Лабораторна робота 1

from sklearn.datasets import load_iris
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns


def print_hist(values, title=''):
    plt.hist(values, bins=10)
    plt.title(f'Діаграма розподілу {title}')
    plt.xlabel('Значення')
    plt.ylabel('Частота')
    plt.show()

def info(df):
    info_df = pd.DataFrame({'MIN': [df['sepal_length'].min(),
                                    df['sepal_width'].min(),
                                    df['petal_length'].min(),
                                    df['petal_width'].min()],
                            'MAX': [df['sepal_length'].max(),
                                    df['sepal_width'].max(),
                                    df['petal_length'].max(),
                                    df['petal_width'].max()],
                            'MEAN': [df['sepal_length'].mean(),
                                     df['sepal_width'].mean(),
                                     df['petal_length'].mean(),
                                     df['petal_width'].mean()],
                            'VARIANCE': [df['sepal_length'].var(),
                                         df['sepal_width'].var(),
                                         df['petal_length'].var(),
                                         df['petal_width'].var()]}, index=df.columns[:-1])
    return info_df

def normalization(df):
    for i in range(len(df.columns) - 1):
        df.iloc[:, i] = (df.iloc[:, i] - df[df.columns[i]].mean()) / df[df.columns[i]].var()

    return df



iris = pd.read_csv('iris.csv')
sns.pairplot(iris, hue='species')
plt.show()

info_iris = info(iris)
print(info_iris)

for i in range(len(iris.columns) - 1):
    print_hist(iris.iloc[:, i], iris.columns[i])

iris_log = iris.copy()

iris_log.iloc[:, :-1] = np.log(iris_log.iloc[:, :-1])
info_iris_log = info(iris_log)
print(info_iris_log)

for i in range(len(iris.columns) - 1):
    print_hist(iris_log.iloc[:, i], f'log {iris.columns[i]}')

iris_log_norm = iris_log.copy()
iris_log_norm = normalization(iris_log_norm)
info_iris_log_norm = info(iris_log_norm)
print(info_iris_log_norm)

for i in range(len(iris.columns) - 1):
    print_hist(iris_log_norm.iloc[:, i], f'log-norm {iris.columns[i]}')