-
Notifications
You must be signed in to change notification settings - Fork 30
/
trainer.py
203 lines (163 loc) · 7.18 KB
/
trainer.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
import os
import utility
import torch
from decimal import Decimal
class Trainer():
def __init__(self, args, loader, my_model, my_loss, ckp):
self.args = args
self.scale = args.scale
self.ckp = ckp
self.loader_train = loader.loader_train
self.loader_test = loader.loader_test
self.model = my_model
self.loss = my_loss
self.optimizer = utility.make_optimizer(args, self.model)
self.scheduler = utility.make_scheduler(args, self.optimizer)
if self.args.load != '.':
self.optimizer.load_state_dict(
torch.load(os.path.join(ckp.dir, 'optimizer.pt'))
)
for _ in range(len(ckp.log)): self.scheduler.step()
self.error_last = 1e8
def train(self):
self.scheduler.step()
self.loss.step()
epoch = self.scheduler.last_epoch + 1
# lr schedule
lr = self.args.lr * (2 ** -(epoch // 200))
for param_group in self.optimizer.param_groups:
param_group['lr'] = lr
self.ckp.write_log('[Epoch {}]\tLearning rate: {:.2e}'.format(epoch, Decimal(lr)))
self.loss.start_log()
self.model.train()
timer_data, timer_model = utility.timer(), utility.timer()
for batch, (lr, hr, _, idx_scale) in enumerate(self.loader_train):
lr, hr = self.prepare(lr, hr)
_, _, outH, outW = hr.size()
# update tau for gumbel softmax
tau = max(1 - (epoch - 1) / 500, 0.4)
for m in self.model.modules():
if hasattr(m, '_set_tau'):
m._set_tau(tau)
# inference
self.optimizer.zero_grad()
sr, sparsity = self.model(lr, idx_scale)
# losses
loss_SR = self.loss(sr, hr)
loss_sparsity = sparsity.mean()
lambda0 = 0.1
lambda_sparsity = min((epoch - 1) / 50, 1) * lambda0
loss = loss_SR + lambda_sparsity * loss_sparsity
# backpropagation
if loss.item() < self.args.skip_threshold * self.error_last:
loss.backward()
self.optimizer.step()
else:
print('Skip this batch {}! (Loss: {})'.format(
batch + 1, loss.item()
))
timer_model.hold()
if (batch + 1) % self.args.print_every == 0:
self.ckp.write_log('[{}/{}]\t{}\t[Sparsity:{:.3f}]\t{:.1f}+{:.1f}s'.format(
(batch + 1) * self.args.batch_size,
len(self.loader_train.dataset),
self.loss.display_loss(batch),
float(sparsity.round().mean().data.cpu()),
timer_model.release(),
timer_data.release()))
timer_data.tic()
self.loss.end_log(len(self.loader_train))
self.error_last = self.loss.log[-1, -1]
target = self.model
torch.save(
target.state_dict(),
os.path.join(self.ckp.dir, 'model', 'model_{}.pt'.format(epoch))
)
def test(self):
epoch = self.scheduler.last_epoch + 1
self.ckp.write_log('\nEvaluation:')
self.ckp.add_log(torch.zeros(1, len(self.scale)))
self.model.eval()
timer_test = utility.timer()
with torch.no_grad():
for idx_scale, scale in enumerate(self.scale):
self.loader_test.dataset.set_scale(idx_scale)
eval_acc = 0
eval_acc_ssim = 0
# Kernel split
# Note that, this part of code does not need to be executed at each run.
# After training, one can run this part of code once and save the splitted kernels.
for m in self.model.modules():
if hasattr(m, '_prepare'):
m._prepare()
for idx_img, (lr, hr, filename, _) in enumerate(self.loader_test):
no_eval = (hr.nelement() == 1)
if not no_eval:
lr, hr = self.prepare(lr, hr)
else:
lr, = self.prepare(lr)
lr, hr = self.crop_border(lr, hr, scale)
sr = self.model(lr, idx_scale)
# run a second time to record inference time
for idx_img, (lr, hr, filename, _) in enumerate(self.loader_test):
filename = filename[0]
no_eval = (hr.nelement() == 1)
if not no_eval:
lr, hr = self.prepare(lr, hr)
else:
lr, = self.prepare(lr)
lr, hr = self.crop_border(lr, hr, scale)
timer_test.tic()
sr = self.model(lr, idx_scale)
timer_test.hold()
sr = utility.quantize(sr, self.args.rgb_range)
hr = utility.quantize(hr, self.args.rgb_range)
save_list = [sr]
if not no_eval:
eval_acc += utility.calc_psnr(
sr, hr, scale, self.args.rgb_range,
benchmark=self.loader_test.dataset.benchmark
)
eval_acc_ssim += utility.calc_ssim(
sr, hr, scale,
benchmark=self.loader_test.dataset.benchmark
)
if self.args.save_results:
self.ckp.save_results(filename, save_list, scale)
self.ckp.log[-1, idx_scale] = eval_acc / len(self.loader_test)
best = self.ckp.log.max(0)
self.ckp.write_log(
'[{} x{}] {:.4f}s\tPSNR: {:.3f} SSIM: {:.4f} (Best: {:.3f} @epoch {})'.format(
self.args.data_test,
scale,
timer_test.release()/len(self.loader_test),
eval_acc / len(self.loader_test),
eval_acc_ssim / len(self.loader_test),
best[0][idx_scale],
best[1][idx_scale] + 1
))
# self.ckp.write_log(
# 'Total time: {:.2f}s\n'.format(timer_test.release()/len(self.loader_test)), refresh=True
# )
if not self.args.test_only:
self.ckp.save(self, epoch, is_best=(best[1][0] + 1 == epoch))
def prepare(self, *args):
device = torch.device('cpu' if self.args.cpu else 'cuda')
def _prepare(tensor):
if self.args.precision == 'half': tensor = tensor.half()
return tensor.to(device)
return [_prepare(a) for a in args]
def crop_border(self, img_lr, img_hr, scale):
N, C, H_lr, W_lr = img_lr.size()
H = H_lr //2 *2
W = W_lr //2 *2
img_lr = img_lr[:, :, :H, :W]
img_hr = img_hr[:, :, :round(scale * H), :round(scale * W)]
return img_lr, img_hr
def terminate(self):
if self.args.test_only:
self.test()
return True
else:
epoch = self.scheduler.last_epoch + 1
return epoch >= self.args.epochs