minchyeom/Aesthetic-Anime-Art

THE INPUT IMAGE MUST HAVE RGB CHANNELS. IT WILL NOT WORK WITH RGBA CHANNELS!

Usage

import torch
import torch.nn as nn
import torch.optim as optim
import torchvision.transforms as transforms
from PIL import Image

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

class CNN(nn.Module):
    def __init__(self, hidden_size=512):
        super(CNN, self).__init__()
        self.conv1 = nn.Conv2d(3, 16, kernel_size=3, padding=1)
        self.conv2 = nn.Conv2d(16, 32, kernel_size=3, padding=1)
        self.fc1 = nn.Linear(32 * 192 * 192, hidden_size)
        self.fc2 = nn.Linear(hidden_size, 2)

    def forward(self, x):
        x = torch.relu(self.conv1(x))
        x = torch.max_pool2d(x, kernel_size=2, stride=2)
        x = torch.relu(self.conv2(x))
        x = torch.max_pool2d(x, kernel_size=2, stride=2)
        x = x.view(-1, 32 * 192 * 192)
        x = torch.relu(self.fc1(x))
        x = self.fc2(x)
        return x

model = CNN().to(device).half()
criterion = nn.CrossEntropyLoss()
optimizer = optim.Adam(model.parameters(), lr=2.5e-5)

transform = transforms.Compose([
    transforms.Resize((768, 768)),
    transforms.ToTensor(),
    transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])

def infer(model, image_path):
    model.eval()
    image = Image.open(image_path)
    image = transform(image).unsqueeze(0).to(device).half()
    with torch.no_grad():
        output = model(image)
    predicted_class = torch.argmax(output).item()
    return predicted_class

checkpoint = torch.load('half_precision_model_checkpoint.pth')
model.load_state_dict(checkpoint['model_state_dict'])
optimizer.load_state_dict(checkpoint['optimizer_state_dict'])
epoch = checkpoint['epoch']
loss = checkpoint['loss']

image_path = 'good.jpg'
predicted_class = infer(model, image_path)
if int(predicted_class) == 0:
    print('Predicted class: Bad Image')
elif int(predicted_class) == 1:
    print('Predicted class: Good Image')