-
Notifications
You must be signed in to change notification settings - Fork 0
/
train.py
324 lines (277 loc) · 14.1 KB
/
train.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
import random
import numpy as np
import torch
from torch.utils.data import DataLoader
from torch.nn.parallel import DistributedDataParallel
from transformers import AdamW
try:
from apex import amp
except ImportError:
pass
import os
import json
import yaml
from argparse import ArgumentParser
from collections import defaultdict
from tqdm import tqdm
from torch.utils.tensorboard import SummaryWriter
from models import ModelRegistry
from datasets import DatasetRegistry
from evaluation import EvaluatorRegistry
from utils import CheckpointSaver, get_logger, get_save_dir, get_data_sizes, get_parameter_groups, get_lr_scheduler
def get_args():
parser = ArgumentParser()
parser.add_argument('--name',
type=str,
required=True,
help='Experiment name.')
parser.add_argument('--model',
type=str,
default='roberta-base',
help='Pretrained model name or path.')
parser.add_argument('--task',
type=str,
default='GT',
choices=['MLM', 'GT', 'QA'],
help='Training task. The options are: '
'MLM (Masked Language Modeling), '
'GT (Gapped Text), '
'QA (Question Answering).')
parser.add_argument('--save_dir',
type=str,
default='experiments',
help='Base directory for saving information.')
parser.add_argument('--data_dir',
type=str,
default='data/GT',
help='Directory with training and evaluation data.')
parser.add_argument('--batch_size',
type=int,
default=8,
help='This is the number of training samples processed simultaneously by one GPU. '
'The effective batch size (number of training samples processed per one '
'optimization step) is equal to batch_size * num_gpus * accumulation_steps.')
parser.add_argument('--accumulation_steps',
type=int,
default=1,
help='Number of gradient accumulation steps.')
parser.add_argument('--amp',
default=False,
action='store_true',
help='Use apex amp for mixed precision training.')
parser.add_argument('--amp_opt_level',
type=str,
default='O1',
choices=['O0', 'O1', 'O2', 'O3'],
help='Apex amp optimization level. Only used if amp is True.')
parser.add_argument('--num_epochs',
type=int,
default=1,
help='Number of training epochs.')
parser.add_argument('--max_steps',
type=int,
default=-1,
help='Maximum number of training steps. '
'Can be used to stop training before the end of an epoch.')
parser.add_argument('--learning_rate',
type=float,
default=1e-5,
help='Maximum learning rate for AdamW.')
parser.add_argument('--warmup_proportion',
type=float,
default=0.1,
help='Proportion of training steps to perform linear learning rate warmup for.')
parser.add_argument('--max_checkpoints',
type=int,
default=10,
help='Maximum number of model and optimizer checkpoints to keep.')
parser.add_argument('--eval_every',
type=int,
default=50000,
help='Evaluate model after processing this many training samples.')
parser.add_argument('--do_not_eval_after_epoch',
default=False,
action='store_true',
help='Do not evaluate model at the end of every epoch.')
parser.add_argument('--weight_decay',
type=float,
default=0,
help='Regularization.')
parser.add_argument('--seed',
type=int,
default=111,
help='Random seed.')
parser.add_argument('--local_rank',
type=int,
default=-1,
help='Local rank for distributed training. '
'This argument is provided by torch.distributed.launch.')
args = parser.parse_args()
return args
def train(args, logger, tb_writer):
logger.info('Args: {}'.format(json.dumps(vars(args), indent=4, sort_keys=True)))
if args.local_rank in [-1, 0]:
with open(os.path.join(args.save_dir, 'args.yaml'), 'w') as file:
yaml.safe_dump(vars(args), file, sort_keys=False)
device_id = args.local_rank if args.local_rank != -1 else 0
device = torch.device('cuda', device_id)
logger.warning(f'Using GPU {args.local_rank}.')
world_size = torch.distributed.get_world_size() if args.local_rank != -1 else 1
logger.info(f'Total number of GPUs used: {world_size}.')
effective_batch_size = args.batch_size * world_size * args.accumulation_steps
logger.info(f'Effective batch size: {effective_batch_size}.')
num_train_samples_per_epoch, num_dev_samples, num_unique_train_epochs = get_data_sizes(data_dir=args.data_dir,
num_epochs=args.num_epochs,
logger=logger)
num_optimization_steps = sum(num_train_samples_per_epoch) // world_size // args.batch_size // \
args.accumulation_steps
if args.max_steps > 0:
num_optimization_steps = min(num_optimization_steps, args.max_steps)
logger.info(f'Total number of optimization steps: {num_optimization_steps}.')
# Set random seed
logger.info(f'Using random seed {args.seed}.')
random.seed(args.seed)
np.random.seed(args.seed)
torch.manual_seed(args.seed)
torch.cuda.manual_seed_all(args.seed)
# Get model
if args.local_rank not in [-1, 0]:
torch.distributed.barrier()
logger.info(f'Loading model {args.model} for task {args.task}...')
model = ModelRegistry.get_model(args.task).from_pretrained(args.model)
if args.local_rank in [-1, 0]:
with open(os.path.join(args.save_dir, 'config.json'), 'w') as file:
json.dump(model.config.__dict__, file)
if args.local_rank == 0:
torch.distributed.barrier()
model.to(device)
# Get optimizer
logger.info('Creating optimizer...')
parameter_groups = get_parameter_groups(model)
optimizer = AdamW(parameter_groups, lr=args.learning_rate, weight_decay=args.weight_decay, eps=1e-8)
scheduler = get_lr_scheduler(optimizer, num_steps=num_optimization_steps, warmup_proportion=args.warmup_proportion)
if args.amp:
amp.register_half_function(torch, 'einsum')
model, optimizer = amp.initialize(model, optimizer, opt_level=args.amp_opt_level)
if args.local_rank != -1:
model = DistributedDataParallel(model,
device_ids=[args.local_rank],
output_device=args.local_rank,
find_unused_parameters=True)
# Get dev data loader
dev_data_file = os.path.join(args.data_dir, f'dev.jsonl.gz')
logger.info(f'Creating dev dataset from {dev_data_file}...')
dev_dataset = DatasetRegistry.get_dataset(args.task)(data_file=dev_data_file,
data_size=num_dev_samples,
local_rank=-1)
dev_loader = DataLoader(dev_dataset,
batch_size=2 * args.batch_size,
num_workers=1,
collate_fn=dev_dataset.collate_fn)
# Get evaluator
evaluator = EvaluatorRegistry.get_evaluator(args.task)(data_loader=dev_loader,
logger=logger,
tb_writer=tb_writer,
device=device,
world_size=world_size,
args=args)
# Get saver
saver = CheckpointSaver(save_dir=args.save_dir,
max_checkpoints=args.max_checkpoints,
primary_metric=evaluator.primary_metric,
maximize_metric=evaluator.maximize_metric,
logger=logger)
global_step = 0
samples_processed = 0
# Train
logger.info('Training...')
samples_till_eval = args.eval_every
for epoch in range(1, args.num_epochs + 1):
# Get train data loader for current epoch
train_data_file_num = ((epoch - 1) % num_unique_train_epochs) + 1
train_data_file = os.path.join(args.data_dir, f'epoch_{train_data_file_num}.jsonl.gz')
logger.info(f'Creating training dataset from {train_data_file}...')
train_dataset = DatasetRegistry.get_dataset(args.task)(train_data_file,
data_size=num_train_samples_per_epoch[epoch - 1],
local_rank=args.local_rank,
world_size=world_size)
train_loader = DataLoader(train_dataset,
batch_size=args.batch_size,
num_workers=1,
collate_fn=train_dataset.collate_fn)
logger.info(f'Starting epoch {epoch}...')
model.train()
model.zero_grad()
loss_values = defaultdict(float)
samples_till_end = (num_optimization_steps - global_step) * effective_batch_size
samples_in_cur_epoch = min([len(train_loader.dataset), samples_till_end])
disable_progress_bar = (args.local_rank not in [-1, 0])
with tqdm(total=samples_in_cur_epoch, disable=disable_progress_bar) as progress_bar:
for step, batch in enumerate(train_loader, 1):
batch = {name: tensor.to(device) for name, tensor in batch.items()}
current_batch_size = batch['input_ids'].shape[0]
outputs = model(**batch)
loss, current_loss_values = outputs[:2]
loss = loss / args.accumulation_steps
for name, value in current_loss_values.items():
loss_values[name] += value / args.accumulation_steps
if args.amp:
with amp.scale_loss(loss, optimizer) as scaled_loss:
scaled_loss.backward()
else:
loss.backward()
samples_processed += current_batch_size * world_size
samples_till_eval -= current_batch_size * world_size
progress_bar.update(current_batch_size * world_size)
if step % args.accumulation_steps == 0:
current_lr = scheduler.get_last_lr()[0]
if args.amp:
torch.nn.utils.clip_grad_norm_(amp.master_params(optimizer), 1.0)
else:
torch.nn.utils.clip_grad_norm_(model.parameters(), 1.0)
optimizer.step()
scheduler.step()
model.zero_grad()
global_step += 1
# Log info
progress_bar.set_postfix(epoch=epoch, step=global_step, lr=current_lr, **loss_values)
if args.local_rank in [-1, 0]:
tb_writer.add_scalar('train/LR', current_lr, global_step)
for name, value in loss_values.items():
tb_writer.add_scalar(f'train/{name}', value, global_step)
loss_values = {name: 0 for name in loss_values}
if global_step == args.max_steps:
logger.info('Reached maximum number of optimization steps.')
break
if samples_till_eval <= 0:
samples_till_eval = args.eval_every
eval_results = evaluator.evaluate(model, global_step)
if args.local_rank in [-1, 0]:
saver.save(model, global_step, eval_results)
if not args.do_not_eval_after_epoch:
eval_results = evaluator.evaluate(model, global_step)
if args.local_rank in [-1, 0]:
saver.save(model, global_step, eval_results)
if __name__ == '__main__':
args = get_args()
if args.local_rank != -1:
torch.cuda.set_device(args.local_rank)
torch.distributed.init_process_group(backend='nccl')
if args.local_rank in [-1, 0]:
args.save_dir = get_save_dir(args.save_dir, args.name)
logger = get_logger(args.save_dir, args.name, log_file=f'log_0.txt')
logger.info(f'Results will be saved to {args.save_dir}.')
tb_writer = SummaryWriter(args.save_dir)
else:
torch.distributed.barrier()
args.save_dir = get_save_dir(args.save_dir, args.name, use_existing_dir=True)
logger = get_logger(args.save_dir, args.name, verbose=False, log_file=f'log_{args.local_rank}.txt')
tb_writer = None
if args.local_rank == 0:
torch.distributed.barrier()
try:
train(args, logger, tb_writer)
except:
logger.exception('An error occured...')
if tb_writer is not None:
tb_writer.close()