app.py

import os
import torch
import torch.nn as nn
from torchvision.utils import make_grid
import gradio as gr

# ------------------------
# CONFIG
# ------------------------
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
NOISE_DIM = 100
IMAGE_SIZE = 64   # Must match your GAN training
DEFAULT_NUM_SAMPLES = 16
GENERATOR_PATH = "generator_final.pth"
LOGO_PATH = "kyo.png"

# ------------------------
# Generator Definition
# ------------------------
class Generator(nn.Module):
    def __init__(self):
        super().__init__()
        self.net = nn.Sequential(
            nn.Linear(NOISE_DIM, 256),
            nn.ReLU(True),
            nn.Linear(256, 512),
            nn.ReLU(True),
            nn.Linear(512, 1024),
            nn.ReLU(True),
            nn.Linear(1024, 3 * IMAGE_SIZE * IMAGE_SIZE),
            nn.Tanh()
        )

    def forward(self, x):
        return self.net(x).view(-1, 3, IMAGE_SIZE, IMAGE_SIZE)

# ------------------------
# Load trained generator
# ------------------------
def load_generator():
    if not os.path.exists(GENERATOR_PATH):
        raise FileNotFoundError(
            f"Checkpoint not found: {GENERATOR_PATH}. Make sure it exists in your Hugging Face Space."
        )
    model = Generator().to(DEVICE)
    state_dict = torch.load(GENERATOR_PATH, map_location=DEVICE)
    model.load_state_dict(state_dict)
    model.eval()
    return model

G = load_generator()

# ------------------------
# Generate graffiti images
# ------------------------
def generate_graffiti(num_samples, seed):
    num_samples = int(num_samples)
    seed = int(seed)

    if num_samples < 1 or num_samples > 32:
        raise gr.Error("Number of samples must be between 1 and 32.")

    torch.manual_seed(seed)
    if torch.cuda.is_available():
        torch.cuda.manual_seed_all(seed)

    noise = torch.randn(num_samples, NOISE_DIM, device=DEVICE)

    with torch.no_grad():
        fake_images = G(noise).cpu()

    fake_images = fake_images * 0.5 + 0.5
    fake_images = fake_images.clamp(0, 1)

    nrow = min(4, num_samples)
    grid = make_grid(fake_images, nrow=nrow)
    grid_img = grid.permute(1, 2, 0).numpy()

    example_img = fake_images[0].permute(1, 2, 0).numpy()

    return grid_img, example_img

# ------------------------
# Gradio UI
# ------------------------
with gr.Blocks(title="KYO") as demo:
    if os.path.exists(LOGO_PATH):
        gr.Image(value=LOGO_PATH, show_label=False, width=180)

    gr.Markdown("# KYO")
    gr.Markdown("Generate graffiti-style images using PyTorch GAN.")

    with gr.Row():
        num_samples = gr.Slider(
            minimum=1,
            maximum=32,
            step=1,
            value=DEFAULT_NUM_SAMPLES,
            label="Number of images"
        )
        seed = gr.Number(value=42, precision=0, label="Random seed")

    generate_btn = gr.Button("Generate")

    with gr.Row():
        grid_output = gr.Image(label="Generated Grid")
        sample_output = gr.Image(label="First Sample")

    generate_btn.click(
        fn=generate_graffiti,
        inputs=[num_samples, seed],
        outputs=[grid_output, sample_output]
    )

# ------------------------
# Launch app
# ------------------------
if __name__ == "__main__":
    demo.launch()