app.py

import gradio as gr
import numpy as np
from tensorflow.keras.models import load_model
from PIL import Image
import requests
import os

MODEL_URL = "https://huggingface.co/liriope/PlantDiseaseDetection/resolve/main/model.h5"
MODEL_PATH = "model.h5"

# ✅ 1. Scarica il modello se non esiste ancora localmente
if not os.path.exists(MODEL_PATH):
    print("📥 Downloading model from Hugging Face...")
    response = requests.get(MODEL_URL)
    with open(MODEL_PATH, "wb") as f:
        f.write(response.content)
    print("✅ Model downloaded and saved locally!")

# ✅ 2. Carica il modello
model = load_model(MODEL_PATH)
print("✅ Model loaded successfully!")

# ✅ 3. Funzione di predizione
def predict(image):
    img = image.convert("RGB").resize((224, 224))
    arr = np.expand_dims(np.array(img) / 255.0, axis=0)
    preds = model.predict(arr)[0]
    class_idx = np.argmax(preds)
    confidence = float(preds[class_idx])
    return {f"Predicted class: {class_idx}": confidence}

# ✅ 4. Interfaccia Gradio
demo = gr.Interface(
    fn=predict,
    inputs=gr.Image(label="Upload a leaf photo"),
    outputs=gr.Label(num_top_classes=3),
    title="🌿 Plant Health Checker AI",
    description="Upload a photo of a leaf and detect potential plant diseases using a deep learning model.",
)

demo.launch()