utils/models.py

import torch.nn as nn
import torch

class VGGEncoder(nn.Module):
    def __init__(self, vgg_path):
        super(VGGEncoder, self).__init__()

        self.vgg = nn.Sequential(
            nn.Conv2d(3, 3, (1, 1)),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(3, 64, (3, 3)),
            nn.ReLU(),  # relu1-1
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(64, 64, (3, 3)),
            nn.ReLU(),  # relu1-2
            nn.MaxPool2d((2, 2), (2, 2), (0, 0), ceil_mode=True),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(64, 128, (3, 3)),
            nn.ReLU(),  # relu2-1
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(128, 128, (3, 3)),
            nn.ReLU(),  # relu2-2
            nn.MaxPool2d((2, 2), (2, 2), (0, 0), ceil_mode=True),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(128, 256, (3, 3)),
            nn.ReLU(),  # relu3-1
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(256, 256, (3, 3)),
            nn.ReLU(),  # relu3-2
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(256, 256, (3, 3)),
            nn.ReLU(),  # relu3-3
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(256, 256, (3, 3)),
            nn.ReLU(),  # relu3-4
            nn.MaxPool2d((2, 2), (2, 2), (0, 0), ceil_mode=True),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(256, 512, (3, 3)),
            nn.ReLU(),  # relu4-1, this is the last layer used
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(512, 512, (3, 3)),
            nn.ReLU(),  # relu4-2
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(512, 512, (3, 3)),
            nn.ReLU(),  # relu4-3
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(512, 512, (3, 3)),
            nn.ReLU(),  # relu4-4
            nn.MaxPool2d((2, 2), (2, 2), (0, 0), ceil_mode=True),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(512, 512, (3, 3)),
            nn.ReLU(),  # relu5-1
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(512, 512, (3, 3)),
            nn.ReLU(),  # relu5-2
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(512, 512, (3, 3)),
            nn.ReLU(),  # relu5-3
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(512, 512, (3, 3)),
            nn.ReLU()  # relu5-4
        )
        self.vgg.load_state_dict(torch.load(vgg_path)) # load the pre-trained VGG model weights from the specified path
        self.vgg = nn.Sequential(*list(self.vgg.children())[:31]) # keep only the layers up to relu4-1 (inclusive) since those are the layers used for feature extraction in style transfer
        enc_layers = list(self.vgg.children())# create a list of the layers in the VGG model to facilitate splitting it into four sequential blocks corresponding to the different stages of the network (enc_1, enc_2, enc_3, enc_4)
        self.enc_1 = nn.Sequential(*enc_layers[:4])
        self.enc_2 = nn.Sequential(*enc_layers[4:11])
        self.enc_3 = nn.Sequential(*enc_layers[11:18])
        self.enc_4 = nn.Sequential(*enc_layers[18:31])

        for name in ['enc_1', 'enc_2', 'enc_3', 'enc_4']: #
            for param in getattr(self, name).parameters():
                param.requires_grad = False

    def forward(self, input, is_test=False):
        h1 = self.enc_1(input)
        h2 = self.enc_2(h1)
        h3 = self.enc_3(h2)
        h4 = self.enc_4(h3)
        if is_test:
            return h4
        return h1, h2, h3, h4


class Decoder(nn.Module):
    def __init__(self):
        super(Decoder, self).__init__()
        self.net = nn.Sequential(
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(512, 256, (3, 3)),
            nn.ReLU(),
            nn.Upsample(scale_factor=2, mode='nearest'),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(256, 256, (3, 3)),
            nn.ReLU(),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(256, 256, (3, 3)),
            nn.ReLU(),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(256, 256, (3, 3)),
            nn.ReLU(),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(256, 128, (3, 3)),
            nn.ReLU(),
            nn.Upsample(scale_factor=2, mode='nearest'),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(128, 128, (3, 3)),
            nn.ReLU(),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(128, 64, (3, 3)),
            nn.ReLU(),
            nn.Upsample(scale_factor=2, mode='nearest'),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(64, 64, (3, 3)),
            nn.ReLU(),
            nn.ReflectionPad2d((1, 1, 1, 1)),
            nn.Conv2d(64, 3, (3, 3)),         
        )

    def forward(self, input):
        return self.net(input)