main.py

# main.py
import pandas as pd
from data_processing import load_and_preprocess_data, group_transactions
from category_mapping import map_category
from utils import clean_category

def main():
    # Load and preprocess data
    train_data, test_data = load_and_preprocess_data()

    # Process and map categories
    grouped_transactions = group_transactions(train_data)
    
    # Replace 'your-api-key' with the actual GPT-3.5 API key
    gpt_api_key = 'from env'
    matched_transactions = map_category(test_data, grouped_transactions, gpt_api_key)

    # Calculate match percentage
    match_percentage = (sum(1 for t in matched_transactions if t.predicted_category is not None) / 220) * 100
    print(f"Match percentage on the new test set: {match_percentage:.2f}%")

    # Prepare and export results
    results = pd.DataFrame([{
        'date': t.date,
        'name': t.name,
        'memo': t.memo,
        'amount': t.amount,
        'transaction_type': t.transaction_type,
        'cleaned_memo': t.cleaned_memo,
        'cleaned_name': t.cleaned_name,
        'predicted_category': clean_category(t.predicted_category),
        'matching_steps': t.matching_steps,
        'MatchId': getattr(t, 'MatchId', None),
        'source': t.source,
        'normalized_source': t.normalized_source,
        'purpose': getattr(t, 'purpose', None),
        'generated_name': getattr(t, 'generated_name', None)
    } for t in matched_transactions])
    
    output_file_path = 'ctrlf3-match.xlsx'
    results.to_excel(output_file_path, index=False)
    print(f'File saved to {output_file_path}')

if __name__ == "__main__":
    main()