-
Notifications
You must be signed in to change notification settings - Fork 3
/
main.py
172 lines (152 loc) · 7.7 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
import os
import argparse
import datetime
import random
import json
import time
from pathlib import Path
from tensorboardX import SummaryWriter
from copy import deepcopy
from inference import infer
from visualize import visual
import numpy as np
import torch
torch.multiprocessing.set_sharing_strategy('file_system')
from torch.utils.data import DataLoader, DistributedSampler
import data
#from mmdet import datasets
import util.misc as utils
from data import build
from engine import evaluate, train_one_epoch
from models import build_model
########################################
from args import add_management_args, add_experiment_args, add_bayes_args
import logging
def main(args):
# setup logger
utils.setup_logger('base', args.output_dir, 'train', level=logging.INFO,
screen=True, tofile=True)
logger = logging.getLogger('base')
utils.init_distributed_mode(args)
writer = SummaryWriter(log_dir=args.output_dir + '/summary')
logger.info(args)
device = torch.device(args.device)
# fix the seed for reproducibility
seed = args.seed + utils.get_rank()
torch.manual_seed(seed)
np.random.seed(seed)
random.seed(seed)
model, criterion, postprocessors, visualizer = build_model(args)
model.to(device)
logger.info(model)
model_without_ddp = model
if args.distributed:
model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.gpu])
model_without_ddp = model.module
n_parameters = sum(p.numel() for p in model.parameters() if p.requires_grad)
logger.info('number of params:{}'.format( n_parameters))
param_dicts = [{"params": [p for n, p in model_without_ddp.named_parameters() if p.requires_grad]}]
optimizer = torch.optim.Adam(param_dicts, lr=args.lr,
weight_decay=args.weight_decay)
lr_scheduler = torch.optim.lr_scheduler.StepLR(optimizer, args.lr_drop)
logger.info('Building training dataset...')
dataset_train_dict = build(image_set='train', args=args)
num_train = [len(v) for v in dataset_train_dict.values()]
logger.info('Number of training images: {}'.format(sum(num_train)))
logger.info('Building validation dataset...')
dataset_val_dict = build(image_set='val', args=args)
num_val = [len(v) for v in dataset_val_dict.values()]
logger.info('Number of validation images: {}'.format(sum(num_val)))
if args.distributed:
sampler_train_dict = {k : DistributedSampler(v) for k, v in dataset_train_dict.items()}
sampler_val_dict = {k: DistributedSampler(v, shuffle=False) for k, v in dataset_val_dict.items()}
else:
sampler_train_dict = {k : torch.utils.data.RandomSampler(v) for k, v in dataset_train_dict.items()}
sampler_val_dict = {k : torch.utils.data.SequentialSampler(v) for k, v in dataset_val_dict.items()}
batch_sampler_train = {
k : torch.utils.data.BatchSampler(v, args.batch_size, drop_last=True) for k, v in sampler_train_dict.items()
}
dataloader_train_dict = {
k : DataLoader(v1, batch_sampler=v2, collate_fn=utils.collate_fn, num_workers=args.num_workers)
for (k, v1), v2 in zip(dataset_train_dict.items(), batch_sampler_train.values())
}
dataloader_val_dict = {
k : DataLoader(v1, args.batch_size, sampler=v2, drop_last=False, collate_fn=utils.collate_fn, num_workers=args.num_workers)
for (k, v1), v2 in zip(dataset_val_dict.items(), sampler_val_dict.values())
}
if args.frozen_weights is not None:
checkpoint = torch.load(args.frozen_weights, map_location='cpu')
model_without_ddp.whst.load_state_dict(checkpoint['model'])
output_dir = Path(args.output_dir)
if args.resume:
checkpoint = torch.load(args.resume, map_location='cpu')
model_without_ddp.load_state_dict(checkpoint['model'])
if not args.eval and 'optimizer' in checkpoint and 'epoch' in checkpoint:
#if not args.eval and 'optimizer' in checkpoint and 'lr_scheduler' in checkpoint and 'epoch' in checkpoint:
optimizer.load_state_dict(checkpoint['optimizer'])
lr_scheduler.load_state_dict(checkpoint['lr_scheduler'])
args.start_epoch = checkpoint['epoch'] + 1
if args.eval:
#test_stats = evaluate(model, criterion, postprocessors, dataloader_val_dict, device, args.output_dir, visualizer, 0, writer)
test_df = infer(model, args.model, args.dataset, args.sequence, dataloader_val_dict, args.output_dir, device)
#test_df = visual(model, args.model, dataloader_val_dict, args.output_dir, device)
else:
logger.info("Start training")
best_dic = None
best_dice = None
start_time = time.time()
min_delta=0.1
patience = 40
hist_loss = 1e16
patience_counter = 0
for epoch in range(args.start_epoch, args.epochs):
# optimizer.param_groups[0]['lr'] = clr.cyclic_learning_rate(epoch, mode='exp_range', gamma=1)
train_stats = train_one_epoch(model, criterion, dataloader_train_dict, optimizer, device, epoch,args)
test_stats = evaluate(model, criterion, postprocessors, dataloader_val_dict, device, args.output_dir, visualizer, epoch, writer)
#test_df = infer(model, criterion, dataloader_val_dict, device)
dice_score = test_stats["loss_AvgDice"]
logger.info('dice score:{}'.format(dice_score))
if args.output_dir:
checkpoint_paths = [output_dir / 'checkpoint.pth']
if best_dice == None or dice_score > best_dice:
best_dice = dice_score
best_dic = deepcopy(test_stats)
logger.info("Update best model!")
checkpoint_paths.append(output_dir / 'best_checkpoint.pth')
if dice_score > 0.81:
logger.info("Update high dice score model!")
file_name = str(dice_score)[0:6]+'new_checkpoint.pth'
checkpoint_paths.append(output_dir / file_name)
# extra checkpoint before LR drop and every 100 epochs
if (epoch + 1) % args.lr_drop == 0 or (epoch + 1) % 100 == 0:
checkpoint_paths.append(output_dir / f'checkpoint{epoch:04}.pth')
for checkpoint_path in checkpoint_paths:
utils.save_on_master({
'model': model_without_ddp.state_dict(),
'optimizer': optimizer.state_dict(),
'lr_scheduler': lr_scheduler.state_dict(),
'epoch': epoch,
'args': args,
}, checkpoint_path)
log_stats = {**{f'train_{k}': v for k, v in train_stats.items()},
**{f'test_{k}': v for k, v in test_stats.items()},
'epoch': epoch,
'n_parameters': n_parameters}
if args.output_dir and utils.is_main_process():
with (output_dir / "log.txt").open("a") 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)))
logger.info('Training time {}'.format(total_time_str))
if __name__ == '__main__':
parser = argparse.ArgumentParser('BayeSeg training and evaluation', allow_abbrev=False)
add_experiment_args(parser)
add_management_args(parser)
add_bayes_args(parser)
args = parser.parse_args()
if args.output_dir:
Path(args.output_dir).mkdir(parents=True, exist_ok=True)
os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
os.environ["CUDA_VISIBLE_DEVICES"] = "{}".format(args.GPU_ids)
print(torch.cuda.is_available())
main(args)