load_datasets.py

import pandas as pd

def load_cancer():
    names = ['id'] + ['fut' + str(i) for i in range(1, 10)] +  ['malignant']
    df = pd.read_csv('data/cancer/breast-cancer-wisconsin.data.cleaned', sep=',', names=names)
    df.drop(columns=['id'], inplace=True)
    df['malignant'] = df['malignant'] == 4
    return df

def load_wine():
    df1 = pd.read_csv('data/wine/winequality-red.csv', sep=';')
#    df1.insert(0, 'type', 1)
    df2 = pd.read_csv('data/wine/winequality-white.csv', sep=';')
#    df2.insert(0, 'type', 2)
    df1['red'] = 1
    df2['red'] = 0
    df = df1.append(df2, ignore_index=True)
#    df = df1
#    df['quality'] = df['quality'] > 5
    return df

def load_mushroom():
    names = ['classification'] + [f'f{i}' for i in range(1,23)]
    df = pd.read_csv('data/mushroom/agaricus-lepiota.data', names=names)
    df.drop(columns=['f11'], inplace=True)
#    df.drop(columns=[f'f{i}' for i in range(6,23)], inplace=True)
    df = df.astype("category")
    for col in df:
        df[col] = df[col].cat.codes
    df['classification'] = df.pop('classification')
    return df

if __name__=='__main__':
#    print(load_cancer())
    print(load_wine())
#    print(load_mushroom())