bayesian_goty.py

import re

from compute_bayesian_rating import choose_prior, compute_bayesian_score


def load_input(filename, file_encoding='utf8'):
    data = []

    with open(filename, encoding=file_encoding) as f:
        for line in f.readlines():
            line = line.strip()
            # Remove empty lines and comments
            if len(line) > 0 and line[0:2] != '# ':
                data.append(line)

    return data


def parse_data(data):
    observations = {}

    for element in data:
        my_list = element.rsplit('(')
        if not (len(my_list) == 2):
            raise AssertionError()

        # Split at '|' to have the rank of the game at the head of the list
        first_part = my_list[0].split('|')
        # Remove the rank of the game
        game_name_list = first_part[1:]
        # Concatenate the remaining elements with '|'
        game_name = '|'.join(game_name_list)  # Remove leading and trailing whitespaces
        game_name = game_name.strip()

        second_part = my_list[1]
        # Reference: https://stackoverflow.com/a/1059601
        tokens = re.split('\\W+', second_part)
        rating_sum = int(tokens[1 + tokens.index('Score')])
        num_votes = int(tokens[1 + tokens.index('Votes')])

        observations[game_name] = {}
        observations[game_name]['score'] = rating_sum / num_votes
        observations[game_name]['num_votes'] = num_votes

    return observations


def compute_ranking(observations, prior):
    for game_name, game in observations.items():
        observations[game_name]['bayesian_score'] = compute_bayesian_score(game, prior)

    ranking = sorted(
        observations.keys(),
        key=lambda x: observations[x]['bayesian_score'],
        reverse=True,
    )

    return ranking


def print_ranking(ranking, observations, prior):
    print('Game of the Year Votes\n')

    for rank, entry in enumerate(ranking):
        game_name = entry.strip()
        sum_scores = (
            observations[game_name]['score'] * observations[game_name]['num_votes']
        )
        votes = observations[game_name]['num_votes']
        aps = observations[game_name]['score']
        bayesian_score = observations[game_name]['bayesian_score']

        sentence = '{0:3} | {1} (Score: {2:.0f} | Votes: {3} | APS: {4:.2f} | Bayesian Rating: {5:.2f})'
        print(
            sentence.format(
                rank + 1,
                game_name,
                sum_scores,
                votes,
                aps,
                bayesian_score,
            ),
        )

    sentence = '\nNB: Bayesian Prior (Score: {0:.2f} | Votes: {1:.2f})'
    print(sentence.format(prior['score'], prior['num_votes']))

    return


def main():
    filename = 'data/steam_resetera_2017_goty.txt'

    data = load_input(filename)

    observations = parse_data(data)

    verbose = False
    prior = choose_prior(observations, verbose)

    ranking = compute_ranking(observations, prior)

    print_ranking(ranking, observations, prior)

    return True


if __name__ == '__main__':
    main()