def __init__(self, channels):
super(ResidualBlock, self).__init__()
self.conv1 = SpectralNorm(
nn.Conv2d(channels, channels, kernel_size=3, padding=1))
self.bn1 = nn.BatchNorm2d(channels)
self.prelu = nn.PReLU()
self.conv2 = SpectralNorm(
nn.Conv2d(channels, channels, kernel_size=3, padding=1))
self.bn2 = nn.BatchNorm2d(channels)
def __init__(self, batch_size, image_size=64, z_dim=100, conv_dim=64):
super(Generator, self).__init__()
self.imsize = image_size
layer1 = []
layer2 = []
layer3 = []
last = []
repeat_num = int(np.log2(self.imsize)) - 3
mult = 2**repeat_num # 8
layer1.append(
SpectralNorm(nn.ConvTranspose2d(z_dim, conv_dim * mult, 4)))
layer1.append(nn.BatchNorm2d(conv_dim * mult))
layer1.append(nn.ReLU())
curr_dim = conv_dim * mult
layer2.append(
SpectralNorm(
nn.ConvTranspose2d(curr_dim, int(curr_dim / 2), 4, 2, 1)))
layer2.append(nn.BatchNorm2d(int(curr_dim / 2)))
layer2.append(nn.ReLU())
curr_dim = int(curr_dim / 2)
layer3.append(
SpectralNorm(
nn.ConvTranspose2d(curr_dim, int(curr_dim / 2), 4, 2, 1)))
layer3.append(nn.BatchNorm2d(int(curr_dim / 2)))
layer3.append(nn.ReLU())
if self.imsize == 64:
layer4 = []
curr_dim = int(curr_dim / 2)
layer4.append(
SpectralNorm(
nn.ConvTranspose2d(curr_dim, int(curr_dim / 2), 4, 2, 1)))
layer4.append(nn.BatchNorm2d(int(curr_dim / 2)))
layer4.append(nn.ReLU())
self.l4 = nn.Sequential(*layer4)
curr_dim = int(curr_dim / 2)
self.l1 = nn.Sequential(*layer1)
self.l2 = nn.Sequential(*layer2)
self.l3 = nn.Sequential(*layer3)
last.append(nn.ConvTranspose2d(curr_dim, 3, 4, 2, 1))
last.append(nn.Tanh())
self.last = nn.Sequential(*last)
self.attn1 = Self_Attn(128, 'relu')
self.attn2 = Self_Attn(64, 'relu')
def __init__(self):
super(Discriminator, self).__init__()
self.conv0 = nn.Sequential(
SpectralNorm(nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1)),
nn.BatchNorm2d(64), nn.ReLU())
self.conv1 = nn.Sequential(
SpectralNorm(nn.Conv2d(64, 64, kernel_size=3, stride=1,
padding=1)), nn.BatchNorm2d(64), nn.ReLU())
self.conv2 = nn.Sequential(
SpectralNorm(nn.Conv2d(64, 64, kernel_size=3, stride=1,
padding=1)), nn.BatchNorm2d(64), nn.ReLU())
self.average_pooling = nn.AdaptiveAvgPool2d(1)
self.linear = SpectralNorm(nn.Linear(64, 1))
def __init__(self):
super(Generator, self).__init__()
self.det_conv0 = nn.Sequential(
SpectralNorm(
nn.Conv2d(3 + 1, 64, kernel_size=3, stride=1, padding=1)),
nn.ReLU())
self.det_conv1 = ResidualBlock(64)
self.det_conv2 = ResidualBlock(64)
self.det_conv3 = ResidualBlock(64)
self.det_conv4 = ResidualBlock(64)
self.det_conv5 = ResidualBlock(64)
self.det_conv6 = ResidualBlock(64)
self.output = nn.Sequential(SpectralNorm(nn.Conv2d(64, 3, 3, 1, 1)))
def __init__(self,
batch_size=64,
image_size=64,
conv_dim=64,
feat_dim=512):
super(Discriminator, self).__init__()
self.imsize = image_size
layer1 = []
layer2 = []
layer3 = []
last = []
layer1.append(SpectralNorm(nn.Conv2d(3, conv_dim, 4, 2, 1)))
layer1.append(nn.LeakyReLU(0.1))
curr_dim = conv_dim
layer2.append(SpectralNorm(nn.Conv2d(curr_dim, curr_dim * 2, 4, 2, 1)))
layer2.append(nn.LeakyReLU(0.1))
curr_dim = curr_dim * 2
layer3.append(SpectralNorm(nn.Conv2d(curr_dim, curr_dim * 2, 4, 2, 1)))
layer3.append(nn.LeakyReLU(0.1))
curr_dim = curr_dim * 2
if self.imsize == 256:
layer4 = []
layer4.append(
SpectralNorm(nn.Conv2d(curr_dim, curr_dim * 2, 4, 2, 1)))
layer4.append(nn.LeakyReLU(0.1))
self.l4 = nn.Sequential(*layer4)
curr_dim = curr_dim * 2
self.l1 = nn.Sequential(*layer1)
self.l2 = nn.Sequential(*layer2)
self.l3 = nn.Sequential(*layer3)
last.append(nn.Conv2d(curr_dim, 1, 4))
self.last = nn.Sequential(*last)
self.attn1 = Self_Attn(256, 'relu')
self.attn2 = Self_Attn(512, 'relu')
self.cn = CondNorm2d(feat_dim, 512)
def __init__(self,
batch_size,
image_size=64,
z_dim=100,
conv_dim=64,
feat_dim=512):
super(Generator32, self).__init__()
self.imsize = image_size
self.z_dim = z_dim
layer1 = []
layer2 = []
layer3 = []
last = []
repeat_num = int(np.log2(self.imsize)) - 3
mult = 2**repeat_num # 8
self.l1 = SpectralNorm(nn.ConvTranspose2d(z_dim, conv_dim * mult, 4))
#layer1.append(nn.BatchNorm2d(conv_d
self.cn1 = CondNorm2d(feat_dim, conv_dim * mult)
curr_dim = conv_dim * mult
self.l2 = SpectralNorm(
nn.ConvTranspose2d(curr_dim, int(curr_dim / 2), 4, 2, 1))
#layer2.append(nn.BatchNorm2d(int(curr_dim / 2)))
self.cn2 = CondNorm2d(feat_dim, int(curr_dim / 2))
curr_dim = int(curr_dim / 2)
self.l3 = SpectralNorm(
nn.ConvTranspose2d(curr_dim, int(curr_dim / 2), 4, 2, 1))
#layer3.append(nn.BatchNorm2d(int(curr_dim / 2)))
self.cn3 = CondNorm2d(feat_dim, int(curr_dim / 2))
curr_dim = int(curr_dim / 2)
self.last = nn.ConvTranspose2d(curr_dim, 3, 4, 2, 1)
self.last_cn = CondNorm2d(feat_dim, 3)
self.attn1 = Self_Attn(128, 'relu')
self.attn2 = Self_Attn(64, 'relu')