Datasets:

Modalities:
Text
ArXiv:
Libraries:
Datasets
License:
PereLluis13 commited on
Commit
568ffc1
1 Parent(s): 18e0a2a

update and add README

Browse files
Files changed (2) hide show
  1. README.md +831 -0
  2. SREDFM.py +13 -9
README.md ADDED
@@ -0,0 +1,831 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ dataset_info:
3
+ - config_name: ar
4
+ features:
5
+ - name: docid
6
+ dtype: string
7
+ - name: title
8
+ dtype: string
9
+ - name: uri
10
+ dtype: string
11
+ - name: text
12
+ dtype: string
13
+ - name: entities
14
+ sequence:
15
+ - name: uri
16
+ dtype: string
17
+ - name: surfaceform
18
+ dtype: string
19
+ - name: type
20
+ dtype: string
21
+ - name: start
22
+ dtype: int32
23
+ - name: end
24
+ dtype: int32
25
+ - name: relations
26
+ sequence:
27
+ - name: subject
28
+ dtype: int32
29
+ - name: predicate
30
+ dtype: string
31
+ - name: object
32
+ dtype: int32
33
+ splits:
34
+ - name: train
35
+ num_bytes: 682284140
36
+ num_examples: 505752
37
+ - name: test
38
+ num_bytes: 9120804
39
+ num_examples: 4387
40
+ - name: validation
41
+ num_bytes: 7497301
42
+ num_examples: 3783
43
+ download_size: 3651950669
44
+ dataset_size: 698902245
45
+ - config_name: ca
46
+ features:
47
+ - name: docid
48
+ dtype: string
49
+ - name: title
50
+ dtype: string
51
+ - name: uri
52
+ dtype: string
53
+ - name: text
54
+ dtype: string
55
+ - name: entities
56
+ sequence:
57
+ - name: uri
58
+ dtype: string
59
+ - name: surfaceform
60
+ dtype: string
61
+ - name: type
62
+ dtype: string
63
+ - name: start
64
+ dtype: int32
65
+ - name: end
66
+ dtype: int32
67
+ - name: relations
68
+ sequence:
69
+ - name: subject
70
+ dtype: int32
71
+ - name: predicate
72
+ dtype: string
73
+ - name: object
74
+ dtype: int32
75
+ splits:
76
+ - name: train
77
+ num_bytes: 421250589
78
+ num_examples: 299706
79
+ - name: test
80
+ num_bytes: 5439773
81
+ num_examples: 2541
82
+ - name: validation
83
+ num_bytes: 3173490
84
+ num_examples: 1532
85
+ download_size: 1513026644
86
+ dataset_size: 429863852
87
+ - config_name: de
88
+ features:
89
+ - name: docid
90
+ dtype: string
91
+ - name: title
92
+ dtype: string
93
+ - name: uri
94
+ dtype: string
95
+ - name: text
96
+ dtype: string
97
+ - name: entities
98
+ sequence:
99
+ - name: uri
100
+ dtype: string
101
+ - name: surfaceform
102
+ dtype: string
103
+ - name: type
104
+ dtype: string
105
+ - name: start
106
+ dtype: int32
107
+ - name: end
108
+ dtype: int32
109
+ - name: relations
110
+ sequence:
111
+ - name: subject
112
+ dtype: int32
113
+ - name: predicate
114
+ dtype: string
115
+ - name: object
116
+ dtype: int32
117
+ splits:
118
+ - name: train
119
+ num_bytes: 1340918444
120
+ num_examples: 1067928
121
+ - name: test
122
+ num_bytes: 10908663
123
+ num_examples: 5649
124
+ - name: validation
125
+ num_bytes: 9075742
126
+ num_examples: 4994
127
+ download_size: 4521091910
128
+ dataset_size: 1360902849
129
+ - config_name: el
130
+ features:
131
+ - name: docid
132
+ dtype: string
133
+ - name: title
134
+ dtype: string
135
+ - name: uri
136
+ dtype: string
137
+ - name: text
138
+ dtype: string
139
+ - name: entities
140
+ sequence:
141
+ - name: uri
142
+ dtype: string
143
+ - name: surfaceform
144
+ dtype: string
145
+ - name: type
146
+ dtype: string
147
+ - name: start
148
+ dtype: int32
149
+ - name: end
150
+ dtype: int32
151
+ - name: relations
152
+ sequence:
153
+ - name: subject
154
+ dtype: int32
155
+ - name: predicate
156
+ dtype: string
157
+ - name: object
158
+ dtype: int32
159
+ splits:
160
+ - name: train
161
+ num_bytes: 139191860
162
+ num_examples: 66012
163
+ - name: test
164
+ num_bytes: 2385490
165
+ num_examples: 861
166
+ - name: validation
167
+ num_bytes: 1852124
168
+ num_examples: 668
169
+ download_size: 579372781
170
+ dataset_size: 143429474
171
+ - config_name: en
172
+ features:
173
+ - name: docid
174
+ dtype: string
175
+ - name: title
176
+ dtype: string
177
+ - name: uri
178
+ dtype: string
179
+ - name: text
180
+ dtype: string
181
+ - name: entities
182
+ sequence:
183
+ - name: uri
184
+ dtype: string
185
+ - name: surfaceform
186
+ dtype: string
187
+ - name: type
188
+ dtype: string
189
+ - name: start
190
+ dtype: int32
191
+ - name: end
192
+ dtype: int32
193
+ - name: relations
194
+ sequence:
195
+ - name: subject
196
+ dtype: int32
197
+ - name: predicate
198
+ dtype: string
199
+ - name: object
200
+ dtype: int32
201
+ splits:
202
+ - name: train
203
+ num_bytes: 3721987473
204
+ num_examples: 2777298
205
+ - name: test
206
+ num_bytes: 13320623
207
+ num_examples: 6685
208
+ - name: validation
209
+ num_bytes: 28009738
210
+ num_examples: 13236
211
+ download_size: 11914987368
212
+ dataset_size: 3763317834
213
+ - config_name: es
214
+ features:
215
+ - name: docid
216
+ dtype: string
217
+ - name: title
218
+ dtype: string
219
+ - name: uri
220
+ dtype: string
221
+ - name: text
222
+ dtype: string
223
+ - name: entities
224
+ sequence:
225
+ - name: uri
226
+ dtype: string
227
+ - name: surfaceform
228
+ dtype: string
229
+ - name: type
230
+ dtype: string
231
+ - name: start
232
+ dtype: int32
233
+ - name: end
234
+ dtype: int32
235
+ - name: relations
236
+ sequence:
237
+ - name: subject
238
+ dtype: int32
239
+ - name: predicate
240
+ dtype: string
241
+ - name: object
242
+ dtype: int32
243
+ splits:
244
+ - name: train
245
+ num_bytes: 921981916
246
+ num_examples: 716002
247
+ - name: test
248
+ num_bytes: 16281846
249
+ num_examples: 8561
250
+ - name: validation
251
+ num_bytes: 4674008
252
+ num_examples: 2177
253
+ download_size: 3570403740
254
+ dataset_size: 942937770
255
+ - config_name: fr
256
+ features:
257
+ - name: docid
258
+ dtype: string
259
+ - name: title
260
+ dtype: string
261
+ - name: uri
262
+ dtype: string
263
+ - name: text
264
+ dtype: string
265
+ - name: entities
266
+ sequence:
267
+ - name: uri
268
+ dtype: string
269
+ - name: surfaceform
270
+ dtype: string
271
+ - name: type
272
+ dtype: string
273
+ - name: start
274
+ dtype: int32
275
+ - name: end
276
+ dtype: int32
277
+ - name: relations
278
+ sequence:
279
+ - name: subject
280
+ dtype: int32
281
+ - name: predicate
282
+ dtype: string
283
+ - name: object
284
+ dtype: int32
285
+ splits:
286
+ - name: train
287
+ num_bytes: 795898612
288
+ num_examples: 876175
289
+ - name: test
290
+ num_bytes: 6030937
291
+ num_examples: 3883
292
+ - name: validation
293
+ num_bytes: 3283158
294
+ num_examples: 2079
295
+ download_size: 3269522484
296
+ dataset_size: 805212707
297
+ - config_name: hi
298
+ features:
299
+ - name: docid
300
+ dtype: string
301
+ - name: title
302
+ dtype: string
303
+ - name: uri
304
+ dtype: string
305
+ - name: text
306
+ dtype: string
307
+ - name: entities
308
+ sequence:
309
+ - name: uri
310
+ dtype: string
311
+ - name: surfaceform
312
+ dtype: string
313
+ - name: type
314
+ dtype: string
315
+ - name: start
316
+ dtype: int32
317
+ - name: end
318
+ dtype: int32
319
+ - name: relations
320
+ sequence:
321
+ - name: subject
322
+ dtype: int32
323
+ - name: predicate
324
+ dtype: string
325
+ - name: object
326
+ dtype: int32
327
+ splits:
328
+ - name: train
329
+ num_bytes: 105751162
330
+ num_examples: 54739
331
+ - name: test
332
+ num_bytes: 1349067
333
+ num_examples: 374
334
+ - name: validation
335
+ num_bytes: 1231818
336
+ num_examples: 405
337
+ download_size: 385810623
338
+ dataset_size: 108332047
339
+ - config_name: it
340
+ features:
341
+ - name: docid
342
+ dtype: string
343
+ - name: title
344
+ dtype: string
345
+ - name: uri
346
+ dtype: string
347
+ - name: text
348
+ dtype: string
349
+ - name: entities
350
+ sequence:
351
+ - name: uri
352
+ dtype: string
353
+ - name: surfaceform
354
+ dtype: string
355
+ - name: type
356
+ dtype: string
357
+ - name: start
358
+ dtype: int32
359
+ - name: end
360
+ dtype: int32
361
+ - name: relations
362
+ sequence:
363
+ - name: subject
364
+ dtype: int32
365
+ - name: predicate
366
+ dtype: string
367
+ - name: object
368
+ dtype: int32
369
+ splits:
370
+ - name: train
371
+ num_bytes: 453523291
372
+ num_examples: 436783
373
+ - name: test
374
+ num_bytes: 3850421
375
+ num_examples: 2175
376
+ - name: validation
377
+ num_bytes: 2261619
378
+ num_examples: 1276
379
+ download_size: 1685172398
380
+ dataset_size: 459635331
381
+ - config_name: ja
382
+ features:
383
+ - name: docid
384
+ dtype: string
385
+ - name: title
386
+ dtype: string
387
+ - name: uri
388
+ dtype: string
389
+ - name: text
390
+ dtype: string
391
+ - name: entities
392
+ sequence:
393
+ - name: uri
394
+ dtype: string
395
+ - name: surfaceform
396
+ dtype: string
397
+ - name: type
398
+ dtype: string
399
+ - name: start
400
+ dtype: int32
401
+ - name: end
402
+ dtype: int32
403
+ - name: relations
404
+ sequence:
405
+ - name: subject
406
+ dtype: int32
407
+ - name: predicate
408
+ dtype: string
409
+ - name: object
410
+ dtype: int32
411
+ splits:
412
+ - name: train
413
+ num_bytes: 746883176
414
+ num_examples: 498156
415
+ - name: test
416
+ num_bytes: 7883474
417
+ num_examples: 3392
418
+ - name: validation
419
+ num_bytes: 7065181
420
+ num_examples: 3106
421
+ download_size: 3186065351
422
+ dataset_size: 761831831
423
+ - config_name: ko
424
+ features:
425
+ - name: docid
426
+ dtype: string
427
+ - name: title
428
+ dtype: string
429
+ - name: uri
430
+ dtype: string
431
+ - name: text
432
+ dtype: string
433
+ - name: entities
434
+ sequence:
435
+ - name: uri
436
+ dtype: string
437
+ - name: surfaceform
438
+ dtype: string
439
+ - name: type
440
+ dtype: string
441
+ - name: start
442
+ dtype: int32
443
+ - name: end
444
+ dtype: int32
445
+ - name: relations
446
+ sequence:
447
+ - name: subject
448
+ dtype: int32
449
+ - name: predicate
450
+ dtype: string
451
+ - name: object
452
+ dtype: int32
453
+ splits:
454
+ - name: train
455
+ num_bytes: 289343227
456
+ num_examples: 228511
457
+ - name: test
458
+ num_bytes: 1756081
459
+ num_examples: 803
460
+ - name: validation
461
+ num_bytes: 1879237
462
+ num_examples: 917
463
+ download_size: 1119778167
464
+ dataset_size: 292978545
465
+ - config_name: nl
466
+ features:
467
+ - name: docid
468
+ dtype: string
469
+ - name: title
470
+ dtype: string
471
+ - name: uri
472
+ dtype: string
473
+ - name: text
474
+ dtype: string
475
+ - name: entities
476
+ sequence:
477
+ - name: uri
478
+ dtype: string
479
+ - name: surfaceform
480
+ dtype: string
481
+ - name: type
482
+ dtype: string
483
+ - name: start
484
+ dtype: int32
485
+ - name: end
486
+ dtype: int32
487
+ - name: relations
488
+ sequence:
489
+ - name: subject
490
+ dtype: int32
491
+ - name: predicate
492
+ dtype: string
493
+ - name: object
494
+ dtype: int32
495
+ splits:
496
+ - name: train
497
+ num_bytes: 730757583
498
+ num_examples: 670667
499
+ - name: test
500
+ num_bytes: 5251744
501
+ num_examples: 2715
502
+ - name: validation
503
+ num_bytes: 4241389
504
+ num_examples: 2188
505
+ download_size: 2591997126
506
+ dataset_size: 740250716
507
+ - config_name: pl
508
+ features:
509
+ - name: docid
510
+ dtype: string
511
+ - name: title
512
+ dtype: string
513
+ - name: uri
514
+ dtype: string
515
+ - name: text
516
+ dtype: string
517
+ - name: entities
518
+ sequence:
519
+ - name: uri
520
+ dtype: string
521
+ - name: surfaceform
522
+ dtype: string
523
+ - name: type
524
+ dtype: string
525
+ - name: start
526
+ dtype: int32
527
+ - name: end
528
+ dtype: int32
529
+ - name: relations
530
+ sequence:
531
+ - name: subject
532
+ dtype: int32
533
+ - name: predicate
534
+ dtype: string
535
+ - name: object
536
+ dtype: int32
537
+ splits:
538
+ - name: train
539
+ num_bytes: 913398908
540
+ num_examples: 688892
541
+ - name: test
542
+ num_bytes: 11628583
543
+ num_examples: 6376
544
+ - name: validation
545
+ num_bytes: 6702413
546
+ num_examples: 3476
547
+ download_size: 3365852789
548
+ dataset_size: 931729904
549
+ - config_name: pt
550
+ features:
551
+ - name: docid
552
+ dtype: string
553
+ - name: title
554
+ dtype: string
555
+ - name: uri
556
+ dtype: string
557
+ - name: text
558
+ dtype: string
559
+ - name: entities
560
+ sequence:
561
+ - name: uri
562
+ dtype: string
563
+ - name: surfaceform
564
+ dtype: string
565
+ - name: type
566
+ dtype: string
567
+ - name: start
568
+ dtype: int32
569
+ - name: end
570
+ dtype: int32
571
+ - name: relations
572
+ sequence:
573
+ - name: subject
574
+ dtype: int32
575
+ - name: predicate
576
+ dtype: string
577
+ - name: object
578
+ dtype: int32
579
+ splits:
580
+ - name: train
581
+ num_bytes: 613573841
582
+ num_examples: 492047
583
+ - name: test
584
+ num_bytes: 8782219
585
+ num_examples: 4313
586
+ - name: validation
587
+ num_bytes: 5878645
588
+ num_examples: 2973
589
+ download_size: 2347987926
590
+ dataset_size: 628234705
591
+ - config_name: ru
592
+ features:
593
+ - name: docid
594
+ dtype: string
595
+ - name: title
596
+ dtype: string
597
+ - name: uri
598
+ dtype: string
599
+ - name: text
600
+ dtype: string
601
+ - name: entities
602
+ sequence:
603
+ - name: uri
604
+ dtype: string
605
+ - name: surfaceform
606
+ dtype: string
607
+ - name: type
608
+ dtype: string
609
+ - name: start
610
+ dtype: int32
611
+ - name: end
612
+ dtype: int32
613
+ - name: relations
614
+ sequence:
615
+ - name: subject
616
+ dtype: int32
617
+ - name: predicate
618
+ dtype: string
619
+ - name: object
620
+ dtype: int32
621
+ splits:
622
+ - name: train
623
+ num_bytes: 645507720
624
+ num_examples: 357966
625
+ - name: test
626
+ num_bytes: 5996262
627
+ num_examples: 2296
628
+ - name: validation
629
+ num_bytes: 5403427
630
+ num_examples: 2107
631
+ download_size: 2754576893
632
+ dataset_size: 656907409
633
+ - config_name: sv
634
+ features:
635
+ - name: docid
636
+ dtype: string
637
+ - name: title
638
+ dtype: string
639
+ - name: uri
640
+ dtype: string
641
+ - name: text
642
+ dtype: string
643
+ - name: entities
644
+ sequence:
645
+ - name: uri
646
+ dtype: string
647
+ - name: surfaceform
648
+ dtype: string
649
+ - name: type
650
+ dtype: string
651
+ - name: start
652
+ dtype: int32
653
+ - name: end
654
+ dtype: int32
655
+ - name: relations
656
+ sequence:
657
+ - name: subject
658
+ dtype: int32
659
+ - name: predicate
660
+ dtype: string
661
+ - name: object
662
+ dtype: int32
663
+ splits:
664
+ - name: train
665
+ num_bytes: 1880140424
666
+ num_examples: 1755721
667
+ - name: test
668
+ num_bytes: 13183100
669
+ num_examples: 7531
670
+ - name: validation
671
+ num_bytes: 5207785
672
+ num_examples: 2987
673
+ download_size: 6790489020
674
+ dataset_size: 1898531309
675
+ - config_name: vi
676
+ features:
677
+ - name: docid
678
+ dtype: string
679
+ - name: title
680
+ dtype: string
681
+ - name: uri
682
+ dtype: string
683
+ - name: text
684
+ dtype: string
685
+ - name: entities
686
+ sequence:
687
+ - name: uri
688
+ dtype: string
689
+ - name: surfaceform
690
+ dtype: string
691
+ - name: type
692
+ dtype: string
693
+ - name: start
694
+ dtype: int32
695
+ - name: end
696
+ dtype: int32
697
+ - name: relations
698
+ sequence:
699
+ - name: subject
700
+ dtype: int32
701
+ - name: predicate
702
+ dtype: string
703
+ - name: object
704
+ dtype: int32
705
+ splits:
706
+ - name: train
707
+ num_bytes: 328894365
708
+ num_examples: 287553
709
+ - name: test
710
+ num_bytes: 4348571
711
+ num_examples: 1824
712
+ - name: validation
713
+ num_bytes: 3437184
714
+ num_examples: 1461
715
+ download_size: 1301938106
716
+ dataset_size: 336680120
717
+ - config_name: zh
718
+ features:
719
+ - name: docid
720
+ dtype: string
721
+ - name: title
722
+ dtype: string
723
+ - name: uri
724
+ dtype: string
725
+ - name: text
726
+ dtype: string
727
+ - name: entities
728
+ sequence:
729
+ - name: uri
730
+ dtype: string
731
+ - name: surfaceform
732
+ dtype: string
733
+ - name: type
734
+ dtype: string
735
+ - name: start
736
+ dtype: int32
737
+ - name: end
738
+ dtype: int32
739
+ - name: relations
740
+ sequence:
741
+ - name: subject
742
+ dtype: int32
743
+ - name: predicate
744
+ dtype: string
745
+ - name: object
746
+ dtype: int32
747
+ splits:
748
+ - name: train
749
+ num_bytes: 478317655
750
+ num_examples: 387461
751
+ - name: test
752
+ num_bytes: 5324982
753
+ num_examples: 2667
754
+ - name: validation
755
+ num_bytes: 3554687
756
+ num_examples: 1816
757
+ download_size: 2440525684
758
+ dataset_size: 487197324
759
+ - config_name: all_languages
760
+ features:
761
+ - name: docid
762
+ dtype: string
763
+ - name: title
764
+ dtype: string
765
+ - name: uri
766
+ dtype: string
767
+ - name: lan
768
+ dtype: string
769
+ - name: text
770
+ dtype: string
771
+ - name: entities
772
+ sequence:
773
+ - name: uri
774
+ dtype: string
775
+ - name: surfaceform
776
+ dtype: string
777
+ - name: type
778
+ dtype: string
779
+ - name: start
780
+ dtype: int32
781
+ - name: end
782
+ dtype: int32
783
+ - name: relations
784
+ sequence:
785
+ - name: subject
786
+ dtype: int32
787
+ - name: predicate
788
+ dtype: string
789
+ - name: object
790
+ dtype: int32
791
+ splits:
792
+ - name: train
793
+ num_bytes: 15282608600
794
+ num_examples: 12167369
795
+ - name: test
796
+ num_bytes: 133244838
797
+ num_examples: 67033
798
+ - name: validation
799
+ num_bytes: 104736032
800
+ num_examples: 51181
801
+ download_size: 56989165879
802
+ dataset_size: 15520589470
803
+ task_categories:
804
+ - token-classification
805
+ language:
806
+ - ar
807
+ - ca
808
+ - de
809
+ - el
810
+ - en
811
+ - es
812
+ - fr
813
+ - hi
814
+ - it
815
+ - ja
816
+ - ko
817
+ - nl
818
+ - pl
819
+ - pt
820
+ - ru
821
+ - sv
822
+ - vi
823
+ - zh
824
+ size_categories:
825
+ - 10M<n<100M
826
+ ---
827
+ # RED<sup>FM</sup>: a Filtered and Multilingual Relation Extraction Dataset
828
+
829
+ This is the automatically-filtered dataset the 2023 ACL paper [RED^{FM}: a Filtered and Multilingual Relation Extraction Dataset](https://github.com/Babelscape/rebel/blob/main/docs/).
830
+
831
+ Dataset Card is coming soon.
SREDFM.py CHANGED
@@ -102,10 +102,10 @@ class SREDFM(datasets.GeneratorBasedBuilder):
102
  "uri": datasets.Value("string"),
103
  "lan": datasets.Value("string"),
104
  "text": datasets.Value("string"),
105
- "entities": datasets.Sequence(feature={'uri': datasets.Value(dtype='string'), 'surfaceform': datasets.Value(dtype='string'), 'type': datasets.Value(dtype='string'), 'start': datasets.Value(dtype='int32'), 'end': datasets.Value(dtype='int32')}),
106
- "relations": datasets.Sequence(feature={'subject': datasets.Value(dtype='int32'),
107
- 'predicate': datasets.Value(dtype='string'),
108
- 'object': datasets.Value(dtype='int32')}),
109
  }
110
  )
111
  else:
@@ -115,10 +115,10 @@ class SREDFM(datasets.GeneratorBasedBuilder):
115
  "title": datasets.Value("string"),
116
  "uri": datasets.Value("string"),
117
  "text": datasets.Value("string"),
118
- "entities": datasets.Sequence(feature={'uri': datasets.Value(dtype='string'), 'surfaceform': datasets.Value(dtype='string'), 'type': datasets.Value(dtype='string'), 'start': datasets.Value(dtype='int32'), 'end': datasets.Value(dtype='int32')}),
119
- "relations": datasets.Sequence(feature={'subject': datasets.Value(dtype='int32'),
120
- 'predicate': datasets.Value(dtype='string'),
121
- 'object': datasets.Value(dtype='int32')}),
122
  }
123
  )
124
  return datasets.DatasetInfo(
@@ -207,7 +207,9 @@ class SREDFM(datasets.GeneratorBasedBuilder):
207
  "end": relation["object"]["boundaries"][1],
208
  "type": relation["object"]["type"],
209
  }),
210
- })
 
 
211
  yield data["docid"]+ '-' + data["lan"], {
212
  "docid": data["docid"],
213
  "title": data["title"],
@@ -252,6 +254,8 @@ class SREDFM(datasets.GeneratorBasedBuilder):
252
  "type": relation["object"]["type"],
253
  }),
254
  })
 
 
255
  yield data["docid"], {
256
  "docid": data["docid"],
257
  "title": data["title"],
 
102
  "uri": datasets.Value("string"),
103
  "lan": datasets.Value("string"),
104
  "text": datasets.Value("string"),
105
+ "entities": [{'uri': datasets.Value(dtype='string'), 'surfaceform': datasets.Value(dtype='string'), 'type': datasets.Value(dtype='string'), 'start': datasets.Value(dtype='int32'), 'end': datasets.Value(dtype='int32')}],
106
+ "relations": [{'subject': datasets.Value(dtype='int32'),
107
+ 'predicate': datasets.Value(dtype='string'),
108
+ 'object': datasets.Value(dtype='int32')}],
109
  }
110
  )
111
  else:
 
115
  "title": datasets.Value("string"),
116
  "uri": datasets.Value("string"),
117
  "text": datasets.Value("string"),
118
+ "entities": [{'uri': datasets.Value(dtype='string'), 'surfaceform': datasets.Value(dtype='string'), 'type': datasets.Value(dtype='string'), 'start': datasets.Value(dtype='int32'), 'end': datasets.Value(dtype='int32')}],
119
+ "relations": [{'subject': datasets.Value(dtype='int32'),
120
+ 'predicate': datasets.Value(dtype='string'),
121
+ 'object': datasets.Value(dtype='int32')}],
122
  }
123
  )
124
  return datasets.DatasetInfo(
 
207
  "end": relation["object"]["boundaries"][1],
208
  "type": relation["object"]["type"],
209
  }),
210
+ })
211
+ if len(relations) == 0:
212
+ continue
213
  yield data["docid"]+ '-' + data["lan"], {
214
  "docid": data["docid"],
215
  "title": data["title"],
 
254
  "type": relation["object"]["type"],
255
  }),
256
  })
257
+ if len(relations) == 0:
258
+ continue
259
  yield data["docid"], {
260
  "docid": data["docid"],
261
  "title": data["title"],