commit be3e041321aad552978fe670bf68c5cc1e44f185
Author: Michał Kaleta <s486867@labs.wmi.amu.edu.pl>
Date:   Sat Mar 25 12:49:56 2023 +0100

    first commit

diff --git a/main.py b/main.py
new file mode 100644
index 0000000..f4011b1
--- /dev/null
+++ b/main.py
@@ -0,0 +1,27 @@
+from kaggle.api.kaggle_api_extended import KaggleApi
+import zipfile
+from sklearn.model_selection import train_test_split
+import pandas as pd
+pd.set_option('display.max_columns', 100)
+
+api = KaggleApi()
+api.authenticate()
+api.dataset_download_files('rkiattisak/sports-car-prices-dataset', path='./data')
+with zipfile.ZipFile('./data/sports-car-prices-dataset.zip', 'r') as zip_ref:
+    zip_ref.extractall('./data')
+
+cars = pd.read_csv('./data/Sport car price.csv')
+
+cars.dropna(inplace=True)
+
+random_seed = 42
+train_data, test_data = train_test_split(cars, test_size=0.2, random_state=random_seed)
+train_data, dev_data = train_test_split(train_data, test_size=0.25, random_state=random_seed)
+
+train_stats = train_data.describe(include='all')
+print(f"\nTraining set statistics:\n{train_stats}")
+dev_stats = dev_data.describe(include='all')
+print(f"\nDevelopment set statistics:\n{dev_stats}")
+test_stats = test_data.describe(include='all')
+print(f"\nTest set statistics:\n{test_stats}")
+