-
Notifications
You must be signed in to change notification settings - Fork 16
/
DST.py
672 lines (586 loc) · 30.8 KB
/
DST.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
import os, random, argparse, time, logging, json, tqdm
import numpy as np
from copy import deepcopy
from collections import OrderedDict
import torch
from itertools import chain
from copy import deepcopy
from utils import _ReaderBase
from damd_multiwoz import ontology
from damd_multiwoz.db_ops import MultiWozDB
from damd_multiwoz.config import global_config as cfg
from damd_multiwoz.eval import MultiWozEvaluator
from transformers import (AdamW, T5Tokenizer, T5ForConditionalGeneration, BartTokenizer, BartForConditionalGeneration, WEIGHTS_NAME,CONFIG_NAME, get_linear_schedule_with_warmup)
class BartTokenizer(BartTokenizer):
def encode(self,text,add_special_tokens=False):
encoded_inputs = self.encode_plus(text,add_special_tokens=False)
return encoded_inputs["input_ids"]
class BART_DST(BartForConditionalGeneration):
def __init__(self, config):
super().__init__(config)
def inference(
self,
tokenizer,
reader,
prev,
input_ids=None,
attention_mask=None,
turn_domain=None,
):
dst_outputs = self.generate(input_ids=input_ids,
attention_mask=attention_mask,
eos_token_id=tokenizer.encode("<eos_b>")[0],
decoder_start_token_id=self.config.decoder_start_token_id,
max_length=200,
min_length=1,
num_beams=1,
length_penalty=1.0,
)
dst_outputs = dst_outputs.tolist()
# DST_UPDATE -> DST
#check whether need to add eos
#dst_outputs = [dst+tokenizer.encode("<eos_b>") for dst in dst_outputs]
batch_size = input_ids.shape[0]
constraint_dict_updates = [reader.bspan_to_constraint_dict(tokenizer.decode(dst_outputs[i])) for i in range(batch_size)]
if prev['bspn']:
# update the belief state
dst_outputs = [reader.update_bspn(prev_bspn=prev['bspn'][i], bspn_update=dst_outputs[i]) for i in range(batch_size)]
return dst_outputs
class T5_DST(T5ForConditionalGeneration):
def __init__(self, config):
super().__init__(config)
def inference(
self,
tokenizer,
reader,
prev,
input_ids=None,
attention_mask=None,
turn_domain=None,
):
dst_outputs = self.generate(input_ids=input_ids,
attention_mask=attention_mask,
eos_token_id=tokenizer.encode("<eos_b>")[0],
decoder_start_token_id=self.config.decoder_start_token_id,
max_length=200,
)
dst_outputs = dst_outputs.tolist()
# DST_UPDATE -> DST
#check whether need to add eos
#dst_outputs = [dst+tokenizer.encode("<eos_b>") for dst in dst_outputs]
batch_size = input_ids.shape[0]
constraint_dict_updates = [reader.bspan_to_constraint_dict(tokenizer.decode(dst_outputs[i])) for i in range(batch_size)]
if prev['bspn']:
# update the belief state
dst_outputs = [reader.update_bspn(prev_bspn=prev['bspn'][i], bspn_update=dst_outputs[i]) for i in range(batch_size)]
return dst_outputs
class Vocab(object):
def __init__(self, model, tokenizer):
self.special_tokens = ["pricerange", "<pad>", "<go_r>", "<unk>", "<go_b>", "<go_a>", "<eos_u>", "<eos_r>", "<eos_b>", "<eos_a>", "<go_d>",
"[restaurant]","[hotel]","[attraction]","[train]","[taxi]","[police]","[hospital]","[general]","[inform]","[request]",
"[nooffer]","[recommend]","[select]","[offerbook]","[offerbooked]","[nobook]","[bye]","[greet]","[reqmore]","[welcome]",
"[value_name]","[value_choice]","[value_area]","[value_price]","[value_type]","[value_reference]","[value_phone]","[value_address]",
"[value_food]","[value_leave]","[value_postcode]","[value_id]","[value_arrive]","[value_stars]","[value_day]","[value_destination]",
"[value_car]","[value_departure]","[value_time]","[value_people]","[value_stay]","[value_pricerange]","[value_department]", "[db_state0]","[db_state1]","[db_state2]","[db_state3]","[db_state4]","<None>"]
self.attr_special_tokens = {'pad_token': '<pad>',
'additional_special_tokens': ["pricerange", "<go_r>", "<unk>", "<go_b>", "<go_a>", "<eos_u>", "<eos_r>", "<eos_b>", "<eos_a>", "<go_d>",
"[restaurant]","[hotel]","[attraction]","[train]","[taxi]","[police]","[hospital]","[general]","[inform]","[request]",
"[nooffer]","[recommend]","[select]","[offerbook]","[offerbooked]","[nobook]","[bye]","[greet]","[reqmore]","[welcome]",
"[value_name]","[value_choice]","[value_area]","[value_price]","[value_type]","[value_reference]","[value_phone]","[value_address]",
"[value_food]","[value_leave]","[value_postcode]","[value_id]","[value_arrive]","[value_stars]","[value_day]","[value_destination]",
"[value_car]","[value_departure]","[value_time]","[value_people]","[value_stay]","[value_pricerange]","[value_department]","[db_state0]","[db_state1]","[db_state2]","[db_state3]","[db_state4]","<None>"]}
self.tokenizer = tokenizer
self.vocab_size = self.add_special_tokens_(model, tokenizer)
def add_special_tokens_(self, model, tokenizer):
""" Add special tokens to the tokenizer and the model if they have not already been added. """
#orig_num_tokens = model.config.vocab_size
orig_num_tokens = len(tokenizer)
num_added_tokens = tokenizer.add_special_tokens(self.attr_special_tokens) # doesn't add if they are already there
if num_added_tokens > 0:
model.resize_token_embeddings(new_num_tokens=orig_num_tokens + num_added_tokens)
return orig_num_tokens + num_added_tokens
def encode(self, word):
""" customize for damd script """
return self.tokenizer.encode(word)[0]
def sentence_encode(self, word_list):
""" customize for damd script """
return self.tokenizer.encode(" ".join(word_list))
def decode(self, idx):
""" customize for damd script """
return self.tokenizer.decode(idx)
def sentence_decode(self, index_list, eos=None):
""" customize for damd script """
l = self.tokenizer.decode(index_list)
l = l.split()
if not eos or eos not in l:
text = ' '.join(l)
else:
idx = l.index(eos)
text = ' '.join(l[:idx])
return text
class MultiWozReader(_ReaderBase):
def __init__(self, vocab=None, args=None):
super().__init__()
self.db = MultiWozDB(cfg.dbs)
self.args = args
self.domain_files = json.loads(open(cfg.domain_file_path, 'r').read())
self.slot_value_set = json.loads(open(cfg.slot_value_set_path, 'r').read())
test_list = [l.strip().lower() for l in open(cfg.test_list, 'r').readlines()]
dev_list = [l.strip().lower() for l in open(cfg.dev_list, 'r').readlines()]
self.dev_files, self.test_files = {}, {}
for fn in test_list:
self.test_files[fn.replace('.json', '')] = 1
for fn in dev_list:
self.dev_files[fn.replace('.json', '')] = 1
self.vocab = vocab
self.vocab_size = vocab.vocab_size
self._load_data()
def _load_data(self, save_temp=False):
self.data = json.loads(open(cfg.data_path+cfg.data_file, 'r', encoding='utf-8').read().lower())
self.train, self.dev, self.test = [] , [], []
for fn, dial in self.data.items():
if 'all' in cfg.exp_domains or self.exp_files.get(fn):
if self.dev_files.get(fn):
self.dev.append(self._get_encoded_data(fn, dial))
elif self.test_files.get(fn):
self.test.append(self._get_encoded_data(fn, dial))
else:
self.train.append(self._get_encoded_data(fn, dial))
random.shuffle(self.train)
random.shuffle(self.dev)
random.shuffle(self.test)
def _get_encoded_data(self, fn, dial):
encoded_dial = []
dial_context = []
delete_op = self.vocab.tokenizer.encode("<None>") #[32157]
prev_constraint_dict = {}
for idx, t in enumerate(dial['log']):
enc = {}
enc['dial_id'] = fn
dial_context.append( self.vocab.tokenizer.encode(t['user']) + self.vocab.tokenizer.encode('<eos_u>') )
enc['resp_nodelex'] = self.vocab.tokenizer.encode(t['resp_nodelex']) + self.vocab.tokenizer.encode('<eos_r>')
enc['user'] = list(chain(*dial_context[-self.args.context_window:])) # here we use user to represent dialogue history
enc['bspn'] = self.vocab.tokenizer.encode(t['constraint']) + self.vocab.tokenizer.encode('<eos_b>')
constraint_dict = self.bspan_to_constraint_dict(t['constraint'])
update_bspn = self.check_update(prev_constraint_dict, constraint_dict)
enc['update_bspn'] = self.vocab.tokenizer.encode(update_bspn)
encoded_dial.append(enc)
prev_constraint_dict = constraint_dict
dial_context.append( enc['resp_nodelex'] )
return encoded_dial
def check_update(self, prev_constraint_dict, constraint_dict):
update_dict = {}
if prev_constraint_dict==constraint_dict:
return '<eos_b>'
for domain in constraint_dict:
if domain in prev_constraint_dict:
for slot in constraint_dict[domain]:
if constraint_dict[domain].get(slot) != prev_constraint_dict[domain].get(slot):
if domain not in update_dict:
update_dict[domain] = {}
update_dict[domain][slot] = constraint_dict[domain].get(slot)
# if delete is needed
# if len(prev_constraint_dict[domain])>len(constraint_dict[domain]):
for slot in prev_constraint_dict[domain]:
if constraint_dict[domain].get(slot) is None:
update_dict[domain][slot] = "<None>"
else:
update_dict[domain] = deepcopy(constraint_dict[domain])
update_bspn= self.constraint_dict_to_bspan(update_dict)
return update_bspn
def constraint_dict_to_bspan(self, constraint_dict):
if not constraint_dict:
return "<eos_b>"
update_bspn=""
for domain in constraint_dict:
if len(update_bspn)==0:
update_bspn += f"[{domain}]"
else:
update_bspn += f" [{domain}]"
for slot in constraint_dict[domain]:
update_bspn += f" {slot} {constraint_dict[domain][slot]}"
update_bspn += f" <eos_b>"
return update_bspn
def bspan_to_constraint_dict(self, bspan, bspn_mode = 'bspn'):
# add decoded(str) here
bspan = bspan.split() if isinstance(bspan, str) else bspan
constraint_dict = {}
domain = None
conslen = len(bspan)
for idx, cons in enumerate(bspan):
cons = self.vocab.decode(cons) if type(cons) is not str else cons
if cons == "[slot]":
continue
if cons == '<eos_b>':
break
if '[' in cons:
if cons[1:-1] not in ontology.all_domains:
continue
domain = cons[1:-1]
elif cons in ontology.get_slot:
if domain is None:
continue
if cons == 'people':
# handle confusion of value name "people's portraits..." and slot people
try:
ns = bspan[idx+1]
ns = self.vocab.decode(ns) if type(ns) is not str else ns
if ns == "'s":
continue
except:
continue
if not constraint_dict.get(domain):
constraint_dict[domain] = {}
if bspn_mode == 'bsdx':
constraint_dict[domain][cons] = 1
continue
vidx = idx+1
if vidx == conslen:
break
vt_collect = []
vt = bspan[vidx]
vt = self.vocab.decode(vt) if type(vt) is not str else vt
while vidx < conslen and vt != '<eos_b>' and '[' not in vt and vt not in ontology.get_slot:
vt_collect.append(vt)
vidx += 1
if vidx == conslen:
break
vt = bspan[vidx]
vt = self.vocab.decode(vt) if type(vt) is not str else vt
if vt_collect:
constraint_dict[domain][cons] = ' '.join(vt_collect)
return constraint_dict
def dspan_to_domain(self, dspan):
domains = {}
dspan = dspan.split() if isinstance(dspan, str) else dspan
for d in dspan:
dom = self.vocab.decode(d) if type(d) is not str else d
if dom != '<eos_d>':
domains[dom] = 1
else:
break
return domains
def convert_batch(self, batch, prev, first_turn=False, dst_start_token = 0):
"""
user: dialogue history ['user']
input: previous dialogue state + dialogue history
output1: dialogue state update ['update_bspn'] or current dialogue state ['bspn']
"""
inputs = {}
pad_token = self.vocab.tokenizer.encode("<pad>")[0]
batch_size = len(batch['user'])
# input: previous dialogue state + dialogue history
input_ids = []
if first_turn:
for i in range(batch_size):
input_ids.append(self.vocab.tokenizer.encode('<eos_b>') + batch['user'][i])
else:
for i in range(batch_size):
input_ids.append(prev['bspn'][i] + batch['user'][i])
input_ids, masks = self.padInput(input_ids, pad_token)
inputs["input_ids"] = torch.tensor(input_ids,dtype=torch.long)
inputs["masks"] = torch.tensor(masks,dtype=torch.long)
if self.args.noupdate_dst:
# here we use state_update denote the belief span (bspn)...
state_update, state_input = self.padOutput(batch['bspn'], pad_token)
else:
state_update, state_input = self.padOutput(batch['update_bspn'], pad_token)
inputs["state_update"] = torch.tensor(state_update,dtype=torch.long) # batch_size, seq_len
inputs["state_input"] = torch.tensor(np.concatenate( (np.ones((batch_size,1))*dst_start_token , state_input[:,:-1]), axis=1 ) ,dtype=torch.long)
# for k in inputs:
# if k=="masks":
# print(k)
# print(inputs[k])
# else:
# print(k)
# print(inputs[k].tolist())
# print(k)
# print(self.vocab.tokenizer.decode(inputs[k].tolist()[0]))
return inputs
def padOutput(self, sequences, pad_token):
lengths = [len(s) for s in sequences]
num_samples = len(lengths)
max_len = max(lengths)
output_ids = np.ones((num_samples, max_len)) * (-100) #-100 ignore by cross entropy
decoder_inputs = np.ones((num_samples, max_len)) * pad_token
for idx, s in enumerate(sequences):
trunc = s[:max_len]
output_ids[idx, :lengths[idx]] = trunc
decoder_inputs[idx, :lengths[idx]] = trunc
return output_ids, decoder_inputs
def padInput(self, sequences, pad_token):
lengths = [len(s) for s in sequences]
num_samples = len(lengths)
max_len = max(lengths)
input_ids = np.ones((num_samples, max_len)) * pad_token
masks = np.zeros((num_samples, max_len))
for idx, s in enumerate(sequences):
trunc = s[-max_len:]
input_ids[idx, :lengths[idx]] = trunc
masks[idx, :lengths[idx]] = 1
return input_ids, masks
def update_bspn(self, prev_bspn, bspn_update):
constraint_dict_update = self.bspan_to_constraint_dict(self.vocab.tokenizer.decode(bspn_update) )
if not constraint_dict_update:
return prev_bspn
constraint_dict = self.bspan_to_constraint_dict(self.vocab.tokenizer.decode(prev_bspn) )
for domain in constraint_dict_update:
if domain not in constraint_dict:
constraint_dict[domain] = {}
for slot, value in constraint_dict_update[domain].items():
if value=="<None>": #delete the slot
_ = constraint_dict[domain].pop(slot, None)
else:
constraint_dict[domain][slot]=value
updated_bspn = self.vocab.tokenizer.encode(self.constraint_dict_to_bspan(constraint_dict))
return updated_bspn
def wrap_result(self, result_dict, eos_syntax=None):
decode_fn = self.vocab.sentence_decode
results = []
eos_syntax = ontology.eos_tokens if not eos_syntax else eos_syntax
field = ['dial_id', 'turn_num', 'user', 'bspn_gen','bspn']
for dial_id, turns in result_dict.items():
entry = {'dial_id': dial_id, 'turn_num': len(turns)}
# customize for the eval, always skip the first turn, so we create a dummy
for prop in field[2:]:
entry[prop] = ''
results.append(entry)
for turn_no, turn in enumerate(turns):
entry = {'dial_id': dial_id}
for key in field:
if key in ['dial_id']:
continue
v = turn.get(key, '')
if key == 'turn_domain':
v = ' '.join(v)
entry[key] = decode_fn(v, eos=eos_syntax[key]) if key in eos_syntax and v != '' else v
results.append(entry)
return results, field
class Model(object):
def __init__(self, args, test=False):
if args.back_bone=="t5":
self.tokenizer = T5Tokenizer.from_pretrained(args.model_path if test else args.pretrained_checkpoint)
self.model = T5_DST.from_pretrained(args.model_path if test else args.pretrained_checkpoint)
elif args.back_bone=="bart":
self.tokenizer = BartTokenizer.from_pretrained(args.model_path if test else args.pretrained_checkpoint)
self.model = BART_DST.from_pretrained(args.model_path if test else args.pretrained_checkpoint)
vocab = Vocab(self.model, self.tokenizer)
self.reader = MultiWozReader(vocab,args)
self.evaluator = MultiWozEvaluator(self.reader) # evaluator class
self.optim = AdamW(self.model.parameters(), lr=args.lr)
self.args = args
self.model.to(args.device)
def load_model(self):
# model_state_dict = torch.load(checkpoint)
# start_model.load_state_dict(model_state_dict)
if self.args.back_bone=="t5":
self.model = T5_DST.from_pretrained(self.args.model_path)
elif self.args.back_bone=="bart":
self.model = BART_DST.from_pretrained(self.args.model_path)
self.model.to(self.args.device)
def train(self):
btm = time.time()
step = 0
prev_min_loss = 1000
print(f"vocab_size:{self.model.config.vocab_size}")
torch.save(self.args, self.args.model_path + '/model_training_args.bin')
self.tokenizer.save_pretrained(self.args.model_path)
self.model.config.to_json_file(os.path.join(self.args.model_path, CONFIG_NAME))
self.model.train()
# lr scheduler
lr_lambda = lambda epoch: self.args.lr_decay ** epoch
scheduler = torch.optim.lr_scheduler.LambdaLR(self.optim, lr_lambda=lr_lambda)
for epoch in range(cfg.epoch_num):
log_dst = 0
log_cnt = 0
sw = time.time()
data_iterator = self.reader.get_batches('train')
for iter_num, dial_batch in enumerate(data_iterator):
py_prev = {'pv_bspn': None}
for turn_num, turn_batch in enumerate(dial_batch):
first_turn = (turn_num==0)
inputs = self.reader.convert_batch(turn_batch, py_prev, first_turn=first_turn, dst_start_token=self.model.config.decoder_start_token_id)
for k in inputs:
inputs[k] = inputs[k].to(self.args.device)
outputs = self.model(input_ids=inputs["input_ids"],
attention_mask=inputs["masks"],
decoder_input_ids=inputs["state_input"],
lm_labels=inputs["state_update"]
)
dst_loss = outputs[0]
py_prev['bspn'] = turn_batch['bspn']
total_loss = dst_loss / self.args.gradient_accumulation_steps
total_loss.backward()
torch.nn.utils.clip_grad_norm_(self.model.parameters(), self.args.max_norm)
if step % self.args.gradient_accumulation_steps == 0:
self.optim.step()
self.optim.zero_grad()
step+=1
log_dst +=float(dst_loss.item())
log_cnt += 1
if (iter_num+1)%cfg.report_interval==0:
logging.info(
'iter:{} [bspn] loss: {:.2f} time: {:.1f} turn:{} '.format(iter_num+1,
log_dst/(log_cnt+ 1e-8),
time.time()-btm,
turn_num+1))
epoch_sup_loss = log_dst/(log_cnt+ 1e-8)
do_test = False
valid_loss = self.validate(do_test=do_test)
logging.info('epoch: %d, train loss: %.3f, valid loss: %.3f, total time: %.1fmin' % (epoch+1, epoch_sup_loss,
valid_loss, (time.time()-sw)/60))
if valid_loss <= prev_min_loss:
early_stop_count = cfg.early_stop_count
prev_min_loss = valid_loss
torch.save(self.model.state_dict(), os.path.join(self.args.model_path, WEIGHTS_NAME))
logging.info('Model saved')
#self.save_model(epoch)
else:
early_stop_count -= 1
scheduler.step()
logging.info('epoch: %d early stop countdown %d' % (epoch+1, early_stop_count))
if not early_stop_count:
self.load_model()
print('result preview...')
file_handler = logging.FileHandler(os.path.join(self.args.model_path, 'eval_log%s.json'%cfg.seed))
logging.getLogger('').addHandler(file_handler)
logging.info(str(cfg))
self.eval()
return
self.load_model()
print('result preview...')
file_handler = logging.FileHandler(os.path.join(self.args.model_path, 'eval_log%s.json'%cfg.seed))
logging.getLogger('').addHandler(file_handler)
logging.info(str(cfg))
self.eval()
def validate(self, data='dev', do_test=False):
self.model.eval()
valid_loss, count = 0, 0
data_iterator = self.reader.get_batches(data)
result_collection = {}
for batch_num, dial_batch in enumerate(data_iterator):
py_prev = {'bspn': None}
for turn_num, turn_batch in enumerate(dial_batch):
first_turn = (turn_num==0)
inputs = self.reader.convert_batch(turn_batch, py_prev, first_turn=first_turn, dst_start_token=self.model.config.decoder_start_token_id)
for k in inputs:
inputs[k] = inputs[k].to(self.args.device)
dst_outputs = self.model.inference(tokenizer=self.tokenizer, reader=self.reader, prev=py_prev, input_ids=inputs['input_ids'],attention_mask=inputs["masks"])
turn_batch['bspn_gen'] = dst_outputs
py_prev['bspn'] = dst_outputs
result_collection.update(self.reader.inverse_transpose_batch(dial_batch))
results, _ = self.reader.wrap_result(result_collection)
# print(results)
jg, slot_f1, slot_acc, slot_cnt, slot_corr = self.evaluator.dialog_state_tracking_eval(results, bspn_mode='bspn')
logging.info('validation DST join goal: %2.1f slot_f1: %2.1f slot_acc: %2.1f'%(jg, slot_f1, slot_acc))
self.model.train()
if do_test:
print('result preview...')
self.eval()
return 100-jg
def eval(self, data='test'):
self.model.eval()
self.reader.result_file = None
result_collection = {}
data_iterator = self.reader.get_batches(data)
for batch_num, dial_batch in tqdm.tqdm(enumerate(data_iterator)):
py_prev = {'bspn': None}
for turn_num, turn_batch in enumerate(dial_batch):
first_turn = (turn_num==0)
inputs = self.reader.convert_batch(turn_batch, py_prev, first_turn=first_turn, dst_start_token=self.model.config.decoder_start_token_id)
for k in inputs:
inputs[k] = inputs[k].to(self.args.device)
dst_outputs = self.model.inference(tokenizer=self.tokenizer, reader=self.reader, prev=py_prev, input_ids=inputs['input_ids'],attention_mask=inputs["masks"])
turn_batch['bspn_gen'] = dst_outputs
py_prev['bspn'] = dst_outputs
result_collection.update(self.reader.inverse_transpose_batch(dial_batch))
results, field = self.reader.wrap_result(result_collection)
jg, slot_f1, slot_acc, slot_cnt, slot_corr = self.evaluator.dialog_state_tracking_eval(results, bspn_mode='bspn')
logging.info('test DST join goal: %2.1f slot_f1: %2.1f slot_acc: %2.1f'%(jg, slot_f1, slot_acc))
self.args.model_path
with open(os.path.join(self.args.model_path, 'result.txt'), 'w') as f:
f.write('test DST join goal: %2.1f slot_f1: %2.1f slot_acc: %2.1f'%(jg, slot_f1, slot_acc))
# self.reader.metric_record(metric_results)
self.model.train()
return None
def count_params(self):
module_parameters = filter(lambda p: p.requires_grad, self.m.parameters())
param_cnt = int(sum([np.prod(p.size()) for p in module_parameters]))
print('total trainable params: %d' % param_cnt)
return param_cnt
def parse_arg_cfg(args):
if args.cfg:
for pair in args.cfg:
k, v = tuple(pair.split('='))
dtype = type(getattr(cfg, k))
if dtype == type(None):
raise ValueError()
if dtype is bool:
v = False if v == 'False' else True
elif dtype is list:
v = v.split(',')
if k=='cuda_device':
v = [int(no) for no in v]
else:
v = dtype(v)
setattr(cfg, k, v)
return
def main():
if not os.path.exists('./experiments_DST'):
os.mkdir('./experiments_DST')
parser = argparse.ArgumentParser()
parser.add_argument('--mode')
parser.add_argument('--cfg', nargs='*')
parser.add_argument("--device", type=str, default="cuda" if torch.cuda.is_available() else "cpu", help="Device (cuda or cpu)")
parser.add_argument("--max_norm", type=float, default=1.0, help="Clipping gradient norm")
parser.add_argument("--lr", type=float, default=6e-4, help="Learning rate")
parser.add_argument("--gradient_accumulation_steps", type=int, default=2, help="Accumulate gradients on several steps")
parser.add_argument("--pretrained_checkpoint", type=str, default="t5-small", help="Path, url or short name of the model")
parser.add_argument("--model_path", type=str, default="")
parser.add_argument("--context_window", type=int, default=5, help="how many previous turns for model input")
parser.add_argument("--lr_decay", type=float, default=0.8, help="Learning rate decay")
parser.add_argument("--back_bone", type=str, default="t5", help="choose t5 or bart")
parser.add_argument("--noupdate_dst", action='store_true', help="dont use update base DST")
args = parser.parse_args()
cfg.mode = args.mode
if args.mode == 'test':
parse_arg_cfg(args)
cfg_load = json.loads(open(os.path.join(args.model_path, 'exp_cfg.json'), 'r').read())
for k, v in cfg_load.items():
if k in ['mode', 'cuda', 'cuda_device', 'eval_per_domain', 'use_true_pv_resp',
'use_true_prev_bspn','use_true_prev_aspn','use_true_curr_bspn','use_true_curr_aspn',
'name_slot_unable', 'book_slot_unable','count_req_dials_only','log_time', 'model_path',
'result_path', 'model_parameters', 'multi_gpu', 'use_true_bspn_for_ctr_eval', 'nbest',
'limit_bspn_vocab', 'limit_aspn_vocab', 'same_eval_as_cambridge', 'beam_width',
'use_true_domain_for_ctr_eval', 'use_true_prev_dspn', 'aspn_decode_mode',
'beam_diverse_param', 'same_eval_act_f1_as_hdsa', 'topk_num', 'nucleur_p',
'act_selection_scheme', 'beam_penalty_type', 'record_mode']:
continue
setattr(cfg, k, v)
cfg.result_path = os.path.join(args.model_path, 'result.csv')
else:
parse_arg_cfg(args)
if args.model_path=="":
args.model_path = 'experiments_DST/{}_sd{}_lr{}_bs{}_sp{}_dc{}_cw{}_model_{}_noupdate{}/'.format('-'.join(cfg.exp_domains), cfg.seed, args.lr, cfg.batch_size,
cfg.early_stop_count, args.lr_decay, args.context_window, args.pretrained_checkpoint, args.noupdate_dst)
if not os.path.exists(args.model_path):
os.makedirs(args.model_path)
cfg.result_path = os.path.join(args.model_path, 'result.csv')
cfg.eval_load_path = args.model_path
cfg._init_logging_handler(args.mode)
torch.manual_seed(cfg.seed)
torch.cuda.manual_seed(cfg.seed)
random.seed(cfg.seed)
np.random.seed(cfg.seed)
#cfg.model_parameters = m.count_params()
logging.info(str(cfg))
if args.mode == 'train':
with open(os.path.join(args.model_path, 'exp_cfg.json'), 'w') as f:
json.dump(cfg.__dict__, f, indent=2)
m = Model(args)
m.train()
elif args.mode == 'test':
m = Model(args,test=True)
m.eval(data='test')
if __name__ == '__main__':
main()