Automatically change dtypes for numerical columns to the minimum it can use

Improve Pandas Memory Efficiency

## source: https://www.kaggle.com/jeru666/did-you-think-of-these-features

def change_datatype(df):
    int_cols = list(df.select_dtypes(include=['int']).columns)
    for col in int_cols:
        if ((np.max(df[col]) <= 127) and(np.min(df[col] >= -128))):
            df[col] = df[col].astype(np.int8)
        elif ((np.max(df[col]) <= 32767) and(np.min(df[col] >= -32768))):
            df[col] = df[col].astype(np.int16)
        elif ((np.max(df[col]) <= 2147483647) and(np.min(df[col] >= -2147483648))):
            df[col] = df[col].astype(np.int32)
        else:
            df[col] = df[col].astype(np.int64)

change_datatype(df_transactions)

def change_datatype_float(df):
    float_cols = list(df.select_dtypes(include=['float']).columns)
    for col in float_cols:
        df[col] = df[col].astype(np.float32)
        
change_datatype_float(df_transactions)

mem = df_transactions.memory_usage(index=True).sum()
print(mem/ 1024**2," MB")