-
Notifications
You must be signed in to change notification settings - Fork 6
/
Copy pathmain_pretrain.py
358 lines (298 loc) · 15.8 KB
/
main_pretrain.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
# ------------------------------------------------------------------------
# SiameseIM
# Copyright (c) SenseTime. All Rights Reserved.
# ------------------------------------------------------------------------
# Modified from MAE (https://github.com/facebookresearch/mae)
# Copyright (c) Meta Platforms, Inc. and affiliates. All Rights Reserved.
# ------------------------------------------------------------------------
# References:
# timm: https://github.com/rwightman/pytorch-image-models/tree/master/timm
# DeiT: https://github.com/facebookresearch/deit
# ------------------------------------------------------------------------
import argparse
import datetime
import json
import numpy as np
import os
import time
from pathlib import Path
import torch
import torch.distributed as dist
import torch.backends.cudnn as cudnn
from torch.utils.tensorboard import SummaryWriter
import torchvision.transforms as transforms
import torchvision.datasets as datasets
import timm
assert timm.__version__ == "0.6.12" # version check
from timm.optim.optim_factory import param_groups_weight_decay
from timm.optim import create_optimizer
import util.misc as misc
from util.misc import NativeScalerWithGradNormCount as NativeScaler
from util.augmentation import RandomResizedCrop, GaussianBlur, SingleRandomResizedCrop, RandomHorizontalFlip, Solarize
from util.datasets import ImagenetWithMask
import models_sim
from engine_pretrain import train_one_epoch
def get_args_parser():
parser = argparse.ArgumentParser('MAE pre-training', add_help=False)
parser.add_argument('--batch_size', default=64, type=int,
help='Batch size per GPU (effective batch size is batch_size * accum_iter * # gpus')
parser.add_argument('--epochs', default=400, type=int)
parser.add_argument('--accum_iter', default=1, type=int,
help='Accumulate gradient iterations (for increasing the effective batch size under memory constraints)')
# Model parameters
parser.add_argument('--model', default='mae_vit_large_patch16', type=str, metavar='MODEL',
help='Name of model to train')
parser.add_argument('--input_size', default=224, type=int,
help='images input size')
parser.add_argument('--mask_ratio', default=0.75, type=float,
help='Masking ratio (percentage of removed patches).')
parser.add_argument('--norm_pix_loss', action='store_true',
help='Use (per-patch) normalized pixels as targets for computing loss')
parser.set_defaults(norm_pix_loss=False)
parser.add_argument('--use_abs_pos_emb', default=True, action='store_true')
parser.add_argument('--disable_abs_pos_emb', dest='use_abs_pos_emb', action='store_false')
parser.add_argument('--use_shared_rel_pos_bias', default=False, action='store_true')
# Optimizer parameters
parser.add_argument('--weight_decay', type=float, default=0.05,
help='weight decay (default: 0.05)')
parser.add_argument('--lr', type=float, default=None, metavar='LR',
help='learning rate (absolute lr)')
parser.add_argument('--blr', type=float, default=1e-3, metavar='LR',
help='base learning rate: absolute_lr = base_lr * total_batch_size / 256')
parser.add_argument('--min_lr', type=float, default=0., metavar='LR',
help='lower lr bound for cyclic schedulers that hit 0')
parser.add_argument('--warmup_epochs', type=int, default=40, metavar='N',
help='epochs to warmup LR')
# Dataset parameters
parser.add_argument('--data_path', default='/datasets01/imagenet_full_size/061417/', type=str,
help='dataset path')
parser.add_argument('--output_dir', default='./output_dir',
help='path where to save, empty for no saving')
parser.add_argument('--log_dir', default='./output_dir',
help='path where to tensorboard log')
parser.add_argument('--device', default='cuda',
help='device to use for training / testing')
parser.add_argument('--seed', default=0, type=int)
parser.add_argument('--resume', default='',
help='resume from checkpoint')
parser.add_argument('--start_epoch', default=0, type=int, metavar='N',
help='start epoch')
parser.add_argument('--num_workers', default=10, type=int)
parser.add_argument('--pin_mem', action='store_true',
help='Pin CPU memory in DataLoader for more efficient (sometimes) transfer to GPU.')
parser.add_argument('--no_pin_mem', action='store_false', dest='pin_mem')
# parser.set_defaults(pin_mem=True)
# distributed training parameters
parser.add_argument('--world_size', default=1, type=int,
help='number of distributed processes')
parser.add_argument('--local_rank', default=-1, type=int)
parser.add_argument('--dist_on_itp', action='store_true')
parser.add_argument('--dist_url', default='env://',
help='url used to set up distributed training')
# SiameseIM parameters
# data
parser.add_argument('--crop_min', default=0.2, type=float)
parser.add_argument('--use_tcs_dataset', default=False, action='store_true')
# model
parser.add_argument('--decoder_embed_dim', default=512, type=int)
parser.add_argument('--drop_path_rate', default=0.0, type=float)
parser.add_argument('--init_values', default=None, type=float)
parser.add_argument('--projector_depth', default=2, type=int)
parser.add_argument('--predictor_depth', default=4, type=int)
parser.add_argument('--use_proj_ln', default=False, action='store_true')
parser.add_argument('--use_pred_ln', default=False, action='store_true')
parser.add_argument('--train_patch_embed', default=False, action='store_true')
parser.add_argument('--online_ln', default=False, action='store_true', help='also use frozen LN in online branch')
parser.add_argument('--loss_type', default='mae')
parser.add_argument('--neg_weight', default=0.02, type=float)
parser.add_argument('--with_blockwise_mask', default=False, action='store_true')
parser.add_argument('--blockwise_num_masking_patches', default=75, type=int)
# hyper-parameter
parser.add_argument('--mm', default=0.996, type=float)
parser.add_argument('--mmschedule', default='const')
parser.add_argument('--lambda_F', default=50, type=float) # may no need
parser.add_argument('--T', default=0.2, type=float) # check
parser.add_argument('--clip_grad', default=None, type=float)
parser.add_argument('--beta2', default=0.95, type=float)
# misc
parser.add_argument('--auto_resume', default=True)
parser.add_argument('--save_freq', default=50, type=int)
parser.add_argument('--save_latest_freq', default=1, type=int)
parser.add_argument('--fp32', default=False, action='store_true')
parser.add_argument('--amp_growth_interval', default=2000, type=int)
return parser
class DataAugmentationForSIM(object):
def __init__(self, args):
self.args = args
self.random_resized_crop = SingleRandomResizedCrop(args.input_size, scale=(args.crop_min, 1.0), interpolation=3)
self.random_flip = RandomHorizontalFlip()
self.color_transform1 = transforms.Compose([
transforms.RandomApply([
transforms.ColorJitter(0.4, 0.4, 0.2, 0.1) # not strengthened
], p=0.8),
transforms.RandomGrayscale(p=0.2),
transforms.RandomApply([GaussianBlur([.1, 2.])], p=1.0),
])
self.color_transform2 = transforms.Compose([
transforms.RandomApply([
transforms.ColorJitter(0.4, 0.4, 0.2, 0.1) # not strengthened
], p=0.8),
transforms.RandomGrayscale(p=0.2),
transforms.RandomApply([GaussianBlur([.1, 2.])], p=0.1),
transforms.RandomApply([Solarize()], p=0.2),
])
self.format_transform = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize(
mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
def __call__(self, image):
spatial_image1, flip1 = self.random_flip(image)
spatial_image2, flip2 = self.random_flip(image)
spatial_image1, i1, j1, h1, w1, W = self.random_resized_crop(spatial_image1)
spatial_image2, i2, j2, h2, w2, W = self.random_resized_crop(spatial_image2)
color_image1 = self.color_transform1(spatial_image1)
color_image2 = self.color_transform2(spatial_image2)
relative_flip = (flip1 and not flip2) or (flip2 and not flip1)
return self.format_transform(color_image1), self.format_transform(color_image2), \
(i2-i1)/h1, (j2-j1)/w1, h2/h1, w2/w1, relative_flip, (W-j1-j2)/w1
def __repr__(self):
repr = "(DataAugmentation,\n"
repr += " transform = %s,\n" % str(self.random_resized_crop) + str(self.random_flip) + str(self.color_transform1) + str(self.format_transform)
repr += ")"
return repr
def main(args):
misc.init_distributed_mode(args) # need change to torch.engine
print('job dir: {}'.format(os.path.dirname(os.path.realpath(__file__))))
print("{}".format(args).replace(', ', ',\n'))
device = torch.device(args.device)
# fix the seed for reproducibility
seed = args.seed + misc.get_rank()
torch.manual_seed(seed)
np.random.seed(seed)
cudnn.benchmark = True
# disable tf32
torch.backends.cuda.matmul.allow_tf32 = False
torch.backends.cudnn.allow_tf32 = False
# build augmentation and dataset
if args.loss_type in ['sim']:
transform_train = DataAugmentationForSIM(args)
else:
transform_train = transforms.Compose([
transforms.RandomResizedCrop(args.input_size, scale=(0.2, 1.0), interpolation=3), # 3 is bicubic
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])])
if not args.use_tcs_dataset:
dataset_train = datasets.ImageFolder(os.path.join(args.data_path, 'train'), transform=transform_train)
dataset_train = ImagenetWithMask(os.path.join(args.data_path, 'train'),
transform=transform_train,
with_blockwise_mask=args.with_blockwise_mask,
blockwise_num_masking_patches=args.blockwise_num_masking_patches)
else: # for internal use only
from util.tcs_datasets import ImagenetTCSDataset
dataset_train = ImagenetTCSDataset('train',
's3://imagenet',
use_tcs=True,
transform=transform_train,
with_blockwise_mask=args.with_blockwise_mask,
blockwise_num_masking_patches=args.blockwise_num_masking_patches,
local_rank=int(os.environ['LOCAL_RANK']),
local_size=int(os.environ['LOCAL_SIZE']),
tcs_conf_path='./petreloss.conf')
print(dataset_train)
# build dataloader
if True: # args.distributed:
num_tasks = misc.get_world_size()
global_rank = misc.get_rank()
sampler_train = torch.utils.data.DistributedSampler(
dataset_train, num_replicas=num_tasks, rank=global_rank, shuffle=True
)
print("Sampler_train = %s" % str(sampler_train))
else:
sampler_train = torch.utils.data.RandomSampler(dataset_train)
if global_rank == 0 and args.log_dir is not None:
os.makedirs(args.log_dir, exist_ok=True)
log_writer = SummaryWriter(log_dir=args.log_dir)
else:
log_writer = None
data_loader_train = torch.utils.data.DataLoader(
dataset_train, sampler=sampler_train,
batch_size=args.batch_size,
num_workers=args.num_workers,
pin_memory=args.pin_mem,
drop_last=True,
)
# build model
model = models_sim.__dict__[args.model](norm_pix_loss=args.norm_pix_loss, args=args)
model.to(device)
model_without_ddp = model
eff_batch_size = args.batch_size * args.accum_iter * misc.get_world_size()
if args.lr is None: # only base_lr is specified
args.lr = args.blr * eff_batch_size / 256
print("base lr: %.2e" % (args.lr * 256 / eff_batch_size))
print("actual lr: %.2e" % args.lr)
print("accumulate grad iterations: %d" % args.accum_iter)
print("effective batch size: %d" % eff_batch_size)
if args.distributed:
model = torch.nn.SyncBatchNorm.convert_sync_batchnorm(model)
model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.gpu], find_unused_parameters=True)
model_without_ddp = model.module
print("Model = %s" % str(model_without_ddp))
# build optimizer
# following timm: set wd as 0 for bias and norm layers
param_groups = param_groups_weight_decay(model_without_ddp, args.weight_decay)
optimizer = torch.optim.AdamW(param_groups, lr=args.lr, betas=(0.9, args.beta2))
print(optimizer)
loss_scaler = NativeScaler(enabled=(not args.fp32), growth_interval=args.amp_growth_interval)
misc.auto_load_model(args=args, model_without_ddp=model_without_ddp, optimizer=optimizer,
loss_scaler=loss_scaler)
# start training
print(f"Start training for {args.epochs} epochs")
start_time = time.time()
for epoch in range(args.start_epoch, args.epochs):
epoch_start_time = time.time()
if args.distributed:
data_loader_train.sampler.set_epoch(epoch)
train_stats = train_one_epoch(
model, data_loader_train,
optimizer, device, epoch, loss_scaler,
log_writer=log_writer,
args=args
)
dist.barrier()
# save ckpt
if args.output_dir and ((epoch+1) % args.save_freq == 0 or epoch + 1 == args.epochs):
misc.save_model(
args=args, model=model, model_without_ddp=model_without_ddp, optimizer=optimizer,
loss_scaler=loss_scaler, epoch=epoch)
if (epoch+1) % args.save_latest_freq == 0:
misc.save_model(
args=args, model=model, model_without_ddp=model_without_ddp, optimizer=optimizer,
loss_scaler=loss_scaler, epoch=epoch, latest=True)
# log information
log_stats = {**{f'train_{k}': v for k, v in train_stats.items()},
'epoch': epoch,}
if args.output_dir and misc.is_main_process():
if log_writer is not None:
log_writer.flush()
with open(os.path.join(args.output_dir, "log.txt"), mode="a", encoding="utf-8") as f:
f.write(json.dumps(log_stats) + "\n")
if misc.is_main_process():
epoch_total_time = time.time() - epoch_start_time
now = datetime.datetime.today()
eta = now + datetime.timedelta(seconds=(args.epochs-epoch-1)*int(epoch_total_time))
next_50_ep = ((epoch + 1) // 50 + 1) * 50
eta_to_next_50 =now + datetime.timedelta(seconds=(next_50_ep - epoch - 1) * int(epoch_total_time))
print(f"ETA to {args.epochs:4d}ep:\t{str(eta)}")
print(f"ETA to {next_50_ep:4d}ep:\t{str(eta_to_next_50)}")
dist.barrier()
total_time = time.time() - start_time
total_time_str = str(datetime.timedelta(seconds=int(total_time)))
print('Training time {}'.format(total_time_str))
if __name__ == '__main__':
args = get_args_parser()
args = args.parse_args()
if args.output_dir:
Path(args.output_dir).mkdir(parents=True, exist_ok=True)
main(args)