nn.PixelShuffle(2),
nn.Conv2d(G,
args.n_colors,
kSize,
padding=(kSize - 1) // 2,
stride=1)
])
else:
raise ValueError("scale must be 2 or 3 or 4.")
def forward(self, x):
f__1 = self.SFENet1(x)
x = self.SFENet2(f__1)
RDBs_out = []
for i in range(self.D):
x = self.RDBs[i](x)
RDBs_out.append(x)
x = self.GFF(torch.cat(RDBs_out, 1))
x += f__1
return self.UPNet(x)
if __name__ == '__main__':
from help_func.my_torchsummary import summary
from option import args
model = RDN(args)
summary(model, (3, 192, 192))
# global feature fusion
self.gff = nn.Sequential(
nn.Conv2d(self.G * self.D, self.G0, kernel_size=1),
nn.Conv2d(self.G0, self.G0, kernel_size=3, padding=3 // 2))
self.output = nn.Conv2d(self.G0,
output_num_channels,
kernel_size=3,
padding=3 // 2)
def forward(self, x):
sfe1 = self.sfe1(x)
sfe2 = self.sfe2(sfe1)
x = sfe2
local_features = []
for i in range(self.D):
x = self.rdbs[i](x)
local_features.append(x)
x = self.gff(torch.cat(local_features,
1)) + sfe1 # global residual learning
x = self.output(x)
return x
if __name__ == '__main__':
from help_func.my_torchsummary import summary
net = RDN(input_num_channels=3, output_num_channels=3)
summary(net, (3, 64, 64), device='cpu')
2, stride=1),
nn.PixelShuffle(2),
nn.Conv2d(G, G * 4, kSize, padding=(kSize - 1) // 2, stride=1),
nn.PixelShuffle(2),
nn.Conv2d(G, 3, kSize, padding=(kSize - 1) // 2, stride=1)
])
else:
raise ValueError("scale must be 2 or 3 or 4.")
def forward(self, x):
_x = self.SFENet1_input4(x)
_x = self.SFENet2(_x)
RDBs_out = []
for i in range(self.D):
_x = self.RDBs[i](_x)
RDBs_out.append(_x)
_x = self.GFF(torch.cat(RDBs_out, 1))
# x += f__1
return self.UPNet(_x) + x[:, :3, ...]
if __name__ == '__main__':
from help_func.my_torchsummary import summary
from option import args
model = RDN(args)
summary(model, (4, 64, 64))