article-identify-hardbounce-ml_2_aux_functions.py

def oversampleSMOTE(X, y):
    '''
    Resample a dataset using SMOTE oversample

    Input:
        X = dataframe with x variables (explanatory variables)
        y = dataframe with y variable (variable to predict)

    Output:
        df[0] = X dataframe resampled
        df[1] = y dataframe resampled
    '''

    from imblearn.over_sampling import SMOTE
    import pandas as pd

    sm = SMOTE(random_state=123)
    X_resampled, y_resampled = sm.fit_resample(X, y.ravel())

    # Get column names
    X_cols = X.columns.values
    y_cols = [y.name]

    return pd.DataFrame(X_resampled, columns=X_cols) , pd.DataFrame(y_resampled, columns=y_cols)

article-identify-hardbounce-ml_2_remove-imbalance.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import libraries\n",
    "from functions import aux_functions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Oversample dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_vars = ['monthsSinceRegDate', 'age', 'percNumbersInEmailUser', 'emailUserCharQty', \n",
    "          'emailDomain_cat_gmail.com', 'emailDomain_cat_hotmail.com',\n",
    "          'emailDomainPiece1_com',\n",
    "          'emailDomainPiece2_br', 'emailDomainPiece2_missing']\n",
    "\n",
    "X = hardbounce_dummy[x_vars]\n",
    "\n",
    "y = hardbounce_dummy['flgHardBounce_n']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_resampled, y_resampled = aux_functions.oversampleSMOTE(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "flgHardBounce_n\n",
       "0    0.5\n",
       "1    0.5\n",
       "Name: __dummy__, dtype: float64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aux_functions.freqTable(\n",
    "    y_resampled['flgHardBounce_n']\n",
    "    ,\"\"\n",
    "    ,False\n",
    "    ,\"columns\"\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}