performance.py

import yaml
import matplotlib.pyplot as plt


# rename
def calculate_performance():
    # Load data from YAML file
    yml_path = "./performance.yml"
    with open(yml_path, "r") as file:
        performances = yaml.safe_load(file)

    # Extract data for each metric
    f1_data = performances['f1']
    mae_data = performances['mae']

    # Sort metrics based on scores
    sorted_f1_data = sorted(f1_data.items(), key=lambda x: x[1])
    sorted_mae_data = sorted(mae_data.items(), key=lambda x: x[1], reverse=True)

    # Extract sorted keys and values
    f1_keys = [item[0] for item in sorted_f1_data]
    f1_values = [item[1] for item in sorted_f1_data]

    mae_keys = [item[0] for item in sorted_mae_data]
    mae_values = [item[1] for item in sorted_mae_data]

    # Plotting
    fig, axs = plt.subplots(1, 2, figsize=(12, 6))

    # Subplot 1 for F1 metrics
    axs[0].barh(f1_keys, f1_values, color='skyblue')
    axs[0].set_xlabel('F1 Scores')
    axs[0].set_title('F1 Metrics')
    axs[0].grid(True)  # Add grid lines

    # Subplot 2 for MAE metrics
    axs[1].barh(mae_keys, mae_values, color='salmon')
    axs[1].set_xlabel('MAE Scores')
    axs[1].set_title('MAE Metrics')
    axs[1].grid(True)  # Add grid lines

    plt.tight_layout()
    plt.savefig("performance.pdf")