Hugging Face's logo

Datasets:
facebook
/
xnli

Modalities:
Text
Formats:
parquet
Languages:
Arabic
Bulgarian
German
Size:
1M - 10M
Libraries:
Datasets
Dask
Dataset card Viewer Files Community
10
albertvillanova HF staff commited on
Commit
d0b69a6
•
1 Parent(s): 83ecd0d

Delete legacy dataset_infos.json

Browse files
Files changed (1) hide show
  1. dataset_infos.json +0 -947
dataset_infos.json DELETED
@@ -1,947 +0,0 @@
1
- {
2
- "ar": {
3
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
4
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
5
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
6
- "license": "",
7
- "features": {
8
- "premise": {
9
- "dtype": "string",
10
- "_type": "Value"
11
- },
12
- "hypothesis": {
13
- "dtype": "string",
14
- "_type": "Value"
15
- },
16
- "label": {
17
- "names": [
18
- "entailment",
19
- "neutral",
20
- "contradiction"
21
- ],
22
- "_type": "ClassLabel"
23
- }
24
- },
25
- "builder_name": "xnli",
26
- "dataset_name": "xnli",
27
- "config_name": "ar",
28
- "version": {
29
- "version_str": "1.1.0",
30
- "description": "",
31
- "major": 1,
32
- "minor": 1,
33
- "patch": 0
34
- },
35
- "splits": {
36
- "train": {
37
- "name": "train",
38
- "num_bytes": 107399614,
39
- "num_examples": 392702,
40
- "dataset_name": null
41
- },
42
- "test": {
43
- "name": "test",
44
- "num_bytes": 1294553,
45
- "num_examples": 5010,
46
- "dataset_name": null
47
- },
48
- "validation": {
49
- "name": "validation",
50
- "num_bytes": 633001,
51
- "num_examples": 2490,
52
- "dataset_name": null
53
- }
54
- },
55
- "download_size": 59215902,
56
- "dataset_size": 109327168,
57
- "size_in_bytes": 168543070
58
- },
59
- "bg": {
60
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
61
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
62
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
63
- "license": "",
64
- "features": {
65
- "premise": {
66
- "dtype": "string",
67
- "_type": "Value"
68
- },
69
- "hypothesis": {
70
- "dtype": "string",
71
- "_type": "Value"
72
- },
73
- "label": {
74
- "names": [
75
- "entailment",
76
- "neutral",
77
- "contradiction"
78
- ],
79
- "_type": "ClassLabel"
80
- }
81
- },
82
- "builder_name": "xnli",
83
- "dataset_name": "xnli",
84
- "config_name": "bg",
85
- "version": {
86
- "version_str": "1.1.0",
87
- "description": "",
88
- "major": 1,
89
- "minor": 1,
90
- "patch": 0
91
- },
92
- "splits": {
93
- "train": {
94
- "name": "train",
95
- "num_bytes": 125973225,
96
- "num_examples": 392702,
97
- "dataset_name": null
98
- },
99
- "test": {
100
- "name": "test",
101
- "num_bytes": 1573034,
102
- "num_examples": 5010,
103
- "dataset_name": null
104
- },
105
- "validation": {
106
- "name": "validation",
107
- "num_bytes": 774061,
108
- "num_examples": 2490,
109
- "dataset_name": null
110
- }
111
- },
112
- "download_size": 66117878,
113
- "dataset_size": 128320320,
114
- "size_in_bytes": 194438198
115
- },
116
- "de": {
117
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
118
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
119
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
120
- "license": "",
121
- "features": {
122
- "premise": {
123
- "dtype": "string",
124
- "_type": "Value"
125
- },
126
- "hypothesis": {
127
- "dtype": "string",
128
- "_type": "Value"
129
- },
130
- "label": {
131
- "names": [
132
- "entailment",
133
- "neutral",
134
- "contradiction"
135
- ],
136
- "_type": "ClassLabel"
137
- }
138
- },
139
- "builder_name": "xnli",
140
- "dataset_name": "xnli",
141
- "config_name": "de",
142
- "version": {
143
- "version_str": "1.1.0",
144
- "description": "",
145
- "major": 1,
146
- "minor": 1,
147
- "patch": 0
148
- },
149
- "splits": {
150
- "train": {
151
- "name": "train",
152
- "num_bytes": 84684140,
153
- "num_examples": 392702,
154
- "dataset_name": null
155
- },
156
- "test": {
157
- "name": "test",
158
- "num_bytes": 996488,
159
- "num_examples": 5010,
160
- "dataset_name": null
161
- },
162
- "validation": {
163
- "name": "validation",
164
- "num_bytes": 494604,
165
- "num_examples": 2490,
166
- "dataset_name": null
167
- }
168
- },
169
- "download_size": 55973883,
170
- "dataset_size": 86175232,
171
- "size_in_bytes": 142149115
172
- },
173
- "el": {
174
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
175
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
176
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
177
- "license": "",
178
- "features": {
179
- "premise": {
180
- "dtype": "string",
181
- "_type": "Value"
182
- },
183
- "hypothesis": {
184
- "dtype": "string",
185
- "_type": "Value"
186
- },
187
- "label": {
188
- "names": [
189
- "entailment",
190
- "neutral",
191
- "contradiction"
192
- ],
193
- "_type": "ClassLabel"
194
- }
195
- },
196
- "builder_name": "xnli",
197
- "dataset_name": "xnli",
198
- "config_name": "el",
199
- "version": {
200
- "version_str": "1.1.0",
201
- "description": "",
202
- "major": 1,
203
- "minor": 1,
204
- "patch": 0
205
- },
206
- "splits": {
207
- "train": {
208
- "name": "train",
209
- "num_bytes": 139753358,
210
- "num_examples": 392702,
211
- "dataset_name": null
212
- },
213
- "test": {
214
- "name": "test",
215
- "num_bytes": 1704785,
216
- "num_examples": 5010,
217
- "dataset_name": null
218
- },
219
- "validation": {
220
- "name": "validation",
221
- "num_bytes": 841226,
222
- "num_examples": 2490,
223
- "dataset_name": null
224
- }
225
- },
226
- "download_size": 74551247,
227
- "dataset_size": 142299369,
228
- "size_in_bytes": 216850616
229
- },
230
- "en": {
231
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
232
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
233
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
234
- "license": "",
235
- "features": {
236
- "premise": {
237
- "dtype": "string",
238
- "_type": "Value"
239
- },
240
- "hypothesis": {
241
- "dtype": "string",
242
- "_type": "Value"
243
- },
244
- "label": {
245
- "names": [
246
- "entailment",
247
- "neutral",
248
- "contradiction"
249
- ],
250
- "_type": "ClassLabel"
251
- }
252
- },
253
- "builder_name": "xnli",
254
- "dataset_name": "xnli",
255
- "config_name": "en",
256
- "version": {
257
- "version_str": "1.1.0",
258
- "description": "",
259
- "major": 1,
260
- "minor": 1,
261
- "patch": 0
262
- },
263
- "splits": {
264
- "train": {
265
- "name": "train",
266
- "num_bytes": 74444026,
267
- "num_examples": 392702,
268
- "dataset_name": null
269
- },
270
- "test": {
271
- "name": "test",
272
- "num_bytes": 875134,
273
- "num_examples": 5010,
274
- "dataset_name": null
275
- },
276
- "validation": {
277
- "name": "validation",
278
- "num_bytes": 433463,
279
- "num_examples": 2490,
280
- "dataset_name": null
281
- }
282
- },
283
- "download_size": 50627367,
284
- "dataset_size": 75752623,
285
- "size_in_bytes": 126379990
286
- },
287
- "es": {
288
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
289
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
290
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
291
- "license": "",
292
- "features": {
293
- "premise": {
294
- "dtype": "string",
295
- "_type": "Value"
296
- },
297
- "hypothesis": {
298
- "dtype": "string",
299
- "_type": "Value"
300
- },
301
- "label": {
302
- "names": [
303
- "entailment",
304
- "neutral",
305
- "contradiction"
306
- ],
307
- "_type": "ClassLabel"
308
- }
309
- },
310
- "builder_name": "xnli",
311
- "dataset_name": "xnli",
312
- "config_name": "es",
313
- "version": {
314
- "version_str": "1.1.0",
315
- "description": "",
316
- "major": 1,
317
- "minor": 1,
318
- "patch": 0
319
- },
320
- "splits": {
321
- "train": {
322
- "name": "train",
323
- "num_bytes": 81383284,
324
- "num_examples": 392702,
325
- "dataset_name": null
326
- },
327
- "test": {
328
- "name": "test",
329
- "num_bytes": 969813,
330
- "num_examples": 5010,
331
- "dataset_name": null
332
- },
333
- "validation": {
334
- "name": "validation",
335
- "num_bytes": 478422,
336
- "num_examples": 2490,
337
- "dataset_name": null
338
- }
339
- },
340
- "download_size": 53677157,
341
- "dataset_size": 82831519,
342
- "size_in_bytes": 136508676
343
- },
344
- "fr": {
345
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
346
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
347
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
348
- "license": "",
349
- "features": {
350
- "premise": {
351
- "dtype": "string",
352
- "_type": "Value"
353
- },
354
- "hypothesis": {
355
- "dtype": "string",
356
- "_type": "Value"
357
- },
358
- "label": {
359
- "names": [
360
- "entailment",
361
- "neutral",
362
- "contradiction"
363
- ],
364
- "_type": "ClassLabel"
365
- }
366
- },
367
- "builder_name": "xnli",
368
- "dataset_name": "xnli",
369
- "config_name": "fr",
370
- "version": {
371
- "version_str": "1.1.0",
372
- "description": "",
373
- "major": 1,
374
- "minor": 1,
375
- "patch": 0
376
- },
377
- "splits": {
378
- "train": {
379
- "name": "train",
380
- "num_bytes": 85808779,
381
- "num_examples": 392702,
382
- "dataset_name": null
383
- },
384
- "test": {
385
- "name": "test",
386
- "num_bytes": 1029239,
387
- "num_examples": 5010,
388
- "dataset_name": null
389
- },
390
- "validation": {
391
- "name": "validation",
392
- "num_bytes": 510104,
393
- "num_examples": 2490,
394
- "dataset_name": null
395
- }
396
- },
397
- "download_size": 55968680,
398
- "dataset_size": 87348122,
399
- "size_in_bytes": 143316802
400
- },
401
- "hi": {
402
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
403
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
404
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
405
- "license": "",
406
- "features": {
407
- "premise": {
408
- "dtype": "string",
409
- "_type": "Value"
410
- },
411
- "hypothesis": {
412
- "dtype": "string",
413
- "_type": "Value"
414
- },
415
- "label": {
416
- "names": [
417
- "entailment",
418
- "neutral",
419
- "contradiction"
420
- ],
421
- "_type": "ClassLabel"
422
- }
423
- },
424
- "builder_name": "xnli",
425
- "dataset_name": "xnli",
426
- "config_name": "hi",
427
- "version": {
428
- "version_str": "1.1.0",
429
- "description": "",
430
- "major": 1,
431
- "minor": 1,
432
- "patch": 0
433
- },
434
- "splits": {
435
- "train": {
436
- "name": "train",
437
- "num_bytes": 170593964,
438
- "num_examples": 392702,
439
- "dataset_name": null
440
- },
441
- "test": {
442
- "name": "test",
443
- "num_bytes": 2073073,
444
- "num_examples": 5010,
445
- "dataset_name": null
446
- },
447
- "validation": {
448
- "name": "validation",
449
- "num_bytes": 1023915,
450
- "num_examples": 2490,
451
- "dataset_name": null
452
- }
453
- },
454
- "download_size": 70908548,
455
- "dataset_size": 173690952,
456
- "size_in_bytes": 244599500
457
- },
458
- "ru": {
459
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
460
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
461
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
462
- "license": "",
463
- "features": {
464
- "premise": {
465
- "dtype": "string",
466
- "_type": "Value"
467
- },
468
- "hypothesis": {
469
- "dtype": "string",
470
- "_type": "Value"
471
- },
472
- "label": {
473
- "names": [
474
- "entailment",
475
- "neutral",
476
- "contradiction"
477
- ],
478
- "_type": "ClassLabel"
479
- }
480
- },
481
- "builder_name": "xnli",
482
- "dataset_name": "xnli",
483
- "config_name": "ru",
484
- "version": {
485
- "version_str": "1.1.0",
486
- "description": "",
487
- "major": 1,
488
- "minor": 1,
489
- "patch": 0
490
- },
491
- "splits": {
492
- "train": {
493
- "name": "train",
494
- "num_bytes": 129859615,
495
- "num_examples": 392702,
496
- "dataset_name": null
497
- },
498
- "test": {
499
- "name": "test",
500
- "num_bytes": 1603466,
501
- "num_examples": 5010,
502
- "dataset_name": null
503
- },
504
- "validation": {
505
- "name": "validation",
506
- "num_bytes": 786442,
507
- "num_examples": 2490,
508
- "dataset_name": null
509
- }
510
- },
511
- "download_size": 70702606,
512
- "dataset_size": 132249523,
513
- "size_in_bytes": 202952129
514
- },
515
- "sw": {
516
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
517
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
518
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
519
- "license": "",
520
- "features": {
521
- "premise": {
522
- "dtype": "string",
523
- "_type": "Value"
524
- },
525
- "hypothesis": {
526
- "dtype": "string",
527
- "_type": "Value"
528
- },
529
- "label": {
530
- "names": [
531
- "entailment",
532
- "neutral",
533
- "contradiction"
534
- ],
535
- "_type": "ClassLabel"
536
- }
537
- },
538
- "builder_name": "xnli",
539
- "dataset_name": "xnli",
540
- "config_name": "sw",
541
- "version": {
542
- "version_str": "1.1.0",
543
- "description": "",
544
- "major": 1,
545
- "minor": 1,
546
- "patch": 0
547
- },
548
- "splits": {
549
- "train": {
550
- "name": "train",
551
- "num_bytes": 69285725,
552
- "num_examples": 392702,
553
- "dataset_name": null
554
- },
555
- "test": {
556
- "name": "test",
557
- "num_bytes": 871651,
558
- "num_examples": 5010,
559
- "dataset_name": null
560
- },
561
- "validation": {
562
- "name": "validation",
563
- "num_bytes": 429850,
564
- "num_examples": 2490,
565
- "dataset_name": null
566
- }
567
- },
568
- "download_size": 45564152,
569
- "dataset_size": 70587226,
570
- "size_in_bytes": 116151378
571
- },
572
- "th": {
573
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
574
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
575
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
576
- "license": "",
577
- "features": {
578
- "premise": {
579
- "dtype": "string",
580
- "_type": "Value"
581
- },
582
- "hypothesis": {
583
- "dtype": "string",
584
- "_type": "Value"
585
- },
586
- "label": {
587
- "names": [
588
- "entailment",
589
- "neutral",
590
- "contradiction"
591
- ],
592
- "_type": "ClassLabel"
593
- }
594
- },
595
- "builder_name": "xnli",
596
- "dataset_name": "xnli",
597
- "config_name": "th",
598
- "version": {
599
- "version_str": "1.1.0",
600
- "description": "",
601
- "major": 1,
602
- "minor": 1,
603
- "patch": 0
604
- },
605
- "splits": {
606
- "train": {
607
- "name": "train",
608
- "num_bytes": 176062892,
609
- "num_examples": 392702,
610
- "dataset_name": null
611
- },
612
- "test": {
613
- "name": "test",
614
- "num_bytes": 2147015,
615
- "num_examples": 5010,
616
- "dataset_name": null
617
- },
618
- "validation": {
619
- "name": "validation",
620
- "num_bytes": 1061160,
621
- "num_examples": 2490,
622
- "dataset_name": null
623
- }
624
- },
625
- "download_size": 77222045,
626
- "dataset_size": 179271067,
627
- "size_in_bytes": 256493112
628
- },
629
- "tr": {
630
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
631
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
632
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
633
- "license": "",
634
- "features": {
635
- "premise": {
636
- "dtype": "string",
637
- "_type": "Value"
638
- },
639
- "hypothesis": {
640
- "dtype": "string",
641
- "_type": "Value"
642
- },
643
- "label": {
644
- "names": [
645
- "entailment",
646
- "neutral",
647
- "contradiction"
648
- ],
649
- "_type": "ClassLabel"
650
- }
651
- },
652
- "builder_name": "xnli",
653
- "dataset_name": "xnli",
654
- "config_name": "tr",
655
- "version": {
656
- "version_str": "1.1.0",
657
- "description": "",
658
- "major": 1,
659
- "minor": 1,
660
- "patch": 0
661
- },
662
- "splits": {
663
- "train": {
664
- "name": "train",
665
- "num_bytes": 71637140,
666
- "num_examples": 392702,
667
- "dataset_name": null
668
- },
669
- "test": {
670
- "name": "test",
671
- "num_bytes": 934934,
672
- "num_examples": 5010,
673
- "dataset_name": null
674
- },
675
- "validation": {
676
- "name": "validation",
677
- "num_bytes": 459308,
678
- "num_examples": 2490,
679
- "dataset_name": null
680
- }
681
- },
682
- "download_size": 48509680,
683
- "dataset_size": 73031382,
684
- "size_in_bytes": 121541062
685
- },
686
- "ur": {
687
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
688
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
689
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
690
- "license": "",
691
- "features": {
692
- "premise": {
693
- "dtype": "string",
694
- "_type": "Value"
695
- },
696
- "hypothesis": {
697
- "dtype": "string",
698
- "_type": "Value"
699
- },
700
- "label": {
701
- "names": [
702
- "entailment",
703
- "neutral",
704
- "contradiction"
705
- ],
706
- "_type": "ClassLabel"
707
- }
708
- },
709
- "builder_name": "xnli",
710
- "dataset_name": "xnli",
711
- "config_name": "ur",
712
- "version": {
713
- "version_str": "1.1.0",
714
- "description": "",
715
- "major": 1,
716
- "minor": 1,
717
- "patch": 0
718
- },
719
- "splits": {
720
- "train": {
721
- "name": "train",
722
- "num_bytes": 96441486,
723
- "num_examples": 392702,
724
- "dataset_name": null
725
- },
726
- "test": {
727
- "name": "test",
728
- "num_bytes": 1416241,
729
- "num_examples": 5010,
730
- "dataset_name": null
731
- },
732
- "validation": {
733
- "name": "validation",
734
- "num_bytes": 699952,
735
- "num_examples": 2490,
736
- "dataset_name": null
737
- }
738
- },
739
- "download_size": 46682785,
740
- "dataset_size": 98557679,
741
- "size_in_bytes": 145240464
742
- },
743
- "vi": {
744
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
745
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
746
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
747
- "license": "",
748
- "features": {
749
- "premise": {
750
- "dtype": "string",
751
- "_type": "Value"
752
- },
753
- "hypothesis": {
754
- "dtype": "string",
755
- "_type": "Value"
756
- },
757
- "label": {
758
- "names": [
759
- "entailment",
760
- "neutral",
761
- "contradiction"
762
- ],
763
- "_type": "ClassLabel"
764
- }
765
- },
766
- "builder_name": "xnli",
767
- "dataset_name": "xnli",
768
- "config_name": "vi",
769
- "version": {
770
- "version_str": "1.1.0",
771
- "description": "",
772
- "major": 1,
773
- "minor": 1,
774
- "patch": 0
775
- },
776
- "splits": {
777
- "train": {
778
- "name": "train",
779
- "num_bytes": 101417430,
780
- "num_examples": 392702,
781
- "dataset_name": null
782
- },
783
- "test": {
784
- "name": "test",
785
- "num_bytes": 1190217,
786
- "num_examples": 5010,
787
- "dataset_name": null
788
- },
789
- "validation": {
790
- "name": "validation",
791
- "num_bytes": 590680,
792
- "num_examples": 2490,
793
- "dataset_name": null
794
- }
795
- },
796
- "download_size": 57690058,
797
- "dataset_size": 103198327,
798
- "size_in_bytes": 160888385
799
- },
800
- "zh": {
801
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
802
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
803
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
804
- "license": "",
805
- "features": {
806
- "premise": {
807
- "dtype": "string",
808
- "_type": "Value"
809
- },
810
- "hypothesis": {
811
- "dtype": "string",
812
- "_type": "Value"
813
- },
814
- "label": {
815
- "names": [
816
- "entailment",
817
- "neutral",
818
- "contradiction"
819
- ],
820
- "_type": "ClassLabel"
821
- }
822
- },
823
- "builder_name": "xnli",
824
- "dataset_name": "xnli",
825
- "config_name": "zh",
826
- "version": {
827
- "version_str": "1.1.0",
828
- "description": "",
829
- "major": 1,
830
- "minor": 1,
831
- "patch": 0
832
- },
833
- "splits": {
834
- "train": {
835
- "name": "train",
836
- "num_bytes": 72224841,
837
- "num_examples": 392702,
838
- "dataset_name": null
839
- },
840
- "test": {
841
- "name": "test",
842
- "num_bytes": 777929,
843
- "num_examples": 5010,
844
- "dataset_name": null
845
- },
846
- "validation": {
847
- "name": "validation",
848
- "num_bytes": 384851,
849
- "num_examples": 2490,
850
- "dataset_name": null
851
- }
852
- },
853
- "download_size": 48269855,
854
- "dataset_size": 73387621,
855
- "size_in_bytes": 121657476
856
- },
857
- "all_languages": {
858
- "description": "XNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n",
859
- "citation": "@InProceedings{conneau2018xnli,\n author = {Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin},\n title = {XNLI: Evaluating Cross-lingual Sentence Representations},\n booktitle = {Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing},\n year = {2018},\n publisher = {Association for Computational Linguistics},\n location = {Brussels, Belgium},\n}",
860
- "homepage": "https://www.nyu.edu/projects/bowman/xnli/",
861
- "license": "",
862
- "features": {
863
- "premise": {
864
- "languages": [
865
- "ar",
866
- "bg",
867
- "de",
868
- "el",
869
- "en",
870
- "es",
871
- "fr",
872
- "hi",
873
- "ru",
874
- "sw",
875
- "th",
876
- "tr",
877
- "ur",
878
- "vi",
879
- "zh"
880
- ],
881
- "_type": "Translation"
882
- },
883
- "hypothesis": {
884
- "languages": [
885
- "ar",
886
- "bg",
887
- "de",
888
- "el",
889
- "en",
890
- "es",
891
- "fr",
892
- "hi",
893
- "ru",
894
- "sw",
895
- "th",
896
- "tr",
897
- "ur",
898
- "vi",
899
- "zh"
900
- ],
901
- "num_languages": 15,
902
- "_type": "TranslationVariableLanguages"
903
- },
904
- "label": {
905
- "names": [
906
- "entailment",
907
- "neutral",
908
- "contradiction"
909
- ],
910
- "_type": "ClassLabel"
911
- }
912
- },
913
- "builder_name": "xnli",
914
- "dataset_name": "xnli",
915
- "config_name": "all_languages",
916
- "version": {
917
- "version_str": "1.1.0",
918
- "description": "",
919
- "major": 1,
920
- "minor": 1,
921
- "patch": 0
922
- },
923
- "splits": {
924
- "train": {
925
- "name": "train",
926
- "num_bytes": 1581471691,
927
- "num_examples": 392702,
928
- "dataset_name": null
929
- },
930
- "test": {
931
- "name": "test",
932
- "num_bytes": 19387432,
933
- "num_examples": 5010,
934
- "dataset_name": null
935
- },
936
- "validation": {
937
- "name": "validation",
938
- "num_bytes": 9566179,
939
- "num_examples": 2490,
940
- "dataset_name": null
941
- }
942
- },
943
- "download_size": 963942271,
944
- "dataset_size": 1610425302,
945
- "size_in_bytes": 2574367573
946
- }
947
- }