app.py

import gradio as gr
import numpy as np
import joblib
from sklearn.preprocessing import StandardScaler 


model = joblib.load("final_rf_model.pkl")
scaler = StandardScaler()  

class_labels = {
    0: 'Insufficient Weight',
    1: 'Normal Weight',
    2: 'Overweight Level I',
    3: 'Overweight Level II',
    4: 'Obesity Type I',
    5: 'Obesity Type II',
    6: 'Obesity Type III'
}

# Function to make predictions
def predict_obesity(weight, height, age, fcvc, gender, ncp, ch2o, faf, tue, fhwow,
                    caec_sometimes, calc_no, calc_sometimes, caec_frequently,
                    alcohol_choice, mtrans_choice, favc):
    
    # Prepare input data for prediction
    input_data = np.array([[weight, height, age, fcvc, 1 if gender == "Male" else 0, ncp, ch2o, faf, tue, fhwow,
                             1 if caec_sometimes else 0, 1 if calc_no else 0, 1 if calc_sometimes else 0,
                             1 if caec_frequently else 0, 1 if alcohol_choice == "Yes" else 0,
                             1 if favc else 0, 1 if mtrans_choice == "Automobile" else 0]])

    # Scale the appropriate input values
    input_data[:, 0:4] = scaler.fit_transform(input_data[:, 0:4])  
    input_data[:, 5:10] = scaler.fit_transform(input_data[:, 5:10])  

    # Make prediction
    prediction = model.predict(input_data)

    # Map the numeric prediction to the corresponding label
    predicted_label = class_labels.get(prediction[0], "Unknown Class")
    return predicted_label

# Custom CSS for better styling
custom_css = """
<style>
    .gradio-container {
        background-color: #0a0a2c;
        background-image: 
            linear-gradient(45deg, #0a0a2c 25%, #12124a 25%, #12124a 50%, #0a0a2c 50%, #0a0a2c 75%, #12124a 75%, #12124a 100%);
        background-size: 56.57px 56.57px;
        border-radius: 15px;
        padding: 30px;
        box-shadow: 0 0 20px rgba(66, 220, 219, 0.3),
                    0 0 40px rgba(233, 30, 99, 0.2);
        border: 1px solid rgba(66, 220, 219, 0.3);
    }

    .title {
        font-family: 'Orbitron', sans-serif;
        font-size: 36px;
        font-weight: bold;
        color: #00fff2;
        text-align: center;
        margin-bottom: 30px;
        text-transform: uppercase;
        letter-spacing: 3px;
        text-shadow: 0 0 10px rgba(0, 255, 242, 0.5),
                     0 0 20px rgba(0, 255, 242, 0.3),
                     0 0 30px rgba(0, 255, 242, 0.1);
    }

    .description {
        font-family: 'Rajdhani', sans-serif;
        font-size: 18px;
        color: #b4f8fc;
        text-align: center;
        margin-bottom: 30px;
        line-height: 1.6;
        text-shadow: 0 0 5px rgba(180, 248, 252, 0.3);
    }

    /* Input fields styling */
    input[type="number"] {
        background-color: rgba(16, 16, 44, 0.9);
        border: 2px solid #00fff2;
        border-radius: 8px;
        padding: 12px;
        color: #fff;
        font-family: 'Rajdhani', sans-serif;
        transition: all 0.3s ease;
        box-shadow: 0 0 10px rgba(0, 255, 242, 0.2);
    }

    input[type="number"]:focus {
        border-color: #ff2e63;
        box-shadow: 0 0 15px rgba(255, 46, 99, 0.4);
        outline: none;
    }

    /* Radio and Checkbox styling */
    input[type="radio"],
    input[type="checkbox"] {
        accent-color: #ff2e63;
    }

    .input-container label {
        color: #b4f8fc;
        font-family: 'Rajdhani', sans-serif;
        font-size: 16px;
        margin-bottom: 8px;
        display: block;
    }

    /* Button styling */
    button {
        background: linear-gradient(45deg, #ff2e63, #00fff2);
        color: #fff;
        border: none;
        padding: 15px 30px;
        border-radius: 8px;
        cursor: pointer;
        font-family: 'Orbitron', sans-serif;
        font-size: 18px;
        text-transform: uppercase;
        letter-spacing: 2px;
        transition: all 0.3s ease;
        box-shadow: 0 0 15px rgba(255, 46, 99, 0.3),
                    0 0 30px rgba(0, 255, 242, 0.2);
    }

    button:hover {
        transform: translateY(-2px);
        box-shadow: 0 0 20px rgba(255, 46, 99, 0.5),
                    0 0 40px rgba(0, 255, 242, 0.3);
    }

    /* Output label styling */
    .output-label {
        background: rgba(16, 16, 44, 0.9);
        border: 2px solid #ff2e63;
        border-radius: 8px;
        padding: 20px;
        color: #00fff2;
        font-family: 'Orbitron', sans-serif;
        font-size: 24px;
        text-align: center;
        margin-top: 20px;
        box-shadow: 0 0 15px rgba(255, 46, 99, 0.3);
    }

    /* Add cyberpunk grid lines to the background */
    .gradio-container::before {
        content: '';
        position: absolute;
        top: 0;
        left: 0;
        right: 0;
        bottom: 0;
        background: 
            linear-gradient(90deg, rgba(66, 220, 219, 0.1) 1px, transparent 1px),
            linear-gradient(0deg, rgba(66, 220, 219, 0.1) 1px, transparent 1px);
        background-size: 20px 20px;
        pointer-events: none;
    }

    /* Add some hover effects to input containers */
    .input-container:hover {
        transform: translateX(5px);
        transition: transform 0.3s ease;
    }

    /* Scrollbar styling */
    ::-webkit-scrollbar {
        width: 10px;
        background: #0a0a2c;
    }

    ::-webkit-scrollbar-thumb {
        background: linear-gradient(45deg, #ff2e63, #00fff2);
        border-radius: 5px;
    }
</style>
"""

# Gradio interface
iface = gr.Interface(
    fn=predict_obesity,
    inputs=[
        gr.Number(label="Weight (40-160 kg)"),
        gr.Number(label="Height (1-2 m)"),
        gr.Number(label="Age (10-60 years)"),
        gr.Number(label="FCVC (Frequency of Vegetable Consumption 1-4)"),
        gr.Radio(choices=["Male", "Female"], label="Gender"),
        gr.Number(label="NCP (Number of meals per day 1-3)"),
        gr.Number(label="CH2O (Water Consumption 1-3)"),
        gr.Number(label="FAF (Physical Activity Frequency 1-4)"),
        gr.Number(label="TUE (Time Spent on Exercise 1-4)"),
        gr.Number(label="FHWOW (Family History with OverWeight)"),
        gr.Radio(choices=["No", "Sometimes", "Frequently"], label="Alcohol Consumption"),
        gr.Radio(choices=["Public Transportation", "Automobile"], label="Transportation Method"),
        gr.Checkbox(label="FAVC (Frequent Consumption of High-Calorie Foods)"),
    ],
    outputs=gr.Label(label="Predicted Obesity Level"),
    title="Obesity Level Estimator",
    description="Enter the features related to eating habits and physical condition to estimate obesity levels.",
    css=custom_css
)

# Launch the interface
iface.launch(share=True)