import gradio as gr
import numpy as np
import joblib

# ==============================
# LOAD MODELS
# ==============================
scaler   = joblib.load("scaler.pkl")
le_label = joblib.load("le_label.pkl")
le_type  = joblib.load("le_type.pkl")

rf_label_model = joblib.load("rf_label_model.pkl")
rf_type_model  = joblib.load("rf_type_model.pkl")

# ==============================
# PREDICTION FUNCTION
# ==============================
def predict(N, P, K, temperature, humidity, ph):
    sample = np.array([[N, P, K, temperature, humidity, ph]])
    sample_scaled = scaler.transform(sample)

    pred_label = le_label.inverse_transform(
        rf_label_model.predict(sample_scaled)
    )[0]

    pred_type = le_type.inverse_transform(
        rf_type_model.predict(sample_scaled)
    )[0]

    return f"🌱 Culture: {pred_label}\n🌍 Soil Type: {pred_type}"

# ==============================
# GRADIO INTERFACE
# ==============================
interface = gr.Interface(
    fn=predict,
    inputs=[
        gr.Number(label="Nitrogen (N)"),
        gr.Number(label="Phosphorus (P)"),
        gr.Number(label="Potassium (K)"),
        gr.Number(label="Temperature (°C)"),
        gr.Number(label="Humidity (%)"),
        gr.Number(label="pH"),
    ],
    outputs="text",
    title="Crop & Soil Prediction Model",
    description="Prediction using Random Forest model trained in Kaggle"
)

interface.launch()