dummy_var.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Mon Oct  8 12:54:46 2018

@author: claudia
"""

import pandas as pd
import numpy as np
from sklearn.preprocessing import LabelEncoder, OneHotEncoder

def preprocessData(df):
    label_encoder = LabelEncoder()
    dummy_encoder = OneHotEncoder()
    pdf = pd.DataFrame()
    for att in df.columns:
        if df[att].dtype == np.float64 or df[att].dtype == np.int64:
            pdf = pd.concat([pdf, df[att]], axis=1)
        else:
            df[att] = label_encoder.fit_transform(df[att])
            # Fitting One Hot Encoding on train data
            temp = dummy_encoder.fit_transform(df[att].values.reshape(-1,1)).toarray()
            # Changing encoded features into a dataframe with new column names
            temp = pd.DataFrame(temp,
                                columns=[(att + "_" + str(i)) for i in df[att].value_counts().index])
            # In side by side concatenation index values should be same
            # Setting the index values similar to the data frame
            temp = temp.set_index(df.index.values)
            # adding the new One Hot Encoded varibales to the dataframe
            pdf = pd.concat([pdf, temp], axis=1)
    return pdf