-
Notifications
You must be signed in to change notification settings - Fork 4
/
train.py
136 lines (114 loc) · 5.08 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
import argparse
import os
import numpy as np
import PIL
from tensorboardX import SummaryWriter
import torch
import torch.nn as nn
from torch.utils import data
from torchvision import transforms
from torchvision.utils import make_grid, save_image
from tqdm import tqdm
from data import DS, InfiniteSampler
from loss import ConsistencyLoss, calc_gan_loss
from model import InpaintNet, FeaturePatchDiscriminator, PatchDiscriminator
parser = argparse.ArgumentParser()
parser.add_argument('--root', type=str, default='./root')
parser.add_argument('--save_dir', type=str, default='./snapshots')
parser.add_argument('--lr', type=float, default=2e-4, help="adam: learning rate")
parser.add_argument("--b1", type=float, default=0.5, help="adam: decay of first order momentum of gradient")
parser.add_argument("--b2", type=float, default=0.999, help="adam: decay of first order momentum of gradient")
parser.add_argument('--max_iter', type=int, default=1000000)
parser.add_argument('--batch_size', type=int, default=1)
parser.add_argument('--n_threads', type=int, default=16)
parser.add_argument('--save_model_interval', type=int, default=10000)
parser.add_argument('--vis_interval', type=int, default=1000)
parser.add_argument('--log_interval', type=int, default=10)
parser.add_argument('--image_size', type=int, default=256)
parser.add_argument('--resume', type=int)
args = parser.parse_args()
use_cuda = torch.cuda.is_available()
device = torch.device('cuda' if use_cuda else 'cpu')
if use_cuda:
torch.backends.cudnn.benchmark = True
if not os.path.exists(args.save_dir):
os.makedirs('{:s}/ckpt'.format(args.save_dir))
writer = SummaryWriter()
size = (args.image_size, args.image_size)
train_tf = transforms.Compose([
transforms.Resize(size),
transforms.RandomHorizontalFlip(),
transforms.RandomGrayscale(),
transforms.ToTensor(),
])
train_set = DS(args.root, train_tf)
iterator_train = iter(data.DataLoader(
train_set,
batch_size=args.batch_size,
sampler=InfiniteSampler(len(train_set)),
num_workers=args.n_threads))
print(len(train_set))
g_model = InpaintNet().to(device)
fd_model = FeaturePatchDiscriminator().to(device)
pd_model = PatchDiscriminator().to(device)
l1 = nn.L1Loss().to(device)
cons = ConsistencyLoss().to(device)
start_iter = 0
g_optimizer = torch.optim.Adam(
g_model.parameters(),
args.lr, (args.b1, args.b2))
fd_optimizer = torch.optim.Adam(
fd_model.parameters(),
args.lr, (args.b1, args.b2))
pd_optimizer = torch.optim.Adam(
pd_model.parameters(),
args.lr, (args.b1, args.b2))
if args.resume:
g_checkpoint = torch.load(f'{args.save_dir}/ckpt/G_{args.resume}.pth', map_location=device)
g_model.load_state_dict(g_checkpoint)
fd_checkpoint = torch.load(f'{args.save_dir}/ckpt/FD_{args.resume}.pth', map_location=device)
fd_model.load_state_dict(fd_checkpoint)
pd_checkpoint = torch.load(f'{args.save_dir}/ckpt/PD_{args.resume}.pth', map_location=device)
pd_model.load_state_dict(pd_checkpoint)
print('Models restored')
for i in tqdm(range(start_iter, args.max_iter)):
img, mask = [x.to(device) for x in next(iterator_train)]
img = 2. * img - 1. # [0,1] -> [-1,1]
masked = img * (1. - mask)
coarse_result, refine_result, csa, csa_d = g_model(masked, mask)
fg_loss, fd_loss = calc_gan_loss(fd_model, refine_result, img)
pg_loss, pd_loss = calc_gan_loss(pd_model, refine_result, img)
recon_loss = l1(coarse_result, img) + l1(refine_result, img)
gan_loss = fg_loss + pg_loss
cons_loss = cons(csa, csa_d, img, mask)
total_loss = 1*recon_loss + 0.01*cons_loss + 0.002*gan_loss
g_optimizer.zero_grad()
total_loss.backward(retain_graph=True)
g_optimizer.step()
fd_optimizer.zero_grad()
fd_loss.backward(retain_graph=True)
fd_optimizer.step()
pd_optimizer.zero_grad()
pd_loss.backward()
pd_optimizer.step()
if (i + 1) % args.save_model_interval == 0 or (i + 1) == args.max_iter:
#torch.save(g_model.state_dict(), f'{args.save_dir}/ckpt/G_{i + 1}.pth')
#torch.save(fd_model.state_dict(), f'{args.save_dir}/ckpt/FD_{i + 1}.pth')
#torch.save(pd_model.state_dict(), f'{args.save_dir}/ckpt/PD_{i + 1}.pth')
torch.save(g_model.state_dict(), f'{args.save_dir}/ckpt/G_10000.pth')
torch.save(fd_model.state_dict(), f'{args.save_dir}/ckpt/FD_10000.pth')
torch.save(pd_model.state_dict(), f'{args.save_dir}/ckpt/PD_10000.pth')
if (i + 1) % args.log_interval == 0:
writer.add_scalar('g_loss/recon_loss', recon_loss.item(), i + 1)
writer.add_scalar('g_loss/cons_loss', cons_loss.item(), i + 1)
writer.add_scalar('g_loss/gan_loss', gan_loss.item(), i + 1)
writer.add_scalar('g_loss/total_loss', total_loss.item(), i + 1)
writer.add_scalar('d_loss/fd_loss', fd_loss.item(), i + 1)
writer.add_scalar('d_loss/pd_loss', pd_loss.item(), i + 1)
def denorm(x):
out = (x + 1) / 2 # [-1,1] -> [0,1]
return out.clamp_(0, 1)
if (i + 1) % args.vis_interval == 0:
ims = torch.cat([img, masked, refine_result], dim=3)
writer.add_images('raw_masked_refine', denorm(ims), i + 1)
writer.close()