File size: 6,461 Bytes
f884940 | 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 | """
Copyright (C) 2019 NVIDIA Corporation. All rights reserved.
Licensed under the CC BY-NC-SA 4.0 license (https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode).
"""
import torch
import torch.nn as nn
class InstanceNorm(nn.Module):
def __init__(self, epsilon=1e-8):
"""
@notice: avoid in-place ops.
https://discuss.pytorch.org/t/encounter-the-runtimeerror-one-of-the-variables-needed-for-gradient-computation-has-been-modified-by-an-inplace-operation/836/3
"""
super(InstanceNorm, self).__init__()
self.epsilon = epsilon
def forward(self, x):
x = x - torch.mean(x, (2, 3), True)
tmp = torch.mul(x, x) # or x ** 2
tmp = torch.rsqrt(torch.mean(tmp, (2, 3), True) + self.epsilon)
return x * tmp
class ApplyStyle(nn.Module):
"""
@ref: https://github.com/lernapparat/lernapparat/blob/master/style_gan/pytorch_style_gan.ipynb
"""
def __init__(self, latent_size, channels):
super(ApplyStyle, self).__init__()
self.linear = nn.Linear(latent_size, channels * 2)
def forward(self, x, latent):
style = self.linear(latent) # style => [batch_size, n_channels*2]
shape = [-1, 2, x.size(1), 1, 1]
style = style.view(shape) # [batch_size, 2, n_channels, ...]
#x = x * (style[:, 0] + 1.) + style[:, 1]
x = x * (style[:, 0] * 1 + 1.) + style[:, 1] * 1
return x
class ResnetBlock_Adain(nn.Module):
def __init__(self, dim, latent_size, padding_type, activation=nn.ReLU(True)):
super(ResnetBlock_Adain, self).__init__()
p = 0
conv1 = []
if padding_type == 'reflect':
conv1 += [nn.ReflectionPad2d(1)]
elif padding_type == 'replicate':
conv1 += [nn.ReplicationPad2d(1)]
elif padding_type == 'zero':
p = 1
else:
raise NotImplementedError('padding [%s] is not implemented' % padding_type)
conv1 += [nn.Conv2d(dim, dim, kernel_size=3, padding = p), InstanceNorm()]
self.conv1 = nn.Sequential(*conv1)
self.style1 = ApplyStyle(latent_size, dim)
self.act1 = activation
p = 0
conv2 = []
if padding_type == 'reflect':
conv2 += [nn.ReflectionPad2d(1)]
elif padding_type == 'replicate':
conv2 += [nn.ReplicationPad2d(1)]
elif padding_type == 'zero':
p = 1
else:
raise NotImplementedError('padding [%s] is not implemented' % padding_type)
conv2 += [nn.Conv2d(dim, dim, kernel_size=3, padding=p), InstanceNorm()]
self.conv2 = nn.Sequential(*conv2)
self.style2 = ApplyStyle(latent_size, dim)
def forward(self, x, dlatents_in_slice):
y = self.conv1(x)
y = self.style1(y, dlatents_in_slice)
y = self.act1(y)
y = self.conv2(y)
y = self.style2(y, dlatents_in_slice)
out = x + y
return out
class Generator_Adain_Upsample(nn.Module):
def __init__(self, input_nc, output_nc, latent_size, n_blocks=6, deep=False,
norm_layer=nn.BatchNorm2d,
padding_type='reflect'):
assert (n_blocks >= 0)
super(Generator_Adain_Upsample, self).__init__()
activation = nn.ReLU(True)
self.deep = deep
self.first_layer = nn.Sequential(nn.ReflectionPad2d(3), nn.Conv2d(input_nc, 64, kernel_size=7, padding=0),
norm_layer(64), activation)
### downsample
self.down1 = nn.Sequential(nn.Conv2d(64, 128, kernel_size=3, stride=2, padding=1),
norm_layer(128), activation)
self.down2 = nn.Sequential(nn.Conv2d(128, 256, kernel_size=3, stride=2, padding=1),
norm_layer(256), activation)
self.down3 = nn.Sequential(nn.Conv2d(256, 512, kernel_size=3, stride=2, padding=1),
norm_layer(512), activation)
if self.deep:
self.down4 = nn.Sequential(nn.Conv2d(512, 512, kernel_size=3, stride=2, padding=1),
norm_layer(512), activation)
### resnet blocks
BN = []
for i in range(n_blocks):
BN += [
ResnetBlock_Adain(512, latent_size=latent_size, padding_type=padding_type, activation=activation)]
self.BottleNeck = nn.Sequential(*BN)
if self.deep:
self.up4 = nn.Sequential(
nn.Upsample(scale_factor=2, mode='bilinear',align_corners=False),
nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),
nn.BatchNorm2d(512), activation
)
self.up3 = nn.Sequential(
nn.Upsample(scale_factor=2, mode='bilinear',align_corners=False),
nn.Conv2d(512, 256, kernel_size=3, stride=1, padding=1),
nn.BatchNorm2d(256), activation
)
self.up2 = nn.Sequential(
nn.Upsample(scale_factor=2, mode='bilinear',align_corners=False),
nn.Conv2d(256, 128, kernel_size=3, stride=1, padding=1),
nn.BatchNorm2d(128), activation
)
self.up1 = nn.Sequential(
nn.Upsample(scale_factor=2, mode='bilinear',align_corners=False),
nn.Conv2d(128, 64, kernel_size=3, stride=1, padding=1),
nn.BatchNorm2d(64), activation
)
self.last_layer = nn.Sequential(nn.ReflectionPad2d(3), nn.Conv2d(64, output_nc, kernel_size=7, padding=0))
def forward(self, input, dlatents):
x = input # 3*224*224
skip1 = self.first_layer(x)
skip2 = self.down1(skip1)
skip3 = self.down2(skip2)
if self.deep:
skip4 = self.down3(skip3)
x = self.down4(skip4)
else:
x = self.down3(skip3)
bot = []
bot.append(x)
features = []
for i in range(len(self.BottleNeck)):
x = self.BottleNeck[i](x, dlatents)
bot.append(x)
if self.deep:
x = self.up4(x)
features.append(x)
x = self.up3(x)
features.append(x)
x = self.up2(x)
features.append(x)
x = self.up1(x)
features.append(x)
x = self.last_layer(x)
# x = (x + 1) / 2
# return x, bot, features, dlatents
return x |