app.py

"""
Gradio web app for Crop Recommendation System
Deploy to Hugging Face Spaces
"""

import json
import joblib
import numpy as np
import pandas as pd
import gradio as gr
from pathlib import Path

# Load model and metadata
MODEL_PATH = "model.pkl"
META_PATH = "model_meta.json"

def load_model_and_metadata():
    """Load trained model and metadata."""
    if not Path(MODEL_PATH).exists():
        raise FileNotFoundError(f"Model not found at {MODEL_PATH}")
    if not Path(META_PATH).exists():
        raise FileNotFoundError(f"Metadata not found at {META_PATH}")
    
    model = joblib.load(MODEL_PATH)
    with open(META_PATH) as f:
        meta = json.load(f)
    
    return model, meta

# Load model once at startup
try:
    model, meta = load_model_and_metadata()
    numeric_cols = meta["numeric_cols"]
    label_classes = meta["label_classes"]
except Exception as e:
    raise RuntimeError(f"Failed to load model: {e}")

def predict_crop(N, P, K, temperature, humidity, ph, rainfall):
    """
    Predict crop recommendation based on soil and weather parameters.
    
    Args:
        N: Nitrogen content in soil (0-140)
        P: Phosphorus content in soil (5-145)
        K: Potassium content in soil (5-205)
        temperature: Average temperature in Celsius (8-44)
        humidity: Relative humidity in % (14-100)
        ph: pH of soil (3.5-10)
        rainfall: Annual rainfall in mm (20-3000)
    
    Returns:
        Tuple of (recommended_crop, confidence, top_3_dict)
    """
    # Create input DataFrame
    input_data = pd.DataFrame({
        'N': [N],
        'P': [P],
        'K': [K],
        'temperature': [temperature],
        'humidity': [humidity],
        'ph': [ph],
        'rainfall': [rainfall]
    })
    
    # Predict
    prediction_encoded = model.predict(input_data)[0]
    probabilities = model.predict_proba(input_data)[0]
    
    # Map to crop name
    recommended_crop = label_classes[prediction_encoded]
    confidence = probabilities[prediction_encoded]
    
    # Get top 3
    top_3_indices = np.argsort(probabilities)[::-1][:3]
    top_3_dict = {}
    for rank, idx in enumerate(top_3_indices, 1):
        crop = label_classes[idx]
        conf = probabilities[idx]
        top_3_dict[f"{rank}. {crop}"] = f"{conf:.2%}"
    
    return recommended_crop, f"{confidence:.2%}", top_3_dict

# Create Gradio interface
with gr.Blocks(title="Crop Recommendation System", theme=gr.themes.Soft()) as demo:
    gr.Markdown("""
    # 🌾 Crop Recommendation System
    
    Enter soil and weather parameters to get AI-powered crop recommendations with confidence scores.
    """)
    
    with gr.Row():
        with gr.Column(scale=1):
            gr.Markdown("### Soil Parameters")
            N = gr.Slider(label="Nitrogen (N)", minimum=0, maximum=140, value=90, step=1)
            P = gr.Slider(label="Phosphorus (P)", minimum=5, maximum=145, value=42, step=1)
            K = gr.Slider(label="Potassium (K)", minimum=5, maximum=205, value=43, step=1)
        
        with gr.Column(scale=1):
            gr.Markdown("### Weather Parameters")
            temperature = gr.Slider(label="Temperature (°C)", minimum=8, maximum=44, value=21, step=0.1)
            humidity = gr.Slider(label="Humidity (%)", minimum=14, maximum=100, value=82, step=1)
            ph = gr.Slider(label="Soil pH", minimum=3.5, maximum=10, value=6.5, step=0.1)
            rainfall = gr.Slider(label="Annual Rainfall (mm)", minimum=20, maximum=3000, value=203, step=10)
    
    predict_btn = gr.Button("🔍 Get Crop Recommendation", variant="primary", size="lg")
    
    with gr.Row():
        with gr.Column(scale=2):
            recommended = gr.Textbox(
                label="🌾 Recommended Crop",
                interactive=False,
                text_align="center",
                scale=1
            )
            confidence = gr.Textbox(
                label="✅ Confidence",
                interactive=False,
                text_align="center",
                scale=1
            )
        
        with gr.Column(scale=1):
            top_3 = gr.JSON(label="📈 Top 3 Recommendations", interactive=False)
    
    predict_btn.click(
        fn=predict_crop,
        inputs=[N, P, K, temperature, humidity, ph, rainfall],
        outputs=[recommended, confidence, top_3]
    )
    
    gr.Markdown("""
    ---
    ### Parameter Ranges (based on training data)
    - **Nitrogen (N)**: 0-140 kg/ha
    - **Phosphorus (P)**: 5-145 kg/ha
    - **Potassium (K)**: 5-205 kg/ha
    - **Temperature**: 8-44°C
    - **Humidity**: 14-100%
    - **pH**: 3.5-10
    - **Rainfall**: 20-3000 mm/year
    """)

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