Yelpdata_663 / Yelpdata_663.py
Johnnyeee's picture
My commit
1baf50d
raw
history blame
6.82 kB
# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
#
# 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.
# TODO: Address all TODOs and remove all explanatory comments
"""TODO: Add a description here."""
import csv
import json
import os
from typing import List
import datasets
import logging
# TODO: Add BibTeX citation
# Find for instance the citation on arxiv or on the dataset repo/website
_CITATION = """\
@InProceedings{huggingface:dataset,
title = {A great new dataset},
author={huggingface, Inc.
},
year={2020}
}
"""
# TODO: Add description of the dataset here
# You can copy an official description
_DESCRIPTION = """\
This new dataset is designed to solve this great NLP task and is crafted with a lot of care.
"""
# TODO: Add a link to an official homepage for the dataset here
_HOMEPAGE = "https://www.yelp.com/dataset/download"
# TODO: Add the licence for the dataset here if you can find it
_LICENSE = ""
import json
import datasets
from random import random
class YelpDataset(datasets.GeneratorBasedBuilder):
"""Yelp Dataset focusing on restaurant reviews and business information."""
VERSION = datasets.Version("1.1.0")
BUILDER_CONFIGS = [
datasets.BuilderConfig(name="restaurants", version=VERSION, description="This part of the dataset covers a wide range of restaurants"),
]
DEFAULT_CONFIG_NAME = "restaurants"
_URL = "https://yelpdata.s3.us-west-2.amazonaws.com/"
_URLS = {
"business": _URL + "yelp_academic_dataset_business.json",
"review": _URL + "yelp_academic_dataset_review.json",
}
def _info(self):
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=datasets.Features({
"business_id": datasets.Value("string"),
"name": datasets.Value("string"),
"address": datasets.Value("string"),
"city": datasets.Value("string"),
"state": datasets.Value("string"),
"postal_code": datasets.Value("string"),
"latitude": datasets.Value("float"),
"longitude": datasets.Value("float"),
"stars_x": datasets.Value("float"),
"review_count": datasets.Value("float"),
"is_open": datasets.Value("float"),
"categories": datasets.Value("string"),
"hours": datasets.Value("string"),
"review_id": datasets.Value("string"),
"user_id": datasets.Value("string"),
"stars_y": datasets.Value("float"),
"useful": datasets.Value("float"),
"funny": datasets.Value("float"),
"cool": datasets.Value("float"),
"text": datasets.Value("string"),
"date": datasets.Value("string"),
"attributes": datasets.Value("string"),
}),
supervised_keys=None,
homepage="https://www.yelp.com/dataset/download",
citation=_CITATION,
license=_LICENSE,
)
def _split_generators(self, dl_manager: datasets.DownloadManager):
"""Returns SplitGenerators."""
downloaded_files = dl_manager.download_and_extract(self._URLS)
return [
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"files": downloaded_files, "split": "train"}),
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"files": downloaded_files, "split": "test"}),
]
def _generate_examples(self, files, split):
"""Yields examples as (key, example) tuples."""
business_path, review_path = files["business"], files["review"]
# Load businesses and filter for restaurants
with open(business_path, encoding="utf-8") as f:
businesses = {}
for line in f:
business = json.loads(line)
# Check if 'categories' is not None and contains "Restaurants"
if business.get("categories") and "Restaurants" in business["categories"]:
businesses[business['business_id']] = business
# Generate examples with an attempted 80/20 split for train/test
with open(review_path, encoding="utf-8") as f:
for line in f:
review = json.loads(line)
business_id = review['business_id']
if business_id in businesses:
business = businesses[business_id]
example = {
"business_id": business['business_id'],
"name": business.get("name", ""),
"address": business.get("address", ""),
"city": business.get("city", ""),
"state": business.get("state", ""),
"postal_code": business.get("postal_code", ""),
"latitude": business.get("latitude", None),
"longitude": business.get("longitude", None),
"stars_x": business.get("stars", None),
"review_count": business.get("review_count", None),
"is_open": business.get("is_open", None),
"categories": business.get("categories", ""),
"hours": json.dumps(business.get("hours", {})), # Storing hours as a JSON string
"review_id": review.get("review_id", ""),
"user_id": review.get("user_id", ""),
"stars_y": review.get("stars", None),
"useful": review.get("useful", None),
"funny": review.get("funny", None),
"cool": review.get("cool", None),
"text": review.get("text", ""),
"date": review.get("date", ""),
"attributes": json.dumps(business.get("attributes", {})), # Storing attributes as a JSON string
}
# Randomly assign to split based on an 80/20 ratio
if (split == 'train' and random() < 0.8) or (split == 'test' and random() >= 0.8):
yield review['review_id'], example