PyTorch风格迁移算法

PyTorch风格迁移算法

本算法主要是基于Pytorch库以及VGG19模型的风格迁移算法,主要依赖的库包括:argparse,torch,torchvision,Pillow。
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
from __future__ import division
from torchvision import models
from torchvision import transforms
from PIL import Image
import argparse
import torch
import torchvision
import torch.nn as nn
import numpy as np
import mysql.connector

# 设备配置
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

def load_image(image_path, transform=None, max_size=None, shape=None):
"""Load an image and convert it to a torch tensor."""
image = Image.open(image_path).convert('RGB')

if max_size:
scale = max_size / max(image.size)
size = np.array(image.size) * scale
image = image.resize(size.astype(int), Image.ANTIALIAS)

if shape:
image = image.resize(shape, Image.LANCZOS)

if transform:
image = transform(image).unsqueeze(0)

return image.to(device)


class VGGNet(nn.Module):
def __init__(self):
"""Select conv1_1 ~ conv5_1 activation maps."""
super(VGGNet, self).__init__()
self.select = ['0', '5', '10', '19', '28']
self.vgg = models.vgg19(pretrained=True).features

def forward(self, x):
"""Extract multiple convolutional feature maps."""
features = []
for name, layer in self.vgg._modules.items():
x = layer(x)
if name in self.select:
features.append(x)
return features


def main(config):

transform = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize(mean=(0.485, 0.456, 0.406),
std=(0.229, 0.224, 0.225))])

# 加载内容与风格图
# 调整两者为相同大小
content = load_image(config.content, transform, max_size=config.max_size)
style = load_image(config.style, transform, shape=[content.size(2), content.size(3)])

# 初始化输出图像(使用内容图)
target = content.clone().requires_grad_(True)

optimizer = torch.optim.Adam([target], lr=config.lr, betas=[0.5, 0.999])

vgg = VGGNet().to(device).eval()

for step in range(config.total_step):

# 提取特征向量
target_features = vgg(target)
content_features = vgg(content)
style_features = vgg(style)

style_loss = 0
content_loss = 0
for f1, f2, f3 in zip(target_features, content_features, style_features):
# 计算内容损失
content_loss += torch.mean((f1 - f2)**2)

# 重塑卷积特征映射
_, c, h, w = f1.size()
f1 = f1.view(c, h * w)
f3 = f3.view(c, h * w)

# 计算gram矩阵
f1 = torch.mm(f1, f1.t())
f3 = torch.mm(f3, f3.t())

# 计算风格损失
style_loss += torch.mean((f1 - f3)**2) / (c * h * w)

# 计算总损失并优化
loss = content_loss + config.style_weight * style_loss
optimizer.zero_grad()
loss.backward()
optimizer.step()

if (step+1) % config.log_step == 0:
print ('Step [{}/{}], Content Loss: {:.4f}, Style Loss: {:.4f}'
.format(step+1, config.total_step, content_loss.item(), style_loss.item()))

if (step+1) % config.sample_step == 0:
# 保存图案
denorm = transforms.Normalize((-2.12, -2.04, -1.80), (4.37, 4.46, 4.44))
img = target.clone().squeeze()
img = denorm(img).clamp_(0, 1)
torchvision.utils.save_image(img, 'output-{}.png'.format(step+1))
if(step+1)==config.total_step:
# 保存图案
denorm = transforms.Normalize((-2.12, -2.04, -1.80), (4.37, 4.46, 4.44))
img = target.clone().squeeze()
img = denorm(img).clamp_(0, 1)
torchvision.utils.save_image(img, '{}'.format(config.name))


if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('--content', type=str, default='png/content.png')
parser.add_argument('--style', type=str, default='png/style.png')
parser.add_argument('--max_size', type=int, default=400)
parser.add_argument('--total_step', type=int, default=2000)
parser.add_argument('--log_step', type=int, default=10)
parser.add_argument('--sample_step', type=int, default=500)
parser.add_argument('--style_weight', type=float, default=100)
parser.add_argument('--lr', type=float, default=0.003)
parser.add_argument('--user', type=str, default='null')
parser.add_argument('--name', type=str, default='test')
config = parser.parse_args()
print(config)
torch.cuda.empty_cache()
main(config)
mydb = mysql.connector.connect(
host="",
user="",
passwd="",
database=""
)
mycursor = mydb.cursor()

sql = ""
val = ("", config.name)

mycursor.execute(sql, val)

mydb.commit()

print(mycursor.rowcount, " 条记录被修改")

评论

Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now

×