models.py

import torch.nn as nn
import torch.nn.functional as F
import torch

def weights_init_normal(m):
    classname = m.__class__.__name__
    if classname.find("Conv") != -1:
        torch.nn.init.normal_(m.weight.data, 0.0, 0.02)
        if hasattr(m, "bias") and m.bias is not None:
            torch.nn.init.constant_(m.bias.data, 0.0)
    elif classname.find("BatchNorm2d") != -1:
        torch.nn.init.normal_(m.weight.data, 1.0, 0.02)
        torch.nn.init.constant_(m.bias.data, 0.0)


# RESNET
class ResidualBlock(nn.Module):
    def __init__(self, in_features):
        super(ResidualBlock, self).__init__()

        self.block = nn.Sequential(
            nn.ReflectionPad2d(1),
            nn.Conv2d(in_features, in_features, 3),
            nn.InstanceNorm2d(in_features),
            nn.ReLU(inplace=True),
            nn.ReflectionPad2d(1),
            nn.Conv2d(in_features, in_features, 3),
            nn.InstanceNorm2d(in_features),
        )

    def forward(self, x):
        return x + self.block(x)


# GENERATOR
class Generator(nn.Module):
    def __init__(self, input_shape, num_residual_blocks):
        super(Generator, self).__init__()

        channels = input_shape[0]

        # Initial convolution block
        out_features = 32
        downsampling = [
            nn.ReflectionPad2d(channels),
            nn.Conv2d(channels, out_features, 7),
            nn.InstanceNorm2d(out_features),
            nn.ReLU(inplace=True),
        ]
        in_features = out_features

        # Downsampling
        for _ in range(2):
            out_features *= 2
            downsampling += [
                nn.Conv2d(in_features, out_features, 3, stride=2, padding=1),
                nn.InstanceNorm2d(out_features),
                nn.ReLU(inplace=True),
            ]
            in_features = out_features

        self.src_d = nn.Sequential(*downsampling)
        self.ref_e = nn.Sequential(*downsampling)
        self.ref_f = nn.Sequential(*downsampling)
        self.ref_l = nn.Sequential(*downsampling)

        # Residual blocks
        res_block = []
        out_features = out_features * 4
        for _ in range(num_residual_blocks):
            res_block += [ResidualBlock(out_features)]
        
        self.res_block = nn.Sequential(*res_block)

        # Upsampling
        in_features = out_features

        upsampling = []
        for _ in range(2):
            out_features //= 2
            upsampling += [
                nn.Conv2d(in_features, out_features, 3, stride=1, padding=1),
                nn.InstanceNorm2d(out_features),
                nn.ReLU(inplace=True),
            ]
            in_features = out_features
        for _ in range(2):
            out_features //= 2
            upsampling += [
                nn.Upsample(scale_factor=2),
                nn.Conv2d(in_features, out_features, 3, stride=1, padding=1),
                nn.InstanceNorm2d(out_features),
                nn.ReLU(inplace=True),
            ]
            in_features = out_features

        # Output layer
        upsampling += [nn.ReflectionPad2d(channels), nn.Conv2d(out_features, channels, 7), nn.Tanh()]

        self.src_u = nn.Sequential(*upsampling)

    def forward(self, src, ref_e, ref_l, ref_f):
        src_1 = self.src_d(src)
        ref_1 = self.ref_e(ref_e)
        ref_2 = self.ref_f(ref_l)
        ref_3 = self.ref_l(ref_f)
        input_res = torch.cat([src_1, ref_1, ref_2, ref_3], 1)
        output_res = self.res_block(input_res)
        out = self.src_u(output_res)
        return out


# DISCRIMINATOR
class Discriminator(nn.Module):
    def __init__(self, input_shape):
        super(Discriminator, self).__init__()

        channels, height, width = input_shape

        # Calculate output shape of image discriminator (PatchGAN)
        self.output_shape = (1, height // 2 ** 4, width // 2 ** 4)

        def discriminator_block(in_filters, out_filters, normalize=True):
            """Returns downsampling layers of each discriminator block"""
            layers = [nn.Conv2d(in_filters, out_filters, 4, stride=2, padding=1)]
            if normalize:
                layers.append(nn.InstanceNorm2d(out_filters))
            layers.append(nn.LeakyReLU(0.2, inplace=True))
            return layers

        self.model = nn.Sequential(
            *discriminator_block(channels, 64, normalize=False),
            *discriminator_block(64, 128),
            *discriminator_block(128, 256),
            *discriminator_block(256, 512),
            nn.ZeroPad2d((1, 0, 1, 0)),
            nn.Conv2d(512, 1, 4, padding=1)
        )

    def forward(self, img):
        return self.model(img)