Read and plot Learning Mall coursework results from JSON

read_lm.py

"""
Read and plot Learning Mall coursework results from JSON
========================================================
"""

import json
import re

import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns


RESPONSE_KEY_START = "q"
DURATION_KEY = "duration"
GRADE_KEY = "grade10000"

DELETE = [("lastname", "Overall average")]
REGEX = r"(?:(\d+)\s*hour)?\s*(?:(\d+)\s+mins)?\s*(?:(\d+)\s+secs)?"


def count_questions(data):
    return len([k for k in data[0] if k.startswith(RESPONSE_KEY_START)])


def number_correct_answers(entry, n):
    return int(float(entry[GRADE_KEY]) * n / 100.)


def time_in_minutes(entry):
    match = re.match(REGEX, entry[DURATION_KEY])
    hours, minutes, seconds = match.groups(default=0)
    return float(hours) * 60. + float(minutes) + float(seconds) / 60.


def keep(entry):
    for k, v in DELETE:
        if entry[k] == v:
            return False
    return True


def json_to_df(json_file_name):

    with open(json_file_name) as f:
        data = json.load(f)[0]

    data = [e for e in data if keep(e)]

    n = count_questions(data)

    for e in data:
        e["number_correct_answers"] = number_correct_answers(e, n)
        e["time_in_minutes"] = time_in_minutes(e)

    return pd.DataFrame(data)


def summary_plot(df):

    fig, ax = plt.subplots(3)

    sns.histplot(df, x="number_correct_answers", hue="department", stat="percent", multiple="stack", discrete=True, shrink=0.5, ax=ax[0])
    sns.histplot(df, x="time_in_minutes", hue="department", multiple="stack", shrink=0.5, ax=ax[1])
    sns.scatterplot(df, x="time_in_minutes", y="number_correct_answers", hue="department", ax=ax[2])

    return fig, ax


if __name__ == "__main__":

    import sys

    json_file_name = sys.argv[1]
    df = json_to_df(json_file_name)

    summary_plot(df)
    plt.show()