result.py

import argparse
import pandas as pd
import joblib
from sklearn.metrics import f1_score, precision_score, recall_score, classification_report, confusion_matrix
import json


def evaluate_results():
    dataset_path = "data/train.csv"
    model_path = "model/best_model.pkl"
    
    # Load dataset and model
    df = pd.read_csv(dataset_path)
    model = joblib.load(model_path)

    features = [
        'product_category_1',
        'product_category_2',
        'user_depth',
        'age_level',
        'city_development_index',
        'var_1',
        'gender'
    ]

    # Prepare the data
    X = df[features]
    y = df['is_click']
    X = pd.get_dummies(X, columns=['gender'], drop_first=True)

    # Train-test split
    _, X_test, _, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

    # Predict
    y_pred = model.predict(X_test)

    # Calculate metrics
    f1 = f1_score(y_test, y_pred)
    precision = precision_score(y_test, y_pred, zero_division=True)
    recall = recall_score(y_test, y_pred)

    # Save results
    metrics = {
        'f1_score': f1,
        'precision': precision,
        'recall': recall
    }

    with open('results/metrics.json', 'w') as f:
        json.dump(metrics, f, indent=4)

    # Classification report and confusion matrix
    report = classification_report(y_test, y_pred)
    matrix = confusion_matrix(y_test, y_pred)

    with open('results/classification_report.txt', 'w') as f:
        f.write(report)

    with open('results/confusion_matrix.txt', 'w') as f:
        f.write(str(matrix))

    print("Results saved in the 'results' folder.")

    
# def main():
#     parser = argparse.ArgumentParser()
#     parser.add_argument('--results-path', type=str, required=True, help='Path to results file')
#     args = parser.parse_args()

#     print(f"Analyzing results from {args.results_path}")
#     # Add analysis logic here

# if __name__ == '__main__':
#     main()