5.1 UNSW_Malicious_ results  with augmentation and aggregation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## This file makes machine learning application for aggregation packets for UNSW ACM SOSR 2019 Attack & Benign Dataset with 4 diffirent group size (3, 6, 9, 12)\n",
    "\n",
    "\n",
    "### Used machine learning algorithms: RF (Random Forest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "from numpy import array\n",
    "from random import random\n",
    "from sklearn import metrics\n",
    "from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis as QDA\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier\n",
    "from sklearn.metrics import average_precision_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.naive_bayes import BernoulliNB#57\n",
    "from sklearn.naive_bayes import GaussianNB#52\n",
    "from sklearn.naive_bayes import MultinomialNB#56\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.metrics import accuracy_score\n",
    "import csv\n",
    "import math\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import os\n",
    "import pandas as pd\n",
    "import sklearn\n",
    "import time\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-------------"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  importing relevant libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "loop=\"UNSW_Malicious_h_train_sk.csv\"\n",
    "loop2=\"UNSW_Malicious_h_test_sk.csv\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>ARP</th>\n",
       "      <th>LLC</th>\n",
       "      <th>EAPOL</th>\n",
       "      <th>IP</th>\n",
       "      <th>ICMP</th>\n",
       "      <th>ICMP6</th>\n",
       "      <th>TCP</th>\n",
       "      <th>UDP</th>\n",
       "      <th>TCP_w_size</th>\n",
       "      <th>HTTP</th>\n",
       "      <th>...</th>\n",
       "      <th>IP_padding</th>\n",
       "      <th>IP_add_count</th>\n",
       "      <th>IP_ralert</th>\n",
       "      <th>Portcl_src</th>\n",
       "      <th>Portcl_dst</th>\n",
       "      <th>Pck_size</th>\n",
       "      <th>Pck_rawdata</th>\n",
       "      <th>Entropy</th>\n",
       "      <th>Label</th>\n",
       "      <th>Mac</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4033</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ACK Flooding</td>\n",
       "      <td>bc:1c:81:4b:ae:ba</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7527</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ACK Flooding</td>\n",
       "      <td>bc:1c:81:4b:ae:ba</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7938</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ACK Flooding</td>\n",
       "      <td>bc:1c:81:4b:ae:ba</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8937</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ACK Flooding</td>\n",
       "      <td>bc:1c:81:4b:ae:ba</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2442</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>ACK Flooding</td>\n",
       "      <td>bc:1c:81:4b:ae:ba</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>309995</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>153</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>lifx</td>\n",
       "      <td>d0:73:d5:01:83:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>309996</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1460</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>74</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>125</td>\n",
       "      <td>1</td>\n",
       "      <td>6.071098</td>\n",
       "      <td>lifx</td>\n",
       "      <td>d0:73:d5:01:83:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>309997</th>\n",
       "      <td>1</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>...</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>0.000000</td>\n",
       "      <td>lifx</td>\n",
       "      <td>d0:73:d5:01:83:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>309998</th>\n",
       "      <td>1</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>...</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>0.000000</td>\n",
       "      <td>lifx</td>\n",
       "      <td>d0:73:d5:01:83:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>309999</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>117</td>\n",
       "      <td>1</td>\n",
       "      <td>4.207220</td>\n",
       "      <td>lifx</td>\n",
       "      <td>d0:73:d5:01:83:08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>310000 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        ARP  LLC  EAPOL  IP  ICMP  ICMP6  TCP  UDP  TCP_w_size  HTTP  ...  \\\n",
       "0         0    0      0   1     0      0    1    0        4033     0  ...   \n",
       "1         0    0      0   1     0      0    1    0        7527     0  ...   \n",
       "2         0    0      0   1     0      0    1    0        7938     0  ...   \n",
       "3         0    0      0   1     0      0    1    0        8937     0  ...   \n",
       "4         0    0      0   1     0      0    1    0        2442     0  ...   \n",
       "...     ...  ...    ...  ..   ...    ...  ...  ...         ...   ...  ...   \n",
       "309995    0    0      0   1     0      0    0    1           0     0  ...   \n",
       "309996    0    0      0   1     0      0    1    0        1460     0  ...   \n",
       "309997    1    0      0   0     0      0    0    0           0     0  ...   \n",
       "309998    1    0      0   0     0      0    0    0           0     0  ...   \n",
       "309999    0    0      1   0     0      0    0    0           0     0  ...   \n",
       "\n",
       "        IP_padding  IP_add_count  IP_ralert  Portcl_src  Portcl_dst  Pck_size  \\\n",
       "0                0            32          0           2           3        40   \n",
       "1                0            32          0           2           3        40   \n",
       "2                0            32          0           2           3        40   \n",
       "3                0            32          0           2           3        40   \n",
       "4                0            32          0           2           3        40   \n",
       "...            ...           ...        ...         ...         ...       ...   \n",
       "309995           0           153          0           2           1        58   \n",
       "309996           0            74          0           2           3       125   \n",
       "309997           0             0          0           0           0         0   \n",
       "309998           0             0          0           0           0         0   \n",
       "309999           0             0          0           0           0       117   \n",
       "\n",
       "        Pck_rawdata   Entropy         Label                Mac  \n",
       "0                 0  0.000000  ACK Flooding  bc:1c:81:4b:ae:ba  \n",
       "1                 0  0.000000  ACK Flooding  bc:1c:81:4b:ae:ba  \n",
       "2                 0  0.000000  ACK Flooding  bc:1c:81:4b:ae:ba  \n",
       "3                 0  0.000000  ACK Flooding  bc:1c:81:4b:ae:ba  \n",
       "4                 0  0.000000  ACK Flooding  bc:1c:81:4b:ae:ba  \n",
       "...             ...       ...           ...                ...  \n",
       "309995            0  0.000000          lifx  d0:73:d5:01:83:08  \n",
       "309996            1  6.071098          lifx  d0:73:d5:01:83:08  \n",
       "309997            0  0.000000          lifx  d0:73:d5:01:83:08  \n",
       "309998            0  0.000000          lifx  d0:73:d5:01:83:08  \n",
       "309999            1  4.207220          lifx  d0:73:d5:01:83:08  \n",
       "\n",
       "[310000 rows x 27 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(loop)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Discovering Labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['ACK Flooding',\n",
       " 'Ezviz',\n",
       " 'HTTP Flooding',\n",
       " 'Host Discovery',\n",
       " 'Netatmo Camera',\n",
       " 'Nugu',\n",
       " 'Samsung Camera',\n",
       " 'Smurf',\n",
       " 'Snmp',\n",
       " 'Ssdp',\n",
       " 'TP Link Plug',\n",
       " 'TcpSynReflection',\n",
       " 'Telnet Bruteforce',\n",
       " 'UDP Flooding',\n",
       " 'WEMO Motion Sensor',\n",
       " 'WEMO Power Switch',\n",
       " 'chromecast',\n",
       " 'lifx']"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def target_names():\n",
    "    name=loop\n",
    "    df = pd.read_csv(name)\n",
    "    target_names=sorted(list(df[df.columns[-2]].unique()))\n",
    "    return target_names\n",
    "target_names=target_names()\n",
    "target_names"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hyperparameters of machine learning algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "ml_list={\"Random Forest R\":RandomForestClassifier(bootstrap=True, class_weight=None, criterion='entropy',\n",
    "                       max_depth=27.0, max_features=1, max_leaf_nodes=None,\n",
    "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
    "                       min_samples_leaf=1, min_samples_split=10,\n",
    "                       min_weight_fraction_leaf=0.0, n_estimators=69,\n",
    "                       n_jobs=None, oob_score=False, random_state=None,\n",
    "                       verbose=0, warm_start=False)}\n",
    "                          "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Aggregation Algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "altime=0\n",
    "def most_frequent(List): \n",
    "    return max(set(List), key = List.count) \n",
    "\n",
    "def split(a, n):\n",
    "    k, m = divmod(len(a), n)\n",
    "    return (a[i * k + min(i, m):(i + 1) * k + min(i + 1, m)] for i in range(n))\n",
    "\n",
    "\n",
    "def merged(mac,y_test,predict):\n",
    "    step=12\n",
    "    second=time.time()\n",
    "    mac_test=[]\n",
    "    for i in y_test.index:\n",
    "        mac_test.append(mac[i])\n",
    "\n",
    "    d_list=sorted(list(mac.unique()))\n",
    "    devices={}\n",
    "    for i in d_list:\n",
    "        devices[i]=[]    \n",
    "\n",
    "\n",
    "    new_y=[0]*len(y_test)\n",
    "\n",
    "    for i,ii in enumerate (mac_test):\n",
    "        devices[ii].append(i)\n",
    "        #devices[]\n",
    "    for i in devices:\n",
    "        \n",
    "        a = [devices[i][j:j + step] for j in range(0, len(devices[i]), step)]  \n",
    "        for ii in a:\n",
    "            step_list=[]\n",
    "            for iii in ii:\n",
    "                step_list.append(predict[iii])\n",
    "            add=most_frequent(list(step_list))\n",
    "            for iii in ii:\n",
    "                new_y[iii]=add\n",
    "                \n",
    "    return new_y,time.time()-second"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculation of evaluations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def score(altime,train_time,test_time,predict,y_test,class_based_results):\n",
    "    precision=[]\n",
    "    recall=[]\n",
    "    f1=[]\n",
    "    accuracy=[]\n",
    "    total_time=[]\n",
    "    kappa=[]\n",
    "    accuracy_b=[]\n",
    "    \n",
    "    rc=sklearn.metrics.recall_score(y_test, predict,average= \"macro\")\n",
    "    pr=sklearn.metrics.precision_score(y_test, predict,average= \"macro\")\n",
    "    f_1=sklearn.metrics.f1_score(y_test, predict,average= \"macro\")        \n",
    "    report = classification_report(y_test, predict, target_names=target_names,output_dict=True)\n",
    "    cr = pd.DataFrame(report).transpose()\n",
    "    if class_based_results.empty:\n",
    "        class_based_results =cr\n",
    "    else:\n",
    "        class_based_results = class_based_results.add(cr, fill_value=0)\n",
    "    precision.append(float(pr))\n",
    "    recall.append(float(rc))\n",
    "    f1.append(float(f_1))\n",
    "    accuracy_b.append(balanced_accuracy_score( y_test,predict))\n",
    "    accuracy.append(accuracy_score(y_test, predict))\n",
    "    #clf.score(X_test, y_test))\n",
    "    #print(balanced_accuracy_score( y_test,predict))\n",
    "    #t_time.append(float((time.time()-second)) )\n",
    "    kappa.append(round(float(sklearn.metrics.cohen_kappa_score(y_test, predict, \n",
    "    labels=None, weights=None, sample_weight=None)),15))\n",
    "    print ('%-15s %-3s %-3s %-6s  %-5s %-5s %-5s %-5s %-5s %-5s %-8s %-8s%-8s%-8s' % (dname,i,cv,ii[0:6],str(round(np.mean(accuracy),2)),str(round(np.mean(accuracy_b),2)),\n",
    "        str(round(np.mean(precision),2)), str(round(np.mean(recall),2)),str(round(np.mean(f1),2)), \n",
    "        str(round(np.mean(kappa),2)),str(round(np.mean(train_time),2)),str(round(np.mean(test_time),2)),str(round(np.mean(test_time)+np.mean(train_time),2)),str(round(np.mean(altime),2))))\n",
    "    lines=(str(dname)+\",\"+str(i)+\",\"+str(cv)+\",\"+str(ii)+\",\"+str(round(np.mean(accuracy),15))+\",\"+str(round(np.mean(accuracy_b),15))+\",\"+str(round(np.mean(precision),15))+\",\"+ str(round(np.mean(recall),15))+\",\"+str(round(np.mean(f1),15))+\",\"+str(round(np.mean(kappa),15))+\",\"+str(round(np.mean(train_time),15))+\",\"+str(round(np.mean(test_time),15))+\",\"+str(round(np.mean(test_time)+np.mean(train_time),15))+str(altime)+\"\\n\")\n",
    "    return lines,class_based_results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Machine learning applications (100 repetitions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "## GROUP SIZE 1 (= individual packets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset         T   CV  ML alg  Acc   b_Acc Prec  Rec   F1    kap   tra-T    test-T  total   al-time \n",
      "train.csv       0   1   Random  0.98  0.96  0.98  0.96  0.97  0.98  6.46     0.78    7.24    0.0     \n",
      "train.csv       1   2   Random  0.98  0.96  0.98  0.96  0.97  0.98  5.76     0.65    6.41    0.0     \n",
      "train.csv       2   3   Random  0.98  0.96  0.98  0.96  0.97  0.98  5.86     0.65    6.52    0.0     \n",
      "train.csv       3   4   Random  0.98  0.96  0.98  0.96  0.97  0.98  5.65     0.65    6.3     0.0     \n",
      "train.csv       4   5   Random  0.98  0.96  0.98  0.96  0.97  0.98  5.72     0.64    6.37    0.0     \n",
      "train.csv       5   6   Random  0.98  0.96  0.98  0.96  0.97  0.98  5.86     0.64    6.51    0.0     \n",
      "train.csv       6   7   Random  0.98  0.96  0.98  0.96  0.97  0.98  5.72     0.63    6.35    0.0     \n",
      "train.csv       7   8   Random  0.98  0.96  0.98  0.96  0.97  0.98  5.63     0.64    6.27    0.0     \n",
      "train.csv       8   9   Random  0.98  0.96  0.98  0.96  0.97  0.98  5.77     0.63    6.4     0.0     \n",
      "train.csv       9   10  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.79     0.63    6.43    0.0     \n",
      "train.csv       10  11  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.76     0.63    6.38    0.0     \n",
      "train.csv       11  12  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.71     0.63    6.34    0.0     \n",
      "train.csv       12  13  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.74     0.63    6.37    0.0     \n",
      "train.csv       13  14  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.66     0.63    6.29    0.0     \n",
      "train.csv       14  15  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.7      0.64    6.34    0.0     \n",
      "train.csv       15  16  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.75     0.64    6.39    0.0     \n",
      "train.csv       16  17  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.76     0.64    6.4     0.0     \n",
      "train.csv       17  18  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.64     0.64    6.28    0.0     \n",
      "train.csv       18  19  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.61     0.63    6.23    0.0     \n",
      "train.csv       19  20  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.63     0.63    6.26    0.0     \n",
      "train.csv       20  21  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.68     0.64    6.32    0.0     \n",
      "train.csv       21  22  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.65     0.64    6.29    0.0     \n",
      "train.csv       22  23  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.75     0.63    6.38    0.0     \n",
      "train.csv       23  24  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.68     0.64    6.32    0.0     \n",
      "train.csv       24  25  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.7      0.64    6.34    0.0     \n",
      "train.csv       25  26  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.68     0.64    6.32    0.0     \n",
      "train.csv       26  27  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.71     0.63    6.34    0.0     \n",
      "train.csv       27  28  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.8      0.64    6.44    0.0     \n",
      "train.csv       28  29  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.64     0.64    6.27    0.0     \n",
      "train.csv       29  30  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.7      0.64    6.34    0.0     \n",
      "train.csv       30  31  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.72     0.63    6.35    0.0     \n",
      "train.csv       31  32  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.74     0.67    6.41    0.0     \n",
      "train.csv       32  33  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.69     0.64    6.33    0.0     \n",
      "train.csv       33  34  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.7      0.67    6.36    0.0     \n",
      "train.csv       34  35  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.79     0.64    6.43    0.0     \n",
      "train.csv       35  36  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.64     0.64    6.28    0.0     \n",
      "train.csv       36  37  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.6      0.64    6.24    0.0     \n",
      "train.csv       37  38  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.69     0.64    6.32    0.0     \n",
      "train.csv       38  39  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.65     0.64    6.29    0.0     \n",
      "train.csv       39  40  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.77     0.63    6.41    0.0     \n",
      "train.csv       40  41  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.69     0.64    6.32    0.0     \n",
      "train.csv       41  42  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.63     0.64    6.27    0.0     \n",
      "train.csv       42  43  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.69     0.63    6.32    0.0     \n",
      "train.csv       43  44  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.59     0.64    6.23    0.0     \n",
      "train.csv       44  45  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.7      0.65    6.35    0.0     \n",
      "train.csv       45  46  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.71     0.64    6.34    0.0     \n",
      "train.csv       46  47  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.78     0.64    6.42    0.0     \n",
      "train.csv       47  48  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.71     0.63    6.34    0.0     \n",
      "train.csv       48  49  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.62     0.64    6.27    0.0     \n",
      "train.csv       49  50  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.76     0.64    6.4     0.0     \n",
      "train.csv       50  51  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.57     0.63    6.2     0.0     \n",
      "train.csv       51  52  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.62     0.65    6.27    0.0     \n",
      "train.csv       52  53  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.67     0.63    6.29    0.0     \n",
      "train.csv       53  54  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.7      0.64    6.34    0.0     \n",
      "train.csv       54  55  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.74     0.63    6.37    0.0     \n",
      "train.csv       55  56  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.72     0.65    6.36    0.0     \n",
      "train.csv       56  57  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.76     0.64    6.4     0.0     \n",
      "train.csv       57  58  Random  0.98  0.96  0.98  0.96  0.97  0.97  5.69     0.63    6.32    0.0     \n",
      "train.csv       58  59  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.69     0.64    6.33    0.0     \n",
      "train.csv       59  60  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.73     0.63    6.36    0.0     \n",
      "train.csv       60  61  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.66     0.64    6.3     0.0     \n",
      "train.csv       61  62  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.65     0.64    6.29    0.0     \n",
      "train.csv       62  63  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.62     0.64    6.27    0.0     \n",
      "train.csv       63  64  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.61     0.63    6.24    0.0     \n",
      "train.csv       64  65  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.67     0.64    6.31    0.0     \n",
      "train.csv       65  66  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.66     0.63    6.29    0.0     \n",
      "train.csv       66  67  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.82     0.63    6.45    0.0     \n",
      "train.csv       67  68  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.58     0.63    6.21    0.0     \n",
      "train.csv       68  69  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.77     0.63    6.39    0.0     \n",
      "train.csv       69  70  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.69     0.63    6.32    0.0     \n",
      "train.csv       70  71  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.81     0.64    6.45    0.0     \n",
      "train.csv       71  72  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.65     0.64    6.29    0.0     \n",
      "train.csv       72  73  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.73     0.64    6.37    0.0     \n",
      "train.csv       73  74  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.7      0.64    6.34    0.0     \n",
      "train.csv       74  75  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.74     0.63    6.38    0.0     \n",
      "train.csv       75  76  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.65     0.65    6.3     0.0     \n",
      "train.csv       76  77  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.71     0.64    6.35    0.0     \n",
      "train.csv       77  78  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.7      0.65    6.36    0.0     \n",
      "train.csv       78  79  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.8      0.64    6.44    0.0     \n",
      "train.csv       79  80  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.73     0.64    6.36    0.0     \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train.csv       80  81  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.61     0.64    6.25    0.0     \n",
      "train.csv       81  82  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.66     0.64    6.3     0.0     \n",
      "train.csv       82  83  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.65     0.63    6.28    0.0     \n",
      "train.csv       83  84  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.83     0.64    6.47    0.0     \n",
      "train.csv       84  85  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.81     0.65    6.46    0.0     \n",
      "train.csv       85  86  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.68     0.63    6.32    0.0     \n",
      "train.csv       86  87  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.67     0.64    6.31    0.0     \n",
      "train.csv       87  88  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.68     0.64    6.31    0.0     \n",
      "train.csv       88  89  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.66     0.64    6.3     0.0     \n",
      "train.csv       89  90  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.67     0.63    6.31    0.0     \n",
      "train.csv       90  91  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.64     0.63    6.28    0.0     \n",
      "train.csv       91  92  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.62     0.62    6.25    0.0     \n",
      "train.csv       92  93  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.66     0.63    6.29    0.0     \n",
      "train.csv       93  94  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.75     0.64    6.39    0.0     \n",
      "train.csv       94  95  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.69     0.63    6.32    0.0     \n",
      "train.csv       95  96  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.81     0.64    6.44    0.0     \n",
      "train.csv       96  97  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.59     0.64    6.22    0.0     \n",
      "train.csv       97  98  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.56     0.64    6.2     0.0     \n",
      "train.csv       98  99  Random  0.98  0.96  0.98  0.96  0.97  0.98  5.76     0.64    6.4     0.0     \n",
      "train.csv       99  100 Random  0.98  0.96  0.98  0.96  0.97  0.98  5.66     0.64    6.29    0.0     \n",
      "                    precision    recall  f1-score       support\n",
      "ACK Flooding         0.999603  1.000000  0.999802   6000.000000\n",
      "Ezviz                0.999399  0.998320  0.998859   3000.000000\n",
      "HTTP Flooding        1.000000  1.000000  1.000000   6000.000000\n",
      "Host Discovery       0.820569  1.000000  0.901442   6000.000000\n",
      "Netatmo Camera       0.997382  0.901457  0.946996   3000.000000\n",
      "Nugu                 0.999240  0.990423  0.994812   3000.000000\n",
      "Samsung Camera       0.994298  0.896167  0.942685   3000.000000\n",
      "Smurf                1.000000  1.000000  1.000000   9000.000000\n",
      "Snmp                 1.000000  1.000000  1.000000   3000.000000\n",
      "Ssdp                 0.999332  0.999833  0.999583   6000.000000\n",
      "TP Link Plug         0.973336  0.779120  0.865462   3000.000000\n",
      "TcpSynReflection     0.998007  0.999944  0.998975  18000.000000\n",
      "Telnet Bruteforce    0.999983  1.000000  0.999992   6000.000000\n",
      "UDP Flooding         1.000000  1.000000  1.000000   6000.000000\n",
      "WEMO Motion Sensor   0.988710  0.959980  0.974133   3000.000000\n",
      "WEMO Power Switch    0.973301  0.881583  0.925174   3000.000000\n",
      "chromecast           0.996551  0.975660  0.985995   3000.000000\n",
      "lifx                 0.831494  0.912763  0.870233   3000.000000\n",
      "accuracy             0.977252  0.977252  0.977252      0.977252\n",
      "macro avg            0.976178  0.960847  0.966897  93000.000000\n",
      "weighted avg         0.980024  0.977252  0.977414  93000.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1008 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ths = open(\"1_08_2020_normal.csv\", \"w\")\n",
    "ths.write(\"Dataset,T,CV,ML algorithm,Acc,b_Acc,Precision, Recall , F1-score, kappa ,tra-Time,test-Time,total-Time),Al-Time)\\n\")\n",
    "repetition=100\n",
    "\n",
    "\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "from sklearn.preprocessing import Normalizer\n",
    "print ('%-15s %-3s %-3s %-6s  %-5s %-5s %-5s %-5s %-5s %-5s %-8s %-8s%-8s%-8s'%\n",
    "           (\"Dataset\",\"T\",\"CV\",\"ML alg\",\"Acc\",\"b_Acc\",\"Prec\", \"Rec\" , \"F1\", \"kap\" ,\"tra-T\",\"test-T\",\"total\",\"al-time\"))\n",
    "\n",
    "for ii in ml_list:\n",
    "    class_based_results=pd.DataFrame()#\"\" #pd.DataFrame(0, index=np.arange((len(target_names)+3)), columns=[\"f1-score\",\"precision\",\"recall\",\"support\"])\n",
    "    cm=pd.DataFrame()\n",
    "    cv=0\n",
    "    for i in range(repetition):\n",
    "\n",
    "\n",
    "        #TRAIN\n",
    "        df = pd.read_csv(loop)\n",
    "        m_train=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_train =df[df.columns[0:-1]]\n",
    "        X_train=np.array(X_train)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_train=df[df.columns[-1]].cat.codes  \n",
    "\n",
    "        #TEST\n",
    "        df = pd.read_csv(loop2)#,header=None )\n",
    "        m_test=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_test =df[df.columns[0:-1]]\n",
    "        X_test=np.array(X_test)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_test=df[df.columns[-1]].cat.codes\n",
    "        \n",
    "\n",
    "\n",
    "        dname=loop  \n",
    "        results_y=[]\n",
    "        cv+=1\n",
    "        results_y.append(y_test)\n",
    "\n",
    "\n",
    "        precision=[]\n",
    "        recall=[]\n",
    "        f1=[]\n",
    "        accuracy=[]\n",
    "        train_time=[]\n",
    "        test_time=[]\n",
    "        total_time=[]\n",
    "        kappa=[]\n",
    "        accuracy_b=[]\n",
    "\n",
    "            #machine learning algorithm is applied in this section\n",
    "        clf = ml_list[ii]#choose algorithm from ml_list dictionary\n",
    "        second=time.time()\n",
    "        clf.fit(X_train, y_train)\n",
    "        train_time.append(float((time.time()-second)) )\n",
    "        second=time.time()\n",
    "        predict =clf.predict(X_test)\n",
    "        test_time.append(float((time.time()-second)) )\n",
    "        #normal\n",
    "        lines,class_based_results=score(altime,train_time,test_time,predict,y_test,class_based_results)\n",
    "        #merged\n",
    "        #predict=merged(m_test,y_test,predict)\n",
    "        #lines,class_based_results=score(train_time,test_time,predict,y_test,class_based_results)\n",
    "\n",
    "        ths.write (lines)\n",
    "\n",
    "        df_cm = pd.DataFrame(confusion_matrix(y_test, predict))\n",
    "        results_y.append(predict)\n",
    "        if cm.empty:\n",
    "            cm =df_cm\n",
    "        else:\n",
    "            cm = cm.add(df_cm, fill_value=0)\n",
    "\n",
    "\n",
    "print(class_based_results/(repetition)) \n",
    "graph_name=\"normal_confusion matrix\" +str(ii)+\".pdf\"          \n",
    "plt.figure(figsize = (20,14))\n",
    "sns.heatmap(cm,xticklabels=target_names, yticklabels=target_names, annot=True)\n",
    "plt.savefig(graph_name,bbox_inches='tight')#, dpi=400)\n",
    "plt.show()\n",
    "#print(cm)\n",
    "print(\"\\n\\n\\n\") \n",
    "ths.close()  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GROUP SIZE 12"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset         T   CV  ML alg  Acc   b_Acc Prec  Rec   F1    kap   tra-T    test-T  total   al-time \n",
      "train.csv       0   1   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.64    6.37    0.9     \n",
      "train.csv       1   2   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.63    6.3     0.88    \n",
      "train.csv       2   3   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.76     0.64    6.39    0.87    \n",
      "train.csv       3   4   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.58     0.63    6.21    0.87    \n",
      "train.csv       4   5   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.64    6.39    0.89    \n",
      "train.csv       5   6   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.62     0.63    6.25    0.88    \n",
      "train.csv       6   7   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.59     0.63    6.21    0.88    \n",
      "train.csv       7   8   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.62    6.26    0.88    \n",
      "train.csv       8   9   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.64    6.41    0.87    \n",
      "train.csv       9   10  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.64    6.39    0.97    \n",
      "train.csv       10  11  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.63    6.36    0.88    \n",
      "train.csv       11  12  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.65    6.3     0.87    \n",
      "train.csv       12  13  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.63    6.31    0.91    \n",
      "train.csv       13  14  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.63    6.29    0.89    \n",
      "train.csv       14  15  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.66    6.32    0.95    \n",
      "train.csv       15  16  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.32    0.87    \n",
      "train.csv       16  17  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.7     6.49    0.93    \n",
      "train.csv       17  18  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.93     0.63    6.56    0.91    \n",
      "train.csv       18  19  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.63    6.33    0.87    \n",
      "train.csv       19  20  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.78     0.65    6.43    0.87    \n",
      "train.csv       20  21  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.64    6.36    0.87    \n",
      "train.csv       21  22  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.32    0.88    \n",
      "train.csv       22  23  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.63    6.34    0.87    \n",
      "train.csv       23  24  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.32    0.87    \n",
      "train.csv       24  25  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.64    6.41    0.87    \n",
      "train.csv       25  26  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.63    6.32    0.86    \n",
      "train.csv       26  27  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.63    6.34    0.87    \n",
      "train.csv       27  28  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.64    6.29    0.93    \n",
      "train.csv       28  29  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.31    0.86    \n",
      "train.csv       29  30  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.65    6.37    0.88    \n",
      "train.csv       30  31  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.63    6.28    0.87    \n",
      "train.csv       31  32  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.64    6.36    0.87    \n",
      "train.csv       32  33  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.66    6.36    0.87    \n",
      "train.csv       33  34  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.64    6.42    0.86    \n",
      "train.csv       34  35  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.63    6.34    0.86    \n",
      "train.csv       35  36  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.35    0.86    \n",
      "train.csv       36  37  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.55     0.63    6.18    0.87    \n",
      "train.csv       37  38  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.65    6.38    0.87    \n",
      "train.csv       38  39  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.65    6.41    0.86    \n",
      "train.csv       39  40  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.64    6.28    0.86    \n",
      "train.csv       40  41  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.85     0.65    6.5     0.86    \n",
      "train.csv       41  42  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.31    0.85    \n",
      "train.csv       42  43  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.64    6.39    0.86    \n",
      "train.csv       43  44  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.64    6.31    0.87    \n",
      "train.csv       44  45  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.86    \n",
      "train.csv       45  46  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.81     0.63    6.44    0.87    \n",
      "train.csv       46  47  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.6      0.63    6.23    0.88    \n",
      "train.csv       47  48  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.32    0.86    \n",
      "train.csv       48  49  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.86    \n",
      "train.csv       49  50  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.82     0.65    6.46    0.88    \n",
      "train.csv       50  51  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.62     0.64    6.27    0.88    \n",
      "train.csv       51  52  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.86    \n",
      "train.csv       52  53  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.31    0.87    \n",
      "train.csv       53  54  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.65    6.35    0.88    \n",
      "train.csv       54  55  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.64    6.38    0.87    \n",
      "train.csv       55  56  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.64    6.39    0.86    \n",
      "train.csv       56  57  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.64    6.28    0.87    \n",
      "train.csv       57  58  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.32    0.87    \n",
      "train.csv       58  59  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.63     0.65    6.28    0.88    \n",
      "train.csv       59  60  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.64    6.3     0.86    \n",
      "train.csv       60  61  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.31    0.86    \n",
      "train.csv       61  62  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.65    6.4     0.87    \n",
      "train.csv       62  63  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.78     0.63    6.42    0.88    \n",
      "train.csv       63  64  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.86    \n",
      "train.csv       64  65  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.31    0.87    \n",
      "train.csv       65  66  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.65    6.43    0.87    \n",
      "train.csv       66  67  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.63    6.34    0.88    \n",
      "train.csv       67  68  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.64    6.34    0.89    \n",
      "train.csv       68  69  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.63     0.63    6.26    0.88    \n",
      "train.csv       69  70  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.65    6.38    0.89    \n",
      "train.csv       70  71  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.85     0.64    6.49    0.87    \n",
      "train.csv       71  72  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.87    \n",
      "train.csv       72  73  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.64    6.38    0.88    \n",
      "train.csv       73  74  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.32    0.87    \n",
      "train.csv       74  75  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.64    6.42    0.9     \n",
      "train.csv       75  76  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.63    6.29    0.86    \n",
      "train.csv       76  77  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.76     0.64    6.39    0.86    \n",
      "train.csv       77  78  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.64    6.38    0.86    \n",
      "train.csv       78  79  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.63    6.32    0.86    \n",
      "train.csv       79  80  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.63    6.33    0.86    \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train.csv       80  81  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.65    6.32    0.87    \n",
      "train.csv       81  82  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.64    6.34    0.86    \n",
      "train.csv       82  83  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.65    6.32    0.86    \n",
      "train.csv       83  84  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.61     0.64    6.25    0.86    \n",
      "train.csv       84  85  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.64    6.36    0.87    \n",
      "train.csv       85  86  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.31    0.93    \n",
      "train.csv       86  87  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.64    6.29    0.87    \n",
      "train.csv       87  88  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.63    6.36    0.88    \n",
      "train.csv       88  89  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.63    6.29    0.87    \n",
      "train.csv       89  90  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.63    6.35    0.86    \n",
      "train.csv       90  91  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.66    6.45    0.86    \n",
      "train.csv       91  92  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.64    6.38    0.87    \n",
      "train.csv       92  93  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.63    6.35    0.86    \n",
      "train.csv       93  94  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.94     0.65    6.59    0.91    \n",
      "train.csv       94  95  Random  1.0   1.0   1.0   1.0   1.0   1.0   6.22     0.77    6.99    0.92    \n",
      "train.csv       95  96  Random  1.0   1.0   1.0   1.0   1.0   1.0   7.22     0.67    7.89    0.87    \n",
      "train.csv       96  97  Random  1.0   1.0   1.0   1.0   1.0   1.0   6.43     0.65    7.07    0.87    \n",
      "train.csv       97  98  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.64    6.39    0.87    \n",
      "train.csv       98  99  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.64    6.37    0.87    \n",
      "train.csv       99  100 Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.65    6.29    0.87    \n",
      "                    precision    recall  f1-score       support\n",
      "ACK Flooding         1.000000  1.000000  1.000000   6000.000000\n",
      "Ezviz                1.000000  1.000000  1.000000   3000.000000\n",
      "HTTP Flooding        1.000000  1.000000  1.000000   6000.000000\n",
      "Host Discovery       1.000000  1.000000  1.000000   6000.000000\n",
      "Netatmo Camera       1.000000  1.000000  1.000000   3000.000000\n",
      "Nugu                 1.000000  1.000000  1.000000   3000.000000\n",
      "Samsung Camera       1.000000  1.000000  1.000000   3000.000000\n",
      "Smurf                1.000000  1.000000  1.000000   9000.000000\n",
      "Snmp                 1.000000  1.000000  1.000000   3000.000000\n",
      "Ssdp                 1.000000  1.000000  1.000000   6000.000000\n",
      "TP Link Plug         1.000000  0.992000  0.995984   3000.000000\n",
      "TcpSynReflection     1.000000  1.000000  1.000000  18000.000000\n",
      "Telnet Bruteforce    1.000000  1.000000  1.000000   6000.000000\n",
      "UDP Flooding         1.000000  1.000000  1.000000   6000.000000\n",
      "WEMO Motion Sensor   1.000000  1.000000  1.000000   3000.000000\n",
      "WEMO Power Switch    1.000000  1.000000  1.000000   3000.000000\n",
      "chromecast           1.000000  1.000000  1.000000   3000.000000\n",
      "lifx                 0.992063  1.000000  0.996016   3000.000000\n",
      "accuracy             0.999742  0.999742  0.999742      0.999742\n",
      "macro avg            0.999559  0.999556  0.999556  93000.000000\n",
      "weighted avg         0.999744  0.999742  0.999742  93000.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1008 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ths = open(\"1_08_2020_merged_12.csv\", \"w\")\n",
    "ths.write(\"Dataset,T,CV,ML algorithm,Acc,b_Acc,Precision, Recall , F1-score, kappa ,tra-Time,test-Time,total-Time),Al-Time)\\n\")\n",
    "repetition=100\n",
    "\n",
    "\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "from sklearn.preprocessing import Normalizer\n",
    "print ('%-15s %-3s %-3s %-6s  %-5s %-5s %-5s %-5s %-5s %-5s %-8s %-8s%-8s%-8s'%\n",
    "           (\"Dataset\",\"T\",\"CV\",\"ML alg\",\"Acc\",\"b_Acc\",\"Prec\", \"Rec\" , \"F1\", \"kap\" ,\"tra-T\",\"test-T\",\"total\",\"al-time\"))\n",
    "\n",
    "for ii in ml_list:\n",
    "    class_based_results=pd.DataFrame()#\"\" #pd.DataFrame(0, index=np.arange((len(target_names)+3)), columns=[\"f1-score\",\"precision\",\"recall\",\"support\"])\n",
    "    cm=pd.DataFrame()\n",
    "    cv=0\n",
    "    for i in range(repetition):\n",
    "\n",
    "\n",
    "        #TRAIN\n",
    "        df = pd.read_csv(loop)\n",
    "        m_train=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_train =df[df.columns[0:-1]]\n",
    "        X_train=np.array(X_train)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_train=df[df.columns[-1]].cat.codes  \n",
    "\n",
    "        #TEST\n",
    "        df = pd.read_csv(loop2)#,header=None )\n",
    "        m_test=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_test =df[df.columns[0:-1]]\n",
    "        X_test=np.array(X_test)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_test=df[df.columns[-1]].cat.codes\n",
    "        \n",
    "\n",
    "\n",
    "        dname=loop  \n",
    "        results_y=[]\n",
    "        cv+=1\n",
    "        results_y.append(y_test)\n",
    "\n",
    "\n",
    "        precision=[]\n",
    "        recall=[]\n",
    "        f1=[]\n",
    "        accuracy=[]\n",
    "        train_time=[]\n",
    "        test_time=[]\n",
    "        total_time=[]\n",
    "        kappa=[]\n",
    "        accuracy_b=[]\n",
    "\n",
    "            #machine learning algorithm is applied in this section\n",
    "        clf = ml_list[ii]#choose algorithm from ml_list dictionary\n",
    "        second=time.time()\n",
    "        clf.fit(X_train, y_train)\n",
    "        train_time.append(float((time.time()-second)) )\n",
    "        second=time.time()\n",
    "        predict =clf.predict(X_test)\n",
    "        test_time.append(float((time.time()-second)) )\n",
    "        #normal\n",
    "        #lines,class_based_results=score(train_time,test_time,predict,y_test,class_based_results)\n",
    "        #merged\n",
    "        predict,altime=merged(m_test,y_test,predict)\n",
    "        lines,class_based_results=score(altime,train_time,test_time,predict,y_test,class_based_results)\n",
    "\n",
    "        ths.write (lines)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "        df_cm = pd.DataFrame(confusion_matrix(y_test, predict))\n",
    "        results_y.append(predict)\n",
    "        if cm.empty:\n",
    "            cm =df_cm\n",
    "        else:\n",
    "            cm = cm.add(df_cm, fill_value=0)\n",
    "\n",
    "\n",
    "print(class_based_results/(repetition)) \n",
    "graph_name=\"merged_confusion matrix\" +str(ii)+\".pdf\"     \n",
    "plt.figure(figsize = (20,14))\n",
    "sns.heatmap(cm,xticklabels=target_names, yticklabels=target_names, annot=True)\n",
    "plt.savefig(graph_name,bbox_inches='tight')#, dpi=400)\n",
    "plt.show()\n",
    "#print(cm)\n",
    "print(\"\\n\\n\\n\") \n",
    "ths.close()  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GROUP SIZE 9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def merged(mac,y_test,predict):\n",
    "    step=8\n",
    "    second=time.time()\n",
    "    mac_test=[]\n",
    "    for i in y_test.index:\n",
    "        mac_test.append(mac[i])\n",
    "\n",
    "    d_list=sorted(list(mac.unique()))\n",
    "    devices={}\n",
    "    for i in d_list:\n",
    "        devices[i]=[]    \n",
    "\n",
    "\n",
    "    new_y=[0]*len(y_test)\n",
    "\n",
    "    for i,ii in enumerate (mac_test):\n",
    "        devices[ii].append(i)\n",
    "        #devices[]\n",
    "    for i in devices:\n",
    "        \n",
    "        a = [devices[i][j:j + step] for j in range(0, len(devices[i]), step)]  \n",
    "        for ii in a:\n",
    "            step_list=[]\n",
    "            for iii in ii:\n",
    "                step_list.append(predict[iii])\n",
    "            add=most_frequent(list(step_list))\n",
    "            for iii in ii:\n",
    "                new_y[iii]=add\n",
    "                \n",
    "    return new_y,time.time()-second"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset         T   CV  ML alg  Acc   b_Acc Prec  Rec   F1    kap   tra-T    test-T  total   al-time \n",
      "train.csv       0   1   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.62     0.64    6.26    0.9     \n",
      "train.csv       1   2   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.65    6.36    0.87    \n",
      "train.csv       2   3   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.65    6.3     0.87    \n",
      "train.csv       3   4   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.64    6.44    0.86    \n",
      "train.csv       4   5   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.65    6.36    0.87    \n",
      "train.csv       5   6   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.78     0.63    6.41    0.88    \n",
      "train.csv       6   7   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.64    6.37    0.88    \n",
      "train.csv       7   8   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.64    6.29    0.88    \n",
      "train.csv       8   9   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.63    6.33    0.87    \n",
      "train.csv       9   10  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.64    6.3     0.89    \n",
      "train.csv       10  11  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.64    6.36    0.87    \n",
      "train.csv       11  12  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.35    0.87    \n",
      "train.csv       12  13  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.64    6.36    0.87    \n",
      "train.csv       13  14  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.31    0.88    \n",
      "train.csv       14  15  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.65    6.34    0.88    \n",
      "train.csv       15  16  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.64    6.34    0.87    \n",
      "train.csv       16  17  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.62     0.64    6.26    0.88    \n",
      "train.csv       17  18  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.82     0.64    6.46    0.88    \n",
      "train.csv       18  19  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.64    6.37    0.92    \n",
      "train.csv       19  20  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.35    0.88    \n",
      "train.csv       20  21  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.65    6.31    0.88    \n",
      "train.csv       21  22  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.64    6.3     0.88    \n",
      "train.csv       22  23  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.65    6.36    0.88    \n",
      "train.csv       23  24  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.63    6.34    0.92    \n",
      "train.csv       24  25  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.8      0.64    6.44    0.88    \n",
      "train.csv       25  26  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.65    6.29    0.88    \n",
      "train.csv       26  27  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.58     0.64    6.21    0.87    \n",
      "train.csv       27  28  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.63    6.35    0.89    \n",
      "train.csv       28  29  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.63    6.29    0.88    \n",
      "train.csv       29  30  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.64    6.34    0.89    \n",
      "train.csv       30  31  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.63     0.63    6.27    0.88    \n",
      "train.csv       31  32  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.64    6.28    0.88    \n",
      "train.csv       32  33  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.63     0.63    6.26    0.91    \n",
      "train.csv       33  34  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.83     0.63    6.46    0.88    \n",
      "train.csv       34  35  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.59     0.62    6.22    0.91    \n",
      "train.csv       35  36  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.64    6.3     0.88    \n",
      "train.csv       36  37  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.86    \n",
      "train.csv       37  38  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.63    6.28    0.89    \n",
      "train.csv       38  39  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.32    0.87    \n",
      "train.csv       39  40  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.64    6.38    0.87    \n",
      "train.csv       40  41  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.54     0.64    6.19    0.89    \n",
      "train.csv       41  42  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.63    6.38    0.87    \n",
      "train.csv       42  43  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.56     0.63    6.19    0.86    \n",
      "train.csv       43  44  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.63    6.31    0.87    \n",
      "train.csv       44  45  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.78     0.65    6.42    0.87    \n",
      "train.csv       45  46  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.86    \n",
      "train.csv       46  47  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.62    6.39    0.87    \n",
      "train.csv       47  48  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.68    6.36    0.87    \n",
      "train.csv       48  49  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.63    6.38    0.88    \n",
      "train.csv       49  50  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.65    6.39    0.9     \n",
      "train.csv       50  51  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.63    6.32    0.86    \n",
      "train.csv       51  52  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.62    6.34    0.87    \n",
      "train.csv       52  53  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.63    6.36    0.86    \n",
      "train.csv       53  54  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.63    6.29    0.86    \n",
      "train.csv       54  55  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.63    6.41    0.91    \n",
      "train.csv       55  56  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.64    6.38    0.87    \n",
      "train.csv       56  57  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.63    6.34    0.88    \n",
      "train.csv       57  58  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.57     0.62    6.19    0.88    \n",
      "train.csv       58  59  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.76     0.64    6.4     0.89    \n",
      "train.csv       59  60  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.63    6.26    0.87    \n",
      "train.csv       60  61  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.63     0.65    6.28    0.87    \n",
      "train.csv       61  62  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.64    6.39    0.87    \n",
      "train.csv       62  63  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.64    6.34    0.89    \n",
      "train.csv       63  64  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.35    0.88    \n",
      "train.csv       64  65  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.85     0.66    6.5     0.88    \n",
      "train.csv       65  66  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.34    0.87    \n",
      "train.csv       66  67  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.64    6.38    0.88    \n",
      "train.csv       67  68  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.64    6.36    0.89    \n",
      "train.csv       68  69  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.82     0.63    6.45    0.88    \n",
      "train.csv       69  70  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.65    6.45    0.87    \n",
      "train.csv       70  71  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.63    6.34    0.87    \n",
      "train.csv       71  72  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.6      0.64    6.25    0.92    \n",
      "train.csv       72  73  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.33    0.89    \n",
      "train.csv       73  74  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.63    6.36    0.88    \n",
      "train.csv       74  75  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.63    6.32    0.88    \n",
      "train.csv       75  76  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.78     0.64    6.42    0.87    \n",
      "train.csv       76  77  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.81     0.64    6.45    0.88    \n",
      "train.csv       77  78  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.88    \n",
      "train.csv       78  79  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.83     0.64    6.47    0.88    \n",
      "train.csv       79  80  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.63    6.4     0.88    \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train.csv       80  81  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.63     0.64    6.27    0.87    \n",
      "train.csv       81  82  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.64    6.35    0.88    \n",
      "train.csv       82  83  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.3     0.88    \n",
      "train.csv       83  84  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.32    0.87    \n",
      "train.csv       84  85  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.65    6.41    0.87    \n",
      "train.csv       85  86  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.35    0.89    \n",
      "train.csv       86  87  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.81     0.64    6.46    0.87    \n",
      "train.csv       87  88  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.65    6.3     0.87    \n",
      "train.csv       88  89  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.65    6.31    0.87    \n",
      "train.csv       89  90  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.8      0.64    6.44    0.9     \n",
      "train.csv       90  91  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.76     0.64    6.4     0.88    \n",
      "train.csv       91  92  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.76     0.64    6.4     0.88    \n",
      "train.csv       92  93  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.8      0.64    6.44    0.89    \n",
      "train.csv       93  94  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.63     0.65    6.28    0.86    \n",
      "train.csv       94  95  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.63    6.32    0.86    \n",
      "train.csv       95  96  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.62     0.64    6.26    0.87    \n",
      "train.csv       96  97  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.32    0.87    \n",
      "train.csv       97  98  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.67    6.34    0.92    \n",
      "train.csv       98  99  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.32    0.88    \n",
      "train.csv       99  100 Random  1.0   1.0   1.0   1.0   1.0   1.0   5.81     0.64    6.45    0.88    \n",
      "                    precision    recall  f1-score       support\n",
      "ACK Flooding         1.000000  1.000000  1.000000   6000.000000\n",
      "Ezviz                1.000000  1.000000  1.000000   3000.000000\n",
      "HTTP Flooding        1.000000  1.000000  1.000000   6000.000000\n",
      "Host Discovery       0.993377  1.000000  0.996678   6000.000000\n",
      "Netatmo Camera       1.000000  0.992000  0.995984   3000.000000\n",
      "Nugu                 1.000000  1.000000  1.000000   3000.000000\n",
      "Samsung Camera       1.000000  1.000000  1.000000   3000.000000\n",
      "Smurf                1.000000  1.000000  1.000000   9000.000000\n",
      "Snmp                 1.000000  1.000000  1.000000   3000.000000\n",
      "Ssdp                 1.000000  1.000000  1.000000   6000.000000\n",
      "TP Link Plug         1.000000  0.977840  0.988795   3000.000000\n",
      "TcpSynReflection     1.000000  1.000000  1.000000  18000.000000\n",
      "Telnet Bruteforce    1.000000  1.000000  1.000000   6000.000000\n",
      "UDP Flooding         1.000000  1.000000  1.000000   6000.000000\n",
      "WEMO Motion Sensor   1.000000  1.000000  1.000000   3000.000000\n",
      "WEMO Power Switch    1.000000  0.994667  0.997326   3000.000000\n",
      "chromecast           1.000000  1.000000  1.000000   3000.000000\n",
      "lifx                 0.978322  1.000000  0.989042   3000.000000\n",
      "accuracy             0.998855  0.998855  0.998855      0.998855\n",
      "macro avg            0.998428  0.998028  0.998213  93000.000000\n",
      "weighted avg         0.998873  0.998855  0.998855  93000.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1008 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ths = open(\"1_08_2020_merged_9.csv\", \"w\")\n",
    "ths.write(\"Dataset,T,CV,ML algorithm,Acc,b_Acc,Precision, Recall , F1-score, kappa ,tra-Time,test-Time,total-Time),Al-Time)\\n\")\n",
    "repetition=100\n",
    "\n",
    "\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "from sklearn.preprocessing import Normalizer\n",
    "print ('%-15s %-3s %-3s %-6s  %-5s %-5s %-5s %-5s %-5s %-5s %-8s %-8s%-8s%-8s'%\n",
    "           (\"Dataset\",\"T\",\"CV\",\"ML alg\",\"Acc\",\"b_Acc\",\"Prec\", \"Rec\" , \"F1\", \"kap\" ,\"tra-T\",\"test-T\",\"total\",\"al-time\"))\n",
    "\n",
    "for ii in ml_list:\n",
    "    class_based_results=pd.DataFrame()#\"\" #pd.DataFrame(0, index=np.arange((len(target_names)+3)), columns=[\"f1-score\",\"precision\",\"recall\",\"support\"])\n",
    "    cm=pd.DataFrame()\n",
    "    cv=0\n",
    "    for i in range(repetition):\n",
    "\n",
    "\n",
    "        #TRAIN\n",
    "        df = pd.read_csv(loop)\n",
    "        m_train=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_train =df[df.columns[0:-1]]\n",
    "        X_train=np.array(X_train)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_train=df[df.columns[-1]].cat.codes  \n",
    "\n",
    "        #TEST\n",
    "        df = pd.read_csv(loop2)#,header=None )\n",
    "        m_test=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_test =df[df.columns[0:-1]]\n",
    "        X_test=np.array(X_test)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_test=df[df.columns[-1]].cat.codes\n",
    "        \n",
    "\n",
    "\n",
    "        dname=loop  \n",
    "        results_y=[]\n",
    "        cv+=1\n",
    "        results_y.append(y_test)\n",
    "\n",
    "\n",
    "        precision=[]\n",
    "        recall=[]\n",
    "        f1=[]\n",
    "        accuracy=[]\n",
    "        train_time=[]\n",
    "        test_time=[]\n",
    "        total_time=[]\n",
    "        kappa=[]\n",
    "        accuracy_b=[]\n",
    "\n",
    "            #machine learning algorithm is applied in this section\n",
    "        clf = ml_list[ii]#choose algorithm from ml_list dictionary\n",
    "        second=time.time()\n",
    "        clf.fit(X_train, y_train)\n",
    "        train_time.append(float((time.time()-second)) )\n",
    "        second=time.time()\n",
    "        predict =clf.predict(X_test)\n",
    "        test_time.append(float((time.time()-second)) )\n",
    "        #normal\n",
    "        #lines,class_based_results=score(train_time,test_time,predict,y_test,class_based_results)\n",
    "        #merged\n",
    "        predict,altime=merged(m_test,y_test,predict)\n",
    "        lines,class_based_results=score(altime,train_time,test_time,predict,y_test,class_based_results)\n",
    "\n",
    "        ths.write (lines)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "        df_cm = pd.DataFrame(confusion_matrix(y_test, predict))\n",
    "        results_y.append(predict)\n",
    "        if cm.empty:\n",
    "            cm =df_cm\n",
    "        else:\n",
    "            cm = cm.add(df_cm, fill_value=0)\n",
    "\n",
    "\n",
    "print(class_based_results/(repetition)) \n",
    "graph_name=\"merged_confusion matrix\" +str(ii)+\".pdf\"     \n",
    "plt.figure(figsize = (20,14))\n",
    "sns.heatmap(cm,xticklabels=target_names, yticklabels=target_names, annot=True)\n",
    "plt.savefig(graph_name,bbox_inches='tight')#, dpi=400)\n",
    "plt.show()\n",
    "#print(cm)\n",
    "print(\"\\n\\n\\n\") \n",
    "ths.close()  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GROUP SIZE 6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def merged(mac,y_test,predict):\n",
    "    step=6\n",
    "    second=time.time()\n",
    "    mac_test=[]\n",
    "    for i in y_test.index:\n",
    "        mac_test.append(mac[i])\n",
    "\n",
    "    d_list=sorted(list(mac.unique()))\n",
    "    devices={}\n",
    "    for i in d_list:\n",
    "        devices[i]=[]    \n",
    "\n",
    "\n",
    "    new_y=[0]*len(y_test)\n",
    "\n",
    "    for i,ii in enumerate (mac_test):\n",
    "        devices[ii].append(i)\n",
    "        #devices[]\n",
    "    for i in devices:\n",
    "        \n",
    "        a = [devices[i][j:j + step] for j in range(0, len(devices[i]), step)]  \n",
    "        for ii in a:\n",
    "            step_list=[]\n",
    "            for iii in ii:\n",
    "                step_list.append(predict[iii])\n",
    "            add=most_frequent(list(step_list))\n",
    "            for iii in ii:\n",
    "                new_y[iii]=add\n",
    "                \n",
    "    return new_y,time.time()-second"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset         T   CV  ML alg  Acc   b_Acc Prec  Rec   F1    kap   tra-T    test-T  total   al-time \n",
      "train.csv       0   1   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.34    0.89    \n",
      "train.csv       1   2   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.85     0.64    6.49    0.87    \n",
      "train.csv       2   3   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.64    6.43    0.87    \n",
      "train.csv       3   4   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.6      0.65    6.25    0.88    \n",
      "train.csv       4   5   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.66    6.35    0.87    \n",
      "train.csv       5   6   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.33    0.89    \n",
      "train.csv       6   7   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.62     0.67    6.29    0.88    \n",
      "train.csv       7   8   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.63     0.64    6.27    0.89    \n",
      "train.csv       8   9   Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.64    6.34    0.91    \n",
      "train.csv       9   10  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.62    6.32    0.9     \n",
      "train.csv       10  11  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.9     \n",
      "train.csv       11  12  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.65    6.4     0.88    \n",
      "train.csv       12  13  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.64    6.43    0.89    \n",
      "train.csv       13  14  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.72     0.64    6.36    0.9     \n",
      "train.csv       14  15  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.31    0.88    \n",
      "train.csv       15  16  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.78     0.64    6.42    0.9     \n",
      "train.csv       16  17  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.35    0.88    \n",
      "train.csv       17  18  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.65    6.41    0.92    \n",
      "train.csv       18  19  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.76     0.64    6.4     0.92    \n",
      "train.csv       19  20  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.63    6.33    0.89    \n",
      "train.csv       20  21  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.64    6.41    0.89    \n",
      "train.csv       21  22  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.34    0.91    \n",
      "train.csv       22  23  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.8      0.64    6.44    0.88    \n",
      "train.csv       23  24  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.98     0.65    6.63    0.9     \n",
      "train.csv       24  25  Random  1.0   1.0   1.0   1.0   1.0   1.0   6.31     0.68    6.99    1.03    \n",
      "train.csv       25  26  Random  1.0   1.0   1.0   1.0   1.0   1.0   6.49     0.71    7.2     0.88    \n",
      "train.csv       26  27  Random  1.0   1.0   1.0   1.0   1.0   1.0   6.26     0.8     7.06    1.02    \n",
      "train.csv       27  28  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.94     0.7     6.64    0.89    \n",
      "train.csv       28  29  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.8      0.64    6.44    0.92    \n",
      "train.csv       29  30  Random  1.0   1.0   1.0   1.0   1.0   1.0   6.4      0.87    7.27    0.92    \n",
      "train.csv       30  31  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.8      0.64    6.45    0.88    \n",
      "train.csv       31  32  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.63    6.27    0.87    \n",
      "train.csv       32  33  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.63    6.37    0.88    \n",
      "train.csv       33  34  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.65    6.34    0.89    \n",
      "train.csv       34  35  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.63    6.36    0.89    \n",
      "train.csv       35  36  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.65    6.42    0.88    \n",
      "train.csv       36  37  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.82     0.64    6.47    0.87    \n",
      "train.csv       37  38  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.64    6.36    0.89    \n",
      "train.csv       38  39  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.83     0.64    6.46    0.94    \n",
      "train.csv       39  40  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.31    0.88    \n",
      "train.csv       40  41  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.63    6.37    0.88    \n",
      "train.csv       41  42  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.62     0.64    6.26    0.88    \n",
      "train.csv       42  43  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.58     0.64    6.22    0.88    \n",
      "train.csv       43  44  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.63    6.38    0.89    \n",
      "train.csv       44  45  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.34    0.87    \n",
      "train.csv       45  46  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.62    6.3     0.88    \n",
      "train.csv       46  47  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.87     0.65    6.51    0.88    \n",
      "train.csv       47  48  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.64    6.36    0.88    \n",
      "train.csv       48  49  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.81     0.64    6.45    0.92    \n",
      "train.csv       49  50  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.31    0.88    \n",
      "train.csv       50  51  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.81     0.64    6.45    0.9     \n",
      "train.csv       51  52  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.65    6.43    0.88    \n",
      "train.csv       52  53  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.64    6.28    0.88    \n",
      "train.csv       53  54  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.78     0.64    6.42    0.87    \n",
      "train.csv       54  55  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.64    6.31    0.88    \n",
      "train.csv       55  56  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.3     0.88    \n",
      "train.csv       56  57  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.64    6.35    0.88    \n",
      "train.csv       57  58  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.63    6.42    0.91    \n",
      "train.csv       58  59  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.61     0.62    6.23    0.88    \n",
      "train.csv       59  60  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.78     0.64    6.42    0.88    \n",
      "train.csv       60  61  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.61     0.63    6.24    0.88    \n",
      "train.csv       61  62  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.7      0.63    6.33    0.88    \n",
      "train.csv       62  63  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.63    6.31    0.87    \n",
      "train.csv       63  64  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.64    6.3     0.88    \n",
      "train.csv       64  65  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.63    6.32    0.89    \n",
      "train.csv       65  66  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.85     0.64    6.49    0.89    \n",
      "train.csv       66  67  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.83     0.64    6.47    0.88    \n",
      "train.csv       67  68  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.65    6.33    0.91    \n",
      "train.csv       68  69  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.63    6.27    0.88    \n",
      "train.csv       69  70  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.64    6.28    0.87    \n",
      "train.csv       70  71  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.33    0.87    \n",
      "train.csv       71  72  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.65    6.39    0.88    \n",
      "train.csv       72  73  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.86     0.65    6.51    0.87    \n",
      "train.csv       73  74  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.63     0.64    6.27    0.88    \n",
      "train.csv       74  75  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.66    6.41    0.87    \n",
      "train.csv       75  76  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.64    6.28    0.86    \n",
      "train.csv       76  77  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.64    6.3     0.87    \n",
      "train.csv       77  78  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.74     0.65    6.38    0.91    \n",
      "train.csv       78  79  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.32    0.87    \n",
      "train.csv       79  80  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.64     0.64    6.28    0.88    \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train.csv       80  81  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.65    6.39    0.93    \n",
      "train.csv       81  82  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.76     0.64    6.39    0.87    \n",
      "train.csv       82  83  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.8      0.64    6.44    0.87    \n",
      "train.csv       83  84  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.73     0.64    6.37    0.88    \n",
      "train.csv       84  85  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.63    6.3     0.89    \n",
      "train.csv       85  86  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.63    6.28    0.88    \n",
      "train.csv       86  87  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.82     0.64    6.45    0.88    \n",
      "train.csv       87  88  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.67     0.64    6.31    0.92    \n",
      "train.csv       88  89  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.71     0.63    6.34    0.86    \n",
      "train.csv       89  90  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.68     0.64    6.32    0.88    \n",
      "train.csv       90  91  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.69     0.64    6.32    0.87    \n",
      "train.csv       91  92  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.77     0.64    6.41    0.88    \n",
      "train.csv       92  93  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.64    6.44    0.87    \n",
      "train.csv       93  94  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.66     0.64    6.3     0.88    \n",
      "train.csv       94  95  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.78     0.63    6.41    0.88    \n",
      "train.csv       95  96  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.8      0.63    6.43    0.88    \n",
      "train.csv       96  97  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.75     0.65    6.4     0.88    \n",
      "train.csv       97  98  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.79     0.64    6.44    0.92    \n",
      "train.csv       98  99  Random  1.0   1.0   1.0   1.0   1.0   1.0   5.65     0.64    6.28    0.9     \n",
      "train.csv       99  100 Random  1.0   1.0   1.0   1.0   1.0   1.0   5.54     0.64    6.17    0.89    \n",
      "                    precision    recall  f1-score       support\n",
      "ACK Flooding         1.000000  1.000000  1.000000   6000.000000\n",
      "Ezviz                1.000000  1.000000  1.000000   3000.000000\n",
      "HTTP Flooding        1.000000  1.000000  1.000000   6000.000000\n",
      "Host Discovery       0.987167  1.000000  0.993542   6000.000000\n",
      "Netatmo Camera       1.000000  0.990000  0.994975   3000.000000\n",
      "Nugu                 1.000000  1.000000  1.000000   3000.000000\n",
      "Samsung Camera       1.000000  1.000000  1.000000   3000.000000\n",
      "Smurf                1.000000  1.000000  1.000000   9000.000000\n",
      "Snmp                 1.000000  1.000000  1.000000   3000.000000\n",
      "Ssdp                 1.000000  1.000000  1.000000   6000.000000\n",
      "TP Link Plug         1.000000  0.952200  0.975514   3000.000000\n",
      "TcpSynReflection     1.000000  1.000000  1.000000  18000.000000\n",
      "Telnet Bruteforce    1.000000  1.000000  1.000000   6000.000000\n",
      "UDP Flooding         1.000000  1.000000  1.000000   6000.000000\n",
      "WEMO Motion Sensor   1.000000  1.000000  1.000000   3000.000000\n",
      "WEMO Power Switch    1.000000  0.986000  0.992951   3000.000000\n",
      "chromecast           1.000000  1.000000  1.000000   3000.000000\n",
      "lifx                 0.954295  0.998000  0.975658   3000.000000\n",
      "accuracy             0.997619  0.997619  0.997619      0.997619\n",
      "macro avg            0.996748  0.995900  0.996258  93000.000000\n",
      "weighted avg         0.997698  0.997619  0.997619  93000.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1008 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ths = open(\"1_08_2020_merged_6.csv\", \"w\")\n",
    "ths.write(\"Dataset,T,CV,ML algorithm,Acc,b_Acc,Precision, Recall , F1-score, kappa ,tra-Time,test-Time,total-Time),Al-Time)\\n\")\n",
    "repetition=100\n",
    "\n",
    "\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "from sklearn.preprocessing import Normalizer\n",
    "print ('%-15s %-3s %-3s %-6s  %-5s %-5s %-5s %-5s %-5s %-5s %-8s %-8s%-8s%-8s'%\n",
    "           (\"Dataset\",\"T\",\"CV\",\"ML alg\",\"Acc\",\"b_Acc\",\"Prec\", \"Rec\" , \"F1\", \"kap\" ,\"tra-T\",\"test-T\",\"total\",\"al-time\"))\n",
    "\n",
    "for ii in ml_list:\n",
    "    class_based_results=pd.DataFrame()#\"\" #pd.DataFrame(0, index=np.arange((len(target_names)+3)), columns=[\"f1-score\",\"precision\",\"recall\",\"support\"])\n",
    "    cm=pd.DataFrame()\n",
    "    cv=0\n",
    "    for i in range(repetition):\n",
    "\n",
    "\n",
    "        #TRAIN\n",
    "        df = pd.read_csv(loop)\n",
    "        m_train=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_train =df[df.columns[0:-1]]\n",
    "        X_train=np.array(X_train)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_train=df[df.columns[-1]].cat.codes  \n",
    "\n",
    "        #TEST\n",
    "        df = pd.read_csv(loop2)#,header=None )\n",
    "        m_test=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_test =df[df.columns[0:-1]]\n",
    "        X_test=np.array(X_test)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_test=df[df.columns[-1]].cat.codes\n",
    "        \n",
    "\n",
    "\n",
    "        dname=loop  \n",
    "        results_y=[]\n",
    "        cv+=1\n",
    "        results_y.append(y_test)\n",
    "\n",
    "\n",
    "        precision=[]\n",
    "        recall=[]\n",
    "        f1=[]\n",
    "        accuracy=[]\n",
    "        train_time=[]\n",
    "        test_time=[]\n",
    "        total_time=[]\n",
    "        kappa=[]\n",
    "        accuracy_b=[]\n",
    "\n",
    "            #machine learning algorithm is applied in this section\n",
    "        clf = ml_list[ii]#choose algorithm from ml_list dictionary\n",
    "        second=time.time()\n",
    "        clf.fit(X_train, y_train)\n",
    "        train_time.append(float((time.time()-second)) )\n",
    "        second=time.time()\n",
    "        predict =clf.predict(X_test)\n",
    "        test_time.append(float((time.time()-second)) )\n",
    "        #normal\n",
    "        #lines,class_based_results=score(train_time,test_time,predict,y_test,class_based_results)\n",
    "        #merged\n",
    "        predict,altime=merged(m_test,y_test,predict)\n",
    "        lines,class_based_results=score(altime,train_time,test_time,predict,y_test,class_based_results)\n",
    "\n",
    "        ths.write (lines)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "        df_cm = pd.DataFrame(confusion_matrix(y_test, predict))\n",
    "        results_y.append(predict)\n",
    "        if cm.empty:\n",
    "            cm =df_cm\n",
    "        else:\n",
    "            cm = cm.add(df_cm, fill_value=0)\n",
    "\n",
    "\n",
    "print(class_based_results/(repetition)) \n",
    "graph_name=\"merged_confusion matrix\" +str(ii)+\".pdf\"     \n",
    "plt.figure(figsize = (20,14))\n",
    "sns.heatmap(cm,xticklabels=target_names, yticklabels=target_names, annot=True)\n",
    "plt.savefig(graph_name,bbox_inches='tight')#, dpi=400)\n",
    "plt.show()\n",
    "#print(cm)\n",
    "print(\"\\n\\n\\n\") \n",
    "ths.close()  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GROUP SIZE 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def merged(mac,y_test,predict):\n",
    "    step=3\n",
    "    second=time.time()\n",
    "    mac_test=[]\n",
    "    for i in y_test.index:\n",
    "        mac_test.append(mac[i])\n",
    "\n",
    "    d_list=sorted(list(mac.unique()))\n",
    "    devices={}\n",
    "    for i in d_list:\n",
    "        devices[i]=[]    \n",
    "\n",
    "\n",
    "    new_y=[0]*len(y_test)\n",
    "\n",
    "    for i,ii in enumerate (mac_test):\n",
    "        devices[ii].append(i)\n",
    "        #devices[]\n",
    "    for i in devices:\n",
    "        \n",
    "        a = [devices[i][j:j + step] for j in range(0, len(devices[i]), step)]  \n",
    "        for ii in a:\n",
    "            step_list=[]\n",
    "            for iii in ii:\n",
    "                step_list.append(predict[iii])\n",
    "            add=most_frequent(list(step_list))\n",
    "            for iii in ii:\n",
    "                new_y[iii]=add\n",
    "                \n",
    "    return new_y,time.time()-second"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset         T   CV  ML alg  Acc   b_Acc Prec  Rec   F1    kap   tra-T    test-T  total   al-time \n",
      "train.csv       0   1   Random  0.99  0.99  0.99  0.99  0.99  0.99  5.83     0.64    6.47    0.96    \n",
      "train.csv       1   2   Random  0.99  0.99  0.99  0.99  0.99  0.99  5.76     0.64    6.41    0.9     \n",
      "train.csv       2   3   Random  0.99  0.99  0.99  0.99  0.99  0.99  5.79     0.64    6.43    0.89    \n",
      "train.csv       3   4   Random  0.99  0.99  0.99  0.99  0.99  0.99  5.73     0.64    6.36    0.93    \n",
      "train.csv       4   5   Random  0.99  0.99  0.99  0.99  0.99  0.99  5.71     0.64    6.35    0.89    \n",
      "train.csv       5   6   Random  0.99  0.99  0.99  0.99  0.99  0.99  5.71     0.64    6.34    0.89    \n",
      "train.csv       6   7   Random  0.99  0.99  0.99  0.99  0.99  0.99  5.78     0.64    6.42    0.9     \n",
      "train.csv       7   8   Random  0.99  0.99  0.99  0.99  0.99  0.99  5.74     0.65    6.39    0.93    \n",
      "train.csv       8   9   Random  0.99  0.99  0.99  0.99  0.99  0.99  5.77     0.64    6.41    0.89    \n",
      "train.csv       9   10  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.69     0.64    6.33    0.91    \n",
      "train.csv       10  11  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.72     0.64    6.36    0.92    \n",
      "train.csv       11  12  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.98     0.65    6.63    0.88    \n",
      "train.csv       12  13  Random  0.99  0.99  0.99  0.99  0.99  0.99  6.53     0.66    7.19    0.91    \n",
      "train.csv       13  14  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.82     0.64    6.47    0.9     \n",
      "train.csv       14  15  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.77     0.64    6.4     0.94    \n",
      "train.csv       15  16  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.66     0.64    6.3     0.89    \n",
      "train.csv       16  17  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.7      0.62    6.32    0.89    \n",
      "train.csv       17  18  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.75     0.63    6.38    0.94    \n",
      "train.csv       18  19  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.8      0.64    6.45    0.89    \n",
      "train.csv       19  20  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.6      0.64    6.24    0.9     \n",
      "train.csv       20  21  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.71     0.63    6.34    0.89    \n",
      "train.csv       21  22  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.75     0.64    6.39    0.93    \n",
      "train.csv       22  23  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.68     0.64    6.32    0.89    \n",
      "train.csv       23  24  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.72     0.64    6.36    0.88    \n",
      "train.csv       24  25  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.77     0.64    6.42    0.92    \n",
      "train.csv       25  26  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.8      0.64    6.45    0.9     \n",
      "train.csv       26  27  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.69     0.63    6.32    0.88    \n",
      "train.csv       27  28  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.63     0.64    6.27    0.92    \n",
      "train.csv       28  29  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.7      0.64    6.34    0.88    \n",
      "train.csv       29  30  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.74     0.64    6.39    0.89    \n",
      "train.csv       30  31  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.64     0.64    6.28    0.89    \n",
      "train.csv       31  32  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.68     0.64    6.32    0.94    \n",
      "train.csv       32  33  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.73     0.63    6.36    0.89    \n",
      "train.csv       33  34  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.66     0.63    6.3     0.96    \n",
      "train.csv       34  35  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.68     0.64    6.32    0.93    \n",
      "train.csv       35  36  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.74     0.63    6.36    0.88    \n",
      "train.csv       36  37  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.71     0.64    6.35    0.9     \n",
      "train.csv       37  38  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.74     0.63    6.37    0.87    \n",
      "train.csv       38  39  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.73     0.64    6.37    0.92    \n",
      "train.csv       39  40  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.66     0.64    6.29    0.89    \n",
      "train.csv       40  41  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.82     0.64    6.46    0.89    \n",
      "train.csv       41  42  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.85     0.64    6.49    0.94    \n",
      "train.csv       42  43  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.72     0.64    6.35    0.89    \n",
      "train.csv       43  44  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.68     0.63    6.31    0.9     \n",
      "train.csv       44  45  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.68     0.64    6.32    0.92    \n",
      "train.csv       45  46  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.7      0.62    6.32    0.9     \n",
      "train.csv       46  47  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.63     0.65    6.28    0.88    \n",
      "train.csv       47  48  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.71     0.64    6.34    0.89    \n",
      "train.csv       48  49  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.69     0.65    6.35    0.99    \n",
      "train.csv       49  50  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.69     0.64    6.33    0.9     \n",
      "train.csv       50  51  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.79     0.64    6.44    0.9     \n",
      "train.csv       51  52  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.73     0.64    6.38    0.93    \n",
      "train.csv       52  53  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.74     0.63    6.37    0.87    \n",
      "train.csv       53  54  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.8      0.65    6.45    0.88    \n",
      "train.csv       54  55  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.64     0.64    6.27    0.91    \n",
      "train.csv       55  56  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.68     0.64    6.32    0.93    \n",
      "train.csv       56  57  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.66     0.64    6.3     0.89    \n",
      "train.csv       57  58  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.67     0.64    6.32    0.87    \n",
      "train.csv       58  59  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.75     0.65    6.4     0.92    \n",
      "train.csv       59  60  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.64     0.64    6.28    0.92    \n",
      "train.csv       60  61  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.76     0.64    6.4     0.89    \n",
      "train.csv       61  62  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.71     0.64    6.35    0.94    \n",
      "train.csv       62  63  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.77     0.64    6.41    0.89    \n",
      "train.csv       63  64  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.63     0.64    6.27    0.88    \n",
      "train.csv       64  65  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.77     0.65    6.42    0.9     \n",
      "train.csv       65  66  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.62     0.64    6.26    0.92    \n",
      "train.csv       66  67  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.71     0.63    6.35    0.89    \n",
      "train.csv       67  68  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.65     0.64    6.28    0.9     \n",
      "train.csv       68  69  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.77     0.67    6.44    0.92    \n",
      "train.csv       69  70  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.73     0.64    6.37    0.89    \n",
      "train.csv       70  71  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.75     0.64    6.39    0.9     \n",
      "train.csv       71  72  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.73     0.64    6.37    0.92    \n",
      "train.csv       72  73  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.67     0.63    6.3     0.92    \n",
      "train.csv       73  74  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.72     0.64    6.35    0.88    \n",
      "train.csv       74  75  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.68     0.65    6.33    0.89    \n",
      "train.csv       75  76  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.69     0.64    6.33    0.92    \n",
      "train.csv       76  77  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.74     0.64    6.38    0.92    \n",
      "train.csv       77  78  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.66     0.64    6.3     0.94    \n",
      "train.csv       78  79  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.77     0.64    6.41    0.92    \n",
      "train.csv       79  80  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.83     0.64    6.47    0.9     \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train.csv       80  81  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.69     0.63    6.32    0.89    \n",
      "train.csv       81  82  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.69     0.63    6.32    0.89    \n",
      "train.csv       82  83  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.62     0.64    6.26    0.93    \n",
      "train.csv       83  84  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.73     0.64    6.37    0.91    \n",
      "train.csv       84  85  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.72     0.64    6.36    0.91    \n",
      "train.csv       85  86  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.64     0.64    6.28    0.94    \n",
      "train.csv       86  87  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.72     0.65    6.36    0.89    \n",
      "train.csv       87  88  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.67     0.64    6.32    0.9     \n",
      "train.csv       88  89  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.66     0.64    6.29    0.89    \n",
      "train.csv       89  90  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.78     0.64    6.42    0.93    \n",
      "train.csv       90  91  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.7      0.64    6.35    0.88    \n",
      "train.csv       91  92  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.67     0.64    6.31    0.88    \n",
      "train.csv       92  93  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.65     0.64    6.29    0.93    \n",
      "train.csv       93  94  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.74     0.64    6.38    0.88    \n",
      "train.csv       94  95  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.65     0.65    6.3     0.88    \n",
      "train.csv       95  96  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.7      0.64    6.34    0.89    \n",
      "train.csv       96  97  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.83     0.64    6.47    0.91    \n",
      "train.csv       97  98  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.71     0.64    6.35    0.95    \n",
      "train.csv       98  99  Random  0.99  0.99  0.99  0.99  0.99  0.99  5.68     0.64    6.32    0.89    \n",
      "train.csv       99  100 Random  0.99  0.99  0.99  0.99  0.99  0.99  5.69     0.64    6.33    0.96    \n",
      "                    precision    recall  f1-score       support\n",
      "ACK Flooding         0.999650  1.000000  0.999825   6000.000000\n",
      "Ezviz                0.999001  1.000000  0.999500   3000.000000\n",
      "HTTP Flooding        1.000000  1.000000  1.000000   6000.000000\n",
      "Host Discovery       0.956014  1.000000  0.977512   6000.000000\n",
      "Netatmo Camera       1.000000  0.977000  0.988366   3000.000000\n",
      "Nugu                 1.000000  1.000000  1.000000   3000.000000\n",
      "Samsung Camera       0.998977  0.976430  0.987575   3000.000000\n",
      "Smurf                1.000000  1.000000  1.000000   9000.000000\n",
      "Snmp                 1.000000  1.000000  1.000000   3000.000000\n",
      "Ssdp                 1.000000  1.000000  1.000000   6000.000000\n",
      "TP Link Plug         0.996316  0.880910  0.935061   3000.000000\n",
      "TcpSynReflection     0.999833  1.000000  0.999917  18000.000000\n",
      "Telnet Bruteforce    1.000000  1.000000  1.000000   6000.000000\n",
      "UDP Flooding         1.000000  1.000000  1.000000   6000.000000\n",
      "WEMO Motion Sensor   1.000000  0.996000  0.997996   3000.000000\n",
      "WEMO Power Switch    1.000000  0.965280  0.982333   3000.000000\n",
      "chromecast           0.994773  0.999000  0.996882   3000.000000\n",
      "lifx                 0.890888  0.980990  0.933768   3000.000000\n",
      "accuracy             0.992762  0.992762  0.992762      0.992762\n",
      "macro avg            0.990858  0.987534  0.988819  93000.000000\n",
      "weighted avg         0.993235  0.992762  0.992763  93000.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1008 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ths = open(\"1_08_2020_merged_3.csv\", \"w\")\n",
    "ths.write(\"Dataset,T,CV,ML algorithm,Acc,b_Acc,Precision, Recall , F1-score, kappa ,tra-Time,test-Time,total-Time),Al-Time)\\n\")\n",
    "repetition=100\n",
    "\n",
    "\n",
    "from sklearn.metrics import balanced_accuracy_score\n",
    "from sklearn.preprocessing import Normalizer\n",
    "print ('%-15s %-3s %-3s %-6s  %-5s %-5s %-5s %-5s %-5s %-5s %-8s %-8s%-8s%-8s'%\n",
    "           (\"Dataset\",\"T\",\"CV\",\"ML alg\",\"Acc\",\"b_Acc\",\"Prec\", \"Rec\" , \"F1\", \"kap\" ,\"tra-T\",\"test-T\",\"total\",\"al-time\"))\n",
    "\n",
    "for ii in ml_list:\n",
    "    class_based_results=pd.DataFrame()#\"\" #pd.DataFrame(0, index=np.arange((len(target_names)+3)), columns=[\"f1-score\",\"precision\",\"recall\",\"support\"])\n",
    "    cm=pd.DataFrame()\n",
    "    cv=0\n",
    "    for i in range(repetition):\n",
    "\n",
    "\n",
    "        #TRAIN\n",
    "        df = pd.read_csv(loop)\n",
    "        m_train=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_train =df[df.columns[0:-1]]\n",
    "        X_train=np.array(X_train)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_train=df[df.columns[-1]].cat.codes  \n",
    "\n",
    "        #TEST\n",
    "        df = pd.read_csv(loop2)#,header=None )\n",
    "        m_test=df[\"Mac\"]\n",
    "        del df[\"Mac\"]\n",
    "        X_test =df[df.columns[0:-1]]\n",
    "        X_test=np.array(X_test)\n",
    "        df[df.columns[-1]] = df[df.columns[-1]].astype('category')\n",
    "        y_test=df[df.columns[-1]].cat.codes\n",
    "        \n",
    "\n",
    "\n",
    "        dname=loop  \n",
    "        results_y=[]\n",
    "        cv+=1\n",
    "        results_y.append(y_test)\n",
    "\n",
    "\n",
    "        precision=[]\n",
    "        recall=[]\n",
    "        f1=[]\n",
    "        accuracy=[]\n",
    "        train_time=[]\n",
    "        test_time=[]\n",
    "        total_time=[]\n",
    "        kappa=[]\n",
    "        accuracy_b=[]\n",
    "\n",
    "            #machine learning algorithm is applied in this section\n",
    "        clf = ml_list[ii]#choose algorithm from ml_list dictionary\n",
    "        second=time.time()\n",
    "        clf.fit(X_train, y_train)\n",
    "        train_time.append(float((time.time()-second)) )\n",
    "        second=time.time()\n",
    "        predict =clf.predict(X_test)\n",
    "        test_time.append(float((time.time()-second)) )\n",
    "        #normal\n",
    "        #lines,class_based_results=score(train_time,test_time,predict,y_test,class_based_results)\n",
    "        #merged\n",
    "        predict,altime=merged(m_test,y_test,predict)\n",
    "        lines,class_based_results=score(altime,train_time,test_time,predict,y_test,class_based_results)\n",
    "\n",
    "        ths.write (lines)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "        df_cm = pd.DataFrame(confusion_matrix(y_test, predict))\n",
    "        results_y.append(predict)\n",
    "        if cm.empty:\n",
    "            cm =df_cm\n",
    "        else:\n",
    "            cm = cm.add(df_cm, fill_value=0)\n",
    "\n",
    "\n",
    "print(class_based_results/(repetition)) \n",
    "graph_name=\"merged_confusion matrix\" +str(ii)+\".pdf\"     \n",
    "plt.figure(figsize = (20,14))\n",
    "sns.heatmap(cm,xticklabels=target_names, yticklabels=target_names, annot=True)\n",
    "plt.savefig(graph_name,bbox_inches='tight')#, dpi=400)\n",
    "plt.show()\n",
    "#print(cm)\n",
    "print(\"\\n\\n\\n\") \n",
    "ths.close()  "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}