system HF staff commited on
Commit
8a92be9
1 Parent(s): f843faa

Update files from the datasets library (from 1.6.1)

Browse files

Release notes: https://github.com/huggingface/datasets/releases/tag/1.6.1

Files changed (1) hide show
  1. README.md +228 -10
README.md CHANGED
@@ -1,4 +1,68 @@
1
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  ---
3
 
4
  # Dataset Card for "glue"
@@ -39,9 +103,55 @@
39
 
40
  ### Dataset Summary
41
 
42
- GLUE, the General Language Understanding Evaluation benchmark
43
- (https://gluebenchmark.com/) is a collection of resources for training,
44
- evaluating, and analyzing natural language understanding systems.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
45
 
46
  ### Supported Tasks
47
 
@@ -65,7 +175,12 @@ We show detailed information for up to 5 configurations of the dataset.
65
 
66
  An example of 'test' looks as follows.
67
  ```
68
-
 
 
 
 
 
69
  ```
70
 
71
  #### cola
@@ -76,7 +191,11 @@ An example of 'test' looks as follows.
76
 
77
  An example of 'train' looks as follows.
78
  ```
79
-
 
 
 
 
80
  ```
81
 
82
  #### mnli
@@ -87,7 +206,12 @@ An example of 'train' looks as follows.
87
 
88
  An example of 'train' looks as follows.
89
  ```
90
-
 
 
 
 
 
91
  ```
92
 
93
  #### mnli_matched
@@ -96,9 +220,14 @@ An example of 'train' looks as follows.
96
  - **Size of the generated dataset:** 3.52 MB
97
  - **Total amount of disk used:** 301.82 MB
98
 
99
- An example of 'validation' looks as follows.
100
  ```
101
-
 
 
 
 
 
102
  ```
103
 
104
  #### mnli_mismatched
@@ -107,11 +236,44 @@ An example of 'validation' looks as follows.
107
  - **Size of the generated dataset:** 3.73 MB
108
  - **Total amount of disk used:** 302.02 MB
109
 
110
- An example of 'validation' looks as follows.
111
  ```
112
-
 
 
 
 
 
113
  ```
114
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
115
  ### Data Fields
116
 
117
  The data fields are the same among all splits.
@@ -145,6 +307,34 @@ The data fields are the same among all splits.
145
  - `label`: a classification label, with possible values including `entailment` (0), `neutral` (1), `contradiction` (2).
146
  - `idx`: a `int32` feature.
147
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
148
  ### Data Splits Sample Size
149
 
150
  #### ax
@@ -177,6 +367,34 @@ The data fields are the same among all splits.
177
  |---------------|---------:|---:|
178
  |mnli_mismatched| 9832|9847|
179
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
180
  ## Dataset Creation
181
 
182
  ### Curation Rationale
 
1
  ---
2
+ annotations_creators:
3
+ - unknown
4
+ language_creators:
5
+ - unknown
6
+ languages:
7
+ - en
8
+ licenses:
9
+ - cc-by-4-0
10
+ multilinguality:
11
+ - monolingual
12
+ size_categories:
13
+ - 10K<n<100K
14
+ source_datasets:
15
+ - unknown
16
+ task_categories:
17
+ ax:
18
+ - text-classification
19
+ cola:
20
+ - text-classification
21
+ mnli:
22
+ - text-classification
23
+ mnli_matched:
24
+ - text-classification
25
+ mnli_mismatched:
26
+ - text-classification
27
+ mrpc:
28
+ - text-classification
29
+ qnli:
30
+ - text-classification
31
+ qqp:
32
+ - text-classification
33
+ rte:
34
+ - text-classification
35
+ sst2:
36
+ - text-classification
37
+ stsb:
38
+ - text-scoring
39
+ wnli:
40
+ - text-classification
41
+ task_ids:
42
+ ax:
43
+ - natural-language-inference
44
+ cola:
45
+ - acceptability-classification
46
+ mnli:
47
+ - natural-language-inference
48
+ mnli_matched:
49
+ - natural-language-inference
50
+ mnli_mismatched:
51
+ - natural-language-inference
52
+ mrpc:
53
+ - text-classification-other-paraphrase-identification
54
+ qnli:
55
+ - text-classification-other-qa-nli
56
+ qqp:
57
+ - text-classification-other-paraphrase-identification
58
+ rte:
59
+ - natural-language-inference
60
+ sst2:
61
+ - sentiment-classification
62
+ stsb:
63
+ - semantic-similarity-scoring
64
+ wnli:
65
+ - text-classification-other-coreference-nli
66
  ---
67
 
68
  # Dataset Card for "glue"
 
103
 
104
  ### Dataset Summary
105
 
106
+ GLUE, the General Language Understanding Evaluation benchmark (https://gluebenchmark.com/) is a collection of resources for training, evaluating, and analyzing natural language understanding systems.
107
+
108
+ #### ax
109
+
110
+ A manually-curated evaluation dataset for fine-grained analysis of system performance on a broad range of linguistic phenomena. This dataset evaluates sentence understanding through Natural Language Inference (NLI) problems. Use a model trained on MulitNLI to produce predictions for this dataset.
111
+
112
+ #### cola
113
+
114
+ The Corpus of Linguistic Acceptability consists of English acceptability judgments drawn from books and journal articles on linguistic theory. Each example is a sequence of words annotated with whether it is a grammatical English sentence.
115
+
116
+ #### mnli
117
+
118
+ The Multi-Genre Natural Language Inference Corpus is a crowdsourced collection of sentence pairs with textual entailment annotations. Given a premise sentence and a hypothesis sentence, the task is to predict whether the premise entails the hypothesis (entailment), contradicts the hypothesis (contradiction), or neither (neutral). The premise sentences are gathered from ten different sources, including transcribed speech, fiction, and government reports. The authors of the benchmark use the standard test set, for which they obtained private labels from the RTE authors, and evaluate on both the matched (in-domain) and mismatched (cross-domain) section. They also uses and recommend the SNLI corpus as 550k examples of auxiliary training data.
119
+
120
+ #### mnli_matched
121
+
122
+ The matched validation and test splits from MNLI. See the "mnli" BuilderConfig for additional information.
123
+
124
+ #### mnli_mismatched
125
+
126
+ The mismatched validation and test splits from MNLI. See the "mnli" BuilderConfig for additional information.
127
+
128
+ #### mrpc
129
+
130
+ The Microsoft Research Paraphrase Corpus (Dolan & Brockett, 2005) is a corpus of sentence pairs automatically extracted from online news sources, with human annotations for whether the sentences in the pair are semantically equivalent.
131
+
132
+ #### qnli
133
+
134
+ The Stanford Question Answering Dataset is a question-answering dataset consisting of question-paragraph pairs, where one of the sentences in the paragraph (drawn from Wikipedia) contains the answer to the corresponding question (written by an annotator). The authors of the benchmark convert the task into sentence pair classification by forming a pair between each question and each sentence in the corresponding context, and filtering out pairs with low lexical overlap between the question and the context sentence. The task is to determine whether the context sentence contains the answer to the question. This modified version of the original task removes the requirement that the model select the exact answer, but also removes the simplifying assumptions that the answer is always present in the input and that lexical overlap is a reliable cue.
135
+
136
+ #### qqp
137
+
138
+ The Quora Question Pairs2 dataset is a collection of question pairs from the community question-answering website Quora. The task is to determine whether a pair of questions are semantically equivalent.
139
+
140
+ #### rte
141
+
142
+ The Recognizing Textual Entailment (RTE) datasets come from a series of annual textual entailment challenges. The authors of the benchmark combined the data from RTE1 (Dagan et al., 2006), RTE2 (Bar Haim et al., 2006), RTE3 (Giampiccolo et al., 2007), and RTE5 (Bentivogli et al., 2009). Examples are constructed based on news and Wikipedia text. The authors of the benchmark convert all datasets to a two-class split, where for three-class datasets they collapse neutral and contradiction into not entailment, for consistency.
143
+
144
+ #### sst2
145
+
146
+ The Stanford Sentiment Treebank consists of sentences from movie reviews and human annotations of their sentiment. The task is to predict the sentiment of a given sentence. It uses the two-way (positive/negative) class split, with only sentence-level labels.
147
+
148
+ #### stsb
149
+
150
+ The Semantic Textual Similarity Benchmark (Cer et al., 2017) is a collection of sentence pairs drawn from news headlines, video and image captions, and natural language inference data. Each pair is human-annotated with a similarity score from 1 to 5.
151
+
152
+ #### wnli
153
+
154
+ The Winograd Schema Challenge (Levesque et al., 2011) is a reading comprehension task in which a system must read a sentence with a pronoun and select the referent of that pronoun from a list of choices. The examples are manually constructed to foil simple statistical methods: Each one is contingent on contextual information provided by a single word or phrase in the sentence. To convert the problem into sentence pair classification, the authors of the benchmark construct sentence pairs by replacing the ambiguous pronoun with each possible referent. The task is to predict if the sentence with the pronoun substituted is entailed by the original sentence. They use a small evaluation set consisting of new examples derived from fiction books that was shared privately by the authors of the original corpus. While the included training set is balanced between two classes, the test set is imbalanced between them (65% not entailment). Also, due to a data quirk, the development set is adversarial: hypotheses are sometimes shared between training and development examples, so if a model memorizes the training examples, they will predict the wrong label on corresponding development set example. As with QNLI, each example is evaluated separately, so there is not a systematic correspondence between a model's score on this task and its score on the unconverted original task. The authors of the benchmark call converted dataset WNLI (Winograd NLI).
155
 
156
  ### Supported Tasks
157
 
 
175
 
176
  An example of 'test' looks as follows.
177
  ```
178
+ {
179
+ "premise": "The cat sat on the mat.",
180
+ "hypothesis": "The cat did not sit on the mat.",
181
+ "label": -1,
182
+ "idx: 0
183
+ }
184
  ```
185
 
186
  #### cola
 
191
 
192
  An example of 'train' looks as follows.
193
  ```
194
+ {
195
+ "sentence": "Our friends won't buy this analysis, let alone the next one we propose.",
196
+ "label": 1,
197
+ "id": 0
198
+ }
199
  ```
200
 
201
  #### mnli
 
206
 
207
  An example of 'train' looks as follows.
208
  ```
209
+ {
210
+ "premise": "Conceptually cream skimming has two basic dimensions - product and geography.",
211
+ "hypothesis": "Product and geography are what make cream skimming work.",
212
+ "label": 1,
213
+ "idx": 0
214
+ }
215
  ```
216
 
217
  #### mnli_matched
 
220
  - **Size of the generated dataset:** 3.52 MB
221
  - **Total amount of disk used:** 301.82 MB
222
 
223
+ An example of 'test' looks as follows.
224
  ```
225
+ {
226
+ "premise": "Hierbas, ans seco, ans dulce, and frigola are just a few names worth keeping a look-out for.",
227
+ "hypothesis": "Hierbas is a name worth looking out for.",
228
+ "label": -1,
229
+ "idx": 0
230
+ }
231
  ```
232
 
233
  #### mnli_mismatched
 
236
  - **Size of the generated dataset:** 3.73 MB
237
  - **Total amount of disk used:** 302.02 MB
238
 
239
+ An example of 'test' looks as follows.
240
  ```
241
+ {
242
+ "premise": "What have you decided, what are you going to do?",
243
+ "hypothesis": "So what's your decision?,
244
+ "label": -1,
245
+ "idx": 0
246
+ }
247
  ```
248
 
249
+ #### mrpc
250
+
251
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
252
+
253
+ #### qnli
254
+
255
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
256
+
257
+ #### qqp
258
+
259
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
260
+
261
+ #### rte
262
+
263
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
264
+
265
+ #### sst2
266
+
267
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
268
+
269
+ #### stsb
270
+
271
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
272
+
273
+ #### wnli
274
+
275
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
276
+
277
  ### Data Fields
278
 
279
  The data fields are the same among all splits.
 
307
  - `label`: a classification label, with possible values including `entailment` (0), `neutral` (1), `contradiction` (2).
308
  - `idx`: a `int32` feature.
309
 
310
+ #### mrpc
311
+
312
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
313
+
314
+ #### qnli
315
+
316
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
317
+
318
+ #### qqp
319
+
320
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
321
+
322
+ #### rte
323
+
324
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
325
+
326
+ #### sst2
327
+
328
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
329
+
330
+ #### stsb
331
+
332
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
333
+
334
+ #### wnli
335
+
336
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
337
+
338
  ### Data Splits Sample Size
339
 
340
  #### ax
 
367
  |---------------|---------:|---:|
368
  |mnli_mismatched| 9832|9847|
369
 
370
+ #### mrpc
371
+
372
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
373
+
374
+ #### qnli
375
+
376
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
377
+
378
+ #### qqp
379
+
380
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
381
+
382
+ #### rte
383
+
384
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
385
+
386
+ #### sst2
387
+
388
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
389
+
390
+ #### stsb
391
+
392
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
393
+
394
+ #### wnli
395
+
396
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
397
+
398
  ## Dataset Creation
399
 
400
  ### Curation Rationale