import os
import pandas

from utils import missing_value_handler, scaling_handler

# STEP 1: handle missing values + remove id column + robust scaling

if not os.path.exists("./data/Ketamin_icp_cleaned.csv"):
    data_path = "./data/Ketamine_icp.csv"
    data_frame_imputed = missing_value_handler(data_path)

    data_frame = scaling_handler(data_frame_imputed, method="robust_scaling")

    data_frame.to_csv("./data/Ketamin_icp_cleaned.csv", index=False)
else:
    data_frame = pandas.read_csv("./data/Ketamin_icp_cleaned.csv")


# STEP 2: Train and Test models