histogram.py

import pandas as pd
import matplotlib.pyplot as plt

from src.Compute import Compute
from src.min_max_scaler import fit_transform

def main():
    """
    Read a dataset, normalize numerical columns, and analyze score distributions across Hogwarts houses.
    The score distributions are also analyzed by computing standard deviation of scores of each course.
    The most homogeneous score is the smallest standard deviation
    Plots histograms to visualize score distributions for each course.

    Raises:
        Exception: if error occurs during file reading or computation.

    """
    try:
        dataset = pd.read_csv('./datasets/dataset_train.csv')
        houses = dataset['Hogwarts House'].unique()

        # Normalize dataset
        numerical_columns = dataset.columns[6:]
        normalized_data = fit_transform(dataset[numerical_columns])
        dataset[numerical_columns] = normalized_data

        # Compute how scores spread out for each course
        courses_std = []
        for i, column in enumerate(dataset.columns[6:]):
            courses_std.append(Compute.std(dataset[column].dropna()))
            print(f"{column}: {courses_std[i]}")

        min_std_index = courses_std.index(Compute.min(courses_std))
        print(f"{dataset.columns[6:][min_std_index]} has the most homogeneous score distribution between all four houses.")

        color_map = {
            'Gryffindor': 'red',
            'Hufflepuff': 'yellow',
            'Ravenclaw': 'blue',
            'Slytherin': 'green'
        }

        # plot histograms
        fig, axs = plt.subplots(nrows=2, ncols=7, figsize=(25, 6))
        axs = axs.flatten()

        for i, column in enumerate(dataset.columns[6:]):

            for house in houses:
                category_values = dataset[dataset['Hogwarts House'] == house][column]
                axs[i].hist(category_values, alpha=0.5, density=True, color=color_map[house])

            axs[i].set_title(f'{column}')
            axs[i].set_xlabel('Score')
            axs[i].set_ylabel('Frequency')

        fig.delaxes(axs[13])
        fig.legend(houses, loc='lower right')
        plt.tight_layout()
        plt.show()
    except Exception as error:
        print(f"error: {error}")

if __name__ == "__main__":
    main()