-
Notifications
You must be signed in to change notification settings - Fork 3
/
train.py
140 lines (124 loc) · 6.21 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
import argparse
import logging
import os
import random
import numpy as np
import torch
import torch.backends.cudnn as cudnn
from networks.WaveFormer import Model
from networks.WaveFormerCompact import Model as ModelCompact
from trainer import trainer_synapse
import warnings
warnings.filterwarnings('ignore')
parser = argparse.ArgumentParser()
parser.add_argument('--root_path', type=str,
default='/images/PublicDataset/Transunet_synaps/project_TransUNet/data/Synapse/train_npz', help='root dir for data')
parser.add_argument('--test_path', type=str,
default='/images/PublicDataset/Transunet_synaps/project_TransUNet/data/Synapse/test_vol_h5', help='root dir for data')
parser.add_argument('--dataset', type=str,
default='Synapse', help='experiment_name')
parser.add_argument('--dstr_fast', type=bool,
default=False, help='SynapseDatasetFast: will load all data into RAM')
parser.add_argument('--en_lnum', type=int,
default=3, help='en_lnum: Laplacian layers (Pyramid) for the encoder')
parser.add_argument('--br_lnum', type=int,
default=3, help='br_lnum: Laplacian layers (Pyramid) for the bridge')
parser.add_argument('--de_lnum', type=int,
default=3, help='de_lnum: Laplacian layers (Pyramid) for the decoder')
parser.add_argument('--compact', type=bool,
default=False, help='compact with 3 blocks insted of 4 blocks')
parser.add_argument('--continue_tr', type=bool,
default=False, help='continue training from the last saved epoch')
parser.add_argument('--optimizer', type=str,
default='SGD', help='optimizer: [SGD, AdamW])')
parser.add_argument('--dice_loss_weight', type=float,
default=0.6, help="You need to determine <x> (default=0.6): => [loss = (1-x)*ce_loss + x*dice_loss]")
parser.add_argument('--list_dir', type=str,
default='./lists/lists_Synapse', help='list dir')
parser.add_argument('--num_classes', type=int,
default=9, help='output channel of network')
parser.add_argument('--output_dir', type=str,
default='./model_out',help='output dir')
parser.add_argument('--max_iterations', type=int,
default=90000, help='maximum epoch number to train')
parser.add_argument('--max_epochs', type=int,
default=400, help='maximum epoch number to train')
parser.add_argument('--batch_size', type=int,
default=20, help='batch_size per gpu')
parser.add_argument('--num_workers', type=int,
default=4, help='num_workers')
parser.add_argument('--eval_interval', type=int,
default=20, help='eval_interval')
parser.add_argument('--model_name', type=str,
default='WaveFormer', help='model_name')
parser.add_argument('--n_gpu', type=int, default=1, help='total gpu')
parser.add_argument('--bridge_layers', type=int, default=1, help='number of bridge layers')
parser.add_argument('--deterministic', type=int, default=1,
help='whether use deterministic training')
parser.add_argument('--base_lr', type=float, default=0.05,
help='segmentation network learning rate')
parser.add_argument('--img_size', type=int,
default=224, help='input patch size of network input')
parser.add_argument('--z_spacing', type=int,
default=1, help='z_spacing')
parser.add_argument('--seed', type=int,
default=1234, help='random seed')
# parser.add_argument('--cfg', type=str, required=True, metavar="FILE", help='path to config file', )
parser.add_argument(
"--opts",
help="Modify config options by adding 'KEY VALUE' pairs. ",
default=None,
nargs='+',
)
parser.add_argument('--zip', action='store_true', help='use zipped dataset instead of folder dataset')
parser.add_argument('--cache-mode', type=str, default='part', choices=['no', 'full', 'part'],
help='no: no cache, '
'full: cache all data, '
'part: sharding the dataset into nonoverlapping pieces and only cache one piece')
parser.add_argument('--resume', help='resume from checkpoint')
parser.add_argument('--accumulation-steps', type=int, help="gradient accumulation steps")
parser.add_argument('--use-checkpoint', action='store_true',
help="whether to use gradient checkpointing to save memory")
parser.add_argument('--amp-opt-level', type=str, default='O1', choices=['O0', 'O1', 'O2'],
help='mixed precision opt level, if O0, no amp is used')
parser.add_argument('--tag', help='tag of experiment')
parser.add_argument('--eval', action='store_true', help='Perform evaluation only')
parser.add_argument('--throughput', action='store_true', help='Test throughput only')
args = parser.parse_args()
args.output_dir = args.output_dir + f'/{args.model_name}'
os.makedirs(args.output_dir, exist_ok=True)
if __name__ == "__main__":
os.environ["CUDA_VISIBLE_DEVICES"] = '0'
if not args.deterministic:
cudnn.benchmark = True
cudnn.deterministic = False
else:
cudnn.benchmark = False
cudnn.deterministic = True
# random.seed(args.seed)
# np.random.seed(args.seed)
# torch.manual_seed(args.seed)
# torch.cuda.manual_seed(args.seed)
dataset_name = args.dataset
if args.batch_size != 24 and args.batch_size % 5 == 0:
args.base_lr *= args.batch_size / 24
if not os.path.exists(args.output_dir):
os.makedirs(args.output_dir)
if args.compact:
net = ModelCompact(
num_classes=args.num_classes,
bridge_layers=args.bridge_layers,
encoder_L_num=args.en_lnum,
bridge_L_num=args.br_lnum,
decoder_L_num=args.de_lnum
).cuda(0)
else:
net = Model(
num_classes=args.num_classes,
bridge_layers=args.bridge_layers,
encoder_L_num=args.en_lnum,
bridge_L_num=args.br_lnum,
decoder_L_num=args.de_lnum
).cuda(0)
trainer = {'Synapse': trainer_synapse,}
trainer[dataset_name](args, net, args.output_dir)