1_ip2vec_preprocess.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "from random import randint\n",
    "import pandas as pd\n",
    "import pickle\n",
    "import collections\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\akarsh\\appdata\\local\\programs\\python\\python37\\lib\\site-packages\\IPython\\core\\interactiveshell.py:3051: DtypeWarning: Columns (8) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  interactivity=interactivity, compiler=compiler, result=result)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(260400, 16)\n",
      "['UDP  ' 'TCP  ' 'ICMP ' 'IGMP '] \n",
      "\n",
      "['normal' 'victim' 'attacker'] \n",
      "\n",
      "['......' '.AP...' '.A....' '.A..S.' '.A...F' '.A..SF' '...R..' '.AP.S.'\n",
      " '.AP.SF' '.AP..F' '....S.' '.A.R..' '.APRSF' '.A.R.F' '.APR..' '.APR.F'\n",
      " '.APRS.' '.A.RS.' '...RS.' '.A.RSF'] 20 \n",
      "\n",
      "(260391, 16)\n",
      "['UDP  ' 'TCP  ' 'ICMP ']\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date first seen</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Proto</th>\n",
       "      <th>Src IP Addr</th>\n",
       "      <th>Src Pt</th>\n",
       "      <th>Dst IP Addr</th>\n",
       "      <th>Dst Pt</th>\n",
       "      <th>Packets</th>\n",
       "      <th>Bytes</th>\n",
       "      <th>Flows</th>\n",
       "      <th>Flags</th>\n",
       "      <th>Tos</th>\n",
       "      <th>class</th>\n",
       "      <th>attackType</th>\n",
       "      <th>attackID</th>\n",
       "      <th>attackDescription</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-03-20 11:03:29.789</td>\n",
       "      <td>0.00</td>\n",
       "      <td>UDP</td>\n",
       "      <td>192.168.220.10</td>\n",
       "      <td>42968</td>\n",
       "      <td>DNS</td>\n",
       "      <td>53.0</td>\n",
       "      <td>2</td>\n",
       "      <td>176</td>\n",
       "      <td>1</td>\n",
       "      <td>......</td>\n",
       "      <td>0</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-03-20 16:54:11.971</td>\n",
       "      <td>0.64</td>\n",
       "      <td>TCP</td>\n",
       "      <td>10086_192</td>\n",
       "      <td>80</td>\n",
       "      <td>192.168.220.14</td>\n",
       "      <td>43392.0</td>\n",
       "      <td>6</td>\n",
       "      <td>3230</td>\n",
       "      <td>1</td>\n",
       "      <td>.AP...</td>\n",
       "      <td>32</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-04-14 18:19:42.203</td>\n",
       "      <td>0.00</td>\n",
       "      <td>TCP</td>\n",
       "      <td>16569_82</td>\n",
       "      <td>80</td>\n",
       "      <td>192.168.220.8</td>\n",
       "      <td>41440.0</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>1</td>\n",
       "      <td>.A....</td>\n",
       "      <td>32</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-03-18 11:08:00.553</td>\n",
       "      <td>0.00</td>\n",
       "      <td>TCP</td>\n",
       "      <td>10296_238</td>\n",
       "      <td>443</td>\n",
       "      <td>192.168.200.9</td>\n",
       "      <td>59692.0</td>\n",
       "      <td>2</td>\n",
       "      <td>154</td>\n",
       "      <td>1</td>\n",
       "      <td>.AP...</td>\n",
       "      <td>32</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-03-18 17:51:09.047</td>\n",
       "      <td>0.00</td>\n",
       "      <td>TCP</td>\n",
       "      <td>192.168.220.14</td>\n",
       "      <td>52236</td>\n",
       "      <td>10056_243</td>\n",
       "      <td>443.0</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>1</td>\n",
       "      <td>.A....</td>\n",
       "      <td>0</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Date first seen  Duration  Proto     Src IP Addr  Src Pt  \\\n",
       "0  2017-03-20 11:03:29.789      0.00  UDP    192.168.220.10   42968   \n",
       "1  2017-03-20 16:54:11.971      0.64  TCP         10086_192      80   \n",
       "2  2017-04-14 18:19:42.203      0.00  TCP          16569_82      80   \n",
       "3  2017-03-18 11:08:00.553      0.00  TCP         10296_238     443   \n",
       "4  2017-03-18 17:51:09.047      0.00  TCP    192.168.220.14   52236   \n",
       "\n",
       "      Dst IP Addr   Dst Pt  Packets     Bytes  Flows   Flags  Tos   class  \\\n",
       "0             DNS     53.0        2       176      1  ......    0  normal   \n",
       "1  192.168.220.14  43392.0        6      3230      1  .AP...   32  normal   \n",
       "2   192.168.220.8  41440.0        1        66      1  .A....   32  normal   \n",
       "3   192.168.200.9  59692.0        2       154      1  .AP...   32  normal   \n",
       "4       10056_243    443.0        1        66      1  .A....    0  normal   \n",
       "\n",
       "  attackType attackID attackDescription  \n",
       "0        ---      ---               ---  \n",
       "1        ---      ---               ---  \n",
       "2        ---      ---               ---  \n",
       "3        ---      ---               ---  \n",
       "4        ---      ---               ---  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nsamples = 65100\n",
    "df_week1 = pd.read_csv(\"C:/Users/Akarsh/Downloads/WISENT-CIDDS-001/CIDDS-001/traffic/OpenStack/CIDDS-001-internal-week1.csv\").sample(n=nsamples)\n",
    "df_week2 = pd.read_csv(\"C:/Users/Akarsh/Downloads/WISENT-CIDDS-001/CIDDS-001/traffic/OpenStack/CIDDS-001-internal-week2.csv\").sample(n=nsamples)\n",
    "df_week3 = pd.read_csv(\"C:/Users/Akarsh/Downloads/WISENT-CIDDS-001/CIDDS-001/traffic/OpenStack/CIDDS-001-internal-week3.csv\").sample(n=nsamples)\n",
    "df_week4 = pd.read_csv(\"C:/Users/Akarsh/Downloads/WISENT-CIDDS-001/CIDDS-001/traffic/OpenStack/CIDDS-001-internal-week4.csv\").sample(n=nsamples)\n",
    "\n",
    "frames = [df_week1, df_week2, df_week3, df_week4]\n",
    "train_data = pd.concat(frames)\n",
    "print(train_data.shape)\n",
    "\n",
    "del df_week1, df_week2, df_week3, df_week4\n",
    "\n",
    "train_data = train_data.sample(frac=1).reset_index(drop=True)\n",
    "\n",
    "print(train_data['Proto'].unique(), '\\n')\n",
    "print(train_data['class'].unique(), '\\n')\n",
    "print(train_data['Flags'].unique(), len(train_data['Flags'].unique()), '\\n')\n",
    "\n",
    "indexNames = train_data[train_data['Proto'].str.match('IGMP', na=False)].index\n",
    "train_data.drop(indexNames, inplace=True)\n",
    "print(train_data.shape)\n",
    "print(train_data['Proto'].unique())\n",
    "\n",
    "display(train_data.head())\n",
    "del indexNames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cleaning data...\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Day</th>\n",
       "      <th>Time</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Proto</th>\n",
       "      <th>Src IP Addr</th>\n",
       "      <th>Src Pt</th>\n",
       "      <th>Dst IP Addr</th>\n",
       "      <th>Dst Pt</th>\n",
       "      <th>Packets</th>\n",
       "      <th>Bytes</th>\n",
       "      <th>Flags</th>\n",
       "      <th>class</th>\n",
       "      <th>attackType</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>11:03:29.789</td>\n",
       "      <td>0.00</td>\n",
       "      <td>UDP</td>\n",
       "      <td>192.168.220.10</td>\n",
       "      <td>42968</td>\n",
       "      <td>DNS</td>\n",
       "      <td>53.0</td>\n",
       "      <td>2</td>\n",
       "      <td>176</td>\n",
       "      <td>......</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>16:54:11.971</td>\n",
       "      <td>0.64</td>\n",
       "      <td>TCP</td>\n",
       "      <td>10086_192</td>\n",
       "      <td>80</td>\n",
       "      <td>192.168.220.14</td>\n",
       "      <td>43392.0</td>\n",
       "      <td>6</td>\n",
       "      <td>3230</td>\n",
       "      <td>.AP...</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>18:19:42.203</td>\n",
       "      <td>0.00</td>\n",
       "      <td>TCP</td>\n",
       "      <td>16569_82</td>\n",
       "      <td>80</td>\n",
       "      <td>192.168.220.8</td>\n",
       "      <td>41440.0</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>.A....</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>11:08:00.553</td>\n",
       "      <td>0.00</td>\n",
       "      <td>TCP</td>\n",
       "      <td>10296_238</td>\n",
       "      <td>443</td>\n",
       "      <td>192.168.200.9</td>\n",
       "      <td>59692.0</td>\n",
       "      <td>2</td>\n",
       "      <td>154</td>\n",
       "      <td>.AP...</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>17:51:09.047</td>\n",
       "      <td>0.00</td>\n",
       "      <td>TCP</td>\n",
       "      <td>192.168.220.14</td>\n",
       "      <td>52236</td>\n",
       "      <td>10056_243</td>\n",
       "      <td>443.0</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>.A....</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Day          Time  Duration Proto     Src IP Addr  Src Pt     Dst IP Addr  \\\n",
       "0    0  11:03:29.789      0.00   UDP  192.168.220.10   42968             DNS   \n",
       "1    0  16:54:11.971      0.64   TCP       10086_192      80  192.168.220.14   \n",
       "2    4  18:19:42.203      0.00   TCP        16569_82      80   192.168.220.8   \n",
       "3    5  11:08:00.553      0.00   TCP       10296_238     443   192.168.200.9   \n",
       "4    5  17:51:09.047      0.00   TCP  192.168.220.14   52236       10056_243   \n",
       "\n",
       "    Dst Pt  Packets  Bytes   Flags   class attackType  \n",
       "0     53.0        2    176  ......  normal        ---  \n",
       "1  43392.0        6   3230  .AP...  normal        ---  \n",
       "2  41440.0        1     66  .A....  normal        ---  \n",
       "3  59692.0        2    154  .AP...  normal        ---  \n",
       "4    443.0        1     66  .A....  normal        ---  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Day</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Src Pt</th>\n",
       "      <th>Dst Pt</th>\n",
       "      <th>Packets</th>\n",
       "      <th>Bytes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>260391.000000</td>\n",
       "      <td>260391.000000</td>\n",
       "      <td>260391.000000</td>\n",
       "      <td>260391.000000</td>\n",
       "      <td>260391.000000</td>\n",
       "      <td>2.603910e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2.366288</td>\n",
       "      <td>0.126139</td>\n",
       "      <td>24759.687209</td>\n",
       "      <td>24402.750514</td>\n",
       "      <td>16.475462</td>\n",
       "      <td>2.094636e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.613342</td>\n",
       "      <td>0.710818</td>\n",
       "      <td>25148.739218</td>\n",
       "      <td>25130.514122</td>\n",
       "      <td>994.716635</td>\n",
       "      <td>2.179878e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.200000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>6.600000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8082.000000</td>\n",
       "      <td>8082.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.400000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>0.032000</td>\n",
       "      <td>50232.500000</td>\n",
       "      <td>50076.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.440000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>111.785000</td>\n",
       "      <td>65534.000000</td>\n",
       "      <td>65535.000000</td>\n",
       "      <td>165278.000000</td>\n",
       "      <td>4.868538e+08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Day       Duration         Src Pt         Dst Pt  \\\n",
       "count  260391.000000  260391.000000  260391.000000  260391.000000   \n",
       "mean        2.366288       0.126139   24759.687209   24402.750514   \n",
       "std         1.613342       0.710818   25148.739218   25130.514122   \n",
       "min         0.000000       0.000000       0.000000       0.000000   \n",
       "25%         1.000000       0.000000      80.000000      80.000000   \n",
       "50%         2.000000       0.000000    8082.000000    8082.000000   \n",
       "75%         4.000000       0.032000   50232.500000   50076.000000   \n",
       "max         6.000000     111.785000   65534.000000   65535.000000   \n",
       "\n",
       "             Packets         Bytes  \n",
       "count  260391.000000  2.603910e+05  \n",
       "mean       16.475462  2.094636e+04  \n",
       "std       994.716635  2.179878e+06  \n",
       "min         1.000000  4.200000e+01  \n",
       "25%         1.000000  6.600000e+01  \n",
       "50%         2.000000  1.400000e+02  \n",
       "75%         3.000000  4.440000e+02  \n",
       "max    165278.000000  4.868538e+08  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 260391 entries, 0 to 260390\n",
      "Data columns (total 13 columns):\n",
      "Day            260391 non-null int64\n",
      "Time           260391 non-null object\n",
      "Duration       260391 non-null float64\n",
      "Proto          260391 non-null object\n",
      "Src IP Addr    260391 non-null object\n",
      "Src Pt         260391 non-null int64\n",
      "Dst IP Addr    260391 non-null object\n",
      "Dst Pt         260391 non-null float64\n",
      "Packets        260391 non-null int64\n",
      "Bytes          260391 non-null int64\n",
      "Flags          260391 non-null object\n",
      "class          260391 non-null object\n",
      "attackType     260391 non-null object\n",
      "dtypes: float64(2), int64(4), object(7)\n",
      "memory usage: 25.8+ MB\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "None"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 260391 entries, 0 to 260390\n",
      "Data columns (total 13 columns):\n",
      "Day            260391 non-null int64\n",
      "Time           260391 non-null object\n",
      "Duration       260391 non-null float64\n",
      "Proto          260391 non-null object\n",
      "Src IP Addr    260391 non-null object\n",
      "Src Pt         260391 non-null int64\n",
      "Dst IP Addr    260391 non-null object\n",
      "Dst Pt         260391 non-null int64\n",
      "Packets        260391 non-null int64\n",
      "Bytes          260391 non-null int64\n",
      "Flags          260391 non-null object\n",
      "class          260391 non-null object\n",
      "attackType     260391 non-null object\n",
      "dtypes: float64(1), int64(5), object(7)\n",
      "memory usage: 25.8+ MB\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "None"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Day</th>\n",
       "      <th>Time</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Proto</th>\n",
       "      <th>Src IP Addr</th>\n",
       "      <th>Src Pt</th>\n",
       "      <th>Dst IP Addr</th>\n",
       "      <th>Dst Pt</th>\n",
       "      <th>Packets</th>\n",
       "      <th>Bytes</th>\n",
       "      <th>Flags</th>\n",
       "      <th>class</th>\n",
       "      <th>attackType</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>11:03:29.789</td>\n",
       "      <td>0.00</td>\n",
       "      <td>UDP</td>\n",
       "      <td>192.168.220.10</td>\n",
       "      <td>42968</td>\n",
       "      <td>DNS</td>\n",
       "      <td>53</td>\n",
       "      <td>2</td>\n",
       "      <td>176</td>\n",
       "      <td>......</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>16:54:11.971</td>\n",
       "      <td>0.64</td>\n",
       "      <td>TCP</td>\n",
       "      <td>10086_192</td>\n",
       "      <td>80</td>\n",
       "      <td>192.168.220.14</td>\n",
       "      <td>43392</td>\n",
       "      <td>6</td>\n",
       "      <td>3230</td>\n",
       "      <td>.AP...</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>18:19:42.203</td>\n",
       "      <td>0.00</td>\n",
       "      <td>TCP</td>\n",
       "      <td>16569_82</td>\n",
       "      <td>80</td>\n",
       "      <td>192.168.220.8</td>\n",
       "      <td>41440</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>.A....</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>11:08:00.553</td>\n",
       "      <td>0.00</td>\n",
       "      <td>TCP</td>\n",
       "      <td>10296_238</td>\n",
       "      <td>443</td>\n",
       "      <td>192.168.200.9</td>\n",
       "      <td>59692</td>\n",
       "      <td>2</td>\n",
       "      <td>154</td>\n",
       "      <td>.AP...</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>17:51:09.047</td>\n",
       "      <td>0.00</td>\n",
       "      <td>TCP</td>\n",
       "      <td>192.168.220.14</td>\n",
       "      <td>52236</td>\n",
       "      <td>10056_243</td>\n",
       "      <td>443</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>.A....</td>\n",
       "      <td>normal</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Day          Time  Duration Proto     Src IP Addr  Src Pt     Dst IP Addr  \\\n",
       "0    0  11:03:29.789      0.00   UDP  192.168.220.10   42968             DNS   \n",
       "1    0  16:54:11.971      0.64   TCP       10086_192      80  192.168.220.14   \n",
       "2    4  18:19:42.203      0.00   TCP        16569_82      80   192.168.220.8   \n",
       "3    5  11:08:00.553      0.00   TCP       10296_238     443   192.168.200.9   \n",
       "4    5  17:51:09.047      0.00   TCP  192.168.220.14   52236       10056_243   \n",
       "\n",
       "   Dst Pt  Packets  Bytes   Flags   class attackType  \n",
       "0      53        2    176  ......  normal        ---  \n",
       "1   43392        6   3230  .AP...  normal        ---  \n",
       "2   41440        1     66  .A....  normal        ---  \n",
       "3   59692        2    154  .AP...  normal        ---  \n",
       "4     443        1     66  .A....  normal        ---  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# clean data\n",
    "def clean_data(train_data):\n",
    "    print(\"cleaning data...\")\n",
    "    process_data = []\n",
    "    for row in train_data.values:\n",
    "        date  = row[0].strip()\n",
    "        datetime_object = datetime.strptime(date, \"%Y-%m-%d %H:%M:%S.%f\")\n",
    "        week_day = datetime_object.weekday()\n",
    "        time  = row[0].strip().split(\" \")[1]\n",
    "        res = []\n",
    "        res.append(week_day)\n",
    "        res.append(time)\n",
    "        for j in range(1,14):\n",
    "            if j == 9 or j == 11:\n",
    "                continue\n",
    "            if j == 8:\n",
    "                if type(row[j]) == int:\n",
    "                    res.append(int(row[j]))\n",
    "                else:\n",
    "                    bytzes = row[j].strip()\n",
    "                    if \"M\" in bytzes:\n",
    "                        bytzes = float( bytzes.split(\" \")[0] )\n",
    "                        bytzes = int( bytzes * 1024 * 1024 )# Conv to Bytes\n",
    "                    res.append(int(bytzes))\n",
    "            else:\n",
    "                if type(row[j]) == str:\n",
    "                    res.append(row[j].strip())\n",
    "                else:\n",
    "                    res.append(row[j])\n",
    "        process_data.append(res)\n",
    "    return process_data\n",
    "\n",
    "col_names = ['Day', 'Time', 'Duration', 'Proto', 'Src IP Addr', 'Src Pt', 'Dst IP Addr', 'Dst Pt', 'Packets', 'Bytes', 'Flags', 'class', 'attackType']\n",
    "process_data = clean_data(train_data)\n",
    "updated_train = pd.DataFrame(process_data, columns = col_names)\n",
    "\n",
    "display(updated_train.head())\n",
    "display(updated_train.describe())\n",
    "display(updated_train.info())\n",
    "\n",
    "updated_train['Dst Pt'] = updated_train['Dst Pt'].astype('int64')\n",
    "display(updated_train.info())\n",
    "display(updated_train.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_data.hist(bins=15, color='steelblue', edgecolor='black', linewidth=1.0,\n",
    "           xlabelsize=8, ylabelsize=8, grid=False)\n",
    "\n",
    "plt.tight_layout(rect=(0, 0, 1.2, 1.2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnEAAADlCAYAAAA85aFEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAZrklEQVR4nO3debRlZXnn8e/PKhDLCIJVtKCUFSUKNB1BwIgpJqMYWNoC0UQk9hIHjKnENjEarNhG09IEh8QEVqIMtpFoJ4JDDGKCKFMhiAWiZUAMSZhJABnKQqCAevqP/dbicLnDqVjnnrsv389ae92932HvZx/Pwqfed79np6qQJElSvzxh3AFIkiRp05nESZIk9ZBJnCRJUg+ZxEmSJPWQSZwkSVIPmcRJkiT1kEmcpMeVJOcnOWnccUjST8skTtKUknwqSbXtwST/muQjSZ68Gc5dSV69OeLcnJIcOHDPg9vHxh2bJA1aOO4AJM155wKvB7YA9gNOBZ4MvG2yxkm2qKoHZy+8kfmvwJ0Dx/dO1ijJE4BU1cOzEpUkNY7ESZrJA1X171V1Y1V9FvgMcBg8atTq0CSXJVkPvLzVvTXJtUnWt79v2XjCJNe13TNa/+sG6qbs1+q3TvKXSW5Ncn+Sq5P82kD9EUnWJHkgyY1J/iBJ/hP3fVu7743bj9v535zk7iSvTPJPwHrg5wbqrm5xXZPk7YPXTvLcJBe2+h8kOaTt/3qr37l9HnsM9FnYyg4bKNspyeeS3JXkziRnJXnOQP0Hk1yZ5Kg2ero2yReSbDfhszw6yfdbDP+e5LRW/ukkX5rQdkGSm5O8/T/xWUoaAUfiJG2q++hG5QadALwTuBb4cZLDgZOA3wHOoUvs/iLJv1fV3wP7ALcBbwHOAh4GmKlfS4i+CmwLHA38EHgesFXrvxdwBvBBumRzH+ATwFrgxM34GSwCjm3x/wi4OcnbgP8F/DZwBfDzwCl0Sd7H24jdl4D/AF4E/AzwMR77WU4ryc8A57ftgHb+3wfOTbJbVd3Xmj4HOAJ4FbA18DfA/wZWtPOsAD4CvAc4u7U5sPU9BfhGkv9SVf/Ryn4ZeBrw15sSr6QRqio3Nze3STfgU8BZA8cvBO4A/rYdHwgU8CsT+l0MfHKSc60aOC7g1ZvSD3gZsAHYdYp4PwN8Y0LZ+4GbBo7PB06a5p433tO6CdvSVv/mVv/8Cf1uBo6cUPZ7wPfa/qHAQ8AzJ7nWr7fjndvxHgNtFrayw9rxMcAP6KZwB9vcDRzRjj9IN/37lIE2fwj8oO0HuBX44DSfw9XA7w0cfx74m3F/J93c3B7ZnE6VNJNfTrIuyf3AJcCFdKNNg1ZPON6VLiEbtArYbYZrzdRvT+DWqrp6E/s/I8nWM1x7ooOAPQa2Wwbq1gPf23iQZAdgR+C09lmtS7KOLpnaOM25K3BDVd00cJ5L6RK0TbEXXbL344Hr3E03kvacgXbXVZsCbm4Btm/7OwBPB74+zXVOBd7Y7m8x8ErgtE2MVdIIOZ0qaSYX0o3+PAjcUpMvWpjsof/JkpNhEpbp+s30bFumucamJkv/VlV3TFF3X1UNnm/jP4jfAnxriutOFntNKN8wSduJ061PAC4HjprkfD8a2J/4v1MNxDnMM4J/BfyfJC8C9qVLAqdL+iTNMkfiJM3kJ1V1bVVdP0UCN5mrgeUTypYDVw0cPwgs2MR+VwA7JNl1iuteNUX/myaMSm1ut9A96/bs9lkNbv8yENvSJDsO9HvRhPPc1v7uMFC2x4Q2V9AtpLhtkmvdtYnx/tJUDVoC+3d0o3FHA5+qqg1TtZc0+xyJkzQKH6ZbeXo53QKFX6YbOTpioM11wC8luYBuBexdQ/T7Ot1I1+eT/A7dwoadgSdX1ZeAjwLfTvJ+4LN0CxveCawc3a1CVVW75p8kWQv8A90I2l7A06vqBOAf6RZ+nJ7kd+kWNnyER0bfqKp1SVYDx7YVu9sBx0+43Ontnv4uyfuAm4CldCuGT6yqfx0y3uOADyW5nW5hw88AB1XVnww0PYVu4ckWdNOpkuYQR+IkbXYtofptulWmVwH/E/jN6lambvROuufObgS+M0y/NhJ0CN1zb39NN3L3Z8CWrf4K4DXArwDfB/64bSN/Q0NVfZxu2vkNwHfppqHfDPxbq3+YLtHaErgM+L90iy4mjm6+ge4f2KuBvwDeO+E66+h+r+8GusUGP2jnegrds3HDxnsi3ef7G8A/0a36nTjCeS7dAohzq+r6Yc8taXbk0Y91SJJmU1sw8uaqmnM/3ZFkEd3U61ur6m/HHY+kR3M6VZL0KO037Z5ON1q6lm7ET9IcYxInSZro2cA/0011H11VD405HkmTcDpVkiSph1zYIEmS1EMmcZIkST1kEidJktRDJnGSJEk9ZBInSZLUQyZxkiRJPWQSJ0mS1EMmcZIkST1kEidJktRDJnGSJEk9ZBInSZLUQyZxkiRJPWQSJ0mS1EMLxx3AbFu8eHEtW7Zs3GFIkiTN6PLLL7+jqpZMVve4S+KWLVvG6tWrxx2GJEnSjJJcP1Wd06mSJEk9ZBInSZLUQyZxkiRJPWQSJ0mS1EMmcZIkST1kEidJktRDj7ufGJktOz5zKbfefOO4w1CP7fCMnbjlphvGHYYkaY4yiRuRW2++kYP/6Kxxh6EeO+d9rxh3CJKkOczpVEmSpB4yiZMkSeohkzhJkqQeMomTJEnqIZM4SZKkHjKJkyRJ6iGTOEmSpB4yiZMkSeohkzhJkqQeMomTJEnqoZElcUn2T3JJkguT/L8kWyTZN8k327Zva7coyZlJViU5bqD/itbunCTbt7LntvNdkuTwVrYgySmt/ylJFozqniRJkuaKUY7E/QtwUFXtD1wH/ArwYeCwtn2otXsTcEFVLQd2SbJ7kqcBRwLLgROBd7e2xwMrgJcAK5MsBA4F1rb+a4FDRnhPkiRJc8LIkriqurmq7m+H64EndsV1W1XdBmxIshVdonZ2a3dWO34hcF5VbQC+Cuzb6pdV1Zqqug+4Cth5iv6SJEnz2sifiUvys3SjY+cD9wxU3Q1sB2w7UP6Ysqp6CNhy4+mG7C9JkjSvjTSJS7ItcDrweuA2YJuB6m2AO4G7BsofU9aecVvf6jcM2X9iHMckWZ1k9e233/7T35gkSdKYjXJhwxOBzwF/UFXXtClQkixOshhY0KZbLwYObt0OaceXAQckSau7tNVfn2S3Ng27G3DtFP0fpapOrqq9q2rvJUuWjOJ2JUmSZtXCEZ77jcAewAe6XIyTgWOBL7f6d7W/pwGnJ3kdsKqq1gAkOQNYBdxLN5IHsBI4tcV9QlU9lOQrwKuSXAT8kEeej5MkSZq3RpbEVdVfAn85SdWLJ7S7Fzhikv4n0q1MHSy7BthvQtnDdCtcJUmSHjf8sV9JkqQeMomTJEnqIZM4SZKkHjKJkyRJ6iGTOEmSpB4yiZMkSeohkzhJkqQeMomTJEnqIZM4SZKkHjKJkyRJ6iGTOEmSpB4yiZMkSeohkzhJkqQeMomTJEnqIZM4SZKkHjKJkyRJ6iGTOEmSpB4yiZMkSeohkzhJkqQeMomTJEnqIZM4SZKkHjKJkyRJ6iGTOEmSpB4yiZMkSeohkzhJkqQeMomTJEnqoaGSuCS7b+qJk2yX5LIk65Ls0crOT7Kq/f2tVrYoyZmt/LiB/iuSfDPJOUm2b2XPTXJhkkuSHN7KFiQ5pfU/JcmCTY1VkiSpb4Ydift4S8h+M8lTh+zzY+BQ4MwJ5a+oqgOr6qR2/CbggqpaDuySZPckTwOOBJYDJwLvbm2PB1YALwFWJlnYrrG29V8LHDJkfJIkSb01VBLXEqSjgJ2A1Uk+m+RlM/R5sKrumFgMfDnJV5I8r5UtB85u+2e14xcC51XVBuCrwL6tfllVramq+4CrgJ2n6C9JkjSvLRy2YVX9c5L3AquBPwf2TBJgZVV9YcjTvLqqfpRkL+DjwEHAtsA9rf5uYFe6EbV72nUfSrJlq8/Aue4Gtpuk/3YTL5rkGOAYgKVLlw4ZqiRJ0tw17DNxP5/kT4Gr6aYyX1lVu7b9Px32YlX1o/b3cuAprfguYJu2vw1w52BZe8ZtfavfMHC6x7QdKJt43ZOrau+q2nvJkiXDhitJkjRnDftM3EnAFcDzq2pFVV0BUFW3AO8d9mJJtm5/nwU81IovBg5u+4e048uAA9pI38HApa3++iS7JdkK2A24dor+kiRJ89qw06mHAvdV1cMASZ4AbFVVP6mq06fqlORcumRrlySfBN6a5Cd0yeM7WrPTgNOTvA5YVVVrWt8zgFXAvcDrW9uVwKkt7hPaVOtXgFcluQj4IY88HydJkjRvDZvEnQu8FFjXjhcB5wAvnq5TVb10QtHJk7S5FzhikvIT6VamDpZdA+w3oexhuhWukiRJjxvDTqduVVUbEzja/qLRhCRJkqSZDJvE3ZvkBRsP2urS+0YTkiRJkmYy7HTqO4AzktzSjncAfm00IUmSJGkmQyVxVfXtJLsAz6P7rbYfVNWDI41MkiRJUxr6x36BfYBlrc+eSaiqT48kKkmSJE1rqCQuyenAc4ArgYdbcQEmcZIkSWMw7Ejc3sBuVVWjDEaSJEnDGXZ16veBp48yEEmSJA1v2JG4xcBVSS4DHthYWFX/fSRRSZIkaVrDJnHvH2UQkiRJ2jTD/sTIBe2l9T9XVecmWQQsGG1okiRJmspQz8QleQtwJvCJVvQM4EujCkqSJEnTG3ZhwwrgF4G1AFX1z8D2owpKkiRJ0xs2iXugqtZvPEiykO534iRJkjQGwyZxFyRZCTwpycuAM4C/H11YkiRJms6wSdyxwO3AGuCtwNnAe0cVlCRJkqY37OrUDcApbZMkSdKYDfvu1H9jkmfgqurZmz0iSZIkzWhT3p260VbAa4DtNn84kiRJGsZQz8RV1Y8Gtpur6mPAS0YcmyRJkqYw7HTqCwYOn0A3MveUkUQkSZKkGQ07nfrRgf2HgOuAX93s0UiSJGkow65OPWjUgUiSJGl4w06n/u509VX1J5snHEmSJA1jU1an7gN8uR2/ErgQuHEUQUmSJGl6w76xYTHwgqp6Z1W9E9gLeGZVfaCqPjBZhyTbJbksyboke7SyfZN8s237trJFSc5MsirJcQP9V7R25yTZvpU9N8mFSS5JcngrW5DklNb/lCQL/vMfhyRJUj8Mm8QtBdYPHK8Hls3Q58fAocCZA2UfBg5r24da2ZuAC6pqObBLkt2TPA04ElgOnAi8u7U9HlhB9/MmK5MsbNdY2/qvBQ4Z8p4kSZJ6a9jp1NOBy5J8ke7NDYcDn56uQ1U9CNyRBIAkT+qK67Z2vCHJVnSJ2srW7ax2fD1wXlVtSPJVune3AiyrqjWt/1XAzq392QP9X97+SpIkzVvDrk49riVT+7Wio6vqO5t4rW2BewaO76Z768Ng+d3ArnQjave0az+UZMtWnyH6P+ZNEkmOAY4BWLp06SaGLUmSNPcMO50KsIhu2vLPgJuS/OwmXusuYJuB422AOyeUP6asPeO2cSp3w5D9H6WqTq6qvatq7yVLlmxi2JIkSXPPUElckj8Efh94TyvaAvjrTblQVd3XzrU4yWJgQVXdD1wMHNyaHdKOLwMOSDcXezBwaau/PslubRp2N+DaKfpLkiTNa8M+E3c4sCdwBUBV3ZJkxtduJTmXLtnaJckn6Z5t2/gzJe9qf08DTk/yOmDVwDNvZwCrgHuB17e2K4FTW9wntKnWrwCvSnIR8EMeeT5OkiRp3ho2iVtfVZWkAJI8eZhOVfXSSYpfPKHNvcARk/Q9kW5l6mDZNTzyXN7GsofpVrhKkiQ9bgz7TNznknwCeGqStwDnAqeMLixJkiRNZ9jVqR9J8jK6VaPPA95XVV8baWSSJEma0oxJXFsd+o9tatTETZIkaQ6YcTq1PXP2kyTbzNRWkiRJs2PYhQ33A2uSfI1utSgAVfX2kUQlSZKkaQ2bxH2lbZIkSZoDpk3ikiytqhuq6q9mKyBJkiTNbKZn4r60cSfJ50cciyRJkoY0UxI3+ML5Z48yEEmSJA1vpiSuptiXJEnSGM20sOH5SdbSjcg9qe3Tjquqth5pdJIkSZrUtElcVS2YrUAkSZI0vGHfnSpJkqQ5xCROkiSph0ziJEmSesgkTpIkqYdM4iRJknrIJE6SJKmHTOIkSZJ6yCROkiSph0ziJEmSesgkTpIkqYdM4iRJknrIJE6SJKmHTOIkSZJ6aFaTuCTLktyR5Py27Zxk3yTfbNu+rd2iJGcmWZXkuIH+K1q7c5Js38qem+TCJJckOXw270eSJGlcxjESt6qqDmzbtcCHgcPa9qHW5k3ABVW1HNglye5JngYcCSwHTgTe3doeD6wAXgKsTLJwFu9FkiRpLMaRxL0oyUVJPprkKUBV1W1VdRuwIclWdIna2a39We34hcB5VbUB+Cqwb6tfVlVrquo+4Cpg51m9G0mSpDGY7STuVmDnqtqvHf8OcM9A/d3AdsC2A+WPKauqh4AtW30m6S9JkjSvzWoSV1UPVNW6dvi3wLOAbQaabAPcCdw1UP6YsiQLgPWtfsMk/R8lyTFJVidZffvtt2+mu5EkSRqf2V7YsPXA4QHA91v54iSLgQVVdT9wMXBwa3dIO74MOCBJWt2lrf76JLu1adjdgGsnXreqTq6qvatq7yVLlozi1iRJkmbVbC8C2D/JB4B76aZWj6ZLzr7c6t/V/p4GnJ7kdXQLIdYAJDkDWNX6v761XQmcSncvJ7SpVkmSpHltVpO4qjqLbqHCoIuBF09ody9wxCT9T6RbmTpYdg2w38S2kiRJ85k/9itJktRDJnGSJEk9ZBInSZLUQyZxkiRJPWQSJ0mS1EO+Z1TSUHZ85lJuvfnGcYehHtvhGTtxy003jDsMad4wiZM0lFtvvpGD/2jiLwRJwzvnfa8YdwjSvOJ0qiRJUg+ZxEmSJPWQSZwkSVIPmcRJkiT1kEmcJElSD5nESZIk9ZBJnCRJUg+ZxEmSJPWQSZwkSVIPmcRJkiT1kEmcJElSD5nESZIk9ZBJnCRJUg+ZxEmSJPWQSZwkSVIPmcRJkiT1kEmcJElSD5nESZIk9dC8SeKSHJdkVZIzkywadzySJEmjNC+SuCS7A7tU1XLgAuCNYw5JkiRppOZFEgfsB5zd9s8Clo8xFkmSpJGbL0nctsA9bf9uYLsxxiJJkjRyqapxx/BTS/I2YH1VnZZkGfDHVfXagfpjgGPa4fOAa2Y9SE20GLhj3EFIm5Hfac1Hfq/H71lVtWSyivmSxP034L1V9WstoVtYVSeOOy5NLcnqqtp73HFIm4vfac1Hfq/ntoXjDmBzqKo1Sf41yUXA7cDrxx2TJEnSKM2LJA6gqt4z7hgkSZJmy3xZ2KD+OXncAUibmd9pzUd+r+ewefFMnCRJ0uONI3GSJEk9ZBKnkUiyTZLz27ZuYH//JGcnuSDJRUmOaO03trk8yaHjjl+aKMmBST42cPyKJO9Pcm2S85J8I8kXk+wy0P6m9r2+KMmzxxe99GhJliX5UpItk3w0yYXte3pyq/9Ukm8NtD8qSSV5qt/tuWPeLGzQ3FJV9wAHAiS5sqoOTLIFcBlwVFVd1Y73bV2ubW12BL7GI2/gkOa6dVV1EECSFwBnJNn4kwxnVtU7krwGOJZHfq9SmiveA/y4qvYHSPJLA3WVZFlVXQe8CrhyoM7v9hzgSJxm0y8Aa6rqKoCqerCqLhxsUFW3AIvGEZz006qqK+j+j26fCVXfA3aa/YikGb0WOH7jQVV9faDuTOA1SZ4ChEfejDTI7/YYmcRpNj0DuHm6Bkl2A340O+FII3Ez3Xd90EHAD8YQizSTJ1bVA1PUnUc3o/Iq4MtTtPG7PUZOp2o23Qy8Yoq6nZOcDzwEvG3WIpKGdz/wxIHjrYD7Jmn3DOAsuv++vjrJHnQ/Qr5i5BFKm+6BJFMlcg8DN9B9d18OvGmgzu/2HOBInGbTt4Dd22gbSRYm2a/VXVtVB1bVS6vq2+MLUZrSD4E9k2z8x+8BwHcHGyR5PrAnsLoVndm+16+pqttmL1RpaH9D90wbAEkOmlB/GvDVqlo7odzv9hzgSJxmTVU9mOS1wMeSLAIWAB8dc1jSUKrqziSfAC5M8iDw3ar6hyQnJTkPKGAt8KtVdX+SscYrDel44Pj22kqAq+mmUQGoqtU88o8SzTH+2K8kSVIPOZ0qSZLUQyZxkiRJPWQSJ0mS1EMmcZIkST1kEidJktRDJnGSHreSPJzkyiTfT3JG++mbTem/cgQxvT/J723u80qaf0ziJD2e3VdVe1TV7sB64DcGK9OZ7r+Tmz2Jk6RhmcRJUuciute/LUtydZK/AK4AdkpyZJI1bcTuBIAkfww8qY3kfaaV/W5r8/0k79h44iT/I8n3knw3yemt7FlJvt7Kv55k6ezfsqQ+840Nkh732qu0DgH+oRU9Dzi6qn4zyY7ACcBewF3AOUkOq6pjk/xWVe3RzrEXcDTwC0CAbyW5gG6E7w+AX6yqO5Js165xEvDpqvqrJG8E/hw4bFZuWNK84EicpMezJyW5ku61QjfQvScS4PqqurTt7wOcX1W3V9VDwGeA/Sc513Lgi1V1b1WtA74A7Ae8hO49k3dA9/qu1n5f4LNt//TWX5KG5kicpMez+zaOpG3U3nl672DRkOeaql3o3qs6E9+BKGmTOBInSdP7FnBAksVJFgBHAhe0ugeTbNH2LwQOS7IoyZOBw+mes/s68KtJngYwMJ36TeC1bf8oYNXob0XSfOJInCRNo6puTfIe4Dy6UbWzq+rvWvXJwPeSXFFVRyX5FHBZqzu1qr4DkOQ44IIkDwPfAd4AvB34ZJJ3AbfTPU8nSUNLlSP4kiRJfeN0qiRJUg+ZxEmSJPWQSZwkSVIPmcRJkiT1kEmcJElSD5nESZIk9ZBJnCRJUg+ZxEmSJPXQ/wdrDeETSC/zawAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_freq(data, title_):\n",
    "    fig = plt.figure(figsize = (10, 3))\n",
    "    title = fig.suptitle(title_ + \" Frequency\", fontsize=14)\n",
    "    fig.subplots_adjust(top=0.85, wspace=0.3)\n",
    "\n",
    "    ax = fig.add_subplot(1, 1, 1)\n",
    "    ax.set_xlabel(title_)\n",
    "    ax.set_ylabel(\"Frequency\") \n",
    "    w_q = data.value_counts()\n",
    "    w_q = (list(w_q.index), list(w_q.values))\n",
    "    ax.tick_params(axis='both', which='major', labelsize=8.5)\n",
    "    bar = ax.bar(w_q[0], w_q[1], color='steelblue', \n",
    "            edgecolor='black', linewidth=1)\n",
    "\n",
    "plot_freq(updated_train['Proto'], \"Protocol\")\n",
    "plot_freq(updated_train['Flags'], \"Flags\")\n",
    "plot_freq(updated_train['class'], \"Class\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, ax = plt.subplots(figsize=(10, 6))\n",
    "corr = updated_train.corr()\n",
    "hm = sns.heatmap(round(corr,2), annot=True, ax=ax, cmap=\"coolwarm\",fmt='.2f',\n",
    "                 linewidths=.05)\n",
    "f.subplots_adjust(top=0.93)\n",
    "t= f.suptitle('CIDDS-01 Attributes Correlation Heatmap', fontsize=14)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "updated_train.hist(bins=15, color='steelblue', edgecolor='black', linewidth=1.0,\n",
    "           xlabelsize=8, ylabelsize=8, grid=False)\n",
    "\n",
    "plt.tight_layout(rect=(0, 0, 1.2, 1.2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>srcIP</th>\n",
       "      <th>srcPt</th>\n",
       "      <th>dstIP</th>\n",
       "      <th>dstPt</th>\n",
       "      <th>proto</th>\n",
       "      <th>packets</th>\n",
       "      <th>bytes</th>\n",
       "      <th>duration</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>192.168.220.10</td>\n",
       "      <td>42968_p</td>\n",
       "      <td>DNS</td>\n",
       "      <td>53_p</td>\n",
       "      <td>UDP</td>\n",
       "      <td>2_k</td>\n",
       "      <td>176_b</td>\n",
       "      <td>0.0_d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10086_192</td>\n",
       "      <td>80_p</td>\n",
       "      <td>192.168.220.14</td>\n",
       "      <td>43392_p</td>\n",
       "      <td>TCP</td>\n",
       "      <td>6_k</td>\n",
       "      <td>3230_b</td>\n",
       "      <td>0.64_d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>16569_82</td>\n",
       "      <td>80_p</td>\n",
       "      <td>192.168.220.8</td>\n",
       "      <td>41440_p</td>\n",
       "      <td>TCP</td>\n",
       "      <td>1_k</td>\n",
       "      <td>66_b</td>\n",
       "      <td>0.0_d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10296_238</td>\n",
       "      <td>443_p</td>\n",
       "      <td>192.168.200.9</td>\n",
       "      <td>59692_p</td>\n",
       "      <td>TCP</td>\n",
       "      <td>2_k</td>\n",
       "      <td>154_b</td>\n",
       "      <td>0.0_d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>192.168.220.14</td>\n",
       "      <td>52236_p</td>\n",
       "      <td>10056_243</td>\n",
       "      <td>443_p</td>\n",
       "      <td>TCP</td>\n",
       "      <td>1_k</td>\n",
       "      <td>66_b</td>\n",
       "      <td>0.0_d</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            srcIP    srcPt           dstIP    dstPt proto packets   bytes  \\\n",
       "0  192.168.220.10  42968_p             DNS     53_p   UDP     2_k   176_b   \n",
       "1       10086_192     80_p  192.168.220.14  43392_p   TCP     6_k  3230_b   \n",
       "2        16569_82     80_p   192.168.220.8  41440_p   TCP     1_k    66_b   \n",
       "3       10296_238    443_p   192.168.200.9  59692_p   TCP     2_k   154_b   \n",
       "4  192.168.220.14  52236_p       10056_243    443_p   TCP     1_k    66_b   \n",
       "\n",
       "  duration  \n",
       "0    0.0_d  \n",
       "1   0.64_d  \n",
       "2    0.0_d  \n",
       "3    0.0_d  \n",
       "4    0.0_d  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(260391, 8)\n"
     ]
    }
   ],
   "source": [
    "# convert to IP2Vec format\n",
    "new_col_names = [\"srcIP\", \"srcPt\", \"dstIP\", \"dstPt\", \"proto\", \"packets\", \"bytes\", \"duration\"]\n",
    "\n",
    "ip2vec_format = []\n",
    "for row in updated_train.values:\n",
    "    proto = row[3].strip()\n",
    "    srcIP = row[4].strip()\n",
    "    srcPt = str(row[5]).strip() + \"_p\"\n",
    "    dstIP = row[6].strip()\n",
    "    dstPt = str(row[7]).strip() + \"_p\"\n",
    "    bytez = str(row[9]).strip() + \"_b\"\n",
    "    packets = str(row[8]).strip() + \"_k\"\n",
    "    duration = str(row[2]).strip() + \"_d\"\n",
    "    ip2vec_format.append([srcIP, srcPt, dstIP, dstPt, proto, packets, bytez, duration])\n",
    "\n",
    "ip2vec_train = pd.DataFrame(ip2vec_format, columns = new_col_names)\n",
    "display(ip2vec_train.head())\n",
    "print(ip2vec_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "437\n"
     ]
    }
   ],
   "source": [
    "def save2file(filename, data):\n",
    "    pickle_out = open(\"C:/Users/Akarsh/Downloads/DP_scripts/store_emb/\" + filename + \".pickle\", \"wb\")\n",
    "    pickle.dump(data, pickle_out, protocol=pickle.HIGHEST_PROTOCOL)\n",
    "    pickle_out.close()\n",
    "\n",
    "main_str = str(randint(0, 1000))\n",
    "print(main_str)\n",
    "\n",
    "filename = \"updated_train_\"  + main_str\n",
    "save2file(filename, updated_train)\n",
    "\n",
    "filename = \"ip2vec_train_\" + main_str\n",
    "save2file(filename, ip2vec_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data and hyperparams saved...\n"
     ]
    }
   ],
   "source": [
    "save = True\n",
    "if save:\n",
    "    df = pd.DataFrame.from_records([{'operation': 'ip2vec_preprocess', 'main_str': main_str, 'num_samples': nsamples}])\n",
    "    df.to_csv(\"C:/Users/Akarsh/Downloads/DP_scripts/store_emb/store_params.csv\", mode='a', index=False)\n",
    "    print('data and hyperparams saved...')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}