exploratory_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# KDD Cup 1999 \n",
    "# http://kdd.ics.uci.edu/databases/kddcup99/kddcup99.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature list"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Table 1: Basic features of individual TCP connections.\n",
    "|feature name | description | type |\n",
    "|-------------|-------------|------|\n",
    "|duration |length (number of seconds) of the connection | continuous |\n",
    "|protocol_type |type of the protocol, e.g. tcp, udp, etc. | discrete |\n",
    "|service | network service on the destination, e.g., http, telnet, etc. | discrete |\n",
    "|src_bytes | number of data bytes from source to destination | continuous |\n",
    "|dst_bytes | number of data bytes from destination to source | continuous |\n",
    "|flag | normal or error status of the connection | discrete |\n",
    "|land | 1 if connection is from/to the same host/port; 0 otherwise | discrete |\n",
    "|wrong_fragment | number of ''wrong'' fragments | continuous |\n",
    "|urgent | number of urgent packets | continuous |"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Table 2: Content features within a connection suggested by domain knowledge.\n",
    "| feature name\t| description | type |\n",
    "|---------------|-------------|------|\n",
    "| hot | number of ''hot'' indicators | continuous |\n",
    "| num_failed_logins | number of failed login attempts | continuous |\n",
    "| logged_in | 1 if successfully logged in; 0 otherwise | discrete |\n",
    "| num_compromised | number of ''compromised'' conditions | continuous |\n",
    "| root_shell | 1 if root shell is obtained; 0 otherwise | discrete |\n",
    "| su_attempted | 1 if ''su root'' command attempted; 0 otherwise | discrete |\n",
    "| num_root | number of ''root'' accesses | continuous |\n",
    "| num_file_creations | number of file creation operations | continuous |\n",
    "| num_shells | number of shell prompts | continuous |\n",
    "| num_access_files | number of operations on access control files | continuous |\n",
    "| num_outbound_cmds\t| number of outbound commands in an ftp session | continuous |\n",
    "| is_hot_login | 1 if the login belongs to the ''hot'' list; 0 otherwise | discrete |\n",
    "| is_guest_login | 1 if the login is a ''guest''login; 0 otherwise | discrete |"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "#%matplotlib notebook\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.patches as mpatches\n",
    "import constants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "data_10_percent = 'kddcup.data_10_percent'\n",
    "data_full = 'kddcup.data'\n",
    "data = pd.read_csv(data_10_percent, names=constants.names)\n",
    "\n",
    "# Remove Traffic features computed using a two-second time window\n",
    "data.drop(constants.traffic_features, inplace=True, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\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>root_shell</th>\n",
       "      <th>su_attempted</th>\n",
       "      <th>num_root</th>\n",
       "      <th>num_file_creations</th>\n",
       "      <th>num_shells</th>\n",
       "      <th>num_access_files</th>\n",
       "      <th>num_outbound_cmds</th>\n",
       "      <th>is_host_login</th>\n",
       "      <th>is_guest_login</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>tcp</td>\n",
       "      <td>http</td>\n",
       "      <td>SF</td>\n",
       "      <td>181</td>\n",
       "      <td>5450</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>normal.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>tcp</td>\n",
       "      <td>http</td>\n",
       "      <td>SF</td>\n",
       "      <td>239</td>\n",
       "      <td>486</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>normal.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>tcp</td>\n",
       "      <td>http</td>\n",
       "      <td>SF</td>\n",
       "      <td>235</td>\n",
       "      <td>1337</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>normal.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>tcp</td>\n",
       "      <td>http</td>\n",
       "      <td>SF</td>\n",
       "      <td>219</td>\n",
       "      <td>1337</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>normal.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>tcp</td>\n",
       "      <td>http</td>\n",
       "      <td>SF</td>\n",
       "      <td>217</td>\n",
       "      <td>2032</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>normal.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   duration protocol_type service flag  src_bytes  dst_bytes  land  \\\n",
       "0         0           tcp    http   SF        181       5450     0   \n",
       "1         0           tcp    http   SF        239        486     0   \n",
       "2         0           tcp    http   SF        235       1337     0   \n",
       "3         0           tcp    http   SF        219       1337     0   \n",
       "4         0           tcp    http   SF        217       2032     0   \n",
       "\n",
       "   wrong_fragment  urgent  hot   ...     root_shell  su_attempted  num_root  \\\n",
       "0               0       0    0   ...              0             0         0   \n",
       "1               0       0    0   ...              0             0         0   \n",
       "2               0       0    0   ...              0             0         0   \n",
       "3               0       0    0   ...              0             0         0   \n",
       "4               0       0    0   ...              0             0         0   \n",
       "\n",
       "   num_file_creations  num_shells  num_access_files  num_outbound_cmds  \\\n",
       "0                   0           0                 0                  0   \n",
       "1                   0           0                 0                  0   \n",
       "2                   0           0                 0                  0   \n",
       "3                   0           0                 0                  0   \n",
       "4                   0           0                 0                  0   \n",
       "\n",
       "   is_host_login  is_guest_login    label  \n",
       "0              0               0  normal.  \n",
       "1              0               0  normal.  \n",
       "2              0               0  normal.  \n",
       "3              0               0  normal.  \n",
       "4              0               0  normal.  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>duration</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>num_failed_logins</th>\n",
       "      <th>logged_in</th>\n",
       "      <th>num_compromised</th>\n",
       "      <th>root_shell</th>\n",
       "      <th>su_attempted</th>\n",
       "      <th>num_root</th>\n",
       "      <th>num_file_creations</th>\n",
       "      <th>num_shells</th>\n",
       "      <th>num_access_files</th>\n",
       "      <th>num_outbound_cmds</th>\n",
       "      <th>is_host_login</th>\n",
       "      <th>is_guest_login</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>494021.000000</td>\n",
       "      <td>4.940210e+05</td>\n",
       "      <td>4.940210e+05</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.000000</td>\n",
       "      <td>494021.0</td>\n",
       "      <td>494021.0</td>\n",
       "      <td>494021.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>47.979302</td>\n",
       "      <td>3.025610e+03</td>\n",
       "      <td>8.685324e+02</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.006433</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.034519</td>\n",
       "      <td>0.000152</td>\n",
       "      <td>0.148247</td>\n",
       "      <td>0.010212</td>\n",
       "      <td>0.000111</td>\n",
       "      <td>0.000036</td>\n",
       "      <td>0.011352</td>\n",
       "      <td>0.001083</td>\n",
       "      <td>0.000109</td>\n",
       "      <td>0.001008</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>707.746472</td>\n",
       "      <td>9.882181e+05</td>\n",
       "      <td>3.304000e+04</td>\n",
       "      <td>0.006673</td>\n",
       "      <td>0.134805</td>\n",
       "      <td>0.005510</td>\n",
       "      <td>0.782103</td>\n",
       "      <td>0.015520</td>\n",
       "      <td>0.355345</td>\n",
       "      <td>1.798326</td>\n",
       "      <td>0.010551</td>\n",
       "      <td>0.007793</td>\n",
       "      <td>2.012718</td>\n",
       "      <td>0.096416</td>\n",
       "      <td>0.011020</td>\n",
       "      <td>0.036482</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.037211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</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",
       "      <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>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.500000e+01</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",
       "      <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>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.200000e+02</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",
       "      <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>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.032000e+03</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",
       "      <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>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>58329.000000</td>\n",
       "      <td>6.933756e+08</td>\n",
       "      <td>5.155468e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>884.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>993.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            duration     src_bytes     dst_bytes           land  \\\n",
       "count  494021.000000  4.940210e+05  4.940210e+05  494021.000000   \n",
       "mean       47.979302  3.025610e+03  8.685324e+02       0.000045   \n",
       "std       707.746472  9.882181e+05  3.304000e+04       0.006673   \n",
       "min         0.000000  0.000000e+00  0.000000e+00       0.000000   \n",
       "25%         0.000000  4.500000e+01  0.000000e+00       0.000000   \n",
       "50%         0.000000  5.200000e+02  0.000000e+00       0.000000   \n",
       "75%         0.000000  1.032000e+03  0.000000e+00       0.000000   \n",
       "max     58329.000000  6.933756e+08  5.155468e+06       1.000000   \n",
       "\n",
       "       wrong_fragment         urgent            hot  num_failed_logins  \\\n",
       "count   494021.000000  494021.000000  494021.000000      494021.000000   \n",
       "mean         0.006433       0.000014       0.034519           0.000152   \n",
       "std          0.134805       0.005510       0.782103           0.015520   \n",
       "min          0.000000       0.000000       0.000000           0.000000   \n",
       "25%          0.000000       0.000000       0.000000           0.000000   \n",
       "50%          0.000000       0.000000       0.000000           0.000000   \n",
       "75%          0.000000       0.000000       0.000000           0.000000   \n",
       "max          3.000000       3.000000      30.000000           5.000000   \n",
       "\n",
       "           logged_in  num_compromised     root_shell   su_attempted  \\\n",
       "count  494021.000000    494021.000000  494021.000000  494021.000000   \n",
       "mean        0.148247         0.010212       0.000111       0.000036   \n",
       "std         0.355345         1.798326       0.010551       0.007793   \n",
       "min         0.000000         0.000000       0.000000       0.000000   \n",
       "25%         0.000000         0.000000       0.000000       0.000000   \n",
       "50%         0.000000         0.000000       0.000000       0.000000   \n",
       "75%         0.000000         0.000000       0.000000       0.000000   \n",
       "max         1.000000       884.000000       1.000000       2.000000   \n",
       "\n",
       "            num_root  num_file_creations     num_shells  num_access_files  \\\n",
       "count  494021.000000       494021.000000  494021.000000     494021.000000   \n",
       "mean        0.011352            0.001083       0.000109          0.001008   \n",
       "std         2.012718            0.096416       0.011020          0.036482   \n",
       "min         0.000000            0.000000       0.000000          0.000000   \n",
       "25%         0.000000            0.000000       0.000000          0.000000   \n",
       "50%         0.000000            0.000000       0.000000          0.000000   \n",
       "75%         0.000000            0.000000       0.000000          0.000000   \n",
       "max       993.000000           28.000000       2.000000          8.000000   \n",
       "\n",
       "       num_outbound_cmds  is_host_login  is_guest_login  \n",
       "count           494021.0       494021.0   494021.000000  \n",
       "mean                 0.0            0.0        0.001387  \n",
       "std                  0.0            0.0        0.037211  \n",
       "min                  0.0            0.0        0.000000  \n",
       "25%                  0.0            0.0        0.000000  \n",
       "50%                  0.0            0.0        0.000000  \n",
       "75%                  0.0            0.0        0.000000  \n",
       "max                  0.0            0.0        1.000000  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Categorical features to numeric labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "protocol_type : {0: 'icmp', 1: 'tcp', 2: 'udp'}\n",
      "service : {0: 'IRC', 1: 'X11', 2: 'Z39_50', 3: 'auth', 4: 'bgp', 5: 'courier', 6: 'csnet_ns', 7: 'ctf', 8: 'daytime', 9: 'discard', 10: 'domain', 11: 'domain_u', 12: 'echo', 13: 'eco_i', 14: 'ecr_i', 15: 'efs', 16: 'exec', 17: 'finger', 18: 'ftp', 19: 'ftp_data', 20: 'gopher', 21: 'hostnames', 22: 'http', 23: 'http_443', 24: 'imap4', 25: 'iso_tsap', 26: 'klogin', 27: 'kshell', 28: 'ldap', 29: 'link', 30: 'login', 31: 'mtp', 32: 'name', 33: 'netbios_dgm', 34: 'netbios_ns', 35: 'netbios_ssn', 36: 'netstat', 37: 'nnsp', 38: 'nntp', 39: 'ntp_u', 40: 'other', 41: 'pm_dump', 42: 'pop_2', 43: 'pop_3', 44: 'printer', 45: 'private', 46: 'red_i', 47: 'remote_job', 48: 'rje', 49: 'shell', 50: 'smtp', 51: 'sql_net', 52: 'ssh', 53: 'sunrpc', 54: 'supdup', 55: 'systat', 56: 'telnet', 57: 'tftp_u', 58: 'tim_i', 59: 'time', 60: 'urh_i', 61: 'urp_i', 62: 'uucp', 63: 'uucp_path', 64: 'vmnet', 65: 'whois'}\n",
      "flag : {0: 'OTH', 1: 'REJ', 2: 'RSTO', 3: 'RSTOS0', 4: 'RSTR', 5: 'S0', 6: 'S1', 7: 'S2', 8: 'S3', 9: 'SF', 10: 'SH'}\n"
     ]
    }
   ],
   "source": [
    "from sklearn import preprocessing\n",
    "\n",
    "le_dicts = {}\n",
    "\n",
    "for categorical_name in constants.categorical_names:\n",
    "    le = preprocessing.LabelEncoder()\n",
    "    le.fit(data[categorical_name])\n",
    "    le_dicts[categorical_name] = dict(zip(le.transform(le.classes_), le.classes_))\n",
    "    print(categorical_name, ':', le_dicts[categorical_name])\n",
    "    data[categorical_name + '_num'] = le.fit_transform(data[categorical_name])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Discrete feature analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "smurf.              280790\n",
       "neptune.            107201\n",
       "normal.              97278\n",
       "back.                 2203\n",
       "satan.                1589\n",
       "ipsweep.              1247\n",
       "portsweep.            1040\n",
       "warezclient.          1020\n",
       "teardrop.              979\n",
       "pod.                   264\n",
       "nmap.                  231\n",
       "guess_passwd.           53\n",
       "buffer_overflow.        30\n",
       "land.                   21\n",
       "warezmaster.            20\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: label, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['label'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# data normal = 0, attack = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    396743\n",
       "0     97278\n",
       "Name: label_binary_num, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['label_binary_num'] = data.label.apply(lambda label: 0 if label == 'normal.' else 1)\n",
    "data['label_binary_num'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f305aa07400>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEWCAYAAACaBstRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njYiIqJbQiIiIagmNiIioltCIiIhq/x9+2asRF6dS1wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f306167a780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data['label_four'] = data.label.apply(lambda label: constants.label_to_four_attack_class[label])\n",
    "data['label_four_num'] = data.label_four.apply(lambda label: constants.five_classes_to_num[label])\n",
    "pd.value_counts(data['label_four'], sort=True).plot.bar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f305a98b240>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEKCAYAAADuEgmxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nf9TlxFYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3066d810b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#all data\n",
    "pd.value_counts(data['protocol_type'], sort=True).plot.bar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEHCAYAAABSjBpvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3058966208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#all data according to label_binary\n",
    "pd.pivot_table(data[['protocol_type_num', 'label_binary_num']].assign(count=1),\n",
    "               index=['label_binary_num'],\n",
    "               columns=['protocol_type_num'],\n",
    "               aggfunc='count').plot(kind='bar', color=constants.my_colors)\n",
    "\n",
    "handles = [mpatches.Patch(label=le_dicts['protocol_type'][i],\n",
    "                          color=constants.my_colors[i]) for i in sorted(le_dicts['protocol_type'])]\n",
    "plt.legend(handles=handles, loc=2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"3\" halign=\"left\">count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>protocol_type</th>\n",
       "      <th>icmp</th>\n",
       "      <th>tcp</th>\n",
       "      <th>udp</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>label_binary_num</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1288</td>\n",
       "      <td>76813</td>\n",
       "      <td>19177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>282314</td>\n",
       "      <td>113252</td>\n",
       "      <td>1177</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   count               \n",
       "protocol_type       icmp     tcp    udp\n",
       "label_binary_num                       \n",
       "0                   1288   76813  19177\n",
       "1                 282314  113252   1177"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.pivot_table(data[['protocol_type', 'label_binary_num']].assign(count=1),\n",
    "               index=['label_binary_num'],\n",
    "               columns=['protocol_type'],\n",
    "               aggfunc='count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ecr_i       281400\n",
       "private     110893\n",
       "http         64293\n",
       "smtp          9723\n",
       "other         7237\n",
       "domain_u      5863\n",
       "ftp_data      4721\n",
       "eco_i         1642\n",
       "ftp            798\n",
       "finger         670\n",
       "Name: service, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['service'].value_counts()[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f30588a5080>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIcAAAFZCAYAAAACfU9jAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAJuz/Axx2+vbjatrRAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f30588aab38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#all data by service\n",
    "pd.value_counts(data['service'], sort=True).mask(lambda x: x < 200)\\\n",
    "                                           .dropna()\\\n",
    "                                           .plot(kind='bar', logy=True, figsize=(20, 5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3058679c18>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3AAAAJvCAYAAAA6HTMtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3058746fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=2, ncols=1, figsize=(15, 10))\n",
    "\n",
    "pd.pivot_table(data[['service_num', 'service', 'label_binary_num']].assign(count=1),\n",
    "               index=['service'],\n",
    "               columns=['label_binary_num'],\n",
    "               aggfunc='count')['count'][0].mask(lambda x: x < 200)\\\n",
    "                                           .dropna().sort_values(ascending=False).plot(kind='bar',\n",
    "                                                                                       logy=True,\n",
    "                                                                                       ax=axes[0])\n",
    "\n",
    "pd.pivot_table(data[['service_num', 'service', 'label_binary_num']].assign(count=1),\n",
    "               index=['service'],\n",
    "               columns=['label_binary_num'],\n",
    "               aggfunc='count')['count'][1].mask(lambda x: x < 200)\\\n",
    "                                           .dropna().sort_values(ascending=False).plot(kind='bar',\n",
    "                                                                                       logy=True,\n",
    "                                                                                       ax=axes[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>label_binary_num</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>service</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>http</th>\n",
       "      <td>61886.0</td>\n",
       "      <td>2407.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>smtp</th>\n",
       "      <td>9598.0</td>\n",
       "      <td>125.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>private</th>\n",
       "      <td>7366.0</td>\n",
       "      <td>103527.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>domain_u</th>\n",
       "      <td>5862.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>other</th>\n",
       "      <td>5632.0</td>\n",
       "      <td>1605.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ftp_data</th>\n",
       "      <td>3798.0</td>\n",
       "      <td>923.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>urp_i</th>\n",
       "      <td>537.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>finger</th>\n",
       "      <td>468.0</td>\n",
       "      <td>202.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>eco_i</th>\n",
       "      <td>389.0</td>\n",
       "      <td>1253.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ftp</th>\n",
       "      <td>373.0</td>\n",
       "      <td>425.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ecr_i</th>\n",
       "      <td>345.0</td>\n",
       "      <td>281055.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>auth</th>\n",
       "      <td>220.0</td>\n",
       "      <td>108.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>telnet</th>\n",
       "      <td>219.0</td>\n",
       "      <td>294.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pop_3</th>\n",
       "      <td>79.0</td>\n",
       "      <td>123.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time</th>\n",
       "      <td>52.0</td>\n",
       "      <td>105.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IRC</th>\n",
       "      <td>42.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>X11</th>\n",
       "      <td>9.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>domain</th>\n",
       "      <td>3.0</td>\n",
       "      <td>113.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tim_i</th>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ssh</th>\n",
       "      <td>1.0</td>\n",
       "      <td>104.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>shell</th>\n",
       "      <td>1.0</td>\n",
       "      <td>111.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    count          \n",
       "label_binary_num        0         1\n",
       "service                            \n",
       "http              61886.0    2407.0\n",
       "smtp               9598.0     125.0\n",
       "private            7366.0  103527.0\n",
       "domain_u           5862.0       1.0\n",
       "other              5632.0    1605.0\n",
       "ftp_data           3798.0     923.0\n",
       "urp_i               537.0       1.0\n",
       "finger              468.0     202.0\n",
       "eco_i               389.0    1253.0\n",
       "ftp                 373.0     425.0\n",
       "ecr_i               345.0  281055.0\n",
       "auth                220.0     108.0\n",
       "telnet              219.0     294.0\n",
       "pop_3                79.0     123.0\n",
       "time                 52.0     105.0\n",
       "IRC                  42.0       1.0\n",
       "X11                   9.0       2.0\n",
       "domain                3.0     113.0\n",
       "tim_i                 2.0       5.0\n",
       "ssh                   1.0     104.0\n",
       "shell                 1.0     111.0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# without NA in any column\n",
    "pd.pivot_table(data[['service', 'label_binary_num']].assign(count=1),\n",
    "               index=['service'],\n",
    "               columns=['label_binary_num'],\n",
    "               aggfunc='count').sort_values(('count', 0), ascending=False).dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f30586f3c50>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3AAAAFQCAYAAAD+2ZzIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3058681ba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.value_counts(data['flag'], sort=True).plot(kind='bar', logy=True, figsize=(15, 5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3AAAAE9CAYAAABKs4BMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3X10VVWa5/Hfk5uYgGhAkyZIgESuWKQgdZGM1bTLqjgW\nNFQbsGzHwDijcaGsKgbUGmuVNbatYkNZPepYIlQVabF5KUFou2QIQ+mMvJUjtJrEW+HNkOCIBCKv\nTVyxoxCy5w+uqYAm3HBPcjjJ97NWljn7nvOc380/+Kyzz97mnBMAAAAA4OKX5HcAAAAAAEB8aOAA\nAAAAICBo4AAAAAAgIGjgAAAAACAgaOAAAAAAICBo4AAAAAAgIGjgAAAAACAgaOAAAAAAICBo4AAA\nAAAgIJL9DiBJGRkZLicnx+8YAAAAAOCLioqKo865zPOd52sDZ2ZFkorC4bDKy8v9jAIAAAAAvjGz\nffGc5+sUSudcmXNuRnp6up8xAAAAACAQeAcOAAAAAAKCBg4AAAAAAuKiWMQEAAAAQLCdOnVKdXV1\n+vzzz/2OclFLS0tTdna2UlJSLuj6i2YREwAAAADBVVdXp8suu0w5OTkyM7/jXJScczp27Jjq6uqU\nm5t7QTVYxAQAAABAwj7//HNdeeWVNG8dMDNdeeWVCT2l5B04AAAAAJ6geTu/RP9GNHAAAAAAEBA0\ncAAAAAA8l5UlmXn3k5V1/nvW1dVpypQpuuaaazR8+HA98MADeuONNxSJRBSJRNSvXz9de+21ikQi\nuuuuu7R582bdcsstZ9UoKSnRq6++2kV/lcT52sCZWZGZlTY0NPgZAwAAAIDHDh3q3nrOOd122226\n9dZbVVNToz179qixsVFvvvmmotGootGoCgoK9PLLLysajWrZsmXeBuwmvq5C6Zwrk1RWUFBwn585\nAAAAAC/szbzRs1rDj7zlWa3eYOPGjUpLS9M999wjSQqFQnruueeUm5urOXPmqG/fvj4n9Ab7wAEA\nAAAIvJ07d2rs2LFnjV1++eUaOnSoamtrlZ+f/7XXvfXWW4pEIq3HH3/88VemVV5MaOAAAAAA9Fo3\n3nij1q1b13pcUlLiX5g4sIgJAAAAgMDLy8tTRUXFWWOffvqpPv74Y4XDYZ9SeY8GDgAAAEDg3Xzz\nzfq3f/u31sVJTp8+rYceekglJSU95v03iQYOAAAAQBcYOLB765mZXnvtNf3TP/2TrrnmGo0YMUJp\naWn6+c9/7m0Qn5lzzu8MKigocOXl5X7HAAAAABLSm1eh3L17t0aOHOl3jED4ur+VmVU45wrOdy37\nwAEAAABAQPjawDnnypxzM9LT0/2MAQAAAACBwDtwAAAAABAQNHAAAAAAEBA0cAAAAAAQEDRwAAAA\nABAQNHAAAAAAvJeVJZl595OVdd5bhkIhRSIRjRo1SkVFRTpx4oQk6aOPPlKfPn0UiURaf77c8Dsn\nJ0dHjx7t0j+Fl5L9DgAAAACgBzp0qNvr9enTR9FoVJJ09913a+HChfqbv/kbSdLw4cNbPwsynsAB\nAAAA6HHGjRunAwcO+B3DczRwAAAAAHqU06dPa8OGDZo8eXLr2N69e8+aQvnWW2/5mPDCMYUSAAAA\nQI/Q1NSkSCSiAwcOaOTIkRo/fnzrZ0yhbIeZFZrZW2b2GzMr9Lo+AAAAAHydL9+B27dvn5xzWrhw\nod+RPBdXA2dmL5nZYTPbcc74RDOrNrNaM/tZbNhJapSUJqnO27gAAAAA0LG+fftq/vz5evbZZ9Xc\n3Ox3HE/F+wRuiaSJbQfMLCRpoaRJkvIkTTOzPElvOecmSXpY0hzvogIAAAAIjIEDfa03ZswY5efn\na+XKlZK++g7c/PnzJUnNzc1KTU31NmsXiusdOOfcH8ws55zh6yXVOuc+lCQze0XSFOfcrtjn/yqp\n3b+Emc2QNEOShg4d2rnUAAAAAC5un3zS7bdsbGw867isrKz196ampq+cf+TIETnndNlll3V5Nq8k\nsojJYEn72xzXSfq2md0m6S8l9Ze0oL2LnXOlkkolqaCgwCWQAwBa7c280bNaw48Ec3UqAABwfmvX\nrtVPf/pTPfXUU35H6RTPV6F0zv1O0u/iOdfMiiQVhcNhr2MAAAAAQLsmT5581jYDQZFIA3dA0pA2\nx9mxsbg558oklRUUFNyXQA4AAAB0kc222bNaha7Qs1pAb5XINgLvSbrGzHLN7BJJUyWt7UwBMysy\ns9KGhoYEYgAAAABA7xDvNgIrJW2TdK2Z1ZnZdOdcs6RZkt6QtFvSaufczs7c3DlX5pybkZ6e3tnc\nAAAAANDrxLsK5bR2xtdLWu9pIgAAAADA1/J8EZPOYBETAAAAoGfKeiZLhz475Fm9gZcO1Cc/6Xhr\nglAopNGjR6u5uVm5ublavny5+vfvr5aWFj344IPauHGjzExpaWlavXq1pk6dqi+++ELHjx9XU1OT\nBg8eLElas2aNBgwYoNmzZ2vr1q1yzumGG27QCy+8IL9nDybyDlzCmEIJAAAA9ExeNm/x1uvTp4+i\n0ah27NihK664QgsXLpQkrVq1SgcPHlRVVZW2b9+u1157Tf3799c777yjaDSqJ598UsXFxYpGo4pG\no8rJydH06dN19dVXq7a2Vnv37lVubq7uvfdeT7/ThfD1CRwAAAAAdIVx48apqqpKklRfX69BgwYp\nKenM86vs7OwOr62trVVFRYVWrVrVOvbYY48pHA5r7969Gj58eNcFPw9fn8CxCiUAAAAAr50+fVob\nNmxo3eftjjvuUFlZmSKRiB566CG9//77HV6/a9cuRSIRhUKh1rFQKKRIJKKdOzu1bqPnmEIJAAAA\noEdoampSJBJRVlaWDh06pPHjx0s688SturpaTz31lJKSknTzzTdrw4YNPqe9ML42cAAAAADglS/f\ngdu3b5+cc63vwElSamqqJk2apKefflqPPPKI1qxZ026dvLw8RaNRtbS0tI61tLQoGo0qLy+vS7/D\n+TCFEgAAAECP0rdvX82fP1/PPvusmpubVVlZqYMHD0o604hVVVVp2LBh7V4fDoc1ZswYzZ07t3Vs\n7ty5uu666+T3CvpMoQQAAADguYGXDvS13pgxY5Sfn6+VK1fq8OHDKioq0qhRo5Sfn6/k5GTNmjWr\nw+sXL16sPXv2aPjw4Ro+fLj27NmjxYsXJ/IVPMEqlAAAAAA8d74927pCY2PjWcdlZWWtv0+cOLHd\n60pKSlRSUnLW2IABA/Tb3/7W03xe4B04AAAAAAgIGjgAAAAACAgWMQEAAACAgGAREwAAAAAICKZQ\nAgAAAEBA0MABAAAAQECwjQAAAAAAz739dpZOnTrkWb2UlIG64YaOtyYIhUIaPXq0mpublZubq+XL\nl6t///5qaWnRgw8+qI0bN8rMlJaWptWrV2vq1Kn64osvdPz4cTU1NWnw4MGSpDVr1mjAgAGaPXu2\ntm7dKuecbrjhBr3wwgtKT09vt15ubq4qKipUUlKipqYmff/739fzzz8vM/Ps78AiJgAAAAA852Xz\nFm+9Pn36KBqNaseOHbriiiu0cOFCSdKqVat08OBBVVVVafv27XrttdfUv39/vfPOO4pGo3ryySdV\nXFysaDSqaDSqnJwcTZ8+XVdffbVqa2u1d+9e5ebm6t577+2wniT96Ec/0j/8wz+opqZGNTU1ev31\n1z39O7CICQAAAIAeZ9y4cTpw4IAkqb6+XoMGDVJS0pn2Jzs7WwMGDGj32traWlVUVOhv//ZvW8ce\ne+wxlZeXa+/eve3Wq6+v16effqo///M/l5nprrvu0po1azz9XrwDBwAAAKBHOX36tDZs2KDJkydL\nku644w6VlZUpEonooYce0vvvv9/h9bt27VIkElEoFGodC4VCikQi2rlzZ7v1Dhw4oOzs7NZrsrOz\nW5tIr9DAAQAAAOgRmpqaFIlElJWVpUOHDmn8+PGSzjRS1dXVeuqpp5SUlKSbb75ZGzZsuOD7eF2v\nM2jgAAAAAPQIX74Dt2/fPjnnWt+Bk6TU1FRNmjRJTz/9tB555JEOpzbm5eUpGo2qpaWldaylpUXR\naFR5eXnt1hs8eLDq6upar6mrq2tdGMUrNHAAAAAAepS+fftq/vz5evbZZ9Xc3KzKykodPHhQ0plG\nrKqqSsOGDWv3+nA4rDFjxmju3LmtY3PnztV1112ncDjcbr1Bgwbp8ssv17/8y7/IOadly5ZpypQp\nnn43GjgAAAAAnktJGehrvTFjxig/P18rV67U4cOHVVRUpFGjRik/P1/JycmaNWtWh9cvXrxYe/bs\n0fDhwzV8+HDt2bNHixcvlqQO6/3qV7/Svffeq3A4rOHDh2vSpEkX9oXbYc45TwteiIKCAldeXu53\nDAA9wN7MGz2rNfzIW57VAnAR8XA/Jl0E/x/V1TbbZs9qFbpCz2pdrHrzv0O7d+/WyJEj/Y4RCF/3\ntzKzCudcwfmuZR84AAAAAAgI9oEDAAAAgIDgHTgAAAAACAgaOAAAAAAICBo4AAAAAAgIGjgAAAAA\nCIhkvwMAAAAA6Hmy3n5bh06d8qzewJQUfXLDDR2eEwqFNHr0aDU3Nys3N1fLly9X//791dLSogcf\nfFAbN26UmSktLU2rV6/W1KlT9cUXX+j48eNqamrS4MGDJUlr1qxRYWGhLrvsMpmZBgwYoGXLlnW4\n+Xd34QkcAAAAAM952bzFW69Pnz6KRqPasWOHrrjiCi1cuFCStGrVKh08eFBVVVXavn27XnvtNfXv\n31/vvPOOotGonnzySRUXFysajSoajSonJ0eStGnTJlVVVamwsFBz58719PtcKBo4AAAAAD3OuHHj\ndODAAUlSfX29Bg0apKSkM+1Pdna2BgwYcEG1/EYDBwAAAKBHOX36tDZs2KDJkydLku644w6VlZUp\nEonooYce0vvvv9+peq+//rpuvfXWrojaaTRwAAAAAHqEpqYmRSIRZWVl6dChQxo/frykM0/cqqur\n9dRTTykpKUk333yzNmzYcN56N910kwYPHqzf//73mjZtWlfHj0uXNHBmdqmZlZvZLV1RHwAAAADO\n9eU7cPv27ZNzrvUdOElKTU3VpEmT9PTTT+uRRx7RmjVrzltv06ZN2rdvnyKRiB5//PGujB63uBo4\nM3vJzA6b2Y5zxieaWbWZ1ZrZz9p89LCk1V4GBQAAAIB49O3bV/Pnz9ezzz6r5uZmVVZW6uDBg5Kk\nlpYWVVVVxb2iZHJysn75y19q2bJlOn78eFfGjku8T+CWSJrYdsDMQpIWSpokKU/SNDPLM7PxknZJ\nOuxhTgAAAAABMjAlxdd6Y8aMUX5+vlauXKnDhw+rqKhIo0aNUn5+vpKTkzVr1qy4aw0aNEjTpk07\n64meX+LaB8459wczyzln+HpJtc65DyXJzF6RNEVSP0mX6kxT12Rm651zLefWNLMZkmZI0tChQy80\nPwAAAICL0Pn2bOsKjY2NZx2XlZW1/j5x4sRzT29VUlKikpKSs8Y++uijs45feOGFhPN5IZGNvAdL\n2t/muE7St51zsyTJzEokHf265k2SnHOlkkolqaCgwCWQAwAAAAB6hUQauA4555ac7xwzK5JUFA6H\nuyoGAAAAAPQYiaxCeUDSkDbH2bGxuDnnypxzM9LT0xOIAQAAAAC9QyIN3HuSrjGzXDO7RNJUSWs7\nU8DMisystKGhIYEYAAAAANA7xLuNwEpJ2yRda2Z1ZjbdOdcsaZakNyTtlrTaObezMzfnCRwAAAAA\nxC/eVSi/dttx59x6Ses9TQQAAAAA+FpdtohJPFjEBAAAAOiZ3s56W6cOnfKsXsrAFN3wScdbE8yb\nN08rVqxQKBRSUlKSFi1apD/7sz/T1KlTdezYMY0dO1bLly/XJZdc4lmu7pbIO3AJYwolAAAA0DN5\n2bzFU2/btm1at26dKisrVVVVpTfffFNDhgzRww8/rB//+Meqra3VgAEDtHjxYk9zdTdfGzgAAAAA\n8EJ9fb0yMjKUmpoqScrIyNCgQYO0ceNG3X777ZKku+++W2vWrPEzZsJ8beBYhRIAAACAFyZMmKD9\n+/drxIgRmjlzprZs2aJjx46pf//+Sk4+8+ZYdna2Dhzo1M5nFx2mUAIAAAAIvH79+qmiokKlpaXK\nzMxUcXGxlixZ4ncsz/m6iAl6KTPvajnnXS0AAAAEWigUUmFhoQoLCzV69GgtXbpUJ06cUHNzs5KT\nk1VXV6fBgwf7HTMhvAMHAAAAIPCqq6tVU1PTehyNRjVs2DDddNNNevXVVyVJS5cu1ZQpU/yK6Am2\nEQAAAADguZSBKZ5vI9CRxsZGzZ49WydOnFBycrLC4bBKS0v16aefaurUqXr00Uc1ZswYTZ8+3bNM\nfvC1gXPOlUkqKygouM/PHAAAAAC8db4927w2duxYbd269SvjGRkZevfdd7s1S1diCiUAAAAABAQN\nHAAAAAAEBPvAAQAAAEBAsA8cAAAAAAQEUygBAAAAICBo4AAAAAAgIGjgAAAAAHguKytLZubZT1ZW\n1nnvOW/ePH3zm99Ufn6+IpGI3nnnHS1YsEDhcFhmpqNHj3bDN+9avu4DBwAAAKBnOnToULfW27Zt\nm9atW6fKykqlpqbq6NGjOnnypC655BLdcsstKiws9DSPX3xt4MysSFJROBz2MwYAAACAgKuvr1dG\nRoZSU1MlndnAW5KuuuoqP2N5jlUoAQAAAATehAkTtH//fo0YMUIzZ87Uli1b/I7UJXgHDgAAAEDg\n9evXTxUVFSotLVVmZqaKi4u1ZMkSv2N5jnfgAAAAAPQIoVBIhYWFKiws1OjRo7V06VKVlJT4HctT\nPIEDAAAAEHjV1dWqqalpPY5Goxo2bJiPiboGDRwAAAAAzw0cOLBb6zU2Nuruu+9WXl6e8vPztWvX\nLj3xxBOaP3++srOzVVdXp/z8fN17772e5upuTKEEAAAA4LlPPvmkW+83duxYbd269Svj999/v+6/\n//5uzdKVeAIHAAAAAAHhawNnZkVmVtrQ0OBnDAAAAAAIBPaBAwAAAICAYAolAAAAAAQEDRwAAAAA\nBAQNHAAAAAAEBNsIAAAAAPDe1qh0qtm7einJ0l9EOjxl3rx5WrFihUKhkJKSkrRo0SLNnz9f5eXl\nSklJ0fXXX69FixYpJSXFu1zdjCdwAAAAALznZfMWR71t27Zp3bp1qqysVFVVld58800NGTJEd955\npz744ANt375dTU1NevHFF73N1c14AgcAAAAg8Orr65WRkaHU1FRJUkZGhiTpqquuaj3n+uuvV11d\nnS/5vMITOAAAAACBN2HCBO3fv18jRozQzJkztWXLlrM+P3XqlJYvX66JEyf6lNAbNHAAAAAAAq9f\nv36qqKhQaWmpMjMzVVxcrCVLlrR+PnPmTH3nO9/RjTfe6F9ID3g+hdLMRkp6QFKGpA3OuV97fQ8A\nAAAAOFcoFFJhYaEKCws1evRoLV26VCUlJZozZ46OHDmiRYsW+R0xYXE9gTOzl8zssJntOGd8oplV\nm1mtmf1Mkpxzu51zP5R0h6QbvI8MAAAAAGerrq5WTU1N63E0GtWwYcP04osv6o033tDKlSuVlBT8\nCYjxPoFbImmBpGVfDphZSNJCSeMl1Ul6z8zWOud2mdlkST+StNzbuAAAAAACISXZ+20EOtDY2KjZ\ns2frxIkTSk5OVjgcVmlpqbKysjRs2DCNGzdOknTbbbfpscce8y5XN4urgXPO/cHMcs4Zvl5SrXPu\nQ0kys1ckTZG0yzm3VtJaM/tfklZ8XU0zmyFphiQNHTr0gsIDAAAAuEidZ882r40dO1Zbt279ynhz\ns8fbGfgskXfgBkva3+a4TtK3zaxQ0m2SUiWtb+9i51yppFJJKigocAnkAAAAAIBewfNFTJxzmyVt\njudcMyuSVBQOh72OAQAAAAA9TiIN3AFJQ9ocZ8fG4uacK5NUVlBQcF8COQAAANCGbd7sWa1NnlUC\n4IVElmF5T9I1ZpZrZpdImipprTexAAAAAADnincbgZWStkm61szqzGy6c65Z0ixJb0jaLWm1c25n\nZ25uZkVmVtrQ0NDZ3AAAAADQ68S7CuW0dsbXq4OFSuKoyxRKAAAAAIiT54uYdAaLmAAAAAA903/K\n+h86cegzz+r1H3ipfvvJf+3wnHnz5mnFihUKhUJKSkrSokWLVFpaqvLycjnnNGLECC1ZskT9+vXz\nLFd383UrcudcmXNuRnp6up8xAAAAAHjMy+Ytnnrbtm3TunXrVFlZqaqqKr355psaMmSInnvuOf3x\nj39UVVWVhg4dqgULFniaq7v5+gQOAAAAALxQX1+vjIwMpaamSpIyMjLO+tw5p6amJpmZH/E84+sT\nOBYxAQAAAOCFCRMmaP/+/RoxYoRmzpypLVu2tH52zz33KCsrSx988IFmz57tY8rEMYUSAAAAQOD1\n69dPFRUVKi0tVWZmpoqLi7VkyRJJ0j/+4z/q4MGDGjlypFatWuVv0AT52sABAAAAgFdCoZAKCws1\nZ84cLViwQP/8z/981mdTp049ayyIaOAAAAAABF51dbVqampaj6PRqIYOHara2lpJZ96BW7t2rb7x\njW/4FdETbCMAAABwgWyOd4shuMedZ7WAi0H/gZd6vo1ARxobGzV79mydOHFCycnJCofD+s1vfqMf\n/OAH+vTTT+Wc07e+9S39+te/9iyTH3xt4NjIGwAAAOiZzrdnm9fGjh2rrVu3fmX87bff7tYcXY0p\nlAAAAAAQEDRwAAAAABAQ7AMHAAAAAAHBPnAAAAAAEBBMoQQAAACAgKCBAwAAAICA8HUbAQAAAAA9\n00d5U3T6yHHP6oUyr1DOrv/Z4Tnz5s3TihUrFAqFlJSUpEWLFunhhx9WfX29+vTpI0l69NFHdfvt\nt3uWq7vRwAEAAADwnJfNWzz1tm3bpnXr1qmyslKpqak6evSoTp48KUl6+eWXVVBQ4Gkev/jawJlZ\nkaSicDjsZwwAAAAAAVdfX6+MjAylpqZKkjIyMnxO1DVYhRIAAABA4E2YMEH79+/XiBEjNHPmTG3Z\nsqX1szvvvFORSESRSETHjh3zMWXimEIJAAAAIPD69euniooKvfXWW9q0aZOKi4v1i1/8QhJTKAEA\nAADgohMKhVRYWKjCwkKNHj1aS5cu9TuS59hGAAAAAEDgVVdXq6ampvU4Go1q2LBhPibqGjRwAAAA\nADwXyryiW+s1Njbq7rvvVl5envLz87Vr1y498cQTnma4GDCFEgAAAIDnzrdnm9fGjh2rrVu3fmV8\n8+bN3Zqjq/EEDgAAAAACwtcGzsyKzKy0oaHBzxgAAAAAEAjsAwcAAAAAAcEUSgAAAAAICBo4AAAA\nAAgIGjgAAAAACAi2EQAAAADguWeeeUafffaZZ/UuvfRS/eQnP+nwnHnz5mnFihUKhUJKSkrSokWL\n9PDDD+uZZ55RQUGBJOmjjz7SLbfcoh07dniWrTvRwAEAAADwnJfNWzz1tm3bpnXr1qmyslKpqak6\nevSoTp486WmGiwENHAAAAIDAq6+vV0ZGhlJTUyVJGRkZPifqGjRwAAAAAAJvwoQJevLJJzVixAh9\n73vfU3Fxsb773e9Kku6880716dNHknTy5EklJQV3KZDgJgcAAACAmH79+qmiokKlpaXKzMxUcXGx\nlixZIkl6+eWXFY1GFY1GtX79en+DJqhLnsCZ2a2S/krS5ZIWO+f+d1fcBwAAAAC+FAqFVFhYqMLC\nQo0ePVpLly71O5Ln4n4CZ2YvmdlhM9txzvhEM6s2s1oz+5kkOefWOOfuk/RDScXeRgYAAACAs1VX\nV6umpqb1OBqNatiwYT4m6hqdeQK3RNICScu+HDCzkKSFksZLqpP0npmtdc7tip3yaOxzAAAAAJ0w\nZ84cz2o9/vjjntWK16WXXur5NgIdaWxs1OzZs3XixAklJycrHA6rtLRUt99+u2cZLgZxN3DOuT+Y\nWc45w9dLqnXOfShJZvaKpClmtlvSLyT93jlX6VFWAAAAAAFxvj3bvDZ27Fht3br1K+ObN28+6zgn\nJyewe8BJiS9iMljS/jbHdbGx2ZK+J+l2M/vh111oZjPMrNzMyo8cOZJgDAAAAADo+bpkERPn3HxJ\n889zTqmkUkkqKChwXZEDAAAAAHqSRJ/AHZA0pM1xdmwsLmZWZGalDQ0NCcYAAAAA4DfneC5zPon+\njRJt4N6TdI2Z5ZrZJZKmSlob78XOuTLn3Iz09PQEYwAAAADwU1pamo4dO0YT1wHnnI4dO6a0tLQL\nrhH3FEozWympUFKGmdVJetw5t9jMZkl6Q1JI0kvOuZ2dqFkkqSgcDncuNQAAAICLSnZ2turq6sT6\nFh1LS0tTdnb2BV/fmVUop7Uzvl7SBW1n7pwrk1RWUFBw34VcDwAAAODikJKSotzcXL9j9HiJTqEE\nAAAAAHSTLlmFMl5MoUSibI55Vss9znxtAAAAXNx8fQLHIiYAAAAAED+mUAIAAABAQPjawLEPHAAA\nAADEjymUAAAAABAQvi5iAqB7mXm46AubdAIAAHQ7VqEELnKbbbPfEQAAAHCRYAolAAAAAAQEq1AC\nAAAAQEDQwAEAAABAQNDAAQAAAEBAsA8cAAAAAAQEi5gAAAAAQEAwhRIAAAAAAoIGDgAAAAACggYO\nAAAAAAIi2c+bm1mRpKJwOOxnDMBztnmzZ7U2eVYJAAAAQcciJgAAAAAQEEyhBAAAAICAoIEDAAAA\ngICggQMAAACAgKCBAwAAAICAoIEDAAAAgICggQMAAACAgPC1gTOzIjMrbWho8DMGAAAAAAQC+8AB\nAAAAQEAwhRIAAAAAAiLZ7wAAAmpLuXe1vlvgXS0AAIAejCdwAAAAABAQNHAAAAAAEBA0cAAAAAAQ\nEDRwAAAAABAQNHAAAAAAEBA0cAAAAAAQEDRwAAAAABAQnjdwZna1mS02s1e9rg0AAAAAvVlcG3mb\n2UuSbpF02Dk3qs34REnPSwpJetE59wvn3IeSptPAAQAAxG/zZvOw2iYPawG4mMT7BG6JpIltB8ws\nJGmhpEmS8iRNM7M8T9MBAAAAAFrF1cA55/4g6fg5w9dLqnXOfeicOynpFUlT4r2xmc0ws3IzKz9y\n5EjcgQGJeXhZAAAI+klEQVQAAACgt0rkHbjBkva3Oa6TNNjMrjSz30gaY2b/rb2LnXOlzrkC51xB\nZmZmAjEAAAAAoHeI6x24znDOHZP0w3jONbMiSUXhcNjrGAAAAADQ4yTyBO6ApCFtjrNjY3FzzpU5\n52akp6cnEAMAAAAAeodEGrj3JF1jZrlmdomkqZLWdqaAmRWZWWlDQ0MCMQAAAACgd4irgTOzlZK2\nSbrWzOrMbLpzrlnSLElvSNotabVzbmdnbs4TOAAAAACIX1zvwDnnprUzvl7Sek8TAQAAAAC+lueL\nmHQGi5gAkKRb7O88q/V8hmelAFxEzMM9rp13peCnLeXe1fpugXe1gC6WyDtwCWMKJQAAAADEz9cG\nDgAAAAAQP18bOFahBAAAAID4MYUSAAAAAAKCKZQAAAAAEBA0cAAAAAAQELwDBwAAAAABwTtwAAAA\nABAQTKEEAAAAgICggQMAAACAgOAdOAAAAAAICN6BAwAAAICAYAolAAAAAAQEDRwAAAAABAQNHAAA\nAAAEBA0cAAAAAAREsp83N7MiSUXhcNjPGAAAAOgGZuZZLbf5Pc9q3WJ/51mt5zM8KwV8LVahBAAA\nAICAYAolAAAAAAQEDRwAAAAABAQNHAAAAAAEBA0cAAAAAAQEDRwAAAAABAQNHAAAAAAEhK8NnJkV\nmVlpQ0ODnzEAAAAAIBDYBw4AAAAAAoIplAAAAAAQEDRwAAAAABAQNHAAAAAAEBA0cAAAAAAQEDRw\nAAAAABAQNHAAAAAAEBA0cAAAAAAQEDRwAAAAABAQyV4XNLNLJf1K0klJm51zL3t9DwAAAADojeJ6\nAmdmL5nZYTPbcc74RDOrNrNaM/tZbPg2Sa865+6TNNnjvAAAAADQa8U7hXKJpIltB8wsJGmhpEmS\n8iRNM7M8SdmS9sdOO+1NTAAAAABAXA2cc+4Pko6fM3y9pFrn3IfOuZOSXpE0RVKdzjRxHdY3sxlm\nVm5m5UeOHOl8cnQrM+9+AAAAAFyYRBYxGaw/PWmTzjRugyX9TtJfm9mvJZW1d7FzrtQ5V+CcK8jM\nzEwgBgAAAAD0Dp4vYuKc+0zSPfGca2ZFkorC4bDXMQAAAACgx0nkCdwBSUPaHGfHxuLmnCtzzs1I\nT09PIAYAAAAA9A6JNHDvSbrGzHLN7BJJUyWt9SYWAAAAAOBc8W4jsFLSNknXmlmdmU13zjVLmiXp\nDUm7Ja12zu3szM3NrMjMShsaGjqbGwAAAAB6nbjegXPOTWtnfL2k9Rd6c+dcmaSygoKC+y60BgAA\nAAD0FolMoUwYT+AAAAAAIH6+NnAsYgIAAAAA8fO1gQMAAAAAxI8plAAAAAAQEEyhBAAAAICAYAol\nAAAAAASEOef8ziAzOyJpn985gADLkHTU7xAAgF6Nf4uAxAxzzmWe76SLooEDkBgzK3fOFfidAwDQ\ne/FvEdA9mEIJAAAAAAFBAwcAAAAAAUEDB/QMpX4HAAD0evxbBHQD3oEDAAAAgIDgCRwAAAAABAQN\nHAAAAAAEBA0cAAAAAAREst8BAHSemX1D0hRJg2NDByStdc7t9i8VAAAAuhpP4ICAMbOHJb0iySS9\nG/sxSSvN7Gd+ZgMAwMzu8TsD0JOxCiUQMGa2R9I3nXOnzhm/RNJO59w1/iQDAEAys4+dc0P9zgH0\nVEyhBIKnRdJVkvadMz4o9hkAAF3KzKra+0jSwO7MAvQ2NHBA8DwoaYOZ1UjaHxsbKiksaZZvqQAA\nvclASX8p6V/PGTdJW7s/DtB70MABAeOce93MRki6XmcvYvKec+60f8kAAL3IOkn9nHPRcz8ws83d\nHwfoPXgHDgAAAAACglUoAQAAACAgaOAAAAAAICBo4AAAAAAgIGjgAABdwswaz/N5jpnt6GTNJWZ2\neweff2RmGV8z/kMzu6sz9wIA4GLEKpQAgB7POfcbL+qYWbJzrtmLWgAAXAiewAEAupSZ9TOzDWZW\naWbbzWxKm4+TzexlM9ttZq+aWd/YNWPNbIuZVZjZG2Y2qBO3/GnsPu+aWThW7wkz+0ns981m9vex\nz/eY2Y2x8RwzeyuWs9LM/iI2XhgbXytpl5k9aWYPtvl+88zsgXa+e2Hsfq+a2Qex72qxz1qfFppZ\nwZdLr8eyLo3dc5+Z3WZm/z32nV43s5RO/C0AAD0MDRwAoKt9LukHzrnrJN0k6dkvmxhJ10r6lXNu\npKRPJc2MNSgvSLrdOTdW0kuS5nXifg3OudGSFkj6ZTvnJDvnrpf0oKTHY2OHJY2P5SyWNL/N+ddJ\nesA5NyKW5y5JMrMkSVMl/baDPGNi98mTdLWkG+L4DsMl/XtJk2O1N8W+U5Okv4rjegBAD8UUSgBA\nVzNJPzez70hq0ZkN6AfGPtvvnHs79vtvJd0v6XVJoyT9n1ifF5JU34n7rWzz3+faOed3sf9WSMqJ\n/Z4iaYGZRSSdljSizfnvOuf+nyQ55z4ys2NmNib2Pd53zh3rIM+7zrk6STKzaOx+//c83+H3zrlT\nZrZdZ77/67Hx7W3yAgB6IRo4AEBXu1NSpqSxsabkI0lpsc/cOec6nWn4djrnxl3g/Vw7v7f1Rey/\np/Wnfwt/LOmQpG/pzAyVz9uc/9k5178oqURSls48kevIF21+b3u/Zv1pJkyazvaFJDnnWszslHPu\ny+/RIv7tBoBejSmUAICuli7pcKx5u0nSsDafDTWzLxu1/6gzT6aqJWV+OW5mKWb2zU7cr7jNf7d1\nMme9c65F0n/WmSdf7XlN0kRJ/07SG524R1sfSRob+/2vL7AGAKCXoYEDAHS1lyUVxKYD3iXpgzaf\nVUv6L2a2W9IASb92zp2UdLukvzezP0qKSvqLTtxvgJlVSXpAZ56qxetXku6O3fMb+upTt1axjJsk\nrXbOne7EPdqaI+l5MyvXmSdzAACcl/1pVgYAAIhHbPGSSkn/wTlX43ceAEDvwRM4AAA6wczyJNVK\n2kDzBgDobjyBAwAEjpm9Jin3nOGHnXMX+j5aonlGS1p+zvAXzrlv+5EHANBz0cABAAAAQEAwhRIA\nAAAAAoIGDgAAAAACggYOAAAAAAKCBg4AAAAAAuL/A5UF4oG7161kAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f305849d7f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#flag according to label_binary\n",
    "pd.pivot_table(data[['flag_num', 'label_binary_num']].assign(count=1),\n",
    "               index=['label_binary_num'],\n",
    "               columns=['flag_num'],\n",
    "               aggfunc='count').plot(kind='bar', color=constants.my_colors, logy=True, legend=False, figsize=(15, 5))\n",
    "\n",
    "handles = [mpatches.Patch(label=le_dicts['flag'][i],\n",
    "                          color=constants.my_colors[i]) for i in sorted(le_dicts['flag'])]\n",
    "plt.legend(handles=handles)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"11\" halign=\"left\">count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>flag</th>\n",
       "      <th>OTH</th>\n",
       "      <th>REJ</th>\n",
       "      <th>RSTO</th>\n",
       "      <th>RSTOS0</th>\n",
       "      <th>RSTR</th>\n",
       "      <th>S0</th>\n",
       "      <th>S1</th>\n",
       "      <th>S2</th>\n",
       "      <th>S3</th>\n",
       "      <th>SF</th>\n",
       "      <th>SH</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>label_binary_num</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>5341.0</td>\n",
       "      <td>67.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>31.0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>54.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>91709.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7.0</td>\n",
       "      <td>21534.0</td>\n",
       "      <td>512.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>872.0</td>\n",
       "      <td>86956.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>286731.0</td>\n",
       "      <td>107.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 count                                                     \\\n",
       "flag               OTH      REJ   RSTO RSTOS0   RSTR       S0    S1    S2   \n",
       "label_binary_num                                                            \n",
       "0                  1.0   5341.0   67.0    NaN   31.0     51.0  54.0  17.0   \n",
       "1                  7.0  21534.0  512.0   11.0  872.0  86956.0   3.0   7.0   \n",
       "\n",
       "                                        \n",
       "flag               S3        SF     SH  \n",
       "label_binary_num                        \n",
       "0                 7.0   91709.0    NaN  \n",
       "1                 3.0  286731.0  107.0  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.pivot_table(data[['flag', 'label_binary_num']].assign(count=1),\n",
    "               index=['label_binary_num'],\n",
    "               columns=['flag'],\n",
    "               aggfunc='count')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Correlation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "logged_in            -0.795282\n",
       "protocol_type_num    -0.616601\n",
       "flag_num             -0.155672\n",
       "service_num          -0.131723\n",
       "duration             -0.118014\n",
       "num_access_files     -0.054268\n",
       "dst_bytes            -0.037709\n",
       "is_guest_login       -0.032299\n",
       "num_file_creations   -0.018671\n",
       "num_shells           -0.014951\n",
       "num_root             -0.011006\n",
       "su_attempted         -0.008789\n",
       "hot                  -0.006327\n",
       "root_shell           -0.005871\n",
       "num_compromised      -0.005046\n",
       "urgent               -0.001498\n",
       "num_failed_logins    -0.001060\n",
       "src_bytes             0.000936\n",
       "land                  0.002542\n",
       "wrong_fragment        0.023630\n",
       "num_outbound_cmds          NaN\n",
       "is_host_login              NaN\n",
       "dtype: float64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Corr with binary label\n",
    "data.drop(constants.categorical_names + constants.label_names, axis=1).corrwith(data.label_binary_num).sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "logged_in            -0.789552\n",
       "protocol_type_num    -0.622297\n",
       "service_num          -0.141445\n",
       "flag_num             -0.135043\n",
       "duration             -0.128845\n",
       "num_access_files     -0.054097\n",
       "is_guest_login       -0.052703\n",
       "dst_bytes            -0.043191\n",
       "hot                  -0.032488\n",
       "num_file_creations   -0.019785\n",
       "num_shells           -0.016380\n",
       "num_root             -0.010988\n",
       "root_shell           -0.010185\n",
       "num_failed_logins    -0.010071\n",
       "su_attempted         -0.008964\n",
       "num_compromised      -0.005057\n",
       "urgent               -0.003045\n",
       "src_bytes            -0.002457\n",
       "land                  0.002647\n",
       "wrong_fragment        0.024343\n",
       "num_outbound_cmds          NaN\n",
       "is_host_login              NaN\n",
       "dtype: float64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Corr with 5 labels \n",
    "data.drop(constants.categorical_names + constants.label_names, axis=1).corrwith(data.label_four_num).sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f305851ef60>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAawAAAGpCAYAAADRBQIfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFgAt/H9CvVun7RYoygAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f30584a0c18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# corr heatmap\n",
    "# last 2 are label_binary_num and label_four_num that's why it so hot\n",
    "plt.figure(figsize=(7,7))\n",
    "plt.matshow(data.drop(constants.categorical_names + \\\n",
    "                      ['label', 'label_four'] + \\\n",
    "                      constants.names_without_changes, axis=1).corr(), fignum=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 duration\n",
      "1 src_bytes\n",
      "2 dst_bytes\n",
      "3 land\n",
      "4 wrong_fragment\n",
      "5 urgent\n",
      "6 hot\n",
      "7 num_failed_logins\n",
      "8 logged_in\n",
      "9 num_compromised\n",
      "10 root_shell\n",
      "11 su_attempted\n",
      "12 num_root\n",
      "13 num_file_creations\n",
      "14 num_shells\n",
      "15 num_access_files\n",
      "16 is_guest_login\n",
      "17 protocol_type_num\n",
      "18 service_num\n",
      "19 flag_num\n",
      "20 label_binary_num\n",
      "21 label_four_num\n"
     ]
    }
   ],
   "source": [
    "for i, elem in enumerate(data.drop(constants.categorical_names + \\\n",
    "                                   ['label', 'label_four'] + \\\n",
    "                                   constants.names_without_changes, axis=1).columns.tolist()):\n",
    "    print(i, elem)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Feature importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "\n",
    "forest = ExtraTreesClassifier(n_estimators=500,\n",
    "                              random_state=42)\n",
    "\n",
    "data_test = data.drop(constants.categorical_names +\\\n",
    "                      constants.label_names +\\\n",
    "                      constants.names_without_changes, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA20AAAD8CAYAAADkIEyxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAMDAlDYAAICBKW0AAAADU9oAAAAG9v8B3gg2iEpULX0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f30547aef60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#for 2 labels\n",
    "forest.fit(data_test, data['label_binary_num'])\n",
    "\n",
    "importances = forest.feature_importances_\n",
    "indices = np.argsort(importances)[::-1]\n",
    "\n",
    "plt.figure(figsize=(15, 4))\n",
    "plt.bar(range(data_test.shape[1]), importances[indices], color=\"r\", align=\"center\")\n",
    "plt.xticks(range(data_test.shape[1]), indices)\n",
    "plt.xlim([-1, data_test.shape[1]])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA20AAAD8CAYAAADkIEyxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f30546f8f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#for 5 labels\n",
    "forest.fit(data_test, data['label_four_num'])\n",
    "\n",
    "importances = forest.feature_importances_\n",
    "indices = np.argsort(importances)[::-1]\n",
    "\n",
    "plt.figure(figsize=(15, 4))\n",
    "plt.bar(range(data_test.shape[1]), importances[indices], color=\"r\", align=\"center\")\n",
    "plt.xticks(range(data_test.shape[1]), indices)\n",
    "plt.xlim([-1, data_test.shape[1]])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 duration\n",
      "1 src_bytes\n",
      "2 dst_bytes\n",
      "3 land\n",
      "4 wrong_fragment\n",
      "5 urgent\n",
      "6 hot\n",
      "7 num_failed_logins\n",
      "8 logged_in\n",
      "9 num_compromised\n",
      "10 root_shell\n",
      "11 su_attempted\n",
      "12 num_root\n",
      "13 num_file_creations\n",
      "14 num_shells\n",
      "15 num_access_files\n",
      "16 is_guest_login\n",
      "17 protocol_type_num\n",
      "18 service_num\n",
      "19 flag_num\n"
     ]
    }
   ],
   "source": [
    "for i, elem in enumerate(data_test.columns.tolist()):\n",
    "    print(i, elem)"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}