acm_imbalance_algorithms.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Imbalanced Learning Example\n",
    "\n",
    "This notebook works through a real-life example (the Kaggle 'Give me some credit' dataset). For more information, see https://www.kaggle.com/c/GiveMeSomeCredit .\n",
    "\n",
    "We'll be training linear classifiers on the data, including logistic regression, linear SVMs, and decision trees. The scikit-learn API provides a 'weights' option for these, and they're relatively fast to train."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Import useful libraries used in the notebook\n",
    "import time\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "# pd.set_option('display.mpl_style', 'default') # Make the graphs a bit prettier\n",
    "\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn import metrics\n",
    "from sklearn.cross_validation import cross_val_score\n",
    "from sklearn.cross_validation import StratifiedKFold\n",
    "\n",
    "# Pull in the models we'll be building\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "# Functions are defined in this module\n",
    "from acm_imbalanced_library import *\n",
    "\n",
    "# Visualization\n",
    "from sklearn.manifold import TSNE\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "# Show plots inline \n",
    "%matplotlib inline\n",
    "\n",
    "# Auto-reload external modules\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data exploration\n",
    "\n",
    "We'll clean up the CSV file after loading it by converting the columns to the appropriate types. Then we'll print out some summary statistics and pair-wise distributions to get a feel for the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>SeriousDlqin2yrs</th>\n",
       "      <th>RevolvingUtilizationOfUnsecuredLines</th>\n",
       "      <th>age</th>\n",
       "      <th>NumberOfTime30-59DaysPastDueNotWorse</th>\n",
       "      <th>DebtRatio</th>\n",
       "      <th>MonthlyIncome</th>\n",
       "      <th>NumberOfOpenCreditLinesAndLoans</th>\n",
       "      <th>NumberOfTimes90DaysLate</th>\n",
       "      <th>NumberRealEstateLoansOrLines</th>\n",
       "      <th>NumberOfTime60-89DaysPastDueNotWorse</th>\n",
       "      <th>NumberOfDependents</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.766127</td>\n",
       "      <td>45</td>\n",
       "      <td>2</td>\n",
       "      <td>0.802982</td>\n",
       "      <td>9120.0</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0.957151</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0.121876</td>\n",
       "      <td>2600.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.658180</td>\n",
       "      <td>38</td>\n",
       "      <td>1</td>\n",
       "      <td>0.085113</td>\n",
       "      <td>3042.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0.233810</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>0.036050</td>\n",
       "      <td>3300.0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0.907239</td>\n",
       "      <td>49</td>\n",
       "      <td>1</td>\n",
       "      <td>0.024926</td>\n",
       "      <td>63588.0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0  SeriousDlqin2yrs  RevolvingUtilizationOfUnsecuredLines  age  \\\n",
       "0           1                 1                              0.766127   45   \n",
       "1           2                 0                              0.957151   40   \n",
       "2           3                 0                              0.658180   38   \n",
       "3           4                 0                              0.233810   30   \n",
       "4           5                 0                              0.907239   49   \n",
       "\n",
       "   NumberOfTime30-59DaysPastDueNotWorse  DebtRatio  MonthlyIncome  \\\n",
       "0                                     2   0.802982         9120.0   \n",
       "1                                     0   0.121876         2600.0   \n",
       "2                                     1   0.085113         3042.0   \n",
       "3                                     0   0.036050         3300.0   \n",
       "4                                     1   0.024926        63588.0   \n",
       "\n",
       "   NumberOfOpenCreditLinesAndLoans  NumberOfTimes90DaysLate  \\\n",
       "0                               13                        0   \n",
       "1                                4                        0   \n",
       "2                                2                        1   \n",
       "3                                5                        0   \n",
       "4                                7                        0   \n",
       "\n",
       "   NumberRealEstateLoansOrLines  NumberOfTime60-89DaysPastDueNotWorse  \\\n",
       "0                             6                                     0   \n",
       "1                             0                                     0   \n",
       "2                             0                                     0   \n",
       "3                             0                                     0   \n",
       "4                             1                                     0   \n",
       "\n",
       "   NumberOfDependents  \n",
       "0                 2.0  \n",
       "1                 1.0  \n",
       "2                 0.0  \n",
       "3                 0.0  \n",
       "4                 0.0  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Read in the dataset, look at the column names and NAs\n",
    "credit_df = pd.read_csv('cs-training.csv')\n",
    "credit_df.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>target</th>\n",
       "      <th>revolving_pct</th>\n",
       "      <th>age</th>\n",
       "      <th>past_due_30_59</th>\n",
       "      <th>debt_ratio</th>\n",
       "      <th>monthly_income</th>\n",
       "      <th>lines</th>\n",
       "      <th>past_due_gt_90</th>\n",
       "      <th>mortgages</th>\n",
       "      <th>past_due_60_89</th>\n",
       "      <th>dependents</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.766127</td>\n",
       "      <td>45</td>\n",
       "      <td>2</td>\n",
       "      <td>0.802982</td>\n",
       "      <td>9120.0</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0.957151</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0.121876</td>\n",
       "      <td>2600.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.658180</td>\n",
       "      <td>38</td>\n",
       "      <td>1</td>\n",
       "      <td>0.085113</td>\n",
       "      <td>3042.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0.233810</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>0.036050</td>\n",
       "      <td>3300.0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0.907239</td>\n",
       "      <td>49</td>\n",
       "      <td>1</td>\n",
       "      <td>0.024926</td>\n",
       "      <td>63588.0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ID  target  revolving_pct  age  past_due_30_59  debt_ratio  monthly_income  \\\n",
       "0   1       1       0.766127   45               2    0.802982          9120.0   \n",
       "1   2       0       0.957151   40               0    0.121876          2600.0   \n",
       "2   3       0       0.658180   38               1    0.085113          3042.0   \n",
       "3   4       0       0.233810   30               0    0.036050          3300.0   \n",
       "4   5       0       0.907239   49               1    0.024926         63588.0   \n",
       "\n",
       "   lines  past_due_gt_90  mortgages  past_due_60_89  dependents  \n",
       "0     13               0          6               0           2  \n",
       "1      4               0          0               0           1  \n",
       "2      2               1          0               0           0  \n",
       "3      5               0          0               0           0  \n",
       "4      7               0          1               0           0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Clean up the imported raw CSV file. Rename columns, fill in NA values,\n",
    "# and change dependents to integer.\n",
    "credit_df = cleanCreditDf(credit_df)\n",
    "\n",
    "# Check how the data looks now\n",
    "credit_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>target</th>\n",
       "      <th>revolving_pct</th>\n",
       "      <th>age</th>\n",
       "      <th>past_due_30_59</th>\n",
       "      <th>debt_ratio</th>\n",
       "      <th>monthly_income</th>\n",
       "      <th>lines</th>\n",
       "      <th>past_due_gt_90</th>\n",
       "      <th>mortgages</th>\n",
       "      <th>past_due_60_89</th>\n",
       "      <th>dependents</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>1.500000e+05</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>75000.500000</td>\n",
       "      <td>0.066840</td>\n",
       "      <td>6.048438</td>\n",
       "      <td>52.295207</td>\n",
       "      <td>0.421033</td>\n",
       "      <td>353.005076</td>\n",
       "      <td>5.348139e+03</td>\n",
       "      <td>8.452760</td>\n",
       "      <td>0.265973</td>\n",
       "      <td>1.018240</td>\n",
       "      <td>0.240387</td>\n",
       "      <td>0.737413</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>43301.414527</td>\n",
       "      <td>0.249746</td>\n",
       "      <td>249.755371</td>\n",
       "      <td>14.771866</td>\n",
       "      <td>4.192781</td>\n",
       "      <td>2037.818523</td>\n",
       "      <td>1.315206e+04</td>\n",
       "      <td>5.145951</td>\n",
       "      <td>4.169304</td>\n",
       "      <td>1.129771</td>\n",
       "      <td>4.155179</td>\n",
       "      <td>1.107021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>37500.750000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.029867</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.175074</td>\n",
       "      <td>1.550000e+03</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>75000.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.154181</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.366508</td>\n",
       "      <td>4.357500e+03</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>112500.250000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.559046</td>\n",
       "      <td>63.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.868254</td>\n",
       "      <td>7.400000e+03</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>150000.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>50708.000000</td>\n",
       "      <td>109.000000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>329664.000000</td>\n",
       "      <td>3.008750e+06</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>54.000000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>20.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  ID         target  revolving_pct            age  \\\n",
       "count  150000.000000  150000.000000  150000.000000  150000.000000   \n",
       "mean    75000.500000       0.066840       6.048438      52.295207   \n",
       "std     43301.414527       0.249746     249.755371      14.771866   \n",
       "min         1.000000       0.000000       0.000000       0.000000   \n",
       "25%     37500.750000       0.000000       0.029867      41.000000   \n",
       "50%     75000.500000       0.000000       0.154181      52.000000   \n",
       "75%    112500.250000       0.000000       0.559046      63.000000   \n",
       "max    150000.000000       1.000000   50708.000000     109.000000   \n",
       "\n",
       "       past_due_30_59     debt_ratio  monthly_income          lines  \\\n",
       "count   150000.000000  150000.000000    1.500000e+05  150000.000000   \n",
       "mean         0.421033     353.005076    5.348139e+03       8.452760   \n",
       "std          4.192781    2037.818523    1.315206e+04       5.145951   \n",
       "min          0.000000       0.000000    0.000000e+00       0.000000   \n",
       "25%          0.000000       0.175074    1.550000e+03       5.000000   \n",
       "50%          0.000000       0.366508    4.357500e+03       8.000000   \n",
       "75%          0.000000       0.868254    7.400000e+03      11.000000   \n",
       "max         98.000000  329664.000000    3.008750e+06      58.000000   \n",
       "\n",
       "       past_due_gt_90      mortgages  past_due_60_89     dependents  \n",
       "count   150000.000000  150000.000000   150000.000000  150000.000000  \n",
       "mean         0.265973       1.018240        0.240387       0.737413  \n",
       "std          4.169304       1.129771        4.155179       1.107021  \n",
       "min          0.000000       0.000000        0.000000       0.000000  \n",
       "25%          0.000000       0.000000        0.000000       0.000000  \n",
       "50%          0.000000       1.000000        0.000000       0.000000  \n",
       "75%          0.000000       2.000000        0.000000       1.000000  \n",
       "max         98.000000      54.000000       98.000000      20.000000  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get some summary statistics from the dataframe\n",
    "credit_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAwcAAALFCAYAAACWD59EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVByIiIiIiAgA/wOxTgFYGs38CgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10259b690>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot out the distributions to see what's going on, colour code by target\n",
    "import seaborn as sns; sns.set(style=\"ticks\", color_codes=True)\n",
    "g = sns.pairplot(credit_df[['age', 'debt_ratio', 'lines', 'target']], \n",
    "                 hue='target', palette=sns.diverging_palette(220,20,n=2))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model training and evaluation\n",
    "\n",
    "Now we've got an idea of the data structure, we can train some linear classifiers to check their relative performance. Before doing this, we'll need to remove:\n",
    "\n",
    "* target - The model is supposed to tell use the target !\n",
    "* ID - This is a unique identifier, and (should be) arbitrary is Kaggle have properly shuffled the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Separate out the ID and target values from the dataset\n",
    "X, y, id_val = removeIDTargetFromCreditDf(credit_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Default logistic regression accuracy  = 0.933740, AUC = 0.696980\n",
      "Balanced logistic regression accuracy = 0.783020, AUC = 0.790229\n"
     ]
    }
   ],
   "source": [
    "# Train a logistic regression model using the balanced class weights \n",
    "\n",
    "default_log_reg = LogisticRegression(class_weight=None)\n",
    "balanced_log_reg = LogisticRegression(class_weight='balanced')\n",
    "\n",
    "# Important to use a stratified K folds to make sure representative\n",
    "# proportion of minority examples are used in each fold\n",
    "cv = StratifiedKFold(y, n_folds=5, shuffle=True)\n",
    "\n",
    "scores = cross_val_score(estimator=default_log_reg, X=X, y=y, cv=cv, scoring='accuracy')\n",
    "def_log_reg_acc = np.mean(scores)\n",
    "\n",
    "scores = cross_val_score(estimator=balanced_log_reg, X=X, y=y, cv=cv, scoring='accuracy')\n",
    "bal_log_reg_acc = np.mean(scores)\n",
    "\n",
    "scores = cross_val_score(estimator=default_log_reg, X=X, y=y, cv=cv, scoring='roc_auc')\n",
    "def_log_reg_roc = np.mean(scores)\n",
    "\n",
    "scores = cross_val_score(estimator=balanced_log_reg, X=X, y=y, cv=cv, scoring='roc_auc')\n",
    "bal_log_reg_roc = np.mean(scores)\n",
    "\n",
    "print 'Default logistic regression accuracy  = {:.6f}, AUC = {:.6f}'.format(def_log_reg_acc, def_log_reg_roc)\n",
    "print 'Balanced logistic regression accuracy = {:.6f}, AUC = {:.6f}'.format(bal_log_reg_acc, bal_log_reg_roc)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Default linear SVM ROC = 0.547332\n",
      "Balanced linear SVM ROC = 0.556079\n"
     ]
    }
   ],
   "source": [
    "# Train a linear SVM model using the balanced class weights \n",
    "\n",
    "default_lin_svm = LinearSVC(class_weight=None)\n",
    "balanced_lin_svm = LinearSVC(class_weight='balanced')\n",
    "\n",
    "scores = cross_val_score(estimator=default_lin_svm, X=X, y=y, cv=cv,  scoring='roc_auc')\n",
    "def_lin_svm_roc = np.mean(scores)\n",
    "\n",
    "scores = cross_val_score(estimator=balanced_lin_svm, X=X, y=y, cv=cv, scoring='roc_auc')\n",
    "bal_lin_svm_roc = np.mean(scores)\n",
    "\n",
    "print 'Default linear SVM ROC = {:.6f}'.format(def_lin_svm_roc)\n",
    "print 'Balanced linear SVM ROC = {:.6f}'.format(bal_lin_svm_roc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Default decision tree ROC = 0.611893\n",
      "Balanced decision tree ROC = 0.601773\n"
     ]
    }
   ],
   "source": [
    "# Train a Decision tree classifier\n",
    "\n",
    "default_tree = DecisionTreeClassifier(class_weight=None)\n",
    "balanced_tree = DecisionTreeClassifier(class_weight='balanced')\n",
    "\n",
    "scores = cross_val_score(estimator=default_tree, X=X, y=y, cv=cv, scoring='roc_auc')\n",
    "def_tree_roc = np.mean(scores)\n",
    "\n",
    "scores = cross_val_score(estimator=balanced_tree, X=X, y=y, cv=cv, scoring='roc_auc')\n",
    "bal_tree_roc = np.mean(scores)\n",
    "\n",
    "print 'Default decision tree ROC = {:.6f}'.format(def_tree_roc)\n",
    "print 'Balanced decision tree ROC = {:.6f}'.format(bal_tree_roc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pre-processed datasets\n",
    "\n",
    "R has a package designed to deal with imbalanced data, called 'unbalanced'. For more information see here: https://cran.r-project.org/web/packages/unbalanced/unbalanced.pdf.\n",
    "\n",
    "The examples in the previous notebook are designed to be easy to read and modify, so they take a long time to process ~MBs of data. \n",
    "\n",
    "The R dataframes have a different order of columns, so a separate \"clean\" function is needed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading R pre-processed dataframes\n",
      "Creating logistic regression setup\n",
      "Running evaluation of processed dataframes\n",
      "-> 1 of 4\n",
      "-> 2 of 4\n",
      "-> 3 of 4\n",
      "-> 4 of 4\n",
      "Baseline logistic regression AUC = 0.790229\n",
      "CNN AUC = 0.817713\n",
      "OSS AUC = 0.821785\n",
      "Tomek AUC = 0.821941\n",
      "SMOTE AUC = 0.909409\n"
     ]
    }
   ],
   "source": [
    "# Load in the R pre-processed dataframes, check AUC on each\n",
    "print 'Loading R pre-processed dataframes'\n",
    "credit_df_cnn = pd.read_csv('cs-training-CNN.csv')\n",
    "credit_df_oss = pd.read_csv('cs-training-OSS.csv')\n",
    "credit_df_smote = pd.read_csv('cs-training-smote.csv')\n",
    "credit_df_tomek = pd.read_csv('cs-training-tomek.csv')\n",
    "\n",
    "print 'Creating logistic regression setup'\n",
    "bal_log_reg = LogisticRegression(class_weight='balanced')\n",
    "# cv = StratifiedKFold(y, n_folds=5, shuffle=True)\n",
    "\n",
    "eval_data = [credit_df_cnn, \n",
    "             credit_df_oss, \n",
    "             credit_df_smote, \n",
    "             credit_df_tomek]\n",
    "\n",
    "eval_title = ['CNN', 'OSS', 'SMOTE', 'Tomek']\n",
    "\n",
    "eval_results = {}\n",
    "\n",
    "print 'Running evaluation of processed dataframes'\n",
    "for data_idx, data in enumerate(eval_data):\n",
    "    print '-> {} of {}'.format(data_idx+1, len(eval_data))\n",
    "    data = cleanRProcessedCreditDf(data)\n",
    "    X, y, id_val = removeIDTargetFromCreditDf(data)\n",
    "    cv = StratifiedKFold(y, n_folds=5, shuffle=True)\n",
    "    scores = cross_val_score(estimator=bal_log_reg, X=X, y=y, cv=cv, scoring='roc_auc')\n",
    "    eval_results[eval_title[data_idx]] = np.mean(scores)\n",
    "    \n",
    "print 'Baseline logistic regression AUC = {:.6f}'.format(bal_log_reg_roc)\n",
    "\n",
    "\n",
    "for result, value in eval_results.items():\n",
    "    print '{} AUC = {:.6f}'.format(result, value)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Summary\n",
    "\n",
    "In this notebook we evaluated some linear classifiers, noting:\n",
    "\n",
    "* How their performance changes when optimizing for AUC instead of accuracy.\n",
    "* How the weights can be used to improve ROC performance.\n",
    "* Their performance on the Kaggle dataset.\n",
    "\n",
    "After this, we loaded pre-processed datasets which were generated using the R 'unbalanced' package. See ```acm_imbalanced_dataset.R``` for more information. We then re-ran using the best performing linear classifier to check the AUC increase.\n",
    "\n",
    "The best performing combination was Logistic Regression with balanced weights running on a SMOTE-processed dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf4AAAGACAYAAABMRwCUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEEJDpPALIYQQGvL/MYvs+dimNIcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1195fa150>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show a bar chart of the ROC results for each of the algorithms\n",
    "\n",
    "ROC_labels = ['logreg', 'lin_svm', 'tree', 'logreg\\n+CNN', \n",
    "            'logreg\\n+OSS', 'logreg\\n+SMOTE', 'logreg\\n+Tomek']\n",
    "ROC_vals = [bal_log_reg_roc, bal_lin_svm_roc, bal_tree_roc, \n",
    "            eval_results['CNN'], eval_results['OSS'], \n",
    "            eval_results['SMOTE'], eval_results['Tomek']]\n",
    "\n",
    "width = 0.5\n",
    "ind = np.arange(len(ROC_labels))\n",
    "base_roc = [bal_log_reg_roc, bal_lin_svm_roc, bal_tree_roc, \n",
    "            bal_log_reg_roc, bal_log_reg_roc, \n",
    "            bal_log_reg_roc, bal_log_reg_roc]\n",
    "inc_roc = [0,0,0, eval_results['CNN']-bal_log_reg_roc, \n",
    "           eval_results['OSS']-bal_log_reg_roc, \n",
    "            eval_results['SMOTE']-bal_log_reg_roc, \n",
    "           eval_results['Tomek']-bal_log_reg_roc]\n",
    "\n",
    "p1 = plt.bar(ind, base_roc, color='b')\n",
    "p2 = plt.bar(ind, inc_roc, color='g', bottom = base_roc)\n",
    "plt.ylabel('AUC', fontsize=18)\n",
    "plt.title('AUC by algorithm and preprocessing type', fontsize=20)\n",
    "plt.xticks(ind + width/2, ROC_labels, fontsize=14)\n",
    "plt.yticks(np.arange(0, 1.3, 0.1))\n",
    "plt.legend((p1[0], p2[0]), ('Balanced algorithm', 'Preprocessing'), fontsize=14)\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}