app.py

import pandas as pd
import pickle
import gradio as gr
from sklearn.preprocessing import LabelEncoder
import os

# --- Load dataset ---
df_original = pd.read_csv('AB_ELEVATORS_DATASET_final.csv')

categorical_cols_original = [
    'BRAND NAME', 'LIFT_TYPE', 'DOOR_FRAME', 'DOOR_TYPE', 'CABIN_MATERIAL', 'MOTOR_TYPE'
]

# --- Create encoders ---
encoders = {}
for col in categorical_cols_original:
    le = LabelEncoder()
    le.fit(df_original[col])
    encoders[col] = le

categorical_options = {
    col: list(df_original[col].unique()) for col in categorical_cols_original
}

# --- Load model ---
with open("elevator_model.pkl", "rb") as f:
    model = pickle.load(f)

# --- Prediction function ---
def predict_price(
    brand_name_str,
    lift_type_str,
    door_frame_str,
    door_type_str,
    cabin_material_str,
    motor_type_str,
    passengers_capacity,
    speed_mps,
    floors
):
    try:
        # Encode inputs
        input_data = pd.DataFrame([[ 
            encoders['BRAND NAME'].transform([brand_name_str])[0],
            encoders['LIFT_TYPE'].transform([lift_type_str])[0],
            encoders['DOOR_FRAME'].transform([door_frame_str])[0],
            encoders['DOOR_TYPE'].transform([door_type_str])[0],
            encoders['CABIN_MATERIAL'].transform([cabin_material_str])[0],
            encoders['MOTOR_TYPE'].transform([motor_type_str])[0],
            int(passengers_capacity),
            float(speed_mps),
            int(floors)
        ]], columns=[
            'brand_name', 'lift_type', 'door_frame', 'door_type',
            'cabin_material', 'motor_type',
            'passengers_capacity', 'speed_mps', 'floors'
        ])

        prediction = model.predict(input_data)[0]
        price = float(prediction)

        return (
            f"### 💰 Estimated Price: ₹ {price:,.2f}",
            "✅ Quote generated successfully"
        )

    except Exception as e:
        return f"❌ Error: {str(e)}", "⚠️ Please check inputs"

#Function to get image path from dropdown
def get_image_path(category, selection):
    if selection is None:
        return None
    
    # Convert dropdown value → UPPERCASE filename
    filename = selection.upper().replace(" ", "_") + ".jpg"
    
    path = os.path.join("Images", category, filename)
    
    return path if os.path.exists(path) else None

# --- Custom UI using Blocks ---
with gr.Blocks(theme=gr.themes.Soft(), title="AB Elevators") as demo:

    gr.Markdown("""
    <div style="text-align:center; padding:20px;">
        <h1 style="color:gold; text-shadow:2px 2px 5px black;">
            🚀 AB ELEVATORS
        </h1>
        <h3>Smart Lift Price Prediction System</h3>
        <p>Get instant quotations based on your requirements</p>
    </div>
    """)


    with gr.Row():
        with gr.Column():
            brand = gr.Dropdown(categorical_options['BRAND NAME'], label="🏢 Brand")
            lift = gr.Dropdown(categorical_options['LIFT_TYPE'], label="🛗 Lift Type")
            lift_img = gr.Image(label="Lift Type Preview", height=150) #Lift_Type Image Display
            door_frame = gr.Dropdown(categorical_options['DOOR_FRAME'], label="🚪 Door Frame")

        with gr.Column():
            door_type = gr.Dropdown(categorical_options['DOOR_TYPE'], label="🚪 Door Type")
            door_img = gr.Image(label="Door Type Preview", height=150)
            cabin = gr.Dropdown(categorical_options['CABIN_MATERIAL'], label="🧱 Cabin Material")
            motor = gr.Dropdown(categorical_options['MOTOR_TYPE'], label="⚙️ Motor Type")
            motor_img = gr.Image(label="Motor Type Preview", height=150)

    gr.Markdown("### 🔢 Configuration")

    with gr.Row():
        passengers = gr.Slider(
            minimum=int(df_original['PASSENGERS CAPACITY'].min()),
            maximum=int(df_original['PASSENGERS CAPACITY'].max()),
            step=1,
            label="👥 Passengers Capacity",
            value=int(df_original['PASSENGERS CAPACITY'].mean())
        )

        speed = gr.Slider(
            minimum=float(df_original['SPEED_MPS'].min()),
            maximum=float(df_original['SPEED_MPS'].max()),
            step=0.1,
            label="⚡ Speed (MPS)",
            value=float(df_original['SPEED_MPS'].mean())
        )

        floors = gr.Slider(
            minimum=int(df_original['FLOORS'].min()),
            maximum=int(df_original['FLOORS'].max()),
            step=1,
            label="🏢 Floors",
            value=int(df_original['FLOORS'].mean())
        )

    predict_btn = gr.Button("💡 Generate Quote", variant="primary")

    output_price = gr.Markdown()
    output_status = gr.Markdown()

    # --- Image Updates ---
    lift.change(
        fn=lambda x: get_image_path("Lift_Type", x),
        inputs=lift,
        outputs=lift_img
    )

    door_type.change(
        fn=lambda x: get_image_path("Door_Type", x),
        inputs=door_type,
        outputs=door_img
    )

    motor.change(
        fn=lambda x: get_image_path("Motor_Type", x),
        inputs=motor,
        outputs=motor_img
    )    
    
    predict_btn.click(
        fn=predict_price,
        inputs=[brand, lift, door_frame, door_type, cabin, motor,
                passengers, speed, floors],
        outputs=[output_price, output_status]
    )

    gr.Markdown("""
    <hr>
    <div style="text-align:center;">
        <p>✨ Designed for AB Elevators | AI Powered Pricing</p>
    </div>
    """)

# --- Launch ---
demo.launch()