app.py

# Import necessary libraries
import numpy as np
import joblib  # For loading the serialized model
import pandas as pd  # For data manipulation
from flask import Flask, request, jsonify  # For creating the Flask API
##from flask_cors import CORS

# Initialize the Flask application
rental_price_predictor_api = Flask("Airbnb Rental Price Predictor")
#CORS(rental_price_predictor_api)  # Enable CORS globally

# Load the trained machine learning model
model = joblib.load("rental_price_prediction_model_v1_0.joblib")

# Define a route for the home page (GET request)
@rental_price_predictor_api.get('/')
def home():
    return "Welcome to the Airbnb Rental Price Prediction API!"

# Endpoint for single property prediction
@rental_price_predictor_api.post('/v1/rental')
def predict_rental_price():
    property_data = request.get_json()

    sample = {
        'room_type': property_data['room_type'],
        'accommodates': property_data['accommodates'],
        'bathrooms': property_data['bathrooms'],
        'cancellation_policy': property_data['cancellation_policy'],
        'cleaning_fee': property_data['cleaning_fee'],
        'instant_bookable': property_data['instant_bookable'],
        'review_scores_rating': property_data['review_scores_rating'],
        'bedrooms': property_data['bedrooms'],
        'beds': property_data['beds']
    }

    input_data = pd.DataFrame([sample])

    predicted_log_price = model.predict(input_data)[0]
    predicted_price = round(float(np.exp(predicted_log_price)), 2)

    return jsonify({'Predicted Price (in dollars)': predicted_price})

# ✅ Fixed endpoint for batch prediction
@rental_price_predictor_api.post('/v1/rentalbatch')
def predict_rental_price_batch():
    """
    Expects a CSV file with one property per row.
    Returns JSON: a list of dicts with `id` and predicted price.
    """
    if 'file' not in request.files:
        return jsonify({"error": "No file uploaded"}), 400

    file = request.files['file']
    input_data = pd.read_csv(file)

    if 'id' not in input_data.columns:
        return jsonify({"error": "Missing 'id' column in uploaded CSV"}), 400

    predicted_log_prices = model.predict(input_data)
    predicted_prices = [round(float(np.exp(lp)), 2) for lp in predicted_log_prices]

    results = [
        {"id": pid, "Predicted Price (in dollars)": price}
        for pid, price in zip(input_data['id'], predicted_prices)
    ]

    return jsonify(results)

# Run the Flask application in debug mode if this script is executed directly
if __name__ == '__main__':
    rental_price_predictor_api.run(debug=True)