app.py

import os
import gradio as gr
from transformers import AutoImageProcessor, AutoModelForImageClassification
from PIL import Image
import torch
import numpy as np
import requests
import logging
from dotenv import load_dotenv  # Load .env file

# Load environment variables
load_dotenv()
HUGGINGFACE_API_KEY = os.getenv("HUGGINGFACE_API_KEY")

# Configure Logging
logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")

# Load Model & Processor
model_name = "linkanjarad/mobilenet_v2_1.0_224-plant-disease-identification"
try:
    processor = AutoImageProcessor.from_pretrained(model_name, use_fast=True)
    model = AutoModelForImageClassification.from_pretrained(model_name)
    logging.info("✅ Model and processor loaded successfully.")
except Exception as e:
    logging.error(f"❌ Failed to load model: {str(e)}")
    raise RuntimeError("Failed to load the model. Please check the logs for details.")

# Function to Get AI-Powered Treatment Suggestions
def get_treatment_suggestions(disease_name):
    url = "https://api-inference.huggingface.co/models/OpenAGI/agriculture-gpt"
    headers = {"Authorization": f"Bearer {HUGGINGFACE_API_KEY}"}
    data = {"inputs": f"What are the treatment options for {disease_name} in plants?"}

    try:
        response = requests.post(url, headers=headers, json=data)
        if response.status_code == 200:
            return response.json()[0]["generated_text"]
        else:
            return f"API Error: {response.status_code}"
    except Exception as e:
        return "Error retrieving treatment details."

# Define Prediction Function
def predict(image):
    try:
        image = Image.fromarray(np.uint8(image)).convert("RGB")
        inputs = processor(images=image, return_tensors="pt")
        with torch.no_grad():
            outputs = model(**inputs)
        logits = outputs.logits
        predicted_class_idx = logits.argmax(-1).item()
        predicted_label = model.config.id2label[predicted_class_idx]
        
        # Get AI-generated treatment suggestions
        treatment = get_treatment_suggestions(predicted_label)
        
        return f"Predicted Disease: {predicted_label}\nTreatment: {treatment}"
    except Exception as e:
        logging.error(f"Prediction failed: {str(e)}")
        return f"❌ Prediction failed: {str(e)}"

# Gradio Interface
iface = gr.Interface(
    fn=predict,
    inputs=gr.Image(type="numpy", label="Upload or capture plant image"),
    outputs=gr.Textbox(label="Result"),
    title="AI-Powered Plant Disease Detector",
    description="Upload a plant leaf image to detect diseases and get AI-powered treatment suggestions.",
    allow_flagging="never",
)

# Launch Gradio App
iface.launch()