app.py

import gradio as gr
from PIL import Image
import pathlib
import os
import shutil
import zipfile
import huggingface_hub
from huggingface_hub import hf_hub_download
from autogluon.multimodal import MultiModalPredictor
import pandas as pd

# Model config
MODEL_REPO_ID = "cassieli226/sign-identification-automl"
ZIP_FILENAME = "autogluon_predictor_dir.zip"
CACHE_DIR = pathlib.Path("hf_assets")
EXTRACT_DIR = CACHE_DIR / "predictor_native"

# Download & load the native predictor
def _prepare_predictor_dir() -> str:
    CACHE_DIR.mkdir(parents=True, exist_ok=True)
    local_zip = hf_hub_download(
        repo_id=MODEL_REPO_ID,
        filename=ZIP_FILENAME,
        repo_type="model",
        local_dir=str(CACHE_DIR),
        local_dir_use_symlinks=False,
    )
    if EXTRACT_DIR.exists():
        shutil.rmtree(EXTRACT_DIR)
    EXTRACT_DIR.mkdir(parents=True, exist_ok=True)
    with zipfile.ZipFile(local_zip, "r") as zf:
        zf.extractall(str(EXTRACT_DIR))
    contents = list(EXTRACT_DIR.iterdir())
    predictor_root = contents[0] if (len(contents) == 1 and contents[0].is_dir()) else EXTRACT_DIR
    return str(predictor_root)

PREDICTOR_DIR = _prepare_predictor_dir()
PREDICTOR = MultiModalPredictor.load(PREDICTOR_DIR)

clf = PREDICTOR  # MultiModalPredictor

def predict_image(image, confidence_threshold=0.5, return_probabilities=False):
    if image is None:
        return "Please upload an image", ""

    try:
        # Wrap the image in a dict for MultiModalPredictor
        input_df = pd.DataFrame([{"image": image}])

        # Make prediction
        prediction = clf.predict(input_df)[0]
        detailed = f"**Predicted Class:** {prediction}"

        if return_probabilities:
            try:
                probs = clf.predict_proba(input_df)
                prob_dict = probs.iloc[0].to_dict()
                sorted_probs = sorted(prob_dict.items(), key=lambda x: x[1], reverse=True)
                detailed += "\n\n**Prediction Probabilities:**\n"
                for class_name, prob in sorted_probs[:5]:
                    if prob >= confidence_threshold:
                        detailed += f"- {class_name}: {prob:.3f}\n"
            except Exception as e:
                detailed += f"\n\nNote: Could not retrieve probabilities ({str(e)})"

        return prediction, detailed

    except Exception as e:
        return f"Error: {str(e)}", f"**Error:** {str(e)}"

# --- Gradio UI ---
with gr.Blocks(title="Image Classifier", theme=gr.themes.Soft()) as demo:
    gr.Markdown("""
    # 🖼️ AI Image Classifier

    ## Upload an image to get AI-powered classification results
    """)

    with gr.Row():
        with gr.Column():
            image_input = gr.Image(type="pil", label="Upload Image", height=300)
            gr.Markdown("### 🔧 Inference Parameters")
            confidence_threshold = gr.Slider(0.0, 1.0, value=0.1, step=0.05, label="Confidence Threshold")
            return_probabilities = gr.Checkbox(label="Show Prediction Probabilities", value=True)
            predict_btn = gr.Button("🎯 Classify Image", variant="primary", size="lg")

        with gr.Column():
            prediction_output = gr.Textbox(label="Prediction Result", interactive=False, lines=1)
            detailed_output = gr.Markdown(label="Detailed Analysis", value="Results will appear here...")

    predict_btn.click(
        fn=predict_image,
        inputs=[image_input, confidence_threshold, return_probabilities],
        outputs=[prediction_output, detailed_output]
    )

    gr.Markdown("""
    ---
    *Built with AutoGluon and Gradio*
    """)

if __name__ == "__main__":
    demo.launch()