-
Notifications
You must be signed in to change notification settings - Fork 0
/
.trainer_18.11.2018.py
215 lines (175 loc) · 7.3 KB
/
.trainer_18.11.2018.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
import math
from decimal import Decimal
import datetime
import utility
import torch
from torch.autograd import Variable
class Trainer():
def __init__(self, loader, ckp, args):
self.args = args
self.scale = args.scale
self.loader_train, self.loader_test = loader
self.model, self.loss, self.optimizer, self.scheduler = ckp.load()
self.ckp = ckp
self.log_training = 0
self.log_test = 0
self.error_last = 1e8
self.tmp_loss = 1000 # Used to save model best
self.test_time = [0,0]
def _scale_change(self, idx_scale, testset=None):
if len(self.scale) > 1:
if self.args.n_GPUs == 1:
self.model.set_scale(idx_scale)
else:
self.model.module.set_scale(idx_scale)
if testset is not None:
testset.dataset.set_scale(idx_scale)
def train(self):
self.scheduler.step()
epoch = self.scheduler.last_epoch + 1
lr = self.scheduler.get_lr()[0]
self.ckp.write_log(
'\n[Epoch {}]\tLearning rate: {:.2e}'.format(epoch, Decimal(lr)))
self.ckp.add_log(torch.zeros(1, len(self.loss)))
self.model.train()
#timer_train = utility.timer() # My code 22.4.2018
timer_data, timer_model = utility.timer(), utility.timer()
for batch, ((lr, labels), hr, idx_scale) in enumerate(self.loader_train):
lr, hr = self.prepare([lr, hr])
self._scale_change(idx_scale)
timer_data.hold()
timer_model.tic()
self.optimizer.zero_grad()
sr = self.model(lr)
loss = self._calc_loss(sr, hr, labels)
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()
self.tmp_loss = loss.item()
if (batch + 1) % self.args.print_every == 0:
self.ckp.write_log('[{}/{}]\t{}\t{:.1f}+{:.1f}s'.format(
(batch + 1) * self.args.batch_size,
len(self.loader_train.dataset),
self._display_loss(batch),
timer_model.release(),
timer_data.release()))
timer_data.tic()
self.ckp.log_training[-1, :] /= len(self.loader_train)
self.error_last = self.ckp.log_training[-1, :][0]
# My code 22.4.2018
#self.ckp.write_log(
# 'Time: {:.2f}s\n'.format(timer_train.toc()), refresh=True
#)
# End my code
def test(self, test_only=False, starttime=0):
epoch = self.scheduler.last_epoch + 1
self.ckp.write_log('Evaluation:')
self.ckp.add_log(torch.zeros(1, len(self.scale)), False)
self.model.eval()
# We can use custom forward function
def _test_forward(x, scale):
if self.args.self_ensemble:
return utility.x8_forward(x, self.model, self.args.precision)
elif self.args.chop_forward:
return utility.chop_forward(x, self.model, scale)
else:
return self.model(x)
set_name = self.args.data_test
for idx_scale, scale in enumerate(self.scale):
eval_acc = 0
self._scale_change(idx_scale, self.loader_test)
for idx_img, ((lr,_), hr, _) in enumerate(self.loader_test):
no_eval = isinstance(hr[0], torch._six.string_classes)
if no_eval:
lr = self.prepare([lr])[0]
filename = hr[0]
else:
lr, hr = self.prepare([lr, hr])
filename = idx_img + 1
rgb_range = self.args.rgb_range
timer_test = utility.timer()
sr = _test_forward(lr, scale)
self.test_time[1] += timer_test.toc()
self.test_time[0] += 1
sr = utility.quantize(sr, rgb_range)
if no_eval:
save_list = [sr]
else:
eval_acc += utility.calc_PSNR(
sr,
hr.div(rgb_range),
set_name,
scale
)
save_list = [sr, lr.div(rgb_range), hr.div(rgb_range)]
if self.args.save_results:
self.ckp.save_results(filename, save_list, scale)
self.ckp.log_test[-1, idx_scale] = eval_acc / len(self.loader_test)
best = self.ckp.log_test.max(0)
performance = 'PSNR: {:.3f}'.format(
self.ckp.log_test[-1, idx_scale]
)
self.ckp.write_log(
'[{} x{}]\t{} (Best: {:.3f} from epoch {})'.format(
set_name,
scale,
performance,
best[0][idx_scale],
best[1][idx_scale] + 1
)
)
is_best = (best[1][0] + 1 == epoch)
if test_only:
self.ckp.write_log(
'Total time: {:.3f}s\r\nAvg. time: {:.3f}s\n'.format(self.test_time[1],self.test_time[1]/self.test_time[0]), refresh=True
)
elif starttime != 0:
now = datetime.datetime.now()
elapsed = now - starttime
est = now + (elapsed / epoch) * (self.args.epochs - epoch)
self.ckp.write_log("Elapsed: {}\n".format(str(elapsed)))
print('Will finish: {}\n'.format(est.strftime('%d-%m-%Y-%H:%M:%S')))
self.ckp.save(self, epoch, is_best=is_best)
def prepare(self, l, volatile=False):
device = torch.device('cpu' if self.args.no_cuda else 'cuda')
def _prepare(tensor):
if self.args.precision == 'half': tensor = tensor.half()
return tensor.to(device)
return [_prepare(_l) for _l in l]
def _calc_loss(self, sr, hr, labels):
loss_list = []
for i, l in enumerate(self.loss):
if isinstance(sr, list):
if isinstance(hr, list):
loss = l['function'](sr[i], hr[i])
else:
loss = l['function'](sr[i], hr)
else:
if l['type'] == 'ML1':
loss = l['function'](sr, hr, labels)
else:
loss = l['function'](sr, hr)
loss_list.append(l['weight'] * loss)
self.ckp.log_training[-1, i] += loss.item()
loss_total = sum(loss_list)
if len(self.loss) > 1:
self.ckp.log_training[-1, -1] += loss_total.data[0]
return loss_total
def _display_loss(self, batch):
n_samples = batch + 1
log = [
'[{}: {:.4f}] '.format(t['type'], l / n_samples) \
for l, t in zip(self.ckp.log_training[-1], self.loss)]
return ''.join(log)
def terminate(self):
if self.args.test_only:
self.test(self.args.test_only)
return True
else:
epoch = self.scheduler.last_epoch + 1
return epoch >= self.args.epochs