-
Notifications
You must be signed in to change notification settings - Fork 8
/
main.py
346 lines (297 loc) · 13.5 KB
/
main.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
#!/usr/bin/env python3
import gc
import os
import sys
import time
import torch
import data_utils
import models
from data_utils import to_torch
from eval_metric import mrr
from model_utils import get_eval_string
from model_utils import get_gold_pred_str
from model_utils import get_output_index
from model_utils import metric_dicts
from model_utils import fine_grained_eval
from tensorboardX import SummaryWriter
from torch import optim
import numpy as np
import random
from tqdm import tqdm
sys.path.insert(0, './resources')
import constant
from config_parser import get_logger
from config_parser import read_args
from label_corr import build_concurr_matrix
def get_data_gen(dataname, mode, args, vocab_set, goal, eval_epoch=1):
dataset = data_utils.TypeDataset(constant.FILE_ROOT + dataname, lstm_type=args.lstm_type,
goal=goal, vocab=vocab_set)
if mode == 'train':
data_gen = dataset.get_batch(args.batch_size, args.num_epoch, forever=False, eval_data=False,
simple_mention=not args.enhanced_mention, shuffle=True)
elif mode == 'dev':
if args.goal == 'onto':
eval_batch_size = 2202
else:
eval_batch_size = 1998
data_gen = dataset.get_batch(eval_batch_size, 1, forever=True, eval_data=True,
simple_mention=not args.enhanced_mention)
else:
if args.goal == "onto":
if 'dev' in dataname:
eval_batch_size = 2202
else:
eval_batch_size = 8963
else:
eval_batch_size = 1998
# eval_batch_size = 20
data_gen = dataset.get_batch(eval_batch_size, eval_epoch, forever=False, eval_data=True,
simple_mention=not args.enhanced_mention)
return data_gen
def get_joint_datasets(args):
vocab = data_utils.get_vocab(args.embed_source)
train_gen_list = []
valid_gen_list = []
if args.mode == 'train':
if not args.remove_open and not args.only_crowd:
train_gen_list.append(
#`("open", get_data_gen('train/open*.json', 'train', args, vocab, "open")))
("open", get_data_gen('distant_supervision/headword_train.json', 'train', args, vocab, "open")))
valid_gen_list.append(("open", get_data_gen('distant_supervision/headword_dev.json', 'dev', args, vocab, "open")))
if not args.remove_el and not args.only_crowd:
valid_gen_list.append(
("wiki",
get_data_gen('distant_supervision/el_dev.json', 'dev', args, vocab, "wiki" if args.multitask else "open")))
train_gen_list.append(
("wiki",
get_data_gen('distant_supervision/el_train.json', 'train', args, vocab, "wiki" if args.multitask else "open")))
#get_data_gen('train/el_train.json', 'train', args, vocab, "wiki" if args.multitask else "open")))
if args.add_crowd or args.only_crowd:
train_gen_list.append(
("open", get_data_gen('crowd/train_m.json', 'train', args, vocab, "open")))
crowd_dev_gen = get_data_gen('crowd/dev.json', 'dev', args, vocab, "open")
return train_gen_list, valid_gen_list, crowd_dev_gen
def get_datasets(data_lists, args, eval_epoch=1):
data_gen_list = []
vocab_set = data_utils.get_vocab(args.embed_source)
for dataname, mode, goal in data_lists:
data_gen_list.append(get_data_gen(dataname, mode, args, vocab_set, goal, eval_epoch))
return data_gen_list
def _train(args):
logger = get_logger(args)
if args.data_setup == 'joint':
train_gen_list, val_gen_list, crowd_dev_gen = get_joint_datasets(args)
else:
train_fname = args.train_data
dev_fname = args.dev_data
data_gens = get_datasets([(train_fname, 'train', args.goal),
(dev_fname, 'dev', args.goal)], args)
train_gen_list = [(args.goal, data_gens[0])]
val_gen_list = [(args.goal, data_gens[1])]
if args.goal == 'onto':
validation_log = SummaryWriter(os.path.join(constant.EXP_ROOT_ONTO, args.model_id, "log", "validation"))
else:
validation_log = SummaryWriter(os.path.join(constant.EXP_ROOT, args.model_id, "log", "validation"))
model = models.Model(args, constant.ANSWER_NUM_DICT[args.goal])
model.cuda()
total_loss = 0
start_time = time.time()
init_time = time.time()
optimizer = optim.Adam(model.parameters(), lr=args.learning_rate)
if args.use_lr_schedule:
scheduler = optim.lr_scheduler.MultiStepLR(optimizer, [1000], gamma=0.1)
if args.load:
load_model(args.reload_model_name, constant.EXP_ROOT, args.model_id, model, optimizer)
best_f1 = 0
logger.info('Start training......')
while True:
model.batch_num += 1 # single batch composed of all train signal passed by.
if args.use_lr_schedule:
scheduler.step()
for (type_name, data_gen) in train_gen_list:
try:
batch = next(data_gen)
batch, _ = to_torch(batch)
except StopIteration:
logger.info(type_name + " finished at " + str(model.batch_num))
torch.save({'state_dict': model.state_dict(), 'optimizer': optimizer.state_dict()},
'{0:s}/{1:s}.pt'.format(constant.EXP_ROOT, args.model_id))
return
optimizer.zero_grad()
loss, output_logits = model(batch, type_name)
loss.backward()
total_loss += loss.item()
optimizer.step()
if model.batch_num % args.log_period == 0 and model.batch_num > 0:
gc.collect()
cur_loss = float(1.0 * loss.item())
elapsed = time.time() - start_time
train_loss_str = ('|loss {0:3f} | at {1:d}step | @ {2:.2f} ms/batch'.format(cur_loss, model.batch_num,elapsed * 1000 / args.log_period))
start_time = time.time()
logger.info(train_loss_str)
if model.batch_num % args.eval_period == 0 and model.batch_num > 0:
eval_start = time.time()
logger.info('---- eval at step {0:d} ---'.format(model.batch_num))
if args.goal == 'onto':
val_type = "onto"
feed_dict = next(val_gen_list[0][1])
EXP_ROOT = constant.EXP_ROOT_ONTO
else:
val_type = "open"
feed_dict = next(crowd_dev_gen)
EXP_ROOT = constant.EXP_ROOT
eval_batch, _ = to_torch(feed_dict)
total_eval_loss, gold_preds = evaluate_batch(model.batch_num, eval_batch, model, val_type, args.goal)
eval_result, output_str = metric_dicts(gold_preds)
if args.use_lr_schedule:
scheduler.step(eval_result['ma_f1'])
if eval_result['ma_f1'] > 0.78 or args.goal == "open":
if eval_result['ma_f1'] > best_f1 or model.batch_num > 10000:
# added for regularization based baselines
if args.add_regu and model.batch_num < 8000:
break
if eval_result['ma_f1'] > best_f1:
best_f1 = eval_result['ma_f1']
save_fname = '{0:s}/{1:s}_{2:f}.pt'.format(EXP_ROOT, args.model_id, eval_result['ma_f1'])
torch.save({'state_dict': model.state_dict(), 'optimizer': optimizer.state_dict()}, save_fname)
logger.critical(
'Found best. Total {0:.2f} minutes have passed, saving at {1:s} '.format((time.time() - init_time) / 60, save_fname))
elif args.goal != "open":
save_fname = '{0:s}/{1:s}_{2:f}.pt'.format(EXP_ROOT, args.model_id, eval_result['ma_f1'])
torch.save({'state_dict': model.state_dict(), 'optimizer': optimizer.state_dict()}, save_fname)
logger.critical(
'Found best. Total {0:.2f} minutes have passed, saving at {1:s} '.format((time.time() - init_time) / 60, save_fname))
logger.info('eval loss total: ' + str(total_eval_loss))
logger.info('eval performance: ' + output_str)
validation_log.add_scalar('eval_crowd_loss', total_eval_loss, model.batch_num)
validation_log.add_scalar('eval_crowd_mi_f1', eval_result["f1"], model.batch_num)
validation_log.add_scalar('eval_crowd_ma_f1', eval_result["ma_f1"], model.batch_num)
validation_log.add_scalar('eval_crowd_ma_p', eval_result["ma_precision"], model.batch_num)
validation_log.add_scalar('eval_crowd_ma_recall', eval_result["ma_recall"], model.batch_num)
logger.info('Eval time clipse {}s'.format(time.time() - eval_start))
if model.batch_num > args.max_batch:
break
# Training finished!
torch.save({'state_dict': model.state_dict(), 'optimizer': optimizer.state_dict()},
'{0:s}/{1:s}.pt'.format(constant.EXP_ROOT, args.model_id))
def evaluate_batch(batch_num, eval_batch, model, val_type_name, goal):
model.eval()
loss, output_logits = model(eval_batch, val_type_name)
output_index = get_output_index(output_logits)
# eval_loss = loss.data.cpu().clone()[0]
eval_loss = loss.item()
# eval_loss_str = 'Eval loss: {0:.7f} at step {1:d}'.format(eval_loss, model.batch_num)
gold_pred = get_gold_pred_str(output_index, eval_batch['y'].data.cpu().clone(), goal)
# eval_accu = sum([set(y) == set(yp) for y, yp in gold_pred]) * 1.0 / len(gold_pred)
# eval_accus = [set(y) == set(yp) for y, yp in gold_pred]
# tensorboard.add_validation_scalar('eval_acc_' + val_type_name, eval_accu, model.batch_num)
# tensorboard.add_validation_scalar('eval_loss_' + val_type_name, eval_loss, model.batch_num)
# eval_str = get_eval_string(gold_pred)
# print(val_type_name + ":" +eval_loss_str)
# print(gold_pred[:3])
# print(val_type_name+":"+ eval_str)
# logging.info(val_type_name + ":" + eval_loss_str)
# logging.info(val_type_name +":" + eval_str)
model.train()
return eval_loss, gold_pred
def load_model(reload_model_name, save_dir, model_id, model, optimizer=None):
if reload_model_name:
model_file_name = '{0:s}/{1:s}.pt'.format(save_dir, reload_model_name)
else:
model_file_name = '{0:s}/{1:s}.pt'.format(save_dir, model_id)
checkpoint = torch.load(model_file_name)
model.load_state_dict(checkpoint['state_dict'])
if optimizer:
optimizer.load_state_dict(checkpoint['optimizer'])
else:
total_params = 0
# Log params
for k in checkpoint['state_dict']:
elem = checkpoint['state_dict'][k]
param_s = 1
for size_dim in elem.size():
param_s = size_dim * param_s
print(k, elem.size())
total_params += param_s
param_str = ('Number of total parameters..{0:d}'.format(total_params))
print(param_str)
print('Loading model from ... {0:s}'.format(model_file_name))
def visualize(args):
saved_path = constant.EXP_ROOT
model = models.Model(args, constant.ANSWER_NUM_DICT[args.goal])
model.cuda()
model.eval()
model.load_state_dict(torch.load(saved_path + '/' + args.model_id + '_best.pt')["state_dict"])
label2id = constant.ANS2ID_DICT["open"]
visualize = SummaryWriter("../visualize/" + args.model_id)
# label_list = ["person", "leader", "president", "politician", "organization", "company", "athlete","adult", "male", "man", "television_program", "event"]
label_list = list(label2id.keys())
ids = [label2id[_] for _ in label_list]
if args.gcn:
# connection_matrix = model.decoder.label_matrix + model.decoder.weight * model.decoder.affinity
connection_matrix = model.decoder.label_matrix + model.decoder.weight * model.decoder.affinity
label_vectors = model.decoder.transform(connection_matrix.mm(model.decoder.linear.weight) / connection_matrix.sum(1, keepdim=True))
else:
label_vectors = model.decoder.linear.weight.data
interested_vectors = torch.index_select(label_vectors, 0, torch.tensor(ids).to(torch.device("cuda")))
visualize.add_embedding(interested_vectors, metadata=label_list, label_img=None)
def _test(args):
assert args.load
test_fname = args.eval_data
model = models.Model(args, constant.ANSWER_NUM_DICT[args.goal])
model.cuda()
model.eval()
# load_model(args.reload_model_name, constant.EXP_ROOT, args.model_id, model)
if args.goal == "onto":
saved_path = constant.EXP_ROOT_ONTO
else:
saved_path = constant.EXP_ROOT
model.load_state_dict(torch.load(saved_path + '/' + args.model_id + '_best.pt')["state_dict"])
data_gens = get_datasets([(test_fname, 'test', args.goal)], args, eval_epoch=1)
for name, dataset in [(test_fname, data_gens[0])]:
print('Processing... ' + name)
batch = next(dataset)
eval_batch, _ = to_torch(batch)
loss, output_logits = model(eval_batch, args.goal)
threshes = np.arange(0,1,0.02)
# threshes = [0.65, 0.68, 0.7, 0.71]
# threshes = [0.5]
p_and_r = []
for thresh in tqdm(threshes):
total_gold_pred = []
total_probs = []
total_ys = []
print('\nthresh {}'.format(thresh))
output_index = get_output_index(output_logits, thresh)
output_prob = model.sigmoid_fn(output_logits).data.cpu().clone().numpy()
y = eval_batch['y'].data.cpu().clone().numpy()
gold_pred = get_gold_pred_str(output_index, y, args.goal)
total_probs.extend(output_prob)
total_ys.extend(y)
total_gold_pred.extend(gold_pred)
# mrr_val = mrr(total_probs, total_ys)
# json.dump(gold_pred, open('nomulti_predictions.json', 'w'))
# np.save('y', total_ys)
# np.save('probs', total_probs)
# print('mrr_value: ', mrr_val)
# result, eval_str = metric_dicts(total_gold_pred)
result, eval_str = fine_grained_eval(total_gold_pred)
# fine_grained_eval(total_gold_pred)
p_and_r.append([result["ma_precision"], result["ma_recall"]])
print(eval_str)
np.save(saved_path + '/{}_pr_else_dev'.format(args.model_id), p_and_r)
if __name__ == '__main__':
config = read_args()
# fix random seed
np.random.seed(config.seed)
random.seed(config.seed)
torch.cuda.manual_seed(config.seed)
torch.cuda.manual_seed_all(config.seed)
if config.mode == 'train':
_train(config)
elif config.mode == 'test':
_test(config)
elif config.mode == 'visual':
visualize(config)
else:
raise ValueError("invalid value for 'mode': {}".format(config.mode))