Untitled

import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
from sklearn.model_selection import RepeatedKFold

# Klasyfikatory
from sklearn.naive_bayes import GaussianNB
from sklearn.linear_model import LogisticRegression

matrix_labels = ['gaussian_nb', 'logistic_regression']
datasets = ["bupa", "coil2000", "cryotherapy", "ecoli4", "german", "glass2", "glass4", "glass5", "haberman", "heart", "ionosphere", "iris", "liver", "mammographic", "monk-2", "phoneme", "pima", "popfailures", "ring", "sonar", "soybean", "spambase", "spectfheart", "titanic", "twonorm", "vowel0", "waveform", "wine", "wisconsin"]

def prepare_dataset(set_name):
    dataset = np.genfromtxt(f"dataset/{set_name}.csv", delimiter=",")
    x = dataset[:, :-1]
    y = dataset[:, -1].astype(int)
    return x,y

def split_data(x, y):
    x_train, x_test, y_train, y_test = train_test_split(
        x, y,
        test_size=.3,
        random_state=42
    )
    return x_train, x_test, y_train, y_test

def get_gaussian(x_train, y_train, x_test):
    clf = GaussianNB()
    clf.fit(x_train, y_train)
    predict = clf.predict(x_test)
    return predict

def get_logistic_regression(x_train, y_train, x_test):
    logreg_clf = LogisticRegression()
    logreg_clf.fit(x_train, y_train)
    predict = logreg_clf.predict(x_test)
    return predict

def test(x, y):
    kf = RepeatedKFold(n_splits=5, n_repeats=2, random_state=1234)
    numpy = []
    numpy.append(matrix_labels)

    for train_index, test_index in kf.split(x):
        x_train, x_test = x[train_index], x[test_index]
        y_train, y_test = y[train_index], y[test_index]
        predict_gaussian = get_gaussian(x_train, y_train, x_test)
        predict_logistic_regression = get_logistic_regression(x_train, y_train, x_test)
        numpy.append([
            accuracy_score(y_test, predict_gaussian),
            accuracy_score(y_test, predict_logistic_regression)
        ])
    return numpy

def get_score(numpy):
    mean_score = np.mean(numpy)
    std_score = np.std(numpy)
    return "Accuracy score: %.3f (%.3f)" % (mean_score, std_score)

for dataset in datasets:
    x,y = prepare_dataset(dataset)
    x_train, x_test, y_train, y_test = split_data(x,y)
    result = test(x, y)
    print(get_score(result))