-
Notifications
You must be signed in to change notification settings - Fork 3
/
utils_prover.py
737 lines (613 loc) · 28.8 KB
/
utils_prover.py
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
from __future__ import absolute_import, division, print_function
import csv
import logging
import os
import sys
from io import open
import json
from nltk.tokenize import sent_tokenize
import numpy as np
from proof_utils import get_proof_graph, get_proof_graph_with_fail
logger = logging.getLogger(__name__)
class InputExample(object):
"""A single training/test example for simple sequence classification."""
def __init__(self, guid, text_a, text_b=None, label=None):
"""Constructs a InputExample.
Args:
guid: Unique id for the example.
text_a: string. The untokenized text of the first sequence. For single
sequence tasks, only this sequence must be specified.
text_b: (Optional) string. The untokenized text of the second sequence.
Only must be specified for sequence pair tasks.
label: (Optional) string. The label of the example. This should be
specified for train and dev examples, but not for test examples.
"""
self.guid = guid
self.text_a = text_a
self.text_b = text_b
self.label = label
class RRInputExampleQA(object):
def __init__(self, id, context, question, label):
self.id = id
self.context = context
self.question = question
self.label = label
class RRInputExample(object):
def __init__(self, id, context, question, node_label, edge_label, label):
self.id = id
self.context = context
self.question = question
self.node_label = node_label
self.edge_label = edge_label
self.label = label
class InputFeatures(object):
"""A single set of features of data."""
def __init__(self, input_ids, input_mask, segment_ids, label_id):
self.input_ids = input_ids
self.input_mask = input_mask
self.segment_ids = segment_ids
self.label_id = label_id
class RRFeaturesQA(object):
def __init__(self, id, input_ids, input_mask, segment_ids, label_id):
self.id = id
self.input_ids = input_ids
self.input_mask = input_mask
self.segment_ids = segment_ids
self.label_id = label_id
class RRFeatures(object):
def __init__(self, id, input_ids, input_mask, segment_ids, proof_offset, node_label, edge_label, label_id):
self.id = id
self.input_ids = input_ids
self.input_mask = input_mask
self.segment_ids = segment_ids
self.proof_offset = proof_offset
self.node_label = node_label
self.edge_label = edge_label
self.label_id = label_id
class DataProcessor(object):
"""Base class for data converters for sequence classification data sets."""
def get_train_examples(self, data_dir):
"""Gets a collection of `InputExample`s for the train set."""
raise NotImplementedError()
def get_dev_examples(self, data_dir):
"""Gets a collection of `InputExample`s for the dev set."""
raise NotImplementedError()
def get_test_examples(self, data_dir):
"""Gets a collection of `InputExample`s for the test set."""
raise NotImplementedError()
def get_labels(self):
"""Gets the list of labels for this data set."""
raise NotImplementedError()
@classmethod
def _read_tsv(cls, input_file, quotechar=None):
"""Reads a tab separated value file."""
with open(input_file, "r", encoding="utf-8-sig") as f:
reader = csv.reader(f, delimiter="\t", quotechar=quotechar)
lines = []
for line in reader:
if sys.version_info[0] == 2:
line = list(unicode(cell, 'utf-8') for cell in line)
lines.append(line)
return lines
@classmethod
def _read_jsonl(cls, input_file):
"""Reads a tab separated value file."""
records = []
with open(input_file, "r", encoding="utf-8-sig") as f:
for line in f:
records.append(json.loads(line))
return records
class RRProcessorQA(DataProcessor):
def get_train_examples(self, data_dir):
return self._create_examples(
self._read_jsonl(os.path.join(data_dir, "train.jsonl")),
self._read_jsonl(os.path.join(data_dir, "meta-train.jsonl")))
def get_dev_examples(self, data_dir):
# Change these to test paths for test results
return self._create_examples(
self._read_jsonl(os.path.join(data_dir, "dev.jsonl")),
self._read_jsonl(os.path.join(data_dir, "meta-dev.jsonl")))
'''
return self._create_examples_leave_one_out(
self._read_jsonl(os.path.join(data_dir, "leave-one-out.jsonl"))
)
'''
def get_test_examples(self, data_dir):
return self._create_examples(
self._read_jsonl(os.path.join(data_dir, "test.jsonl")),
self._read_jsonl(os.path.join(data_dir, "meta-test.jsonl")))
def get_labels(self):
return [True, False]
def _create_examples(self, records, meta_records):
examples = []
for (i, (record, meta_record)) in enumerate(zip(records, meta_records)):
print(i)
assert record["id"] == meta_record["id"]
context = record["context"]
for (j, question) in enumerate(record["questions"]):
id = question["id"]
label = question["label"]
meta_data = meta_record["questions"]["Q" + str(j + 1)]
proofs = meta_data["proofs"]
#if "CWA" in proofs:
# continue
#if question["meta"]["QDep"] != 3:
# continue
question = question["text"]
examples.append(RRInputExampleQA(id, context, question, label))
return examples
def _create_examples_leave_one_out(self, records):
examples = []
for record in records:
examples.append(RRInputExampleQA(record["id"], record["context"], record["question"], record["label"]))
return examples
class RRProcessor(DataProcessor):
def get_train_examples(self, data_dir):
return self._create_examples(
self._read_jsonl(os.path.join(data_dir, "train.jsonl")),
self._read_jsonl(os.path.join(data_dir, "meta-train.jsonl")), "train")
def get_dev_examples(self, data_dir):
# Change these to test paths for test results
return self._create_examples(
self._read_jsonl(os.path.join(data_dir, "dev.jsonl")),
self._read_jsonl(os.path.join(data_dir, "meta-dev.jsonl")), "dev")
def get_test_examples(self, data_dir):
return self._create_examples(
self._read_jsonl(os.path.join(data_dir, "test.jsonl")),
self._read_jsonl(os.path.join(data_dir, "meta-test.jsonl")), "test")
def get_labels(self):
return [True, False]
# Unconstrained training, use this for ablation
def _get_node_edge_label_unconstrained(self, proofs, sentence_scramble, nfact, nrule):
proof = proofs.split("OR")[0]
#print(proof)
node_label = [0] * (nfact + nrule + 1)
edge_label = np.zeros((nfact+nrule+1, nfact+nrule+1), dtype=int)
if "FAIL" in proof:
nodes, edges = get_proof_graph_with_fail(proof)
else:
nodes, edges = get_proof_graph(proof)
#print(nodes)
#print(edges)
component_index_map = {}
for (i, index) in enumerate(sentence_scramble):
if index <= nfact:
component = "triple" + str(index)
else:
component = "rule" + str(index-nfact)
component_index_map[component] = i
for node in nodes:
if node != "NAF":
index = component_index_map[node]
else:
index = nfact+nrule
node_label[index] = 1
edges = list(set(edges))
for edge in edges:
if edge[0] != "NAF":
start_index = component_index_map[edge[0]]
else:
start_index = nfact+nrule
if edge[1] != "NAF":
end_index = component_index_map[edge[1]]
else:
end_index = nfact+nrule
edge_label[start_index][end_index] = 1
return node_label, list(edge_label.flatten())
def _get_node_edge_label_constrained(self, proof, sentence_scramble, nfact, nrule):
#print(proof)
node_label = [0] * (nfact + nrule + 1)
edge_label = np.zeros((nfact + nrule + 1, nfact + nrule + 1), dtype=int)
if "FAIL" in proof:
nodes, edges = get_proof_graph_with_fail(proof)
else:
nodes, edges = get_proof_graph(proof)
# print(nodes)
# print(edges)
component_index_map = {}
for (i, index) in enumerate(sentence_scramble):
if index <= nfact:
component = "triple" + str(index)
else:
component = "rule" + str(index - nfact)
component_index_map[component] = i
component_index_map["NAF"] = nfact+nrule
for node in nodes:
index = component_index_map[node]
node_label[index] = 1
edges = list(set(edges))
for edge in edges:
start_index = component_index_map[edge[0]]
end_index = component_index_map[edge[1]]
edge_label[start_index][end_index] = 1
# Mask impossible edges
for i in range(len(edge_label)):
for j in range(len(edge_label)):
# Ignore diagonal
if i == j:
edge_label[i][j] = -100
continue
# Ignore edges between non-nodes
if node_label[i] == 0 or node_label[j] == 0:
edge_label[i][j] = -100
continue
is_fact_start = False
is_fact_end = False
if i == len(edge_label)-1 or sentence_scramble[i] <= nfact:
is_fact_start = True
if j == len(edge_label)-1 or sentence_scramble[j] <= nfact:
is_fact_end = True
# No edge between fact/NAF -> fact/NAF
if is_fact_start and is_fact_end:
edge_label[i][j] = -100
continue
# No edge between Rule -> fact/NAF
if not is_fact_start and is_fact_end:
edge_label[i][j] = -100
continue
return node_label, list(edge_label.flatten())
def _create_examples(self, records, meta_records, split):
examples = []
for (i, (record, meta_record)) in enumerate(zip(records, meta_records)):
#print(i)
assert record["id"] == meta_record["id"]
context = record["context"]
sentence_scramble = record["meta"]["sentenceScramble"]
for (j, question) in enumerate(record["questions"]):
# Uncomment to train/evaluate at a certain depth
#if question["meta"]["QDep"] != 5:
# continue
# Uncomment to test at a specific subset of Birds-Electricity dataset
#if not record["id"].startswith("AttPosElectricityRB4"):
# continue
id = question["id"]
label = question["label"]
question = question["text"]
meta_data = meta_record["questions"]["Q"+str(j+1)]
assert (question == meta_data["question"])
proofs = meta_data["proofs"]
nfact = meta_record["NFact"]
nrule = meta_record["NRule"]
for proof in proofs.split("OR"):
node_label, edge_label = self._get_node_edge_label_constrained(proof, sentence_scramble, nfact, nrule)
examples.append(RRInputExample(id, context, question, node_label, edge_label, label))
if split == "dev" or split == "test":
break
return examples
def convert_examples_to_features(examples,
label_list,
max_seq_length,
tokenizer,
output_mode,
cls_token_at_end=False,
pad_on_left=False,
cls_token='[CLS]',
sep_token='[SEP]',
sep_token_extra=False,
pad_token=0,
sequence_a_segment_id=0,
sequence_b_segment_id=1,
cls_token_segment_id=1,
pad_token_segment_id=0,
mask_padding_with_zero=True):
""" Loads a data file into a list of `InputBatch`s
`cls_token_at_end` define the location of the CLS token:
- False (Default, BERT/XLM pattern): [CLS] + A + [SEP] + B + [SEP]
- True (XLNet/GPT pattern): A + [SEP] + B + [SEP] + [CLS]
`cls_token_segment_id` define the segment id associated to the CLS token (0 for BERT, 2 for XLNet)
"""
label_map = {label : i for i, label in enumerate(label_list)}
features = []
for (ex_index, example) in enumerate(examples):
if ex_index % 10000 == 0:
logger.info("Writing example %d of %d" % (ex_index, len(examples)))
tokens_a = tokenizer.tokenize(example.text_a)
tokens_b = None
special_tokens_count = 3 if sep_token_extra else 2
if example.text_b:
tokens_b = tokenizer.tokenize(example.text_b)
# Modifies `tokens_a` and `tokens_b` in place so that the total
# length is less than the specified length.
# Account for [CLS], [SEP], [SEP] with "- 3"
_truncate_seq_pair(tokens_a, tokens_b, max_seq_length - special_tokens_count - 1)
else:
# Account for [CLS] and [SEP] with "- 2" or "-3" for RoBERTa
if len(tokens_a) > max_seq_length - special_tokens_count:
tokens_a = tokens_a[:(max_seq_length - special_tokens_count)]
# The convention in BERT is:
# (a) For sequence pairs:
# tokens: [CLS] is this jack ##son ##ville ? [SEP] no it is not . [SEP]
# type_ids: 0 0 0 0 0 0 0 0 1 1 1 1 1 1
# (b) For single sequences:
# tokens: [CLS] the dog is hairy . [SEP]
# type_ids: 0 0 0 0 0 0 0
#
# Where "type_ids" are used to indicate whether this is the first
# sequence or the second sequence. The embedding vectors for `type=0` and
# `type=1` were learned during pre-training and are added to the wordpiece
# embedding vector (and position vector). This is not *strictly* necessary
# since the [SEP] token unambiguously separates the sequences, but it makes
# it easier for the model to learn the concept of sequences.
#
# For classification tasks, the first vector (corresponding to [CLS]) is
# used as as the "sentence vector". Note that this only makes sense because
# the entire model is fine-tuned.
tokens = tokens_a + [sep_token]
if sep_token_extra:
# roberta uses an extra separator b/w pairs of sentences
tokens += [sep_token]
segment_ids = [sequence_a_segment_id] * len(tokens)
if tokens_b:
tokens += tokens_b + [sep_token]
segment_ids += [sequence_b_segment_id] * (len(tokens_b) + 1)
if cls_token_at_end:
tokens = tokens + [cls_token]
segment_ids = segment_ids + [cls_token_segment_id]
else:
tokens = [cls_token] + tokens
segment_ids = [cls_token_segment_id] + segment_ids
input_ids = tokenizer.convert_tokens_to_ids(tokens)
# The mask has 1 for real tokens and 0 for padding tokens. Only real
# tokens are attended to.
input_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
# Zero-pad up to the sequence length.
padding_length = max_seq_length - len(input_ids)
if pad_on_left:
input_ids = ([pad_token] * padding_length) + input_ids
input_mask = ([0 if mask_padding_with_zero else 1] * padding_length) + input_mask
segment_ids = ([pad_token_segment_id] * padding_length) + segment_ids
else:
input_ids = input_ids + ([pad_token] * padding_length)
input_mask = input_mask + ([0 if mask_padding_with_zero else 1] * padding_length)
segment_ids = segment_ids + ([pad_token_segment_id] * padding_length)
assert len(input_ids) == max_seq_length
assert len(input_mask) == max_seq_length
assert len(segment_ids) == max_seq_length
if output_mode == "classification":
label_id = label_map[example.label]
elif output_mode == "regression":
label_id = float(example.label)
else:
raise KeyError(output_mode)
if ex_index < 5:
logger.info("*** Example ***")
logger.info("guid: %s" % (example.guid))
logger.info("tokens: %s" % " ".join(
[str(x) for x in tokens]))
logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
logger.info("input_mask: %s" % " ".join([str(x) for x in input_mask]))
logger.info("segment_ids: %s" % " ".join([str(x) for x in segment_ids]))
logger.info("label: %s (id = %d)" % (example.label, label_id))
features.append(
InputFeatures(input_ids=input_ids,
input_mask=input_mask,
segment_ids=segment_ids,
label_id=label_id))
return features
def convert_examples_to_features_RR_QA(examples,
label_list,
max_seq_length,
tokenizer,
output_mode,
cls_token_at_end=False,
pad_on_left=False,
cls_token='[CLS]',
sep_token='[SEP]',
sep_token_extra=False,
pad_token=0,
sequence_a_segment_id=0,
sequence_b_segment_id=1,
cls_token_segment_id=1,
pad_token_segment_id=0,
mask_padding_with_zero=True):
label_map = {label : i for i, label in enumerate(label_list)}
features = []
max_size = 0
for (ex_index, example) in enumerate(examples):
if ex_index % 10000 == 0:
logger.info("Writing example %d of %d" % (ex_index, len(examples)))
sentences = sent_tokenize(example.context)
context_tokens = tokenizer.tokenize(example.context)
'''
for sentence in sentences:
sentence_tokens = tokenizer.tokenize(sentence)
context_tokens.extend(sentence_tokens)
'''
max_size = max(max_size, len(context_tokens))
question_tokens = tokenizer.tokenize(example.question)
special_tokens_count = 3 if sep_token_extra else 2
_truncate_seq_pair(context_tokens, question_tokens, max_seq_length - special_tokens_count - 1)
tokens = context_tokens + [sep_token]
if sep_token_extra:
# roberta uses an extra separator b/w pairs of sentences
tokens += [sep_token]
segment_ids = [sequence_a_segment_id] * len(tokens)
tokens += question_tokens + [sep_token]
segment_ids += [sequence_b_segment_id] * (len(question_tokens) + 1)
if cls_token_at_end:
tokens = tokens + [cls_token]
segment_ids = segment_ids + [cls_token_segment_id]
else:
tokens = [cls_token] + tokens
segment_ids = [cls_token_segment_id] + segment_ids
input_ids = tokenizer.convert_tokens_to_ids(tokens)
# The mask has 1 for real tokens and 0 for padding tokens. Only real
# tokens are attended to.
input_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
# Zero-pad up to the sequence length.
padding_length = max_seq_length - len(input_ids)
if pad_on_left:
input_ids = ([pad_token] * padding_length) + input_ids
input_mask = ([0 if mask_padding_with_zero else 1] * padding_length) + input_mask
segment_ids = ([pad_token_segment_id] * padding_length) + segment_ids
else:
input_ids = input_ids + ([pad_token] * padding_length)
input_mask = input_mask + ([0 if mask_padding_with_zero else 1] * padding_length)
segment_ids = segment_ids + ([pad_token_segment_id] * padding_length)
assert len(input_ids) == max_seq_length
assert len(input_mask) == max_seq_length
assert len(segment_ids) == max_seq_length
label_id = label_map[example.label]
if ex_index < 5:
logger.info("*** Example ***")
logger.info("id: %s" % (example.id))
logger.info("tokens: %s" % " ".join(
[str(x) for x in tokens]))
logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
logger.info("input_mask: %s" % " ".join([str(x) for x in input_mask]))
logger.info("segment_ids: %s" % " ".join([str(x) for x in segment_ids]))
logger.info("label: %s (id = %d)" % (example.label, label_id))
features.append(
RRFeaturesQA(id=id,
input_ids=input_ids,
input_mask=input_mask,
segment_ids=segment_ids,
label_id=label_id))
return features
def convert_examples_to_features_RR(examples,
label_list,
max_seq_length,
max_node_length,
max_edge_length,
tokenizer,
output_mode,
cls_token_at_end=False,
pad_on_left=False,
cls_token='[CLS]',
sep_token='[SEP]',
sep_token_extra=False,
pad_token=0,
sequence_a_segment_id=0,
sequence_b_segment_id=1,
cls_token_segment_id=1,
pad_token_segment_id=0,
mask_padding_with_zero=True):
label_map = {label : i for i, label in enumerate(label_list)}
features = []
max_size = 0
for (ex_index, example) in enumerate(examples):
if ex_index % 10000 == 0:
logger.info("Writing example %d of %d" % (ex_index, len(examples)))
sentences = sent_tokenize(example.context)
context_tokens = []
proof_offset = []
for sentence in sentences:
sentence_tokens = tokenizer.tokenize(sentence)
context_tokens.extend(sentence_tokens)
proof_offset.append(len(context_tokens))
max_size = max(max_size, len(context_tokens))
question_tokens = tokenizer.tokenize(example.question)
special_tokens_count = 3 if sep_token_extra else 2
_truncate_seq_pair(context_tokens, question_tokens, max_seq_length - special_tokens_count - 1)
tokens = context_tokens + [sep_token]
if sep_token_extra:
# roberta uses an extra separator b/w pairs of sentences
tokens += [sep_token]
segment_ids = [sequence_a_segment_id] * len(tokens)
tokens += question_tokens + [sep_token]
segment_ids += [sequence_b_segment_id] * (len(question_tokens) + 1)
if cls_token_at_end:
tokens = tokens + [cls_token]
segment_ids = segment_ids + [cls_token_segment_id]
else:
tokens = [cls_token] + tokens
segment_ids = [cls_token_segment_id] + segment_ids
input_ids = tokenizer.convert_tokens_to_ids(tokens)
# The mask has 1 for real tokens and 0 for padding tokens. Only real
# tokens are attended to.
input_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
# Zero-pad up to the sequence length.
padding_length = max_seq_length - len(input_ids)
if pad_on_left:
input_ids = ([pad_token] * padding_length) + input_ids
input_mask = ([0 if mask_padding_with_zero else 1] * padding_length) + input_mask
segment_ids = ([pad_token_segment_id] * padding_length) + segment_ids
else:
input_ids = input_ids + ([pad_token] * padding_length)
input_mask = input_mask + ([0 if mask_padding_with_zero else 1] * padding_length)
segment_ids = segment_ids + ([pad_token_segment_id] * padding_length)
proof_offset = proof_offset + [0] * (max_node_length - len(proof_offset))
node_label = example.node_label
node_label = node_label + [-100] * (max_node_length - len(node_label))
edge_label = example.edge_label
edge_label = edge_label + [-100] * (max_edge_length - len(edge_label))
assert len(input_ids) == max_seq_length
assert len(input_mask) == max_seq_length
assert len(segment_ids) == max_seq_length
assert len(proof_offset) == max_node_length
assert len(node_label) == max_node_length
assert len(edge_label) == max_edge_length
label_id = label_map[example.label]
if ex_index < 5:
logger.info("*** Example ***")
logger.info("id: %s" % (example.id))
logger.info("tokens: %s" % " ".join(
[str(x) for x in tokens]))
logger.info("input_ids: %s" % " ".join([str(x) for x in input_ids]))
logger.info("input_mask: %s" % " ".join([str(x) for x in input_mask]))
logger.info("segment_ids: %s" % " ".join([str(x) for x in segment_ids]))
logger.info("proof_offset: %s" % " ".join([str(x) for x in proof_offset]))
logger.info("node_label: %s" % " ".join([str(x) for x in node_label]))
logger.info("edge_label: %s" % " ".join([str(x) for x in edge_label]))
logger.info("label: %s (id = %d)" % (example.label, label_id))
features.append(
RRFeatures(id=id,
input_ids=input_ids,
input_mask=input_mask,
segment_ids=segment_ids,
proof_offset=proof_offset,
node_label=node_label,
edge_label=edge_label,
label_id=label_id))
return features
def _truncate_seq_pair(tokens_a, tokens_b, max_length):
"""Truncates a sequence pair in place to the maximum length."""
# This is a simple heuristic which will always truncate the longer sequence
# one token at a time. This makes more sense than truncating an equal percent
# of tokens from each, since if one sequence is very short then each token
# that's truncated likely contains more information than a longer sequence.
while True:
total_length = len(tokens_a) + len(tokens_b)
if total_length <= max_length:
break
if len(tokens_a) > len(tokens_b):
tokens_a.pop()
else:
tokens_b.pop()
def _truncate_seq_triple(tokens_a, tokens_b, tokens_c, max_length):
"""Truncates a sequence pair in place to the maximum length."""
# This is a simple heuristic which will always truncate the longer sequence
# one token at a time. This makes more sense than truncating an equal percent
# of tokens from each, since if one sequence is very short then each token
# that's truncated likely contains more information than a longer sequence.
while True:
total_length = len(tokens_a) + len(tokens_b) + len(tokens_c)
if total_length <= max_length:
break
max_len = max(len(tokens_a), len(tokens_b), len(tokens_c))
if max_len == len(tokens_a):
tokens_a.pop()
elif max_len == len(tokens_b):
tokens_b.pop()
else:
tokens_c.pop()
def simple_accuracy(preds, labels):
return (preds == labels).mean()
def compute_metrics(task_name, preds, labels):
assert len(preds) == len(labels)
if task_name == "rr" or task_name == "rr_qa":
return {"acc": simple_accuracy(preds, labels)}
else:
raise KeyError(task_name)
processors = {
"rr": RRProcessor,
"rr_qa": RRProcessorQA
}
output_modes = {
"rr": "classification",
"rr_qa": "classification"
}
GLUE_TASKS_NUM_LABELS = {
"rr": 2,
"rr_qa": 2
}