Data-raw/model.py

import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import DataLoader
import torchvision.transforms as transforms
from torchvision.datasets import ImageFolder

# Define the autoencoder architecture
class Autoencoder(nn.Module):
    def __init__(self):
        super(Autoencoder, self).__init__()
        # Encoder layers
        self.encoder = nn.Sequential(
            nn.Conv2d(3, 16, kernel_size=3, stride=2, padding=1),  # 3x256x256 -> 16x128x128
            nn.LeakyReLU(inplace=True),
            nn.Conv2d(16, 32, kernel_size=3, stride=2, padding=1),  # 16x128x128 -> 32x64x64
            nn.LeakyReLU(inplace=True),
            nn.Conv2d(32, 64, kernel_size=3, stride=2, padding=1),  # 32x64x64 -> 64x32x32
            nn.LeakyReLU(inplace=True)
        )
        # Decoder layers
        self.decoder = nn.Sequential(
            nn.ConvTranspose2d(64, 32, kernel_size=3, stride=2, padding=1, output_padding=1),  # 64x32x32 -> 32x64x64
            nn.LeakyReLU(inplace=True),
            nn.ConvTranspose2d(32, 16, kernel_size=3, stride=2, padding=1, output_padding=1),  # 32x64x64 -> 16x128x128
            nn.LeakyReLU(inplace=True),
            nn.ConvTranspose2d(16, 3, kernel_size=3, stride=2, padding=1, output_padding=1),  # 16x128x128 -> 3x256x256
            nn.Tanh()
        )

    def forward(self, x):
        x = self.encoder(x)
        x = self.decoder(x)
        return x