-
Notifications
You must be signed in to change notification settings - Fork 1
/
main.py
173 lines (146 loc) · 6.06 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
173
import argparse
import os
import torch
import random
import torch.backends.cudnn as cudnn
import torch.nn as nn
import torch.optim as optim
from torch.autograd import Variable
from torch.utils.data import DataLoader
from model.VDSR import Net
from model.Asym_VDSR import Asym_Net
from model.Back_Asym_VDSR import Back_Net
from model.Front_Asym_VDSR import Front_Net
from dataset_h5 import Read_dataset_h5
import numpy as np
import math
from eval_main import eval_model
# Training settings
parser = argparse.ArgumentParser(description="PyTorch VDSR")
parser.add_argument("--batchSize", type=int, default=64)
parser.add_argument("--nEpochs", type=int, default=80)
parser.add_argument("--lr", type=float, default=0.1)
parser.add_argument("--step", type=int, default=10)
parser.add_argument("--cuda", action="store_true")
parser.add_argument("--start-epoch", default=1, type=int)
parser.add_argument("--clip", type=float, default=0.4)
parser.add_argument("--threads", type=int, default=1)
parser.add_argument("--momentum", default=0.9, type=float)
parser.add_argument("--weight-decay", "--wd", default=1e-4, type=float)
parser.add_argument('--pretrained', default='', type=str)
parser.add_argument("--gpus", default="0", type=str)
parser.add_argument("--optimizer", default="SGD", type=str)
parser.add_argument("--net", default="VDSR", type=str)
def main():
global opt, model, optimizer # opt, model global
opt = parser.parse_args() # opt < parser
print(opt)
cuda = opt.cuda
if cuda:
print("=> use gpu id: '{}'".format(opt.gpus))
os.environ["CUDA_VISIBLE_DEVICES"] = opt.gpus # set gpu
if not torch.cuda.is_available():
raise Exception(
"No GPU found or Wrong gpu id, please run without --cuda")
opt.seed = random.randint(1, 10000)
print("Random Seed: ", opt.seed)
torch.manual_seed(opt.seed) # set seed
if cuda:
torch.cuda.manual_seed(opt.seed)
cudnn.benchmark = True # find optimal algorithms for hardware
print("===> Loading datasets")
train_set = Read_dataset_h5("data/train.h5")
training_data_loader = DataLoader(
dataset=train_set, num_workers=opt.threads, batch_size=opt.batchSize, shuffle=True) # read to DataLoader
print("===> Building model")
if opt.net == 'VDSR':
model = Net() # net
elif opt.net == 'Asym_VDSR':
model = Asym_Net() # net
elif opt.net == 'Back_Asym_VDSR':
model = Back_Net() # net
elif opt.net == 'Front_Asym_VDSR':
model = Front_Net() # net
criterion = nn.MSELoss(size_average=False) # set loss
epoch = opt.start_epoch
# optionally copy weights from a checkpoint
if opt.pretrained:
if os.path.isfile(opt.pretrained):
print("=> loading model '{}'".format(opt.pretrained))
checkpoint = torch.load(opt.pretrained)
model.load_state_dict(checkpoint['model_state_dict'])
epoch = checkpoint['epoch'] + 1 # load model
else:
print("=> no model found at '{}'".format(opt.pretrained))
print("===> Setting GPU")
if cuda:
model = model.cuda()
criterion = criterion.cuda() # set model&loss for use gpu
print("===> Setting Optimizer")
if opt.optimizer == 'SGD':
optimizer = optim.SGD(model.parameters(), lr=opt.lr,
momentum=opt.momentum, weight_decay=opt.weight_decay)
elif opt.optimizer == 'Adam':
if opt.lr == 0.1:
optimizer = optim.Adam(model.parameters())
else:
optimizer = optim.Adam(model.parameters(), lr=opt.lr)
else:
print("=> input 'SGD' or 'Adam', not '{}'".format(opt.optimizer))
if opt.pretrained:
if os.path.isfile(opt.pretrained):
optimizer.load_state_dict(checkpoint['optimizer_state_dict'])
print("=> start epoch '{}'".format(epoch))
print("===> Training")
for epoch_ in range(epoch, opt.nEpochs + 1):
train(training_data_loader, optimizer, model, criterion, epoch_)
def adjust_learning_rate(optimizer, epoch):
lr = opt.lr * (0.1 ** (epoch // opt.step))
return lr
def PSNR(loss):
psnr = 10 * np.log10(1 / (loss + 1e-10))
return psnr
def train(training_data_loader, optimizer, model, criterion, epoch):
if opt.optimizer == 'SGD':
lr = adjust_learning_rate(optimizer, epoch-1)
for param_group in optimizer.param_groups:
param_group["lr"] = lr
print("Epoch = {}, lr = {}".format(epoch, optimizer.param_groups[0]["lr"]))
model.train()
for iteration, batch in enumerate(training_data_loader, 1):
optimizer.zero_grad()
input, label = Variable(batch[0], requires_grad=False), Variable(
batch[1], requires_grad=False)
total_loss = 0
if opt.cuda:
input = input.cuda() / 255.
label = label.cuda() / 255.
output = model(input)
loss = criterion(output, label)
total_loss += loss.item()
loss.backward()
if opt.optimizer == 'SGD':
torch.nn.utils.clip_grad_norm_(model.parameters(), opt.clip)
optimizer.step()
if iteration % 100 == 0:
print("===> Epoch[{}]({}/{}): Loss: {:.10f}, PSNR : {:.10f}".format(
epoch, iteration, len(training_data_loader), loss.item(), PSNR(loss.item())))
epoch_loss = total_loss/len(training_data_loader)
psnr = PSNR(epoch_loss)
print("===> Epoch[{}]: loss : {:.10f} ,PSNR : {:.10f}".format(
epoch, epoch_loss, psnr))
save_checkpoint(model, epoch, optimizer)
eval_model(model, "Set5", opt.cuda, opt.gpus)
def save_checkpoint(model, epoch, optimizer):
model_out_path = "checkpoint/{}/{}_{}_epoch_{}.tar".format(
opt.net, opt.net, opt.optimizer, epoch)
if not os.path.exists("checkpoint/{}/".format(opt.net)):
os.makedirs("checkpoint/{}/".format(opt.net))
torch.save({
'epoch': epoch,
'model_state_dict': model.state_dict(),
'optimizer_state_dict': optimizer.state_dict()
}, model_out_path)
print("Checkpoint saved to {}".format(model_out_path))
if __name__ == "__main__":
main()