map) for anno in data], dtype=np.float32) if max_anno is not None: anno_len = len(data) if anno_len > max_anno: raise ValueError(f'More annotations than maximum allowed [{anno_len}/{max_anno}]') z_np = np.zeros((max_anno-anno_len, 5), dtype=np.float32) z_np[:, 0] = -1 if anno_len > 0: return torch.from_numpy(np.concatenate((anno_np, z_np))) else: return torch.from_numpy(z_np) else: return torch.from_numpy(anno_np) @staticmethod def _tf_anno(anno, dimension, class_label_map): """ Transforms brambox annotation to list """ net_w, net_h = dimension if class_label_map is not None: cls = class_label_map.index(anno.class_label) else: try: cls = int(anno.class_label) except ValueError: cls = 0 cx = (anno.x_top_left + (anno.width / 2)) / net_w cy = (anno.y_top_left + (anno.height / 2)) / net_h w = anno.width / net_w h = anno.height / net_h return [cls, cx, cy, w, h]
mit
caronc/nzb-subliminal
Subliminal/guessit/transfo/guess_episode_details.py
7
2809
#!/usr/bin/env python # -*- coding: utf-8 -*- # # GuessIt - A library for guessing information from filenames # Copyright (c) 2013 Nicolas Wack <wackou@gmail.com> # # GuessIt is free software; you can redistribute it and/or modify it under # the terms of the Lesser GNU General Public License as published by # the Free Software Foundation; either version 3 of the License, or # (at your option) any later version. # # GuessIt is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # Lesser GNU General Public License for more details. # # You should have received a copy of the Lesser GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. # from __future__ import absolute_import, division, print_function, unicode_literals from guessit.plugins.transformers import Transformer from guessit.matcher import found_guess from guessit.containers import PropertiesContainer import itertools class GuessEpisodeDetails(Transformer): def __init__(self): Transformer.__init__(self, -205) self.container = PropertiesContainer() self.container.register_property('episodeDetails', 'Special', 'Bonus', 'Omake', 'Ova', 'Oav', 'Pilot', 'Unaired') self.container.register_property('episodeDetails', 'Extras?', canonical_form='Extras') def guess_details(self, string, node=None, options=None): properties = self.container.find_properties(string, node, options, 'episodeDetails', multiple=True) guesses = self.container.as_guess(properties, multiple=True) return guesses def second_pass_options(self, mtree, options=None): if not mtree.guess.get('type', '').startswith('episode'): for unidentified_leaf in mtree.unidentified_leaves(): properties = self.container.find_properties(unidentified_leaf.value, unidentified_leaf, options, 'episodeDetails') guess = self.container.as_guess(properties) if guess: return {'type': 'episode'} return None def supported_properties(self): return self.container.get_supported_properties() def process(self, mtree, options=None): if (mtree.guess.get('type', '').startswith('episode') and (not mtree.info.get('episodeNumber') or mtree.info.get('season') == 0)): for leaf in itertools.chain(mtree.leaves_containing('title'), mtree.unidentified_leaves()): guesses = self.guess_details(leaf.value, leaf, options) for guess in guesses: found_guess(leaf, guess, update_guess=False) return None
gpl-3.0
luo66/scikit-learn
examples/mixture/plot_gmm_sin.py
247
2747
""" ================================= Gaussian Mixture Model Sine Curve ================================= This example highlights the advantages of the Dirichlet Process: complexity control and dealing with sparse data. The dataset is formed by 100 points loosely spaced following a noisy sine curve. The fit by the GMM class, using the expectation-maximization algorithm to fit a mixture of 10 Gaussian components, finds too-small components and very little structure. The fits by the Dirichlet process, however, show that the model can either learn a global structure for the data (small alpha) or easily interpolate to finding relevant local structure (large alpha), never falling into the problems shown by the GMM class. """ import itertools import numpy as np from scipy import linalg import matplotlib.pyplot as plt import matplotlib as mpl from sklearn import mixture from sklearn.externals.six.moves import xrange # Number of samples per component n_samples = 100 # Generate random sample following a sine curve np.random.seed(0) X = np.zeros((n_samples, 2)) step = 4 * np.pi / n_samples for i in xrange(X.shape[0]): x = i * step - 6 X[i, 0] = x + np.random.normal(0, 0.1) X[i, 1] = 3 * (np.sin(x) + np.random.normal(0, .2)) color_iter = itertools.cycle(['r', 'g', 'b', 'c', 'm']) for i, (clf, title) in enumerate([ (mixture.GMM(n_components=10, covariance_type='full', n_iter=100), "Expectation-maximization"), (mixture.DPGMM(n_components=10, covariance_type='full', alpha=0.01, n_iter=100), "Dirichlet Process,alpha=0.01"), (mixture.DPGMM(n_components=10, covariance_type='diag', alpha=100., n_iter=100), "Dirichlet Process,alpha=100.")]): clf.fit(X) splot = plt.subplot(3, 1, 1 + i) Y_ = clf.predict(X) for i, (mean, covar, color) in enumerate(zip( clf.means_, clf._get_covars(), color_iter)): v, w = linalg.eigh(covar) u = w[0] / linalg.norm(w[0]) # as the DP will not use every component it has access to # unless it needs it, we shouldn't plot the redundant # components. if not np.any(Y_ == i): continue plt.scatter(X[Y_ == i, 0], X[Y_ == i, 1], .8, color=color) # Plot an ellipse to show the Gaussian component angle = np.arctan(u[1] / u[0]) angle = 180 * angle / np.pi # convert to degrees ell = mpl.patches.Ellipse(mean, v[0], v[1], 180 + angle, color=color) ell.set_clip_box(splot.bbox) ell.set_alpha(0.5) splot.add_artist(ell) plt.xlim(-6, 4 * np.pi - 6) plt.ylim(-5, 5) plt.title(title) plt.xticks(()) plt.yticks(()) plt.show()
bsd-3-clause
automl/auto-sklearn
examples/80_extending/example_extending_regression.py
1
5223
""" ================================================ Extending Auto-Sklearn with Regression Component ================================================ The following example demonstrates how to create a new regression component for using in auto-sklearn. """ from typing import Optional from pprint import pprint from ConfigSpace.configuration_space import ConfigurationSpace from ConfigSpace.hyperparameters import ( UniformFloatHyperparameter, UniformIntegerHyperparameter, CategoricalHyperparameter, ) from ConfigSpace.conditions import EqualsCondition import sklearn.metrics from autosklearn.askl_typing import FEAT_TYPE_TYPE import autosklearn.regression import autosklearn.pipeline.components.regression from autosklearn.pipeline.components.base import AutoSklearnRegressionAlgorithm from autosklearn.pipeline.constants import ( SPARSE, DENSE, SIGNED_DATA, UNSIGNED_DATA, PREDICTIONS, ) from sklearn.datasets import load_diabetes from sklearn.model_selection import train_test_split ############################################################################ # Implement kernel ridge regression component for auto-sklearn # ============================================================ class KernelRidgeRegression(AutoSklearnRegressionAlgorithm): def __init__(self, alpha, kernel, gamma, degree, coef0, random_state=None): self.alpha = alpha self.kernel = kernel self.gamma = gamma self.degree = degree self.coef0 = coef0 self.random_state = random_state self.estimator = None def fit(self, X, y): self.alpha = float(self.alpha) self.gamma = float(self.gamma) self.degree = int(self.degree) self.coef0 = float(self.coef0) import sklearn.kernel_ridge self.estimator = sklearn.kernel_ridge.KernelRidge( alpha=self.alpha, kernel=self.kernel, gamma=self.gamma, degree=self.degree, coef0=self.coef0, ) self.estimator.fit(X, y) return self def predict(self, X): if self.estimator is None: raise NotImplementedError return self.estimator.predict(X) @staticmethod def get_properties(dataset_properties=None): return { "shortname": "KRR", "name": "Kernel Ridge Regression", "handles_regression": True, "handles_classification": False, "handles_multiclass": False, "handles_multilabel": False, "handles_multioutput": True, "is_deterministic": True, "input": (SPARSE, DENSE, UNSIGNED_DATA, SIGNED_DATA), "output": (PREDICTIONS,), } @staticmethod def get_hyperparameter_search_space( feat_type: Optional[FEAT_TYPE_TYPE] = None, dataset_properties=None ): cs = ConfigurationSpace() alpha = UniformFloatHyperparameter( name="alpha", lower=10**-5, upper=1, log=True, default_value=1.0 ) kernel = CategoricalHyperparameter( name="kernel", # We restrict ourselves to two possible kernels for this example choices=["polynomial", "rbf"], default_value="polynomial", ) gamma = UniformFloatHyperparameter( name="gamma", lower=0.00001, upper=1, default_value=0.1, log=True ) degree = UniformIntegerHyperparameter( name="degree", lower=2, upper=5, default_value=3 ) coef0 = UniformFloatHyperparameter( name="coef0", lower=1e-2, upper=1e2, log=True, default_value=1, ) cs.add_hyperparameters([alpha, kernel, gamma, degree, coef0]) degree_condition = EqualsCondition(degree, kernel, "polynomial") coef0_condition = EqualsCondition(coef0, kernel, "polynomial") cs.add_conditions([degree_condition, coef0_condition]) return cs # Add KRR component to auto-sklearn. autosklearn.pipeline.components.regression.add_regressor(KernelRidgeRegression) cs = KernelRidgeRegression.get_hyperparameter_search_space() print(cs) ############################################################################ # Generate data # ============= X, y = load_diabetes(return_X_y=True) X_train, X_test, y_train, y_test = train_test_split(X, y) ############################################################################ # Fit the model using KRR # ======================= reg = autosklearn.regression.AutoSklearnRegressor( time_left_for_this_task=30, per_run_time_limit=10, include={"regressor": ["KernelRidgeRegression"]}, # Bellow two flags are provided to speed up calculations # Not recommended for a real implementation initial_configurations_via_metalearning=0, smac_scenario_args={"runcount_limit": 5}, ) reg.fit(X_train, y_train) ############################################################################ # Print prediction score and statistics # ===================================== y_pred = reg.predict(X_test) print("r2 score: ", sklearn.metrics.r2_score(y_pred, y_test)) pprint(reg.show_models(), indent=4)
bsd-3-clause
GauthamGoli/quantify-2016
Machine Learning - Bond Liquidity Prediction/final_code.py
1
5973
# importing various modules that would be required in the program import pandas as pd import numpy as np import dateutil.parser as dateparser from datetime import datetime import dateutil.parser as dateparser from datetime import datetime from sklearn.ensemble import RandomForestRegressor from sklearn.preprocessing import Imputer, LabelEncoder # Preprocessing the ML_Bond_metadata.csv file after initial treatment in excel. data = pd.read_csv('ML_Bond_metadata_corrected_dates.csv') # print data.isnull().sum() numerical_fields = [ 'coupon', 'amtIssued', 'amtOutstanding', ] categorical_fields = [ 'issuer', 'Market', 'collateralType', 'couponFrequency' 'couponType', 'industryGroup', 'industrySector', 'industrySubGroup', 'maturityType', 'securityType', 'paymentRank', '144aflag', 'ratingAgency1Rating', 'ratingAgency2Rating', 'ratingAgency1Watch', 'ratingAgency2Watch' ] date_fields = [ 'issueDate', 'maturity', 'ratingAgency1EffectiveDate', 'ratingAgency2EffectiveDate' ] # The difference in the amounts is a new feature added to the data to give better insights data['AmtDiff'] = data['amtIssued'] - data['amtOutstanding'] # The duration between issue and maturity data['DateDiff'] = data['maturity'] - data['IssueDate'] # Imputing values in the columns where NANs are found for i in ['issueDate','maturity','ratingAgency1EffectiveDate','ratingAgency1EffectiveDate']: data[i] = data[i].fillna(data.median()) # Changing the value of couponFrequency from NAN to 0 if coupon is also zero temp = [] for i in range(leng): if data['coupon'].iloc[i] == 0.00: temp.append(i) for i in temp: j = j+1 data.set_value(i,'couponFrequency',0) # For Cleaning Categorical Data for i in ['couponFrequency']: data[i] = LabelEncoder().fit_transform(data[i]) # For Cleaning Numeric Data for i in ['AmtDiff','DateDiff','amtOutstanding', 'amtIssued']: data[i] = StandardScaler().fit_transform(data[i]) pd.to_csv('metadata_clean.csv') # functions which will be used for cleaning the dataset.csv file def get_cluster(x): if x == 0: return 'A' elif x == 1: return 'B' elif x == 2: return 'C' elif x == 3: return 'D' elif x == 4: return 'E' elif x == 5: return 'F' elif x == 6: return 'G' elif x == 7: return 'H' elif x == 8: return 'I' elif x == 9: return 'J' # Starting of the training time i.e. 16 March 2016 is considered as the epoch def get_days(x): dt = datetime.strptime(x, "%d%b%Y") epoch = datetime(2016,3,16) return int((dt-epoch).total_seconds()/86400) def get_time(x): if x.find('pm') == -1 and x.find('am') == -1: dt = datetime.strptime(x[:26], "%a %d%b%Y %H:%M:%S.%f") else: dt = datetime.strptime(x, "%a %d%b%Y %I:%M:%S.%f %p") epoch = datetime(2016,3,16) diff = dt - epoch return int(diff.total_seconds()) def get_side_back(x): if x == 0: return 'S' elif x == 1: return 'B' def get_isin(x): return int(x[4:]) def correct_time(x): return x[:9]+'20'+x[9:] # Clustering of the Bonds # importing the module containg the python implementation of k-ptototype algorithm from kmodes import kprototypes X = data.as_matrix() kproto = kprototypes.KPrototypes(n_clusters=10, init='Cao', verbose=2) clusters = kproto.fit_predict(X, categorical=[0, 2, 5, 6, 7, 8, 9, 10, 11 ,12, 13, 14, 15, 16, 18, 19]) # New column has been created in the dataset for clusters data['cluster'] = clusters # saving the results to a csv file data.to_csv('metadata_clean_cluster_10.csv') # creating a temporary dataframe temp_df = pd.DataFrame(columns=['isin','cluster']) temp_df['isin'] = temp_df['isin'] temp_df['cluster'] = temp_df['cluster'] data1 = pd.read_csv('dataset.csv') # price culd not be modelled data1 = data1.drop(['price']) data1['isin'] = data1['isin'].apply(get_isin) data1['time'] = data1['time'].apply(correct_time) data1['time'] = data1['time'].apply(get_time) data1['date'] = data1['date'].apply(get_days) data1 = data1.drop(['time'],axis=1) # sequence to compress the various trades in a day into one single entry isins = data1['isin'].unique() for i in isins: temp = data1[data1['isin'] == i] temp_dates = temp['date'].unique() for j in temp_dates: temp1 = temp[temp['date'] == j] temp_side = temp1['side'].unique() for k in temp_side: temp2 = temp1[temp1['side'] == k] sum_vol = temp2['volume'].sum() res1 = pd.DataFrame([[i,sum_vol,k,j]],columns=['isin','volume','side','date']) res = res.append(res1) data1 = res # merge tables to include cluster data in the dataset also data1 = data1.merge(temp_df, on='isin', how='left') data1.to_csv('dataset_clean_cluster10.csv') data1['cluster'] = data1['cluster'].apply(get_cluster) # dividing the data into training and validation test data1['is_t_data'] = np.random.uniform(0,1,len(data))<=0.75 train, validate = data1[data1['is_t_data']==True], data1[data1['is_t_data']==False] # for the purpose of validation tests train = train.drop(['is_t_data'],axis=1) validate = validate.drop(['is_t_data'],axis=1) model = MixedLM.from_formula('volume ~ side + date + cluster', data = data1, re_formula = 'date', groups = train['isin']) result = model.fit() print result.summary() # template was created for easy access of output parameters out = pd.read_csv('output_template.csv') out['dummy'] = 0 dummy, X = dmatrices('dummy ~ side + date - 1', data=out, return_type='dataframe') values = model.predict(X) # values are returned as an numpy ndarray and after careful addition in excel, the csv file is uploaded.
mit
MohammedWasim/scikit-learn
sklearn/neighbors/base.py
71
31147
"""Base and mixin classes for nearest neighbors""" # Authors: Jake Vanderplas <vanderplas@astro.washington.edu> # Fabian Pedregosa <fabian.pedregosa@inria.fr> # Alexandre Gramfort <alexandre.gramfort@inria.fr> # Sparseness support by Lars Buitinck <L.J.Buitinck@uva.nl> # Multi-output support by Arnaud Joly <a.joly@ulg.ac.be> # # License: BSD 3 clause (C) INRIA, University of Amsterdam import warnings from abc import ABCMeta, abstractmethod import numpy as np from scipy.sparse import csr_matrix, issparse from .ball_tree import BallTree from .kd_tree import KDTree from ..base import BaseEstimator from ..metrics import pairwise_distances from ..metrics.pairwise import PAIRWISE_DISTANCE_FUNCTIONS from ..utils import check_X_y, check_array, _get_n_jobs, gen_even_slices from ..utils.fixes import argpartition from ..utils.validation import DataConversionWarning from ..utils.validation import NotFittedError from ..externals import six from ..externals.joblib import Parallel, delayed VALID_METRICS = dict(ball_tree=BallTree.valid_metrics, kd_tree=KDTree.valid_metrics, # The following list comes from the # sklearn.metrics.pairwise doc string brute=(list(PAIRWISE_DISTANCE_FUNCTIONS.keys()) + ['braycurtis', 'canberra', 'chebyshev', 'correlation', 'cosine', 'dice', 'hamming', 'jaccard', 'kulsinski', 'mahalanobis', 'matching', 'minkowski', 'rogerstanimoto', 'russellrao', 'seuclidean', 'sokalmichener', 'sokalsneath', 'sqeuclidean', 'yule', 'wminkowski'])) VALID_METRICS_SPARSE = dict(ball_tree=[], kd_tree=[], brute=PAIRWISE_DISTANCE_FUNCTIONS.keys()) class NeighborsWarning(UserWarning): pass # Make sure that NeighborsWarning are displayed more than once warnings.simplefilter("always", NeighborsWarning) def _check_weights(weights): """Check to make sure weights are valid""" if weights in (None, 'uniform', 'distance'): return weights elif callable(weights): return weights else: raise ValueError("weights not recognized: should be 'uniform', " "'distance', or a callable function") def _get_weights(dist, weights): """Get the weights from an array of distances and a parameter ``weights`` Parameters =========== dist: ndarray The input distances weights: {'uniform', 'distance' or a callable} The kind of weighting used Returns ======== weights_arr: array of the same shape as ``dist`` if ``weights == 'uniform'``, then returns None """ if weights in (None, 'uniform'): return None elif weights == 'distance': # if user attempts to classify a point that was zero distance from one # or more training points, those training points are weighted as 1.0 # and the other points as 0.0 if dist.dtype is np.dtype(object): for point_dist_i, point_dist in enumerate(dist): # check if point_dist is iterable # (ex: RadiusNeighborClassifier.predict may set an element of # dist to 1e-6 to represent an 'outlier') if hasattr(point_dist, '__contains__') and 0. in point_dist: dist[point_dist_i] = point_dist == 0. else: dist[point_dist_i] = 1. / point_dist else: with np.errstate(divide='ignore'): dist = 1. / dist inf_mask = np.isinf(dist) inf_row = np.any(inf_mask, axis=1) dist[inf_row] = inf_mask[inf_row] return dist elif callable(weights): return weights(dist) else: raise ValueError("weights not recognized: should be 'uniform', " "'distance', or a callable function") class NeighborsBase(six.with_metaclass(ABCMeta, BaseEstimator)): """Base class for nearest neighbors estimators.""" @abstractmethod def __init__(self): pass def _init_params(self, n_neighbors=None, radius=None, algorithm='auto', leaf_size=30, metric='minkowski', p=2, metric_params=None, n_jobs=1, **kwargs): if kwargs: warnings.warn("Passing additional arguments to the metric " "function as **kwargs is deprecated " "and will no longer be supported in 0.18. " "Use metric_params instead.", DeprecationWarning, stacklevel=3) if metric_params is None: metric_params = {} metric_params.update(kwargs) self.n_neighbors = n_neighbors self.radius = radius self.algorithm = algorithm self.leaf_size = leaf_size self.metric = metric self.metric_params = metric_params self.p = p self.n_jobs = n_jobs if algorithm not in ['auto', 'brute', 'kd_tree', 'ball_tree']: raise ValueError("unrecognized algorithm: '%s'" % algorithm) if algorithm == 'auto': if metric == 'precomputed': alg_check = 'brute' else: alg_check = 'ball_tree' else: alg_check = algorithm if callable(metric): if algorithm == 'kd_tree': # callable metric is only valid for brute force and ball_tree raise ValueError( "kd_tree algorithm does not support callable metric '%s'" % metric) elif metric not in VALID_METRICS[alg_check]: raise ValueError("Metric '%s' not valid for algorithm '%s'" % (metric, algorithm)) if self.metric_params is not None and 'p' in self.metric_params: warnings.warn("Parameter p is found in metric_params. " "The corresponding parameter from __init__ " "is ignored.", SyntaxWarning, stacklevel=3) effective_p = metric_params['p'] else: effective_p = self.p if self.metric in ['wminkowski', 'minkowski'] and effective_p < 1: raise ValueError("p must be greater than one for minkowski metric") self._fit_X = None self._tree = None self._fit_method = None def _fit(self, X): if self.metric_params is None: self.effective_metric_params_ = {} else: self.effective_metric_params_ = self.metric_params.copy() effective_p = self.effective_metric_params_.get('p', self.p) if self.metric in ['wminkowski', 'minkowski']: self.effective_metric_params_['p'] = effective_p self.effective_metric_ = self.metric # For minkowski distance, use more efficient methods where available if self.metric == 'minkowski': p = self.effective_metric_params_.pop('p', 2) if p < 1: raise ValueError("p must be greater than one " "for minkowski metric") elif p == 1: self.effective_metric_ = 'manhattan' elif p == 2: self.effective_metric_ = 'euclidean' elif p == np.inf: self.effective_metric_ = 'chebyshev' else: self.effective_metric_params_['p'] = p if isinstance(X, NeighborsBase): self._fit_X = X._fit_X self._tree = X._tree self._fit_method = X._fit_method return self elif isinstance(X, BallTree): self._fit_X = X.data self._tree = X self._fit_method = 'ball_tree' return self elif isinstance(X, KDTree): self._fit_X = X.data self._tree = X self._fit_method = 'kd_tree' return self X = check_array(X, accept_sparse='csr') n_samples = X.shape[0] if n_samples == 0: raise ValueError("n_samples must be greater than 0") if issparse(X): if self.algorithm not in ('auto', 'brute'): warnings.warn("cannot use tree with sparse input: " "using brute force") if self.effective_metric_ not in VALID_METRICS_SPARSE['brute']: raise ValueError("metric '%s' not valid for sparse input" % self.effective_metric_) self._fit_X = X.copy() self._tree = None self._fit_method = 'brute' return self self._fit_method = self.algorithm self._fit_X = X if self._fit_method == 'auto': # A tree approach is better for small number of neighbors, # and KDTree is generally faster when available if ((self.n_neighbors is None or self.n_neighbors < self._fit_X.shape[0] // 2) and self.metric != 'precomputed'): if self.effective_metric_ in VALID_METRICS['kd_tree']: self._fit_method = 'kd_tree' else: self._fit_method = 'ball_tree' else: self._fit_method = 'brute' if self._fit_method == 'ball_tree': self._tree = BallTree(X, self.leaf_size, metric=self.effective_metric_, **self.effective_metric_params_) elif self._fit_method == 'kd_tree': self._tree = KDTree(X, self.leaf_size, metric=self.effective_metric_, **self.effective_metric_params_) elif self._fit_method == 'brute': self._tree = None else: raise ValueError("algorithm = '%s' not recognized" % self.algorithm) if self.n_neighbors is not None: if self.n_neighbors <= 0: raise ValueError( "Expected n_neighbors > 0. Got %d" % self.n_neighbors ) return self @property def _pairwise(self): # For cross-validation routines to split data correctly return self.metric == 'precomputed' class KNeighborsMixin(object): """Mixin for k-neighbors searches""" def kneighbors(self, X=None, n_neighbors=None, return_distance=True): """Finds the K-neighbors of a point. Returns indices of and distances to the neighbors of each point. Parameters ---------- X : array-like, shape (n_query, n_features), \ or (n_query, n_indexed) if metric == 'precomputed' The query point or points. If not provided, neighbors of each indexed point are returned. In this case, the query point is not considered its own neighbor. n_neighbors : int Number of neighbors to get (default is the value passed to the constructor). return_distance : boolean, optional. Defaults to True. If False, distances will not be returned Returns ------- dist : array Array representing the lengths to points, only present if return_distance=True ind : array Indices of the nearest points in the population matrix. Examples -------- In the following example, we construct a NeighborsClassifier class from an array representing our data set and ask who's the closest point to [1,1,1] >>> samples = [[0., 0., 0.], [0., .5, 0.], [1., 1., .5]] >>> from sklearn.neighbors import NearestNeighbors >>> neigh = NearestNeighbors(n_neighbors=1) >>> neigh.fit(samples) # doctest: +ELLIPSIS NearestNeighbors(algorithm='auto', leaf_size=30, ...) >>> print(neigh.kneighbors([[1., 1., 1.]])) # doctest: +ELLIPSIS (array([[ 0.5]]), array([[2]]...)) As you can see, it returns [[0.5]], and [[2]], which means that the element is at distance 0.5 and is the third element of samples (indexes start at 0). You can also query for multiple points: >>> X = [[0., 1., 0.], [1., 0., 1.]] >>> neigh.kneighbors(X, return_distance=False) # doctest: +ELLIPSIS array([[1], [2]]...) """ if self._fit_method is None: raise NotFittedError("Must fit neighbors before querying.") if n_neighbors is None: n_neighbors = self.n_neighbors if X is not None: query_is_train = False X = check_array(X, accept_sparse='csr') else: query_is_train = True X = self._fit_X # Include an extra neighbor to account for the sample itself being # returned, which is removed later n_neighbors += 1 train_size = self._fit_X.shape[0] if n_neighbors > train_size: raise ValueError( "Expected n_neighbors <= n_samples, " " but n_samples = %d, n_neighbors = %d" % (train_size, n_neighbors) ) n_samples, _ = X.shape sample_range = np.arange(n_samples)[:, None] n_jobs = _get_n_jobs(self.n_jobs) if self._fit_method == 'brute': # for efficiency, use squared euclidean distances if self.effective_metric_ == 'euclidean': dist = pairwise_distances(X, self._fit_X, 'euclidean', n_jobs=n_jobs, squared=True) else: dist = pairwise_distances( X, self._fit_X, self.effective_metric_, n_jobs=n_jobs, **self.effective_metric_params_) neigh_ind = argpartition(dist, n_neighbors - 1, axis=1) neigh_ind = neigh_ind[:, :n_neighbors] # argpartition doesn't guarantee sorted order, so we sort again neigh_ind = neigh_ind[ sample_range, np.argsort(dist[sample_range, neigh_ind])] if return_distance: if self.effective_metric_ == 'euclidean': result = np.sqrt(dist[sample_range, neigh_ind]), neigh_ind else: result = dist[sample_range, neigh_ind], neigh_ind else: result = neigh_ind elif self._fit_method in ['ball_tree', 'kd_tree']: if issparse(X): raise ValueError( "%s does not work with sparse matrices. Densify the data, " "or set algorithm='brute'" % self._fit_method) result = Parallel(n_jobs, backend='threading')( delayed(self._tree.query, check_pickle=False)( X[s], n_neighbors, return_distance) for s in gen_even_slices(X.shape[0], n_jobs) ) if return_distance: dist, neigh_ind = tuple(zip(*result)) result = np.vstack(dist), np.vstack(neigh_ind) else: result = np.vstack(result) else: raise ValueError("internal: _fit_method not recognized") if not query_is_train: return result else: # If the query data is the same as the indexed data, we would like # to ignore the first nearest neighbor of every sample, i.e # the sample itself. if return_distance: dist, neigh_ind = result else: neigh_ind = result sample_mask = neigh_ind != sample_range # Corner case: When the number of duplicates are more # than the number of neighbors, the first NN will not # be the sample, but a duplicate. # In that case mask the first duplicate. dup_gr_nbrs = np.all(sample_mask, axis=1) sample_mask[:, 0][dup_gr_nbrs] = False neigh_ind = np.reshape( neigh_ind[sample_mask], (n_samples, n_neighbors - 1)) if return_distance: dist = np.reshape( dist[sample_mask], (n_samples, n_neighbors - 1)) return dist, neigh_ind return neigh_ind def kneighbors_graph(self, X=None, n_neighbors=None, mode='connectivity'): """Computes the (weighted) graph of k-Neighbors for points in X Parameters ---------- X : array-like, shape (n_query, n_features), \ or (n_query, n_indexed) if metric == 'precomputed' The query point or points. If not provided, neighbors of each indexed point are returned. In this case, the query point is not considered its own neighbor. n_neighbors : int Number of neighbors for each sample. (default is value passed to the constructor). mode : {'connectivity', 'distance'}, optional Type of returned matrix: 'connectivity' will return the connectivity matrix with ones and zeros, in 'distance' the edges are Euclidean distance between points. Returns ------- A : sparse matrix in CSR format, shape = [n_samples, n_samples_fit] n_samples_fit is the number of samples in the fitted data A[i, j] is assigned the weight of edge that connects i to j. Examples -------- >>> X = [[0], [3], [1]] >>> from sklearn.neighbors import NearestNeighbors >>> neigh = NearestNeighbors(n_neighbors=2) >>> neigh.fit(X) # doctest: +ELLIPSIS NearestNeighbors(algorithm='auto', leaf_size=30, ...) >>> A = neigh.kneighbors_graph(X) >>> A.toarray() array([[ 1., 0., 1.], [ 0., 1., 1.], [ 1., 0., 1.]]) See also -------- NearestNeighbors.radius_neighbors_graph """ if n_neighbors is None: n_neighbors = self.n_neighbors # kneighbors does the None handling. if X is not None: X = check_array(X, accept_sparse='csr') n_samples1 = X.shape[0] else: n_samples1 = self._fit_X.shape[0] n_samples2 = self._fit_X.shape[0] n_nonzero = n_samples1 * n_neighbors A_indptr = np.arange(0, n_nonzero + 1, n_neighbors) # construct CSR matrix representation of the k-NN graph if mode == 'connectivity': A_data = np.ones(n_samples1 * n_neighbors) A_ind = self.kneighbors(X, n_neighbors, return_distance=False) elif mode == 'distance': A_data, A_ind = self.kneighbors( X, n_neighbors, return_distance=True) A_data = np.ravel(A_data) else: raise ValueError( 'Unsupported mode, must be one of "connectivity" ' 'or "distance" but got "%s" instead' % mode) kneighbors_graph = csr_matrix((A_data, A_ind.ravel(), A_indptr), shape=(n_samples1, n_samples2)) return kneighbors_graph class RadiusNeighborsMixin(object): """Mixin for radius-based neighbors searches""" def radius_neighbors(self, X=None, radius=None, return_distance=True): """Finds the neighbors within a given radius of a point or points. Return the indices and distances of each point from the dataset lying in a ball with size ``radius`` around the points of the query array. Points lying on the boundary are included in the results. The result points are *not* necessarily sorted by distance to their query point. Parameters ---------- X : array-like, (n_samples, n_features), optional The query point or points. If not provided, neighbors of each indexed point are returned. In this case, the query point is not considered its own neighbor. radius : float Limiting distance of neighbors to return. (default is the value passed to the constructor). return_distance : boolean, optional. Defaults to True. If False, distances will not be returned Returns ------- dist : array, shape (n_samples,) of arrays Array representing the distances to each point, only present if return_distance=True. The distance values are computed according to the ``metric`` constructor parameter. ind : array, shape (n_samples,) of arrays An array of arrays of indices of the approximate nearest points from the population matrix that lie within a ball of size ``radius`` around the query points. Examples -------- In the following example, we construct a NeighborsClassifier class from an array representing our data set and ask who's the closest point to [1, 1, 1]: >>> import numpy as np >>> samples = [[0., 0., 0.], [0., .5, 0.], [1., 1., .5]] >>> from sklearn.neighbors import NearestNeighbors >>> neigh = NearestNeighbors(radius=1.6) >>> neigh.fit(samples) # doctest: +ELLIPSIS NearestNeighbors(algorithm='auto', leaf_size=30, ...) >>> rng = neigh.radius_neighbors([[1., 1., 1.]]) >>> print(np.asarray(rng[0][0])) # doctest: +ELLIPSIS [ 1.5 0.5] >>> print(np.asarray(rng[1][0])) # doctest: +ELLIPSIS [1 2] The first array returned contains the distances to all points which are closer than 1.6, while the second array returned contains their indices. In general, multiple points can be queried at the same time. Notes ----- Because the number of neighbors of each point is not necessarily equal, the results for multiple query points cannot be fit in a standard data array. For efficiency, `radius_neighbors` returns arrays of objects, where each object is a 1D array of indices or distances. """ if self._fit_method is None: raise NotFittedError("Must fit neighbors before querying.") if X is not None: query_is_train = False X = check_array(X, accept_sparse='csr') else: query_is_train = True X = self._fit_X if radius is None: radius = self.radius n_samples = X.shape[0] if self._fit_method == 'brute': # for efficiency, use squared euclidean distances if self.effective_metric_ == 'euclidean': dist = pairwise_distances(X, self._fit_X, 'euclidean', squared=True) radius *= radius else: dist = pairwise_distances(X, self._fit_X, self.effective_metric_, **self.effective_metric_params_) neigh_ind_list = [np.where(d <= radius)[0] for d in dist] # See https://github.com/numpy/numpy/issues/5456 # if you want to understand why this is initialized this way. neigh_ind = np.empty(n_samples, dtype='object') neigh_ind[:] = neigh_ind_list if return_distance: dist_array = np.empty(n_samples, dtype='object') if self.effective_metric_ == 'euclidean': dist_list = [np.sqrt(d[neigh_ind[i]]) for i, d in enumerate(dist)] else: dist_list = [d[neigh_ind[i]] for i, d in enumerate(dist)] dist_array[:] = dist_list results = dist_array, neigh_ind else: results = neigh_ind elif self._fit_method in ['ball_tree', 'kd_tree']: if issparse(X): raise ValueError( "%s does not work with sparse matrices. Densify the data, " "or set algorithm='brute'" % self._fit_method) results = self._tree.query_radius(X, radius, return_distance=return_distance) if return_distance: results = results[::-1] else: raise ValueError("internal: _fit_method not recognized") if not query_is_train: return results else: # If the query data is the same as the indexed data, we would like # to ignore the first nearest neighbor of every sample, i.e # the sample itself. if return_distance: dist, neigh_ind = results else: neigh_ind = results for ind, ind_neighbor in enumerate(neigh_ind): mask = ind_neighbor != ind neigh_ind[ind] = ind_neighbor[mask] if return_distance: dist[ind] = dist[ind][mask] if return_distance: return dist, neigh_ind return neigh_ind def radius_neighbors_graph(self, X=None, radius=None, mode='connectivity'): """Computes the (weighted) graph of Neighbors for points in X Neighborhoods are restricted the points at a distance lower than radius. Parameters ---------- X : array-like, shape = [n_samples, n_features], optional The query point or points. If not provided, neighbors of each indexed point are returned. In this case, the query point is not considered its own neighbor. radius : float Radius of neighborhoods. (default is the value passed to the constructor). mode : {'connectivity', 'distance'}, optional Type of returned matrix: 'connectivity' will return the connectivity matrix with ones and zeros, in 'distance' the edges are Euclidean distance between points. Returns ------- A : sparse matrix in CSR format, shape = [n_samples, n_samples] A[i, j] is assigned the weight of edge that connects i to j. Examples -------- >>> X = [[0], [3], [1]] >>> from sklearn.neighbors import NearestNeighbors >>> neigh = NearestNeighbors(radius=1.5) >>> neigh.fit(X) # doctest: +ELLIPSIS NearestNeighbors(algorithm='auto', leaf_size=30, ...) >>> A = neigh.radius_neighbors_graph(X) >>> A.toarray() array([[ 1., 0., 1.], [ 0., 1., 0.], [ 1., 0., 1.]]) See also -------- kneighbors_graph """ if X is not None: X = check_array(X, accept_sparse=['csr', 'csc', 'coo']) n_samples2 = self._fit_X.shape[0] if radius is None: radius = self.radius # construct CSR matrix representation of the NN graph if mode == 'connectivity': A_ind = self.radius_neighbors(X, radius, return_distance=False) A_data = None elif mode == 'distance': dist, A_ind = self.radius_neighbors(X, radius, return_distance=True) A_data = np.concatenate(list(dist)) else: raise ValueError( 'Unsupported mode, must be one of "connectivity", ' 'or "distance" but got %s instead' % mode) n_samples1 = A_ind.shape[0] n_neighbors = np.array([len(a) for a in A_ind]) A_ind = np.concatenate(list(A_ind)) if A_data is None: A_data = np.ones(len(A_ind)) A_indptr = np.concatenate((np.zeros(1, dtype=int), np.cumsum(n_neighbors))) return csr_matrix((A_data, A_ind, A_indptr), shape=(n_samples1, n_samples2)) class SupervisedFloatMixin(object): def fit(self, X, y): """Fit the model using X as training data and y as target values Parameters ---------- X : {array-like, sparse matrix, BallTree, KDTree} Training data. If array or matrix, shape [n_samples, n_features], or [n_samples, n_samples] if metric='precomputed'. y : {array-like, sparse matrix} Target values, array of float values, shape = [n_samples] or [n_samples, n_outputs] """ if not isinstance(X, (KDTree, BallTree)): X, y = check_X_y(X, y, "csr", multi_output=True) self._y = y return self._fit(X) class SupervisedIntegerMixin(object): def fit(self, X, y): """Fit the model using X as training data and y as target values Parameters ---------- X : {array-like, sparse matrix, BallTree, KDTree} Training data. If array or matrix, shape [n_samples, n_features], or [n_samples, n_samples] if metric='precomputed'. y : {array-like, sparse matrix} Target values of shape = [n_samples] or [n_samples, n_outputs] """ if not isinstance(X, (KDTree, BallTree)): X, y = check_X_y(X, y, "csr", multi_output=True) if y.ndim == 1 or y.ndim == 2 and y.shape[1] == 1: if y.ndim != 1: warnings.warn("A column-vector y was passed when a 1d array " "was expected. Please change the shape of y to " "(n_samples, ), for example using ravel().", DataConversionWarning, stacklevel=2) self.outputs_2d_ = False y = y.reshape((-1, 1)) else: self.outputs_2d_ = True self.classes_ = [] self._y = np.empty(y.shape, dtype=np.int) for k in range(self._y.shape[1]): classes, self._y[:, k] = np.unique(y[:, k], return_inverse=True) self.classes_.append(classes) if not self.outputs_2d_: self.classes_ = self.classes_[0] self._y = self._y.ravel() return self._fit(X) class UnsupervisedMixin(object): def fit(self, X, y=None): """Fit the model using X as training data Parameters ---------- X : {array-like, sparse matrix, BallTree, KDTree} Training data. If array or matrix, shape [n_samples, n_features], or [n_samples, n_samples] if metric='precomputed'. """ return self._fit(X)
bsd-3-clause
sgenoud/scikit-learn
examples/plot_digits_pipe.py
5
1781
#!/usr/bin/python # -*- coding: utf-8 -*- """ ========================================================= Pipelining: chaining a PCA and a logistic regression ========================================================= The PCA does an unsupervised dimensionality reduction, while the logistic regression does the prediction. We use a GridSearchCV to set the dimensionality of the PCA """ print __doc__ # Code source: Gael Varoqueux # Modified for Documentation merge by Jaques Grobler # License: BSD import numpy as np import pylab as pl from sklearn import linear_model, decomposition, datasets logistic = linear_model.LogisticRegression() pca = decomposition.PCA() from sklearn.pipeline import Pipeline pipe = Pipeline(steps=[('pca', pca), ('logistic', logistic)]) digits = datasets.load_digits() X_digits = digits.data y_digits = digits.target ############################################################################### # Plot the PCA spectrum pca.fit(X_digits) pl.figure(1, figsize=(4, 3)) pl.clf() pl.axes([.2, .2, .7, .7]) pl.plot(pca.explained_variance_, linewidth=2) pl.axis('tight') pl.xlabel('n_components') pl.ylabel('explained_variance_') ############################################################################### # Prediction from sklearn.grid_search import GridSearchCV n_components = [20, 40, 64] Cs = np.logspace(-4, 4, 3) #Parameters of pipelines can be set using ‘__’ separated parameter names: estimator = GridSearchCV(pipe, dict(pca__n_components=n_components, logistic__C=Cs)) estimator.fit(X_digits, y_digits) pl.axvline(estimator.best_estimator_.named_steps['pca'].n_components, linestyle=':', label='n_components chosen') pl.legend(prop=dict(size=12)) pl.show()
bsd-3-clause
mlperf/training_results_v0.5
v0.5.0/google/cloud_v2.8/resnet-tpuv2-8/code/resnet/model/models/official/transformer/data_download.py
4
14804
# Copyright 2018 The TensorFlow Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """Download and preprocess WMT17 ende training and evaluation datasets.""" from __future__ import absolute_import from __future__ import division from __future__ import print_function import os import random import tarfile # pylint: disable=g-bad-import-order import six from six.moves import urllib from absl import app as absl_app from absl import flags import tensorflow as tf # pylint: enable=g-bad-import-order from official.transformer.utils import tokenizer from official.utils.flags import core as flags_core # Data sources for training/evaluating the transformer translation model. # If any of the training sources are changed, then either: # 1) use the flag `--search` to find the best min count or # 2) update the _TRAIN_DATA_MIN_COUNT constant. # min_count is the minimum number of times a token must appear in the data # before it is added to the vocabulary. "Best min count" refers to the value # that generates a vocabulary set that is closest in size to _TARGET_VOCAB_SIZE. _TRAIN_DATA_SOURCES = [ { "url": "http://data.statmt.org/wmt17/translation-task/" "training-parallel-nc-v12.tgz", "input": "news-commentary-v12.de-en.en", "target": "news-commentary-v12.de-en.de", }, { "url": "http://www.statmt.org/wmt13/training-parallel-commoncrawl.tgz", "input": "commoncrawl.de-en.en", "target": "commoncrawl.de-en.de", }, { "url": "http://www.statmt.org/wmt13/training-parallel-europarl-v7.tgz", "input": "europarl-v7.de-en.en", "target": "europarl-v7.de-en.de", }, ] # Use pre-defined minimum count to generate subtoken vocabulary. _TRAIN_DATA_MIN_COUNT = 6 _EVAL_DATA_SOURCES = [ { "url": "http://data.statmt.org/wmt17/translation-task/dev.tgz", "input": "newstest2013.en", "target": "newstest2013.de", } ] # Vocabulary constants _TARGET_VOCAB_SIZE = 32768 # Number of subtokens in the vocabulary list. _TARGET_THRESHOLD = 327 # Accept vocabulary if size is within this threshold VOCAB_FILE = "vocab.ende.%d" % _TARGET_VOCAB_SIZE # Strings to inclue in the generated files. _PREFIX = "wmt32k" _TRAIN_TAG = "train" _EVAL_TAG = "dev" # Following WMT and Tensor2Tensor conventions, in which the # evaluation datasets are tagged as "dev" for development. # Number of files to split train and evaluation data _TRAIN_SHARDS = 100 _EVAL_SHARDS = 1 def find_file(path, filename, max_depth=5): """Returns full filepath if the file is in path or a subdirectory.""" for root, dirs, files in os.walk(path): if filename in files: return os.path.join(root, filename) # Don't search past max_depth depth = root[len(path) + 1:].count(os.sep) if depth > max_depth: del dirs[:] # Clear dirs return None ############################################################################### # Download and extraction functions ############################################################################### def get_raw_files(raw_dir, data_source): """Return raw files from source. Downloads/extracts if needed. Args: raw_dir: string directory to store raw files data_source: dictionary with {"url": url of compressed dataset containing input and target files "input": file with data in input language "target": file with data in target language} Returns: dictionary with {"inputs": list of files containing data in input language "targets": list of files containing corresponding data in target language } """ raw_files = { "inputs": [], "targets": [], } # keys for d in data_source: input_file, target_file = download_and_extract( raw_dir, d["url"], d["input"], d["target"]) raw_files["inputs"].append(input_file) raw_files["targets"].append(target_file) return raw_files def download_report_hook(count, block_size, total_size): """Report hook for download progress. Args: count: current block number block_size: block size total_size: total size """ percent = int(count * block_size * 100 / total_size) print("\r%d%%" % percent + " completed", end="\r") def download_from_url(path, url): """Download content from a url. Args: path: string directory where file will be downloaded url: string url Returns: Full path to downloaded file """ filename = url.split("/")[-1] found_file = find_file(path, filename, max_depth=0) if found_file is None: filename = os.path.join(path, filename) tf.logging.info("Downloading from %s to %s." % (url, filename)) inprogress_filepath = filename + ".incomplete" inprogress_filepath, _ = urllib.request.urlretrieve( url, inprogress_filepath, reporthook=download_report_hook) # Print newline to clear the carriage return from the download progress. print() tf.gfile.Rename(inprogress_filepath, filename) return filename else: tf.logging.info("Already downloaded: %s (at %s)." % (url, found_file)) return found_file def download_and_extract(path, url, input_filename, target_filename): """Extract files from downloaded compressed archive file. Args: path: string directory where the files will be downloaded url: url containing the compressed input and target files input_filename: name of file containing data in source language target_filename: name of file containing data in target language Returns: Full paths to extracted input and target files. Raises: OSError: if the the download/extraction fails. """ # Check if extracted files already exist in path input_file = find_file(path, input_filename) target_file = find_file(path, target_filename) if input_file and target_file: tf.logging.info("Already downloaded and extracted %s." % url) return input_file, target_file # Download archive file if it doesn't already exist. compressed_file = download_from_url(path, url) # Extract compressed files tf.logging.info("Extracting %s." % compressed_file) with tarfile.open(compressed_file, "r:gz") as corpus_tar: corpus_tar.extractall(path) # Return filepaths of the requested files. input_file = find_file(path, input_filename) target_file = find_file(path, target_filename) if input_file and target_file: return input_file, target_file raise OSError("Download/extraction failed for url %s to path %s" % (url, path)) def txt_line_iterator(path): """Iterate through lines of file.""" with tf.gfile.Open(path) as f: for line in f: yield line.strip() def compile_files(raw_dir, raw_files, tag): """Compile raw files into a single file for each language. Args: raw_dir: Directory containing downloaded raw files. raw_files: Dict containing filenames of input and target data. {"inputs": list of files containing data in input language "targets": list of files containing corresponding data in target language } tag: String to append to the compiled filename. Returns: Full path of compiled input and target files. """ tf.logging.info("Compiling files with tag %s." % tag) filename = "%s-%s" % (_PREFIX, tag) input_compiled_file = os.path.join(raw_dir, filename + ".lang1") target_compiled_file = os.path.join(raw_dir, filename + ".lang2") with tf.gfile.Open(input_compiled_file, mode="w") as input_writer: with tf.gfile.Open(target_compiled_file, mode="w") as target_writer: for i in range(len(raw_files["inputs"])): input_file = raw_files["inputs"][i] target_file = raw_files["targets"][i] tf.logging.info("Reading files %s and %s." % (input_file, target_file)) write_file(input_writer, input_file) write_file(target_writer, target_file) return input_compiled_file, target_compiled_file def write_file(writer, filename): """Write all of lines from file using the writer.""" for line in txt_line_iterator(filename): writer.write(line) writer.write("\n") ############################################################################### # Data preprocessing ############################################################################### def encode_and_save_files( subtokenizer, data_dir, raw_files, tag, total_shards): """Save data from files as encoded Examples in TFrecord format. Args: subtokenizer: Subtokenizer object that will be used to encode the strings. data_dir: The directory in which to write the examples raw_files: A tuple of (input, target) data files. Each line in the input and the corresponding line in target file will be saved in a tf.Example. tag: String that will be added onto the file names. total_shards: Number of files to divide the data into. Returns: List of all files produced. """ # Create a file for each shard. filepaths = [shard_filename(data_dir, tag, n + 1, total_shards) for n in range(total_shards)] if all_exist(filepaths): tf.logging.info("Files with tag %s already exist." % tag) return filepaths tf.logging.info("Saving files with tag %s." % tag) input_file = raw_files[0] target_file = raw_files[1] # Write examples to each shard in round robin order. tmp_filepaths = [fname + ".incomplete" for fname in filepaths] writers = [tf.python_io.TFRecordWriter(fname) for fname in tmp_filepaths] counter, shard = 0, 0 for counter, (input_line, target_line) in enumerate(zip( txt_line_iterator(input_file), txt_line_iterator(target_file))): if counter > 0 and counter % 100000 == 0: tf.logging.info("\tSaving case %d." % counter) example = dict_to_example( {"inputs": subtokenizer.encode(input_line, add_eos=True), "targets": subtokenizer.encode(target_line, add_eos=True)}) writers[shard].write(example.SerializeToString()) shard = (shard + 1) % total_shards for writer in writers: writer.close() for tmp_name, final_name in zip(tmp_filepaths, filepaths): tf.gfile.Rename(tmp_name, final_name) tf.logging.info("Saved %d Examples", counter + 1) return filepaths def shard_filename(path, tag, shard_num, total_shards): """Create filename for data shard.""" return os.path.join( path, "%s-%s-%.5d-of-%.5d" % (_PREFIX, tag, shard_num, total_shards)) def shuffle_records(fname): """Shuffle records in a single file.""" tf.logging.info("Shuffling records in file %s" % fname) # Rename file prior to shuffling tmp_fname = fname + ".unshuffled" tf.gfile.Rename(fname, tmp_fname) reader = tf.python_io.tf_record_iterator(tmp_fname) records = [] for record in reader: records.append(record) if len(records) % 100000 == 0: tf.logging.info("\tRead: %d", len(records)) random.shuffle(records) # Write shuffled records to original file name with tf.python_io.TFRecordWriter(fname) as w: for count, record in enumerate(records): w.write(record) if count > 0 and count % 100000 == 0: tf.logging.info("\tWriting record: %d" % count) tf.gfile.Remove(tmp_fname) def dict_to_example(dictionary): """Converts a dictionary of string->int to a tf.Example.""" features = {} for k, v in six.iteritems(dictionary): features[k] = tf.train.Feature(int64_list=tf.train.Int64List(value=v)) return tf.train.Example(features=tf.train.Features(feature=features)) def all_exist(filepaths): """Returns true if all files in the list exist.""" for fname in filepaths: if not tf.gfile.Exists(fname): return False return True def make_dir(path): if not tf.gfile.Exists(path): tf.logging.info("Creating directory %s" % path) tf.gfile.MakeDirs(path) def main(unused_argv): """Obtain training and evaluation data for the Transformer model.""" make_dir(FLAGS.raw_dir) make_dir(FLAGS.data_dir) # Get paths of download/extracted training and evaluation files. tf.logging.info("Step 1/4: Downloading data from source") train_files = get_raw_files(FLAGS.raw_dir, _TRAIN_DATA_SOURCES) eval_files = get_raw_files(FLAGS.raw_dir, _EVAL_DATA_SOURCES) # Create subtokenizer based on the training files. tf.logging.info("Step 2/4: Creating subtokenizer and building vocabulary") train_files_flat = train_files["inputs"] + train_files["targets"] vocab_file = os.path.join(FLAGS.data_dir, VOCAB_FILE) subtokenizer = tokenizer.Subtokenizer.init_from_files( vocab_file, train_files_flat, _TARGET_VOCAB_SIZE, _TARGET_THRESHOLD, min_count=None if FLAGS.search else _TRAIN_DATA_MIN_COUNT) tf.logging.info("Step 3/4: Compiling training and evaluation data") compiled_train_files = compile_files(FLAGS.raw_dir, train_files, _TRAIN_TAG) compiled_eval_files = compile_files(FLAGS.raw_dir, eval_files, _EVAL_TAG) # Tokenize and save data as Examples in the TFRecord format. tf.logging.info("Step 4/4: Preprocessing and saving data") train_tfrecord_files = encode_and_save_files( subtokenizer, FLAGS.data_dir, compiled_train_files, _TRAIN_TAG, _TRAIN_SHARDS) encode_and_save_files( subtokenizer, FLAGS.data_dir, compiled_eval_files, _EVAL_TAG, _EVAL_SHARDS) for fname in train_tfrecord_files: shuffle_records(fname) def define_data_download_flags(): """Add flags specifying data download arguments.""" flags.DEFINE_string( name="data_dir", short_name="dd", default="/tmp/translate_ende", help=flags_core.help_wrap( "Directory for where the translate_ende_wmt32k dataset is saved.")) flags.DEFINE_string( name="raw_dir", short_name="rd", default="/tmp/translate_ende_raw", help=flags_core.help_wrap( "Path where the raw data will be downloaded and extracted.")) flags.DEFINE_bool( name="search", default=False, help=flags_core.help_wrap( "If set, use binary search to find the vocabulary set with size" "closest to the target size (%d)." % _TARGET_VOCAB_SIZE)) if __name__ == "__main__": tf.logging.set_verbosity(tf.logging.INFO) define_data_download_flags() FLAGS = flags.FLAGS absl_app.run(main)
apache-2.0
mlflow/mlflow
tests/utils/test_requirements_utils.py
1
13714
import os import importlib from unittest import mock import importlib_metadata import pytest import mlflow import mlflow.utils.requirements_utils from mlflow.utils.requirements_utils import ( _is_comment, _is_empty, _is_requirements_file, _strip_inline_comment, _join_continued_lines, _parse_requirements, _prune_packages, _strip_local_version_label, _get_installed_version, _get_pinned_requirement, _infer_requirements, _normalize_package_name, _PyPIPackageIndex, ) def test_is_comment(): assert _is_comment("# comment") assert _is_comment("#") assert _is_comment("### comment ###") assert not _is_comment("comment") assert not _is_comment("") def test_is_empty(): assert _is_empty("") assert not _is_empty(" ") assert not _is_empty("a") def test_is_requirements_file(): assert _is_requirements_file("-r req.txt") assert _is_requirements_file("-r req.txt") assert _is_requirements_file("--requirement req.txt") assert _is_requirements_file("--requirement req.txt") assert not _is_requirements_file("req") def test_strip_inline_comment(): assert _strip_inline_comment("aaa # comment") == "aaa" assert _strip_inline_comment("aaa # comment") == "aaa" assert _strip_inline_comment("aaa # comment") == "aaa" assert _strip_inline_comment("aaa # com1 # com2") == "aaa" # Ensure a URI fragment is not stripped assert ( _strip_inline_comment("git+https://git/repo.git#subdirectory=subdir") == "git+https://git/repo.git#subdirectory=subdir" ) def test_join_continued_lines(): assert list(_join_continued_lines(["a"])) == ["a"] assert list(_join_continued_lines(["a\\", "b"])) == ["ab"] assert list(_join_continued_lines(["a\\", "b\\", "c"])) == ["abc"] assert list(_join_continued_lines(["a\\", " b"])) == ["a b"] assert list(_join_continued_lines(["a\\", " b\\", " c"])) == ["a b c"] assert list(_join_continued_lines(["a\\", "\\", "b"])) == ["ab"] assert list(_join_continued_lines(["a\\", "b", "c\\", "d"])) == ["ab", "cd"] assert list(_join_continued_lines(["a\\", "", "b"])) == ["a", "b"] assert list(_join_continued_lines(["a\\"])) == ["a"] assert list(_join_continued_lines(["\\", "a"])) == ["a"] def test_parse_requirements(request, tmpdir): """ Ensures `_parse_requirements` returns the same result as `pip._internal.req.parse_requirements` """ from pip._internal.req import parse_requirements as pip_parse_requirements from pip._internal.network.session import PipSession root_req_src = """ # No version specifier noverspec no-ver-spec # Version specifiers verspec<1.0 ver-spec == 2.0 # Environment marker env-marker; python_version < "3.8" inline-comm # Inline comment inlinecomm # Inline comment # Git URIs git+https://github.com/git/uri git+https://github.com/sub/dir#subdirectory=subdir # Requirements files -r {relative_req} --requirement {absolute_req} # Constraints files -c {relative_con} --constraint {absolute_con} # Line continuation line-cont\ ==\ 1.0 # Line continuation with spaces line-cont-space \ == \ 1.0 # Line continuation with a blank line line-cont-blank\ # Line continuation at EOF line-cont-eof\ """.strip() try: os.chdir(tmpdir) root_req = tmpdir.join("requirements.txt") # Requirements files rel_req = tmpdir.join("relative_req.txt") abs_req = tmpdir.join("absolute_req.txt") # Constraints files rel_con = tmpdir.join("relative_con.txt") abs_con = tmpdir.join("absolute_con.txt") # pip's requirements parser collapses an absolute requirements file path: # https://github.com/pypa/pip/issues/10121 # As a workaround, use a relative path on Windows. absolute_req = abs_req.basename if os.name == "nt" else abs_req.strpath absolute_con = abs_con.basename if os.name == "nt" else abs_con.strpath root_req.write( root_req_src.format( relative_req=rel_req.basename, absolute_req=absolute_req, relative_con=rel_con.basename, absolute_con=absolute_con, ) ) rel_req.write("rel-req-xxx\nrel-req-yyy") abs_req.write("abs-req-zzz") rel_con.write("rel-con-xxx\nrel-con-yyy") abs_con.write("abs-con-zzz") expected_cons = [ "rel-con-xxx", "rel-con-yyy", "abs-con-zzz", ] expected_reqs = [ "noverspec", "no-ver-spec", "verspec<1.0", "ver-spec == 2.0", 'env-marker; python_version < "3.8"', "inline-comm", "inlinecomm", "git+https://github.com/git/uri", "git+https://github.com/sub/dir#subdirectory=subdir", "rel-req-xxx", "rel-req-yyy", "abs-req-zzz", "line-cont==1.0", "line-cont-space == 1.0", "line-cont-blank", "line-cont-eof", ] parsed_reqs = list(_parse_requirements(root_req.basename, is_constraint=False)) pip_reqs = list(pip_parse_requirements(root_req.basename, session=PipSession())) # Requirements assert [r.req_str for r in parsed_reqs if not r.is_constraint] == expected_reqs assert [r.requirement for r in pip_reqs if not r.constraint] == expected_reqs # Constraints assert [r.req_str for r in parsed_reqs if r.is_constraint] == expected_cons assert [r.requirement for r in pip_reqs if r.constraint] == expected_cons finally: os.chdir(request.config.invocation_dir) def test_normalize_package_name(): assert _normalize_package_name("abc") == "abc" assert _normalize_package_name("ABC") == "abc" assert _normalize_package_name("a-b-c") == "a-b-c" assert _normalize_package_name("a.b.c") == "a-b-c" assert _normalize_package_name("a_b_c") == "a-b-c" assert _normalize_package_name("a--b--c") == "a-b-c" assert _normalize_package_name("a-._b-._c") == "a-b-c" def test_prune_packages(): assert _prune_packages(["mlflow"]) == {"mlflow"} assert _prune_packages(["mlflow", "packaging"]) == {"mlflow"} assert _prune_packages(["mlflow", "scikit-learn"]) == {"mlflow"} def test_capture_imported_modules(): from mlflow.utils._capture_modules import _CaptureImportedModules with _CaptureImportedModules() as cap: # pylint: disable=unused-import import math __import__("pandas") importlib.import_module("numpy") assert "math" in cap.imported_modules assert "pandas" in cap.imported_modules assert "numpy" in cap.imported_modules def test_strip_local_version_label(): assert _strip_local_version_label("1.2.3") == "1.2.3" assert _strip_local_version_label("1.2.3+ab") == "1.2.3" assert _strip_local_version_label("1.2.3rc0+ab") == "1.2.3rc0" assert _strip_local_version_label("1.2.3.dev0+ab") == "1.2.3.dev0" assert _strip_local_version_label("1.2.3.post0+ab") == "1.2.3.post0" assert _strip_local_version_label("invalid") == "invalid" def test_get_installed_version(tmpdir, monkeypatch): import numpy as np import pandas as pd import sklearn assert _get_installed_version("mlflow") == mlflow.__version__ assert _get_installed_version("numpy") == np.__version__ assert _get_installed_version("pandas") == pd.__version__ assert _get_installed_version("scikit-learn", module="sklearn") == sklearn.__version__ not_found_package = tmpdir.join("not_found.py") not_found_package.write("__version__ = '1.2.3'") monkeypatch.syspath_prepend(tmpdir.strpath) with pytest.raises(importlib_metadata.PackageNotFoundError, match=r".+"): importlib_metadata.version("not_found") assert _get_installed_version("not_found") == "1.2.3" def test_get_pinned_requirement(tmpdir, monkeypatch): assert _get_pinned_requirement("mlflow") == f"mlflow=={mlflow.__version__}" assert _get_pinned_requirement("mlflow", version="1.2.3") == "mlflow==1.2.3" not_found_package = tmpdir.join("not_found.py") not_found_package.write("__version__ = '1.2.3'") monkeypatch.syspath_prepend(tmpdir.strpath) with pytest.raises(importlib_metadata.PackageNotFoundError, match=r".+"): importlib_metadata.version("not_found") assert _get_pinned_requirement("not_found") == "not_found==1.2.3" def test_get_pinned_requirement_local_version_label(tmpdir, monkeypatch): package = tmpdir.join("my_package.py") lvl = "abc.def.ghi" # Local version label package.write(f"__version__ = '1.2.3+{lvl}'") monkeypatch.syspath_prepend(tmpdir.strpath) with mock.patch("mlflow.utils.requirements_utils._logger.warning") as mock_warning: req = _get_pinned_requirement("my_package") mock_warning.assert_called_once() (first_pos_arg,) = mock_warning.call_args[0] assert first_pos_arg.startswith( f"Found my_package version (1.2.3+{lvl}) contains a local version label (+{lvl})." ) assert req == "my_package==1.2.3" def test_infer_requirements_excludes_mlflow(): with mock.patch( "mlflow.utils.requirements_utils._capture_imported_modules", return_value=["mlflow", "pytest"], ): mlflow_package = "mlflow-skinny" if "MLFLOW_SKINNY" in os.environ else "mlflow" assert mlflow_package in importlib_metadata.packages_distributions()["mlflow"] assert _infer_requirements("path/to/model", "sklearn") == [f"pytest=={pytest.__version__}"] def test_infer_requirements_prints_warning_for_unrecognized_packages(): with mock.patch( "mlflow.utils.requirements_utils._capture_imported_modules", return_value=["sklearn"], ), mock.patch( "mlflow.utils.requirements_utils._PYPI_PACKAGE_INDEX", _PyPIPackageIndex(date="2022-01-01", package_names=set()), ), mock.patch( "mlflow.utils.requirements_utils._logger.warning" ) as mock_warning: _infer_requirements("path/to/model", "sklearn") mock_warning.assert_called_once() warning_template = mock_warning.call_args[0][0] date, unrecognized_packages = mock_warning.call_args[0][1:3] warning_text = warning_template % (date, unrecognized_packages) assert "not found in the public PyPI package index" in warning_text assert "scikit-learn" in warning_text def test_infer_requirements_does_not_print_warning_for_recognized_packages(): with mock.patch( "mlflow.utils.requirements_utils._capture_imported_modules", return_value=["sklearn"], ), mock.patch( "mlflow.utils.requirements_utils._PYPI_PACKAGE_INDEX", _PyPIPackageIndex(date="2022-01-01", package_names={"scikit-learn"}), ), mock.patch( "mlflow.utils.requirements_utils._logger.warning" ) as mock_warning: _infer_requirements("path/to/model", "sklearn") mock_warning.assert_not_called() def test_capture_imported_modules_scopes_databricks_imports(monkeypatch, tmpdir): from mlflow.utils._capture_modules import _CaptureImportedModules monkeypatch.chdir(tmpdir) monkeypatch.syspath_prepend(str(tmpdir)) databricks_dir = os.path.join(tmpdir, "databricks") os.makedirs(databricks_dir) for file_name in [ "__init__.py", "automl.py", "automl_runtime.py", "automl_foo.py", "model_monitoring.py", "other.py", ]: with open(os.path.join(databricks_dir, file_name), "w"): pass with _CaptureImportedModules() as cap: # pylint: disable=unused-import import databricks import databricks.automl import databricks.automl_foo import databricks.automl_runtime import databricks.model_monitoring assert "databricks.automl" in cap.imported_modules assert "databricks.model_monitoring" in cap.imported_modules assert "databricks" not in cap.imported_modules assert "databricks.automl_foo" not in cap.imported_modules with _CaptureImportedModules() as cap: # pylint: disable=unused-import import databricks.automl import databricks.automl_foo import databricks.automl_runtime import databricks.model_monitoring import databricks.other assert "databricks.automl" in cap.imported_modules assert "databricks.model_monitoring" in cap.imported_modules assert "databricks" in cap.imported_modules assert "databricks.automl_foo" not in cap.imported_modules def test_infer_pip_requirements_scopes_databricks_imports(): mlflow.utils.requirements_utils._MODULES_TO_PACKAGES = None mlflow.utils.requirements_utils._PACKAGES_TO_MODULES = None with mock.patch( "mlflow.utils.requirements_utils._capture_imported_modules", return_value=[ "databricks.automl", "databricks.model_monitoring", "databricks.automl_runtime", ], ), mock.patch( "mlflow.utils.requirements_utils._get_installed_version", return_value="1.0", ), mock.patch( "importlib_metadata.packages_distributions", return_value={ "databricks": ["databricks-automl-runtime", "databricks-model-monitoring", "koalas"], }, ): assert _infer_requirements("path/to/model", "sklearn") == [ "databricks-automl-runtime==1.0", "databricks-model-monitoring==1.0", ] assert mlflow.utils.requirements_utils._MODULES_TO_PACKAGES["databricks"] == ["koalas"]
apache-2.0
jzt5132/scikit-learn
sklearn/feature_selection/tests/test_feature_select.py
102
22297
""" Todo: cross-check the F-value with stats model """ from __future__ import division import itertools import warnings import numpy as np from scipy import stats, sparse from sklearn.utils.testing import assert_equal from sklearn.utils.testing import assert_almost_equal from sklearn.utils.testing import assert_raises from sklearn.utils.testing import assert_true from sklearn.utils.testing import assert_array_equal from sklearn.utils.testing import assert_array_almost_equal from sklearn.utils.testing import assert_not_in from sklearn.utils.testing import assert_less from sklearn.utils.testing import assert_warns from sklearn.utils.testing import ignore_warnings from sklearn.utils.testing import assert_warns_message from sklearn.utils.testing import assert_greater from sklearn.utils.testing import assert_greater_equal from sklearn.utils import safe_mask from sklearn.datasets.samples_generator import (make_classification, make_regression) from sklearn.feature_selection import (chi2, f_classif, f_oneway, f_regression, SelectPercentile, SelectKBest, SelectFpr, SelectFdr, SelectFwe, GenericUnivariateSelect) ############################################################################## # Test the score functions def test_f_oneway_vs_scipy_stats(): # Test that our f_oneway gives the same result as scipy.stats rng = np.random.RandomState(0) X1 = rng.randn(10, 3) X2 = 1 + rng.randn(10, 3) f, pv = stats.f_oneway(X1, X2) f2, pv2 = f_oneway(X1, X2) assert_true(np.allclose(f, f2)) assert_true(np.allclose(pv, pv2)) def test_f_oneway_ints(): # Smoke test f_oneway on integers: that it does raise casting errors # with recent numpys rng = np.random.RandomState(0) X = rng.randint(10, size=(10, 10)) y = np.arange(10) fint, pint = f_oneway(X, y) # test that is gives the same result as with float f, p = f_oneway(X.astype(np.float), y) assert_array_almost_equal(f, fint, decimal=4) assert_array_almost_equal(p, pint, decimal=4) def test_f_classif(): # Test whether the F test yields meaningful results # on a simple simulated classification problem X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) F, pv = f_classif(X, y) F_sparse, pv_sparse = f_classif(sparse.csr_matrix(X), y) assert_true((F > 0).all()) assert_true((pv > 0).all()) assert_true((pv < 1).all()) assert_true((pv[:5] < 0.05).all()) assert_true((pv[5:] > 1.e-4).all()) assert_array_almost_equal(F_sparse, F) assert_array_almost_equal(pv_sparse, pv) def test_f_regression(): # Test whether the F test yields meaningful results # on a simple simulated regression problem X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0) F, pv = f_regression(X, y) assert_true((F > 0).all()) assert_true((pv > 0).all()) assert_true((pv < 1).all()) assert_true((pv[:5] < 0.05).all()) assert_true((pv[5:] > 1.e-4).all()) # again without centering, compare with sparse F, pv = f_regression(X, y, center=False) F_sparse, pv_sparse = f_regression(sparse.csr_matrix(X), y, center=False) assert_array_almost_equal(F_sparse, F) assert_array_almost_equal(pv_sparse, pv) def test_f_regression_input_dtype(): # Test whether f_regression returns the same value # for any numeric data_type rng = np.random.RandomState(0) X = rng.rand(10, 20) y = np.arange(10).astype(np.int) F1, pv1 = f_regression(X, y) F2, pv2 = f_regression(X, y.astype(np.float)) assert_array_almost_equal(F1, F2, 5) assert_array_almost_equal(pv1, pv2, 5) def test_f_regression_center(): # Test whether f_regression preserves dof according to 'center' argument # We use two centered variates so we have a simple relationship between # F-score with variates centering and F-score without variates centering. # Create toy example X = np.arange(-5, 6).reshape(-1, 1) # X has zero mean n_samples = X.size Y = np.ones(n_samples) Y[::2] *= -1. Y[0] = 0. # have Y mean being null F1, _ = f_regression(X, Y, center=True) F2, _ = f_regression(X, Y, center=False) assert_array_almost_equal(F1 * (n_samples - 1.) / (n_samples - 2.), F2) assert_almost_equal(F2[0], 0.232558139) # value from statsmodels OLS def test_f_classif_multi_class(): # Test whether the F test yields meaningful results # on a simple simulated classification problem X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) F, pv = f_classif(X, y) assert_true((F > 0).all()) assert_true((pv > 0).all()) assert_true((pv < 1).all()) assert_true((pv[:5] < 0.05).all()) assert_true((pv[5:] > 1.e-4).all()) def test_select_percentile_classif(): # Test whether the relative univariate feature selection # gets the correct items in a simple classification problem # with the percentile heuristic X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) univariate_filter = SelectPercentile(f_classif, percentile=25) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect(f_classif, mode='percentile', param=25).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) def test_select_percentile_classif_sparse(): # Test whether the relative univariate feature selection # gets the correct items in a simple classification problem # with the percentile heuristic X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) X = sparse.csr_matrix(X) univariate_filter = SelectPercentile(f_classif, percentile=25) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect(f_classif, mode='percentile', param=25).fit(X, y).transform(X) assert_array_equal(X_r.toarray(), X_r2.toarray()) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) X_r2inv = univariate_filter.inverse_transform(X_r2) assert_true(sparse.issparse(X_r2inv)) support_mask = safe_mask(X_r2inv, support) assert_equal(X_r2inv.shape, X.shape) assert_array_equal(X_r2inv[:, support_mask].toarray(), X_r.toarray()) # Check other columns are empty assert_equal(X_r2inv.getnnz(), X_r.getnnz()) ############################################################################## # Test univariate selection in classification settings def test_select_kbest_classif(): # Test whether the relative univariate feature selection # gets the correct items in a simple classification problem # with the k best heuristic X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) univariate_filter = SelectKBest(f_classif, k=5) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect( f_classif, mode='k_best', param=5).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) def test_select_kbest_all(): # Test whether k="all" correctly returns all features. X, y = make_classification(n_samples=20, n_features=10, shuffle=False, random_state=0) univariate_filter = SelectKBest(f_classif, k='all') X_r = univariate_filter.fit(X, y).transform(X) assert_array_equal(X, X_r) def test_select_kbest_zero(): # Test whether k=0 correctly returns no features. X, y = make_classification(n_samples=20, n_features=10, shuffle=False, random_state=0) univariate_filter = SelectKBest(f_classif, k=0) univariate_filter.fit(X, y) support = univariate_filter.get_support() gtruth = np.zeros(10, dtype=bool) assert_array_equal(support, gtruth) X_selected = assert_warns_message(UserWarning, 'No features were selected', univariate_filter.transform, X) assert_equal(X_selected.shape, (20, 0)) def test_select_heuristics_classif(): # Test whether the relative univariate feature selection # gets the correct items in a simple classification problem # with the fdr, fwe and fpr heuristics X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) univariate_filter = SelectFwe(f_classif, alpha=0.01) X_r = univariate_filter.fit(X, y).transform(X) gtruth = np.zeros(20) gtruth[:5] = 1 for mode in ['fdr', 'fpr', 'fwe']: X_r2 = GenericUnivariateSelect( f_classif, mode=mode, param=0.01).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() assert_array_almost_equal(support, gtruth) ############################################################################## # Test univariate selection in regression settings def assert_best_scores_kept(score_filter): scores = score_filter.scores_ support = score_filter.get_support() assert_array_equal(np.sort(scores[support]), np.sort(scores)[-support.sum():]) def test_select_percentile_regression(): # Test whether the relative univariate feature selection # gets the correct items in a simple regression problem # with the percentile heuristic X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0) univariate_filter = SelectPercentile(f_regression, percentile=25) X_r = univariate_filter.fit(X, y).transform(X) assert_best_scores_kept(univariate_filter) X_r2 = GenericUnivariateSelect( f_regression, mode='percentile', param=25).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) X_2 = X.copy() X_2[:, np.logical_not(support)] = 0 assert_array_equal(X_2, univariate_filter.inverse_transform(X_r)) # Check inverse_transform respects dtype assert_array_equal(X_2.astype(bool), univariate_filter.inverse_transform(X_r.astype(bool))) def test_select_percentile_regression_full(): # Test whether the relative univariate feature selection # selects all features when '100%' is asked. X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0) univariate_filter = SelectPercentile(f_regression, percentile=100) X_r = univariate_filter.fit(X, y).transform(X) assert_best_scores_kept(univariate_filter) X_r2 = GenericUnivariateSelect( f_regression, mode='percentile', param=100).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.ones(20) assert_array_equal(support, gtruth) def test_invalid_percentile(): X, y = make_regression(n_samples=10, n_features=20, n_informative=2, shuffle=False, random_state=0) assert_raises(ValueError, SelectPercentile(percentile=-1).fit, X, y) assert_raises(ValueError, SelectPercentile(percentile=101).fit, X, y) assert_raises(ValueError, GenericUnivariateSelect(mode='percentile', param=-1).fit, X, y) assert_raises(ValueError, GenericUnivariateSelect(mode='percentile', param=101).fit, X, y) def test_select_kbest_regression(): # Test whether the relative univariate feature selection # gets the correct items in a simple regression problem # with the k best heuristic X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0, noise=10) univariate_filter = SelectKBest(f_regression, k=5) X_r = univariate_filter.fit(X, y).transform(X) assert_best_scores_kept(univariate_filter) X_r2 = GenericUnivariateSelect( f_regression, mode='k_best', param=5).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) def test_select_heuristics_regression(): # Test whether the relative univariate feature selection # gets the correct items in a simple regression problem # with the fpr, fdr or fwe heuristics X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0, noise=10) univariate_filter = SelectFpr(f_regression, alpha=0.01) X_r = univariate_filter.fit(X, y).transform(X) gtruth = np.zeros(20) gtruth[:5] = 1 for mode in ['fdr', 'fpr', 'fwe']: X_r2 = GenericUnivariateSelect( f_regression, mode=mode, param=0.01).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() assert_array_equal(support[:5], np.ones((5, ), dtype=np.bool)) assert_less(np.sum(support[5:] == 1), 3) def test_select_fdr_regression(): # Test that fdr heuristic actually has low FDR. def single_fdr(alpha, n_informative, random_state): X, y = make_regression(n_samples=150, n_features=20, n_informative=n_informative, shuffle=False, random_state=random_state, noise=10) with warnings.catch_warnings(record=True): # Warnings can be raised when no features are selected # (low alpha or very noisy data) univariate_filter = SelectFdr(f_regression, alpha=alpha) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect( f_regression, mode='fdr', param=alpha).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() num_false_positives = np.sum(support[n_informative:] == 1) num_true_positives = np.sum(support[:n_informative] == 1) if num_false_positives == 0: return 0. false_discovery_rate = (num_false_positives / (num_true_positives + num_false_positives)) return false_discovery_rate for alpha in [0.001, 0.01, 0.1]: for n_informative in [1, 5, 10]: # As per Benjamini-Hochberg, the expected false discovery rate # should be lower than alpha: # FDR = E(FP / (TP + FP)) <= alpha false_discovery_rate = np.mean([single_fdr(alpha, n_informative, random_state) for random_state in range(30)]) assert_greater_equal(alpha, false_discovery_rate) # Make sure that the empirical false discovery rate increases # with alpha: if false_discovery_rate != 0: assert_greater(false_discovery_rate, alpha / 10) def test_select_fwe_regression(): # Test whether the relative univariate feature selection # gets the correct items in a simple regression problem # with the fwe heuristic X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0) univariate_filter = SelectFwe(f_regression, alpha=0.01) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect( f_regression, mode='fwe', param=0.01).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support[:5], np.ones((5, ), dtype=np.bool)) assert_less(np.sum(support[5:] == 1), 2) def test_selectkbest_tiebreaking(): # Test whether SelectKBest actually selects k features in case of ties. # Prior to 0.11, SelectKBest would return more features than requested. Xs = [[0, 1, 1], [0, 0, 1], [1, 0, 0], [1, 1, 0]] y = [1] dummy_score = lambda X, y: (X[0], X[0]) for X in Xs: sel = SelectKBest(dummy_score, k=1) X1 = ignore_warnings(sel.fit_transform)([X], y) assert_equal(X1.shape[1], 1) assert_best_scores_kept(sel) sel = SelectKBest(dummy_score, k=2) X2 = ignore_warnings(sel.fit_transform)([X], y) assert_equal(X2.shape[1], 2) assert_best_scores_kept(sel) def test_selectpercentile_tiebreaking(): # Test if SelectPercentile selects the right n_features in case of ties. Xs = [[0, 1, 1], [0, 0, 1], [1, 0, 0], [1, 1, 0]] y = [1] dummy_score = lambda X, y: (X[0], X[0]) for X in Xs: sel = SelectPercentile(dummy_score, percentile=34) X1 = ignore_warnings(sel.fit_transform)([X], y) assert_equal(X1.shape[1], 1) assert_best_scores_kept(sel) sel = SelectPercentile(dummy_score, percentile=67) X2 = ignore_warnings(sel.fit_transform)([X], y) assert_equal(X2.shape[1], 2) assert_best_scores_kept(sel) def test_tied_pvalues(): # Test whether k-best and percentiles work with tied pvalues from chi2. # chi2 will return the same p-values for the following features, but it # will return different scores. X0 = np.array([[10000, 9999, 9998], [1, 1, 1]]) y = [0, 1] for perm in itertools.permutations((0, 1, 2)): X = X0[:, perm] Xt = SelectKBest(chi2, k=2).fit_transform(X, y) assert_equal(Xt.shape, (2, 2)) assert_not_in(9998, Xt) Xt = SelectPercentile(chi2, percentile=67).fit_transform(X, y) assert_equal(Xt.shape, (2, 2)) assert_not_in(9998, Xt) def test_tied_scores(): # Test for stable sorting in k-best with tied scores. X_train = np.array([[0, 0, 0], [1, 1, 1]]) y_train = [0, 1] for n_features in [1, 2, 3]: sel = SelectKBest(chi2, k=n_features).fit(X_train, y_train) X_test = sel.transform([[0, 1, 2]]) assert_array_equal(X_test[0], np.arange(3)[-n_features:]) def test_nans(): # Assert that SelectKBest and SelectPercentile can handle NaNs. # First feature has zero variance to confuse f_classif (ANOVA) and # make it return a NaN. X = [[0, 1, 0], [0, -1, -1], [0, .5, .5]] y = [1, 0, 1] for select in (SelectKBest(f_classif, 2), SelectPercentile(f_classif, percentile=67)): ignore_warnings(select.fit)(X, y) assert_array_equal(select.get_support(indices=True), np.array([1, 2])) def test_score_func_error(): X = [[0, 1, 0], [0, -1, -1], [0, .5, .5]] y = [1, 0, 1] for SelectFeatures in [SelectKBest, SelectPercentile, SelectFwe, SelectFdr, SelectFpr, GenericUnivariateSelect]: assert_raises(TypeError, SelectFeatures(score_func=10).fit, X, y) def test_invalid_k(): X = [[0, 1, 0], [0, -1, -1], [0, .5, .5]] y = [1, 0, 1] assert_raises(ValueError, SelectKBest(k=-1).fit, X, y) assert_raises(ValueError, SelectKBest(k=4).fit, X, y) assert_raises(ValueError, GenericUnivariateSelect(mode='k_best', param=-1).fit, X, y) assert_raises(ValueError, GenericUnivariateSelect(mode='k_best', param=4).fit, X, y) def test_f_classif_constant_feature(): # Test that f_classif warns if a feature is constant throughout. X, y = make_classification(n_samples=10, n_features=5) X[:, 0] = 2.0 assert_warns(UserWarning, f_classif, X, y) def test_no_feature_selected(): rng = np.random.RandomState(0) # Generate random uncorrelated data: a strict univariate test should # rejects all the features X = rng.rand(40, 10) y = rng.randint(0, 4, size=40) strict_selectors = [ SelectFwe(alpha=0.01).fit(X, y), SelectFdr(alpha=0.01).fit(X, y), SelectFpr(alpha=0.01).fit(X, y), SelectPercentile(percentile=0).fit(X, y), SelectKBest(k=0).fit(X, y), ] for selector in strict_selectors: assert_array_equal(selector.get_support(), np.zeros(10)) X_selected = assert_warns_message( UserWarning, 'No features were selected', selector.transform, X) assert_equal(X_selected.shape, (40, 0))
bsd-3-clause
luo66/scikit-learn
sklearn/feature_selection/tests/test_feature_select.py
102
22297
""" Todo: cross-check the F-value with stats model """ from __future__ import division import itertools import warnings import numpy as np from scipy import stats, sparse from sklearn.utils.testing import assert_equal from sklearn.utils.testing import assert_almost_equal from sklearn.utils.testing import assert_raises from sklearn.utils.testing import assert_true from sklearn.utils.testing import assert_array_equal from sklearn.utils.testing import assert_array_almost_equal from sklearn.utils.testing import assert_not_in from sklearn.utils.testing import assert_less from sklearn.utils.testing import assert_warns from sklearn.utils.testing import ignore_warnings from sklearn.utils.testing import assert_warns_message from sklearn.utils.testing import assert_greater from sklearn.utils.testing import assert_greater_equal from sklearn.utils import safe_mask from sklearn.datasets.samples_generator import (make_classification, make_regression) from sklearn.feature_selection import (chi2, f_classif, f_oneway, f_regression, SelectPercentile, SelectKBest, SelectFpr, SelectFdr, SelectFwe, GenericUnivariateSelect) ############################################################################## # Test the score functions def test_f_oneway_vs_scipy_stats(): # Test that our f_oneway gives the same result as scipy.stats rng = np.random.RandomState(0) X1 = rng.randn(10, 3) X2 = 1 + rng.randn(10, 3) f, pv = stats.f_oneway(X1, X2) f2, pv2 = f_oneway(X1, X2) assert_true(np.allclose(f, f2)) assert_true(np.allclose(pv, pv2)) def test_f_oneway_ints(): # Smoke test f_oneway on integers: that it does raise casting errors # with recent numpys rng = np.random.RandomState(0) X = rng.randint(10, size=(10, 10)) y = np.arange(10) fint, pint = f_oneway(X, y) # test that is gives the same result as with float f, p = f_oneway(X.astype(np.float), y) assert_array_almost_equal(f, fint, decimal=4) assert_array_almost_equal(p, pint, decimal=4) def test_f_classif(): # Test whether the F test yields meaningful results # on a simple simulated classification problem X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) F, pv = f_classif(X, y) F_sparse, pv_sparse = f_classif(sparse.csr_matrix(X), y) assert_true((F > 0).all()) assert_true((pv > 0).all()) assert_true((pv < 1).all()) assert_true((pv[:5] < 0.05).all()) assert_true((pv[5:] > 1.e-4).all()) assert_array_almost_equal(F_sparse, F) assert_array_almost_equal(pv_sparse, pv) def test_f_regression(): # Test whether the F test yields meaningful results # on a simple simulated regression problem X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0) F, pv = f_regression(X, y) assert_true((F > 0).all()) assert_true((pv > 0).all()) assert_true((pv < 1).all()) assert_true((pv[:5] < 0.05).all()) assert_true((pv[5:] > 1.e-4).all()) # again without centering, compare with sparse F, pv = f_regression(X, y, center=False) F_sparse, pv_sparse = f_regression(sparse.csr_matrix(X), y, center=False) assert_array_almost_equal(F_sparse, F) assert_array_almost_equal(pv_sparse, pv) def test_f_regression_input_dtype(): # Test whether f_regression returns the same value # for any numeric data_type rng = np.random.RandomState(0) X = rng.rand(10, 20) y = np.arange(10).astype(np.int) F1, pv1 = f_regression(X, y) F2, pv2 = f_regression(X, y.astype(np.float)) assert_array_almost_equal(F1, F2, 5) assert_array_almost_equal(pv1, pv2, 5) def test_f_regression_center(): # Test whether f_regression preserves dof according to 'center' argument # We use two centered variates so we have a simple relationship between # F-score with variates centering and F-score without variates centering. # Create toy example X = np.arange(-5, 6).reshape(-1, 1) # X has zero mean n_samples = X.size Y = np.ones(n_samples) Y[::2] *= -1. Y[0] = 0. # have Y mean being null F1, _ = f_regression(X, Y, center=True) F2, _ = f_regression(X, Y, center=False) assert_array_almost_equal(F1 * (n_samples - 1.) / (n_samples - 2.), F2) assert_almost_equal(F2[0], 0.232558139) # value from statsmodels OLS def test_f_classif_multi_class(): # Test whether the F test yields meaningful results # on a simple simulated classification problem X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) F, pv = f_classif(X, y) assert_true((F > 0).all()) assert_true((pv > 0).all()) assert_true((pv < 1).all()) assert_true((pv[:5] < 0.05).all()) assert_true((pv[5:] > 1.e-4).all()) def test_select_percentile_classif(): # Test whether the relative univariate feature selection # gets the correct items in a simple classification problem # with the percentile heuristic X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) univariate_filter = SelectPercentile(f_classif, percentile=25) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect(f_classif, mode='percentile', param=25).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) def test_select_percentile_classif_sparse(): # Test whether the relative univariate feature selection # gets the correct items in a simple classification problem # with the percentile heuristic X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) X = sparse.csr_matrix(X) univariate_filter = SelectPercentile(f_classif, percentile=25) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect(f_classif, mode='percentile', param=25).fit(X, y).transform(X) assert_array_equal(X_r.toarray(), X_r2.toarray()) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) X_r2inv = univariate_filter.inverse_transform(X_r2) assert_true(sparse.issparse(X_r2inv)) support_mask = safe_mask(X_r2inv, support) assert_equal(X_r2inv.shape, X.shape) assert_array_equal(X_r2inv[:, support_mask].toarray(), X_r.toarray()) # Check other columns are empty assert_equal(X_r2inv.getnnz(), X_r.getnnz()) ############################################################################## # Test univariate selection in classification settings def test_select_kbest_classif(): # Test whether the relative univariate feature selection # gets the correct items in a simple classification problem # with the k best heuristic X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) univariate_filter = SelectKBest(f_classif, k=5) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect( f_classif, mode='k_best', param=5).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) def test_select_kbest_all(): # Test whether k="all" correctly returns all features. X, y = make_classification(n_samples=20, n_features=10, shuffle=False, random_state=0) univariate_filter = SelectKBest(f_classif, k='all') X_r = univariate_filter.fit(X, y).transform(X) assert_array_equal(X, X_r) def test_select_kbest_zero(): # Test whether k=0 correctly returns no features. X, y = make_classification(n_samples=20, n_features=10, shuffle=False, random_state=0) univariate_filter = SelectKBest(f_classif, k=0) univariate_filter.fit(X, y) support = univariate_filter.get_support() gtruth = np.zeros(10, dtype=bool) assert_array_equal(support, gtruth) X_selected = assert_warns_message(UserWarning, 'No features were selected', univariate_filter.transform, X) assert_equal(X_selected.shape, (20, 0)) def test_select_heuristics_classif(): # Test whether the relative univariate feature selection # gets the correct items in a simple classification problem # with the fdr, fwe and fpr heuristics X, y = make_classification(n_samples=200, n_features=20, n_informative=3, n_redundant=2, n_repeated=0, n_classes=8, n_clusters_per_class=1, flip_y=0.0, class_sep=10, shuffle=False, random_state=0) univariate_filter = SelectFwe(f_classif, alpha=0.01) X_r = univariate_filter.fit(X, y).transform(X) gtruth = np.zeros(20) gtruth[:5] = 1 for mode in ['fdr', 'fpr', 'fwe']: X_r2 = GenericUnivariateSelect( f_classif, mode=mode, param=0.01).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() assert_array_almost_equal(support, gtruth) ############################################################################## # Test univariate selection in regression settings def assert_best_scores_kept(score_filter): scores = score_filter.scores_ support = score_filter.get_support() assert_array_equal(np.sort(scores[support]), np.sort(scores)[-support.sum():]) def test_select_percentile_regression(): # Test whether the relative univariate feature selection # gets the correct items in a simple regression problem # with the percentile heuristic X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0) univariate_filter = SelectPercentile(f_regression, percentile=25) X_r = univariate_filter.fit(X, y).transform(X) assert_best_scores_kept(univariate_filter) X_r2 = GenericUnivariateSelect( f_regression, mode='percentile', param=25).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) X_2 = X.copy() X_2[:, np.logical_not(support)] = 0 assert_array_equal(X_2, univariate_filter.inverse_transform(X_r)) # Check inverse_transform respects dtype assert_array_equal(X_2.astype(bool), univariate_filter.inverse_transform(X_r.astype(bool))) def test_select_percentile_regression_full(): # Test whether the relative univariate feature selection # selects all features when '100%' is asked. X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0) univariate_filter = SelectPercentile(f_regression, percentile=100) X_r = univariate_filter.fit(X, y).transform(X) assert_best_scores_kept(univariate_filter) X_r2 = GenericUnivariateSelect( f_regression, mode='percentile', param=100).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.ones(20) assert_array_equal(support, gtruth) def test_invalid_percentile(): X, y = make_regression(n_samples=10, n_features=20, n_informative=2, shuffle=False, random_state=0) assert_raises(ValueError, SelectPercentile(percentile=-1).fit, X, y) assert_raises(ValueError, SelectPercentile(percentile=101).fit, X, y) assert_raises(ValueError, GenericUnivariateSelect(mode='percentile', param=-1).fit, X, y) assert_raises(ValueError, GenericUnivariateSelect(mode='percentile', param=101).fit, X, y) def test_select_kbest_regression(): # Test whether the relative univariate feature selection # gets the correct items in a simple regression problem # with the k best heuristic X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0, noise=10) univariate_filter = SelectKBest(f_regression, k=5) X_r = univariate_filter.fit(X, y).transform(X) assert_best_scores_kept(univariate_filter) X_r2 = GenericUnivariateSelect( f_regression, mode='k_best', param=5).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support, gtruth) def test_select_heuristics_regression(): # Test whether the relative univariate feature selection # gets the correct items in a simple regression problem # with the fpr, fdr or fwe heuristics X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0, noise=10) univariate_filter = SelectFpr(f_regression, alpha=0.01) X_r = univariate_filter.fit(X, y).transform(X) gtruth = np.zeros(20) gtruth[:5] = 1 for mode in ['fdr', 'fpr', 'fwe']: X_r2 = GenericUnivariateSelect( f_regression, mode=mode, param=0.01).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() assert_array_equal(support[:5], np.ones((5, ), dtype=np.bool)) assert_less(np.sum(support[5:] == 1), 3) def test_select_fdr_regression(): # Test that fdr heuristic actually has low FDR. def single_fdr(alpha, n_informative, random_state): X, y = make_regression(n_samples=150, n_features=20, n_informative=n_informative, shuffle=False, random_state=random_state, noise=10) with warnings.catch_warnings(record=True): # Warnings can be raised when no features are selected # (low alpha or very noisy data) univariate_filter = SelectFdr(f_regression, alpha=alpha) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect( f_regression, mode='fdr', param=alpha).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() num_false_positives = np.sum(support[n_informative:] == 1) num_true_positives = np.sum(support[:n_informative] == 1) if num_false_positives == 0: return 0. false_discovery_rate = (num_false_positives / (num_true_positives + num_false_positives)) return false_discovery_rate for alpha in [0.001, 0.01, 0.1]: for n_informative in [1, 5, 10]: # As per Benjamini-Hochberg, the expected false discovery rate # should be lower than alpha: # FDR = E(FP / (TP + FP)) <= alpha false_discovery_rate = np.mean([single_fdr(alpha, n_informative, random_state) for random_state in range(30)]) assert_greater_equal(alpha, false_discovery_rate) # Make sure that the empirical false discovery rate increases # with alpha: if false_discovery_rate != 0: assert_greater(false_discovery_rate, alpha / 10) def test_select_fwe_regression(): # Test whether the relative univariate feature selection # gets the correct items in a simple regression problem # with the fwe heuristic X, y = make_regression(n_samples=200, n_features=20, n_informative=5, shuffle=False, random_state=0) univariate_filter = SelectFwe(f_regression, alpha=0.01) X_r = univariate_filter.fit(X, y).transform(X) X_r2 = GenericUnivariateSelect( f_regression, mode='fwe', param=0.01).fit(X, y).transform(X) assert_array_equal(X_r, X_r2) support = univariate_filter.get_support() gtruth = np.zeros(20) gtruth[:5] = 1 assert_array_equal(support[:5], np.ones((5, ), dtype=np.bool)) assert_less(np.sum(support[5:] == 1), 2) def test_selectkbest_tiebreaking(): # Test whether SelectKBest actually selects k features in case of ties. # Prior to 0.11, SelectKBest would return more features than requested. Xs = [[0, 1, 1], [0, 0, 1], [1, 0, 0], [1, 1, 0]] y = [1] dummy_score = lambda X, y: (X[0], X[0]) for X in Xs: sel = SelectKBest(dummy_score, k=1) X1 = ignore_warnings(sel.fit_transform)([X], y) assert_equal(X1.shape[1], 1) assert_best_scores_kept(sel) sel = SelectKBest(dummy_score, k=2) X2 = ignore_warnings(sel.fit_transform)([X], y) assert_equal(X2.shape[1], 2) assert_best_scores_kept(sel) def test_selectpercentile_tiebreaking(): # Test if SelectPercentile selects the right n_features in case of ties. Xs = [[0, 1, 1], [0, 0, 1], [1, 0, 0], [1, 1, 0]] y = [1] dummy_score = lambda X, y: (X[0], X[0]) for X in Xs: sel = SelectPercentile(dummy_score, percentile=34) X1 = ignore_warnings(sel.fit_transform)([X], y) assert_equal(X1.shape[1], 1) assert_best_scores_kept(sel) sel = SelectPercentile(dummy_score, percentile=67) X2 = ignore_warnings(sel.fit_transform)([X], y) assert_equal(X2.shape[1], 2) assert_best_scores_kept(sel) def test_tied_pvalues(): # Test whether k-best and percentiles work with tied pvalues from chi2. # chi2 will return the same p-values for the following features, but it # will return different scores. X0 = np.array([[10000, 9999, 9998], [1, 1, 1]]) y = [0, 1] for perm in itertools.permutations((0, 1, 2)): X = X0[:, perm] Xt = SelectKBest(chi2, k=2).fit_transform(X, y) assert_equal(Xt.shape, (2, 2)) assert_not_in(9998, Xt) Xt = SelectPercentile(chi2, percentile=67).fit_transform(X, y) assert_equal(Xt.shape, (2, 2)) assert_not_in(9998, Xt) def test_tied_scores(): # Test for stable sorting in k-best with tied scores. X_train = np.array([[0, 0, 0], [1, 1, 1]]) y_train = [0, 1] for n_features in [1, 2, 3]: sel = SelectKBest(chi2, k=n_features).fit(X_train, y_train) X_test = sel.transform([[0, 1, 2]]) assert_array_equal(X_test[0], np.arange(3)[-n_features:]) def test_nans(): # Assert that SelectKBest and SelectPercentile can handle NaNs. # First feature has zero variance to confuse f_classif (ANOVA) and # make it return a NaN. X = [[0, 1, 0], [0, -1, -1], [0, .5, .5]] y = [1, 0, 1] for select in (SelectKBest(f_classif, 2), SelectPercentile(f_classif, percentile=67)): ignore_warnings(select.fit)(X, y) assert_array_equal(select.get_support(indices=True), np.array([1, 2])) def test_score_func_error(): X = [[0, 1, 0], [0, -1, -1], [0, .5, .5]] y = [1, 0, 1] for SelectFeatures in [SelectKBest, SelectPercentile, SelectFwe, SelectFdr, SelectFpr, GenericUnivariateSelect]: assert_raises(TypeError, SelectFeatures(score_func=10).fit, X, y) def test_invalid_k(): X = [[0, 1, 0], [0, -1, -1], [0, .5, .5]] y = [1, 0, 1] assert_raises(ValueError, SelectKBest(k=-1).fit, X, y) assert_raises(ValueError, SelectKBest(k=4).fit, X, y) assert_raises(ValueError, GenericUnivariateSelect(mode='k_best', param=-1).fit, X, y) assert_raises(ValueError, GenericUnivariateSelect(mode='k_best', param=4).fit, X, y) def test_f_classif_constant_feature(): # Test that f_classif warns if a feature is constant throughout. X, y = make_classification(n_samples=10, n_features=5) X[:, 0] = 2.0 assert_warns(UserWarning, f_classif, X, y) def test_no_feature_selected(): rng = np.random.RandomState(0) # Generate random uncorrelated data: a strict univariate test should # rejects all the features X = rng.rand(40, 10) y = rng.randint(0, 4, size=40) strict_selectors = [ SelectFwe(alpha=0.01).fit(X, y), SelectFdr(alpha=0.01).fit(X, y), SelectFpr(alpha=0.01).fit(X, y), SelectPercentile(percentile=0).fit(X, y), SelectKBest(k=0).fit(X, y), ] for selector in strict_selectors: assert_array_equal(selector.get_support(), np.zeros(10)) X_selected = assert_warns_message( UserWarning, 'No features were selected', selector.transform, X) assert_equal(X_selected.shape, (40, 0))
bsd-3-clause
luo66/scikit-learn
benchmarks/bench_plot_ward.py
288
1260
""" Benchmark scikit-learn's Ward implement compared to SciPy's """ import time import numpy as np from scipy.cluster import hierarchy import pylab as pl from sklearn.cluster import AgglomerativeClustering ward = AgglomerativeClustering(n_clusters=3, linkage='ward') n_samples = np.logspace(.5, 3, 9) n_features = np.logspace(1, 3.5, 7) N_samples, N_features = np.meshgrid(n_samples, n_features) scikits_time = np.zeros(N_samples.shape) scipy_time = np.zeros(N_samples.shape) for i, n in enumerate(n_samples): for j, p in enumerate(n_features): X = np.random.normal(size=(n, p)) t0 = time.time() ward.fit(X) scikits_time[j, i] = time.time() - t0 t0 = time.time() hierarchy.ward(X) scipy_time[j, i] = time.time() - t0 ratio = scikits_time / scipy_time pl.figure("scikit-learn Ward's method benchmark results") pl.imshow(np.log(ratio), aspect='auto', origin="lower") pl.colorbar() pl.contour(ratio, levels=[1, ], colors='k') pl.yticks(range(len(n_features)), n_features.astype(np.int)) pl.ylabel('N features') pl.xticks(range(len(n_samples)), n_samples.astype(np.int)) pl.xlabel('N samples') pl.title("Scikit's time, in units of scipy time (log)") pl.show()
bsd-3-clause
mlflow/mlflow
examples/pyspark_ml_autologging/pipeline.py
1
1185
from pyspark.ml.classification import LogisticRegression from pyspark.ml.feature import VectorAssembler, StandardScaler from pyspark.ml import Pipeline from pyspark.sql import SparkSession from sklearn.datasets import load_iris import mlflow spark = SparkSession.builder.getOrCreate() mlflow.pyspark.ml.autolog() df = load_iris(as_frame=True).frame.rename(columns={"target": "label"}) df = spark.createDataFrame(df) train, test = df.randomSplit([0.8, 0.2]) assembler = VectorAssembler(inputCols=df.columns[:-1], outputCol="features") scaler = StandardScaler(inputCol=assembler.getOutputCol(), outputCol="scaledFeatures") lor = LogisticRegression(maxIter=5, featuresCol=scaler.getOutputCol()) # Non-neseted pipeline pipeline = Pipeline(stages=[assembler, scaler, lor]) with mlflow.start_run(): pipeline_model = pipeline.fit(train) columns = ["features", "prediction"] pipeline_model.transform(test).select(columns).show() # Nested pipeline nested_pipeline = Pipeline(stages=[Pipeline(stages=[assembler, scaler]), lor]) with mlflow.start_run(): nested_pipeline_model = nested_pipeline.fit(train) nested_pipeline_model.transform(test).select(columns).show() spark.stop()
apache-2.0
sgenoud/scikit-learn
examples/plot_feature_selection.py
1
2443
""" =============================== Univariate Feature Selection =============================== An example showing univariate feature selection. Noisy (non informative) features are added to the iris data and univariate feature selection is applied. For each feature, we plot the p-values for the univariate feature selection and the corresponding weights of an SVM. We can see that univariate feature selection selects the informative features and that these have larger SVM weights. In the total set of features, only the 4 first ones are significant. We can see that they have the highest score with univariate feature selection. The SVM attributes small weights to these features, but these weight are non zero. Applying univariate feature selection before the SVM increases the SVM weight attributed to the significant features, and will thus improve classification. """ print __doc__ import numpy as np import pylab as pl from sklearn import datasets, svm from sklearn.feature_selection import SelectPercentile, f_classif ############################################################################### # import some data to play with # The IRIS dataset iris = datasets.load_iris() # Some noisy data not correlated E = np.random.normal(size=(len(iris.data), 35)) # Add the noisy data to the informative features x = np.hstack((iris.data, E)) y = iris.target ############################################################################### pl.figure(1) pl.clf() x_indices = np.arange(x.shape[-1]) ############################################################################### # Univariate feature selection with F-test for feature scoring # We use the default selection function: the 10% most significant features selector = SelectPercentile(f_classif, percentile=10) selector.fit(x, y) scores = -np.log10(selector.scores_) scores /= scores.max() pl.bar(x_indices - .45, scores, width=.3, label=r'Univariate score ($-Log(p_{value})$)', color='g') ############################################################################### # Compare to the weights of an SVM clf = svm.SVC(kernel='linear') clf.fit(x, y) svm_weights = (clf.coef_ ** 2).sum(axis=0) svm_weights /= svm_weights.max() pl.bar(x_indices - .15, svm_weights, width=.3, label='SVM weight', color='r') pl.title("Comparing feature selection") pl.xlabel('Feature number') pl.yticks(()) pl.axis('tight') pl.legend(loc='upper right') pl.show()
bsd-3-clause
rcln/tag.suggestion
code_python27/toolEval/alchemyAPItagger/alchemyeval_strict.py
1
5202
# -*- coding: utf-8 -*- """ Created on Thu Jul 16 18:40:12 2015 @author: ivan """ from __future__ import division #from sklearn.feature_extraction.text import CountVectorizer #from sklearn.feature_extraction.text import TfidfTransformer import argparse from lxml import etree import cPickle import os import time import numpy as np from alchemyapi import AlchemyAPI import json def parseBlogFile(myfile): f=open(myfile) document = etree.fromstring(f.read()) date=document.xpath("date/text()") # print date[0] title=document.xpath("title/text()") # print title[0] author=document.xpath("author/text()") # print author[0] tags=document.xpath("tags_set/tag/text()") # for tag in tags: # print tag.text cats=document.xpath("categories_set/category/text()") # for cat in cats: # print cat.text text=document.xpath("text/text()")[0] # print text return (date, author, tags, cats, text, title) def exploreDir(targetDir): f=[] for (dirpath,dirnames,filenames) in os.walk(targetDir): f.extend(filenames) break return f def readXMLCorpusFrom(thisDirectory): corpusFiles=exploreDir(thisDirectory) corpus=[] tagw=[] #indx=0 for doc in corpusFiles: #print doc [dat,aut,tag,cat,txt,tit]=parseBlogFile(thisDirectory+doc) corpus.append(txt) #print type(tag) tagw.append(tag) #if indx==11: # print tag #print tit #indx+=1 return [corpus,tagw] def loadStopwords(listfile): lf=open(listfile) stopwordlst=[] for word in lf.readlines(): word=word.replace('\n',"") stopwordlst.append(unicode(word,'utf8')) #print word return stopwordlst def evalAvPrecision(prop,tgs): n=len(tgs) precArr=[] #cycle to go through all instances for x in range(n): count=0 #cycle to go through actual tags of instance x for tg in tgs[x]: #cycle to go through proposed tags for instance x for tprop in prop[x]: #if x==11: # print tg.lower()+" "+tprop if tg.lower() == tprop: #if x==11: # print tg.lower()+" "+tprop count+=1 #if x==11: # print strcount if len(tgs[x])> 0 : precArr.append(count/len(prop[x])) return np.mean(precArr), precArr def evalAvRecall(prop,tgs): n=len(tgs) recArr=[] #cycle to go through all instances for x in range(n): count=0 #cycle to go through actual tags of instance x for tg in tgs[x]: #cycle to go through proposed tags for instance x for tprop in prop[x]: if tg.lower()==tprop: count+=1 if len(tgs[x])> 0 : recArr.append(count/len(tgs[x])) return np.mean(recArr), recArr def f1measure(a,b): if (a+b)>0: return 2*(a*b)/(a+b) else: return 0 def mergeList(l1,l2): n=len(l1) l3=[] for x in range(n): l3.append(list(set(l1[x]+l2[x]))) return l3 parser = argparse.ArgumentParser() parser.add_argument('corpusloc', help='corpus location') parser.add_argument('resultfile', help='file storing results') parser.add_argument('--n', dest="nTOpropose", type=int, default=10, help='how many tags will be proposed, default=10') parser.add_argument('--ngram', dest="ngram", type=int, default=2, help='max size of the n-grams considered, default=2') args = parser.parse_args() if args.corpusloc[-1]!='/': args.corpusloc+='/' topn=0-args.nTOpropose #stpwrds=loadStopwords("stopwords.french.list") #print len(stpwrds) start = time.time() #vectorizer = CountVectorizer(ngram_range=(1,args.ngram),lowercase=True,stop_words=stpwrds) [txtCorpus,tagsw]=readXMLCorpusFrom(args.corpusloc) #print tagsw[11] #print len(txtCorpus) tgsugtop=[] alchemyapi = AlchemyAPI() for docX in txtCorpus: #print docX tagsug=[] response = alchemyapi.combined('text', docX) if response['status'] == 'OK': for keyword in response['keywords']: tagsug.append([keyword['text'],keyword['relevance']]) for entity in response['entities']: tagsug.append([entity['text'],entity['relevance']]) for tgsg in sorted(tagsug, key=lambda x: x[1])[:10]: tgsugtop.append(tgsg[0]) print tgsg[0] #-------------------------------------- #f=open('../../results/tfidfsuggestion/'+args.resultfile+"_"+str(args.nTOpropose)+".res", 'wb') f=open("../../../results/tooleval/alchemyAPItagger/"+args.resultfile+"_"+str(args.nTOpropose)+".res", 'wb') cPickle.dump(tgsugtop, f) #cPickle.dump(tagsw, f) f.close() end = time.time() print "processing time: "+str(end - start) start = time.time() print "Evaluation AlchemyAPI tagger:" [avprec2,precs2]=evalAvPrecision(tgsugtop,tagsw) [avrec2,recs2]=evalAvRecall(tgsugtop,tagsw) print "Precision "+str(avprec2) print "Recall "+str(avrec2) print "F1-measure "+str(f1measure(avprec2,avrec2)) end = time.time() print "evaluation time: "+str(end - start)
gpl-2.0
TinghuiWang/pyActLearn
examples/CASAS_Single_Test/b1_lstm_raw.py
1
8331
import os import pickle import logging import argparse import numpy as np import tensorflow as tf from datetime import datetime from pyActLearn.CASAS.data import CASASData from pyActLearn.CASAS.fuel import CASASFuel from pyActLearn.learning.nn.lstm import LSTM from pyActLearn.performance.record import LearningResult from pyActLearn.performance import get_confusion_matrix logger = logging.getLogger(__file__) def training_and_test(token, train_data, test_data, num_classes, result, model, log_dir): """Train and test Args: token (:obj:`str`): token representing this run train_data (:obj:`tuple` of :obj:`numpy.array`): Tuple of training feature and label test_data (:obj:`tuple` of :obj:`numpy.array`): Tuple of testing feature and label num_classes (:obj:`int`): Number of classes result (:obj:`pyActLearn.performance.record.LearningResult`): LearningResult object to hold learning result """ train_y = np.zeros((train_data[1].shape[0], num_classes)) test_y = np.zeros((test_data[1].shape[0], num_classes)) for i in range(train_data[1].shape[0]): train_y[i, train_data[1].flatten()[i]] = 1 for i in range(test_data[1].shape[0]): test_y[i, test_data[1].flatten()[i]] = 1 model.fit(train_data[0], train_y, iter_num=8000, batch_size=100, criterion='monitor_based', summaries_dir=log_dir, test_x=test_data[0], test_y=test_y, summary_interval=100) # Test predicted_y = model.predict(test_data[0]) predicted_proba = model.predict_proba(test_data[0]) # Evaluate the Test and Store Result confusion_matrix = get_confusion_matrix(num_classes=num_classes, label=test_data[1][model.num_steps:].flatten(), predicted=predicted_y) variable_file = os.path.join(log_dir, token + '_save.ckpt') saver.save(model.sess, variable_file) result.add_record(variable_file, key=token, confusion_matrix=confusion_matrix) return predicted_y, predicted_proba def load_and_test(token, test_data, num_classes, result, model): """Load and test Args: token (:obj:`str`): token representing this run test_data (:obj:`tuple` of :obj:`numpy.array`): Tuple of testing feature and label num_classes (:obj:`int`): Number of classes result (:obj:`pyActLearn.performance.record.LearningResult`): LearningResult object to hold learning result """ saver.restore(model.sess, result.get_record_by_key(token)['model']) # Test predicted_y = model.predict(test_data[0]) predicted_proba = model.predict_proba(test_data[0]) return predicted_y, predicted_proba if __name__ == '__main__': args_ok = False parser = argparse.ArgumentParser(description='Run LSTM on single resident CASAS datasets.') parser.add_argument('-d', '--dataset', help='Directory to original datasets') parser.add_argument('-o', '--output', help='Output folder') parser.add_argument('--week', type=int, metavar='N', help='Train on week N-1 and run on week N') parser.add_argument('--h5py', help='HDF5 dataset folder') args = parser.parse_args() # Default parameters log_filename = os.path.basename(__file__).split('.')[0] + \ '-%s.log' % datetime.now().strftime('%y%m%d_%H:%M:%S') # Setup output directory output_dir = args.output if output_dir is not None: output_dir = os.path.abspath(os.path.expanduser(output_dir)) if os.path.exists(output_dir): # Found output_dir, check if it is a directory if not os.path.isdir(output_dir): exit('Output directory %s is found, but not a directory. Abort.' % output_dir) else: # Create directory os.makedirs(output_dir) else: output_dir = '.' log_filename = os.path.join(output_dir, log_filename) # Setup Logging as early as possible logging.basicConfig(level=logging.DEBUG, format='[%(asctime)s] %(name)s:%(levelname)s:%(message)s', handlers=[logging.FileHandler(log_filename), logging.StreamHandler()]) # If dataset is specified, update h5py casas_data_dir = args.dataset if casas_data_dir is not None: casas_data_dir = os.path.abspath(os.path.expanduser(casas_data_dir)) if not os.path.isdir(casas_data_dir): exit('CASAS dataset at %s does not exist. Abort.' % casas_data_dir) # Find h5py dataset first h5py_dir = args.h5py if h5py_dir is not None: h5py_dir = os.path.abspath(os.path.expanduser(h5py_dir)) else: # Default location h5py_dir = os.path.join(output_dir, 'h5py') if os.path.exists(h5py_dir): if not os.path.isdir(h5py_dir): exit('h5py dataset location %s is not a directory. Abort.' % h5py_dir) if not CASASFuel.files_exist(h5py_dir): # Finish check and creating all directory needed - now load datasets if casas_data_dir is not None: casas_data = CASASData(path=casas_data_dir) casas_data.summary() # SVM needs to use statistical feature with per-sensor and normalization casas_data.populate_feature(method='raw', normalized=True, per_sensor=True) casas_data.export_hdf5(h5py_dir) casas_fuel = CASASFuel(dir_name=h5py_dir) # Prepare learning result result_pkl_file = os.path.join(output_dir, 'result.pkl') result = None if os.path.isfile(result_pkl_file): f = open(result_pkl_file, 'rb') result = pickle.load(f) f.close() if result.data != h5py_dir: logger.error('Result pickle file found for different dataset %s' % result.data) exit('Cannot save learning result at %s' % result_pkl_file) else: result = LearningResult(name='LSTM', data=h5py_dir, mode='by_week') num_classes = casas_fuel.get_output_dims() # Open Fuel and get all splits split_list = casas_fuel.get_set_list() # If week is specified if args.week is not None: if 0 < args.week < len(split_list): split_list = [split_list[args.week - 1], split_list[args.week]] # Start training train_names = ('week 24', 'week 23', 'week 22', 'week 21') test_names = ('week 25', 'week 26', 'week 27', 'week 28') test_name = 'single_test' train_set = casas_fuel.get_dataset(train_names, load_in_memory=True) (train_set_data) = train_set.data_sources test_set = casas_fuel.get_dataset(test_names, load_in_memory=True) (test_set_data) = test_set.data_sources # Prepare Back Annotation fp_back_annotated = open(os.path.join(output_dir, 'back_annotated.txt'), 'w') fp_back_probability = open(os.path.join(output_dir, 'back_annotated_proba.txt'), 'w') output_log_dir = os.path.join(output_dir, 'log') if not os.path.isdir(output_log_dir): os.makedirs(output_log_dir) model = LSTM(casas_fuel.get_input_dims(), casas_fuel.get_output_dims(), num_units=200, num_steps=100) saver = tf.train.Saver(max_to_keep=len(split_list)) session = tf.Session() model.sess = session log_dir = output_log_dir if not os.path.isdir(log_dir): os.makedirs(log_dir) # run svm logger.info('Training on %s, Testing on %s' % (str(train_names), str(test_names))) if result.get_record_by_key(test_name) is None: prediction, prediction_proba = training_and_test(test_name, train_set_data, test_set_data, num_classes, result, model=model, log_dir=log_dir) else: prediction, prediction_proba = load_and_test(test_name, test_set_data, num_classes, result, model=model) casas_fuel.back_annotate(fp_back_annotated, prediction=prediction, split_name=test_names) casas_fuel.back_annotate_with_proba(fp_back_probability, prediction_proba, split_name=test_names) train_name = test_name train_set_data = test_set_data fp_back_annotated.close() fp_back_probability.close() f = open(result_pkl_file, 'wb') pickle.dump(obj=result, file=f, protocol=pickle.HIGHEST_PROTOCOL) f.close() result.export_to_xlsx(os.path.join(output_dir, 'result.xlsx'))
bsd-3-clause
lilleswing/deepchem
contrib/one_shot_models/examples/sider_from_tox21_res_one_fold.py
8
2496
""" Train low-data res models on Tox21. Test on SIDER. Test last fold only. """ from __future__ import print_function from __future__ import division from __future__ import unicode_literals import numpy as np import deepchem as dc import tensorflow as tf from datasets import load_sider_convmol from datasets import load_tox21_convmol from datasets import to_numpy_dataset # Number of folds for split K = 4 # Depth of attention module max_depth = 3 # num positive/negative ligands n_pos = 10 n_neg = 10 # Set batch sizes for network test_batch_size = 128 support_batch_size = n_pos + n_neg nb_epochs = 1 n_train_trials = 2000 n_eval_trials = 20 learning_rate = 1e-4 log_every_n_samples = 50 # Number of features on conv-mols n_feat = 75 sider_tasks, sider_dataset, _ = load_sider_convmol() sider_dataset = to_numpy_dataset(sider_dataset) tox21_tasks, tox21_dataset, _ = load_tox21_convmol() tox21_dataset = to_numpy_dataset(tox21_dataset) # Define metric metric = dc.metrics.Metric(dc.metrics.roc_auc_score, mode="classification") # Train support model on train support_model = dc.nn.SequentialSupportGraph(n_feat) # Add layers support_model.add(dc.nn.GraphConv(64, n_feat, activation='relu')) support_model.add(dc.nn.GraphPool()) support_model.add(dc.nn.GraphConv(128, 64, activation='relu')) support_model.add(dc.nn.GraphPool()) support_model.add(dc.nn.GraphConv(64, 128, activation='relu')) support_model.add(dc.nn.GraphPool()) support_model.add(dc.nn.Dense(128, 64, activation='tanh')) support_model.add_test(dc.nn.GraphGather(test_batch_size, activation='tanh')) support_model.add_support( dc.nn.GraphGather(support_batch_size, activation='tanh')) # Apply a residual lstm layer support_model.join( dc.nn.ResiLSTMEmbedding(test_batch_size, support_batch_size, 128, max_depth)) model = dc.models.SupportGraphClassifier( support_model, test_batch_size=test_batch_size, support_batch_size=support_batch_size, learning_rate=learning_rate) model.fit( tox21_dataset, nb_epochs=nb_epochs, n_episodes_per_epoch=n_train_trials, n_pos=n_pos, n_neg=n_neg, log_every_n_samples=log_every_n_samples) mean_scores, std_scores = model.evaluate( sider_dataset, metric, n_pos, n_neg, n_trials=n_eval_trials) print("Mean Scores on evaluation dataset") print(mean_scores) print("Standard Deviations on evaluation dataset") print(std_scores) print("Median of Mean Scores") print(np.median(np.array(mean_scores.values())))
mit
tomsilver/nupic
tests/swarming/nupic/swarming/experiments/spatial_classification/description.py
1
15598
# ---------------------------------------------------------------------- # Numenta Platform for Intelligent Computing (NuPIC) # Copyright (C) 2013, Numenta, Inc. Unless you have an agreement # with Numenta, Inc., for a separate license for this software code, the # following terms and conditions apply: # # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License version 3 as # published by the Free Software Foundation. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. # See the GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program. If not, see http://www.gnu.org/licenses. # # http://numenta.org/licenses/ # ---------------------------------------------------------------------- """ Template file used by the OPF Experiment Generator to generate the actual description.py file by replacing $XXXXXXXX tokens with desired values. This description.py file was generated by: '/Users/ronmarianetti/nupic/eng/lib/python2.6/site-packages/nupicengine/frameworks/opf/expGenerator/ExpGenerator.pyc' """ from nupic.frameworks.opf.expdescriptionapi import ExperimentDescriptionAPI from nupic.frameworks.opf.expdescriptionhelpers import ( updateConfigFromSubConfig, applyValueGettersToContainer, DeferredDictLookup) from nupic.frameworks.opf.clamodelcallbacks import * from nupic.frameworks.opf.metrics import MetricSpec from nupic.frameworks.opf.opfutils import (InferenceType, InferenceElement) from nupic.support import aggregationDivide from nupic.frameworks.opf.opftaskdriver import ( IterationPhaseSpecLearnOnly, IterationPhaseSpecInferOnly, IterationPhaseSpecLearnAndInfer) # Model Configuration Dictionary: # # Define the model parameters and adjust for any modifications if imported # from a sub-experiment. # # These fields might be modified by a sub-experiment; this dict is passed # between the sub-experiment and base experiment # # # NOTE: Use of DEFERRED VALUE-GETTERs: dictionary fields and list elements # within the config dictionary may be assigned futures derived from the # ValueGetterBase class, such as DeferredDictLookup. # This facility is particularly handy for enabling substitution of values in # the config dictionary from other values in the config dictionary, which is # needed by permutation.py-based experiments. These values will be resolved # during the call to applyValueGettersToContainer(), # which we call after the base experiment's config dictionary is updated from # the sub-experiment. See ValueGetterBase and # DeferredDictLookup for more details about value-getters. # # For each custom encoder parameter to be exposed to the sub-experiment/ # permutation overrides, define a variable in this section, using key names # beginning with a single underscore character to avoid collisions with # pre-defined keys (e.g., _dsEncoderFieldName2_N). # # Example: # config = dict( # _dsEncoderFieldName2_N = 70, # _dsEncoderFieldName2_W = 5, # dsEncoderSchema = [ # base=dict( # fieldname='Name2', type='ScalarEncoder', # name='Name2', minval=0, maxval=270, clipInput=True, # n=DeferredDictLookup('_dsEncoderFieldName2_N'), # w=DeferredDictLookup('_dsEncoderFieldName2_W')), # ], # ) # updateConfigFromSubConfig(config) # applyValueGettersToContainer(config) config = { # Type of model that the rest of these parameters apply to. 'model': "CLA", # Version that specifies the format of the config. 'version': 1, # Intermediate variables used to compute fields in modelParams and also # referenced from the control section. 'aggregationInfo': { 'fields': [], 'days': 0, 'hours': 0, 'microseconds': 0, 'milliseconds': 0, 'minutes': 0, 'months': 0, 'seconds': 0, 'weeks': 0, 'years': 0 }, 'predictAheadTime': None, # Model parameter dictionary. 'modelParams': { # The type of inference that this model will perform 'inferenceType': 'NontemporalClassification', 'sensorParams': { # Sensor diagnostic output verbosity control; # if > 0: sensor region will print out on screen what it's sensing # at each step 0: silent; >=1: some info; >=2: more info; # >=3: even more info (see compute() in py/regions/RecordSensor.py) 'verbosity' : 0, # Example: # dsEncoderSchema = [ # DeferredDictLookup('__field_name_encoder'), # ], # # (value generated from DS_ENCODER_SCHEMA) 'encoders': { 'address': { 'fieldname': u'address', 'n': 300, 'name': u'address', 'type': 'SDRCategoryEncoder', 'w': 21 }, '_classifierInput': { 'name': u'_classifierInput', 'fieldname': u'consumption', 'classifierOnly': True, 'clipInput': True, 'maxval': 200, 'minval': 0, 'n': 1500, 'type': 'ScalarEncoder', 'w': 21 }, 'gym': { 'fieldname': u'gym', 'n': 300, 'name': u'gym', 'type': 'SDRCategoryEncoder', 'w': 21 }, 'timestamp_dayOfWeek': { 'dayOfWeek': (7, 3), 'fieldname': u'timestamp', 'name': u'timestamp_dayOfWeek', 'type': 'DateEncoder' }, 'timestamp_timeOfDay': { 'fieldname': u'timestamp', 'name': u'timestamp_timeOfDay', 'timeOfDay': (7, 8), 'type': 'DateEncoder' } }, # A dictionary specifying the period for automatically-generated # resets from a RecordSensor; # # None = disable automatically-generated resets (also disabled if # all of the specified values evaluate to 0). # Valid keys is the desired combination of the following: # days, hours, minutes, seconds, milliseconds, microseconds, weeks # # Example for 1.5 days: sensorAutoReset = dict(days=1,hours=12), # # (value generated from SENSOR_AUTO_RESET) 'sensorAutoReset' : None, }, 'spEnable': False, 'spParams': { # SP diagnostic output verbosity control; # 0: silent; >=1: some info; >=2: more info; 'spVerbosity' : 0, 'globalInhibition': 1, # Number of cell columns in the cortical region (same number for # SP and TP) # (see also tpNCellsPerCol) 'columnCount': 2048, 'inputWidth': 0, # SP inhibition control (absolute value); # Maximum number of active columns in the SP region's output (when # there are more, the weaker ones are suppressed) 'numActiveColumnsPerInhArea': 40, 'seed': 1956, # potentialPct # What percent of the columns's receptive field is available # for potential synapses. At initialization time, we will # choose potentialPct * (2*potentialRadius+1)^2 'potentialPct': 0.5, # The default connected threshold. Any synapse whose # permanence value is above the connected threshold is # a "connected synapse", meaning it can contribute to the # cell's firing. Typical value is 0.10. Cells whose activity # level before inhibition falls below minDutyCycleBeforeInh # will have their own internal synPermConnectedCell # threshold set below this default value. # (This concept applies to both SP and TP and so 'cells' # is correct here as opposed to 'columns') 'synPermConnected': 0.1, 'synPermActiveInc': 0.1, 'synPermInactiveDec': 0.01, }, # Controls whether TP is enabled or disabled; # TP is necessary for making temporal predictions, such as predicting # the next inputs. Without TP, the model is only capable of # reconstructing missing sensor inputs (via SP). 'tpEnable' : False, 'tpParams': { # TP diagnostic output verbosity control; # 0: silent; [1..6]: increasing levels of verbosity # (see verbosity in nupic/trunk/py/nupic/research/TP.py and TP10X*.py) 'verbosity': 0, # Number of cell columns in the cortical region (same number for # SP and TP) # (see also tpNCellsPerCol) 'columnCount': 2048, # The number of cells (i.e., states), allocated per column. 'cellsPerColumn': 32, 'inputWidth': 2048, 'seed': 1960, # Temporal Pooler implementation selector (see _getTPClass in # CLARegion.py). 'temporalImp': 'cpp', # New Synapse formation count # NOTE: If None, use spNumActivePerInhArea # # TODO: need better explanation 'newSynapseCount': 20, # Maximum number of synapses per segment # > 0 for fixed-size CLA # -1 for non-fixed-size CLA # # TODO: for Ron: once the appropriate value is placed in TP # constructor, see if we should eliminate this parameter from # description.py. 'maxSynapsesPerSegment': 32, # Maximum number of segments per cell # > 0 for fixed-size CLA # -1 for non-fixed-size CLA # # TODO: for Ron: once the appropriate value is placed in TP # constructor, see if we should eliminate this parameter from # description.py. 'maxSegmentsPerCell': 128, # Initial Permanence # TODO: need better explanation 'initialPerm': 0.21, # Permanence Increment 'permanenceInc': 0.1, # Permanence Decrement # If set to None, will automatically default to tpPermanenceInc # value. 'permanenceDec' : 0.1, 'globalDecay': 0.0, 'maxAge': 0, # Minimum number of active synapses for a segment to be considered # during search for the best-matching segments. # None=use default # Replaces: tpMinThreshold 'minThreshold': 12, # Segment activation threshold. # A segment is active if it has >= tpSegmentActivationThreshold # connected synapses that are active due to infActiveState # None=use default # Replaces: tpActivationThreshold 'activationThreshold': 16, 'outputType': 'normal', # "Pay Attention Mode" length. This tells the TP how many new # elements to append to the end of a learned sequence at a time. # Smaller values are better for datasets with short sequences, # higher values are better for datasets with long sequences. 'pamLength': 1, }, 'clParams': { 'regionName' : 'CLAClassifierRegion', 'implementation': 'py', # Classifier diagnostic output verbosity control; # 0: silent; [1..6]: increasing levels of verbosity 'clVerbosity' : 0, # This controls how fast the classifier learns/forgets. Higher values # make it adapt faster and forget older patterns faster. 'alpha': 0.001, # This is set after the call to updateConfigFromSubConfig and is # computed from the aggregationInfo and predictAheadTime. 'steps': '0', }, 'anomalyParams': { u'anomalyCacheRecords': None, u'autoDetectThreshold': None, u'autoDetectWaitRecords': None}, 'trainSPNetOnlyIfRequested': False, }, } # end of config dictionary # Adjust base config dictionary for any modifications if imported from a # sub-experiment updateConfigFromSubConfig(config) # Compute predictionSteps based on the predictAheadTime and the aggregation # period, which may be permuted over. if config['predictAheadTime'] is not None: predictionSteps = int(round(aggregationDivide( config['predictAheadTime'], config['aggregationInfo']))) assert (predictionSteps >= 1) config['modelParams']['clParams']['steps'] = str(predictionSteps) # Adjust config by applying ValueGetterBase-derived # futures. NOTE: this MUST be called after updateConfigFromSubConfig() in order # to support value-getter-based substitutions from the sub-experiment (if any) applyValueGettersToContainer(config) control = { # The environment that the current model is being run in "environment": 'nupic', # Input stream specification per py/nupic/frameworks/opf/jsonschema/stream_def.json. # 'dataset' : { u'info': u'testSpatialClassification', u'streams': [ { u'columns': [u'*'], u'info': u'test data', u'source': u'file://swarming/test_data.csv'}], u'version': 1}, # Iteration count: maximum number of iterations. Each iteration corresponds # to one record from the (possibly aggregated) dataset. The task is # terminated when either number of iterations reaches iterationCount or # all records in the (possibly aggregated) database have been processed, # whichever occurs first. # # iterationCount of -1 = iterate over the entire dataset #'iterationCount' : ITERATION_COUNT, # A dictionary containing all the supplementary parameters for inference "inferenceArgs":{u'predictedField': u'consumption', u'predictionSteps': [0]}, # Metrics: A list of MetricSpecs that instantiate the metrics that are # computed for this experiment 'metrics':[ MetricSpec(field=u'consumption', metric='multiStep', inferenceElement='multiStepBestPredictions', params={'window': 1000, 'steps': [0], 'errorMetric': 'avg_err'}) ], # Logged Metrics: A sequence of regular expressions that specify which of # the metrics from the Inference Specifications section MUST be logged for # every prediction. The regex's correspond to the automatically generated # metric labels. This is similar to the way the optimization metric is # specified in permutations.py. 'loggedMetrics': ['.*'], } ################################################################################ ################################################################################ descriptionInterface = ExperimentDescriptionAPI(modelConfig=config, control=control)
gpl-3.0
danielfree/srs
trunk/research/api-server/server.py
13
48099
#!/usr/bin/python ''' The MIT License (MIT) Copyright (c) 2013-2014 winlin Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. ''' """ the api-server is a default demo server for srs to call when srs get some event, for example, when client connect to srs, srs can invoke the http api of the api-server """ import sys # reload sys model to enable the getdefaultencoding method. reload(sys) # set the default encoding to utf-8 # using exec to set the encoding, to avoid error in IDE. exec("sys.setdefaultencoding('utf-8')") assert sys.getdefaultencoding().lower() == "utf-8" import os, json, time, datetime, cherrypy, threading # simple log functions. def trace(msg): date = datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S") print "[%s][trace] %s"%(date, msg) # enable crossdomain access for js-client # define the following method: # def OPTIONS(self, *args, **kwargs) # enable_crossdomain() # invoke this method to enable js to request crossdomain. def enable_crossdomain(): cherrypy.response.headers["Access-Control-Allow-Origin"] = "*" cherrypy.response.headers["Access-Control-Allow-Methods"] = "GET, POST, HEAD, PUT, DELETE" # generate allow headers for crossdomain. allow_headers = ["Cache-Control", "X-Proxy-Authorization", "X-Requested-With", "Content-Type"] cherrypy.response.headers["Access-Control-Allow-Headers"] = ",".join(allow_headers) # error codes definition class Error: # ok, success, completed. success = 0 # error when parse json system_parse_json = 100 # request action invalid request_invalid_action = 200 # cdn node not exists cdn_node_not_exists = 201 ''' handle the clients requests: connect/disconnect vhost/app. ''' class RESTClients(object): exposed = True def GET(self): enable_crossdomain() clients = {} return json.dumps(clients) ''' for SRS hook: on_connect/on_close on_connect: when client connect to vhost/app, call the hook, the request in the POST data string is a object encode by json: { "action": "on_connect", "client_id": 1985, "ip": "192.168.1.10", "vhost": "video.test.com", "app": "live", "pageUrl": "http://www.test.com/live.html" } on_close: when client close/disconnect to vhost/app/stream, call the hook, the request in the POST data string is a object encode by json: { "action": "on_close", "client_id": 1985, "ip": "192.168.1.10", "vhost": "video.test.com", "app": "live" } if valid, the hook must return HTTP code 200(Stauts OK) and response an int value specifies the error code(0 corresponding to success): 0 ''' def POST(self): enable_crossdomain() # return the error code in str code = Error.success req = cherrypy.request.body.read() trace("post to clients, req=%s"%(req)) try: json_req = json.loads(req) except Exception, ex: code = Error.system_parse_json trace("parse the request to json failed, req=%s, ex=%s, code=%s"%(req, ex, code)) return str(code) action = json_req["action"] if action == "on_connect": code = self.__on_connect(json_req) elif action == "on_close": code = self.__on_close(json_req) else: trace("invalid request action: %s"%(json_req["action"])) code = Error.request_invalid_action return str(code) def OPTIONS(self, *args, **kwargs): enable_crossdomain() def __on_connect(self, req): code = Error.success trace("srs %s: client id=%s, ip=%s, vhost=%s, app=%s, tcUrl=%s, pageUrl=%s"%( req["action"], req["client_id"], req["ip"], req["vhost"], req["app"], req["tcUrl"], req["pageUrl"] )) # TODO: process the on_connect event return code def __on_close(self, req): code = Error.success trace("srs %s: client id=%s, ip=%s, vhost=%s, app=%s"%( req["action"], req["client_id"], req["ip"], req["vhost"], req["app"] )) # TODO: process the on_close event return code ''' handle the streams requests: publish/unpublish stream. ''' class RESTStreams(object): exposed = True def GET(self): enable_crossdomain() streams = {} return json.dumps(streams) ''' for SRS hook: on_publish/on_unpublish on_publish: when client(encoder) publish to vhost/app/stream, call the hook, the request in the POST data string is a object encode by json: { "action": "on_publish", "client_id": 1985, "ip": "192.168.1.10", "vhost": "video.test.com", "app": "live", "stream": "livestream" } on_unpublish: when client(encoder) stop publish to vhost/app/stream, call the hook, the request in the POST data string is a object encode by json: { "action": "on_unpublish", "client_id": 1985, "ip": "192.168.1.10", "vhost": "video.test.com", "app": "live", "stream": "livestream" } if valid, the hook must return HTTP code 200(Stauts OK) and response an int value specifies the error code(0 corresponding to success): 0 ''' def POST(self): enable_crossdomain() # return the error code in str code = Error.success req = cherrypy.request.body.read() trace("post to streams, req=%s"%(req)) try: json_req = json.loads(req) except Exception, ex: code = Error.system_parse_json trace("parse the request to json failed, req=%s, ex=%s, code=%s"%(req, ex, code)) return str(code) action = json_req["action"] if action == "on_publish": code = self.__on_publish(json_req) elif action == "on_unpublish": code = self.__on_unpublish(json_req) else: trace("invalid request action: %s"%(json_req["action"])) code = Error.request_invalid_action return str(code) def OPTIONS(self, *args, **kwargs): enable_crossdomain() def __on_publish(self, req): code = Error.success trace("srs %s: client id=%s, ip=%s, vhost=%s, app=%s, stream=%s"%( req["action"], req["client_id"], req["ip"], req["vhost"], req["app"], req["stream"] )) # TODO: process the on_publish event return code def __on_unpublish(self, req): code = Error.success trace("srs %s: client id=%s, ip=%s, vhost=%s, app=%s, stream=%s"%( req["action"], req["client_id"], req["ip"], req["vhost"], req["app"], req["stream"] )) # TODO: process the on_unpublish event return code ''' handle the sessions requests: client play/stop stream ''' class RESTSessions(object): exposed = True def GET(self): enable_crossdomain() sessions = {} return json.dumps(sessions) ''' for SRS hook: on_play/on_stop on_play: when client(encoder) publish to vhost/app/stream, call the hook, the request in the POST data string is a object encode by json: { "action": "on_play", "client_id": 1985, "ip": "192.168.1.10", "vhost": "video.test.com", "app": "live", "stream": "livestream" } on_stop: when client(encoder) stop publish to vhost/app/stream, call the hook, the request in the POST data string is a object encode by json: { "action": "on_stop", "client_id": 1985, "ip": "192.168.1.10", "vhost": "video.test.com", "app": "live", "stream": "livestream" } if valid, the hook must return HTTP code 200(Stauts OK) and response an int value specifies the error code(0 corresponding to success): 0 ''' def POST(self): enable_crossdomain() # return the error code in str code = Error.success req = cherrypy.request.body.read() trace("post to sessions, req=%s"%(req)) try: json_req = json.loads(req) except Exception, ex: code = Error.system_parse_json trace("parse the request to json failed, req=%s, ex=%s, code=%s"%(req, ex, code)) return str(code) action = json_req["action"] if action == "on_play": code = self.__on_play(json_req) elif action == "on_stop": code = self.__on_stop(json_req) else: trace("invalid request action: %s"%(json_req["action"])) code = Error.request_invalid_action return str(code) def OPTIONS(self, *args, **kwargs): enable_crossdomain() def __on_play(self, req): code = Error.success trace("srs %s: client id=%s, ip=%s, vhost=%s, app=%s, stream=%s"%( req["action"], req["client_id"], req["ip"], req["vhost"], req["app"], req["stream"] )) # TODO: process the on_play event return code def __on_stop(self, req): code = Error.success trace("srs %s: client id=%s, ip=%s, vhost=%s, app=%s, stream=%s"%( req["action"], req["client_id"], req["ip"], req["vhost"], req["app"], req["stream"] )) # TODO: process the on_stop event return code global_arm_server_id = os.getpid(); class ArmServer: def __init__(self): global global_arm_server_id global_arm_server_id += 1 self.id = str(global_arm_server_id) self.ip = None self.device_id = None self.public_ip = cherrypy.request.remote.ip self.heartbeat = time.time() self.clients = 0 def dead(self): dead_time_seconds = 20 if time.time() - self.heartbeat > dead_time_seconds: return True return False def json_dump(self): data = {} data["id"] = self.id data["ip"] = self.ip data["device_id"] = self.device_id data["public_ip"] = self.public_ip data["heartbeat"] = self.heartbeat data["heartbeat_h"] = time.strftime("%Y-%m-%d %H:%M:%S",time.localtime(self.heartbeat)) data["summaries"] = "http://%s:1985/api/v1/summaries"%(self.ip) return data ''' the server list ''' class RESTServers(object): exposed = True def __init__(self): self.__nodes = [] self.__last_update = datetime.datetime.now(); self.__lock = threading.Lock() def __get_node(self, device_id): for node in self.__nodes: if node.device_id == device_id: return node return None def __refresh_nodes(self): while len(self.__nodes) > 0: has_dead_node = False for node in self.__nodes: if node.dead(): self.__nodes.remove(node) has_dead_node = True if not has_dead_node: break def __json_dump_nodes(self, peers): data = [] for node in peers: data.append(node.json_dump()) return data def __get_peers_for_play(self, device_id): peers = [] for node in self.__nodes: if node.device_id == device_id: peers.append(node) return peers def __select_peer(self, peers, device_id): target = None for peer in peers: if target is None or target.clients > peer.clients: target = peer if target is None: return None target.clients += 1 return target.ip ''' post to update server ip. request body: the new raspberry-pi server ip. TODO: FIXME: more info. ''' def POST(self): enable_crossdomain() try: self.__lock.acquire() req = cherrypy.request.body.read() trace("post to nodes, req=%s"%(req)) try: json_req = json.loads(req) except Exception, ex: code = Error.system_parse_json trace("parse the request to json failed, req=%s, ex=%s, code=%s"%(req, ex, code)) return json.dumps({"code":code, "data": None}) device_id = json_req["device_id"] node = self.__get_node(device_id) if node is None: node = ArmServer() self.__nodes.append(node) node.ip = json_req["ip"] node.device_id = device_id node.public_ip = cherrypy.request.remote.ip node.heartbeat = time.time() return json.dumps({"code":Error.success, "data": {"id":node.id}}) finally: self.__lock.release() ''' id canbe: pi: the pi demo, raspberry-pi default demo. device_id: the id of device to get. action: canbe play or mgmt, play to play the inest stream, mgmt to get api/v1/versions. stream: the stream to play, for example, live/livestream for http://server:8080/live/livestream.html meeting: the meeting demo. jump to web meeting if index is None. device_id: the id of device to get. local: whether view the local raspberry-pi stream. if "true", redirect to the local(internal) api server. index: the meeting stream index, dynamic get the streams from root.api.v1.chats.get_url_by_index(index) gslb: the gslb to get edge ip device_id: the id of device to get. ingest: deprecated, alias for pi. ''' def GET(self, id=None, action="play", stream="live/livestream", index=None, local="false", device_id=None): enable_crossdomain() try: self.__lock.acquire() self.__refresh_nodes() data = self.__json_dump_nodes(self.__nodes) server_ip = "demo.chnvideo.com" ip = cherrypy.request.remote.ip if type is not None: peers = self.__get_peers_for_play(device_id) if len(peers) > 0: server_ip = self.__select_peer(peers, device_id) # demo, srs meeting urls. if id == "meeting": if index is None: url = "http://%s:8085"%(server_ip) elif local == "true": url = "http://%s:8085/api/v1/servers?id=%s&index=%s&local=false"%(server_ip, id, index) else: rtmp_url = root.api.v1.chats.get_url_by_index(index) if rtmp_url is None: return "meeting stream not found" urls = rtmp_url.replace("...vhost...", "?vhost=").replace("rtmp://", "").split("/") hls_url = "http://%s:8080/%s/%s.m3u8"%(urls[0].strip(":19350").strip(":1935"), urls[1].split("?")[0], urls[2]) return self.__generate_hls(hls_url) # raspberry-pi urls. elif id == "ingest" or id == "pi": if action == "play": url = "http://%s:8080/%s.html"%(server_ip, stream) elif action == "rtmp": url = "../../players/srs_player.html?server=%s&vhost=%s&app=%s&stream=%s&autostart=true"%(server_ip, server_ip, stream.split("/")[0], stream.split("/")[1]) elif action == "hls": hls_url = "http://%s:8080/%s.m3u8"%(server_ip, stream); if stream.startswith("http://"): hls_url = stream; return self.__generate_hls(hls_url.replace(".m3u8.m3u8", ".m3u8")) else: url = "http://%s:8080/api/v1/versions"%(server_ip) elif id == "gslb": return json.dumps({"code":Error.success, "data": { "edge":server_ip, "client":ip, "peers":self.__json_dump_nodes(peers), "streams": { "pi": { "livestream": { "sales-pi-hls": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-arm&stream=live/livestream", "dev-pi-hls": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-dev-arm&stream=live/livestream" }, "cztv": { "sales-pi-hls": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-arm&stream=live/rtmp_cztv01-sd", "dev-pi-hls": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-dev-arm&stream=live/rtmp_cztv01-sd" } }, "hiwifi": { "hls": { "dev-livestream": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-dev-hiwifi&stream=live/livestream", "sales-livestream": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-hiwifi&stream=live/livestream" }, "rtmp":{ "dev-livestream": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-dev-hiwifi&stream=live/livestream", "sales-livestream": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-sales-hiwifi&stream=live/livestream" }, "meiyi": { "rtmp": { "avatar": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-sales-hiwifi&stream=live/avatar", "MenInBlack3": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-sales-hiwifi&stream=live/MenInBlack3", "skyfall": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-sales-hiwifi&stream=live/skyfall", "SpiderMan": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-sales-hiwifi&stream=live/SpiderMan", "thehobbit": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-sales-hiwifi&stream=live/thehobbit", "thorthedarkworld": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-sales-hiwifi&stream=live/thorthedarkworld", "transformers": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-sales-hiwifi&stream=live/transformers" }, "hls": { "avatar": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-hiwifi&stream=live/avatar", "MenInBlack3": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-hiwifi&stream=live/MenInBlack3", "skyfall": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-hiwifi&stream=live/skyfall", "SpiderMan": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-hiwifi&stream=live/SpiderMan", "thehobbit": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-hiwifi&stream=live/thehobbit", "thorthedarkworld": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-hiwifi&stream=live/thorthedarkworld", "transformers": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-sales-hiwifi&stream=live/transformers" } } }, "cubieboard": { "meiyi": { "rtmp": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard&stream=live/livestream", "stream1": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard&stream=live/stream1", "stream2": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard&stream=live/stream2", "stream3": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard&stream=live/stream3", "stream4": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard&stream=live/stream4", "stream5": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard&stream=live/stream5", "stream6": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard&stream=live/stream6", "stream7": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard&stream=live/stream7" }, "hls": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard&stream=live/livestream", "stream1": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard&stream=live/stream1", "stream2": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard&stream=live/stream2", "stream3": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard&stream=live/stream3", "stream4": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard&stream=live/stream4", "stream5": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard&stream=live/stream5", "stream6": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard&stream=live/stream6", "stream7": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard&stream=live/stream7" } }, "meiyi-house": { "rtmp": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-house&stream=live/livestream" }, "hls": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-house&stream=live/livestream" } }, "meiyi-bk": { "rtmp": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/livestream", "stream1": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream1", "stream2": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream2", "stream3": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream3", "stream4": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream4", "stream5": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream5", "stream6": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream6", "stream7": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream7" }, "hls": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/livestream", "stream1": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream1", "stream2": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream2", "stream3": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream3", "stream4": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream4", "stream5": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream5", "stream6": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream6", "stream7": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-bk&stream=live/stream7" } }, "meiyi-dev1": { "rtmp": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-dev1&stream=live/livestream" }, "hls": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-dev1&stream=live/livestream" } }, "meiyi-dev2": { "rtmp": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=rtmp&device_id=chnvideo-meiyi-cubieboard-dev2&stream=live/livestream" }, "hls": { "livesteam": "http://demo.chnvideo.com:8085/api/v1/servers?id=ingest&action=hls&device_id=chnvideo-meiyi-cubieboard-dev2&stream=live/livestream" } } } } }}) # others, default. else: return json.dumps(data) #return "id=%s, action=%s, stream=%s, url=%s, index=%s, local=%s"%(id, action, stream, url, index, local) raise cherrypy.HTTPRedirect(url) finally: self.__lock.release() def DELETE(self, id): enable_crossdomain() raise cherrypy.HTTPError(405, "Not allowed.") def PUT(self, id): enable_crossdomain() raise cherrypy.HTTPError(405, "Not allowed.") def OPTIONS(self, *args, **kwargs): enable_crossdomain() def __generate_hls(self, hls_url): return SrsUtility().hls_html(hls_url) class SrsUtility: def hls_html(self, hls_url): return """ <h1>%s</h1> <video width="640" height="360" autoplay controls autobuffer src="%s" type="application/vnd.apple.mpegurl"> </video>"""%(hls_url, hls_url); global_cdn_id = os.getpid(); class CdnNode: def __init__(self): global global_cdn_id global_cdn_id += 1 self.id = str(global_cdn_id) self.ip = None self.origin = None self.os = None self.srs_status = None self.public_ip = cherrypy.request.remote.ip self.heartbeat = time.time() self.clients = 0 def dead(self): dead_time_seconds = 10 if time.time() - self.heartbeat > dead_time_seconds: return True return False def json_dump(self): data = {} data["id"] = self.id data["ip"] = self.ip data["origin"] = self.origin data["os"] = self.os data["srs_status"] = self.srs_status data["public_ip"] = self.public_ip data["heartbeat"] = self.heartbeat data["heartbeat_h"] = time.strftime("%Y-%m-%d %H:%M:%S",time.localtime(self.heartbeat)) data["clients"] = self.clients data["summaries"] = "http://%s:1985/api/v1/summaries"%(self.ip) return data ''' the cdn nodes list ''' class RESTNodes(object): exposed = True def __init__(self): self.__nodes = [] # @remark, if there is shared data, such as the self.__nodes, # we must use lock for cherrypy, or the cpu of cherrypy will high # and performance suffer. self.__lock = threading.Lock() def __get_node(self, id): for node in self.__nodes: if node.id == id: return node return None def __refresh_nodes(self): while len(self.__nodes) > 0: has_dead_node = False for node in self.__nodes: if node.dead(): self.__nodes.remove(node) has_dead_node = True if not has_dead_node: break def __get_peers(self, target_node): peers = [] for node in self.__nodes: if str(node.id).strip() == str(target_node.id).strip(): continue if node.public_ip == target_node.public_ip and node.srs_status == "running" and node.origin != target_node.ip: peers.append(node) return peers def __get_peers_for_play(self, ip): peers = [] for node in self.__nodes: if node.public_ip == ip and node.srs_status == "running": peers.append(node) return peers def __json_dump_nodes(self, peers): data = [] for node in peers: data.append(node.json_dump()) return data def __select_peer(self, peers, ip): target = None for peer in peers: if target is None or target.clients > peer.clients: target = peer if target is None: return None target.clients += 1 return target.ip def GET(self, type=None, format=None, origin=None, vhost=None, port=None, stream=None, node_id=None): enable_crossdomain() try: self.__lock.acquire() self.__refresh_nodes() data = self.__json_dump_nodes(self.__nodes) ip = cherrypy.request.remote.ip if type is not None: server = origin peers = self.__get_peers_for_play(ip) if len(peers) > 0: server = self.__select_peer(peers, ip) if type == "hls": hls_url = "http://%s:%s/%s.m3u8"%(server, port, stream) hls_url = hls_url.replace(".m3u8.m3u8", ".m3u8") if format == "html": return SrsUtility().hls_html(hls_url) else: #return hls_url raise cherrypy.HTTPRedirect(hls_url) elif type == "rtmp": rtmp_url = "rtmp://%s:%s/%s?vhost=%s/%s"%(server, port, stream.split("/")[0], vhost, stream.split("/")[1]) if format == "html": html = "%s?server=%s&port=%s&vhost=%s&app=%s&stream=%s&autostart=true"%( "http://demo.chnvideo.com:8085/srs/trunk/research/players/srs_player.html", server, port, vhost, stream.split("/")[0], stream.split("/")[1]) #return html raise cherrypy.HTTPRedirect(html) return rtmp_url elif type == "gslb": return json.dumps({"code":Error.success, "data": { "edge":server, "client":ip, "peers":self.__json_dump_nodes(peers), "streams": { "cztv": { "hls": "http://demo.chnvideo.com:8085/api/v1/nodes?type=hls&format=html&origin=demo.chnvideo.com&port=8080&stream=live/rtmp_cztv01-sd", "rtmp": "http://demo.chnvideo.com:8085/api/v1/nodes?type=rtmp&format=html&origin=demo.chnvideo.com&vhost=android&port=1935&stream=live/rtmp_cztv01-sd" }, "livestream": { "hls": "http://demo.chnvideo.com:8085/api/v1/nodes?type=hls&format=html&origin=demo.chnvideo.com&port=8080&stream=live/livestream", "rtmp": "http://demo.chnvideo.com:8085/api/v1/nodes?type=rtmp&format=html&origin=demo.chnvideo.com&vhost=demo.srs.com&port=1935&stream=live/livestream" }, "apk": "http://demo.chnvideo.com/android.srs.apk" } }}) return json.dumps({"code":Error.success, "data": data}) finally: self.__lock.release() def PUT(self): enable_crossdomain() try: self.__lock.acquire() req = cherrypy.request.body.read() trace("put to nodes, req=%s"%(req)) try: json_req = json.loads(req) except Exception, ex: code = Error.system_parse_json trace("parse the request to json failed, req=%s, ex=%s, code=%s"%(req, ex, code)) return json.dumps({"code":code, "data": None}) id = str(json_req["id"]) node = self.__get_node(id) if node is None: code = Error.cdn_node_not_exists trace("cdn node not exists, req=%s, id=%s, code=%s"%(req, id, code)) return json.dumps({"code":code, "data": None}) node.heartbeat = time.time() node.srs_status = str(json_req["srs_status"]) node.ip = str(json_req["ip"]) if "origin" in json_req: node.origin = str(json_req["origin"]); node.public_ip = cherrypy.request.remote.ip # reset if restart. if node.srs_status != "running": node.clients = 0 self.__refresh_nodes() peers = self.__get_peers(node) peers_data = self.__json_dump_nodes(peers) res = json.dumps({"code":Error.success, "data": {"id":node.id, "peers":peers_data}}) trace(res) return res finally: self.__lock.release() def POST(self): enable_crossdomain() try: self.__lock.acquire() req = cherrypy.request.body.read() trace("post to nodes, req=%s"%(req)) try: json_req = json.loads(req) except Exception, ex: code = Error.system_parse_json trace("parse the request to json failed, req=%s, ex=%s, code=%s"%(req, ex, code)) return json.dumps({"code":code, "data": None}) node = CdnNode() node.ip = str(json_req["ip"]); node.os = str(json_req["os"]); if "origin" in json_req: node.origin = str(json_req["origin"]); node.srs_status = str(json_req["srs_status"]) self.__nodes.append(node) self.__refresh_nodes() peers = self.__get_peers(node) peers_data = self.__json_dump_nodes(peers) res = json.dumps({"code":Error.success, "data": {"id":node.id, "peers":peers_data}}) trace(res) return res finally: self.__lock.release() def OPTIONS(self, *args, **kwargs): enable_crossdomain() global_chat_id = os.getpid(); ''' the chat streams, public chat room. ''' class RESTChats(object): exposed = True global_id = 100 def __init__(self): # object fields: # id: an int value indicates the id of user. # username: a str indicates the user name. # url: a str indicates the url of user stream. # agent: a str indicates the agent of user. # join_date: a number indicates the join timestamp in seconds. # join_date_str: a str specifies the formated friendly time. # heatbeat: a number indicates the heartbeat timestamp in seconds. # vcodec: a dict indicates the video codec info. # acodec: a dict indicates the audio codec info. self.__chats = []; self.__chat_lock = threading.Lock(); # dead time in seconds, if exceed, remove the chat. self.__dead_time = 15; ''' get the rtmp url of chat object. None if overflow. ''' def get_url_by_index(self, index): index = int(index) if index is None or index >= len(self.__chats): return None; return self.__chats[index]["url"]; def GET(self): enable_crossdomain() try: self.__chat_lock.acquire(); chats = []; copy = self.__chats[:]; for chat in copy: if time.time() - chat["heartbeat"] > self.__dead_time: self.__chats.remove(chat); continue; chats.append({ "id": chat["id"], "username": chat["username"], "url": chat["url"], "join_date_str": chat["join_date_str"], "heartbeat": chat["heartbeat"], }); finally: self.__chat_lock.release(); return json.dumps({"code":0, "data": {"now": time.time(), "chats": chats}}) def POST(self): enable_crossdomain() req = cherrypy.request.body.read() chat = json.loads(req) global global_chat_id; chat["id"] = global_chat_id global_chat_id += 1 chat["join_date"] = time.time(); chat["heartbeat"] = time.time(); chat["join_date_str"] = time.strftime("%Y-%m-%d %H:%M:%S"); try: self.__chat_lock.acquire(); self.__chats.append(chat) finally: self.__chat_lock.release(); trace("create chat success, id=%s"%(chat["id"])) return json.dumps({"code":0, "data": chat["id"]}) def DELETE(self, id): enable_crossdomain() try: self.__chat_lock.acquire(); for chat in self.__chats: if str(id) != str(chat["id"]): continue self.__chats.remove(chat) trace("delete chat success, id=%s"%(id)) return json.dumps({"code":0, "data": None}) finally: self.__chat_lock.release(); raise cherrypy.HTTPError(405, "Not allowed.") def PUT(self, id): enable_crossdomain() try: self.__chat_lock.acquire(); for chat in self.__chats: if str(id) != str(chat["id"]): continue chat["heartbeat"] = time.time(); trace("heartbeat chat success, id=%s"%(id)) return json.dumps({"code":0, "data": None}) finally: self.__chat_lock.release(); raise cherrypy.HTTPError(405, "Not allowed.") def OPTIONS(self, *args, **kwargs): enable_crossdomain() # HTTP RESTful path. class Root(object): exposed = True def __init__(self): self.api = Api() def GET(self): enable_crossdomain(); return json.dumps({"code":Error.success, "urls":{"api":"the api root"}}) def OPTIONS(self, *args, **kwargs): enable_crossdomain(); # HTTP RESTful path. class Api(object): exposed = True def __init__(self): self.v1 = V1() def GET(self): enable_crossdomain(); return json.dumps({"code":Error.success, "urls": { "v1": "the api version 1.0" } }); def OPTIONS(self, *args, **kwargs): enable_crossdomain(); # HTTP RESTful path. to access as: # http://127.0.0.1:8085/api/v1/clients class V1(object): exposed = True def __init__(self): self.clients = RESTClients() self.streams = RESTStreams() self.sessions = RESTSessions() self.chats = RESTChats() self.servers = RESTServers() self.nodes = RESTNodes() def GET(self): enable_crossdomain(); return json.dumps({"code":Error.success, "urls":{ "clients": "for srs http callback, to handle the clients requests: connect/disconnect vhost/app.", "streams": "for srs http callback, to handle the streams requests: publish/unpublish stream.", "sessions": "for srs http callback, to handle the sessions requests: client play/stop stream", "chats": "for srs demo meeting, the chat streams, public chat room.", "nodes": { "summary": "for srs cdn node", "POST ip=node_ip&os=node_os": "register a new node", "GET": "get the active edge nodes", "GET type=gslb&origin=demo.chnvideo.com": "get the gslb edge ip", "GET type=hls&format=html&origin=demo.chnvideo.com&port=8080&stream=live/livestream": "get the play url, html for hls", "GET type=rtmp&format=html&origin=demo.chnvideo.com&vhost=demo.srs.com&port=1935&stream=live/livestream": "get the play url, for rtmp" }, "servers": { "summary": "for srs raspberry-pi and meeting demo", "GET": "get the current raspberry-pi servers info", "GET id=gslb&device_id=chnvideo-sales-arm": "get the gslb edge ip", "POST ip=node_ip&device_id=device_id": "the new raspberry-pi server info.", "GET id=ingest&action=play&stream=live/livestream": "play the ingest HLS stream on raspberry-pi", "GET id=ingest&action=rtmp&stream=live/livestream": "play the ingest RTMP stream on raspberry-pi", "GET id=ingest&action=hls&stream=live/livestream": "play the ingest HLS stream on raspberry-pi", "GET id=ingest&action=mgmt": "open the HTTP api url of raspberry-pi", "GET id=meeting": "redirect to local raspberry-pi meeting url(local ignored)", "GET id=meeting&local=false&index=0": "play the first(index=0) meeting HLS stream on demo.chnvideo.com(not local)", "GET id=meeting&local=true&index=0": "play the first(index=0) meeting HLS stream on local server(local x86/x64 server), warn: raspberry-pi donot support HLS meeting." } }}); def OPTIONS(self, *args, **kwargs): enable_crossdomain(); ''' main code start. ''' # donot support use this module as library. if __name__ != "__main__": raise Exception("embed not support") # check the user options if len(sys.argv) <= 1: print "SRS api callback server, Copyright (c) 2013-2014 winlin" print "Usage: python %s <port>"%(sys.argv[0]) print " port: the port to listen at." print "For example:" print " python %s 8085"%(sys.argv[0]) print "" print "See also: https://github.com/simple-rtmp-server/srs" sys.exit(1) # parse port from user options. port = int(sys.argv[1]) static_dir = os.path.abspath(os.path.join(os.path.dirname(sys.argv[0]), "static-dir")) trace("api server listen at port: %s, static_dir: %s"%(port, static_dir)) # cherrypy config. conf = { 'global': { 'server.shutdown_timeout': 1, 'server.socket_host': '0.0.0.0', 'server.socket_port': port, 'tools.encode.on': True, 'tools.staticdir.on': True, 'tools.encode.encoding': "utf-8", #'server.thread_pool': 2, # single thread server. }, '/': { 'tools.staticdir.dir': static_dir, 'tools.staticdir.index': "index.html", # for cherrypy RESTful api support 'request.dispatch': cherrypy.dispatch.MethodDispatcher() } } # start cherrypy web engine trace("start cherrypy server") root = Root() cherrypy.quickstart(root, '/', conf)
mit
jzt5132/scikit-learn
examples/linear_model/plot_polynomial_interpolation.py
250
1895
#!/usr/bin/env python """ ======================== Polynomial interpolation ======================== This example demonstrates how to approximate a function with a polynomial of degree n_degree by using ridge regression. Concretely, from n_samples 1d points, it suffices to build the Vandermonde matrix, which is n_samples x n_degree+1 and has the following form: [[1, x_1, x_1 ** 2, x_1 ** 3, ...], [1, x_2, x_2 ** 2, x_2 ** 3, ...], ...] Intuitively, this matrix can be interpreted as a matrix of pseudo features (the points raised to some power). The matrix is akin to (but different from) the matrix induced by a polynomial kernel. This example shows that you can do non-linear regression with a linear model, using a pipeline to add non-linear features. Kernel methods extend this idea and can induce very high (even infinite) dimensional feature spaces. """ print(__doc__) # Author: Mathieu Blondel # Jake Vanderplas # License: BSD 3 clause import numpy as np import matplotlib.pyplot as plt from sklearn.linear_model import Ridge from sklearn.preprocessing import PolynomialFeatures from sklearn.pipeline import make_pipeline def f(x): """ function to approximate by polynomial interpolation""" return x * np.sin(x) # generate points used to plot x_plot = np.linspace(0, 10, 100) # generate points and keep a subset of them x = np.linspace(0, 10, 100) rng = np.random.RandomState(0) rng.shuffle(x) x = np.sort(x[:20]) y = f(x) # create matrix versions of these arrays X = x[:, np.newaxis] X_plot = x_plot[:, np.newaxis] plt.plot(x_plot, f(x_plot), label="ground truth") plt.scatter(x, y, label="training points") for degree in [3, 4, 5]: model = make_pipeline(PolynomialFeatures(degree), Ridge()) model.fit(X, y) y_plot = model.predict(X_plot) plt.plot(x_plot, y_plot, label="degree %d" % degree) plt.legend(loc='lower left') plt.show()
bsd-3-clause
MohammedWasim/scikit-learn
examples/linear_model/plot_polynomial_interpolation.py
250
1895
#!/usr/bin/env python """ ======================== Polynomial interpolation ======================== This example demonstrates how to approximate a function with a polynomial of degree n_degree by using ridge regression. Concretely, from n_samples 1d points, it suffices to build the Vandermonde matrix, which is n_samples x n_degree+1 and has the following form: [[1, x_1, x_1 ** 2, x_1 ** 3, ...], [1, x_2, x_2 ** 2, x_2 ** 3, ...], ...] Intuitively, this matrix can be interpreted as a matrix of pseudo features (the points raised to some power). The matrix is akin to (but different from) the matrix induced by a polynomial kernel. This example shows that you can do non-linear regression with a linear model, using a pipeline to add non-linear features. Kernel methods extend this idea and can induce very high (even infinite) dimensional feature spaces. """ print(__doc__) # Author: Mathieu Blondel # Jake Vanderplas # License: BSD 3 clause import numpy as np import matplotlib.pyplot as plt from sklearn.linear_model import Ridge from sklearn.preprocessing import PolynomialFeatures from sklearn.pipeline import make_pipeline def f(x): """ function to approximate by polynomial interpolation""" return x * np.sin(x) # generate points used to plot x_plot = np.linspace(0, 10, 100) # generate points and keep a subset of them x = np.linspace(0, 10, 100) rng = np.random.RandomState(0) rng.shuffle(x) x = np.sort(x[:20]) y = f(x) # create matrix versions of these arrays X = x[:, np.newaxis] X_plot = x_plot[:, np.newaxis] plt.plot(x_plot, f(x_plot), label="ground truth") plt.scatter(x, y, label="training points") for degree in [3, 4, 5]: model = make_pipeline(PolynomialFeatures(degree), Ridge()) model.fit(X, y) y_plot = model.predict(X_plot) plt.plot(x_plot, y_plot, label="degree %d" % degree) plt.legend(loc='lower left') plt.show()
bsd-3-clause
MohammedWasim/scikit-learn
examples/plot_johnson_lindenstrauss_bound.py
126
7477
r""" ===================================================================== The Johnson-Lindenstrauss bound for embedding with random projections ===================================================================== The `Johnson-Lindenstrauss lemma`_ states that any high dimensional dataset can be randomly projected into a lower dimensional Euclidean space while controlling the distortion in the pairwise distances. .. _`Johnson-Lindenstrauss lemma`: http://en.wikipedia.org/wiki/Johnson%E2%80%93Lindenstrauss_lemma Theoretical bounds ================== The distortion introduced by a random projection `p` is asserted by the fact that `p` is defining an eps-embedding with good probability as defined by: .. math:: (1 - eps) \|u - v\|^2 < \|p(u) - p(v)\|^2 < (1 + eps) \|u - v\|^2 Where u and v are any rows taken from a dataset of shape [n_samples, n_features] and p is a projection by a random Gaussian N(0, 1) matrix with shape [n_components, n_features] (or a sparse Achlioptas matrix). The minimum number of components to guarantees the eps-embedding is given by: .. math:: n\_components >= 4 log(n\_samples) / (eps^2 / 2 - eps^3 / 3) The first plot shows that with an increasing number of samples ``n_samples``, the minimal number of dimensions ``n_components`` increased logarithmically in order to guarantee an ``eps``-embedding. The second plot shows that an increase of the admissible distortion ``eps`` allows to reduce drastically the minimal number of dimensions ``n_components`` for a given number of samples ``n_samples`` Empirical validation ==================== We validate the above bounds on the the digits dataset or on the 20 newsgroups text document (TF-IDF word frequencies) dataset: - for the digits dataset, some 8x8 gray level pixels data for 500 handwritten digits pictures are randomly projected to spaces for various larger number of dimensions ``n_components``. - for the 20 newsgroups dataset some 500 documents with 100k features in total are projected using a sparse random matrix to smaller euclidean spaces with various values for the target number of dimensions ``n_components``. The default dataset is the digits dataset. To run the example on the twenty newsgroups dataset, pass the --twenty-newsgroups command line argument to this script. For each value of ``n_components``, we plot: - 2D distribution of sample pairs with pairwise distances in original and projected spaces as x and y axis respectively. - 1D histogram of the ratio of those distances (projected / original). We can see that for low values of ``n_components`` the distribution is wide with many distorted pairs and a skewed distribution (due to the hard limit of zero ratio on the left as distances are always positives) while for larger values of n_components the distortion is controlled and the distances are well preserved by the random projection. Remarks ======= According to the JL lemma, projecting 500 samples without too much distortion will require at least several thousands dimensions, irrespective of the number of features of the original dataset. Hence using random projections on the digits dataset which only has 64 features in the input space does not make sense: it does not allow for dimensionality reduction in this case. On the twenty newsgroups on the other hand the dimensionality can be decreased from 56436 down to 10000 while reasonably preserving pairwise distances. """ print(__doc__) import sys from time import time import numpy as np import matplotlib.pyplot as plt from sklearn.random_projection import johnson_lindenstrauss_min_dim from sklearn.random_projection import SparseRandomProjection from sklearn.datasets import fetch_20newsgroups_vectorized from sklearn.datasets import load_digits from sklearn.metrics.pairwise import euclidean_distances # Part 1: plot the theoretical dependency between n_components_min and # n_samples # range of admissible distortions eps_range = np.linspace(0.1, 0.99, 5) colors = plt.cm.Blues(np.linspace(0.3, 1.0, len(eps_range))) # range of number of samples (observation) to embed n_samples_range = np.logspace(1, 9, 9) plt.figure() for eps, color in zip(eps_range, colors): min_n_components = johnson_lindenstrauss_min_dim(n_samples_range, eps=eps) plt.loglog(n_samples_range, min_n_components, color=color) plt.legend(["eps = %0.1f" % eps for eps in eps_range], loc="lower right") plt.xlabel("Number of observations to eps-embed") plt.ylabel("Minimum number of dimensions") plt.title("Johnson-Lindenstrauss bounds:\nn_samples vs n_components") # range of admissible distortions eps_range = np.linspace(0.01, 0.99, 100) # range of number of samples (observation) to embed n_samples_range = np.logspace(2, 6, 5) colors = plt.cm.Blues(np.linspace(0.3, 1.0, len(n_samples_range))) plt.figure() for n_samples, color in zip(n_samples_range, colors): min_n_components = johnson_lindenstrauss_min_dim(n_samples, eps=eps_range) plt.semilogy(eps_range, min_n_components, color=color) plt.legend(["n_samples = %d" % n for n in n_samples_range], loc="upper right") plt.xlabel("Distortion eps") plt.ylabel("Minimum number of dimensions") plt.title("Johnson-Lindenstrauss bounds:\nn_components vs eps") # Part 2: perform sparse random projection of some digits images which are # quite low dimensional and dense or documents of the 20 newsgroups dataset # which is both high dimensional and sparse if '--twenty-newsgroups' in sys.argv: # Need an internet connection hence not enabled by default data = fetch_20newsgroups_vectorized().data[:500] else: data = load_digits().data[:500] n_samples, n_features = data.shape print("Embedding %d samples with dim %d using various random projections" % (n_samples, n_features)) n_components_range = np.array([300, 1000, 10000]) dists = euclidean_distances(data, squared=True).ravel() # select only non-identical samples pairs nonzero = dists != 0 dists = dists[nonzero] for n_components in n_components_range: t0 = time() rp = SparseRandomProjection(n_components=n_components) projected_data = rp.fit_transform(data) print("Projected %d samples from %d to %d in %0.3fs" % (n_samples, n_features, n_components, time() - t0)) if hasattr(rp, 'components_'): n_bytes = rp.components_.data.nbytes n_bytes += rp.components_.indices.nbytes print("Random matrix with size: %0.3fMB" % (n_bytes / 1e6)) projected_dists = euclidean_distances( projected_data, squared=True).ravel()[nonzero] plt.figure() plt.hexbin(dists, projected_dists, gridsize=100, cmap=plt.cm.PuBu) plt.xlabel("Pairwise squared distances in original space") plt.ylabel("Pairwise squared distances in projected space") plt.title("Pairwise distances distribution for n_components=%d" % n_components) cb = plt.colorbar() cb.set_label('Sample pairs counts') rates = projected_dists / dists print("Mean distances rate: %0.2f (%0.2f)" % (np.mean(rates), np.std(rates))) plt.figure() plt.hist(rates, bins=50, normed=True, range=(0., 2.)) plt.xlabel("Squared distances rate: projected / original") plt.ylabel("Distribution of samples pairs") plt.title("Histogram of pairwise distance rates for n_components=%d" % n_components) # TODO: compute the expected value of eps and add them to the previous plot # as vertical lines / region plt.show()
bsd-3-clause
tomsilver/nupic
nupic/datafiles/extra/regression/makeDataset.py
9
5327
#! /usr/bin/env python # ---------------------------------------------------------------------- # Numenta Platform for Intelligent Computing (NuPIC) # Copyright (C) 2013, Numenta, Inc. Unless you have an agreement # with Numenta, Inc., for a separate license for this software code, the # following terms and conditions apply: # # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License version 3 as # published by the Free Software Foundation. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. # See the GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program. If not, see http://www.gnu.org/licenses. # # http://numenta.org/licenses/ # ---------------------------------------------------------------------- """ Generate artificial datasets """ import numpy from nupic.data.file import File ########################################################################### def scaleData(data, newScale=[0,100]): minVals = data.min(axis=0) maxVals = data.max(axis=0) data = (data-minVals)*(newScale[1]-newScale[0])/(maxVals-minVals) + newScale[0] return data ########################################################################### def generatePolyData(numDataPoints=100, coefficients=[1, 0], noiseLevel = 0.1, dataScale = [0,100],): xvals = numpy.random.random(numDataPoints) yvals = numpy.polyval(coefficients, xvals) + \ noiseLevel * numpy.random.randn(numDataPoints) data = numpy.vstack((yvals, xvals)).transpose() scaledData = scaleData(data, newScale=dataScale) return scaledData ########################################################################### def generateLinearData(numDataPoints=100, coefficients=[1, 1], noiseLevel = 0.1, dataScale = [0,100],): xvals = numpy.random.random((numDataPoints, len(coefficients))) yvals = (xvals * coefficients).sum(axis=1) + \ noiseLevel * numpy.random.randn(numDataPoints) data = numpy.hstack((yvals.reshape(-1,1), xvals)) scaledData = scaleData(data, newScale=dataScale) return scaledData ########################################################################### def _generateLinearModel(numTrainingRecords, numTestingRecords, coefficients=[1], noiseLevel=0.1, dataScale=[0,100]): """ """ data = generateLinearData(numDataPoints=numTrainingRecords+numTestingRecords, coefficients=coefficients, noiseLevel=noiseLevel, dataScale=dataScale,) trainData = data[:numTrainingRecords] testData = data[numTrainingRecords:] return trainData, testData ########################################################################### def _generateFile(filename, data): """ Parameters: ---------------------------------------------------------------- filename: name of .csv file to generate """ # Create the file print "Creating %s..." % (filename) numRecords, numFields = data.shape fields = [('field%d'%(i+1), 'float', '') for i in range(numFields)] outFile = File(filename, fields) for i in xrange(numRecords): outFile.write(data[i].tolist()) outFile.close() ######################################################################## def generate(model, filenameTrain, filenameTest, numTrainingRecords=10000, numTestingRecords=1000,): """ """ numpy.random.seed(41) # ==================================================================== # Generate the model if model == 'linear0': trainData, testData = _generateLinearModel(numTrainingRecords, numTestingRecords, coefficients=[1], noiseLevel=0.1) #import pylab #pylab.figure() #pylab.plot(trainData[:,1], trainData[:,0], 'b.') ##pylab.figure() #pylab.plot(testData[:,1], testData[:,0],'g.') #pylab.show() elif model == 'linear1': trainData, testData = _generateLinearModel(numTrainingRecords, numTestingRecords, coefficients=[1,1], noiseLevel=0.1) elif model == 'linear2': trainData, testData = _generateLinearModel(numTrainingRecords, numTestingRecords, coefficients=[1,-3]) else: raise RuntimeError("Unsupported model") # ==================================================================== # Generate the training and testing files _generateFile(filename=filenameTrain, data=trainData,) _generateFile(filename=filenameTest, data=testData,)
gpl-3.0
mlperf/training_results_v0.5
v0.5.0/nvidia/submission/code/single_stage_detector/pytorch/ssd300.py
1
7384
# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import torch import torch.nn as nn from base_model import L2Norm, ResNet18, ResNet34, ResNet50 from mlperf_compliance import mlperf_log from mlperf_logger import ssd_print from nhwc import resnet_nhwc from nhwc.conv import Conv2d_NHWC class SSD300(nn.Module): """ Build a SSD module to take 300x300 image input, and output 8732 per class bounding boxes vggt: pretrained vgg16 (partial) model label_num: number of classes (including background 0) """ def __init__(self, label_num, backbone='resnet34', use_nhwc=False, pad_input=False): super(SSD300, self).__init__() self.label_num = label_num self.use_nhwc = use_nhwc self.pad_input = pad_input if backbone == 'resnet18': self.model = ResNet18(self.use_nhwc, self.pad_input) out_channels = 256 out_size = 38 self.out_chan = [out_channels, 512, 512, 256, 256, 128] elif backbone == 'resnet34': self.model = ResNet34(self.use_nhwc, self.pad_input) ssd_print(key=mlperf_log.BACKBONE, value='resnet34') out_channels = 256 out_size = 38 self.out_chan = [out_channels, 512, 512, 256, 256, 256] ssd_print(key=mlperf_log.LOC_CONF_OUT_CHANNELS, value=self.out_chan) elif backbone == 'resnet50': self.model = ResNet50(self.use_nhwc, self.pad_input) out_channels = 1024 out_size = 38 self.l2norm4 = L2Norm() self.out_chan = [out_channels, 1024, 512, 512, 256, 256] else: print('Invalid backbone chosen') self._build_additional_features(out_size, self.out_chan) # after l2norm, conv7, conv8_2, conv9_2, conv10_2, conv11_2 # classifer 1, 2, 3, 4, 5 ,6 self.num_defaults = [4, 6, 6, 6, 4, 4] ssd_print(key=mlperf_log.NUM_DEFAULTS_PER_CELL, value=self.num_defaults) self.loc = [] self.conf = [] for nd, oc in zip(self.num_defaults, self.out_chan): self.loc.append(nn.Conv2d(oc, nd*4, kernel_size=3, padding=1)) self.conf.append(nn.Conv2d(oc, nd*label_num, kernel_size=3, padding=1)) self.loc = nn.ModuleList(self.loc) self.conf = nn.ModuleList(self.conf) # intitalize all weights self._init_weights() def _build_additional_features(self, input_size, input_channels): idx = 0 if input_size == 38: idx = 0 elif input_size == 19: idx = 1 elif input_size == 10: idx = 2 self.additional_blocks = [] if self.use_nhwc: conv_fn = Conv2d_NHWC else: conv_fn = nn.Conv2d # if input_size == 38: self.additional_blocks.append(nn.Sequential( conv_fn(input_channels[idx], 256, kernel_size=1), nn.ReLU(inplace=True), conv_fn(256, input_channels[idx+1], kernel_size=3, padding=1, stride=2), nn.ReLU(inplace=True), )) idx += 1 self.additional_blocks.append(nn.Sequential( conv_fn(input_channels[idx], 256, kernel_size=1), nn.ReLU(inplace=True), conv_fn(256, input_channels[idx+1], kernel_size=3, padding=1, stride=2), nn.ReLU(inplace=True), )) idx += 1 # conv9_1, conv9_2 self.additional_blocks.append(nn.Sequential( conv_fn(input_channels[idx], 128, kernel_size=1), nn.ReLU(inplace=True), conv_fn(128, input_channels[idx+1], kernel_size=3, padding=1, stride=2), nn.ReLU(inplace=True), )) idx += 1 # conv10_1, conv10_2 self.additional_blocks.append(nn.Sequential( conv_fn(input_channels[idx], 128, kernel_size=1), nn.ReLU(inplace=True), conv_fn(128, input_channels[idx+1], kernel_size=3), nn.ReLU(inplace=True), )) idx += 1 # Only necessary in VGG for now if input_size >= 19: # conv11_1, conv11_2 self.additional_blocks.append(nn.Sequential( conv_fn(input_channels[idx], 128, kernel_size=1), nn.ReLU(inplace=True), conv_fn(128, input_channels[idx+1], kernel_size=3), nn.ReLU(inplace=True), )) self.additional_blocks = nn.ModuleList(self.additional_blocks) def _init_weights(self): addn_blocks = [ *self.additional_blocks] layers = [ *self.loc, *self.conf] # Need to handle additional blocks differently in NHWC case due to xavier initialization for layer in addn_blocks: for param in layer.parameters(): if param.dim() > 1: if self.use_nhwc: # xavier_uniform relies on fan-in/-out, so need to use NCHW here to get # correct values (K, R) instead of the correct (K, C) nn.init.xavier_uniform_(param.permute(0, 3, 1, 2).contiguous()) # Now permute correctly-initialized param back to NHWC param = param.permute(0, 2, 3, 1).contiguous() else: nn.init.xavier_uniform_(param) for layer in layers: for param in layer.parameters(): if param.dim() > 1: nn.init.xavier_uniform_(param) # Shape the classifier to the view of bboxes def bbox_view(self, src, loc, conf): ret = [] for s, l, c in zip(src, loc, conf): if self.use_nhwc: s = s.permute(0, 3, 1, 2).contiguous() ret.append((l(s).view(s.size(0), 4, -1), c(s).view(s.size(0), self.label_num, -1))) locs, confs = list(zip(*ret)) locs, confs = torch.cat(locs, 2).contiguous(), torch.cat(confs, 2).contiguous() return locs, confs def forward(self, data): layers = self.model(data) # last result from network goes into additional blocks x = layers[-1] # If necessary, transpose back to NCHW additional_results = [] for i, l in enumerate(self.additional_blocks): x = l(x) additional_results.append(x) # do we need the l2norm on the first result? src = [*layers, *additional_results] # Feature Map 38x38x4, 19x19x6, 10x10x6, 5x5x6, 3x3x4, 1x1x4 locs, confs = self.bbox_view(src, self.loc, self.conf) # For SSD 300, shall return nbatch x 8732 x {nlabels, nlocs} results return locs, confs
apache-2.0