Calculate metric using evaluate instead of datasets
#5
by
Vigneshwaran
- opened
- metrics/scrolls.py +6 -5
metrics/scrolls.py
CHANGED
@@ -3,6 +3,7 @@
|
|
3 |
from collections import defaultdict
|
4 |
from copy import deepcopy
|
5 |
import datasets
|
|
|
6 |
|
7 |
# fmt: off
|
8 |
from .rouge import compute_rouge, postprocess_text as rouge_postprocess_text # From: https://huggingface.co/datasets/tau/scrolls/raw/main/metrics/rouge.py
|
@@ -58,20 +59,20 @@ Examples:
|
|
58 |
predictions = ["exact match example", "hello there", "general kenobi"] # List[str]
|
59 |
references = [["exact match example"], ["hello", "hi there"], ["commander kenobi"]] # List[List[str]]
|
60 |
|
61 |
-
>>> scrolls_metric =
|
62 |
>>> results = scrolls_metric.compute(predictions=predictions, references=references)
|
63 |
>>> print(results)
|
64 |
{'rouge/rouge1': 72.2222, 'rouge/rouge2': 33.3333, 'rouge/rougeL': 72.2222, 'rouge/rougeLsum': 72.2222, 'rouge/geometric_mean': 55.8136,
|
65 |
'num_predicted': 3, 'mean_prediction_length_characters': 14.6667, 'scrolls_score': 55.8136,
|
66 |
'display_keys': ['rouge/rouge1', 'rouge/rouge2', 'rouge/rougeL'], 'display': [72.2222, 33.3333, 72.2222]}
|
67 |
|
68 |
-
>>> scrolls_metric =
|
69 |
>>> results = scrolls_metric.compute(predictions=predictions, references=references)
|
70 |
>>> print(results)
|
71 |
{'exact_match': 33.3333, 'num_predicted': 3, 'mean_prediction_length_characters': 14.6667, 'scrolls_score': 33.3333,
|
72 |
'display_keys': ['exact_match'], 'display': [33.3333]}
|
73 |
|
74 |
-
>>> scrolls_metric =
|
75 |
>>> results = scrolls_metric.compute(predictions=predictions, references=references)
|
76 |
>>> print(results)
|
77 |
{'f1': 72.2222, 'num_predicted': 3, 'mean_prediction_length_characters': 14.6667, 'scrolls_score': 72.2222,
|
@@ -123,7 +124,7 @@ DATASET_TO_METRICS = {
|
|
123 |
|
124 |
|
125 |
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION, _KWARGS_DESCRIPTION)
|
126 |
-
class Scrolls(
|
127 |
def __init__(self, *args, **kwargs):
|
128 |
super().__init__(*args, **kwargs)
|
129 |
|
@@ -173,7 +174,7 @@ class Scrolls(datasets.Metric):
|
|
173 |
self._metrics_to_compute = DATASET_TO_METRICS[self.config_name]["metrics_to_compute"]
|
174 |
|
175 |
def _info(self):
|
176 |
-
return
|
177 |
description=_DESCRIPTION,
|
178 |
citation=_CITATION,
|
179 |
inputs_description=_KWARGS_DESCRIPTION,
|
|
|
3 |
from collections import defaultdict
|
4 |
from copy import deepcopy
|
5 |
import datasets
|
6 |
+
import evaluate
|
7 |
|
8 |
# fmt: off
|
9 |
from .rouge import compute_rouge, postprocess_text as rouge_postprocess_text # From: https://huggingface.co/datasets/tau/scrolls/raw/main/metrics/rouge.py
|
|
|
59 |
predictions = ["exact match example", "hello there", "general kenobi"] # List[str]
|
60 |
references = [["exact match example"], ["hello", "hi there"], ["commander kenobi"]] # List[List[str]]
|
61 |
|
62 |
+
>>> scrolls_metric = evaluate.load(scrolls_metric_path, 'gov_report') # 'gov_report' or any of ["qmsum", "summ_screen_fd"]
|
63 |
>>> results = scrolls_metric.compute(predictions=predictions, references=references)
|
64 |
>>> print(results)
|
65 |
{'rouge/rouge1': 72.2222, 'rouge/rouge2': 33.3333, 'rouge/rougeL': 72.2222, 'rouge/rougeLsum': 72.2222, 'rouge/geometric_mean': 55.8136,
|
66 |
'num_predicted': 3, 'mean_prediction_length_characters': 14.6667, 'scrolls_score': 55.8136,
|
67 |
'display_keys': ['rouge/rouge1', 'rouge/rouge2', 'rouge/rougeL'], 'display': [72.2222, 33.3333, 72.2222]}
|
68 |
|
69 |
+
>>> scrolls_metric = evaluate.load(scrolls_metric_path, 'contract_nli') # 'contract_nli' or "quality"
|
70 |
>>> results = scrolls_metric.compute(predictions=predictions, references=references)
|
71 |
>>> print(results)
|
72 |
{'exact_match': 33.3333, 'num_predicted': 3, 'mean_prediction_length_characters': 14.6667, 'scrolls_score': 33.3333,
|
73 |
'display_keys': ['exact_match'], 'display': [33.3333]}
|
74 |
|
75 |
+
>>> scrolls_metric = evaluate.load(scrolls_metric_path, 'narrative_qa') # 'narrative_qa' or "qasper"
|
76 |
>>> results = scrolls_metric.compute(predictions=predictions, references=references)
|
77 |
>>> print(results)
|
78 |
{'f1': 72.2222, 'num_predicted': 3, 'mean_prediction_length_characters': 14.6667, 'scrolls_score': 72.2222,
|
|
|
124 |
|
125 |
|
126 |
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION, _KWARGS_DESCRIPTION)
|
127 |
+
class Scrolls(evaluate.Metric):
|
128 |
def __init__(self, *args, **kwargs):
|
129 |
super().__init__(*args, **kwargs)
|
130 |
|
|
|
174 |
self._metrics_to_compute = DATASET_TO_METRICS[self.config_name]["metrics_to_compute"]
|
175 |
|
176 |
def _info(self):
|
177 |
+
return evaluate.MetricInfo(
|
178 |
description=_DESCRIPTION,
|
179 |
citation=_CITATION,
|
180 |
inputs_description=_KWARGS_DESCRIPTION,
|