data/app.py

import joblib
import pandas as pd
from flask import Flask, request, jsonify
import os

# Initialize Flask application
app = Flask(__name__)

# Define paths for model and data
MODEL_PATH = 'best_random_forest_model.joblib'
X_TRAIN_PATH = 'data/X_train.csv'

# Load the pre-trained model
try:
    model = joblib.load(MODEL_PATH)
    print(f"Model loaded successfully from {MODEL_PATH}")
except Exception as e:
    print(f"Error loading model: {e}")
    model = None # Set model to None if loading fails

# Load X_train.csv to get column names for consistent feature ordering
try:
    # Adjust path if running locally or in a different environment where current working directory might not be tourism_project
    # In a Docker container, it's expected to be at /app/data/X_train.csv
    X_train_columns = pd.read_csv(X_TRAIN_PATH).columns.tolist()
    print(f"X_train columns loaded successfully from {X_TRAIN_PATH}")
except Exception as e:
    print(f"Error loading X_train columns: {e}")
    X_train_columns = None # Set to None if loading fails

@app.route('/predict', methods=['POST'])
def predict():
    if model is None or X_train_columns is None:
        return jsonify({'error': 'Model or X_train columns not loaded correctly.'}), 500

    try:
        data = request.get_json(force=True)
        if not isinstance(data, list):
            data = [data] # Ensure data is a list of dictionaries for DataFrame conversion

        # Convert input data to pandas DataFrame
        input_df = pd.DataFrame(data)

        # Reindex the input DataFrame to ensure correct column order and handle missing columns
        # Fill any missing columns (e.g., from one-hot encoding) with 0
        # Exclude 'ProdTaken' if it somehow ends up in X_train_columns, although it shouldn't normally
        # For this specific case, X_train_columns should not contain 'ProdTaken' as it's the target
        
        # Filter X_train_columns to ensure no target variable is included
        feature_columns = [col for col in X_train_columns if col != 'ProdTaken']
        
        input_df = input_df.reindex(columns=feature_columns, fill_value=0)

        # Make predictions
        predictions = model.predict(input_df)
        probabilities = model.predict_proba(input_df)[:, 1] # Probability of the positive class

        # Prepare results
        results = []
        for i in range(len(predictions)):
            results.append({
                'prediction': int(predictions[i]),
                'probability': float(probabilities[i])
            })

        return jsonify(results)

    except Exception as e:
        return jsonify({'error': str(e)}), 400

if __name__ == '__main__':
    app.run(host='0.0.0.0', port=8000)