"""
EDEN Plant Disease Detection - Gradio Interface
"""
import gradio as gr
import torch
from PIL import Image
import json
from torchvision import transforms
from modeling import load_model

# Load model and labels
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model, labels_data = load_model(".", device)
idx_to_disease = labels_data["idx_to_disease"]

# Define preprocessing
transform = transforms.Compose([
    transforms.Resize((256, 256)),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])

def predict_disease(image):
    """Predict plant disease from image"""
    if image is None:
        return {}

    # Preprocess
    img_tensor = transform(image).unsqueeze(0).to(device)

    # Inference
    with torch.no_grad():
        outputs = model(img_tensor)
        probabilities = torch.softmax(outputs, dim=1)[0]

    # Get top 5 predictions
    top5_prob, top5_idx = torch.topk(probabilities, 5)

    results = {}
    for prob, idx in zip(top5_prob, top5_idx):
        disease = idx_to_disease[str(idx.item())]
        results[disease] = prob.item()

    return results

# Create Gradio interface
title = " EDEN Plant Disease Detection"
description = """
Detect plant diseases using AI-powered image analysis.
Upload an image of a plant leaf to identify potential diseases.

**Model**: Mob-Res (MobileNetV2 + ResNet hybrid)
**Accuracy**: 99.4%
**Diseases**: Multiple common plant diseases
"""

demo = gr.Interface(
    fn=predict_disease,
    inputs=gr.Image(type="pil", label="Upload Plant Image"),
    outputs=gr.Label(num_top_classes=5, label="Disease Predictions"),
    title=title,
    description=description,
    examples=None
)

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