app.py

# gradio app for Insurance Charges Prediction

import gradio as gr
import pandas as pd
import pickle
import numpy as np

# =====================
# 1. Load the trained model
# =====================
with open("insurance_model.pkl", "rb") as f:
    model = pickle.load(f)

# =====================
# 2. Helper function for BMI category
# =====================
def bmi_category(bmi):
    if bmi < 18.5:
        return "Underweight"
    elif bmi < 25:
        return "Normal"
    elif bmi < 30:
        return "Overweight"
    else:
        return "Obese"

# =====================
# 3. Prediction function
# =====================
def predict_charges(age, bmi, children, sex, smoker, region):
    
    # Compute BMI category
    bmi_cat = bmi_category(bmi)
    
    # Pack into a DataFrame
    input_df = pd.DataFrame([[
        age, bmi, children, sex, smoker, region, bmi_cat
    ]],
    columns=["age", "bmi", "children", "sex", "smoker", "region", "bmi_category"])
    
    # Predict charges
    prediction = model.predict(input_df)[0]
    
    return f"Predicted Insurance Charges: ${prediction:,.2f}"

# =====================
# 4. Gradio interface
# =====================
inputs = [
    gr.Number(label="Age", value=30, precision=0),
    gr.Number(label="BMI", value=25.0, precision=1),
    gr.Number(label="Number of Children", value=0, precision=0),
    gr.Dropdown(["male", "female"], label="Sex"),
    gr.Dropdown(["yes", "no"], label="Smoker"),
    gr.Dropdown(["southwest", "southeast", "northwest", "northeast"], label="Region")
]

app = gr.Interface(
    fn=predict_charges,
    inputs=inputs,
    outputs="text",
    title="Medical Insurance Charges Predictor",
    description="Enter patient details to predict expected medical insurance charges."
)

if __name__ == "__main__":
    # Launch with share=True for a public URL
    app.launch(share=True)