app.py

import cv2
import numpy as np
import gradio as gr
import onnxruntime as rt
from huggingface_hub import hf_hub_download


def predict(img):
    img = img.astype(np.float32) / 255
    s = 768
    h, w = img.shape[:-1]
    h, w = (s, int(s * w / h)) if h > w else (int(s * h / w), s)
    ph, pw = s - h, s - w
    img_input = np.zeros([s, s, 3], dtype=np.float32)
    img_input[ph // 2:ph // 2 + h, pw // 2:pw // 2 + w] = cv2.resize(img, (w, h))
    img_input = np.transpose(img_input, (2, 0, 1))
    img_input = img_input[np.newaxis, :]
    pred = model.run(None, {"img": img_input})[0].item()
    return f"{pred}"


if __name__ == "__main__":
    model_path = hf_hub_download(repo_id="skytnt/anime-aesthetic", filename="model.onnx")
    model = rt.InferenceSession(model_path, providers=['CUDAExecutionProvider', 'CPUExecutionProvider'])
    examples = [[f"examples/{x:02d}.jpg"] for x in range(0, 2)]
    app = gr.Interface(predict, gr.Image(label="input image"), gr.Textbox(label="score"),title="Anime Aesthetic Predict",
                       allow_flagging="never", examples=examples)
    app.launch()