from config import Config

import os
import torch
import torch.nn as nn


class ResidualBlock(nn.Module):
    def __init__(self, in_channels, out_channels):
        super(ResidualBlock, self).__init__()
        
        self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=1, padding=1)
        self.in1 = nn.InstanceNorm2d(out_channels)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=1, padding=1)
        self.in2 = nn.InstanceNorm2d(out_channels)
        
        self.skip = nn.Conv2d(in_channels, out_channels, kernel_size=1, stride=1, padding=0)
        self.skip_in = nn.InstanceNorm2d(out_channels)

    def forward(self, x):
        out = self.conv1(x)
        out = self.in1(out)
        out = self.relu(out)
        out = self.conv2(out)
        out = self.in2(out)
        
        skip = self.skip(x)
        skip = self.skip_in(skip)
        
        return out + skip


class Generator(nn.Module):
    def __init__(self, input_channels=3, output_channels=3):
        super(Generator, self).__init__()

        self.enc1 = nn.Conv2d(input_channels, 64, kernel_size=7, stride=1, padding=3)
        self.enc2 = nn.Conv2d(64, 128, kernel_size=3, stride=2, padding=1)
        self.enc3 = nn.Conv2d(128, 256, kernel_size=3, stride=2, padding=1)

        self.res1 = ResidualBlock(256, 256)
        self.res2 = ResidualBlock(256, 256)
        self.res3 = ResidualBlock(256, 256)
        self.res4 = ResidualBlock(256, 256)
        self.res5 = ResidualBlock(256, 256)
        self.res6 = ResidualBlock(256, 256)
        self.res7 = ResidualBlock(256, 256)
        self.res8 = ResidualBlock(256, 256)
        self.res9 = ResidualBlock(256, 256)

        self.dec1 = nn.ConvTranspose2d(256, 128, kernel_size=3, stride=2, padding=1, output_padding=1)
        self.dec2 = nn.ConvTranspose2d(128, 64, kernel_size=3, stride=2, padding=1, output_padding=1)
        self.dec3 = nn.Conv2d(64, output_channels, kernel_size=7, stride=1, padding=3)

        self.tanh = nn.Tanh()

    def forward(self, x):
        x = self.enc1(x)
        x = self.enc2(x)
        x = self.enc3(x)

        x = self.res1(x)
        x = self.res2(x)
        x = self.res3(x)
        x = self.res4(x)
        x = self.res5(x)
        x = self.res6(x)
        x = self.res7(x)
        x = self.res8(x)
        x = self.res9(x)

        x = self.dec1(x)
        x = self.dec2(x)
        x = self.dec3(x)

        return self.tanh(x)
    

class Discriminator(nn.Module):
    def __init__(self, input_channels=3):
        super(Discriminator, self).__init__()

        self.model = nn.Sequential(
            nn.Conv2d(input_channels, 64, kernel_size=4, stride=2, padding=1),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Conv2d(64, 128, kernel_size=4, stride=2, padding=1),
            nn.InstanceNorm2d(128),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Conv2d(128, 256, kernel_size=4, stride=2, padding=1),
            nn.InstanceNorm2d(256),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Conv2d(256, 512, kernel_size=4, stride=2, padding=1),
            nn.InstanceNorm2d(512),
            nn.LeakyReLU(0.2, inplace=True),
            nn.Conv2d(512, 1, kernel_size=4, stride=1, padding=1)
        )

    def forward(self, x):
        output = self.model(x)
        return output
    
    
class CycleGAN(nn.Module):
    def __init__(self):
        super().__init__()
        
        self.generator_A2B = Generator()
        self.generator_B2A = Generator()
        
        self.discriminator_A = Discriminator()
        self.discriminator_B = Discriminator()    


def get_model(name):
    model = CycleGAN()
    model.load_state_dict(torch.load(os.path.join(Config.WEIGHTS_PATH, f"{name}.pth"), map_location=Config.DEVICE))
    return model