PandasUtility.py

import pandas as pd


def convert_float(df, columns, to_replace=r'\$|,', value=''):
    """
    Convert a series $453,534.65 --> 453.543.65 with sep=','
    @params df (DataFrame): DataFrame to be modified
	@params columns (list): List of columns
	"""

    for col in columns:
        df[col] = df[col].replace(to_replace, value=value, regex=True)
    df[columns] = df[columns].apply(pd.to_numeric, errors='coerce')
    return df


def clean_columns(df):
    """
    Lowers the case of all columns and removes spaces
    """
    df.columns = df.columns.str.lower().str.replace(' ', '_')
    return df


def convert_kmb_float(df, columns, mapping_dict=None):
    # Convert columns containg K - Thousand, M - Million or B - Billion at the end to float
    if mapping_dict is None:
        mapping_dict = {'K': 1e3, 'M': 1e6, 'B': 1e9}

    for col in columns:
        df[col] = df[col].str.replace(' ','').str.upper().str.replace('K|B|M', '').astype(float).mul(df[col].str.upper().str.get(-1).map(mapping_dict))
    return df


def preprocess_df(df, float_cols, **kwrgs):
    df = clean_columns(df)
    df = convert_float(df, float_cols, **kwrgs)
    return df