-
Notifications
You must be signed in to change notification settings - Fork 16
/
Copy pathraw_train.py
286 lines (230 loc) · 14.9 KB
/
raw_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
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
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
import argparse
import logging
import sys
import torch
import torchvision
import torch.nn as nn
import torch.nn.functional as F
from torch.utils.tensorboard import SummaryWriter
import data, models, utils
def main(args):
device = torch.device('cuda:0') if torch.cuda.is_available() else torch.device('cpu')
utils.setup_experiment(args)
utils.init_logging(args)
# Build data loaders, a model and an optimizer
model = models.build_model(args).to(device)
cpf = model.c # channels per frame
mid = args.n_frames // 2
model = nn.DataParallel(model)
print(model)
optimizer = torch.optim.Adam(model.parameters(), lr=args.lr)
scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, milestones=[2,3], gamma=0.5)
logging.info(f"Built a model consisting of {sum(p.numel() for p in model.parameters()):,} parameters")
if args.resume_training:
state_dict = utils.load_checkpoint(args, model, optimizer, scheduler)
global_step = state_dict['last_step']
start_epoch = int(state_dict['last_step']/(403200/state_dict['args'].batch_size))+1
else:
global_step = -1
start_epoch = 0
train_loader, valid_loader, test_loader = data.build_dataset(args.dataset, args.data_path, batch_size=args.batch_size, image_size=args.image_size, stride=args.stride, n_frames=args.n_frames, aug=args.aug)
# Track moving average of loss values
train_meters = {name: utils.RunningAverageMeter(0.98) for name in (["train_loss", "train_psnr", "train_ssim"])}
if args.loss == "loglike":
mean_meters = {name: utils.AverageMeter() for name in (["mean_psnr", "mean_ssim"])}
valid_meters = {name: utils.AverageMeter() for name in (["valid_loss", "valid_psnr", "valid_ssim"])}
writer = SummaryWriter(log_dir=args.experiment_dir) if not args.no_visual else None
for epoch in range(start_epoch, args.num_epochs):
if args.resume_training:
if epoch %10 == 0:
optimizer.param_groups[0]["lr"] /= 2
print('learning rate reduced by factor of 2')
train_bar = utils.ProgressBar(train_loader, epoch)
for meter in train_meters.values():
meter.reset()
if args.loss == "loglike":
for meter in mean_meters.values():
meter.reset()
for batch_id, (inputs, noisy_inputs) in enumerate(train_bar):
model.train()
global_step += 1
inputs = inputs.to(device)
noisy_inputs = noisy_inputs.to(device)
outputs, est_sigma = model(noisy_inputs)
noisy_frame = noisy_inputs[:, (mid*cpf):((mid+1)*cpf), :, :]
if args.blind_noise:
loss = utils.loss_function(outputs, noisy_frame, mode=args.loss, sigma=est_sigma, device=device)
else:
loss = utils.loss_function(outputs, noisy_frame, mode=args.loss, sigma=args.noise_std/255, device=device)
model.zero_grad()
loss.backward()
optimizer.step()
if args.loss == "loglike":
with torch.no_grad():
if args.blind_noise:
outputs, mean_image = utils.post_process(outputs, noisy_frame, model=args.model, sigma=est_sigma, device=device)
else:
outputs, mean_image = utils.post_process(outputs, noisy_frame, model=args.model, sigma=args.noise_std/255, device=device)
train_psnr = utils.psnr(inputs[:, (mid*cpf):((mid+1)*cpf), :, :], outputs, normalized=False, raw=True)
train_ssim = utils.ssim(inputs[:, (mid*cpf):((mid+1)*cpf), :, :], outputs, normalized=False, raw=True)
train_meters["train_loss"].update(loss.item())
train_meters["train_psnr"].update(train_psnr.item())
train_meters["train_ssim"].update(train_ssim.item())
if args.loss == "loglike":
mean_psnr = utils.psnr(inputs[:, (mid*cpf):((mid+1)*cpf), :, :], mean_image, normalized=False, raw=True)
mean_ssim = utils.ssim(inputs[:, (mid*cpf):((mid+1)*cpf), :, :], mean_image, normalized=False, raw=True)
mean_meters["mean_psnr"].update(mean_psnr.item())
mean_meters["mean_ssim"].update(mean_ssim.item())
if args.loss == "loglike":
train_bar.log(dict(**train_meters, **mean_meters, lr=optimizer.param_groups[0]["lr"]), verbose=True)
else:
train_bar.log(dict(**train_meters, lr=optimizer.param_groups[0]["lr"]), verbose=True)
if writer is not None and global_step % args.log_interval == 0:
writer.add_scalar("lr", optimizer.param_groups[0]["lr"], global_step)
writer.add_scalar("loss/train", loss.item(), global_step)
writer.add_scalar("psnr/train", train_psnr.item(), global_step)
writer.add_scalar("ssim/train", train_ssim.item(), global_step)
if args.loss == "loglike":
writer.add_scalar("psnr/mean", mean_psnr.item(), global_step)
writer.add_scalar("ssim/mean", mean_ssim.item(), global_step)
gradients = torch.cat([p.grad.view(-1) for p in model.parameters() if p.grad is not None], dim=0)
writer.add_histogram("gradients", gradients, global_step)
sys.stdout.flush()
if (batch_id+1) % 200 == 0:
if args.loss == "loglike":
logging.info(train_bar.print(dict(**train_meters, **mean_meters, lr=optimizer.param_groups[0]["lr"]))+f" | {batch_id+1} mini-batches ended")
else:
logging.info(train_bar.print(dict(**train_meters, lr=optimizer.param_groups[0]["lr"]))+f" | {batch_id+1} mini-batches ended")
if (batch_id+1) % 5000 == 0:
model.eval()
for meter in valid_meters.values():
meter.reset()
if args.loss == "loglike":
for meter in mean_meters.values():
meter.reset()
valid_bar = utils.ProgressBar(valid_loader)
running_valid_psnr = 0.0
for sample_id, (sample, noisy_inputs) in enumerate(valid_bar):
with torch.no_grad():
sample = sample.to(device)
noisy_inputs = noisy_inputs.to(device)
outputs, est_sigma = model(noisy_inputs)
noisy_frame = noisy_inputs[:, (mid*cpf):((mid+1)*cpf), :, :]
if args.blind_noise:
loss = utils.loss_function(outputs, noisy_frame, mode=args.loss, sigma=est_sigma, device=device)
else:
loss = utils.loss_function(outputs, noisy_frame, mode=args.loss, sigma=args.noise_std/255, device=device)
if args.loss == "loglike":
if args.blind_noise:
outputs, mean_image = utils.post_process(outputs, noisy_frame, model=args.model, sigma=est_sigma, device=device)
else:
outputs, mean_image = utils.post_process(outputs, noisy_frame, model=args.model, sigma=args.noise_std/255, device=device)
valid_psnr = utils.psnr(sample[:, (mid*cpf):((mid+1)*cpf), :, :], outputs, normalized=False, raw=True)
valid_ssim = utils.ssim(sample[:, (mid*cpf):((mid+1)*cpf), :, :], outputs, normalized=False, raw=True)
running_valid_psnr += valid_psnr
valid_meters["valid_loss"].update(loss.item())
valid_meters["valid_psnr"].update(valid_psnr.item())
valid_meters["valid_ssim"].update(valid_ssim.item())
if args.loss == "loglike":
mean_psnr = utils.psnr(sample[:, (mid*cpf):((mid+1)*cpf), :, :], mean_image, normalized=False, raw=True)
mean_ssim = utils.ssim(sample[:, (mid*cpf):((mid+1)*cpf), :, :], mean_image, normalized=False, raw=True)
mean_meters["mean_psnr"].update(mean_psnr.item())
mean_meters["mean_ssim"].update(mean_ssim.item())
running_valid_psnr /= (sample_id+1)
if writer is not None:
writer.add_scalar("psnr/valid", valid_meters['valid_psnr'].avg, global_step)
writer.add_scalar("ssim/valid", valid_meters['valid_ssim'].avg, global_step)
sys.stdout.flush()
if args.loss == "loglike":
logging.info("EVAL:"+train_bar.print(dict(**valid_meters, **mean_meters, lr=optimizer.param_groups[0]["lr"])))
else:
logging.info("EVAL:"+train_bar.print(dict(**valid_meters, lr=optimizer.param_groups[0]["lr"])))
utils.save_checkpoint(args, global_step, model, optimizer, score=valid_meters["valid_loss"].avg, mode="min")
scheduler.step()
if args.loss == "loglike":
logging.info(train_bar.print(dict(**train_meters, **mean_meters, lr=optimizer.param_groups[0]["lr"])))
else:
logging.info(train_bar.print(dict(**train_meters, lr=optimizer.param_groups[0]["lr"])))
if (epoch+1) % args.valid_interval == 0:
model.eval()
for meter in valid_meters.values():
meter.reset()
if args.loss == "loglike":
for meter in mean_meters.values():
meter.reset()
valid_bar = utils.ProgressBar(valid_loader)
running_valid_psnr = 0.0
for sample_id, (sample, noisy_inputs) in enumerate(valid_bar):
with torch.no_grad():
sample = sample.to(device)
noisy_inputs = noisy_inputs.to(device)
outputs, est_sigma = model(noisy_inputs)
noisy_frame = noisy_inputs[:, (mid*cpf):((mid+1)*cpf), :, :]
if args.blind_noise:
loss = utils.loss_function(outputs, noisy_frame, mode=args.loss, sigma=est_sigma, device=device)
else:
loss = utils.loss_function(outputs, noisy_frame, mode=args.loss, sigma=args.noise_std/255, device=device)
if args.loss == "loglike":
if args.blind_noise:
outputs, mean_image = utils.post_process(outputs, noisy_frame, model=args.model, sigma=est_sigma, device=device)
else:
outputs, mean_image = utils.post_process(outputs, noisy_frame, model=args.model, sigma=args.noise_std/255, device=device)
valid_psnr = utils.psnr(sample[:, (mid*cpf):((mid+1)*cpf), :, :], outputs, normalized=False, raw=True)
valid_ssim = utils.ssim(sample[:, (mid*cpf):((mid+1)*cpf), :, :], outputs, normalized=False, raw=True)
running_valid_psnr += valid_psnr
valid_meters["valid_loss"].update(loss.item())
valid_meters["valid_psnr"].update(valid_psnr.item())
valid_meters["valid_ssim"].update(valid_ssim.item())
if args.loss == "loglike":
mean_psnr = utils.psnr(sample[:, (mid*cpf):((mid+1)*cpf), :, :], mean_image, normalized=False, raw=True)
mean_ssim = utils.ssim(sample[:, (mid*cpf):((mid+1)*cpf), :, :], mean_image, normalized=False, raw=True)
mean_meters["mean_psnr"].update(mean_psnr.item())
mean_meters["mean_ssim"].update(mean_ssim.item())
running_valid_psnr /= (sample_id+1)
if writer is not None:
writer.add_scalar("psnr/valid", valid_meters['valid_psnr'].avg, global_step)
writer.add_scalar("ssim/valid", valid_meters['valid_ssim'].avg, global_step)
sys.stdout.flush()
if args.loss == "loglike":
logging.info("EVAL:"+train_bar.print(dict(**valid_meters, **mean_meters, lr=optimizer.param_groups[0]["lr"])))
else:
logging.info("EVAL:"+train_bar.print(dict(**valid_meters, lr=optimizer.param_groups[0]["lr"])))
utils.save_checkpoint(args, global_step, model, optimizer, score=valid_meters["valid_loss"].avg, mode="min")
logging.info(f"Done training! Best PSNR {utils.save_checkpoint.best_score:.3f} obtained after step {utils.save_checkpoint.best_step}.")
def get_args():
parser = argparse.ArgumentParser(allow_abbrev=False)
# Add data arguments
parser.add_argument("--data-path", default="data", help="path to data directory")
parser.add_argument("--dataset", default="RawVideo", help="train dataset name")
parser.add_argument("--aug", default=0, type=int, help="augmentations")
parser.add_argument("--batch-size", default=8, type=int, help="train batch size")
parser.add_argument("--image-size", default=128, type=int, help="image size for train")
parser.add_argument("--n-frames", default=5, type=int, help="number of frames for training")
parser.add_argument("--stride", default=64, type=int, help="stride for patch extraction")
# Add model arguments
parser.add_argument("--model", default="blind-video-net-4", help="model architecture")
# Add loss function
parser.add_argument("--loss", default="mse", help="loss function used for training")
# Add noise arguments
parser.add_argument("--noise_dist", default="G", help="G - Gaussian, P - Poisson")
parser.add_argument("--noise_mode", default="S", help="B - Blind, S - one noise level")
parser.add_argument('--noise_std', default = 30, type = float,
help = 'noise level when mode is S')
parser.add_argument('--min_noise', default = 0, type = float,
help = 'minimum noise level when mode is B')
parser.add_argument('--max_noise', default = 100, type = float,
help = 'maximum noise level when mode is B')
# Add optimization arguments
parser.add_argument("--lr", default=1e-3, type=float, help="learning rate")
parser.add_argument("--num-epochs", default=100, type=int, help="force stop training at specified epoch")
parser.add_argument("--valid-interval", default=1, type=int, help="evaluate every N epochs")
parser.add_argument("--save-interval", default=1, type=int, help="save a checkpoint every N steps")
# Parse twice as model arguments are not known the first time
parser = utils.add_logging_arguments(parser)
args, _ = parser.parse_known_args()
models.MODEL_REGISTRY[args.model].add_args(parser)
args = parser.parse_args()
return args
if __name__ == "__main__":
args = get_args()
main(args)