ParseValAndTrain.ipynb
113 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "aabfb24b",
"metadata": {},
"outputs": [],
"source": [
"COMBO = '/home/kkrasnowska/anaconda3/envs/combo_p39/bin/combo'"
]
},
{
"cell_type": "markdown",
"id": "787fff78",
"metadata": {},
"source": [
"Main model"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "1d9daaa9",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"I0407 10:49:31.448594 140072765682752 archival.py:184] loading archive file model-pdbc/model.tar.gz\n",
"I0407 10:49:31.449148 140072765682752 archival.py:263] extracting archive file model-pdbc/model.tar.gz to temp dir /tmp/tmp_htckuhc\n",
"I0407 10:49:48.075045 140072765682752 params.py:248] dataset_reader.type = conllu\n",
"I0407 10:49:48.075561 140072765682752 params.py:248] dataset_reader.lazy = False\n",
"I0407 10:49:48.075693 140072765682752 params.py:248] dataset_reader.cache_directory = None\n",
"I0407 10:49:48.075764 140072765682752 params.py:248] dataset_reader.max_instances = None\n",
"I0407 10:49:48.075832 140072765682752 params.py:248] dataset_reader.manual_distributed_sharding = False\n",
"I0407 10:49:48.075901 140072765682752 params.py:248] dataset_reader.manual_multi_process_sharding = False\n",
"I0407 10:49:48.076193 140072765682752 params.py:248] dataset_reader.token_indexers.char.type = characters_const_padding\n",
"I0407 10:49:48.076388 140072765682752 params.py:248] dataset_reader.token_indexers.char.namespace = token_characters\n",
"I0407 10:49:48.076621 140072765682752 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:49:48.076697 140072765682752 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:49:48.076790 140072765682752 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:49:48.076939 140072765682752 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:49:48.077063 140072765682752 params.py:248] dataset_reader.token_indexers.char.start_tokens = None\n",
"I0407 10:49:48.077118 140072765682752 params.py:248] dataset_reader.token_indexers.char.end_tokens = None\n",
"I0407 10:49:48.077185 140072765682752 params.py:248] dataset_reader.token_indexers.char.min_padding_length = 32\n",
"I0407 10:49:48.077238 140072765682752 params.py:248] dataset_reader.token_indexers.char.token_min_padding_length = 0\n",
"I0407 10:49:48.077383 140072765682752 params.py:248] dataset_reader.token_indexers.feats.type = feats_indexer\n",
"I0407 10:49:48.077555 140072765682752 params.py:248] dataset_reader.token_indexers.feats.namespace = feats\n",
"I0407 10:49:48.077628 140072765682752 params.py:248] dataset_reader.token_indexers.feats.feature_name = feats_\n",
"I0407 10:49:48.077702 140072765682752 params.py:248] dataset_reader.token_indexers.feats.token_min_padding_length = 0\n",
"I0407 10:49:48.077838 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.type = characters_const_padding\n",
"I0407 10:49:48.078031 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.namespace = token_characters\n",
"I0407 10:49:48.078231 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.byte_encoding = None\n",
"I0407 10:49:48.078300 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.lowercase_characters = False\n",
"I0407 10:49:48.078378 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:49:48.078666 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:49:48.078786 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.start_tokens = None\n",
"I0407 10:49:48.078862 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.end_tokens = None\n",
"I0407 10:49:48.078916 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.min_padding_length = 32\n",
"I0407 10:49:48.078969 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.token_min_padding_length = 0\n",
"I0407 10:49:48.079103 140072765682752 params.py:248] dataset_reader.token_indexers.token.type = pretrained_transformer_mismatched_fixed\n",
"I0407 10:49:48.079328 140072765682752 params.py:248] dataset_reader.token_indexers.token.token_min_padding_length = 0\n",
"I0407 10:49:48.079406 140072765682752 params.py:248] dataset_reader.token_indexers.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:49:48.079461 140072765682752 params.py:248] dataset_reader.token_indexers.token.namespace = tags\n",
"I0407 10:49:48.079525 140072765682752 params.py:248] dataset_reader.token_indexers.token.max_length = None\n",
"I0407 10:49:48.079628 140072765682752 params.py:384] dataset_reader.token_indexers.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:49:51.185825 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.type = single_id\n",
"I0407 10:49:51.186234 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.namespace = upostag\n",
"I0407 10:49:51.186336 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.lowercase_tokens = False\n",
"I0407 10:49:51.186398 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.start_tokens = None\n",
"I0407 10:49:51.186465 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.end_tokens = None\n",
"I0407 10:49:51.186517 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.feature_name = pos_\n",
"I0407 10:49:51.186579 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:49:51.186631 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.token_min_padding_length = 0\n",
"I0407 10:49:51.186791 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.type = single_id\n",
"I0407 10:49:51.186975 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.namespace = xpostag\n",
"I0407 10:49:51.187041 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.lowercase_tokens = False\n",
"I0407 10:49:51.187107 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.start_tokens = None\n",
"I0407 10:49:51.187170 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.end_tokens = None\n",
"I0407 10:49:51.187220 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.feature_name = tag_\n",
"I0407 10:49:51.187275 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:49:51.187334 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.token_min_padding_length = 0\n",
"I0407 10:49:51.187556 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.type = characters_const_padding\n",
"I0407 10:49:51.187731 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.namespace = lemma_characters\n",
"I0407 10:49:51.187935 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:49:51.187995 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:49:51.188073 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:49:51.188217 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:49:51.188334 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.start_tokens = None\n",
"I0407 10:49:51.188398 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.end_tokens = None\n",
"I0407 10:49:51.188460 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.min_padding_length = 32\n",
"I0407 10:49:51.188522 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.token_min_padding_length = 0\n",
"I0407 10:49:51.188614 140072765682752 params.py:248] dataset_reader.features = ['token', 'char']\n",
"I0407 10:49:51.188712 140072765682752 params.py:248] dataset_reader.targets = ['head', 'deprel']\n",
"I0407 10:49:51.188802 140072765682752 params.py:248] dataset_reader.use_sem = False\n",
"I0407 10:49:51.188952 140072765682752 params.py:248] dataset_reader.type = conllu\n",
"I0407 10:49:51.189191 140072765682752 params.py:248] dataset_reader.lazy = False\n",
"I0407 10:49:51.189266 140072765682752 params.py:248] dataset_reader.cache_directory = None\n",
"I0407 10:49:51.189324 140072765682752 params.py:248] dataset_reader.max_instances = None\n",
"I0407 10:49:51.189382 140072765682752 params.py:248] dataset_reader.manual_distributed_sharding = False\n",
"I0407 10:49:51.189436 140072765682752 params.py:248] dataset_reader.manual_multi_process_sharding = False\n",
"I0407 10:49:51.189675 140072765682752 params.py:248] dataset_reader.token_indexers.char.type = characters_const_padding\n",
"I0407 10:49:51.189843 140072765682752 params.py:248] dataset_reader.token_indexers.char.namespace = token_characters\n",
"I0407 10:49:51.190060 140072765682752 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:49:51.190128 140072765682752 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:49:51.190197 140072765682752 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:49:51.190324 140072765682752 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:49:51.190443 140072765682752 params.py:248] dataset_reader.token_indexers.char.start_tokens = None\n",
"I0407 10:49:51.190508 140072765682752 params.py:248] dataset_reader.token_indexers.char.end_tokens = None\n",
"I0407 10:49:51.190564 140072765682752 params.py:248] dataset_reader.token_indexers.char.min_padding_length = 32\n",
"I0407 10:49:51.190627 140072765682752 params.py:248] dataset_reader.token_indexers.char.token_min_padding_length = 0\n",
"I0407 10:49:51.190772 140072765682752 params.py:248] dataset_reader.token_indexers.feats.type = feats_indexer\n",
"I0407 10:49:51.190932 140072765682752 params.py:248] dataset_reader.token_indexers.feats.namespace = feats\n",
"I0407 10:49:51.191003 140072765682752 params.py:248] dataset_reader.token_indexers.feats.feature_name = feats_\n",
"I0407 10:49:51.191065 140072765682752 params.py:248] dataset_reader.token_indexers.feats.token_min_padding_length = 0\n",
"I0407 10:49:51.191206 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.type = characters_const_padding\n",
"I0407 10:49:51.191369 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.namespace = token_characters\n",
"I0407 10:49:51.191561 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.byte_encoding = None\n",
"I0407 10:49:51.191629 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.lowercase_characters = False\n",
"I0407 10:49:51.191706 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:49:51.191827 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:49:51.191938 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.start_tokens = None\n",
"I0407 10:49:51.191999 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.end_tokens = None\n",
"I0407 10:49:51.192067 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.min_padding_length = 32\n",
"I0407 10:49:51.192142 140072765682752 params.py:248] dataset_reader.token_indexers.lemma.token_min_padding_length = 0\n",
"I0407 10:49:51.192281 140072765682752 params.py:248] dataset_reader.token_indexers.token.type = pretrained_transformer_mismatched_fixed\n",
"I0407 10:49:51.192501 140072765682752 params.py:248] dataset_reader.token_indexers.token.token_min_padding_length = 0\n",
"I0407 10:49:51.192575 140072765682752 params.py:248] dataset_reader.token_indexers.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:49:51.192638 140072765682752 params.py:248] dataset_reader.token_indexers.token.namespace = tags\n",
"I0407 10:49:51.192698 140072765682752 params.py:248] dataset_reader.token_indexers.token.max_length = None\n",
"I0407 10:49:51.192795 140072765682752 params.py:384] dataset_reader.token_indexers.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:49:51.194080 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.type = single_id\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"I0407 10:49:51.194318 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.namespace = upostag\n",
"I0407 10:49:51.194404 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.lowercase_tokens = False\n",
"I0407 10:49:51.194471 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.start_tokens = None\n",
"I0407 10:49:51.194532 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.end_tokens = None\n",
"I0407 10:49:51.194586 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.feature_name = pos_\n",
"I0407 10:49:51.194648 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:49:51.194708 140072765682752 params.py:248] dataset_reader.token_indexers.upostag.token_min_padding_length = 0\n",
"I0407 10:49:51.194854 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.type = single_id\n",
"I0407 10:49:51.195033 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.namespace = xpostag\n",
"I0407 10:49:51.195105 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.lowercase_tokens = False\n",
"I0407 10:49:51.195167 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.start_tokens = None\n",
"I0407 10:49:51.195222 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.end_tokens = None\n",
"I0407 10:49:51.195280 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.feature_name = tag_\n",
"I0407 10:49:51.195338 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:49:51.195398 140072765682752 params.py:248] dataset_reader.token_indexers.xpostag.token_min_padding_length = 0\n",
"I0407 10:49:51.195601 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.type = characters_const_padding\n",
"I0407 10:49:51.195774 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.namespace = lemma_characters\n",
"I0407 10:49:51.195971 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:49:51.196039 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:49:51.196113 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:49:51.196244 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:49:51.196364 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.start_tokens = None\n",
"I0407 10:49:51.196430 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.end_tokens = None\n",
"I0407 10:49:51.196492 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.min_padding_length = 32\n",
"I0407 10:49:51.196552 140072765682752 params.py:248] dataset_reader.lemma_indexers.char.token_min_padding_length = 0\n",
"I0407 10:49:51.196640 140072765682752 params.py:248] dataset_reader.features = ['token', 'char']\n",
"I0407 10:49:51.196732 140072765682752 params.py:248] dataset_reader.targets = ['head', 'deprel']\n",
"I0407 10:49:51.196815 140072765682752 params.py:248] dataset_reader.use_sem = False\n",
"I0407 10:49:51.197346 140072765682752 params.py:248] vocabulary.type = from_instances_extended\n",
"I0407 10:49:51.197421 140072765682752 vocabulary.py:323] Loading token dictionary from /tmp/tmp_htckuhc/vocabulary.\n",
"I0407 10:49:51.197736 140072765682752 filelock.py:254] Lock 140069359832176 acquired on /tmp/tmp_htckuhc/vocabulary/.lock\n",
"I0407 10:49:51.198361 140072765682752 filelock.py:317] Lock 140069359832176 released on /tmp/tmp_htckuhc/vocabulary/.lock\n",
"I0407 10:49:51.198865 140072765682752 params.py:248] model.type = semantic_multitask\n",
"I0407 10:49:51.199399 140072765682752 params.py:248] model.text_field_embedder.type = basic\n",
"I0407 10:49:51.199762 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.type = char_embeddings_from_config\n",
"I0407 10:49:51.199955 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.embedding_dim = 64\n",
"I0407 10:49:51.200206 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.input_dim = 64\n",
"I0407 10:49:51.200286 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.filters = [512, 256, 64]\n",
"I0407 10:49:51.200380 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.kernel_size = [3, 3, 3]\n",
"I0407 10:49:51.200467 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.stride = [1, 1, 1]\n",
"I0407 10:49:51.200556 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.padding = [1, 2, 4]\n",
"I0407 10:49:51.200649 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.dilation = [1, 2, 4]\n",
"I0407 10:49:51.200745 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.activations = ['relu', 'relu', 'linear']\n",
"I0407 10:49:51.200886 140072765682752 params.py:248] type = relu\n",
"I0407 10:49:51.201073 140072765682752 params.py:248] type = relu\n",
"I0407 10:49:51.201222 140072765682752 params.py:248] type = linear\n",
"I0407 10:49:51.208180 140072765682752 params.py:248] model.text_field_embedder.token_embedders.char.vocab_namespace = token_characters\n",
"I0407 10:49:51.208718 140072765682752 params.py:248] model.text_field_embedder.token_embedders.token.type = transformers_word_embeddings\n",
"I0407 10:49:51.208946 140072765682752 params.py:248] model.text_field_embedder.token_embedders.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:49:51.209028 140072765682752 params.py:248] model.text_field_embedder.token_embedders.token.projection_dim = 100\n",
"I0407 10:49:51.209110 140072765682752 params.py:248] model.text_field_embedder.token_embedders.token.projection_activation = <function TransformersWordEmbedder.<lambda> at 0x7f646dd85280>\n",
"I0407 10:49:51.209182 140072765682752 params.py:248] model.text_field_embedder.token_embedders.token.projection_dropout_rate = 0.0\n",
"I0407 10:49:51.209239 140072765682752 params.py:248] model.text_field_embedder.token_embedders.token.freeze_transformer = True\n",
"I0407 10:49:51.209295 140072765682752 params.py:248] model.text_field_embedder.token_embedders.token.last_layer_only = True\n",
"I0407 10:49:51.209401 140072765682752 params.py:384] model.text_field_embedder.token_embedders.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:49:51.209471 140072765682752 params.py:248] model.text_field_embedder.token_embedders.token.transformer_kwargs = None\n",
"I0407 10:49:58.747374 140072765682752 params.py:248] model.seq_encoder.type = combo_encoder\n",
"I0407 10:49:58.747746 140072765682752 params.py:248] model.seq_encoder.stacked_bilstm.input_size = 164\n",
"I0407 10:49:58.747819 140072765682752 params.py:248] model.seq_encoder.stacked_bilstm.hidden_size = 512\n",
"I0407 10:49:58.747869 140072765682752 params.py:248] model.seq_encoder.stacked_bilstm.num_layers = 2\n",
"I0407 10:49:58.747919 140072765682752 params.py:248] model.seq_encoder.stacked_bilstm.recurrent_dropout_probability = 0.33\n",
"I0407 10:49:58.747966 140072765682752 params.py:248] model.seq_encoder.stacked_bilstm.layer_dropout_probability = 0.33\n",
"I0407 10:49:58.748013 140072765682752 params.py:248] model.seq_encoder.stacked_bilstm.use_highway = False\n",
"I0407 10:49:59.084017 140072765682752 params.py:248] model.seq_encoder.layer_dropout_probability = 0.33\n",
"I0407 10:49:59.084280 140072765682752 params.py:248] model.use_sample_weight = True\n",
"I0407 10:49:59.084377 140072765682752 params.py:248] model.lemmatizer = None\n",
"I0407 10:49:59.084436 140072765682752 params.py:248] model.upos_tagger = None\n",
"I0407 10:49:59.084487 140072765682752 params.py:248] model.xpos_tagger = None\n",
"I0407 10:49:59.084537 140072765682752 params.py:248] model.semantic_relation = None\n",
"I0407 10:49:59.084585 140072765682752 params.py:248] model.morphological_feat = None\n",
"I0407 10:49:59.084832 140072765682752 params.py:248] model.dependency_relation.type = combo_dependency_parsing_from_vocab\n",
"I0407 10:49:59.085025 140072765682752 params.py:248] model.dependency_relation.vocab_namespace = deprel_labels\n",
"I0407 10:49:59.085301 140072765682752 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.in_features = 1024\n",
"I0407 10:49:59.085365 140072765682752 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.out_features = 512\n",
"I0407 10:49:59.085421 140072765682752 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.activation = tanh\n",
"I0407 10:49:59.085520 140072765682752 params.py:248] type = tanh\n",
"I0407 10:49:59.085608 140072765682752 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.dropout_rate = 0.0\n",
"I0407 10:49:59.089095 140072765682752 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.in_features = 1024\n",
"I0407 10:49:59.089183 140072765682752 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.out_features = 512\n",
"I0407 10:49:59.089244 140072765682752 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.activation = tanh\n",
"I0407 10:49:59.089346 140072765682752 params.py:248] type = tanh\n",
"I0407 10:49:59.089423 140072765682752 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.dropout_rate = 0.0\n",
"I0407 10:49:59.092701 140072765682752 params.py:248] model.dependency_relation.head_predictor.cycle_loss_n = 0\n",
"I0407 10:49:59.092917 140072765682752 params.py:248] model.dependency_relation.head_projection_layer.in_features = 1024\n",
"I0407 10:49:59.092972 140072765682752 params.py:248] model.dependency_relation.head_projection_layer.out_features = 128\n",
"I0407 10:49:59.093022 140072765682752 params.py:248] model.dependency_relation.head_projection_layer.activation = tanh\n",
"I0407 10:49:59.093108 140072765682752 params.py:248] type = tanh\n",
"I0407 10:49:59.093183 140072765682752 params.py:248] model.dependency_relation.head_projection_layer.dropout_rate = 0.25\n",
"I0407 10:49:59.094336 140072765682752 params.py:248] model.dependency_relation.dependency_projection_layer.in_features = 1024\n",
"I0407 10:49:59.094411 140072765682752 params.py:248] model.dependency_relation.dependency_projection_layer.out_features = 128\n",
"I0407 10:49:59.094463 140072765682752 params.py:248] model.dependency_relation.dependency_projection_layer.activation = tanh\n",
"I0407 10:49:59.094551 140072765682752 params.py:248] type = tanh\n",
"I0407 10:49:59.094618 140072765682752 params.py:248] model.dependency_relation.dependency_projection_layer.dropout_rate = 0.25\n",
"I0407 10:49:59.095806 140072765682752 params.py:248] model.enhanced_dependency_relation = None\n",
"I0407 10:49:59.096206 140072765682752 params.py:248] model.regularizer.regexes.0.1.type = l2\n",
"I0407 10:49:59.096345 140072765682752 params.py:248] model.regularizer.regexes.0.1.alpha = 1e-06\n",
"I0407 10:49:59.096471 140072765682752 params.py:248] model.regularizer.regexes.1.1.type = l2\n",
"I0407 10:49:59.096584 140072765682752 params.py:248] model.regularizer.regexes.1.1.alpha = 1e-06\n",
"I0407 10:49:59.096696 140072765682752 params.py:248] model.regularizer.regexes.2.1.type = l2\n",
"I0407 10:49:59.096809 140072765682752 params.py:248] model.regularizer.regexes.2.1.alpha = 1e-06\n",
"I0407 10:49:59.096917 140072765682752 params.py:248] model.regularizer.regexes.3.1.type = l2\n",
"I0407 10:49:59.097025 140072765682752 params.py:248] model.regularizer.regexes.3.1.alpha = 1e-05\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"I0407 10:50:01.854557 140072765682752 archival.py:211] removing temporary unarchived model dir at /tmp/tmp_htckuhc\n",
"reading instances: 2211it [01:52, 19.69it/s]\n"
]
}
],
"source": [
"! {COMBO} --mode predict \\\n",
" --cuda_device 0 \\\n",
" --model_path model-pdbc/model.tar.gz \\\n",
" --input_file connlu/pdbc-validation.conllu \\\n",
" --output_file connlu/pdbc-validation-pred.conllu"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "11f1b7b1",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# text = Dwie dziewczynki opierają się o dach kapliczki , chłopiec wspina się na niego , a trzecia dziewczynka stoi obok .\r\n",
"1\tDwie\tdwa\t_\t_\t_\t3\tsubj\t3:subj\t_\r\n",
"2\tdziewczynki\tdziewczynka\t_\t_\t_\t1\tcomp\t1:comp\t_\r\n",
"3\topierają\topierać\t_\t_\t_\t15\tconjunct\t15:conjunct\t_\r\n",
"4\tsię\tsię\t_\t_\t_\t3\trefl\t3:refl\t_\r\n",
"5\to\to\t_\t_\t_\t3\tcomp\t3:comp\t_\r\n",
"6\tdach\tdach\t_\t_\t_\t5\tcomp\t5:comp\t_\r\n",
"7\tkapliczki\tkapliczka\t_\t_\t_\t6\tadjunct\t6:adjunct\t_\r\n",
"8\t,\t,\t_\t_\t_\t15\tpunct\t15:punct\t_\r\n",
"9\tchłopiec\tchłopiec\t_\t_\t_\t10\tsubj\t10:subj\t_\r\n"
]
}
],
"source": [
"! head connlu/pdbc-validation.conllu"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "8fa72124",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# text = Dwie dziewczynki opierają się o dach kapliczki , chłopiec wspina się na niego , a trzecia dziewczynka stoi obok .\r\n",
"1\tDwie\tdwa\t_\t_\t_\t3\tsubj\t3:subj\t_\r\n",
"2\tdziewczynki\tdziewczynka\t_\t_\t_\t1\tcomp\t1:comp\t_\r\n",
"3\topierają\topierać\t_\t_\t_\t15\tconjunct\t15:conjunct\t_\r\n",
"4\tsię\tsię\t_\t_\t_\t3\trefl\t3:refl\t_\r\n",
"5\to\to\t_\t_\t_\t3\tcomp\t3:comp\t_\r\n",
"6\tdach\tdach\t_\t_\t_\t5\tcomp\t5:comp\t_\r\n",
"7\tkapliczki\tkapliczka\t_\t_\t_\t6\tadjunct\t6:adjunct\t_\r\n",
"8\t,\t,\t_\t_\t_\t15\tpunct\t15:punct\t_\r\n",
"9\tchłopiec\tchłopiec\t_\t_\t_\t10\tsubj\t10:subj\t_\r\n"
]
}
],
"source": [
"! head connlu/pdbc-validation-pred.conllu"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "dde6dd31",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"I0407 10:52:00.220404 139754138821696 archival.py:184] loading archive file model-pdbc/model.tar.gz\n",
"I0407 10:52:00.221079 139754138821696 archival.py:263] extracting archive file model-pdbc/model.tar.gz to temp dir /tmp/tmp2jhqu3i6\n",
"I0407 10:52:16.996590 139754138821696 params.py:248] dataset_reader.type = conllu\n",
"I0407 10:52:16.997079 139754138821696 params.py:248] dataset_reader.lazy = False\n",
"I0407 10:52:16.997236 139754138821696 params.py:248] dataset_reader.cache_directory = None\n",
"I0407 10:52:16.997326 139754138821696 params.py:248] dataset_reader.max_instances = None\n",
"I0407 10:52:16.997391 139754138821696 params.py:248] dataset_reader.manual_distributed_sharding = False\n",
"I0407 10:52:16.997456 139754138821696 params.py:248] dataset_reader.manual_multi_process_sharding = False\n",
"I0407 10:52:16.997756 139754138821696 params.py:248] dataset_reader.token_indexers.char.type = characters_const_padding\n",
"I0407 10:52:16.997950 139754138821696 params.py:248] dataset_reader.token_indexers.char.namespace = token_characters\n",
"I0407 10:52:16.998211 139754138821696 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:52:16.998285 139754138821696 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:52:16.998367 139754138821696 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:52:16.998522 139754138821696 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:52:16.998643 139754138821696 params.py:248] dataset_reader.token_indexers.char.start_tokens = None\n",
"I0407 10:52:16.998707 139754138821696 params.py:248] dataset_reader.token_indexers.char.end_tokens = None\n",
"I0407 10:52:16.998770 139754138821696 params.py:248] dataset_reader.token_indexers.char.min_padding_length = 32\n",
"I0407 10:52:16.998831 139754138821696 params.py:248] dataset_reader.token_indexers.char.token_min_padding_length = 0\n",
"I0407 10:52:16.998980 139754138821696 params.py:248] dataset_reader.token_indexers.feats.type = feats_indexer\n",
"I0407 10:52:16.999143 139754138821696 params.py:248] dataset_reader.token_indexers.feats.namespace = feats\n",
"I0407 10:52:16.999213 139754138821696 params.py:248] dataset_reader.token_indexers.feats.feature_name = feats_\n",
"I0407 10:52:16.999269 139754138821696 params.py:248] dataset_reader.token_indexers.feats.token_min_padding_length = 0\n",
"I0407 10:52:16.999412 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.type = characters_const_padding\n",
"I0407 10:52:16.999578 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.namespace = token_characters\n",
"I0407 10:52:16.999774 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.byte_encoding = None\n",
"I0407 10:52:16.999842 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.lowercase_characters = False\n",
"I0407 10:52:16.999923 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:52:17.000045 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:52:17.000156 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.start_tokens = None\n",
"I0407 10:52:17.000220 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.end_tokens = None\n",
"I0407 10:52:17.000282 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.min_padding_length = 32\n",
"I0407 10:52:17.000344 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.token_min_padding_length = 0\n",
"I0407 10:52:17.000521 139754138821696 params.py:248] dataset_reader.token_indexers.token.type = pretrained_transformer_mismatched_fixed\n",
"I0407 10:52:17.000770 139754138821696 params.py:248] dataset_reader.token_indexers.token.token_min_padding_length = 0\n",
"I0407 10:52:17.000865 139754138821696 params.py:248] dataset_reader.token_indexers.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:52:17.000947 139754138821696 params.py:248] dataset_reader.token_indexers.token.namespace = tags\n",
"I0407 10:52:17.001028 139754138821696 params.py:248] dataset_reader.token_indexers.token.max_length = None\n",
"I0407 10:52:17.001172 139754138821696 params.py:384] dataset_reader.token_indexers.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:52:20.459573 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.type = single_id\n",
"I0407 10:52:20.459947 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.namespace = upostag\n",
"I0407 10:52:20.460046 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.lowercase_tokens = False\n",
"I0407 10:52:20.460119 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.start_tokens = None\n",
"I0407 10:52:20.460172 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.end_tokens = None\n",
"I0407 10:52:20.460235 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.feature_name = pos_\n",
"I0407 10:52:20.460288 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:52:20.460351 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.token_min_padding_length = 0\n",
"I0407 10:52:20.460508 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.type = single_id\n",
"I0407 10:52:20.460695 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.namespace = xpostag\n",
"I0407 10:52:20.460773 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.lowercase_tokens = False\n",
"I0407 10:52:20.460840 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.start_tokens = None\n",
"I0407 10:52:20.460901 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.end_tokens = None\n",
"I0407 10:52:20.460962 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.feature_name = tag_\n",
"I0407 10:52:20.461021 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:52:20.461083 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.token_min_padding_length = 0\n",
"I0407 10:52:20.461313 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.type = characters_const_padding\n",
"I0407 10:52:20.461496 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.namespace = lemma_characters\n",
"I0407 10:52:20.461706 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:52:20.461774 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:52:20.461853 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:52:20.462028 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:52:20.462157 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.start_tokens = None\n",
"I0407 10:52:20.462226 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.end_tokens = None\n",
"I0407 10:52:20.462283 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.min_padding_length = 32\n",
"I0407 10:52:20.462336 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.token_min_padding_length = 0\n",
"I0407 10:52:20.462417 139754138821696 params.py:248] dataset_reader.features = ['token', 'char']\n",
"I0407 10:52:20.462514 139754138821696 params.py:248] dataset_reader.targets = ['head', 'deprel']\n",
"I0407 10:52:20.462607 139754138821696 params.py:248] dataset_reader.use_sem = False\n",
"I0407 10:52:20.462767 139754138821696 params.py:248] dataset_reader.type = conllu\n",
"I0407 10:52:20.463083 139754138821696 params.py:248] dataset_reader.lazy = False\n",
"I0407 10:52:20.463172 139754138821696 params.py:248] dataset_reader.cache_directory = None\n",
"I0407 10:52:20.463237 139754138821696 params.py:248] dataset_reader.max_instances = None\n",
"I0407 10:52:20.463301 139754138821696 params.py:248] dataset_reader.manual_distributed_sharding = False\n",
"I0407 10:52:20.463361 139754138821696 params.py:248] dataset_reader.manual_multi_process_sharding = False\n",
"I0407 10:52:20.463605 139754138821696 params.py:248] dataset_reader.token_indexers.char.type = characters_const_padding\n",
"I0407 10:52:20.463779 139754138821696 params.py:248] dataset_reader.token_indexers.char.namespace = token_characters\n",
"I0407 10:52:20.463980 139754138821696 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:52:20.464051 139754138821696 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:52:20.464129 139754138821696 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:52:20.464254 139754138821696 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:52:20.464366 139754138821696 params.py:248] dataset_reader.token_indexers.char.start_tokens = None\n",
"I0407 10:52:20.464429 139754138821696 params.py:248] dataset_reader.token_indexers.char.end_tokens = None\n",
"I0407 10:52:20.464490 139754138821696 params.py:248] dataset_reader.token_indexers.char.min_padding_length = 32\n",
"I0407 10:52:20.464552 139754138821696 params.py:248] dataset_reader.token_indexers.char.token_min_padding_length = 0\n",
"I0407 10:52:20.464691 139754138821696 params.py:248] dataset_reader.token_indexers.feats.type = feats_indexer\n",
"I0407 10:52:20.464847 139754138821696 params.py:248] dataset_reader.token_indexers.feats.namespace = feats\n",
"I0407 10:52:20.464918 139754138821696 params.py:248] dataset_reader.token_indexers.feats.feature_name = feats_\n",
"I0407 10:52:20.464980 139754138821696 params.py:248] dataset_reader.token_indexers.feats.token_min_padding_length = 0\n",
"I0407 10:52:20.465120 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.type = characters_const_padding\n",
"I0407 10:52:20.465285 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.namespace = token_characters\n",
"I0407 10:52:20.465479 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.byte_encoding = None\n",
"I0407 10:52:20.465544 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.lowercase_characters = False\n",
"I0407 10:52:20.465618 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:52:20.465741 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:52:20.465851 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.start_tokens = None\n",
"I0407 10:52:20.465914 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.end_tokens = None\n",
"I0407 10:52:20.466024 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.min_padding_length = 32\n",
"I0407 10:52:20.466112 139754138821696 params.py:248] dataset_reader.token_indexers.lemma.token_min_padding_length = 0\n",
"I0407 10:52:20.466268 139754138821696 params.py:248] dataset_reader.token_indexers.token.type = pretrained_transformer_mismatched_fixed\n",
"I0407 10:52:20.466485 139754138821696 params.py:248] dataset_reader.token_indexers.token.token_min_padding_length = 0\n",
"I0407 10:52:20.466559 139754138821696 params.py:248] dataset_reader.token_indexers.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:52:20.466621 139754138821696 params.py:248] dataset_reader.token_indexers.token.namespace = tags\n",
"I0407 10:52:20.466682 139754138821696 params.py:248] dataset_reader.token_indexers.token.max_length = None\n",
"I0407 10:52:20.466777 139754138821696 params.py:384] dataset_reader.token_indexers.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:52:20.468071 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.type = single_id\n",
"I0407 10:52:20.468319 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.namespace = upostag\n",
"I0407 10:52:20.468404 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.lowercase_tokens = False\n",
"I0407 10:52:20.468464 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.start_tokens = None\n",
"I0407 10:52:20.468523 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.end_tokens = None\n",
"I0407 10:52:20.468573 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.feature_name = pos_\n",
"I0407 10:52:20.468636 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:52:20.468697 139754138821696 params.py:248] dataset_reader.token_indexers.upostag.token_min_padding_length = 0\n",
"I0407 10:52:20.468832 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.type = single_id\n",
"I0407 10:52:20.469012 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.namespace = xpostag\n",
"I0407 10:52:20.469086 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.lowercase_tokens = False\n",
"I0407 10:52:20.469144 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.start_tokens = None\n",
"I0407 10:52:20.469196 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.end_tokens = None\n",
"I0407 10:52:20.469256 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.feature_name = tag_\n",
"I0407 10:52:20.469320 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:52:20.469382 139754138821696 params.py:248] dataset_reader.token_indexers.xpostag.token_min_padding_length = 0\n",
"I0407 10:52:20.469586 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.type = characters_const_padding\n",
"I0407 10:52:20.469758 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.namespace = lemma_characters\n",
"I0407 10:52:20.469957 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:52:20.470050 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:52:20.470128 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:52:20.470261 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:52:20.470381 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.start_tokens = None\n",
"I0407 10:52:20.470448 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.end_tokens = None\n",
"I0407 10:52:20.470509 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.min_padding_length = 32\n",
"I0407 10:52:20.470579 139754138821696 params.py:248] dataset_reader.lemma_indexers.char.token_min_padding_length = 0\n",
"I0407 10:52:20.470668 139754138821696 params.py:248] dataset_reader.features = ['token', 'char']\n",
"I0407 10:52:20.470764 139754138821696 params.py:248] dataset_reader.targets = ['head', 'deprel']\n",
"I0407 10:52:20.470849 139754138821696 params.py:248] dataset_reader.use_sem = False\n",
"I0407 10:52:20.471387 139754138821696 params.py:248] vocabulary.type = from_instances_extended\n",
"I0407 10:52:20.471461 139754138821696 vocabulary.py:323] Loading token dictionary from /tmp/tmp2jhqu3i6/vocabulary.\n",
"I0407 10:52:20.471798 139754138821696 filelock.py:254] Lock 139750732975216 acquired on /tmp/tmp2jhqu3i6/vocabulary/.lock\n",
"I0407 10:52:20.472387 139754138821696 filelock.py:317] Lock 139750732975216 released on /tmp/tmp2jhqu3i6/vocabulary/.lock\n",
"I0407 10:52:20.472922 139754138821696 params.py:248] model.type = semantic_multitask\n",
"I0407 10:52:20.473455 139754138821696 params.py:248] model.text_field_embedder.type = basic\n",
"I0407 10:52:20.473808 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.type = char_embeddings_from_config\n",
"I0407 10:52:20.474030 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.embedding_dim = 64\n",
"I0407 10:52:20.474286 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.input_dim = 64\n",
"I0407 10:52:20.474377 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.filters = [512, 256, 64]\n",
"I0407 10:52:20.474480 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.kernel_size = [3, 3, 3]\n",
"I0407 10:52:20.474578 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.stride = [1, 1, 1]\n",
"I0407 10:52:20.474673 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.padding = [1, 2, 4]\n",
"I0407 10:52:20.474768 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.dilation = [1, 2, 4]\n",
"I0407 10:52:20.474864 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.activations = ['relu', 'relu', 'linear']\n",
"I0407 10:52:20.475005 139754138821696 params.py:248] type = relu\n",
"I0407 10:52:20.475197 139754138821696 params.py:248] type = relu\n",
"I0407 10:52:20.475347 139754138821696 params.py:248] type = linear\n",
"I0407 10:52:20.481609 139754138821696 params.py:248] model.text_field_embedder.token_embedders.char.vocab_namespace = token_characters\n",
"I0407 10:52:20.482178 139754138821696 params.py:248] model.text_field_embedder.token_embedders.token.type = transformers_word_embeddings\n",
"I0407 10:52:20.482446 139754138821696 params.py:248] model.text_field_embedder.token_embedders.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:52:20.482533 139754138821696 params.py:248] model.text_field_embedder.token_embedders.token.projection_dim = 100\n",
"I0407 10:52:20.482632 139754138821696 params.py:248] model.text_field_embedder.token_embedders.token.projection_activation = <function TransformersWordEmbedder.<lambda> at 0x7f1a3e346280>\n",
"I0407 10:52:20.482703 139754138821696 params.py:248] model.text_field_embedder.token_embedders.token.projection_dropout_rate = 0.0\n",
"I0407 10:52:20.482769 139754138821696 params.py:248] model.text_field_embedder.token_embedders.token.freeze_transformer = True\n",
"I0407 10:52:20.482831 139754138821696 params.py:248] model.text_field_embedder.token_embedders.token.last_layer_only = True\n",
"I0407 10:52:20.482933 139754138821696 params.py:384] model.text_field_embedder.token_embedders.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:52:20.483003 139754138821696 params.py:248] model.text_field_embedder.token_embedders.token.transformer_kwargs = None\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"I0407 10:52:28.699278 139754138821696 params.py:248] model.seq_encoder.type = combo_encoder\n",
"I0407 10:52:28.699747 139754138821696 params.py:248] model.seq_encoder.stacked_bilstm.input_size = 164\n",
"I0407 10:52:28.699841 139754138821696 params.py:248] model.seq_encoder.stacked_bilstm.hidden_size = 512\n",
"I0407 10:52:28.699910 139754138821696 params.py:248] model.seq_encoder.stacked_bilstm.num_layers = 2\n",
"I0407 10:52:28.699976 139754138821696 params.py:248] model.seq_encoder.stacked_bilstm.recurrent_dropout_probability = 0.33\n",
"I0407 10:52:28.700042 139754138821696 params.py:248] model.seq_encoder.stacked_bilstm.layer_dropout_probability = 0.33\n",
"I0407 10:52:28.700106 139754138821696 params.py:248] model.seq_encoder.stacked_bilstm.use_highway = False\n",
"I0407 10:52:29.089101 139754138821696 params.py:248] model.seq_encoder.layer_dropout_probability = 0.33\n",
"I0407 10:52:29.089426 139754138821696 params.py:248] model.use_sample_weight = True\n",
"I0407 10:52:29.089556 139754138821696 params.py:248] model.lemmatizer = None\n",
"I0407 10:52:29.089638 139754138821696 params.py:248] model.upos_tagger = None\n",
"I0407 10:52:29.089704 139754138821696 params.py:248] model.xpos_tagger = None\n",
"I0407 10:52:29.089766 139754138821696 params.py:248] model.semantic_relation = None\n",
"I0407 10:52:29.089827 139754138821696 params.py:248] model.morphological_feat = None\n",
"I0407 10:52:29.090160 139754138821696 params.py:248] model.dependency_relation.type = combo_dependency_parsing_from_vocab\n",
"I0407 10:52:29.090409 139754138821696 params.py:248] model.dependency_relation.vocab_namespace = deprel_labels\n",
"I0407 10:52:29.090762 139754138821696 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.in_features = 1024\n",
"I0407 10:52:29.090843 139754138821696 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.out_features = 512\n",
"I0407 10:52:29.090915 139754138821696 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.activation = tanh\n",
"I0407 10:52:29.091041 139754138821696 params.py:248] type = tanh\n",
"I0407 10:52:29.091149 139754138821696 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.dropout_rate = 0.0\n",
"I0407 10:52:29.096003 139754138821696 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.in_features = 1024\n",
"I0407 10:52:29.096106 139754138821696 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.out_features = 512\n",
"I0407 10:52:29.096185 139754138821696 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.activation = tanh\n",
"I0407 10:52:29.096311 139754138821696 params.py:248] type = tanh\n",
"I0407 10:52:29.096407 139754138821696 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.dropout_rate = 0.0\n",
"I0407 10:52:29.101276 139754138821696 params.py:248] model.dependency_relation.head_predictor.cycle_loss_n = 0\n",
"I0407 10:52:29.101581 139754138821696 params.py:248] model.dependency_relation.head_projection_layer.in_features = 1024\n",
"I0407 10:52:29.101692 139754138821696 params.py:248] model.dependency_relation.head_projection_layer.out_features = 128\n",
"I0407 10:52:29.101771 139754138821696 params.py:248] model.dependency_relation.head_projection_layer.activation = tanh\n",
"I0407 10:52:29.101904 139754138821696 params.py:248] type = tanh\n",
"I0407 10:52:29.102032 139754138821696 params.py:248] model.dependency_relation.head_projection_layer.dropout_rate = 0.25\n",
"I0407 10:52:29.103649 139754138821696 params.py:248] model.dependency_relation.dependency_projection_layer.in_features = 1024\n",
"I0407 10:52:29.103747 139754138821696 params.py:248] model.dependency_relation.dependency_projection_layer.out_features = 128\n",
"I0407 10:52:29.103819 139754138821696 params.py:248] model.dependency_relation.dependency_projection_layer.activation = tanh\n",
"I0407 10:52:29.103948 139754138821696 params.py:248] type = tanh\n",
"I0407 10:52:29.104044 139754138821696 params.py:248] model.dependency_relation.dependency_projection_layer.dropout_rate = 0.25\n",
"I0407 10:52:29.105780 139754138821696 params.py:248] model.enhanced_dependency_relation = None\n",
"I0407 10:52:29.106371 139754138821696 params.py:248] model.regularizer.regexes.0.1.type = l2\n",
"I0407 10:52:29.106555 139754138821696 params.py:248] model.regularizer.regexes.0.1.alpha = 1e-06\n",
"I0407 10:52:29.106724 139754138821696 params.py:248] model.regularizer.regexes.1.1.type = l2\n",
"I0407 10:52:29.106879 139754138821696 params.py:248] model.regularizer.regexes.1.1.alpha = 1e-06\n",
"I0407 10:52:29.107035 139754138821696 params.py:248] model.regularizer.regexes.2.1.type = l2\n",
"I0407 10:52:29.107207 139754138821696 params.py:248] model.regularizer.regexes.2.1.alpha = 1e-06\n",
"I0407 10:52:29.107368 139754138821696 params.py:248] model.regularizer.regexes.3.1.type = l2\n",
"I0407 10:52:29.107544 139754138821696 params.py:248] model.regularizer.regexes.3.1.alpha = 1e-05\n",
"I0407 10:52:32.063793 139754138821696 archival.py:211] removing temporary unarchived model dir at /tmp/tmp2jhqu3i6\n",
"reading instances: 2205it [01:49, 20.15it/s]\n"
]
}
],
"source": [
"! {COMBO} --mode predict \\\n",
" --cuda_device 0 \\\n",
" --model_path model-pdbc/model.tar.gz \\\n",
" --input_file connlu/pdbc-test.conllu \\\n",
" --output_file connlu/pdbc-test-pred.conllu"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "13748ca1",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# text = Mały chłopiec patrzy w bok po ściągnięciu okularów .\r\n",
"1\tMały\tmały\t_\t_\t_\t2\tadjunct\t2:adjunct\t_\r\n",
"2\tchłopiec\tchłopiec\t_\t_\t_\t3\tsubj\t3:subj\t_\r\n",
"3\tpatrzy\tpatrzeć\t_\t_\t_\t0\troot\t0:root\t_\r\n",
"4\tw\tw\t_\t_\t_\t3\tadjunct_adl\t3:adjunct_adl\t_\r\n",
"5\tbok\tbok\t_\t_\t_\t4\tcomp\t4:comp\t_\r\n",
"6\tpo\tpo\t_\t_\t_\t3\tadjunct_temp\t3:adjunct_temp\t_\r\n",
"7\tściągnięciu\tściągnąć\t_\t_\t_\t6\tcomp\t6:comp\t_\r\n",
"8\tokularów\tokulary\t_\t_\t_\t7\tobj\t7:obj\t_\r\n",
"9\t.\t.\t_\t_\t_\t3\tpunct\t3:punct\t_\r\n"
]
}
],
"source": [
"! head connlu/pdbc-test.conllu"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "30021124",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# text = Mały chłopiec patrzy w bok po ściągnięciu okularów .\r\n",
"1\tMały\tmały\t_\t_\t_\t2\tadjunct\t2:adjunct\t_\r\n",
"2\tchłopiec\tchłopiec\t_\t_\t_\t3\tsubj\t3:subj\t_\r\n",
"3\tpatrzy\tpatrzeć\t_\t_\t_\t0\troot\t0:root\t_\r\n",
"4\tw\tw\t_\t_\t_\t3\tcomp\t3:adjunct_adl\t_\r\n",
"5\tbok\tbok\t_\t_\t_\t4\tcomp\t4:comp\t_\r\n",
"6\tpo\tpo\t_\t_\t_\t3\tadjunct_temp\t3:adjunct_temp\t_\r\n",
"7\tściągnięciu\tściągnąć\t_\t_\t_\t6\tcomp\t6:comp\t_\r\n",
"8\tokularów\tokulary\t_\t_\t_\t7\tobj\t7:obj\t_\r\n",
"9\t.\t.\t_\t_\t_\t3\tpunct\t3:punct\t_\r\n"
]
}
],
"source": [
"! head connlu/pdbc-test-pred.conllu"
]
},
{
"cell_type": "markdown",
"id": "99359d8c",
"metadata": {},
"source": [
"Continuous-only model"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "30a66da6",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"I0407 10:54:27.401382 140321380496448 archival.py:184] loading archive file model-pdbc-cont/model.tar.gz\n",
"I0407 10:54:27.402150 140321380496448 archival.py:263] extracting archive file model-pdbc-cont/model.tar.gz to temp dir /tmp/tmpuvesoi4q\n",
"I0407 10:54:43.091615 140321380496448 params.py:248] dataset_reader.type = conllu\n",
"I0407 10:54:43.092000 140321380496448 params.py:248] dataset_reader.lazy = False\n",
"I0407 10:54:43.092082 140321380496448 params.py:248] dataset_reader.cache_directory = None\n",
"I0407 10:54:43.092129 140321380496448 params.py:248] dataset_reader.max_instances = None\n",
"I0407 10:54:43.092173 140321380496448 params.py:248] dataset_reader.manual_distributed_sharding = False\n",
"I0407 10:54:43.092208 140321380496448 params.py:248] dataset_reader.manual_multi_process_sharding = False\n",
"I0407 10:54:43.092409 140321380496448 params.py:248] dataset_reader.token_indexers.char.type = characters_const_padding\n",
"I0407 10:54:43.092535 140321380496448 params.py:248] dataset_reader.token_indexers.char.namespace = token_characters\n",
"I0407 10:54:43.092682 140321380496448 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:54:43.092730 140321380496448 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:54:43.092786 140321380496448 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:54:43.092888 140321380496448 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:54:43.092970 140321380496448 params.py:248] dataset_reader.token_indexers.char.start_tokens = None\n",
"I0407 10:54:43.093014 140321380496448 params.py:248] dataset_reader.token_indexers.char.end_tokens = None\n",
"I0407 10:54:43.093051 140321380496448 params.py:248] dataset_reader.token_indexers.char.min_padding_length = 32\n",
"I0407 10:54:43.093093 140321380496448 params.py:248] dataset_reader.token_indexers.char.token_min_padding_length = 0\n",
"I0407 10:54:43.093198 140321380496448 params.py:248] dataset_reader.token_indexers.feats.type = feats_indexer\n",
"I0407 10:54:43.093306 140321380496448 params.py:248] dataset_reader.token_indexers.feats.namespace = feats\n",
"I0407 10:54:43.093353 140321380496448 params.py:248] dataset_reader.token_indexers.feats.feature_name = feats_\n",
"I0407 10:54:43.093388 140321380496448 params.py:248] dataset_reader.token_indexers.feats.token_min_padding_length = 0\n",
"I0407 10:54:43.093482 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.type = characters_const_padding\n",
"I0407 10:54:43.093593 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.namespace = token_characters\n",
"I0407 10:54:43.093723 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.byte_encoding = None\n",
"I0407 10:54:43.093769 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.lowercase_characters = False\n",
"I0407 10:54:43.093816 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:54:43.093899 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:54:43.093993 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.start_tokens = None\n",
"I0407 10:54:43.094043 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.end_tokens = None\n",
"I0407 10:54:43.094079 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.min_padding_length = 32\n",
"I0407 10:54:43.094121 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.token_min_padding_length = 0\n",
"I0407 10:54:43.094226 140321380496448 params.py:248] dataset_reader.token_indexers.token.type = pretrained_transformer_mismatched_fixed\n",
"I0407 10:54:43.094377 140321380496448 params.py:248] dataset_reader.token_indexers.token.token_min_padding_length = 0\n",
"I0407 10:54:43.094430 140321380496448 params.py:248] dataset_reader.token_indexers.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:54:43.094474 140321380496448 params.py:248] dataset_reader.token_indexers.token.namespace = tags\n",
"I0407 10:54:43.094522 140321380496448 params.py:248] dataset_reader.token_indexers.token.max_length = None\n",
"I0407 10:54:43.094592 140321380496448 params.py:384] dataset_reader.token_indexers.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:54:45.858621 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.type = single_id\n",
"I0407 10:54:45.858990 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.namespace = upostag\n",
"I0407 10:54:45.859087 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.lowercase_tokens = False\n",
"I0407 10:54:45.859157 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.start_tokens = None\n",
"I0407 10:54:45.859210 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.end_tokens = None\n",
"I0407 10:54:45.859268 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.feature_name = pos_\n",
"I0407 10:54:45.859321 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:54:45.859382 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.token_min_padding_length = 0\n",
"I0407 10:54:45.859541 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.type = single_id\n",
"I0407 10:54:45.859729 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.namespace = xpostag\n",
"I0407 10:54:45.859802 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.lowercase_tokens = False\n",
"I0407 10:54:45.859875 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.start_tokens = None\n",
"I0407 10:54:45.859931 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.end_tokens = None\n",
"I0407 10:54:45.859991 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.feature_name = tag_\n",
"I0407 10:54:45.860045 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:54:45.860103 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.token_min_padding_length = 0\n",
"I0407 10:54:45.860332 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.type = characters_const_padding\n",
"I0407 10:54:45.860523 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.namespace = lemma_characters\n",
"I0407 10:54:45.860739 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:54:45.860809 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:54:45.860888 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:54:45.861032 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:54:45.861149 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.start_tokens = None\n",
"I0407 10:54:45.861213 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.end_tokens = None\n",
"I0407 10:54:45.861277 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.min_padding_length = 32\n",
"I0407 10:54:45.861337 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.token_min_padding_length = 0\n",
"I0407 10:54:45.861427 140321380496448 params.py:248] dataset_reader.features = ['token', 'char']\n",
"I0407 10:54:45.861522 140321380496448 params.py:248] dataset_reader.targets = ['head', 'deprel']\n",
"I0407 10:54:45.861611 140321380496448 params.py:248] dataset_reader.use_sem = False\n",
"I0407 10:54:45.861762 140321380496448 params.py:248] dataset_reader.type = conllu\n",
"I0407 10:54:45.862029 140321380496448 params.py:248] dataset_reader.lazy = False\n",
"I0407 10:54:45.862116 140321380496448 params.py:248] dataset_reader.cache_directory = None\n",
"I0407 10:54:45.862177 140321380496448 params.py:248] dataset_reader.max_instances = None\n",
"I0407 10:54:45.862234 140321380496448 params.py:248] dataset_reader.manual_distributed_sharding = False\n",
"I0407 10:54:45.862295 140321380496448 params.py:248] dataset_reader.manual_multi_process_sharding = False\n",
"I0407 10:54:45.862535 140321380496448 params.py:248] dataset_reader.token_indexers.char.type = characters_const_padding\n",
"I0407 10:54:45.862701 140321380496448 params.py:248] dataset_reader.token_indexers.char.namespace = token_characters\n",
"I0407 10:54:45.862900 140321380496448 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:54:45.862966 140321380496448 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:54:45.863043 140321380496448 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:54:45.863168 140321380496448 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:54:45.863281 140321380496448 params.py:248] dataset_reader.token_indexers.char.start_tokens = None\n",
"I0407 10:54:45.863344 140321380496448 params.py:248] dataset_reader.token_indexers.char.end_tokens = None\n",
"I0407 10:54:45.863406 140321380496448 params.py:248] dataset_reader.token_indexers.char.min_padding_length = 32\n",
"I0407 10:54:45.863469 140321380496448 params.py:248] dataset_reader.token_indexers.char.token_min_padding_length = 0\n",
"I0407 10:54:45.863596 140321380496448 params.py:248] dataset_reader.token_indexers.feats.type = feats_indexer\n",
"I0407 10:54:45.863752 140321380496448 params.py:248] dataset_reader.token_indexers.feats.namespace = feats\n",
"I0407 10:54:45.863821 140321380496448 params.py:248] dataset_reader.token_indexers.feats.feature_name = feats_\n",
"I0407 10:54:45.863883 140321380496448 params.py:248] dataset_reader.token_indexers.feats.token_min_padding_length = 0\n",
"I0407 10:54:45.864030 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.type = characters_const_padding\n",
"I0407 10:54:45.864196 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.namespace = token_characters\n",
"I0407 10:54:45.864392 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.byte_encoding = None\n",
"I0407 10:54:45.864460 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.lowercase_characters = False\n",
"I0407 10:54:45.864540 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:54:45.864660 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:54:45.864772 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.start_tokens = None\n",
"I0407 10:54:45.864835 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.end_tokens = None\n",
"I0407 10:54:45.864896 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.min_padding_length = 32\n",
"I0407 10:54:45.864965 140321380496448 params.py:248] dataset_reader.token_indexers.lemma.token_min_padding_length = 0\n",
"I0407 10:54:45.865104 140321380496448 params.py:248] dataset_reader.token_indexers.token.type = pretrained_transformer_mismatched_fixed\n",
"I0407 10:54:45.865323 140321380496448 params.py:248] dataset_reader.token_indexers.token.token_min_padding_length = 0\n",
"I0407 10:54:45.865396 140321380496448 params.py:248] dataset_reader.token_indexers.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:54:45.865460 140321380496448 params.py:248] dataset_reader.token_indexers.token.namespace = tags\n",
"I0407 10:54:45.865518 140321380496448 params.py:248] dataset_reader.token_indexers.token.max_length = None\n",
"I0407 10:54:45.865614 140321380496448 params.py:384] dataset_reader.token_indexers.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:54:45.866884 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.type = single_id\n",
"I0407 10:54:45.867116 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.namespace = upostag\n",
"I0407 10:54:45.867190 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.lowercase_tokens = False\n",
"I0407 10:54:45.867258 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.start_tokens = None\n",
"I0407 10:54:45.867316 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.end_tokens = None\n",
"I0407 10:54:45.867376 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.feature_name = pos_\n",
"I0407 10:54:45.867437 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:54:45.867497 140321380496448 params.py:248] dataset_reader.token_indexers.upostag.token_min_padding_length = 0\n",
"I0407 10:54:45.867640 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.type = single_id\n",
"I0407 10:54:45.867815 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.namespace = xpostag\n",
"I0407 10:54:45.867887 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.lowercase_tokens = False\n",
"I0407 10:54:45.867951 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.start_tokens = None\n",
"I0407 10:54:45.868006 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.end_tokens = None\n",
"I0407 10:54:45.868063 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.feature_name = tag_\n",
"I0407 10:54:45.868122 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:54:45.868181 140321380496448 params.py:248] dataset_reader.token_indexers.xpostag.token_min_padding_length = 0\n",
"I0407 10:54:45.868388 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.type = characters_const_padding\n",
"I0407 10:54:45.868559 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.namespace = lemma_characters\n",
"I0407 10:54:45.868757 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:54:45.868824 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:54:45.868897 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:54:45.869028 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:54:45.869139 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.start_tokens = None\n",
"I0407 10:54:45.869202 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.end_tokens = None\n",
"I0407 10:54:45.869256 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.min_padding_length = 32\n",
"I0407 10:54:45.869315 140321380496448 params.py:248] dataset_reader.lemma_indexers.char.token_min_padding_length = 0\n",
"I0407 10:54:45.869398 140321380496448 params.py:248] dataset_reader.features = ['token', 'char']\n",
"I0407 10:54:45.869489 140321380496448 params.py:248] dataset_reader.targets = ['head', 'deprel']\n",
"I0407 10:54:45.869572 140321380496448 params.py:248] dataset_reader.use_sem = False\n",
"I0407 10:54:45.870136 140321380496448 params.py:248] vocabulary.type = from_instances_extended\n",
"I0407 10:54:45.870218 140321380496448 vocabulary.py:323] Loading token dictionary from /tmp/tmpuvesoi4q/vocabulary.\n",
"I0407 10:54:45.870543 140321380496448 filelock.py:254] Lock 140317974842768 acquired on /tmp/tmpuvesoi4q/vocabulary/.lock\n",
"I0407 10:54:45.871132 140321380496448 filelock.py:317] Lock 140317974842768 released on /tmp/tmpuvesoi4q/vocabulary/.lock\n",
"I0407 10:54:45.871641 140321380496448 params.py:248] model.type = semantic_multitask\n",
"I0407 10:54:45.872183 140321380496448 params.py:248] model.text_field_embedder.type = basic\n",
"I0407 10:54:45.872548 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.type = char_embeddings_from_config\n",
"I0407 10:54:45.872749 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.embedding_dim = 64\n",
"I0407 10:54:45.873004 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.input_dim = 64\n",
"I0407 10:54:45.873091 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.filters = [512, 256, 64]\n",
"I0407 10:54:45.873195 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.kernel_size = [3, 3, 3]\n",
"I0407 10:54:45.873291 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.stride = [1, 1, 1]\n",
"I0407 10:54:45.873384 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.padding = [1, 2, 4]\n",
"I0407 10:54:45.873478 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.dilation = [1, 2, 4]\n",
"I0407 10:54:45.873572 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.activations = ['relu', 'relu', 'linear']\n",
"I0407 10:54:45.873714 140321380496448 params.py:248] type = relu\n",
"I0407 10:54:45.873904 140321380496448 params.py:248] type = relu\n",
"I0407 10:54:45.874098 140321380496448 params.py:248] type = linear\n",
"I0407 10:54:45.880232 140321380496448 params.py:248] model.text_field_embedder.token_embedders.char.vocab_namespace = token_characters\n",
"I0407 10:54:45.880783 140321380496448 params.py:248] model.text_field_embedder.token_embedders.token.type = transformers_word_embeddings\n",
"I0407 10:54:45.881011 140321380496448 params.py:248] model.text_field_embedder.token_embedders.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:54:45.881093 140321380496448 params.py:248] model.text_field_embedder.token_embedders.token.projection_dim = 100\n",
"I0407 10:54:45.881184 140321380496448 params.py:248] model.text_field_embedder.token_embedders.token.projection_activation = <function TransformersWordEmbedder.<lambda> at 0x7f9e50745280>\n",
"I0407 10:54:45.881261 140321380496448 params.py:248] model.text_field_embedder.token_embedders.token.projection_dropout_rate = 0.0\n",
"I0407 10:54:45.881328 140321380496448 params.py:248] model.text_field_embedder.token_embedders.token.freeze_transformer = True\n",
"I0407 10:54:45.881389 140321380496448 params.py:248] model.text_field_embedder.token_embedders.token.last_layer_only = True\n",
"I0407 10:54:45.881492 140321380496448 params.py:384] model.text_field_embedder.token_embedders.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:54:45.881562 140321380496448 params.py:248] model.text_field_embedder.token_embedders.token.transformer_kwargs = None\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"I0407 10:54:52.911276 140321380496448 params.py:248] model.seq_encoder.type = combo_encoder\n",
"I0407 10:54:52.911743 140321380496448 params.py:248] model.seq_encoder.stacked_bilstm.input_size = 164\n",
"I0407 10:54:52.911836 140321380496448 params.py:248] model.seq_encoder.stacked_bilstm.hidden_size = 512\n",
"I0407 10:54:52.911902 140321380496448 params.py:248] model.seq_encoder.stacked_bilstm.num_layers = 2\n",
"I0407 10:54:52.911965 140321380496448 params.py:248] model.seq_encoder.stacked_bilstm.recurrent_dropout_probability = 0.33\n",
"I0407 10:54:52.912029 140321380496448 params.py:248] model.seq_encoder.stacked_bilstm.layer_dropout_probability = 0.33\n",
"I0407 10:54:52.912090 140321380496448 params.py:248] model.seq_encoder.stacked_bilstm.use_highway = False\n",
"I0407 10:54:53.279199 140321380496448 params.py:248] model.seq_encoder.layer_dropout_probability = 0.33\n",
"I0407 10:54:53.279505 140321380496448 params.py:248] model.use_sample_weight = True\n",
"I0407 10:54:53.279624 140321380496448 params.py:248] model.lemmatizer = None\n",
"I0407 10:54:53.279695 140321380496448 params.py:248] model.upos_tagger = None\n",
"I0407 10:54:53.279757 140321380496448 params.py:248] model.xpos_tagger = None\n",
"I0407 10:54:53.279815 140321380496448 params.py:248] model.semantic_relation = None\n",
"I0407 10:54:53.279873 140321380496448 params.py:248] model.morphological_feat = None\n",
"I0407 10:54:53.280155 140321380496448 params.py:248] model.dependency_relation.type = combo_dependency_parsing_from_vocab\n",
"I0407 10:54:53.280393 140321380496448 params.py:248] model.dependency_relation.vocab_namespace = deprel_labels\n",
"I0407 10:54:53.280741 140321380496448 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.in_features = 1024\n",
"I0407 10:54:53.280819 140321380496448 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.out_features = 512\n",
"I0407 10:54:53.280887 140321380496448 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.activation = tanh\n",
"I0407 10:54:53.281012 140321380496448 params.py:248] type = tanh\n",
"I0407 10:54:53.281121 140321380496448 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.dropout_rate = 0.0\n",
"I0407 10:54:53.285843 140321380496448 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.in_features = 1024\n",
"I0407 10:54:53.286010 140321380496448 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.out_features = 512\n",
"I0407 10:54:53.286088 140321380496448 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.activation = tanh\n",
"I0407 10:54:53.286234 140321380496448 params.py:248] type = tanh\n",
"I0407 10:54:53.286334 140321380496448 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.dropout_rate = 0.0\n",
"I0407 10:54:53.290788 140321380496448 params.py:248] model.dependency_relation.head_predictor.cycle_loss_n = 0\n",
"I0407 10:54:53.291093 140321380496448 params.py:248] model.dependency_relation.head_projection_layer.in_features = 1024\n",
"I0407 10:54:53.291184 140321380496448 params.py:248] model.dependency_relation.head_projection_layer.out_features = 128\n",
"I0407 10:54:53.291281 140321380496448 params.py:248] model.dependency_relation.head_projection_layer.activation = tanh\n",
"I0407 10:54:53.291444 140321380496448 params.py:248] type = tanh\n",
"I0407 10:54:53.291567 140321380496448 params.py:248] model.dependency_relation.head_projection_layer.dropout_rate = 0.25\n",
"I0407 10:54:53.293048 140321380496448 params.py:248] model.dependency_relation.dependency_projection_layer.in_features = 1024\n",
"I0407 10:54:53.293147 140321380496448 params.py:248] model.dependency_relation.dependency_projection_layer.out_features = 128\n",
"I0407 10:54:53.293218 140321380496448 params.py:248] model.dependency_relation.dependency_projection_layer.activation = tanh\n",
"I0407 10:54:53.293342 140321380496448 params.py:248] type = tanh\n",
"I0407 10:54:53.293437 140321380496448 params.py:248] model.dependency_relation.dependency_projection_layer.dropout_rate = 0.25\n",
"I0407 10:54:53.295091 140321380496448 params.py:248] model.enhanced_dependency_relation = None\n",
"I0407 10:54:53.295609 140321380496448 params.py:248] model.regularizer.regexes.0.1.type = l2\n",
"I0407 10:54:53.295784 140321380496448 params.py:248] model.regularizer.regexes.0.1.alpha = 1e-06\n",
"I0407 10:54:53.295953 140321380496448 params.py:248] model.regularizer.regexes.1.1.type = l2\n",
"I0407 10:54:53.296107 140321380496448 params.py:248] model.regularizer.regexes.1.1.alpha = 1e-06\n",
"I0407 10:54:53.296261 140321380496448 params.py:248] model.regularizer.regexes.2.1.type = l2\n",
"I0407 10:54:53.296412 140321380496448 params.py:248] model.regularizer.regexes.2.1.alpha = 1e-06\n",
"I0407 10:54:53.296564 140321380496448 params.py:248] model.regularizer.regexes.3.1.type = l2\n",
"I0407 10:54:53.296715 140321380496448 params.py:248] model.regularizer.regexes.3.1.alpha = 1e-05\n",
"I0407 10:54:56.194218 140321380496448 archival.py:211] removing temporary unarchived model dir at /tmp/tmpuvesoi4q\n",
"reading instances: 1980it [01:33, 21.15it/s]\n"
]
}
],
"source": [
"! {COMBO} --mode predict \\\n",
" --cuda_device 0 \\\n",
" --model_path model-pdbc-cont/model.tar.gz \\\n",
" --input_file connlu/pdbc-cont-validation.conllu \\\n",
" --output_file connlu/pdbc-cont-validation-pred.conllu"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "cfe7a3c4",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# text = Dwie dziewczynki opierają się o dach kapliczki , chłopiec wspina się na niego , a trzecia dziewczynka stoi obok .\r\n",
"1\tDwie\tdwa\t_\t_\t_\t3\tsubj\t3:subj\t_\r\n",
"2\tdziewczynki\tdziewczynka\t_\t_\t_\t1\tcomp\t1:comp\t_\r\n",
"3\topierają\topierać\t_\t_\t_\t15\tconjunct\t15:conjunct\t_\r\n",
"4\tsię\tsię\t_\t_\t_\t3\trefl\t3:refl\t_\r\n",
"5\to\to\t_\t_\t_\t3\tcomp\t3:comp\t_\r\n",
"6\tdach\tdach\t_\t_\t_\t5\tcomp\t5:comp\t_\r\n",
"7\tkapliczki\tkapliczka\t_\t_\t_\t6\tadjunct\t6:adjunct\t_\r\n",
"8\t,\t,\t_\t_\t_\t15\tpunct\t15:punct\t_\r\n",
"9\tchłopiec\tchłopiec\t_\t_\t_\t10\tsubj\t10:subj\t_\r\n"
]
}
],
"source": [
"! head connlu/pdbc-cont-validation.conllu"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "7dba9571",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# text = Dwie dziewczynki opierają się o dach kapliczki , chłopiec wspina się na niego , a trzecia dziewczynka stoi obok .\r\n",
"1\tDwie\tdwa\t_\t_\t_\t3\tsubj\t3:subj\t_\r\n",
"2\tdziewczynki\tdziewczynka\t_\t_\t_\t1\tcomp\t1:comp\t_\r\n",
"3\topierają\topierać\t_\t_\t_\t15\tconjunct\t15:conjunct\t_\r\n",
"4\tsię\tsię\t_\t_\t_\t3\trefl\t3:refl\t_\r\n",
"5\to\to\t_\t_\t_\t3\tcomp\t3:comp\t_\r\n",
"6\tdach\tdach\t_\t_\t_\t5\tcomp\t5:comp\t_\r\n",
"7\tkapliczki\tkapliczka\t_\t_\t_\t6\tadjunct\t6:adjunct\t_\r\n",
"8\t,\t,\t_\t_\t_\t15\tpunct\t15:punct\t_\r\n",
"9\tchłopiec\tchłopiec\t_\t_\t_\t10\tsubj\t10:subj\t_\r\n"
]
}
],
"source": [
"! head connlu/pdbc-cont-validation-pred.conllu"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "679601c2",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"I0407 10:56:35.295660 140254825452608 archival.py:184] loading archive file model-pdbc-cont/model.tar.gz\n",
"I0407 10:56:35.296370 140254825452608 archival.py:263] extracting archive file model-pdbc-cont/model.tar.gz to temp dir /tmp/tmpdhtf4et1\n",
"I0407 10:56:52.876630 140254825452608 params.py:248] dataset_reader.type = conllu\n",
"I0407 10:56:52.877122 140254825452608 params.py:248] dataset_reader.lazy = False\n",
"I0407 10:56:52.877243 140254825452608 params.py:248] dataset_reader.cache_directory = None\n",
"I0407 10:56:52.877313 140254825452608 params.py:248] dataset_reader.max_instances = None\n",
"I0407 10:56:52.877380 140254825452608 params.py:248] dataset_reader.manual_distributed_sharding = False\n",
"I0407 10:56:52.877446 140254825452608 params.py:248] dataset_reader.manual_multi_process_sharding = False\n",
"I0407 10:56:52.877737 140254825452608 params.py:248] dataset_reader.token_indexers.char.type = characters_const_padding\n",
"I0407 10:56:52.877938 140254825452608 params.py:248] dataset_reader.token_indexers.char.namespace = token_characters\n",
"I0407 10:56:52.878201 140254825452608 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:56:52.878276 140254825452608 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:56:52.878360 140254825452608 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:56:52.878507 140254825452608 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:56:52.878633 140254825452608 params.py:248] dataset_reader.token_indexers.char.start_tokens = None\n",
"I0407 10:56:52.878702 140254825452608 params.py:248] dataset_reader.token_indexers.char.end_tokens = None\n",
"I0407 10:56:52.878761 140254825452608 params.py:248] dataset_reader.token_indexers.char.min_padding_length = 32\n",
"I0407 10:56:52.878825 140254825452608 params.py:248] dataset_reader.token_indexers.char.token_min_padding_length = 0\n",
"I0407 10:56:52.878969 140254825452608 params.py:248] dataset_reader.token_indexers.feats.type = feats_indexer\n",
"I0407 10:56:52.879144 140254825452608 params.py:248] dataset_reader.token_indexers.feats.namespace = feats\n",
"I0407 10:56:52.879218 140254825452608 params.py:248] dataset_reader.token_indexers.feats.feature_name = feats_\n",
"I0407 10:56:52.879282 140254825452608 params.py:248] dataset_reader.token_indexers.feats.token_min_padding_length = 0\n",
"I0407 10:56:52.879426 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.type = characters_const_padding\n",
"I0407 10:56:52.879594 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.namespace = token_characters\n",
"I0407 10:56:52.879792 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.byte_encoding = None\n",
"I0407 10:56:52.879862 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.lowercase_characters = False\n",
"I0407 10:56:52.879944 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:56:52.880068 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:56:52.880184 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.start_tokens = None\n",
"I0407 10:56:52.880254 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.end_tokens = None\n",
"I0407 10:56:52.880316 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.min_padding_length = 32\n",
"I0407 10:56:52.880378 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.token_min_padding_length = 0\n",
"I0407 10:56:52.880523 140254825452608 params.py:248] dataset_reader.token_indexers.token.type = pretrained_transformer_mismatched_fixed\n",
"I0407 10:56:52.880748 140254825452608 params.py:248] dataset_reader.token_indexers.token.token_min_padding_length = 0\n",
"I0407 10:56:52.880829 140254825452608 params.py:248] dataset_reader.token_indexers.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:56:52.880893 140254825452608 params.py:248] dataset_reader.token_indexers.token.namespace = tags\n",
"I0407 10:56:52.880957 140254825452608 params.py:248] dataset_reader.token_indexers.token.max_length = None\n",
"I0407 10:56:52.881069 140254825452608 params.py:384] dataset_reader.token_indexers.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:56:55.893562 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.type = single_id\n",
"I0407 10:56:55.894115 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.namespace = upostag\n",
"I0407 10:56:55.894256 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.lowercase_tokens = False\n",
"I0407 10:56:55.894343 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.start_tokens = None\n",
"I0407 10:56:55.894395 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.end_tokens = None\n",
"I0407 10:56:55.894465 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.feature_name = pos_\n",
"I0407 10:56:55.894520 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:56:55.894590 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.token_min_padding_length = 0\n",
"I0407 10:56:55.894762 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.type = single_id\n",
"I0407 10:56:55.894958 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.namespace = xpostag\n",
"I0407 10:56:55.895048 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.lowercase_tokens = False\n",
"I0407 10:56:55.895111 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.start_tokens = None\n",
"I0407 10:56:55.895176 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.end_tokens = None\n",
"I0407 10:56:55.895228 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.feature_name = tag_\n",
"I0407 10:56:55.895297 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:56:55.895349 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.token_min_padding_length = 0\n",
"I0407 10:56:55.895593 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.type = characters_const_padding\n",
"I0407 10:56:55.895786 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.namespace = lemma_characters\n",
"I0407 10:56:55.896016 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:56:55.896095 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:56:55.896188 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:56:55.896353 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:56:55.896480 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.start_tokens = None\n",
"I0407 10:56:55.896552 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.end_tokens = None\n",
"I0407 10:56:55.896607 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.min_padding_length = 32\n",
"I0407 10:56:55.896675 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.token_min_padding_length = 0\n",
"I0407 10:56:55.896760 140254825452608 params.py:248] dataset_reader.features = ['token', 'char']\n",
"I0407 10:56:55.896864 140254825452608 params.py:248] dataset_reader.targets = ['head', 'deprel']\n",
"I0407 10:56:55.896962 140254825452608 params.py:248] dataset_reader.use_sem = False\n",
"I0407 10:56:55.897153 140254825452608 params.py:248] dataset_reader.type = conllu\n",
"I0407 10:56:55.897414 140254825452608 params.py:248] dataset_reader.lazy = False\n",
"I0407 10:56:55.897499 140254825452608 params.py:248] dataset_reader.cache_directory = None\n",
"I0407 10:56:55.897570 140254825452608 params.py:248] dataset_reader.max_instances = None\n",
"I0407 10:56:55.897637 140254825452608 params.py:248] dataset_reader.manual_distributed_sharding = False\n",
"I0407 10:56:55.897707 140254825452608 params.py:248] dataset_reader.manual_multi_process_sharding = False\n",
"I0407 10:56:55.897995 140254825452608 params.py:248] dataset_reader.token_indexers.char.type = characters_const_padding\n",
"I0407 10:56:55.898183 140254825452608 params.py:248] dataset_reader.token_indexers.char.namespace = token_characters\n",
"I0407 10:56:55.898398 140254825452608 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:56:55.898473 140254825452608 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:56:55.898542 140254825452608 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:56:55.898677 140254825452608 params.py:248] dataset_reader.token_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:56:55.898799 140254825452608 params.py:248] dataset_reader.token_indexers.char.start_tokens = None\n",
"I0407 10:56:55.898869 140254825452608 params.py:248] dataset_reader.token_indexers.char.end_tokens = None\n",
"I0407 10:56:55.898936 140254825452608 params.py:248] dataset_reader.token_indexers.char.min_padding_length = 32\n",
"I0407 10:56:55.898998 140254825452608 params.py:248] dataset_reader.token_indexers.char.token_min_padding_length = 0\n",
"I0407 10:56:55.899158 140254825452608 params.py:248] dataset_reader.token_indexers.feats.type = feats_indexer\n",
"I0407 10:56:55.899337 140254825452608 params.py:248] dataset_reader.token_indexers.feats.namespace = feats\n",
"I0407 10:56:55.899414 140254825452608 params.py:248] dataset_reader.token_indexers.feats.feature_name = feats_\n",
"I0407 10:56:55.899485 140254825452608 params.py:248] dataset_reader.token_indexers.feats.token_min_padding_length = 0\n",
"I0407 10:56:55.899629 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.type = characters_const_padding\n",
"I0407 10:56:55.899797 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.namespace = token_characters\n",
"I0407 10:56:55.899995 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.byte_encoding = None\n",
"I0407 10:56:55.900055 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.lowercase_characters = False\n",
"I0407 10:56:55.900130 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:56:55.900250 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:56:55.900363 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.start_tokens = None\n",
"I0407 10:56:55.900426 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.end_tokens = None\n",
"I0407 10:56:55.900486 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.min_padding_length = 32\n",
"I0407 10:56:55.900547 140254825452608 params.py:248] dataset_reader.token_indexers.lemma.token_min_padding_length = 0\n",
"I0407 10:56:55.900689 140254825452608 params.py:248] dataset_reader.token_indexers.token.type = pretrained_transformer_mismatched_fixed\n",
"I0407 10:56:55.900916 140254825452608 params.py:248] dataset_reader.token_indexers.token.token_min_padding_length = 0\n",
"I0407 10:56:55.900995 140254825452608 params.py:248] dataset_reader.token_indexers.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:56:55.901061 140254825452608 params.py:248] dataset_reader.token_indexers.token.namespace = tags\n",
"I0407 10:56:55.901125 140254825452608 params.py:248] dataset_reader.token_indexers.token.max_length = None\n",
"I0407 10:56:55.901226 140254825452608 params.py:384] dataset_reader.token_indexers.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:56:55.902561 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.type = single_id\n",
"I0407 10:56:55.902824 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.namespace = upostag\n",
"I0407 10:56:55.902909 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.lowercase_tokens = False\n",
"I0407 10:56:55.902969 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.start_tokens = None\n",
"I0407 10:56:55.903034 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.end_tokens = None\n",
"I0407 10:56:55.903095 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.feature_name = pos_\n",
"I0407 10:56:55.903159 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:56:55.903219 140254825452608 params.py:248] dataset_reader.token_indexers.upostag.token_min_padding_length = 0\n",
"I0407 10:56:55.903364 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.type = single_id\n",
"I0407 10:56:55.903547 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.namespace = xpostag\n",
"I0407 10:56:55.903621 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.lowercase_tokens = False\n",
"I0407 10:56:55.903687 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.start_tokens = None\n",
"I0407 10:56:55.903748 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.end_tokens = None\n",
"I0407 10:56:55.903811 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.feature_name = tag_\n",
"I0407 10:56:55.903868 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.default_value = THIS IS A REALLY UNLIKELY VALUE THAT HAS TO BE A STRING\n",
"I0407 10:56:55.903931 140254825452608 params.py:248] dataset_reader.token_indexers.xpostag.token_min_padding_length = 0\n",
"I0407 10:56:55.904146 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.type = characters_const_padding\n",
"I0407 10:56:55.904325 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.namespace = lemma_characters\n",
"I0407 10:56:55.904539 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.byte_encoding = None\n",
"I0407 10:56:55.904611 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.lowercase_characters = False\n",
"I0407 10:56:55.904691 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.start_tokens = ['__START__']\n",
"I0407 10:56:55.904827 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.character_tokenizer.end_tokens = ['__END__']\n",
"I0407 10:56:55.904946 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.start_tokens = None\n",
"I0407 10:56:55.905013 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.end_tokens = None\n",
"I0407 10:56:55.905084 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.min_padding_length = 32\n",
"I0407 10:56:55.905149 140254825452608 params.py:248] dataset_reader.lemma_indexers.char.token_min_padding_length = 0\n",
"I0407 10:56:55.905237 140254825452608 params.py:248] dataset_reader.features = ['token', 'char']\n",
"I0407 10:56:55.905334 140254825452608 params.py:248] dataset_reader.targets = ['head', 'deprel']\n",
"I0407 10:56:55.905422 140254825452608 params.py:248] dataset_reader.use_sem = False\n",
"I0407 10:56:55.906047 140254825452608 params.py:248] vocabulary.type = from_instances_extended\n",
"I0407 10:56:55.906157 140254825452608 vocabulary.py:323] Loading token dictionary from /tmp/tmpdhtf4et1/vocabulary.\n",
"I0407 10:56:55.906635 140254825452608 filelock.py:254] Lock 140251419626896 acquired on /tmp/tmpdhtf4et1/vocabulary/.lock\n",
"I0407 10:56:55.907354 140254825452608 filelock.py:317] Lock 140251419626896 released on /tmp/tmpdhtf4et1/vocabulary/.lock\n",
"I0407 10:56:55.907914 140254825452608 params.py:248] model.type = semantic_multitask\n",
"I0407 10:56:55.908506 140254825452608 params.py:248] model.text_field_embedder.type = basic\n",
"I0407 10:56:55.908878 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.type = char_embeddings_from_config\n",
"I0407 10:56:55.909080 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.embedding_dim = 64\n",
"I0407 10:56:55.909353 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.input_dim = 64\n",
"I0407 10:56:55.909446 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.filters = [512, 256, 64]\n",
"I0407 10:56:55.909554 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.kernel_size = [3, 3, 3]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"I0407 10:56:55.909654 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.stride = [1, 1, 1]\n",
"I0407 10:56:55.909750 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.padding = [1, 2, 4]\n",
"I0407 10:56:55.909847 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.dilation = [1, 2, 4]\n",
"I0407 10:56:55.909946 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.dilated_cnn_encoder.activations = ['relu', 'relu', 'linear']\n",
"I0407 10:56:55.910176 140254825452608 params.py:248] type = relu\n",
"I0407 10:56:55.910410 140254825452608 params.py:248] type = relu\n",
"I0407 10:56:55.910567 140254825452608 params.py:248] type = linear\n",
"I0407 10:56:55.917278 140254825452608 params.py:248] model.text_field_embedder.token_embedders.char.vocab_namespace = token_characters\n",
"I0407 10:56:55.917941 140254825452608 params.py:248] model.text_field_embedder.token_embedders.token.type = transformers_word_embeddings\n",
"I0407 10:56:55.918267 140254825452608 params.py:248] model.text_field_embedder.token_embedders.token.model_name = allegro/herbert-large-cased\n",
"I0407 10:56:55.918358 140254825452608 params.py:248] model.text_field_embedder.token_embedders.token.projection_dim = 100\n",
"I0407 10:56:55.918458 140254825452608 params.py:248] model.text_field_embedder.token_embedders.token.projection_activation = <function TransformersWordEmbedder.<lambda> at 0x7f8ed1745280>\n",
"I0407 10:56:55.918541 140254825452608 params.py:248] model.text_field_embedder.token_embedders.token.projection_dropout_rate = 0.0\n",
"I0407 10:56:55.918609 140254825452608 params.py:248] model.text_field_embedder.token_embedders.token.freeze_transformer = True\n",
"I0407 10:56:55.918674 140254825452608 params.py:248] model.text_field_embedder.token_embedders.token.last_layer_only = True\n",
"I0407 10:56:55.918785 140254825452608 params.py:384] model.text_field_embedder.token_embedders.token.tokenizer_kwargs.use_fast = False\n",
"I0407 10:56:55.918858 140254825452608 params.py:248] model.text_field_embedder.token_embedders.token.transformer_kwargs = None\n",
"I0407 10:57:03.624983 140254825452608 params.py:248] model.seq_encoder.type = combo_encoder\n",
"I0407 10:57:03.625626 140254825452608 params.py:248] model.seq_encoder.stacked_bilstm.input_size = 164\n",
"I0407 10:57:03.625742 140254825452608 params.py:248] model.seq_encoder.stacked_bilstm.hidden_size = 512\n",
"I0407 10:57:03.625796 140254825452608 params.py:248] model.seq_encoder.stacked_bilstm.num_layers = 2\n",
"I0407 10:57:03.625844 140254825452608 params.py:248] model.seq_encoder.stacked_bilstm.recurrent_dropout_probability = 0.33\n",
"I0407 10:57:03.625942 140254825452608 params.py:248] model.seq_encoder.stacked_bilstm.layer_dropout_probability = 0.33\n",
"I0407 10:57:03.626068 140254825452608 params.py:248] model.seq_encoder.stacked_bilstm.use_highway = False\n",
"I0407 10:57:03.933019 140254825452608 params.py:248] model.seq_encoder.layer_dropout_probability = 0.33\n",
"I0407 10:57:03.933302 140254825452608 params.py:248] model.use_sample_weight = True\n",
"I0407 10:57:03.933391 140254825452608 params.py:248] model.lemmatizer = None\n",
"I0407 10:57:03.933440 140254825452608 params.py:248] model.upos_tagger = None\n",
"I0407 10:57:03.933486 140254825452608 params.py:248] model.xpos_tagger = None\n",
"I0407 10:57:03.933528 140254825452608 params.py:248] model.semantic_relation = None\n",
"I0407 10:57:03.933570 140254825452608 params.py:248] model.morphological_feat = None\n",
"I0407 10:57:03.933835 140254825452608 params.py:248] model.dependency_relation.type = combo_dependency_parsing_from_vocab\n",
"I0407 10:57:03.934096 140254825452608 params.py:248] model.dependency_relation.vocab_namespace = deprel_labels\n",
"I0407 10:57:03.934389 140254825452608 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.in_features = 1024\n",
"I0407 10:57:03.934459 140254825452608 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.out_features = 512\n",
"I0407 10:57:03.934515 140254825452608 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.activation = tanh\n",
"I0407 10:57:03.934614 140254825452608 params.py:248] type = tanh\n",
"I0407 10:57:03.934703 140254825452608 params.py:248] model.dependency_relation.head_predictor.head_projection_layer.dropout_rate = 0.0\n",
"I0407 10:57:03.938141 140254825452608 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.in_features = 1024\n",
"I0407 10:57:03.938247 140254825452608 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.out_features = 512\n",
"I0407 10:57:03.938306 140254825452608 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.activation = tanh\n",
"I0407 10:57:03.938404 140254825452608 params.py:248] type = tanh\n",
"I0407 10:57:03.938489 140254825452608 params.py:248] model.dependency_relation.head_predictor.dependency_projection_layer.dropout_rate = 0.0\n",
"I0407 10:57:03.941669 140254825452608 params.py:248] model.dependency_relation.head_predictor.cycle_loss_n = 0\n",
"I0407 10:57:03.941908 140254825452608 params.py:248] model.dependency_relation.head_projection_layer.in_features = 1024\n",
"I0407 10:57:03.941985 140254825452608 params.py:248] model.dependency_relation.head_projection_layer.out_features = 128\n",
"I0407 10:57:03.942037 140254825452608 params.py:248] model.dependency_relation.head_projection_layer.activation = tanh\n",
"I0407 10:57:03.942123 140254825452608 params.py:248] type = tanh\n",
"I0407 10:57:03.942194 140254825452608 params.py:248] model.dependency_relation.head_projection_layer.dropout_rate = 0.25\n",
"I0407 10:57:03.943288 140254825452608 params.py:248] model.dependency_relation.dependency_projection_layer.in_features = 1024\n",
"I0407 10:57:03.943376 140254825452608 params.py:248] model.dependency_relation.dependency_projection_layer.out_features = 128\n",
"I0407 10:57:03.943423 140254825452608 params.py:248] model.dependency_relation.dependency_projection_layer.activation = tanh\n",
"I0407 10:57:03.943510 140254825452608 params.py:248] type = tanh\n",
"I0407 10:57:03.943577 140254825452608 params.py:248] model.dependency_relation.dependency_projection_layer.dropout_rate = 0.25\n",
"I0407 10:57:03.944838 140254825452608 params.py:248] model.enhanced_dependency_relation = None\n",
"I0407 10:57:03.945286 140254825452608 params.py:248] model.regularizer.regexes.0.1.type = l2\n",
"I0407 10:57:03.945443 140254825452608 params.py:248] model.regularizer.regexes.0.1.alpha = 1e-06\n",
"I0407 10:57:03.945568 140254825452608 params.py:248] model.regularizer.regexes.1.1.type = l2\n",
"I0407 10:57:03.945679 140254825452608 params.py:248] model.regularizer.regexes.1.1.alpha = 1e-06\n",
"I0407 10:57:03.945787 140254825452608 params.py:248] model.regularizer.regexes.2.1.type = l2\n",
"I0407 10:57:03.945892 140254825452608 params.py:248] model.regularizer.regexes.2.1.alpha = 1e-06\n",
"I0407 10:57:03.946047 140254825452608 params.py:248] model.regularizer.regexes.3.1.type = l2\n",
"I0407 10:57:03.946158 140254825452608 params.py:248] model.regularizer.regexes.3.1.alpha = 1e-05\n",
"I0407 10:57:06.549506 140254825452608 archival.py:211] removing temporary unarchived model dir at /tmp/tmpdhtf4et1\n",
"reading instances: 1990it [01:39, 20.00it/s]\n"
]
}
],
"source": [
"! {COMBO} --mode predict \\\n",
" --cuda_device 0 \\\n",
" --model_path model-pdbc-cont/model.tar.gz \\\n",
" --input_file connlu/pdbc-cont-test.conllu \\\n",
" --output_file connlu/pdbc-cont-test-pred.conllu"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "ddc3986b",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# text = Mały chłopiec patrzy w bok po ściągnięciu okularów .\r\n",
"1\tMały\tmały\t_\t_\t_\t2\tadjunct\t2:adjunct\t_\r\n",
"2\tchłopiec\tchłopiec\t_\t_\t_\t3\tsubj\t3:subj\t_\r\n",
"3\tpatrzy\tpatrzeć\t_\t_\t_\t0\troot\t0:root\t_\r\n",
"4\tw\tw\t_\t_\t_\t3\tadjunct_adl\t3:adjunct_adl\t_\r\n",
"5\tbok\tbok\t_\t_\t_\t4\tcomp\t4:comp\t_\r\n",
"6\tpo\tpo\t_\t_\t_\t3\tadjunct_temp\t3:adjunct_temp\t_\r\n",
"7\tściągnięciu\tściągnąć\t_\t_\t_\t6\tcomp\t6:comp\t_\r\n",
"8\tokularów\tokulary\t_\t_\t_\t7\tobj\t7:obj\t_\r\n",
"9\t.\t.\t_\t_\t_\t3\tpunct\t3:punct\t_\r\n"
]
}
],
"source": [
"! head connlu/pdbc-cont-test.conllu"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "34aa16d9",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# text = Mały chłopiec patrzy w bok po ściągnięciu okularów .\r\n",
"1\tMały\tmały\t_\t_\t_\t2\tadjunct\t2:adjunct\t_\r\n",
"2\tchłopiec\tchłopiec\t_\t_\t_\t3\tsubj\t3:subj\t_\r\n",
"3\tpatrzy\tpatrzeć\t_\t_\t_\t0\troot\t0:root\t_\r\n",
"4\tw\tw\t_\t_\t_\t3\tcomp\t3:adjunct_adl\t_\r\n",
"5\tbok\tbok\t_\t_\t_\t4\tcomp\t4:comp\t_\r\n",
"6\tpo\tpo\t_\t_\t_\t3\tadjunct_temp\t3:adjunct_temp\t_\r\n",
"7\tściągnięciu\tściągnąć\t_\t_\t_\t6\tcomp\t6:comp\t_\r\n",
"8\tokularów\tokulary\t_\t_\t_\t7\tobj\t7:obj\t_\r\n",
"9\t.\t.\t_\t_\t_\t3\tpunct\t3:punct\t_\r\n"
]
}
],
"source": [
"! head connlu/pdbc-cont-test-pred.conllu"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "combo_python39",
"language": "python",
"name": "combo_python39"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.16"
}
},
"nbformat": 4,
"nbformat_minor": 5
}