-
Notifications
You must be signed in to change notification settings - Fork 2
/
mae_pre-training.py
250 lines (199 loc) · 10.5 KB
/
mae_pre-training.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
import argparse
import torch
import utils2.misc as misc
import torch.backends.cudnn as cudnn
from torch.utils.tensorboard import SummaryWriter
from pathlib import Path
from torch.utils.data import DistributedSampler
from data import cityscapes
from data.loader_tools import get_joint_transformations, get_standard_transformations
from model.mae_model import mae_vit
from utils2.misc import NativeScalerWithGradNormCount as NativeScaler
import timm.optim.optim_factory as optim_factory
import os
import json
import datetime
import math
import sys
from typing import Iterable
import utils2.lr_sched as lr_sched
from utils2.pos_embed import interpolate_pos_embed
import time
def get_args_parser():
parser = argparse.ArgumentParser('MAE pre-training', add_help=False)
parser.add_argument('--batch_size', default=4, type=int,
help='Batch size per GPU (effective batch size is batch_size * accum_iter * # gpus')
parser.add_argument('--num_epochs', default=100, type=int)
parser.add_argument('--accum_iter', default=1, type=int,
help='Accumulate gradient iterations (for increasing the effective batch size under memory constraints)')
parser.add_argument('--use_pre', default=True, type=bool)
# Model parameters
parser.add_argument("--crop_size", type=int, default=[640, 640],
help="crop size for training and inference slice.")
parser.add_argument("--stride_rate", type=float, default=0.5, help="stride ratio.")
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)
# VIT settings
parser.add_argument("--patch_size", type=int, default=16, help="define the patch size.")
parser.add_argument("--encoder_embed_dim", type=int, default=768, help="dimension for encoder.")
parser.add_argument("--decoder_embed_dim", type=int, default=512, help="dimension for decoder.")
parser.add_argument("--encoder_depth", type=int, default=12, help="depth for encoder.")
parser.add_argument("--decoder_depth", type=int, default=8, help="depth for decoder.")
parser.add_argument("--encoder_num_head", type=int, default=12, help="head number for encoder.")
parser.add_argument("--decoder_num_head", type=int, default=16, help="head number for decoder.")
# 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=0.0005, 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=20, metavar='N',
help='epochs to warmup LR')
# Dataset parameters
parser.add_argument('--root', default="/home/wangbowen/DATA/cityscapes", type=str,
help='dataset path')
parser.add_argument('--output_dir', default='save_model',
help='path where to save, empty for no saving')
parser.add_argument('--log_dir', default='save_model',
help='path where to tensorboard log')
parser.add_argument('--device', default='cuda',
help='device to use for training / testing')
parser.add_argument('--num_workers', default=4, type=int)
# 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')
return parser
def main():
# distribution
misc.init_distributed_mode(args)
device = torch.device(args.device)
cudnn.benchmark = True
model = mae_vit(args)
model.to(device)
model_without_ddp = model
# print("Model = %s" % str(model_without_ddp))
joint_transformations = get_joint_transformations(args)
standard_transformations = get_standard_transformations()
train_set = cityscapes.CityScapes(args, 'fine', 'train', joint_transform=joint_transformations,
standard_transform=standard_transformations)
if args.use_pre:
# use the pre-trained parameter from mae paper
checkpoint = torch.load("save_model/mae_pretrain_vit_base.pth", map_location='cpu')
checkpoint_model = checkpoint['model']
interpolate_pos_embed(model, checkpoint_model)
model_without_ddp.load_state_dict(checkpoint_model, strict=False)
print("load pre-trained model")
if args.distributed:
num_tasks = misc.get_world_size()
global_rank = misc.get_rank()
sampler_train = DistributedSampler(train_set)
else:
sampler_train = torch.utils.data.RandomSampler(train_set)
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
eff_batch_size = args.batch_size * args.accum_iter * misc.get_world_size()
batch_sampler_train = torch.utils.data.BatchSampler(sampler_train, args.batch_size, drop_last=True)
train_loader = torch.utils.data.DataLoader(train_set, batch_sampler=batch_sampler_train, num_workers=args.num_workers, pin_memory=True)
if args.lr is None: # only base_lr is specified
args.lr = args.blr * eff_batch_size / 16
print("base lr: %.2e" % (args.lr * 16 / 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.parallel.DistributedDataParallel(model, device_ids=[args.gpu], find_unused_parameters=True)
model_without_ddp = model.module
# following timm: set wd as 0 for bias and norm layers
param_groups = optim_factory.add_weight_decay(model_without_ddp, args.weight_decay)
optimizer = torch.optim.AdamW(param_groups, lr=args.lr, betas=(0.9, 0.95))
print(optimizer)
loss_scaler = NativeScaler()
print(f"Start training for {args.num_epochs} epochs")
start_time = time.time()
for epoch in range(args.num_epochs):
if args.distributed:
sampler_train.set_epoch(epoch)
if args.output_dir and ((epoch + 1) % 10 == 0 or epoch + 1 == args.num_epochs):
misc.save_model(args=args, model=model, model_without_ddp=model_without_ddp, optimizer=optimizer,
loss_scaler=loss_scaler, epoch=epoch)
train_stats = train_one_epoch(
model, train_loader,
optimizer, device, epoch, loss_scaler,
log_writer=log_writer,
args=args
)
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")
total_time = time.time() - start_time
total_time_str = str(datetime.timedelta(seconds=int(total_time)))
print('Training time {}'.format(total_time_str))
def train_one_epoch(model: torch.nn.Module,
data_loader: Iterable, optimizer: torch.optim.Optimizer,
device: torch.device, epoch: int, loss_scaler,
log_writer=None,
args=None):
model.train(True)
metric_logger = misc.MetricLogger(delimiter=" ")
metric_logger.add_meter('lr', misc.SmoothedValue(window_size=1, fmt='{value:.6f}'))
header = 'Epoch: [{}]'.format(epoch)
print_freq = 20
accum_iter = args.accum_iter
optimizer.zero_grad()
if log_writer is not None:
print('log_dir: {}'.format(log_writer.log_dir))
for data_iter_step, samples in enumerate(metric_logger.log_every(data_loader, print_freq, header)):
# we use a per iteration (instead of per epoch) lr scheduler
if data_iter_step % accum_iter == 0:
lr_sched.adjust_learning_rate(optimizer, data_iter_step / len(data_loader) + epoch, args)
samples = samples["images"].to(device, non_blocking=True)
with torch.cuda.amp.autocast():
loss, _, _ = model(samples, mask_ratio=args.mask_ratio)
loss_value = loss.item()
if not math.isfinite(loss_value):
print("Loss is {}, stopping training".format(loss_value))
sys.exit(2)
loss = loss / accum_iter
loss_scaler(loss, optimizer, parameters=model.parameters(),
update_grad=(data_iter_step + 1) % accum_iter == 0)
if (data_iter_step + 1) % accum_iter == 0:
optimizer.zero_grad()
torch.cuda.synchronize()
metric_logger.update(loss=loss_value)
lr = optimizer.param_groups[0]["lr"]
metric_logger.update(lr=lr)
# loss_value_reduce = misc.all_reduce_mean(loss_value)
# if log_writer is not None and (data_iter_step + 1) % accum_iter == 0:
# """ We use epoch_1000x as the x-axis in tensorboard.
# This calibrates different curves when batch size changes.
# """
# epoch_1000x = int((data_iter_step / len(data_loader) + epoch) * 1000)
# log_writer.add_scalar('train_loss', loss_value_reduce, epoch_1000x)
# log_writer.add_scalar('lr', lr, epoch_1000x)
# gather the stats from all processes
metric_logger.synchronize_between_processes()
print("Averaged stats:", metric_logger)
return {k: meter.global_avg for k, meter in metric_logger.meters.items()}
if __name__ == '__main__':
parser = argparse.ArgumentParser('model mae pre-training', parents=[get_args_parser()])
args = parser.parse_args()
if args.output_dir:
Path(args.output_dir).mkdir(parents=True, exist_ok=True)
main()