kdd resampling.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import imblearn\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Loading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "names = [\"duration\",\"protocol_type\",\"service\",\"flag\",\"src_bytes\",\n",
    "    \"dst_bytes\",\"land\",\"wrong_fragment\",\"urgent\",\"hot\",\"num_failed_logins\",\n",
    "    \"logged_in\",\"num_compromised\",\"root_shell\",\"su_attempted\",\"num_root\",\n",
    "    \"num_file_creations\",\"num_shells\",\"num_access_files\",\"num_outbound_cmds\",\n",
    "    \"is_host_login\",\"is_guest_login\",\"count\",\"srv_count\",\"serror_rate\",\n",
    "    \"srv_serror_rate\",\"rerror_rate\",\"srv_rerror_rate\",\"same_srv_rate\",\n",
    "    \"diff_srv_rate\",\"srv_diff_host_rate\",\"dst_host_count\",\"dst_host_srv_count\",\n",
    "    \"dst_host_same_srv_rate\",\"dst_host_diff_srv_rate\",\"dst_host_same_src_port_rate\",\n",
    "    \"dst_host_srv_diff_host_rate\",\"dst_host_serror_rate\",\"dst_host_srv_serror_rate\",\n",
    "    \"dst_host_rerror_rate\",\"dst_host_srv_rerror_rate\",\"target\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('kddcup.csv', names=names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df.drop('num_outbound_cmds', axis=1, inplace=True)\n",
    "df.drop('is_host_login', axis=1, inplace=True)\n",
    "df['protocol_type'] = df['protocol_type'].astype('category')\n",
    "df['service'] = df['service'].astype('category')\n",
    "df['flag'] = df['flag'].astype('category')\n",
    "cat_columns = df.select_dtypes(['category']).columns\n",
    "df[cat_columns] = df[cat_columns].apply(lambda x: x.cat.codes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df.drop_duplicates(subset=None, keep='first', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(145586, 40)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "target_count = df.target.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "normal.             87832\n",
       "neptune.            51820\n",
       "back.                 968\n",
       "teardrop.             918\n",
       "satan.                906\n",
       "warezclient.          893\n",
       "ipsweep.              651\n",
       "smurf.                641\n",
       "portsweep.            416\n",
       "pod.                  206\n",
       "nmap.                 158\n",
       "guess_passwd.          53\n",
       "buffer_overflow.       30\n",
       "warezmaster.           20\n",
       "land.                  19\n",
       "imap.                  12\n",
       "rootkit.               10\n",
       "loadmodule.             9\n",
       "ftp_write.              8\n",
       "multihop.               7\n",
       "phf.                    4\n",
       "perl.                   3\n",
       "spy.                    2\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target_count"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random under-sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "count_class_normal, count_class_neptune, count_class_back, count_class_teardrop, count_class_satan, count_class_warezclient, count_class_ipsweep, count_class_smurf, count_class_portsweep, count_class_pod, count_class_nmap, count_class_guess_passwd, count_class_buffer_overflow, count_class_warezmaster, count_class_land, count_class_imap, count_class_rootkit, count_class_loadmodule, count_class_ftp_write, count_class_multihop, count_class_phf, count_class_perl, count_class_spy = df.target.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Divide by class\n",
    "df_normal = df[df['target'] == 'normal.']\n",
    "df_neptune = df[df['target'] == 'neptune.']\n",
    "df_back = df[df['target'] == 'back.']\n",
    "df_teardrop = df[df['target'] == 'teardrop.']\n",
    "df_satan = df[df['target'] == 'satan.']\n",
    "df_warezclient = df[df['target'] == 'warezclient.']\n",
    "df_ipsweep = df[df['target'] == 'ipsweep.']\n",
    "df_smurf = df[df['target'] == 'smurf.']\n",
    "df_portsweep = df[df['target'] == 'portsweep.']\n",
    "df_pod = df[df['target'] == 'pod.']\n",
    "df_nmap = df[df['target'] == 'nmap.']\n",
    "df_guess_passwd = df[df['target'] == 'guess_passwd.']\n",
    "df_buffer_overflow = df[df['target'] == 'buffer_overflow.']\n",
    "df_warezmaster = df[df['target'] == 'warezmaster.']\n",
    "df_land = df[df['target'] == 'land.']\n",
    "df_imap = df[df['target'] == 'imap.']\n",
    "df_rootkit = df[df['target'] == 'rootkit.']\n",
    "df_loadmodule = df[df['target'] == 'loadmodule.']\n",
    "df_ftp_write = df[df['target'] == 'ftp_write.']\n",
    "df_multihop = df[df['target'] == 'multihop.']\n",
    "df_phf = df[df['target'] == 'phf.']\n",
    "df_perl = df[df['target'] == 'perl.']\n",
    "df_spy = df[df['target'] == 'spy.']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_normal_under = df_normal.sample(count_class_neptune)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_test_under = pd.concat([df_normal_under, df_neptune], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random under-sampling:\n",
      "neptune.    51820\n",
      "normal.     51820\n",
      "Name: target, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print('Random under-sampling:')\n",
    "print(df_test_under.target.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAErCAYAAAA1wE/rAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMRwkSR3DQZLUMRwkSZ3/D/s3qWalNKtXAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9b2053a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "my_colors = 'rgbkymc'\n",
    "df_test_under.target.value_counts().plot(kind='bar', title='Count (target)',  color=my_colors);\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random oversampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_neptune_over = df_neptune.sample(count_class_normal, replace=True)\n",
    "df_test_over = pd.concat([df_normal, df_neptune_over], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random over-sampling:\n",
      "normal.     87832\n",
      "neptune.    87832\n",
      "Name: target, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAErCAYAAAA1wE/rAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9b21487f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Random over-sampling:')\n",
    "print(df_test_over.target.value_counts())\n",
    "\n",
    "df_test_over.target.value_counts().plot(kind='bar', title='Count (target)', color=my_colors);\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "from imblearn.datasets import make_imbalance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_test = pd.concat([df_loadmodule, df_rootkit], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_test['target'] = df_test['target'].astype('category')\n",
    "cat_columns = df_test.select_dtypes(['category']).columns\n",
    "df_test[cat_columns] = df_test[cat_columns].apply(lambda x: x.cat.codes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>duration</th>\n",
       "      <th>protocol_type</th>\n",
       "      <th>service</th>\n",
       "      <th>flag</th>\n",
       "      <th>src_bytes</th>\n",
       "      <th>dst_bytes</th>\n",
       "      <th>land</th>\n",
       "      <th>wrong_fragment</th>\n",
       "      <th>urgent</th>\n",
       "      <th>hot</th>\n",
       "      <th>...</th>\n",
       "      <th>dst_host_srv_count</th>\n",
       "      <th>dst_host_same_srv_rate</th>\n",
       "      <th>dst_host_diff_srv_rate</th>\n",
       "      <th>dst_host_same_src_port_rate</th>\n",
       "      <th>dst_host_srv_diff_host_rate</th>\n",
       "      <th>dst_host_serror_rate</th>\n",
       "      <th>dst_host_srv_serror_rate</th>\n",
       "      <th>dst_host_rerror_rate</th>\n",
       "      <th>dst_host_srv_rerror_rate</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4049</th>\n",
       "      <td>79</td>\n",
       "      <td>1</td>\n",
       "      <td>56</td>\n",
       "      <td>9</td>\n",
       "      <td>281</td>\n",
       "      <td>1301</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>10</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86219</th>\n",
       "      <td>103</td>\n",
       "      <td>1</td>\n",
       "      <td>56</td>\n",
       "      <td>9</td>\n",
       "      <td>302</td>\n",
       "      <td>8876</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147604</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>5921</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147605</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>5014</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147606</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>2072</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.40</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 40 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        duration  protocol_type  service  flag  src_bytes  dst_bytes  land  \\\n",
       "4049          79              1       56     9        281       1301     0   \n",
       "86219        103              1       56     9        302       8876     0   \n",
       "147604         0              1       19     9          0       5921     0   \n",
       "147605         0              1       19     9          0       5014     0   \n",
       "147606         0              1       19     9          0       2072     0   \n",
       "\n",
       "        wrong_fragment  urgent  hot   ...    dst_host_srv_count  \\\n",
       "4049                 0       0    2   ...                    10   \n",
       "86219                0       0    2   ...                     1   \n",
       "147604               0       0    0   ...                     3   \n",
       "147605               0       0    0   ...                     4   \n",
       "147606               0       0    1   ...                     5   \n",
       "\n",
       "        dst_host_same_srv_rate  dst_host_diff_srv_rate  \\\n",
       "4049                       1.0                     0.0   \n",
       "86219                      1.0                     0.0   \n",
       "147604                     1.0                     0.0   \n",
       "147605                     1.0                     0.0   \n",
       "147606                     1.0                     0.0   \n",
       "\n",
       "        dst_host_same_src_port_rate  dst_host_srv_diff_host_rate  \\\n",
       "4049                            1.0                         0.30   \n",
       "86219                           1.0                         0.00   \n",
       "147604                          1.0                         0.67   \n",
       "147605                          1.0                         0.50   \n",
       "147606                          1.0                         0.40   \n",
       "\n",
       "        dst_host_serror_rate  dst_host_srv_serror_rate  dst_host_rerror_rate  \\\n",
       "4049                     0.0                       0.0                   0.0   \n",
       "86219                    0.0                       0.0                   0.0   \n",
       "147604                   0.0                       0.0                   0.0   \n",
       "147605                   0.0                       0.0                   0.0   \n",
       "147606                   0.0                       0.0                   0.0   \n",
       "\n",
       "        dst_host_srv_rerror_rate  target  \n",
       "4049                         0.1       0  \n",
       "86219                        0.0       0  \n",
       "147604                       0.0       0  \n",
       "147605                       0.0       0  \n",
       "147606                       0.0       0  \n",
       "\n",
       "[5 rows x 40 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data = df_test.values\n",
    "Y = data[:,39]\n",
    "X = data[:,0:39]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(19, 39)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Y = Y.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "sScaler = StandardScaler()\n",
    "rescaleX = sScaler.fit_transform(X)\n",
    "pca = PCA(n_components=2)\n",
    "rescaleX = pca.fit_transform(rescaleX)\n",
    "rescaleX = np.append(rescaleX, Y, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "principalDf = pd.DataFrame(data = rescaleX, columns = ['principal component 1', 'principal component 2', 'target'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>principal component 1</th>\n",
       "      <th>principal component 2</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.074827</td>\n",
       "      <td>2.854827</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.291615</td>\n",
       "      <td>4.483403</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-4.468923</td>\n",
       "      <td>1.584395</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-4.376407</td>\n",
       "      <td>1.320772</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-4.585688</td>\n",
       "      <td>2.086770</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   principal component 1  principal component 2  target\n",
       "0              -0.074827               2.854827     0.0\n",
       "1               1.291615               4.483403     0.0\n",
       "2              -4.468923               1.584395     0.0\n",
       "3              -4.376407               1.320772     0.0\n",
       "4              -4.585688               2.086770     0.0"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "principalDf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x34ce6ad4a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmMAAAGDCAYAAABnZBdiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x34ce6ad518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.scatter(principalDf.iloc[:,0], principalDf.iloc[:,1])\n",
    "plt.title('Principal component 1 vs 2')\n",
    "plt.xlabel(\"pc 1\")\n",
    "plt.ylabel(\"pc 2\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data = principalDf.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>principal component 1</th>\n",
       "      <th>principal component 2</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.074827</td>\n",
       "      <td>2.854827</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.291615</td>\n",
       "      <td>4.483403</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-4.468923</td>\n",
       "      <td>1.584395</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-4.376407</td>\n",
       "      <td>1.320772</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-4.585688</td>\n",
       "      <td>2.086770</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.049923</td>\n",
       "      <td>0.170846</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.196737</td>\n",
       "      <td>-1.027830</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.089203</td>\n",
       "      <td>-1.163269</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>-0.471347</td>\n",
       "      <td>-1.202252</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.292547</td>\n",
       "      <td>-3.185521</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.083273</td>\n",
       "      <td>-2.429047</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>5.185817</td>\n",
       "      <td>2.267475</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.530970</td>\n",
       "      <td>-1.284119</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>4.336565</td>\n",
       "      <td>2.204245</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>-1.160869</td>\n",
       "      <td>0.581018</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>3.383620</td>\n",
       "      <td>-0.159761</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.572774</td>\n",
       "      <td>-3.122641</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>-1.192520</td>\n",
       "      <td>-2.030427</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>-1.682464</td>\n",
       "      <td>-1.948886</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    principal component 1  principal component 2  target\n",
       "0               -0.074827               2.854827     0.0\n",
       "1                1.291615               4.483403     0.0\n",
       "2               -4.468923               1.584395     0.0\n",
       "3               -4.376407               1.320772     0.0\n",
       "4               -4.585688               2.086770     0.0\n",
       "5                0.049923               0.170846     0.0\n",
       "6                0.196737              -1.027830     0.0\n",
       "7                0.089203              -1.163269     0.0\n",
       "8               -0.471347              -1.202252     0.0\n",
       "9                1.292547              -3.185521     1.0\n",
       "10               1.083273              -2.429047     1.0\n",
       "11               5.185817               2.267475     1.0\n",
       "12               0.530970              -1.284119     1.0\n",
       "13               4.336565               2.204245     1.0\n",
       "14              -1.160869               0.581018     1.0\n",
       "15               3.383620              -0.159761     1.0\n",
       "16               0.572774              -3.122641     1.0\n",
       "17              -1.192520              -2.030427     1.0\n",
       "18              -1.682464              -1.948886     1.0"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "principalDf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "label = data[:,2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  1.,  1.,  1.,  1.,\n",
       "        1.,  1.,  1.,  1.,  1.,  1.])"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x34ceccde80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmMAAAF3CAYAAADpZ0xtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAHGhJREFUeJzt3X2wnmV9J/Dvz5BIJGhaTVUSINR1\nUmoCSXoIdBkBkRptkbfdVl36Nm6HcbpO1SlZAXWnOzu8rOmIVjvjMl3Z+lK1qyFYbCfoKgV8axID\nIsV0HQc1h7YGOlHAQwnh2j/ysgmcvBw451zPOefzmTkz57nOnfv5cp4wzzfXfd3XU621AADQx3N6\nBwAAmMmUMQCAjpQxAICOlDEAgI6UMQCAjpQxAICOlDEAgI6UMQCAjpQxAICOlDEAgI6O6h1gLF70\nohe1xYsX944BAHBYmzdvfrC1tuBwx02pMrZ48eJs2rSpdwwAgMOqqu8fyXEuUwIAdKSMAQB0pIwB\nAHQ0pdaMAQAHt3Pnzmzbti2PPfZY7ygzytFHH51FixZl9uzZz+jPK2MAME1s27Ytxx57bBYvXpyq\n6h1nRmit5aGHHsq2bdty0kknPaNzuEwJANPEY489lhe+8IWK2CSqqrzwhS98VrORyhgATCOK2OR7\ntr9zZQwAGDfz5s0bl/P80R/9Uf74j/94XM411vPedtttOf/888f9uQ9GGQMA6EgZA4AZav2W4Zx5\n3Zdy0hWfz5nXfSnrtwyP27lba1mzZk2WLl2aZcuW5dOf/nSS5JFHHsmrX/3qrFy5MsuWLcvNN9+8\n789cffXVWbJkSc4777xs3bp13/g555yTd7zjHTnrrLNy8sknZ+PGjbnkkkvy8pe/PO9+97v3Hfe+\n970vS5cuzdKlS/P+97//iM6795N9HnzwwYz2kYuPPvpo3vzmN+e0007LihUrDsg7XtxNCTDFrd8y\nnLUbtuaBHSM5bv7crFm9JBetWNg7FgNu/ZbhXLnunozs3JUkGd4xkivX3ZMk4/L3Z926dbnrrrty\n991358EHH8xpp52Ws846KwsWLMhNN92U5z//+XnwwQdzxhln5IILLsg3v/nNfOpTn8qWLVvyxBNP\nZOXKlfmlX/qlfeebM2dObr/99nzgAx/IhRdemM2bN+dnf/Zn87KXvSzveMc7cv/99+fGG2/MN77x\njbTWcvrpp+fss8/Ok08+ecjzHs7VV1+dc889Nx/5yEeyY8eOrFq1Kuedd16OOeaYZ/072ksZA5jC\nJvoNlelr7Yat+/7e7DWyc1fWbtg6Ln937rzzzrzpTW/KrFmz8uIXvzhnn312Nm7cmNe97nW56qqr\ncvvtt+c5z3lOhoeH88///M+54447cvHFF+d5z3tekuSCCy444Hx7Hy9btiyveMUr8tKXvjRJ8vM/\n//P54Q9/mDvvvDMXX3zxvpJ0ySWX5I477siTTz55yPMezq233prPfe5z+9aZPfbYY/nBD36Qk08+\n+Zn/cp5CGQOYwib6DZXp64EdI2MaH6vW2qjjn/jEJ7J9+/Zs3rw5s2fPzuLFi/dtC3GouxKf+9zn\nJkme85zn7Pt+7+MnnnjioM93qPMeddRRefLJJ5PkoFtTtNby2c9+NkuWLDno+Z8ta8YAprCJfkNl\n+jpu/twxjY/VWWedlU9/+tPZtWtXtm/fnttvvz2rVq3Kj3/84/zcz/1cZs+enS9/+cv5/ve/v+/4\nm266KSMjI3n44YfzV3/1V2N+vvXr1+enP/1pHn300dx000155StfecjzLl68OJs3b06SfOYznxn1\nvKtXr84HP/jBfWVvy5Ytz+TXcUhmxgCmsOPmz83wKMVrvN5Qmb7WrF5ywCXuJJk7e1bWrB6fGaCL\nL744X/va13LqqaemqvLe9743L3nJS3LppZfm9a9/fYaGhrJ8+fL8wi/8QpJk5cqVecMb3pDly5fn\nxBNPzCtf+coxPd/KlSvzu7/7u1m1alWS5Pd+7/eyYsWKJDnoeS+//PL8xm/8Rj72sY/l3HPPHfW8\n73nPe/L2t789p5xySlprWbx4cW655ZZn8is5qDrUtN6gGRoaanvvegDg6WvGkt1vqNdessxlyhno\nvvvuG9NaJjd/jJ/RfvdVtbm1NnS4P2tmDGAK2/vG6Q2VZ+KiFQv9XRkAyhjAFOcNFaY2C/gBADpS\nxgAAOlLGAAA6UsYAADpSxgCAgXHNNdfs+/7+++/P0qVLn3bMpk2b8gd/8AdJkttuuy1f/epXJy3f\nRFDGAIAJ0Vrb93FDR2r/MnYwQ0ND+ZM/+ZMkyhgAMFVdszD5oxc8/euaZ7dNyv3335+TTz45v//7\nv5+VK1fmYx/7WJYtW5alS5fmne98577jPvnJTz5t/IorrsjIyEiWL1+eSy+99IDzfu9738uKFSuy\ncePG3HbbbTn//PNz//3358Mf/nCuv/76LF++PHfcccezyt6LfcYAYCZ6/JGxjY/B1q1bc+ONN+bd\n7353zjjjjGzevDk/8zM/k9e85jVZv359Vq1alXe+851PG7/uuuvyoQ99KHfddVeS3cVu7/ne+MY3\n5sYbb8zy5ctz2223Jdn92ZJvectbMm/evFx++eXPOncvZsYAgHF14okn5owzzsjGjRtzzjnnZMGC\nBTnqqKNy6aWX5vbbbz/o+Gi2b9+eCy+8MB//+MezfPnySf4vmRzdy1hVzaqqLVU1vp+6CQB0ccwx\nxyTZvWZsNGP5XOwXvOAFOf744/OVr3xlXLINou5lLMnbktzXOwQAML5OP/30/O3f/m0efPDB7Nq1\nK5/85Cdz9tlnH3Q8SWbPnp2dO3fuO8ecOXOyfv36fPSjH81f/MVfPO05jj322Dz88MOT9t80EbqW\nsapalOTXkvxZzxwAwPh76UtfmmuvvTavetWrcuqpp2blypW58MILDzqeJJdddllOOeWUAxbwH3PM\nMbnlllty/fXX5+abbz7gOV7/+tfnpptumtIL+GssU4Xj/uRVn0lybZJjk1zeWjv/UMcPDQ21TZs2\nTUo2AJhq7rvvvpx88slHdvA1C0dfrD9nXnLV8PgGmwFG+91X1ebW2tDh/my3uymr6vwkP2qtba6q\ncw5x3GVJLkuSE044YZLSAcA0p3ANjJ6XKc9MckFV3Z/kU0nOraqPP/Wg1toNrbWh1trQggULJjsj\nAMCE6lbGWmtXttYWtdYWJ3ljki+11n6zVx4AgB4G4W5KAGCc9FwLPlM929/5QJSx1tpth1u8DwAc\n2tFHH52HHnpIIZtErbU89NBDOfroo5/xOXwcEgBME4sWLcq2bduyffv23lFmlKOPPjqLFi16xn9e\nGQOAaWL27Nk56aSTesdgjAbiMiUAwEyljAEAdKSMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHSkjAEA\ndKSMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHSk\njAEAdKSMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHTUrYxV1dFV9XdVdXdV\n3VtV/7VXFgCAXo7q+Nz/muTc1tojVTU7yZ1V9Tetta93zAQAMKm6lbHWWkvyyJ6Hs/d8tV55AAB6\n6LpmrKpmVdVdSX6U5AuttW/0zAMAMNm6lrHW2q7W2vIki5KsqqqlTz2mqi6rqk1VtWn79u2THxIA\nYAINxN2UrbUdSW5L8tpRfnZDa22otTa0YMGCSc8GADCRuq0Zq6oFSXa21nZU1dwk5yX5773yAINv\n/ZbhrN2wNQ/sGMlx8+dmzeoluWjFwt6xAJ6VnndTvjTJn1fVrOyeofvL1totHfMAA2z9luFcue6e\njOzclSQZ3jGSK9fdkyQKGTCl9byb8ltJVvR6fmBqWbth674ittfIzl1Zu2GrMgZMaQOxZgzgcB7Y\nMTKmcYCpQhkDpoTj5s8d0zjAVKGMAVPCmtVLMnf2rAPG5s6elTWrl3RKBDA+ei7gBzhie9eFuZsS\nmG6UMWDKuGjFQuULmHZcpgQA6EgZAwDoSBkDAOhIGQMA6EgZAwDoSBkDAOhIGQMA6EgZAwDoSBkD\nAOhIGQMA6EgZAwDoSBkDAOhIGQMA6EgZAwDoSBkDAOjoqN4BAAAm1DULk8cfefr4nHnJVcOTn+cp\nzIwBANPbaEXsUOOTzMwYADA2Az7TNNWYGQMAxmbAZ5qmGmUMAKAjZQwAoCNlDACY3ubMG9v4JLOA\nfz/rtwxn7YateWDHSI6bPzdrVi/JRSsW9o4FADwbA35TgTK2x/otw7ly3T0Z2bkrSTK8YyRXrrsn\nSRQyANjfnHkHv5uSMetWxqrq+CQfTfKSJE8muaG19oFeedZu2LqviO01snNX1m7YqowBwP4GfKZp\nquk5M/ZEkj9srX2zqo5NsrmqvtBa+/seYR7YMTKmcQCA8dBtAX9r7R9ba9/c8/3DSe5L0m0K6rj5\nc8c0DgAwHgbibsqqWpxkRZJv9MqwZvWSzJ0964CxubNnZc3qJZ0SAQAzQfcF/FU1L8lnk7y9tfaT\nUX5+WZLLkuSEE06YsBx714W5mxIAmEzVWuv35FWzk9ySZENr7X2HO35oaKht2rRp4oMBADxLVbW5\ntTZ0uOO6XaasqkryP5PcdyRFDABgOuq5ZuzMJL+V5NyqumvP1692zAMAMOm6rRlrrd2ZpHo9PwDA\nIBiIuykBAGYqZQwAoCNlDACgI2UMAKAjZQwAoCNlDACgI2UMAKAjZQwAoKPuHxQ+FazfMuwDxAGA\nCaGMHcb6LcO5ct09Gdm5K0kyvGMkV667J0kUMgDgWXOZ8jDWbti6r4jtNbJzV9Zu2NopEQAwnShj\nh/HAjpExjQMAjIUydhjHzZ87pnEAgLFQxg5jzeolmTt71gFjc2fPyprVSzolAgCmEwv4D2PvIn13\nUwIAE0EZOwIXrViofAEAE8JlSgCAjpQxAICOXKZ8BuzIDwCMF2VsjOzIDwCMJ5cpx8iO/ADAeFLG\nxsiO/ADAeFLGxsiO/ADAeFLGxsiO/ADAeLKAf4zsyA8AjCdl7BmwIz8AMF5cpgQA6EgZAwDoSBkD\nAOioaxmrqo9U1Y+q6ts9cwAA9NJ7Zux/JXlt5wwAAN10LWOttduT/EvPDAAAPfWeGQMAmNEGvoxV\n1WVVtamqNm3fvr13HACAcTXwZay1dkNrbai1NrRgwYLecQAAxtXAlzEAgOms99YWn0zytSRLqmpb\nVf3HnnkAACZb18+mbK29qefzAwD05jIlAEBHyhgAQEfKGABAR0dUxqpq9ihjLxr/OAAAM8shy1hV\nvaqqtiV5oKpurarF+/341okMBgAwExxuZuy9SVa31hYkuSHJF6rqjD0/qwlNBgAwAxxua4s5rbV7\nk6S19pmqui/Juqq6Ikmb8HQAANPc4crYzqp6SWvtn5KktXZvVb06yS1JXjbh6QAAprnDlbErkrw4\nyT/tHWitbauqs5O8dSKDAWN0zcLk8UeePj5nXnLV8OTnAeCIHLKMtda+mCRVdUySkdbak3t+9HCS\n6yc4GzAWoxWxQ40DMBCOdJ+x/5Pkefs9fl6SL45/HACAmeVIy9jRrbV9/7ze8/3zDnE8AABH4EjL\n2KNVtXLvg6oaSjIyMZEAAGaOwy3g3+vtSf53VT2Q3VtaHJfkDROWCgBghjjSmbF7knw4yb8meTDJ\n/0hy70SFAp6BOfPGNg7AQDjSmbGPJvlJkqv3PH5Tko8l+fWJCAU8A7avAJiSjrSMLWmtnbrf4y9X\n1d0TEQgAYCY50suUW/b7TMpU1elJvjIxkQAAZo4jnRk7PclvV9UP9jw+Icl9VXVPktZaO2VC0gEA\nTHNHWsZeO6EpAABmqCMqY6217090EACAmehI14wBADABlDEAgI6UMQCAjo50AT/AwFq/ZThrN2zN\nAztGctz8uVmzekkuWrGwdyyAI6KMAVPa+i3DuXLdPRnZuStJMrxjJFeuuydJFDJgSnCZEpjS1m7Y\nuq+I7TWyc1fWbtjaKRHA2ChjwJT2wI6RMY0DDBplDJjSjps/d0zjAINGGQOmtDWrl2Tu7FkHjM2d\nPStrVi/plAhgbLqWsap6bVVtrarvVtUVPbMAU9NFKxbm2kuWZeH8uakkC+fPzbWXLLN4H5gyut1N\nWVWzkvxpkl9Jsi3Jxqr6XGvt73tlAqami1YsVL6AKavnzNiqJN9trX2vtfZ4kk8lubBjHgCASdez\njC1M8sP9Hm/bMwYAMGP0LGM1ylh72kFVl1XVpqratH379kmIBQAweXqWsW1Jjt/v8aIkDzz1oNba\nDa21odba0IIFCyYtHADAZOhZxjYmeXlVnVRVc5K8McnnOuYBAJh03e6mbK09UVVvTbIhyawkH2mt\n3dsrDwBMimsWJo8/8vTxOfOSq4YnPw/ddf2g8NbaXyf5654ZAGBSjVbEDjXOtGcHfgCAjpQxAICO\nlDEAgI6UMQCAjpQxAJhMc+aNbZxpr+vdlAAw49i+gqcwMwYA0JEyBgDQkTIGANCRMgYA0JEyBgDQ\nkTIGANCRMgYA0JEyBgDQkTIGANCRMgYA0JEyBgDQkTIGANCRMgYA0JEyBgDQkTIGANCRMgYA0JEy\nBgDQkTIGANCRMgYA0JEyBgDQkTIGANCRMgYA0JEyBgDQUZcyVlW/XlX3VtWTVTXUIwMAwCDoNTP2\n7SSXJLm90/MDAAyEo3o8aWvtviSpqh5PD8ww67cMZ+2GrXlgx0iOmz83a1YvyUUrFvaOBZCkUxkD\nmCzrtwznynX3ZGTnriTJ8I6RXLnuniRRyICBMGGXKavqi1X17VG+LhzjeS6rqk1VtWn79u0TFReY\nptZu2LqviO01snNX1m7Y2ikRwIEmbGastXbeOJ3nhiQ3JMnQ0FAbj3MCM8cDO0bGNA4w2WxtAUxr\nx82fO6ZxgMnWa2uLi6tqW5JfTvL5qtrQIwcwvazfMpwzr/tSTrri8znzui9l/ZbhrFm9JHNnzzrg\nuLmzZ2XN6iWdUgIcqEsZa63d1Fpb1Fp7bmvtxa211T1yANPH3oX6wztG0nLgQv1rL1mWhfPnppIs\nnD83116yzOJ9YGC4mxKYEJO9ncShFup/5YpzlS9gYCljwLjrsZ2EhfrAVGUBPzDuemwnYaE+MFWZ\nGQPGXY9ZqjWrlxwwG5eM40L9axYmjz/y9PE585Krhp/9+YEZzcwYMO56zFJdtGLhxC3UH62IHWoc\nYAzMjAHjbkJnqQ7hohULLdQHphxlDBh3ewuRD+cGODxlDJgQZqkAjow1YwAAHSljAIczZ97YxgHG\nwGVKgMOxfQUwgcyMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHSkjAEAdKSM\nAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHR0VO8AMK1dszB5/JGnj8+Zl1w1\nPPl5ABg4XWbGqmptVX2nqr5VVTdV1fweOWDCjVbEDjUOwIzT6zLlF5Isba2dkuQfklzZKQcAQFdd\nLlO21m7d7+HXk/z7HjlgWnApFGBKG4QF/G9O8je9Q8CU5VIowJQ2YTNjVfXFJC8Z5Ufvaq3dvOeY\ndyV5IsknDnGey5JcliQnnHDCBCQFAOhnwspYa+28Q/28qn4nyflJXt1aa4c4zw1JbkiSoaGhgx4H\nA2nOvINfQgSAdFozVlWvTfLOJGe31n7aIwNMCmu2ADiMXmvGPpTk2CRfqKq7qurDnXIAAHTV627K\nf9PjeWFacikUYEqzAz9MdS6FAkxpg7C1BQDAjKWMAQB0pIwBAHSkjAEAdKSMAQB0pIwBAHSkjAEA\ndKSMAQB0ZNNXgEF0zcKDf7KCjX5hWjEzBjCIRitihxoHpixlDACgI2UMAKAjZQwAoCNlDACgI2UM\nYBDNmTe2cWDKsrUFwCCyfQXMGGbGAAA6UsYAADpSxgAAOlLGAAA6UsYAADpSxgAAOlLGAAA6UsYA\nADpSxgAAOlLGAAA6UsYAADpSxgAAOlLGAAA66lLGquq/VdW3ququqrq1qo7rkQMAoLdeM2NrW2un\ntNaWJ7klyX/plAMAoKsuZay19pP9Hh6TpPXIAQDQ21G9nriqrk7y20l+nORVvXIAAPQ0YTNjVfXF\nqvr2KF8XJklr7V2tteOTfCLJWw9xnsuqalNVbdq+fftExQUA6KJa63uFsKpOTPL51trSwx07NDTU\nNm3aNAmpAEZxzcLk8UeePj5nXnLV8OTnAQZaVW1urQ0d7rhed1O+fL+HFyT5To8cAGMyWhE71DjA\nEei1Zuy6qlqS5Mkk30/ylk45AAC66lLGWmv/rsfzAkxLLp/ClGYHfoCpzuVTmNKUMQCAjpQxgCM1\nZ97YxgGOQLdNXwGmHOuvgAlgZgwAoCNlDGCqc/kUpjSXKQGmOpdPYUozMwYA0JEyBgDQkTIGANCR\nMgYA0JEyBgDQkTIGANCRMgYA0JEyBgDQkTIGANCRMgYA0JEyBgDQkTIGANBRtdZ6ZzhiVbU9yfd7\n55hAL0ryYO8QjMprM9i8PoPLazO4vDYT78TW2oLDHTSlyth0V1WbWmtDvXPwdF6bweb1GVxem8Hl\ntRkcLlMCAHSkjAEAdKSMDZYbegfgoLw2g83rM7i8NoPLazMgrBkDAOjIzBgAQEfK2ICqqsurqlXV\ni3pnYbeqWltV36mqb1XVTVU1v3emma6qXltVW6vqu1V1Re88/H9VdXxVfbmq7quqe6vqbb0zcaCq\nmlVVW6rqlt5ZZjplbABV1fFJfiXJD3pn4QBfSLK0tXZKkn9IcmXnPDNaVc1K8qdJXpfkF5O8qap+\nsW8q9vNEkj9srZ2c5Iwk/8nrM3DeluS+3iFQxgbV9Un+cxIL+gZIa+3W1toTex5+PcminnnIqiTf\nba19r7X2eJJPJbmwcyb2aK39Y2vtm3u+fzi73/QX9k3FXlW1KMmvJfmz3llQxgZOVV2QZLi1dnfv\nLBzSm5P8Te8QM9zCJD/c7/G2eLMfSFW1OMmKJN/om4T9vD+7/9H/ZO8gJEf1DjATVdUXk7xklB+9\nK8lVSV4zuYnY61CvTWvt5j3HvCu7L8F8YjKz8TQ1ypjZ5AFTVfOSfDbJ21trP+mdh6Sqzk/yo9ba\n5qo6p3celLEuWmvnjTZeVcuSnJTk7qpKdl8G+2ZVrWqt/dMkRpyxDvba7FVVv5Pk/CSvbvaF6W1b\nkuP3e7woyQOdsjCKqpqd3UXsE621db3zsM+ZSS6oql9NcnSS51fVx1trv9k514xln7EBVlX3Jxlq\nrfkg1wFQVa9N8r4kZ7fWtvfOM9NV1VHZfSPFq5MMJ9mY5D+01u7tGowkSe3+F+WfJ/mX1trbe+dh\ndHtmxi5vrZ3fO8tMZs0YHLkPJTk2yReq6q6q+nDvQDPZnpsp3ppkQ3YvDv9LRWygnJnkt5Kcu+f/\nl7v2zMQAT2FmDACgIzNjAAAdKWMAAB0pYwAAHSljAAAdKWMAAB0pYwD7qaoXVtWXq+qRqvpQ7zzA\n9GcHfoADPZbkPUmW7vkCmFBmxoBpr6oWV9V3qurPq+pbVfWZqnpeVZ1WVV+tqrur6u+q6tjW2qOt\ntTuzu5QBTDhlDJgpliS5obV2SpKfZPfu/Z9O8rbW2qlJzksy0jEfMEMpY8BM8cPW2lf2fP/xJKuT\n/GNrbWOStNZ+sucjlgAmlTIGzBRP/ey3n4wyBjDplDFgpjihqn55z/dvSvL1JMdV1WlJUlXHVpWb\nmoBJ54PCgWmvqhYn+esktyf5t0n+b5LfSvKKJB9MMje714ud11p7pKruT/L8JHOS7Ejymtba3096\ncGBGUMaAaW9PGbultWarCmDguEwJANCRmTEAgI7MjAEAdKSMAQB0pIwBAHSkjAEAdKSMAQB0pIwB\nAHT0/wCs1+A+glCLuAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x34ceccd5f8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "plt.figure(figsize = (10, 6))\n",
    "names = ['loadmodule', 'rootkit']\n",
    "colors = ['#1F77B4', '#FF7F0E']\n",
    "markers = ['o', 's']\n",
    "#label = numpy array of target column\n",
    "plt.xlabel('pc1')\n",
    "plt.ylabel('pc2')\n",
    "for i in range(len(names)):\n",
    "    bucket = principalDf[principalDf['target'] == i]\n",
    "    bucket = bucket.iloc[:,[0,1]].values\n",
    "    plt.scatter(bucket[:, 0], bucket[:, 1], c=colors[i], label=names[i], marker=markers[i]) \n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data = principalDf.values\n",
    "Y = data[:,2]\n",
    "X = data[:,0:2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random under-sampling and over-sampling with imbalanced-learn\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removed indexes: [    0     1     2 ..., 47440 22420  1780]\n"
     ]
    }
   ],
   "source": [
    "from imblearn.under_sampling import RandomUnderSampler\n",
    "\n",
    "rus = RandomUnderSampler(return_indices=True)\n",
    "X_rus, Y_rus, id_rus = rus.fit_sample(X, Y)\n",
    "\n",
    "print('Removed indexes:', id_rus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Y_rus = Y_rus.reshape(-1, 1)\n",
    "X_rus = np.append(X_rus, Y_rus, axis=1)\n",
    "principalDf = pd.DataFrame(data = X_rus, columns = ['principal component 1', 'principal component 2', 'target'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9d49fdc50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmMAAAGDCAYAAABnZBdiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/j/l1ZH0xwUQSAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9d49fdeb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "plt.figure(figsize = (10, 6))\n",
    "names = ['back', 'neptune']\n",
    "colors = ['#1F77B4', '#FF7F0E']\n",
    "markers = ['o', 's']\n",
    "#label = numpy array of target column\n",
    "plt.title('Principal component 1 vs 2')\n",
    "plt.xlabel('pc1')\n",
    "plt.ylabel('pc2')\n",
    "for i in range(len(names)):\n",
    "    bucket = principalDf[principalDf['target'] == i]\n",
    "    bucket = bucket.iloc[:,[0,1]].values\n",
    "    plt.scatter(bucket[:, 0], bucket[:, 1], c=colors[i], label=names[i], marker=markers[i]) \n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "target\n",
       "0.0    968\n",
       "1.0    968\n",
       "dtype: int64"
      ]
     },
     "execution_count": 188,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "principalDf.groupby('target').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "50852 new random picked points\n"
     ]
    }
   ],
   "source": [
    "from imblearn.over_sampling import RandomOverSampler\n",
    "\n",
    "ros = RandomOverSampler()\n",
    "X_ros, Y_ros = ros.fit_sample(X, Y)\n",
    "\n",
    "print(X_ros.shape[0] - X.shape[0], 'new random picked points')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Y_ros = Y_ros.reshape(-1, 1)\n",
    "X_ros = np.append(X_ros, Y_ros, axis=1)\n",
    "principalDf = pd.DataFrame(data = X_ros, columns = ['principal component 1', 'principal component 2', 'target'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9be9ba7f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmMAAAGDCAYAAABnZBdiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9be9ba7b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "plt.figure(figsize = (10, 6))\n",
    "names = ['back', 'neptune']\n",
    "colors = ['#1F77B4', '#FF7F0E']\n",
    "markers = ['o', 's']\n",
    "#label = numpy array of target column\n",
    "plt.title('Principal component 1 vs 2')\n",
    "plt.xlabel('pc1')\n",
    "plt.ylabel('pc2')\n",
    "for i in range(len(names)):\n",
    "    bucket = principalDf[principalDf['target'] == i]\n",
    "    bucket = bucket.iloc[:,[0,1]].values\n",
    "    plt.scatter(bucket[:, 0], bucket[:, 1], c=colors[i], label=names[i], marker=markers[i]) \n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "target\n",
       "0.0    51820\n",
       "1.0    51820\n",
       "dtype: int64"
      ]
     },
     "execution_count": 194,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "principalDf.groupby('target').size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Under-sampling: Tomek links"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removed indexes: [    0     1     2 ..., 52785 52786 52787]\n"
     ]
    }
   ],
   "source": [
    "from imblearn.under_sampling import TomekLinks\n",
    "\n",
    "tl = TomekLinks(return_indices=True, ratio='majority')\n",
    "X_tl, Y_tl, id_tl = tl.fit_sample(X, Y)\n",
    "\n",
    "print('Removed indexes:', id_tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(52788,)"
      ]
     },
     "execution_count": 208,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_tl.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Y_tl = Y_tl.reshape(-1, 1)\n",
    "X_tl = np.append(X_tl, Y_tl, axis=1)\n",
    "principalDf = pd.DataFrame(data = X_tl, columns = ['principal component 1', 'principal component 2', 'target'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 201,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9d4fafc50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmMAAAGDCAYAAABnZBdiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9d4faf9e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "plt.figure(figsize = (10, 6))\n",
    "names = ['back', 'neptune']\n",
    "colors = ['#1F77B4', '#FF7F0E']\n",
    "markers = ['o', 's']\n",
    "#label = numpy array of target column\n",
    "plt.title('Principal component 1 vs 2')\n",
    "plt.xlabel('pc1')\n",
    "plt.ylabel('pc2')\n",
    "for i in range(len(names)):\n",
    "    bucket = principalDf[principalDf['target'] == i]\n",
    "    bucket = bucket.iloc[:,[0,1]].values\n",
    "    plt.scatter(bucket[:, 0], bucket[:, 1], c=colors[i], label=names[i], marker=markers[i]) \n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "target\n",
       "0.0      968\n",
       "1.0    51820\n",
       "dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "principalDf.groupby('target').size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Under-sampling: Cluster Centroids"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from imblearn.under_sampling import ClusterCentroids\n",
    "\n",
    "cc = ClusterCentroids(ratio={1: 968, 0: 500})\n",
    "X_cc, Y_cc = cc.fit_sample(X, Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1468, 2)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_cc.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1468,)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_cc.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 29.62892884,  14.88659951],\n",
       "       [ 25.04042331,  12.55011015],\n",
       "       [ 23.27153278,  11.66499162],\n",
       "       ..., \n",
       "       [ -1.54301872,   0.74372408],\n",
       "       [ -1.18151537,   0.98291305],\n",
       "       [  0.40780625,   1.99713311]])"
      ]
     },
     "execution_count": 231,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_cc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Y_cc = Y_cc.reshape(-1, 1)\n",
    "X_cc = np.append(X_cc, Y_cc, axis=1)\n",
    "principalDf = pd.DataFrame(data = X_cc, columns = ['principal component 1', 'principal component 2', 'target'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1468, 3)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "principalDf.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9cdbfb630>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmMAAAGDCAYAAABnZBdiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzt3X+cVXW97/H3h3GQQdQRRa8OIOg1\n8ioKNYaJ55T90DxZUWlm2tHHvUfqpPdkdbgHPdzCrhk3yjp1TvXQY+VJI/zV+APP0Tpo5Q/IwYGQ\nkGsoAoMigWMigw4zn/vHXnvYs2etvdee/WPtH6/n48GDmbXXXuu7F5Pz7vv9fL9fc3cBAAAgGaOS\nbgAAAEAjI4wBAAAkiDAGAACQIMIYAABAgghjAAAACSKMAQAAJIgwBjQYM/t3M7u0yGv8hZltKFF7\nHjGzvynFtQCgFhHGgBpnZpvMrNfMdpvZdjP7sZmNizrf3c9191uKuae7/9bdpxVzDeRmZgvN7NY8\n51xpZp1m9oaZ/aQCbTrSzJaY2TYze9XMHjOzWeW+L1DvCGNAffiQu4+T9DZJp0lakH2CpfC/+fqy\nTdJ1kn5UofuNk/SkpLdLGi/pFknLcoV/APnxH2agjrh7t6R/l3SyNDgE+DUze0zSHknHZQ4Lmtll\nZvaomX3TzF4xs+fN7Nz09cxsfNDTti14vSM4/m4z25px3iYzu9rM/hCc92MzGxO8dpiZ3W9mO4LX\n7jeziXE+j5k1mdk1ZrbRzF4zs1VmNil47QwzezLooXnSzM7IeN8jZnadmT0e9BjeZ2aHm9ltZvbn\n4PwpGee7mf2dmT1nZn8ys8Xp4Gpmo8xsgZm9YGYvm9m/mdmhwWtTgvdeamabg/f+Y8Z1R5nZ/KD9\nO83sdjMbn++9ZvYBSddIujBo/5qIf++73b1D0s48z/FAM+sxs5Mzjk0IelSPNLMjgn+XHjPbZWa/\nDQvu7v6cu9/g7i+6e7+73yhptCR6SYEiEMaAOhIElb+S1JVx+NOS5ko6WNILIW+bJWmDpCMkfUPS\nzWZmwWs/lTRW0kmSjpT07Ry3v1jSOZKOl/QW7e+dGyXpx5KOlTRZUq+kf475kb4o6aLgMx0i6b9L\n2hMEmmWSvivpcEk3KNVDc3jGez+p1GdvC9r0RNCO8ZLWS/pK1r0+Kqldqd7FjwT3kqTLgj9nSTpO\nqd6h7PafqVQgea+kL5vZicHxv5M0R9K7JB0j6RVJ/5Lvve7+H5Kul7TU3ce5+6m5H1Nu7v6GpLuV\nepZpn5D0a3d/WdKXJG2VNEHSUUoFwbx75ZnZDKXC2B+LaR/Q6AhjQH3oMLMeSY9K+rVSv8jTfuLu\n69x9n7v3hbz3BXe/yd37lRp2OlrSUWZ2tKRzJX3W3V9x9z53/3WONvyzu29x912SvqbgF7+773T3\nu9x9j7u/Frz2rpif628kLXD3DZ6yxt13SvqgpGfd/afB51oi6RlJH8p474/dfaO7v6pUb+FGd/+V\nu++TdIekmVn3+r/uvsvdN0v6jvYHl4sl3RD0Cu2WdLWkT5rZARnvvdbde919jaQ1ktLh6TOS/tHd\ntwaBaKGk82O+t9R+pqFh7FPBMUnqU+rf/djg3/m3nmfjYjM7RKmwfm3wjAGM0AH5TwFQA+a4+68i\nXtuS570vpb9w9z1Bp9g4pXqQdrn7KzHbkHmfF5TqCZKZjVWqR+0Dkg4LXj/YzJqCAJjLJEkbQ44f\no+G9fC8o1QuWtj3j696Q77PrnELbH3KvF5T6b+dRGcdeyvh6T8a1j5X0CzMbyHi9P+Z7S225pJag\n4P4lSTMk/SJ4bbFSQfGh4N//RndfFHUhM2uRdJ+kFe7+9TK1F2gY9IwB9S/vcFOELZLGm1lrzPMn\nZXw9Wanicik1BDZN0ix3P0TSXwbHTfltUWqIMds2pYJOpsmSumO2NUxU+7PvNVnSPg0Nd1G2SDrX\n3Vsz/owJavvyGem/W/jF3Ack3a5U79inJN0f9FTK3V9z9y+5+3FK9S5+0czeG3YdMztQUodSz/oz\npWwj0KgIYwBCufuLSg3vfT8owm82s7/M8ZYrzGxiUM91jaSlwfGDleqJ6gley67VyuVfJf0fMzvB\nUk4J6sIekPQWM/uUmR1gZhdK+m+S7i/wY2aaF3zOSZI+n9H+JZK+YGZTLTVrMF3LtS/GNX8o6Wtm\ndqw0WDT/kZjt2S5pSlghfVrw2cdIapLUZGZjsoZAs/1M0oVKDb2mhyhlZueZ2X8NagX/rFTv3bBe\nSzNrlnSnUv+efx0EPABFIowByOXTStUTPSPpZUlX5Tj3Z5IekvRc8Oe64Ph3JLVI+pOkFZL+o4D7\n36BUb85DSoWEmyW1BHVj5ynV67ZT0v+SdJ67/6mAa2e7R9IqSauVmhxwc3D8R0rVRv1G0vOS9kr6\nnzGv+U+S7lVq+O81pT5/3HW57gj+3mlmT0Wcs0CpYDRf0iXB18OWNUlz95WSXldq6PXfM146QdKv\nJO1WaqLD9939kZBLnKHUcz9bqXC9O/jzFzE/E4AQlqdGEwDyMrNNkv4mR91aVTMzl3SCuzMrEEDF\n0TMGAACQIMIYAABAghimBAAASBA9YwAAAAkijAEAACSoplbgP+KII3zKlClJNwMAACCvVatW/cnd\nJ+Q7r6bC2JQpU9TZ2Zl0MwAAAPIys+xt20IxTAkAAJAgwhgAAECCCGMAAAAJqqmasTB9fX3aunWr\n9u7dm3RTasKYMWM0ceJENTc3J90UAACgOghjW7du1cEHH6wpU6bIzJJuTlVzd+3cuVNbt27V1KlT\nk24OAABQHQxT7t27V4cffjhBLAYz0+GHH04vIgAAVaTmw5gkglgBeFYAAFSXughjSdu0aZNOPvnk\noq7xyCOP6LzzzitRiwAAQK0oexgzsx+Z2ctm9nTGsYVm1m1mq4M/f1XudgAAAFSjSvSM/UTSB0KO\nf9vdZwR/HqhAOyRJHV3dmr1ouabOX6bZi5aro6u7JNfdt2+fLr30Up1yyik6//zztWfPHn31q1/V\naaedppNPPllz586Vu0uS/vjHP+p973ufTj31VL3tbW/Txo0bh1zrySef1MyZM/Xcc8+VpG0AAKB6\nlT2MuftvJO0q933i6Ojq1tV3r1V3T69cUndPr66+e21JAtmGDRs0d+5c/f73v9chhxyi73//+7ry\nyiv15JNP6umnn1Zvb6/uv/9+SdLFF1+sK664QmvWrNHjjz+uo48+evA6jz/+uD772c/qnnvu0XHH\nHVd0uwAAwH7l6pQpRpI1Y1ea2e+DYczDok4ys7lm1mlmnTt27Cjqhosf3KDevv4hx3r7+rX4wQ1F\nXVeSJk2apNmzZ0uSLrnkEj366KN6+OGHNWvWLE2fPl3Lly/XunXr9Nprr6m7u1sf/ehHJaXW/Ro7\ndqwkaf369Zo7d67uu+8+TZ48ueg2AQCA/crZKVOMpMLYDyQdL2mGpBclfSvqRHe/0d3b3b19woS8\nG5/ntK2nt6DjhciepWhm+tznPqc777xTa9eu1eWXX669e/cODlWGOfroozVmzBh1dXUV3R4AADBU\nOTtlipFIGHP37e7e7+4Dkm6S9I5K3PeY1paCjhdi8+bNeuKJJyRJS5Ys0ZlnnilJOuKII7R7927d\neeedkqRDDjlEEydOVEdHhyTpjTfe0J49eyRJra2tWrZsma655ho98sgjRbcJAADsV85OmWIkEsbM\n7OiMbz8q6emoc0tp3jnT1NLcNORYS3OT5p0zrehrn3jiibrlllt0yimnaNeuXfrbv/1bXX755Zo+\nfbrmzJmj0047bfDcn/70p/rud7+rU045RWeccYZeeumlwdeOOuoo3Xfffbriiiu0cuXKotsFAABS\nytkpUwzLNWxWkhuYLZH0bklHSNou6SvB9zMkuaRNkj7j7i/mu1Z7e7t3dnYOObZ+/XqdeOKJsdvT\n0dWtxQ9u0LaeXh3T2qJ550zTnJltsd9fDwp9ZgAA1IN0zVjmUGVLc5O+/rHpZckCZrbK3dvznVf2\nvSnd/aKQwzeX+75R5sxsa7jwBQAANPj7v9o6ZWp+o3AAAIC4qrFThu2QAAAAEkQYAwAASBBhDAAA\nIEGEMQAAgAQRxqrI6tWr9cADFdszHQAAVAHCWBUhjAEA0Hgaa2mL69ukN3cPPz56nHTNyDcJ3bRp\nk84991ydeeaZevzxx9XW1qZ77rlH27Zt0xVXXKEdO3Zo7Nixuummm/TWt75Vl112mcaMGaN169Zp\n+/btuuGGG3T22Wfry1/+snp7e/Xoo4/q6quv1vr16zVu3Dj9/d//vSTp5JNP1v333y9JofdraWnR\nxo0bQ+8JAACqU2P1jIUFsVzHC/Dss8/qiiuu0Lp169Ta2qq77rpLc+fO1fe+9z2tWrVK3/zmN/W5\nz31u8PxNmzbp17/+tZYtW6bPfvazGhgY0Fe/+lVdeOGFWr16tS688MKC7ycp5z0BAED1aayesTKa\nOnWqZsyYIUl6+9vfrk2bNunxxx/XBRdcMHjOG2+8Mfj1Jz7xCY0aNUonnHCCjjvuOD3zzDNF32/3\n7t057wkAAKoPYaxEDjzwwMGvm5qatH37drW2tmr16tWh55tZzu8l6YADDtDAwMDg93v37o28X29v\nrwYGBnLeEwAAVJ/GGqasoEMOOURTp07VHXfcIUlyd61Zs2bw9TvuuEMDAwPauHGjnnvuOU2bNk0H\nH3ywXnvttcFzpkyZoqeeekqS9NRTT+n5558v6p4AAKD6EMbK6LbbbtPNN9+sU089VSeddJLuueee\nwdemTZumd73rXTr33HP1wx/+UGPGjNFZZ52lP/zhD5oxY4aWLl2qj3/849q1a5dmzJihH/zgB3rL\nW95S1D0BAED1MXdPug2xtbe3e2dn55Bj69ev14knnhjvAmWaTVmoyy67TOedd57OP//8it0zU0HP\nDADQUBZ0rNWSlVvU764mM100a5KumzM96WbVJDNb5e7t+c5rrJqxCgYuAABqzYKOtbp1xebB7/vd\nB78nkJVPY4WxKvGTn/wk6SYAADDMkpVbIo8TxsqHmjEAACAp1RNWyHGURl2EsVqqe0sazwoAEKUp\nZJmlXMdRGjUfxsaMGaOdO3cSMmJwd+3cuVNjxoxJuikAgCp00axJBR1HadR8zdjEiRO1detW7dix\nI+mm1IQxY8Zo4sSJSTcDAFCF0nVhzKasrJpf2gIAAKAasbQFAAAoSkdXtxY/uEHbenp1TGuL5p0z\nTXNmtiXdrLpDGAMAAMN0dHXr6rvXqrevX5LU3dOrq+9eK0kEshKr+QJ+AABQeosf3DAYxNJ6+/q1\n+MENCbWofhHGAADAMNt6egs6jpFjmBIAgDpSqjqvY1pb1B0SvI5pbSlFM5GBnjEAAOpEus6ru6dX\nrv11Xh1dhe/NPO+caWppbhpyrKW5SfPOmVai1iKNMAYAQJ0oZZ3XnJlt+vrHpquttUUmqa21RV//\n2HSK98uAYUoAAOpEqeu85sxsI3xVAD1jAADUiah6Luq8qhthDACAOkGdV21imBIAgDqRHlJk1fza\nQhgDAKCOUOdVeximBAAASBA9YwAAVAE25W5chDEAABLGptyNjWFKAAASxqbcjY0wBgBAwtiUu7Ex\nTAkAQBnFqQVjU+7GRs8YAABlEnfjbhZrbWyEMQAAyiRuLRibcjc2hikBACix9NBk2NCjFF4LxmKt\njYswBgBACWUvUxGGWjBkYpgSAIASChuazEQtGLLRMwYAQIlMmb8s5+ttrKyPEIQxAABKIE4Qe2z+\neyrUGtQSwhgAADEUu3ckQ5OIQhgDACCPUuwdydAkolDADwBAHuwdiXKiZwwAUFcWdKzVkpVb1O+u\nJjNdNGuSrpszPfZ7b1uxWZ5x7JLTJ7N3JMqKMAYAqBsLOtbq1hWbB7/vdx/8Plcgu/imJ/TYxl2h\nr926YrMOGt2k198cvlxF5nphl5w+eci9M7W2NMdqPxoTYQwAUDeWrNwSevzWFZvVfuz4wbqtzGJ8\nkzSQ57phQSx7vbDr5kzX8zt2Dwt1zaNMCz98UkGfA42FMAYAqBv97pGvpQvu01+na8Ci3zGcBX9H\nzaa87fJ3Fj3rEo2HMAYAqBum6HDV29evq5auLur6LmnTog/mPIc9JlEowhgAoOale6MK6eUCqgVh\nDABQs8JmP5Ybw5AoNcIYAKBksmclzj5+vG67/J1luVf2zMlK+eLtqzUQpL/unl7Nu3ONJBZ1xcix\n6CsAoCTClod4bOMuXXzTEyW7R0dXt2Zc+5CmzF9W9iBmEccHsrrh+vpd1963rqxtQX2jZwwAUBJR\n63RFHY+ro6tbC+9dp57evqKuE1eTmfrd1Tq2Wa/siXfPuOcBYQhjAICq1NHVrWvvW1exoDPKUkGs\nL+j6ImChUghjAICyi1v0vqBjrX62cvOwocBKGHBpIMc6ZbmkV9inuB8jUfaaMTP7kZm9bGZPZxwb\nb2a/NLNng78PK3c7AADlNfv48ZGvXbV0tbp7euVKFb1/YelqLehYO+ScWV/7pW5dkUwQK0Z6hf2O\nrm5dfffaIZ/z6rvXqqOrO+kmospVooD/J5I+kHVsvqT/dPcTJP1n8D0AoIbddvk7cwayTK7UFkUd\nXd26+KYnNGX+Mm1/7c3yNrCE0sX9ba0tWnzBqZozs02LH9wwuKp/Wm9fvxY/uKHyDURNKfswpbv/\nxsymZB3+iKR3B1/fIukRSf9Q7rYAAMorcxmL2YuWq7unN+f5xa6InxRXKog9Nv89g8e2RXzWqOMM\naSItqaUtjnL3FyUp+PvIqBPNbK6ZdZpZ544dOyrWQABAcaJCSL3I/nzHtLaEnhd2nCFNZKr6dcbc\n/UZ3b3f39gkTJiTdHABATFHhpF5kf75550xTS3PTkGMtzU2ad860Ye9lSBOZkgpj283saEkK/n45\noXYAAMpk3jnT1NwUtXRqbQsLWXNmtunrH5uuttYWmVLDmF//2PTQocdChzRR35Ja2uJeSZdKWhT8\nfU9C7QAAlEk6hHzh9tUa4YoRFdU8yjRuzAGR64sdeMAovblvIGd915yZbbHqvo5pbQmtp6v33kSE\nK3sYM7MlShXrH2FmWyV9RakQdruZ/Q9JmyVdUO52AAAqLx1MaqFQ/8J3TFL7seN19d1rhwwhtjQ3\nRfZwjdS8c6aF3idsSBP1r+zDlO5+kbsf7e7N7j7R3W92953u/l53PyH4u7i9MgAAVatWZgjetSpV\nPB93qDFMR1e3Zi9arqnzl2n2ouU5C/IPPGD/r+DDxjaXPPChdrACPwAA2l9A/9j894woFKVnSKZ7\nu9IzJKWhgTT7PEna2zdQZOtRy6p+NiUAAKVkOeYUFFNAH3eGJDMpkY2eMQBAwbI38W5tadbCD58U\n2aM0+/jxemxjdVSkXDxrsh5+ZkfJC+jjzpBkJiWy0TMGAIito6tbM659SFctXT1k1mFPb5/m3bEm\nskaqkK2S4mgepcGaru9cOEObFn1Ql5w+WU1Bt1eTmS45fXLosevmTC9oTbC44i76WsjisGgM5rUw\n3zjQ3t7unZ2dSTcDABpSWK1Ttuwtggp9f5QmM/W7q8lMF82apOvmTC/4GmHtydyO6Ky3TtDDz+wY\n8fZEYZ8vbCZm3PNQ+8xslbu35zuPYUoAQKjssPL6G/vyBqlcQ21htVJxtLY0a/VXzi74fflkrgkW\nVXzf+cKu2AEtfTzffpNxz0PjIIwBAIYJCydx5BpqG0lNVPMo08IPn1Tw+3JZ0LFWS1ZuGdLT9vAz\nO0KL6m9bsVnp8aOo2ZGZsoNWuig/XyCLOg+NgZoxAMAwI+3FevHVXi3oWBv62khqoqYcMbakAWVB\nx1rdumKz+oMSnX533bpic2TYzC7kyTfrMe4G4GwUjkyEMQDAMCOd2Tfg0q0rNocGsrCi+Xyeffn1\nEbUjypKVW4q+RqFDsSxvgXwIYwCAYaJ6sQ4b2zxkdfpREWt2hYWesI204yhlb1F/jklr2R8lajmy\nkQzFsrwFcqFmDAAwTNTeiV/50NC1xKbMXxb6/qjQk72RdnrYMJcv3b5m8L3FSs/KDONKBTBXKiie\n9dYJumtVd0H7R8bdAPzQlmb19A7fkPzQluZYnwP1hTAGABgmzoy/XD1WTbmWuc9w3ZzpecNYv3tk\n4Xz2jM98sxIvmjUp5/3SQSy9PEf7sePzPoPs5THiBLioxxPzsaHOEMYAAKGye7EypQvQo1w0a1Ls\n+8RZnT9dT5Vrva44sx2vmzNdz+/YnfN+3T296ujqHvz8+Z5B5v3vWtWtt00+VCuee2VwtubH3z78\nGj17hveK5TqO+kbNGABAHV3dmr1ouabOX6bZi5bnrdPKNdsyvcp9XLdd/k41x/htlF1PNZIi+I6u\nbj21+dW894ozszHq/o9v3DVktuZdq7qHXYtV+JGJMAYADW4kyyxEFZqbNKLV8RdfMCPvTMvsoDKS\nIvi4S3bECXXFLIdRju2YULsIYwDQ4EbSw1Sqnp10j9wXlq7WgQeM0mFjUwXs2aVTYUFlJG0oZLZi\n1Ln5hmjjXCtsZinbITUuasYAoMGNpIcparZlIT072TVX6dmFrS3NOu/Uo/NuQzSSNkTNdow6N0yu\n3rX0bMw418pVj4bGQhgDgCpQ6KzAUoq7HEOmYvdX7Ojq1pduXxO6zERPb5+W/m6LFl9was7rjaQN\nYQGuuckkl/oG9rclV6jLFVIvPn1ywcthAOY5FsCrNu3t7d7Z2Zl0MwCgpLJ7iKTUL/BKDVtV+v5h\n9wtTrg3Cw4KvFD/UzV60PDS8ppfESDJYo7qY2Sp3b897HmEMAJKV75d7JVQiQKTvEXeYUJI2Lfpg\nSdtQCkmHZ9SOuGGMYUoASFg1bI1T7vqluL1h2WZc+5AWfvikqgo5xQ7RAtkIYwDK7/o26c3dw4+P\nHiddU7p9B2vVSGq2ak3cJSWy9fT2ad4dpdsOqVQovkcpsbQFgPILC2K5jjeYOGtOFbooa7Upppev\nb8BzLrMB1Dp6xgAgYVHDXtL+erLMJRPibPtTbQpZUiJMJYdsgUojjAFAFcge9squsYpa1b1WwljY\nkhLZotbokupryBbIxjAlAFShODVWSfUWjWTINGzF+UtOnzzk+4tPn5xa8ytL8yhjnS7UNXrGAKBI\n5VgWIk7QKndvUdR6XJk9XHGGTLOv8+0LZ0Se237seF173zq9smf/avzVNpsSKDXCGIDyGz0uejZl\njcseTixVPVe+GiuTdNZbJ4z4+vlEfa4DDxgVuY9l2Oct9PkwSxGNiDAGoPzqePmKqE22v3R7ccsx\n5Kuxckl3repW+7HjyxJeoj5XVHuievJybUJO6AJSCGMAUISoENLvPtgD9C8PP6tnX3598LUTjjxI\nv/ziu3NeN3uG5SizYfs4ljPUFFqPFjVkWg0L2gLVjgJ+AChCrrqtVA/Z6iFBTJKeffl1vf+GR/Je\ne87MNj02/z16ftEHNRCxdV25Qk3U5zpsbPOwNdEkac+b+0IL+aOuw+xIYD/CGAAUKHM24Z4396l5\n1PAZgGn9EWs1ZAe0fCodasIWok1/zt6+fmV/4lf29Onqu9cOC2RxFrQFGh1hDAAKkC5I7+7plSsV\nQmSSReexkqh0qMleiqK1pVkyDc5yDMuYvX39umrp6iHLXYQtaTGSDbVrfQcCIBdqxgCgAGEF6X39\nrtaWZr2xb2DIay3NTSPajzFMuTanjrssx2t79w2rWYvS3dM7ZD/JYmdIlmvGKlAtCGMAUICoGq1X\ne/v07QtnDAs22cX7aScceVDB9y71sg8LOtbqthWbh2yz9IWlq9X5wi61Hzt+SACKG8TS+gZcC+9d\nV5L2MiOzPGvZoXoQxgCgAFHrfx3T2hIalubMbNP7b3ik4NmUpRS1eGtmEEvz4Piy379YdK9eT29f\nUe9Pa/QZmfQM1j/CGAAUIGz9r3y1W5UMXtmifpGPaR4VuQ/kYC1cBduYq9cnVwBuBPQM1j8K+AGg\nAKUqSK+UqF/klQhbh41tzntO9oSIdFjMLNBv9BmZjd4z2AjoGQPqzfVt0VsP1fFK+JVUS1v2jPQX\ndtSEhI+/vU13rerOO4TZ3GT6yodOynlOR1e3vnT7mryL2ZZr8kKtaPSewUZAGAPqTVgQy3UcdS3q\nF3lY2EpraW7Swg+nglRYAMrezDvNlBribIsRltI9YlETA7JDZC0F4FIbydA4agthDADqWNQv8syw\n1d3Tq6Zgu6XsIJVrM+9iZviFDZ9motdnv0bvGWwEhDEAqHJRoSdOGMr3i7yYX+jF9FblGj6l12e4\nRu4ZbATmBa4dk6T29nbv7OxMuhlAdVt4aI7XXq1cO1ASHV3dmnfHGvUNDP1v9QlHHqStr+wd0rvU\nPMo0+oBRev3N1LHWlmYt/PBJFf8lHickzl60PHT4tMlM3/rEqQQP1AUzW+Xu7fnOYzYlAFSxq+/+\n/bAgJqX2thy2E8CADwYxKbXO17w71lR066A4syOl6BmSBDE0IoYpgXozelz0bEpUraiFWXv7Boq6\nbt+AD5mZWO6V3OOuiUUdFLAfYQyoNyxfUXPCFma9aunqkl0/XZ9ViZXcC1kTizooIIUwBiSNdcEa\nXr6ZhcVKz0ysxErurIkFFI6aMSBprAvW0Dq6ukPDS6k0j7LBIc9KrOTe6KvlAyNBGAOAhKSHDcul\ntaVZiy/YXxAf1TtVyl6rWtsuCqgGDFMC1ez6NoYq61g5hifbWlv02Pz3hL5WqZXcqQUDCkMYA6oZ\nQ5V1rRwbPecKVsxgBKoTYQwAEhJV7B7FTMq3Tne+YEWvFVB9qBkDksb6X3Wvo6tbsxct19T5yzR7\n0fLBBVDDit1zuXjW5JyvN5kV1U4AyaBnDEjaNd25tzAaCZbLqAodXd1aeO869fT2DR7LXEMsbNhw\nz5v79MqevmHXam1p1nVzpmvlczv17Muvh97volmTyvAp6l+5F8IF8iGMAfUo13IZ6eBHMCuJXJt4\nZxfLZ5p3x+rBIcPMX/xh72tpbtLCD58kSfrlF9+ti296Qo9t3DXkepecPlnXzZlehk9Y3yqxEC6Q\nD2EMKFYpeqGS2MKIyQFFy/WLPN9MyahdjubMbFPnC7u0ZOUW9buryUwff/vQwHbb5e8s3YdoQJkB\nepSZ+rMK8Uq9EC6QD2EMKFZ1ofeKAAASIklEQVQpFm2lh6omRa1onz00WYiOrm7dtap7MCD0u+uu\nVd1qP3Z8ZDhgmC2+7ACdHcTSyjHTFYhCAT8AjFDUTMi4Qez4qx/Qgo6hi77m2rIoTDpcdPf0yrW/\ndy49SQBDxV3bje2bUEmEMaCcFh6a+nM9vRT1pqOrW8XOXex3160rNg8JZIVuWVRoeGt0cXq82L4J\nlUYYAyqh0vVZLJdRdosf3KA8S37FtmTllsGvC92yqBL7TdaTqOfYZMb2TUgMNWNAPcqsQcs1wQAj\nVmjYaQopFE/LPF7olkVRC8cyzBYu6vkSwJCkRMOYmW2S9Jqkfkn73L09yfYAIxI1E7JaMDmgLKJC\nUGtLs97YNzDsl32uOqXMxVoL3bKoUvtN1gu2hEI1Ms+3t0Y5b54KY+3u/qc457e3t3tnZ2d5GwUU\nI9firQtfrVw7UHYLOtbqthWbhwxVpntYpOG/7Bc/uCHn1kdtRYQCZlMC1cnMVsXpaGKYEgAKlF5+\nIjOImTRkPbCwMDTvjjXqGwj/P8DFLDbKfpNAbUu6gN8lPWRmq8xsbsJtAYoXVYdFfVZdCZvB6JIe\nfmZH7jfmmX7JLEigMSXdMzbb3beZ2ZGSfmlmz7j7bzJPCELaXEmaPDn3JrlA4qqhPot9KctuJDMY\nFz+4QX39+ctCmAUJNJ5Ee8bcfVvw98uSfiHpHSHn3Oju7e7ePmHChEo3Eag9pdgRADkVuvyEFD9k\nMQsSaDyxwpiZNYccO6KYG5vZQWZ2cPprSWdLerqYawJAJcw7Z5pampuGHMs3gzFOyIozC7Kjq1uz\nFy3X1PnLNHvRclbaB+pAzjBmZmeZ2VZJ28zsITObkvHyQ0Xe+yhJj5rZGkm/k7TM3f+jyGsCQNnN\nmdmmr39sutpaW2IvFBoW4JpHmQ4b2xz7Gmx9BNSnfDVj35B0jruvM7Pzlarr+rS7r1DeUtTc3P05\nSacWcw2gJPItikr9FUIUOoOxFOtb5dr6qJSzKVkqA6isfGFstLuvkyR3v9PM1ku628zmSyXbCQRI\n1khqrKi/ggoPLcUuQVGJrY/SvW/p0FfMkhsA4skXxvrM7L+4+0uSFPSQvVfS/ZKOL3vrgFpTDTMZ\no3YEYHmNkkoitFRi66NK9b4B2C9fGJuvVG3XS+kD7r7VzN4l6cpyNgwoWFQQSqtEIKqGmYwMn1ZE\nEqGlElsflbv3jSFQYLicYczdfyUNznbsdfeB4KXXJH27zG0DCpMv8IS9fj2/BBpNqcJAJYYMs1Vi\nX8Vy9r4xBAqEi7vo639Kep+k9G+zsUrNpjyjHI0C8srXC1apa6CmlDIMVGLIMEy5tz4qZ+8bQ6BA\nuLiLvo5x98HfWsHXY8vTJCCGkYaozJ6wYoIY9Vc1KVcYKNRI1hqrBSNZtiOuJHoTgVoQt2fsdTN7\nm7s/JUlm1i6J//Wg9pSiJ2zhq8VfA4koZRioxJBhUgrtfYs79JtUbyJQ7eKGsask3WFm25Ra0uIY\nSReWrVVArWImY1UrdRgo95BhLShk6LcSExCAWhQ3jK2V9ENJ50j6s6T7JK0rV6MASbVZ08VMxqpG\nGCi9QurA6rk3EShG3DD2b0qFsK8F318k6aeSLihHowBJtRfEUPUIA6VX6NAvvYnAcHHD2DR3z9y6\n6OFgT0mgduUbUmS4sS4RBkqLOjCgeHHDWJeZnR7sSSkzmyXpsfI1C6gAhhSBojH0CxQvbhibJemv\nzWxz8P1kSevNbK0kd/dTytI6AEBVY+gXKF7cMPaBsrYCSKvFon2gwTH0CxQnVhhz9xfK3RDUiULD\nVPZ+kQQxAECDidszBsRTaJh6c7e08NDU1/mK47MXW02/rxAU4AMAqgxhDNWj3L1irJwPAKhCcfem\nBAAAQBkQxgAAABLEMCWKV8oZkIXs7Rh1btR1AQCoQoQxjEy5lqAoZCFWFm0FANQBhilRuHKuBbbw\n0NT1AQBoEPSMIZ5KLsbKWmMAgAZCGEO0pFfDj7p/9kKxAADUMIYpES7pICZF3z/pdgEAUEL0jGGo\naghhAAA0EHrGsB9BDACAiqNnDCPb47GcWBMMANBACGONrFwhrJDFWNPnU5APAGhQDFOi9AoNVlHB\nLaqHjJ4zAEAdoWeskdRaTRi9ZQCABkAYawS1FsIAAGggDFM2gkoGsfQQIkOJAADEQs9YPSv3LMmF\nr0a/FjbEWG2zNgEAqAKEsXpRC0ORUbMs6UUDADQwwli9qHQQG0mAoiAfAIBhCGMoXK7hSQAAUBDC\nWC1LYmiSIUUAAEqKMFaLkqoPo0cMAICSI4zViqQL9OkRAwCgLAhjtaDSS0LQAwYAQMUQxqoZ63IB\nAFD3CGPViBAGAEDDIIxVE0IYAAANhzBWDQhhAAA0LMJYkpKeIQkAABJHGEsKvWEAAEDSqKQb0HAW\nHpp8EFv4avS6YawnBgBARdEzVinVNiTJpt0AAFQFwlglJN0TBgAAqhZhrJwIYQAAIA9qxsqlWoMY\nNWEAAFQVesZKrZpqw9hjEgCAqkcYK6Vq6g2jBwwAgJpAGCuFauoNGz2OmZIAANQQasaKVU1BTKqu\ntgAAgLwIY8WotiAGAABqDsOUhap0ABs9jsAHAEAdI4zFVeoQli6wD7tmdt0XPXAAANQtwlhcpQpD\nIymwDzu/mmZuAgCAESOMVUqp1/yKGr5kSQsAAGoKYazcyrXUBMtXAABQFxINY2b2AUn/JKlJ0r+6\n+6Ik21NSrPcFAABiSCyMmVmTpH+R9H5JWyU9aWb3uvsfkmpT0QhgAACgQEn2jL1D0h/d/TlJMrOf\nS/qIpOoMY7lqtOIGsKhZkYQ4AAAaVpJhrE3Slozvt0qalX2Smc2VNFeSJk+eXJmWhSlFWIqakcmy\nFQAANKwkV+C3kGM+7ID7je7e7u7tEyZMqECzAAAAKifJMLZV0qSM7ydK2pZQWwAAABKR5DDlk5JO\nMLOpkrolfVLSpxJsT/GoCQMAAAVKLIy5+z4zu1LSg0otbfEjd1+XVHtKgpowAABQoETXGXP3ByQ9\nkGQbKopV8wEAQBZW4K8khioBAECWJAv4AQAAGh5hDAAAIEGEsVKKqv2iJgwAAESgZqyUqAkDAAAF\nomcMAAAgQYQxAACABBHGAAAAEkQYAwAASBBhDAAAIEGEMQAAgAQRxgAAABJEGAMAAEgQYQwAACBB\nhDEAAIAEEcYAAAASxN6UQCO5vk16c/fw46PHsbcqACSEnjGgkYQFsVzHAQBlR89YJnoNAABAhdEz\nloleAwAAUGGEMQAAgAQRxgAAABJEGAMayehxhR0HAJQdBfxAI2EiCgBUHXrGMtFrAAAAKoyesUz0\nGgAAgAqjZwwAACBBhDEAAIAEEcYAAAASRBgDAABIEGEMAAAgQYQxAACABBHGAAAAEkQYAwAASBBh\nDAAAIEGEMQAAgAQRxgAAABJEGAMAAEgQG4UDade3SW/uHn589Dg2kQcAlA09Y0BaWBDLdRwAgBIg\njAEAACSIMAYAAJAgwhgAAECCCGMAAAAJIowBaaPHFXYcAIASYGkLII3lKwAACaBnDAAAIEH0jAFR\nWAQWAFAB9IwBUVgEFgBQAYQxAACABBHGAAAAEkQYAwAASBBhDAAAIEHMpkT9KdUsyNHjoq8DAECJ\nEMZQf3LNglx4aOrrOMGM5SsAABVAGEP9iOoRC8PyFACAKkEYQ20rJIABAFCFKOBHbSOIAQBqHD1j\nqD65CvAlAhgAoK4QxlA98g05EsIAAHWIMIbkJVH3xfIUAIAqkUgYM7OFki6XtCM4dI27P5BEW5Cg\npIrvF75a+XsCABAhyZ6xb7v7NxO8P5KUXu+r0ugRAwBUGYYpUTlJDUeyeCsAoIolGcauNLO/ltQp\n6Uvu/krYSWY2V9JcSZo8eXIFm4eSqWQIYwgSAFBjyrbOmJn9ysyeDvnzEUk/kHS8pBmSXpT0rajr\nuPuN7t7u7u0TJkwoV3NRLpUMYgxBAgBqUNl6xtz9fXHOM7ObJN1frnYgIZWoCaMXDABQB5KaTXm0\nu78YfPtRSU8n0Q6UAdsTAQBQkKRqxr5hZjMkuaRNkj6TUDtQSknNkAQAoIYlEsbc/dNJ3BdlQggD\nAGDE2CgcxSGIAQBQFNYZw8iUuzYsXZyfb9NwAABqHGEMhatUEJNYsBUAUPcIYyhcOYIYK+UDABoU\nYQzJY70wAEADo4AfyRk9jiAGAGh49IwhnlLXiRHCAACQRBhDXKUKYoQwAACGYJgSlcGQJAAAoegZ\nQ3kxSxIAgJwIYygfesIAAMiLMIbSoicMAICCEMYQz+hxuYv4CWEAAIwIYQzxELQAACgLZlMCAAAk\niDAGAACQIMIYAABAgghjAAAACSKMAQAAJIgwBgAAkCDCGAAAQIIIYwAAAAkijAEAACSIMAYAAJAg\nwhgAAECCCGMAAAAJMndPug2xmdkOSS8k3Y4qdYSkPyXdiBrEcyscz2xkeG4jw3MbGZ7byJT6uR3r\n7hPynVRTYQzRzKzT3duTbket4bkVjmc2Mjy3keG5jQzPbWSSem4MUwIAACSIMAYAAJAgwlj9uDHp\nBtQonlvheGYjw3MbGZ7byPDcRiaR50bNGAAAQILoGQMAAEgQYazGmdkHzGyDmf3RzOYn3Z5aYWab\nzGytma02s86k21OtzOxHZvaymT2dcWy8mf3SzJ4N/j4syTZWo4jnttDMuoOfudVm9ldJtrEamdkk\nM3vYzNab2Toz+3xwnJ+5HHI8N37mcjCzMWb2OzNbEzy3a4PjU81sZfDzttTMRpe9LQxT1i4za5L0\n/yS9X9JWSU9Kusjd/5Bow2qAmW2S1O7urMOTg5n9paTdkv7N3U8Ojn1D0i53XxT8H4DD3P0fkmxn\ntYl4bgsl7Xb3bybZtmpmZkdLOtrdnzKzgyWtkjRH0mXiZy5Sjuf2CfEzF8nMTNJB7r7bzJolPSrp\n85K+KOlud/+5mf1Q0hp3/0E520LPWG17h6Q/uvtz7v6mpJ9L+kjCbUIdcfffSNqVdfgjkm4Jvr5F\nqf/oI0PEc0Me7v6iuz8VfP2apPWS2sTPXE45nhty8JTdwbfNwR+X9B5JdwbHK/LzRhirbW2StmR8\nv1X8DzAul/SQma0ys7lJN6bGHOXuL0qpXwKSjky4PbXkSjP7fTCMyVBbDmY2RdJMSSvFz1xsWc9N\n4mcuJzNrMrPVkl6W9EtJGyX1uPu+4JSK/F4ljNU2CznGuHM8s939bZLOlXRFMKwElNMPJB0vaYak\nFyV9K9nmVC8zGyfpLklXufufk25PrQh5bvzM5eHu/e4+Q9JEpUabTgw7rdztIIzVtq2SJmV8P1HS\ntoTaUlPcfVvw98uSfqHU/wgRz/agRiVdq/Jywu2pCe6+PfgP/4Ckm8TPXKigducuSbe5+93BYX7m\n8gh7bvzMxefuPZIekXS6pFYzOyB4qSK/Vwljte1JSScEMz9GS/qkpHsTblPVM7ODgiJXmdlBks6W\n9HTudyHDvZIuDb6+VNI9CbalZqTDROCj4mdumKCg+mZJ6939hoyX+JnLIeq58TOXm5lNMLPW4OsW\nSe9Tqt7uYUnnB6dV5OeN2ZQ1Lpiq/B1JTZJ+5O5fS7hJVc/MjlOqN0ySDpD0M55bODNbIundko6Q\ntF3SVyR1SLpd0mRJmyVd4O4Uq2eIeG7vVmq4yCVtkvSZdB0UUszsTEm/lbRW0kBw+Bql6p/4mYuQ\n47ldJH7mIpnZKUoV6Dcp1Tl1u7t/Nfgd8XNJ4yV1SbrE3d8oa1sIYwAAAMlhmBIAACBBhDEAAIAE\nEcYAAAASRBgDAABIEGEMAAAgQYQxAMhgZoeb2cNmttvM/jnp9gCofwfkPwUAGspeSf9b0snBHwAo\nK3rGANQ9M5tiZs+Y2S3Bpsl3mtlYMzvNzB43szVm9jszO9jdX3f3R5UKZQBQdoQxAI1imqQb3f0U\nSX+WdKWkpZI+7+6nKrUVSm+C7QPQoAhjABrFFnd/LPj6VknnSHrR3Z+UJHf/s7vvS6x1ABoWYQxA\no8je++3PIccAoOIIYwAaxWQze2fw9UWSVkg6xsxOkyQzO9jMmNQEoOLYKBxA3TOzKZIekPQbSWdI\nelbSpyWdJOl7klqUqhd7n7vvNrNNkg6RNFpSj6Sz3f0PFW84gIZAGANQ94Iwdr+7s1QFgKrDMCUA\nAECC6BkDAABIED1jAAAACSKMAQAAJIgwBgAAkCDCGAAAQIIIYwAAAAkijAEAACTo/wMkEKdkPGZk\nLAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xd9cdbfb710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "plt.figure(figsize = (10, 6))\n",
    "names = ['back', 'neptune']\n",
    "colors = ['#1F77B4', '#FF7F0E']\n",
    "markers = ['o', 's']\n",
    "#label = numpy array of target column\n",
    "plt.title('Principal component 1 vs 2')\n",
    "plt.xlabel('pc1')\n",
    "plt.ylabel('pc2')\n",
    "for i in range(len(names)):\n",
    "    bucket = principalDf[principalDf['target'] == i]\n",
    "    bucket = bucket.iloc[:,[0,1]].values\n",
    "    plt.scatter(bucket[:, 0], bucket[:, 1], c=colors[i], label=names[i], marker=markers[i]) \n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "target\n",
       "0.0    968\n",
       "1.0    968\n",
       "dtype: int64"
      ]
     },
     "execution_count": 236,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "principalDf.groupby('target').size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Over-sampling: SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.07482734,  2.85482723],\n",
       "       [ 1.29161521,  4.48340331],\n",
       "       [-4.46892273,  1.58439499],\n",
       "       [-4.37640658,  1.32077222],\n",
       "       [-4.58568763,  2.08676963],\n",
       "       [ 0.04992336,  0.17084613],\n",
       "       [ 0.19673664, -1.02782954],\n",
       "       [ 0.08920261, -1.16326857],\n",
       "       [-0.47134677, -1.2022523 ],\n",
       "       [ 1.2925473 , -3.18552102],\n",
       "       [ 1.08327325, -2.42904712],\n",
       "       [ 5.18581676,  2.26747508],\n",
       "       [ 0.53097045, -1.28411893],\n",
       "       [ 4.33656509,  2.20424516],\n",
       "       [-1.16086933,  0.58101802],\n",
       "       [ 3.38361985, -0.15976077],\n",
       "       [ 0.57277412, -3.12264072],\n",
       "       [-1.19252029, -2.03042714],\n",
       "       [-1.68246398, -1.94888564]])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from imblearn.over_sampling import SMOTE\n",
    "\n",
    "smote = SMOTE(ratio={0: 50, 1: 50})\n",
    "X_sm, Y_sm = smote.fit_sample(X, Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100,)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_sm.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Y_sm = Y_sm.reshape(-1, 1)\n",
    "X_sm = np.append(X_sm, Y_sm, axis=1)\n",
    "principalDf = pd.DataFrame(data = X_sm, columns = ['principal component 1', 'principal component 2', 'target'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x34ce678eb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmMAAAF3CAYAAADpZ0xtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzt3X903HW95/HXu2FC0wYJQhU6paS6\nnlhpoKmh9G4PlB9dg2tpQ9ersujV47o9HtejcC5ZWsC19+zS9hqP+IN7jrfnLqwCAhdsA4KeAAu1\ngIJtSWutNfd6vAWa+iNlDVAIkqbv/SOZkB8zk5lkZj7f78zzcQ6HzjffzrzJlM4rn8/78/mYuwsA\nAABhzAhdAAAAQCUjjAEAAAREGAMAAAiIMAYAABAQYQwAACAgwhgAAEBAhDEAAICACGMAAAABEcYA\nAAACIowBAAAEdFLoAvJxxhlneH19fegyAAAAJrVnz56j7j5nsvtiFcbq6+u1e/fu0GUAAABMysxe\nyOU+pikBAAACIowBAAAERBgDAAAIKFY9YwAAILOBgQEdPnxYb775ZuhSKsrMmTM1b948JRKJKf1+\nwhgAAGXi8OHDOuWUU1RfXy8zC11ORXB3vfzyyzp8+LAWLFgwpedgmhIAgDLx5ptv6vTTTyeIlZCZ\n6fTTT5/WaCRhDACAMkIQK73pfs8JYwAAoGBqa2sL8jwbN27U17/+9YI8V77Pu2PHDq1atargr50J\nYQwAACAgwhgAABWqo6tHy7c8oQXrH9HyLU+oo6unYM/t7mpra9OiRYvU2Nio++67T5J07NgxXX75\n5VqyZIkaGxv14IMPjvyeW265RQ0NDVq5cqW6u7tHrl9yySW67rrrdPHFF2vhwoXatWuX1q5dq/e9\n7326+eabR+77xje+oUWLFmnRokX65je/mdPzpk72OXr0qNIdufj666/rs5/9rC644AI1NTWNqbdQ\nWE0JADHX0dWj9s5uHenr19y6GrW1NKi1KRm6LERcR1ePNmzbr/6BQUlST1+/NmzbL0kF+fOzbds2\n7d27V/v27dPRo0d1wQUX6OKLL9acOXO0fft2veMd79DRo0e1bNkyrV69Ws8//7zuvfdedXV16fjx\n41qyZIk++MEPjjxfdXW1du7cqW9961tas2aN9uzZo3e+851673vfq+uuu06HDh3SHXfcoeeee07u\nrgsvvFArVqzQiRMnsj7vZG655RZddtlluv3229XX16elS5dq5cqVmj179rS/RymEMQCIsWJ/oKJ8\ntXd2j/y5SekfGFR7Z3dB/uw8/fTTuvrqq1VVVaV3v/vdWrFihXbt2qUPf/jDuvHGG7Vz507NmDFD\nPT09+uMf/6innnpKV111lWbNmiVJWr169ZjnSz1ubGzUueeeq7POOkuS9J73vEcvvfSSnn76aV11\n1VUjIWnt2rV66qmndOLEiazPO5lHH31UDz300Eif2ZtvvqkXX3xRCxcunPo3ZxzCGADEWLE/UFG+\njvT153U9X+6e9vrdd9+t3t5e7dmzR4lEQvX19SPbQmRblXjyySdLkmbMmDHy69Tj48ePZ3y9bM97\n0kkn6cSJE5KUcWsKd9cPf/hDNTQ0ZHz+6aJnDABirNgfqChfc+tq8rqer4svvlj33XefBgcH1dvb\nq507d2rp0qV65ZVX9K53vUuJREJPPvmkXnjhhZH7t2/frv7+fr322mv60Y9+lPfrdXR06I033tDr\nr7+u7du366KLLsr6vPX19dqzZ48k6YEHHkj7vC0tLfrOd74zEva6urqm8u3IipExAIixuXU16kkT\nvAr1gYry1dbSMGaKW5JqElVqaynMCNBVV12ln//85zr//PNlZvra176mM888U9dcc42uvPJKNTc3\na/HixXr/+98vSVqyZIk+/vGPa/HixTrnnHN00UUX5fV6S5Ys0Wc+8xktXbpUkvS5z31OTU1NkpTx\nea+//np97GMf05133qnLLrss7fN+5Stf0bXXXqvzzjtP7q76+no9/PDDU/mWZGTZhvWiprm52VOr\nHgAAE3vGpKEP1M1rG5mmrEAHDx7Mq5eJxR+Fk+57b2Z73L15st/LyBgAxFjqg5MPVExFa1OSPysR\nQBgDgJjjAxWINxr4AQAAAiKMAQAABEQYAwAACIgwBgAAEBBhDAAARMamTZtGfn3o0CEtWrRowj27\nd+/Wl770JUnSjh079LOf/axk9RUDYQwAABSFu48cN5Sr0WEsk+bmZn3729+WRBgDAJRYR1ePlm95\nQgvWP6LlW55QR1dP6JIQV5uS0sZTJ/6zaXrbpBw6dEgLFy7UF77wBS1ZskR33nmnGhsbtWjRIt1w\nww0j991zzz0Trq9fv179/f1avHixrrnmmjHP+7vf/U5NTU3atWuXduzYoVWrVunQoUP67ne/q1tv\nvVWLFy/WU089Na3aQ2GfMQCIifG77ff09WvDtv2SxD5jyN9bx/K7nofu7m7dcccduvnmm7Vs2TLt\n2bNHp512mj70oQ+po6NDS5cu1Q033DDh+pYtW3Tbbbdp7969koaCXer5PvGJT+iOO+7Q4sWLtWPH\nDklDZ0t+/vOfV21tra6//vpp1x0KI2MAEBPtnd1jjj2SpP6BQbV3dgeqCEjvnHPO0bJly7Rr1y5d\ncsklmjNnjk466SRdc8012rlzZ8br6fT29mrNmjW66667tHjx4hL/l5RG8DBmZlVm1mVmhT11EwDK\nzJE0B4Jnuw6EMnv2bElDPWPp5HMu9qmnnqqzzz5bzzzzTEFqi6LgYUzSlyUdDF0EAETd3LqavK4D\noV144YX66U9/qqNHj2pwcFD33HOPVqxYkfG6JCUSCQ0MDIw8R3V1tTo6OvT9739fP/jBDya8ximn\nnKLXXnutZP9NxRA0jJnZPEkfkfRPIesAgDhoa2lQTaJqzLWaRJXaWhoCVQRkd9ZZZ2nz5s269NJL\ndf7552vJkiVas2ZNxuuStG7dOp133nljGvhnz56thx9+WLfeeqsefPDBMa9x5ZVXavv27bFu4Ld8\nhgoL/uJmD0jaLOkUSde7+6ps9zc3N/vu3btLUhsARFFHV4/aO7t1pK9fc+tq1NbSQPM+Rhw8eFAL\nFy7M7eZNyfTN+tW10o2s0s1Xuu+9me1x9+bJfm+w1ZRmtkrSn9x9j5ldkuW+dZLWSdL8+fNLVB0A\nRFNrU5LwhcIgcEVGyGnK5ZJWm9khSfdKuszM7hp/k7tvdfdmd2+eM2dOqWsEAAAoqmBhzN03uPs8\nd6+X9AlJT7j7J0PVAwAAEEIUVlMCAIACCdkLXqmm+z2PRBhz9x2TNe8DAIDsZs6cqZdffplAVkLu\nrpdfflkzZ86c8nNwHBIAAGVi3rx5Onz4sHp7e0OXUlFmzpypefPmTfn3E8YAACgTiURCCxYsCF0G\n8hSJaUoAAIBKRRgDAAAIiDAGAAAQEGEMAAAgIMIYAABAQIQxAACAgAhjAAAAARHGAAAAAiKMAQAA\nBEQYAwAACIgwBgAAEBBhDAAAICDCGAAAQECEMQAAgIAIYwAAAAERxgAAAAIijAEAAAREGAMAAAiI\nMAYAABAQYQwAACAgwhgAAEBAhDEAAICACGMAAAABEcYAAAACIowBAAAERBgDAAAIiDAGAAAQEGEM\nAAAgoGBhzMxmmtkvzGyfmR0ws78LVQsAAEAoJwV87b9Iuszdj5lZQtLTZvYTd382YE0AAAAlFSyM\nubtLOjb8MDH8j4eqBwAAIISgPWNmVmVmeyX9SdJj7v5cyHoAAABKLWgYc/dBd18saZ6kpWa2aPw9\nZrbOzHab2e7e3t7SFwkAAFBEkVhN6e59knZIuiLN17a6e7O7N8+ZM6fktQEAABRTsJ4xM5sjacDd\n+8ysRtJKSX8fqh4A0dfR1aONDx1QX/+AJOm0WQl99cpz1dqUDFwZAExdyNWUZ0n6nplVaWiE7p/d\n/eGA9QCIsI6uHrXdv08DJ95e5/PnNwbU9sA+SSKQAYitkKspfympKdTrA4iX9s7uMUEsZWDQ1d7Z\nTRgDEFuR6BkDgMkc6euf0tcAIOoIYwBiYW5dzZS+BgBRRxgDEAttLQ1KzLAJ1xNVpraWhgAVAUBh\nhGzgB4CcpXrCWE0JoNwQxgDERmtTkuAFoOwwTQkAABAQYQwAACAgwhgAAEBAhDEAAICAaOAHgBLq\n6OpRe2e3jvT1a25djdpaGliUAFQ4whgAlEhHV482bNuv/oFBSVJPX782bNsvibM1gUrGNCUAlEh7\nZ/dIEEvpHxhUe2d3oIoARAFhDABKJNMZmpytCVQ2pikBYJpy7QObW1ejnjTBi7M1gcrGyBgQSEdX\nj5ZveUIL1j+i5VueUEdXT+iSMAWpPrCevn653u4DS/d+trU0qCZRNeZaTaKKszWBCkcYAwLI5wMc\n0ZZPH1hrU1Kb1zYqWVcjk5Ssq9HmtY007wMVjmlKIIBsH+B8MMdLvn1gnK8JYDxGxoAAaOQuH5n6\nvegDA5ArwhgQAB/g5YM+MADTRRgDAuADvHzQBwZguugZAwJIfVBzLE55oA8MiKBNSemtY9nvqa6V\nbgy/cIowBgTCBzgAFFAu4Wu8fO8vEsIYAADIT6bgE3KkKSLBairoGQMAAPnJFHxiHIhCIowBAAAE\nRBgDAAAIiDAGAAAqU3Vt6Aok0cA/RkdXD1sNAAAQR9W10VtUkCPC2LCOrh613b9PAydc0tDBzW33\n75MkAhkAAKNlCz6hRDxwZRMsjJnZ2ZK+L+lMSSckbXX3b4WqZ+NDB0aCWMrACdfGhw4QxgAAGC3G\nwSeKQo6MHZf0t+7+vJmdImmPmT3m7r8OUUxf/0Be1wEAAAohWAO/u//e3Z8f/vVrkg5KYggKAABU\nlEispjSzeklNkp4LVcNpsxJ5XQcAACiE4GHMzGol/VDSte7+apqvrzOz3Wa2u7e3t2h1fPXKc5Wo\nsjHXElWmr155btFeEwAAIGgYM7OEhoLY3e6+Ld097r7V3ZvdvXnOnDlFq6W1Kan2j56vZF2NJKnK\nTAODrvbObnV00agIAACKI1gYMzOT9L8lHXT3b4SqY7TWpqTaWhpUk6jSoL+9xcWGbfsJZAAAoChC\nrqZcLulTkvab2d7haze6+48D1qT2zm71DwyOudY/MKj2zm62uAAAIIRNydhu6JqLYGHM3Z+WZJPe\nWGJH+vrTXu/p61dHVw+BDKggnMqBihW18JOulmzXY4Yd+MeZW1ejngyBbMO2/ZLYkR+oBB1dPdqw\nbf/ISHmqZUHi7wAUSNQCz2hlHn6iJvhqyqhJ9Yylk5quBFD+srUsAAVB4MEwRsbGSf3Ee+19e9N+\nPdM0JoDykun/df4OQFnLNFqHoiKMpdHalFR7Z3fa6cq5w1tfAChvmVoW+DsAZW26QSzKU68RxjRl\nBummK2sSVWpraQhUEYBS4u8AYAqKNfVaXZvf9ZhhZCyD1HRlaoSsymxMvwgNvEB5G/13AKspgWGh\nwk+Zj6oRxrJI/aXLiiqgMrU2Jfn/HMVTXZt5Si+KNr4SuoKyRRibBJvAAgCKosxHe5A7esYmwYoq\nAEDFKPPerKhiZGwSdbMS+vMbA2mvAwBQVqY7Whe3qdeIIIxNYvi88JyvAwBQsZh6nRKmKSfxSv/E\nUbFs1wEAAPJBGJtEpg0e2fgRAAAUAmFsEpnOqnzjrePq6GI4FgAATA9hbBKtTUltXtuoupqxDft/\nfmNAG7btJ5ABAIBpIYzloLUpqdknT1zrMHpHfgAAgKlgNWWO2G8MKL6Orh6OHwJQcRgZy1Gmhv0Z\nZkxVAgXQ0dWjDdv2q6evX663jx7j/y8A5Y4wlqNMjfyD7mp7YB8fGMA0ZTt6DADKGWEsR6lG/hk2\n8WsDg66/+9GB0hcFlBFaAQBUKsJYHlqbkjqRYef9dEcmAcgde/oBqFSEMQCRkK4VoCZRpbaWhkAV\nAUBpsJoyT3U1CfWlOQpp/D5kAPKTWjXJakoAlYYwlqeNq89V2/37NDBqvjIxw7Rx9bkBqwLKQ2tT\nkvAFoOIQxvLET+8AAKCQCGNTEMWf3tksEwCAeCKMlYHUZpmpPZpSm2VKIpABABBxrKYsA2yWCQBA\nfDEyVgbYLBMAUBKbktJbxyZer66VbuQkmqkKOjJmZreb2Z/M7Fch64g7NssEAJREuiCW7TpyEnqa\n8v9IuiJwDbHHZpkAAMRX0GlKd99pZvUhaygHbLeBSsUqYgDlgJ6xMhHF7TaAYmIVMYByEXqaclJm\nts7MdpvZ7t7e3tDlAIgIVhEDKBeRHxlz962StkpSc3OzT3I7gArBKmKgiDKtmsykurZ4tVSAyIcx\nAEhnbl2NetIEL1YRAwWQLYhtfKV0dVSI0Ftb3CPp55IazOywmf2XkPUAiI9Qq4g7unq0fMsTWrD+\nES3f8oQ6uthbCcD0hF5NeXXI1wcQXyFWEbNoAEAxME0JILZKvYo426IBwhiAqYr8akoAiAoWDQAo\nBsIYAOSIo8dQMTKtjmTVZFHkNE1pZgl3Hxh37Qx3P1qcsgAgetpaGsb0jEkcPYYyxaHfJZV1ZMzM\nLjWzw5KOmNmj444uerSYhQFA1LQ2JbV5baOSdTUyScm6Gm1e20i/GIBpmWxk7GuSWtz9gJl9VNJj\nZvYpd39WkhW/PACIFo4eA1Bok4Wxanc/IEnu/oCZHZS0zczWS2I3fAAAgGmaLIwNmNmZ7v4HSRoe\nIbtc0sOS3lv06gAAAMrcZGFsvaR3S/pD6oK7HzazFZK+WMzCAOQp01ly1bU04wJAhGUNY+7+uCSZ\n2WxJ/e5+YvhLr0m6tci1AchHprPk8jnsFwBQcrnuM/Z/Jc0a9XiWpMcLXw4AAEBlyTWMzXT3kR+v\nh389K8v9AAAAyEGuYex1M1uSemBmzZI4/wMAAGCacj0o/FpJ95vZEQ1taTFX0seLVhUAAECFyHVk\nbL+k70r6i6Sjkv5R0oFiFQVgCjhLDgBiKdeRse9LelXSLcOPr5Z0p6S/LkZRAKaA7SsAIJZyDWMN\n7n7+qMdPmtm+YhQEAABQSXKdpuwys2WpB2Z2oaRnilMSAABA5ch1ZOxCSX9jZi8OP54v6aCZ7Zfk\n7n5eUaoDAAAoc7mGsSuKWgUAAECFyimMufsLxS4EAACgEuXaMwYAAIAiIIwBAAAERBgDAAAIKNcG\nfgCIrI6uHrV3dutIX7/m1tWoraVBrU3J0GUBQE4IYwBiZXzwuvT9c3TfL17SwAmXJPX09avt/qE9\nqQlkAOKAaUoAsdHR1aMN2/arp69frqHgddezL44EsZSBE66ND3F8LoB4IIwBiI32zm71DwzmdG9f\n/0CRqwGAwmCaEhiH/qPoOtLXH7oEACg4RsaAUdJNg23Ytl8dXT2hS4OkuXU1Od972qxEESsBgMIJ\nGsbM7Aoz6zaz35rZ+pC1AFL6abD+gUG1d3YHqgijtbU0qCZRldO9X73y3CJXAwCFESyMmVmVpH+Q\n9GFJH5B0tZl9IFQ9gJR5GozpsWhobUpq89pGJetqZJLqahJKVNmYe0zSJ5fNZ2oZQGyE7BlbKum3\n7v47STKzeyWtkfTrgDWhws2tq1FPmuCVz/QYiqu1KTkmaNHjByDuQoaxpKSXRj0+LOnCQLUAkoam\nwTZs2z9mqrImUaW2loaAVSGb8eEMAOImZBizNNd8wk1m6yStk6T58+cXuyZUuNSHOiMtAIBSCRnG\nDks6e9TjeZKOjL/J3bdK2ipJzc3NE8IaUGiMtAAASinkaspdkt5nZgvMrFrSJyQ9FLAeAACAkgs2\nMubux83si5I6JVVJut3dOb8EAFDeNiWlt45NvF5dK93InoaVKOgO/O7+Y0k/DlkDgNJh5SOg9EEs\n23WUPY5DAspBDH7STp1ukFqpmjrdQBKBDEBF4zgkoBzE4CdtTjcAgPQYGQNQEsU83YDpTwBxxsgY\ngJLIdIrBdE834HB3AHFHGANQEukO+S7E6QZMfyJ2qmvzu46yxzQlEGG5TL91dPWoNVB9+SjW6QaZ\npjnTnTEKREJEFtUgOghjQETluvqwvbNbK32mau3NiU8SsZ+0i3G6QabD3U3DQZXeMQARRxgDIqq9\ns1u7ZnxatTPHhawHJf3k7S0rjvT1a5Fun/D7TdK/bfxI8Qstklyb8ttaGnTdfXsnHGzrGvoeEsYA\nRB09Y0BEHenrTz/aJY3ZsqJYjfEh5dOU39qUnBDEUgqxUhMAio0wBkRQR1ePZpjldG+xGuNDyrcp\nP1mGgRRA5SCMARGTGhUa9EzjPWO1NiW1eW2jknU1Mg0Fk81rG2M9PZfvnmTlGEgBVA56xoCISTcq\nNJliNMaHlKkpP9NIV7FWagJAKRDGgIihz2lopGv0SlJp8pGucgukACoHYQyImNGjQsdismVFoTHS\nhQk2JdOftVpdy75diD3CGBAxo0eFFv1laMuKmkRV7PvA8sVIF8bIdOh9puujEeQQcYQxIGIYFQIK\nbDpBDigBwhgQQYwKAUDlYGsLAACAgBgZA6KG/hbka+OpWb72SunqADAljIwBUUN/CzBRphXEZb6y\nGJWBkTEAZS3XA8cRcdMZFa6uzTzaDEQAYQxA2UodLZXaPDZ14LgkAlklYXofEcc0JYCyle+B4wAQ\nAiNjAMrG+CnJdOdbShw5BSBaCGNA1NDfMiXppiRNkqe5N9OB4wAQAmEMiBr6W6Yk3ZSkSxMC2WQH\njscS21cAsUbPGICykGnq0SUl62pkw/+utDM+AUQfI2MAykKmHrFkXY2eWX9ZgIoAIDeMjAEoC20t\nDapJVI25VpZTkgDKTpAwZmZ/bWYHzOyEmTWHqAFAeWltSmrz2kamJAHETqhpyl9JWivpHwO9PoAy\n1NqUJHwBiJ0gYczdD0qSmYV4eQAV5uaO/brnuZc06K4qM1194dn6X62NocsCAEk08AMoU6kNYMc3\n9Q+6665nX5QkAhmASChaGDOzxyWdmeZLN7n7g3k8zzpJ6yRp/vz5BaoOQNxlG+0avwFsOvc89xJh\nDEAkFC2MufvKAj3PVklbJam5uTndZtoAKszNHftHRrekiaNd6TaAHW/Q+esEQDSwtQWA2LnnuZey\nXs/l7MkqelYBRESQnjEzu0rSdyTNkfSIme1195YQtQAonvEHd7e1NBRktWOmUa1Bd9WvfySn57j6\nwrOnXQcAFEKo1ZTbJW0P8doASiPdwd0btu2XpGkHsiqzaU0zLn/vO+kXAxAZTFMCKIq/+9GBCX1b\n/QODuva+vVq+5Ql1dE39QPRl7zltSr8vWVejb358se7+r3815dcGgEJjawsABdfR1aM/vzGQ8evT\nHSU79PLkPWGjmaR/2/KRvF8HAEqBkTEABdfe2T3pPf0Dgzndl04uDfqjza2rmdLrAEApMDIGoOBy\nDUv5hqqUuXU1EzZzzaRqhk3/sPBNSemtYxOvV9dKN059uhUAJMIYgCLINSxNdcSqraUh7aau1VWm\ntwbfbuyfXV2lW66awmHhmcLXeLncAwCTIIwBKLi2lga13b9PAycyr3isSVRNecQqFa6mvW1GrqEL\nAIqIMAag4FKhaONDB9TXP9TIPysxQycnqtT3xkBB9hxrbUpOf88yghiACCCMASiKgoSlYmFEDECE\nsJoSQOUhiAGIEMIYAABAQIQxAACAgOgZA4Dxxu8ftvHUcLUAKHuEMaCU2Dw0+ja+EroCABWGaUqg\nlDI1jtNQXlrVtcW9DgB5YGQMQOXJdxSSUUsARcTIGAAAQECEMQAAgIAIYwAAAAERxoBSohEcADAO\nDfxAKdEIDgAYh5ExAACAgAhjAAAAARHGAAAAAiKMAQAABEQYAwAACIgwBgAAEBBbWwDFtCmZ/hDw\n6lq2uQAASCKMAcWVLohluw6UAj8kAJHCNCUAVBp+SAAiJUgYM7N2M/uNmf3SzLabWV2IOoCgNp46\nNEIBAKhooUbGHpO0yN3Pk/QvkjYEqgMIi5EIAKh4QXrG3P3RUQ+flfTREHUAZYH+HwCItSj0jH1W\n0k9CFwHEFv0/ABBrRRsZM7PHJZ2Z5ks3ufuDw/fcJOm4pLuzPM86Seskaf78+UWoFAAqTHVt5tFU\nACVXtDDm7iuzfd3MPi1plaTL3d2zPM9WSVslqbm5OeN9QCRl+tADQmL6GoiUID1jZnaFpBskrXD3\nN0LUAJRE6kMvW18XAKCihdr09TZJJ0t6zMwk6Vl3/3ygWoDiYyQCAJBBqNWU/y7E6wJlif4fAIg1\njkMC4o5RNwCItShsbQEAAFCxCGMAAAABEcYAAAAComcMAOKGI7CAskIYAzBRpX/YR/2/nyOwgLJC\nGAMwUaV/2E/lvz/qAQ5AZBHGAKAQihFgMwU8AGWFBn4AiCqCGFARCGMAEEWbkqErAFAiTFMCqBxx\n6uuayqgYR2ABsUQYAzBRtvMu4xRoxsu1rysO531ufCV0BQAKhDAGYKJsoWrjqemvl1N/01RCZRwC\nHIBIIowBQCFMZ1Qw31WTBDygrNDADwCh5TuqGPXpYAB5IYwBQJwwKgaUHaYpAVSOOPZ10agPlD3C\nGID8xDHQpMR1ei/OK1gBTIowBiA/fPiXXqWfFQqUOXrGACC0TKOKcRhtBDBtjIwBiIdynqqLe/0A\npoWRMQDxwFQdgDLFyBgAFFM5j+gBKAjCGID4S3dEU1TCTiFG9OK8ghXApAhjAMpTOU1fRiFUAiga\nesYAAAACYmQMQDxkmqqrNPSgAWWHMAYgHjIFjXT9YuWMVaVA2WGaEgCKiQ1dAUyCkTEA8Rb1lYZM\nHQKYRJAwZmb/U9IaSSck/UnSZ9z9SIhaAMQcYQdAzIWapmx39/PcfbGkhyX9j0B1AAAABBUkjLn7\nq6MezpbkIeoAgNihBw0oO8F6xszsFkl/I+kVSZeGqgMAYoVpWaDsFG1kzMweN7NfpflnjSS5+03u\nfrakuyV9McvzrDOz3Wa2u7e3t1jlAgAABGHuYWcIzewcSY+4+6LJ7m1ubvbdu3eXoCoASIMNVwHk\nwcz2uHvzZPeFWk35Pnf/1+GHqyX9JkQdAJCXbBuujt58NtW/RXADkINQPWNbzKxBQ1tbvCDp84Hq\nAIDCy7YbPjvlAxgnSBhz9/8U4nUBoCwxfQrEGschAUDccV4lEGuEMQAAgIAIYwCQKzZWBVAEHBQO\nALka33+VrVdLivYB5gAigzDzhrFDAAAEd0lEQVQGAFNFczyAAmCaEgDijvMqgVhjZAwA4o4ROiDW\nGBkDAAAIiDAGAAAQEGEMAAAgIMIYAABAQIQxAACAgAhjAAAAARHGAAAAAiKMAQAABEQYAwAACIgw\nBgAAEBBhDAAAICDCGAAAQEDm7qFryJmZ9Up6IXQdRXSGpKOhi0BavDfRxvsTXbw30cV7U3znuPuc\nyW6KVRgrd2a2292bQ9eBiXhvoo33J7p4b6KL9yY6mKYEAAAIiDAGAAAQEGEsWraGLgAZ8d5EG+9P\ndPHeRBfvTUTQMwYAABAQI2MAAAABEcYiysyuNzM3szNC14IhZtZuZr8xs1+a2XYzqwtdU6UzsyvM\nrNvMfmtm60PXg7eZ2dlm9qSZHTSzA2b25dA1YSwzqzKzLjN7OHQtlY4wFkFmdrak/yDpxdC1YIzH\nJC1y9/Mk/YukDYHrqWhmViXpHyR9WNIHJF1tZh8IWxVGOS7pb919oaRlkv4b70/kfFnSwdBFgDAW\nVbdK+u+SaOiLEHd/1N2PDz98VtK8kPVASyX91t1/5+5vSbpX0prANWGYu//e3Z8f/vVrGvrQT4at\nCilmNk/SRyT9U+haQBiLHDNbLanH3feFrgVZfVbST0IXUeGSkl4a9fiw+LCPJDOrl9Qk6bmwlWCU\nb2roh/4ToQuBdFLoAiqRmT0u6cw0X7pJ0o2SPlTaipCS7b1x9weH77lJQ1Mwd5eyNkxgaa4xmhwx\nZlYr6YeSrnX3V0PXA8nMVkn6k7vvMbNLQtcDwlgQ7r4y3XUza5S0QNI+M5OGpsGeN7Ol7v6HEpZY\nsTK9Nylm9mlJqyRd7uwLE9phSWePejxP0pFAtSANM0toKIjd7e7bQteDEcslrTaz/yhppqR3mNld\n7v7JwHVVLPYZizAzOySp2d05yDUCzOwKSd+QtMLde0PXU+nM7CQNLaS4XFKPpF2S/rO7HwhaGCRJ\nNvQT5fck/T93vzZ0PUhveGTsendfFbqWSkbPGJC72ySdIukxM9trZt8NXVAlG15M8UVJnRpqDv9n\nglikLJf0KUmXDf//snd4JAbAOIyMAQAABMTIGAAAQECEMQAAgIAIYwAAAAERxgAAAAIijAEAAARE\nGAOAUczsdDN70syOmdltoesBUP7YgR8AxnpT0lckLRr+BwCKipExAGXPzOrN7Ddm9j0z+6WZPWBm\ns8zsAjP7mZntM7NfmNkp7v66uz+toVAGAEVHGANQKRokbXX38yS9qqHd+++T9GV3P1/SSkn9AesD\nUKEIYwAqxUvu/szwr++S1CLp9+6+S5Lc/dXhI5YAoKQIYwAqxfiz315Ncw0ASo4wBqBSzDezvxr+\n9dWSnpU018wukCQzO8XMWNQEoOQ4KBxA2TOzekk/lrRT0r+X9K+SPiXpXEnfkVSjoX6xle5+zMwO\nSXqHpGpJfZI+5O6/LnnhACoCYQxA2RsOYw+7O1tVAIgcpikBAAACYmQMAAAgIEbGAAAAAiKMAQAA\nBEQYAwAACIgwBgAAEBBhDAAAICDCGAAAQED/H6Xu9VnoAt9+AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x34ce678e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.clf()\n",
    "plt.figure(figsize = (10, 6))\n",
    "names = ['loadmodule', 'rootkit']\n",
    "colors = ['#1F77B4', '#FF7F0E']\n",
    "markers = ['o', 's']\n",
    "#label = numpy array of target column\n",
    "plt.xlabel('pc1')\n",
    "plt.ylabel('pc2')\n",
    "for i in range(len(names)):\n",
    "    bucket = principalDf[principalDf['target'] == i]\n",
    "    bucket = bucket.iloc[:,[0,1]].values\n",
    "    plt.scatter(bucket[:, 0], bucket[:, 1], c=colors[i], label=names[i], marker=markers[i]) \n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}