import pandas as pd
from sklearn.model_selection import train_test_split
import kaggle
import os

kaggle.api.authenticate()

c_directory = os.getcwd()

kaggle.api.dataset_download_files('kamilpytlak/personal-key-indicators-of-heart-disease', path=f"{c_directory}", unzip=True)


dataset = pd.read_csv("heart_2020_cleaned.csv")
print(dataset.describe(include='all'))
dataset = dataset.dropna()

print(dataset.describe(include='all'))

dataset_train, dataset_test = train_test_split(dataset, test_size=.2, train_size=.8, random_state=1)

print(dataset_train.describe(include='all'))

print("Wielkości:")

print("Zbiór uczący:", dataset_train.shape[0])
print("Zbiór testowy:", dataset_test.shape[0])
print("Łącznie: ", dataset.shape[0])

print(dataset["GenHealth"].value_counts())
print(dataset_train["GenHealth"].value_counts())
print("Średnia BMI -łącznie: ", dataset["BMI"].mean())
print("Odchylenie standardowe BMI - uczący:", dataset_train["BMI"].std())
print("Odchylenie standardowe BMI - łącznie:", dataset["BMI"].std())

print("Mediana BMI:", dataset_test["BMI"].median())

max_bmi = dataset_train["BMI"].max()

print(max_bmi)
dataset_train["BMI"] = dataset_train["BMI"].apply(lambda x: x/max_bmi)
dataset_test["BMI"] = dataset_test["BMI"].apply(lambda x: x/max_bmi)
print(dataset_train["AgeCategory"].value_counts())
print(dataset_train["BMI"])