KAD/zad1/main.py

"""
Komputerowa analiza danych
Zadanie 1
Michał Leśniak 195642
"""
import math
import locale
locale.setlocale(locale.LC_ALL, '')

SPECIES_NAMES = {
    '0': 'setosa',
    '1': 'versicolor',
    '2': 'virginica'
}

ATTRIBUTE_NAMES = ['Długość działki kielicha [cm]',
                   'Szerokość działki kielicha [cm]',
                   'Długość płatka [cm]',
                   'Szerokość płatka [cm]']


def mean(lst):
    sum = 0
    for i in lst:
        sum += i
    return sum/len(lst)


def median(lst, is_even):
    return (lst[len(lst)//2-1]+lst[len(lst)//2])/2 if is_even else lst[len(lst)//2]


def sample_standard_deviation(lst, lst_mean):
    sum = 0
    for x in [(y-lst_mean)**2 for y in lst]:
        sum += x
    return math.sqrt(sum/(len(lst)-1))


def q1(lst, is_even, lst_median):
    idx = len(lst)//2-1 if is_even else len(lst)//2+1
    sorted_list = lst[:idx]
    if is_even:
        sorted_list.append(lst_median)

    return median(sorted_list, is_even)


def q3(lst, is_even, lst_median):
    idx = len(lst)//2+1 if is_even else len(lst)//2
    sorted_list = lst[idx:]
    if is_even:
        sorted_list.insert(0, lst_median)

    return median(sorted_list, is_even)


def min(lst):
    return lst[0]


def max(lst):
    return lst[-1]


def calc_data(lst):
    is_even = len(lst) % 2 == 0
    lst_mean = mean(lst)
    lst_median = median(lst, is_even)
    return min(lst), lst_mean, sample_standard_deviation(lst, lst_mean), lst_median, q1(lst, is_even, lst_median), q3(lst, is_even, lst_median), max(lst)


def calc_species_data(species):
    total = 0
    for x in species.values():
        total += x

    species_list = []
    for key in species:
        species_list.append(
            (SPECIES_NAMES[key], species[key], species[key]/total))

    species_list.append(("Razem", total, 1.0))

    return species_list


def percentage_format(x):
    return locale.format_string('%.1f%%', x*100)


def float_format(x):
    return locale.format_string('%.2f', x)


def print_table(table):
    if not table:
        return
    cell_sizes = [0] * len(table[0])
    for i in range(len(table[0])):
        for x in table:
            if not x:
                continue
            if len(x[i]) > cell_sizes[i]:
                cell_sizes[i] = len(x[i])

    header_row_format = ' | '.join((f'{{:^{x}}}' for x in cell_sizes))
    row_format = f'{{:<{cell_sizes[0]}}} | ' + \
        ' | '.join((f'{{:^{x}}}' for x in cell_sizes[1:]))

    lines = []
    for x in table:
        if x == table[0]:
            lines.append(header_row_format.format(*x))
            continue
        if x is None:
            lines.append(None)
            continue
        lines.append(row_format.format(*x))
    row_size = 0
    for x in lines:
        if x and len(x) > row_size:
            row_size = len(x)
    row_size += 1
    lines.insert(1, '='*row_size)
    lines.insert(0, '='*row_size)
    lines.insert(len(lines), '='*row_size)
    for i in range(len(lines)):
        if lines[i] is None:
            lines[i] = '-'*row_size
    for x in lines:
        print(x)


def print_data(species_data, trait_data):
    species_printable_table = []
    species_printable_table.append(("Gatunek", "Liczebność (%)"))
    for x in species_data:
        if x == species_data[-1]:
            species_printable_table.append(None)
        species_printable_table.append(
            (x[0], "{} ({})".format(x[1], percentage_format(x[2]))))

    trait_printable_table = []
    trait_printable_table.append(
        ("Cecha", "Minimum", "Śr. arytm. (± odch. stand.)", "Mediana (Q1 - Q3)", "Maksimum"))
    for i in range(len(trait_data)):
        trait_printable_table.append((ATTRIBUTE_NAMES[i],
                                      float_format(trait_data[i][0]),
                                      "{} (±{})".format(float_format(
                                          trait_data[i][1]), float_format(trait_data[i][2])),
                                      "{} ({} - {})".format(
                                          float_format(trait_data[i][3]),
                                          float_format(trait_data[i][4]),
                                          float_format(trait_data[i][5])),
                                      float_format(trait_data[i][6])))

    print_table(species_printable_table)
    print()
    print_table(trait_printable_table)


def main():
    # read data
    with open("data.csv", "r") as f:
        data = f.read().splitlines()

    # parse data
    sepal_length_list = []
    sepal_width_list = []
    petal_length_list = []
    petal_width_list = []
    species_dict = {
        '0': 0,
        '1': 0,
        '2': 0,
    }
    for line in data:
        sepal_length, sepal_width, petal_length, petal_width, species = line.split(
            ',')
        sepal_length_list.append(float(sepal_length))
        sepal_width_list.append(float(sepal_width))
        petal_length_list.append(float(petal_length))
        petal_width_list.append(float(petal_width))
        species_dict[species] += 1

    sepal_length_list.sort()
    sepal_width_list.sort()
    petal_length_list.sort()
    petal_width_list.sort()

    # calculate results
    species_data = calc_species_data(species_dict)

    trait_data = []

    for lst in [sepal_length_list, sepal_width_list, petal_length_list, petal_width_list]:
        trait_data.append(calc_data(lst))

    # print results
    print_data(species_data, trait_data)


if __name__ == "__main__":
    import os.path
    assert os.path.isfile('data.csv')
    main()