SDAE_Code_v3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1.Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import time\n",
    "from tensorflow.keras import regularizers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "XL=pd.read_csv('9_510_L_resample_95_5.csv')\n",
    "XU_train=pd.read_csv('9_510_U_resample_95_5.csv')# No resampling done in the unlabeled set, It just denotes the pair of L and U\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Source.IP</th>\n",
       "      <th>Source.Port</th>\n",
       "      <th>Destination.IP</th>\n",
       "      <th>Destination.Port</th>\n",
       "      <th>Protocol</th>\n",
       "      <th>Timestamp</th>\n",
       "      <th>Flow.Duration</th>\n",
       "      <th>Total.Fwd.Packets</th>\n",
       "      <th>Total.Backward.Packets</th>\n",
       "      <th>Fwd.Packet.Length.Max</th>\n",
       "      <th>...</th>\n",
       "      <th>act_data_pkt_fwd</th>\n",
       "      <th>min_seg_size_forward</th>\n",
       "      <th>Active.Mean</th>\n",
       "      <th>Active.Std</th>\n",
       "      <th>Active.Max</th>\n",
       "      <th>Active.Min</th>\n",
       "      <th>Idle.Mean</th>\n",
       "      <th>Idle.Std</th>\n",
       "      <th>Idle.Max</th>\n",
       "      <th>Idle.Min</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>180881158</td>\n",
       "      <td>3128</td>\n",
       "      <td>3232288105</td>\n",
       "      <td>3125</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494314e+09</td>\n",
       "      <td>182</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>55</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>180881370</td>\n",
       "      <td>47719</td>\n",
       "      <td>3003188475</td>\n",
       "      <td>443</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494314e+09</td>\n",
       "      <td>16259595</td>\n",
       "      <td>21</td>\n",
       "      <td>12</td>\n",
       "      <td>258</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>32</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3232238139</td>\n",
       "      <td>55664</td>\n",
       "      <td>180881160</td>\n",
       "      <td>3128</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494314e+09</td>\n",
       "      <td>13470045</td>\n",
       "      <td>15</td>\n",
       "      <td>25</td>\n",
       "      <td>258</td>\n",
       "      <td>...</td>\n",
       "      <td>13</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3232258715</td>\n",
       "      <td>2310</td>\n",
       "      <td>180881161</td>\n",
       "      <td>3128</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494314e+09</td>\n",
       "      <td>258099</td>\n",
       "      <td>6</td>\n",
       "      <td>36</td>\n",
       "      <td>1460</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>180881156</td>\n",
       "      <td>3128</td>\n",
       "      <td>3232243811</td>\n",
       "      <td>50141</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494314e+09</td>\n",
       "      <td>73512428</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>257</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>20</td>\n",
       "      <td>4453.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4453.0</td>\n",
       "      <td>4453.0</td>\n",
       "      <td>73504132.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>73504132.0</td>\n",
       "      <td>73504132.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 75 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Source.IP  Source.Port  Destination.IP  Destination.Port  Protocol  \\\n",
       "0   180881158         3128      3232288105              3125         6   \n",
       "1   180881370        47719      3003188475               443         6   \n",
       "2  3232238139        55664       180881160              3128         6   \n",
       "3  3232258715         2310       180881161              3128         6   \n",
       "4   180881156         3128      3232243811             50141         6   \n",
       "\n",
       "      Timestamp  Flow.Duration  Total.Fwd.Packets  Total.Backward.Packets  \\\n",
       "0  1.494314e+09            182                  2                       0   \n",
       "1  1.494314e+09       16259595                 21                      12   \n",
       "2  1.494314e+09       13470045                 15                      25   \n",
       "3  1.494314e+09         258099                  6                      36   \n",
       "4  1.494314e+09       73512428                  4                       2   \n",
       "\n",
       "   Fwd.Packet.Length.Max  ...  act_data_pkt_fwd  min_seg_size_forward  \\\n",
       "0                     55  ...                 1                    20   \n",
       "1                    258  ...                 4                    32   \n",
       "2                    258  ...                13                    20   \n",
       "3                   1460  ...                 5                    20   \n",
       "4                    257  ...                 3                    20   \n",
       "\n",
       "   Active.Mean  Active.Std  Active.Max  Active.Min   Idle.Mean  Idle.Std  \\\n",
       "0          0.0         0.0         0.0         0.0         0.0       0.0   \n",
       "1          0.0         0.0         0.0         0.0         0.0       0.0   \n",
       "2          0.0         0.0         0.0         0.0         0.0       0.0   \n",
       "3          0.0         0.0         0.0         0.0         0.0       0.0   \n",
       "4       4453.0         0.0      4453.0      4453.0  73504132.0       0.0   \n",
       "\n",
       "     Idle.Max    Idle.Min  \n",
       "0         0.0         0.0  \n",
       "1         0.0         0.0  \n",
       "2         0.0         0.0  \n",
       "3         0.0         0.0  \n",
       "4  73504132.0  73504132.0  \n",
       "\n",
       "[5 rows x 75 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.utils import shuffle\n",
    "XU_train.head()"
   ]
  },
  {
   "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>Source.IP</th>\n",
       "      <th>Source.Port</th>\n",
       "      <th>Destination.IP</th>\n",
       "      <th>Destination.Port</th>\n",
       "      <th>Protocol</th>\n",
       "      <th>Timestamp</th>\n",
       "      <th>Flow.Duration</th>\n",
       "      <th>Total.Fwd.Packets</th>\n",
       "      <th>Total.Backward.Packets</th>\n",
       "      <th>Fwd.Packet.Length.Max</th>\n",
       "      <th>...</th>\n",
       "      <th>min_seg_size_forward</th>\n",
       "      <th>Active.Mean</th>\n",
       "      <th>Active.Std</th>\n",
       "      <th>Active.Max</th>\n",
       "      <th>Active.Min</th>\n",
       "      <th>Idle.Mean</th>\n",
       "      <th>Idle.Std</th>\n",
       "      <th>Idle.Max</th>\n",
       "      <th>Idle.Min</th>\n",
       "      <th>ProtocolName</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>180881370</td>\n",
       "      <td>44176</td>\n",
       "      <td>872483344</td>\n",
       "      <td>80</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494318e+09</td>\n",
       "      <td>263343</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>32</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>AMAZON</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>180881351</td>\n",
       "      <td>58339</td>\n",
       "      <td>877954422</td>\n",
       "      <td>80</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494314e+09</td>\n",
       "      <td>59769277</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>712</td>\n",
       "      <td>...</td>\n",
       "      <td>32</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>AMAZON</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>919780279</td>\n",
       "      <td>443</td>\n",
       "      <td>180881348</td>\n",
       "      <td>55558</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494324e+09</td>\n",
       "      <td>130702</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>542</td>\n",
       "      <td>...</td>\n",
       "      <td>32</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>AMAZON</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>180881351</td>\n",
       "      <td>55012</td>\n",
       "      <td>839964609</td>\n",
       "      <td>80</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494322e+09</td>\n",
       "      <td>188648</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>92</td>\n",
       "      <td>...</td>\n",
       "      <td>32</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>AMAZON</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>877262549</td>\n",
       "      <td>443</td>\n",
       "      <td>180881369</td>\n",
       "      <td>50404</td>\n",
       "      <td>6</td>\n",
       "      <td>1.494314e+09</td>\n",
       "      <td>1984</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>681</td>\n",
       "      <td>...</td>\n",
       "      <td>32</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>AMAZON</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 76 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Source.IP  Source.Port  Destination.IP  Destination.Port  Protocol  \\\n",
       "0  180881370        44176       872483344                80         6   \n",
       "1  180881351        58339       877954422                80         6   \n",
       "2  919780279          443       180881348             55558         6   \n",
       "3  180881351        55012       839964609                80         6   \n",
       "4  877262549          443       180881369             50404         6   \n",
       "\n",
       "      Timestamp  Flow.Duration  Total.Fwd.Packets  Total.Backward.Packets  \\\n",
       "0  1.494318e+09         263343                  9                       8   \n",
       "1  1.494314e+09       59769277                  9                       6   \n",
       "2  1.494324e+09         130702                  3                       3   \n",
       "3  1.494322e+09         188648                  3                       5   \n",
       "4  1.494314e+09           1984                  4                       3   \n",
       "\n",
       "   Fwd.Packet.Length.Max  ...  min_seg_size_forward  Active.Mean  Active.Std  \\\n",
       "0                      1  ...                    32          0.0         0.0   \n",
       "1                    712  ...                    32          0.0         0.0   \n",
       "2                    542  ...                    32          0.0         0.0   \n",
       "3                     92  ...                    32          0.0         0.0   \n",
       "4                    681  ...                    32          0.0         0.0   \n",
       "\n",
       "   Active.Max  Active.Min  Idle.Mean  Idle.Std  Idle.Max  Idle.Min  \\\n",
       "0         0.0         0.0        0.0       0.0       0.0       0.0   \n",
       "1         0.0         0.0        0.0       0.0       0.0       0.0   \n",
       "2         0.0         0.0        0.0       0.0       0.0       0.0   \n",
       "3         0.0         0.0        0.0       0.0       0.0       0.0   \n",
       "4         0.0         0.0        0.0       0.0       0.0       0.0   \n",
       "\n",
       "   ProtocolName  \n",
       "0        AMAZON  \n",
       "1        AMAZON  \n",
       "2        AMAZON  \n",
       "3        AMAZON  \n",
       "4        AMAZON  \n",
       "\n",
       "[5 rows x 76 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "XL.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from imblearn.over_sampling import RandomOverSampler\n",
    "from imblearn.under_sampling import RandomUnderSampler "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "a=XL.groupby('ProtocolName')['ProtocolName'].count().sort_values(ascending=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ProtocolName\n",
       "GOOGLE            5365\n",
       "YOUTUBE           5000\n",
       "YAHOO             5000\n",
       "APPLE             5000\n",
       "APPLE_ITUNES      5000\n",
       "CLOUDFLARE        5000\n",
       "CONTENT_FLASH     5000\n",
       "DROPBOX           5000\n",
       "EBAY              5000\n",
       "FACEBOOK          5000\n",
       "GMAIL             5000\n",
       "HTTP              5000\n",
       "HTTP_CONNECT      5000\n",
       "HTTP_PROXY        5000\n",
       "MICROSOFT         5000\n",
       "MSN               5000\n",
       "OFFICE_365        5000\n",
       "SKYPE             5000\n",
       "SSL               5000\n",
       "TWITTER           5000\n",
       "WHATSAPP          5000\n",
       "WIKIPEDIA         5000\n",
       "WINDOWS_UPDATE    5000\n",
       "AMAZON            5000\n",
       "Name: ProtocolName, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a   # see resampled instances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Function mask_noise\n",
    "\n",
    "Input: numpy array , fraction_delete(between 0 and 1)\n",
    "\n",
    "Removes a fraction_delete fraction of the input array X randomly\n",
    "\n",
    "Output: numpy array after deletions\n",
    "\n",
    "'''\n",
    "def mask_noise(X,fraction_delete=0.2):\n",
    "    mask=np.random.rand(np.shape(X)[0],np.shape(X)[1])\n",
    "    mask=1*(mask>fraction_delete)\n",
    "    return mask*X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Function Gaussian_noise\n",
    "\n",
    "Input: numpy array , sigma\n",
    "\n",
    "adds noise distributed normally with a stddev sigma \n",
    "\n",
    "Output: numpy array after adding noise\n",
    "\n",
    "'''\n",
    "def Gaussian_noise(X,sigma=0.5):\n",
    "    noise = np.random.normal(0,sigma,np.shape(X))\n",
    "    return noise+X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.layers import Input, Dense\n",
    "from keras.models import Model\n",
    "import keras.backend as K\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import regularizers\n",
    "from keras.models import load_model\n",
    "import keras\n",
    "from keras.layers import Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Sparcity Autoencoder's Sparcity Activity Regularization\n",
    "\n",
    "'''\n",
    "def kl_divergence(rho, rho_hat):\n",
    "    return rho * tf.math.log(rho) - rho * tf.math.log(rho_hat+1e-12) + (1 - rho) * tf.math.log(1 - rho) - (1 - rho) * tf.math.log(1 - rho_hat+1e-12)\n",
    "\n",
    "class SparseActivityRegularizer(tf.keras.regularizers.Regularizer):\n",
    "\n",
    "    def __init__(self, p=0.1, sparsityBeta=3):\n",
    "        self.p = p\n",
    "        self.sparsityBeta = sparsityBeta\n",
    "\n",
    "    def __call__(self, x):\n",
    "        regularization = 0            \n",
    "\n",
    "        p_hat = K.mean(x, axis=0)\n",
    "        regularization += self.sparsityBeta * K.sum(kl_divergence(self.p, p_hat))\n",
    "\n",
    "        return regularization\n",
    "\n",
    "    def get_config(self):\n",
    "        return {\"name\": self.__class__.__name__} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.optimizers import RMSprop\n",
    "from keras.optimizers import Adam\n",
    "from keras.optimizers import Adagrad\n",
    "from keras.optimizers import SGD\n",
    "from sklearn.preprocessing import LabelEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "yL_train=XL.iloc[:,-1]\n",
    "XL_train=XL.iloc[:,:-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "yL_train=pd.get_dummies(yL_train)#One-hot_encoding\n",
    "\n",
    "# Split into train,test and val\n",
    "\n",
    "#randomly split into train(80%) and val(20%)\n",
    "XL_train,XL_val,yL_train,yL_val=train_test_split(XL_train,yL_train,random_state=0,test_size=0.2)\n",
    "\n",
    "#randomly split val into val(10%) and test(10%)\n",
    "XL_val,XL_test,yL_val,yL_test=train_test_split(XL_val,yL_val,random_state=0,test_size=0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " Reversing the one hot encoding for each set to be used for ML methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "yL_train2=yL_train[yL_train==1].stack().reset_index().drop(0,1)['level_1'] # reversing the one hot encoding for each set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "yL_val2=yL_val[yL_val==1].stack().reset_index().drop(0,1)['level_1']# reversing the one hot encoding for each set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "yL_test2=yL_test[yL_test==1].stack().reset_index().drop(0,1)['level_1']# reversing the one hot encoding for each set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.preprocessing import normalize\n",
    "\n",
    "#scaling and normalization\n",
    "scaler=StandardScaler()\n",
    "XU_train=scaler.fit_transform(XU_train)\n",
    "XL_train=scaler.fit_transform(XL_train)\n",
    "XL_val=scaler.transform(XL_val)\n",
    "XL_test=scaler.transform(XL_test)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Training the SDAE\n",
    "#### Training the Encoder parts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras_sequential_ascii import keras2ascii\n",
    "from tensorflow.keras import regularizers\n",
    "from keras.models import Sequential"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import backend as K\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Function SDAE\n",
    "Inputs: numpy array X, list layers,list activations,float rho,int n_epochs,string denoise,float fraction_delete\n",
    "\n",
    "denoise is the mode of adding noise\n",
    "rho is the sparcity constraint parameter\n",
    "layers is a list of no. of hidden nodes in each hidden layer\n",
    "activations is a list of activations used in each encoding layer\n",
    "\n",
    "trains each autoencoder layer and outputs all the encoder-layer models.\n",
    "\n",
    "Output: a list of keras models\n",
    "\n",
    "'''\n",
    "def SDAE(X,layers,activations,rho,n_epochs,denoise,fraction_delete=0.2):\n",
    "    layers.insert(0,len(X[0]))\n",
    "    #layers.insert(0,len(X_train.iloc[0]))\n",
    "    ec=[]\n",
    "    for i in range(len(layers)-1):\n",
    "        encoding_dim = layers[i+1]\n",
    "\n",
    "    # this is our input placeholder\n",
    "        input_ = Input(shape=(layers[i],))\n",
    "    # define regulariser\n",
    "        if rho[i] != None: \n",
    "            regulariser = SparseActivityRegularizer(rho[i],3)\n",
    "        else:\n",
    "            regulariser=None\n",
    "\n",
    "        # \"encoded\" is the encoded representation of the input\n",
    "        encoded = Dense(encoding_dim, activation=activations[i],bias_initializer='random_normal',kernel_initializer='he_uniform',activity_regularizer=regulariser,name='encoder'+str(i+1))(input_)\n",
    "        \n",
    "        # \"decoded\" is the lossy reconstruction of the input\n",
    "        decoded = Dense(layers[i],activation=None,name='decoder'+str(i+1))(encoded)\n",
    "\n",
    "        # this model maps an input to its reconstruction\n",
    "        autoencoder = Model(input_, decoded)\n",
    "\n",
    "        # this model maps an input to its encoded representation\n",
    "        encoder = Model(input_, encoded)\n",
    "\n",
    "        # create a placeholder for an encoded (32-dimensional) input\n",
    "        encoded_input = Input(shape=(encoding_dim,))\n",
    "\n",
    "        autoencoder.compile(optimizer=RMSprop(learning_rate=0.001,clipnorm=0.1), loss='mse')\n",
    "        \n",
    "        if denoise =='Gauss':\n",
    "            X_noisy=Gaussian_noise(X,fraction_delete)\n",
    "        else:\n",
    "            X_noisy=mask_noise(X,fraction_delete)\n",
    "        \n",
    "        print('Training Layer ' + str(i+1)+' : ')\n",
    "        autoencoder.fit(X_noisy,X,\n",
    "                epochs=n_epochs,\n",
    "                batch_size=256,\n",
    "                shuffle=True,\n",
    "                validation_data=(X,X))\n",
    "        ec.append(encoder)\n",
    "        X=encoder.predict(X)\n",
    "    print('All Layers Done!')    \n",
    "    return ec       "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Layer 1 : \n",
      "Train on 384366 samples, validate on 384366 samples\n",
      "Epoch 1/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 3.2923 - val_loss: 0.4568\n",
      "Epoch 2/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.4191 - val_loss: 0.3589\n",
      "Epoch 3/10\n",
      "384366/384366 [==============================] - 8s 21us/step - loss: 0.3510 - val_loss: 0.3089\n",
      "Epoch 4/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.3105 - val_loss: 0.2697\n",
      "Epoch 5/10\n",
      "384366/384366 [==============================] - 7s 19us/step - loss: 0.2822 - val_loss: 0.2536\n",
      "Epoch 6/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.2568 - val_loss: 0.2220\n",
      "Epoch 7/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.2377 - val_loss: 0.2188\n",
      "Epoch 8/10\n",
      "384366/384366 [==============================] - 8s 21us/step - loss: 0.2207 - val_loss: 0.2071\n",
      "Epoch 9/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.2062 - val_loss: 0.1963\n",
      "Epoch 10/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.1929 - val_loss: 0.1796\n",
      "Training Layer 2 : \n",
      "Train on 384366 samples, validate on 384366 samples\n",
      "Epoch 1/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.2764 - val_loss: 0.2412\n",
      "Epoch 2/10\n",
      "384366/384366 [==============================] - 8s 21us/step - loss: 0.1803 - val_loss: 0.1757\n",
      "Epoch 3/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.1654 - val_loss: 0.1771\n",
      "Epoch 4/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.1552 - val_loss: 0.1800\n",
      "Epoch 5/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.1470 - val_loss: 0.1582\n",
      "Epoch 6/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.1420 - val_loss: 0.1932\n",
      "Epoch 7/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.1386 - val_loss: 0.1257\n",
      "Epoch 8/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.1363 - val_loss: 0.1735\n",
      "Epoch 9/10\n",
      "384366/384366 [==============================] - 8s 21us/step - loss: 0.1330 - val_loss: 0.1788\n",
      "Epoch 10/10\n",
      "384366/384366 [==============================] - 8s 20us/step - loss: 0.1320 - val_loss: 0.1226\n",
      "Training Layer 3 : \n",
      "Train on 384366 samples, validate on 384366 samples\n",
      "Epoch 1/10\n",
      "384366/384366 [==============================] - 14s 36us/step - loss: 0.7841 - val_loss: 0.5975\n",
      "Epoch 2/10\n",
      "384366/384366 [==============================] - 13s 35us/step - loss: 0.4448 - val_loss: 0.4692\n",
      "Epoch 3/10\n",
      "384366/384366 [==============================] - 13s 35us/step - loss: 0.4390 - val_loss: 0.7912\n",
      "Epoch 4/10\n",
      "384366/384366 [==============================] - 14s 35us/step - loss: 0.4362 - val_loss: 0.4573\n",
      "Epoch 5/10\n",
      "384366/384366 [==============================] - 13s 35us/step - loss: 0.4337 - val_loss: 0.6329\n",
      "Epoch 6/10\n",
      "384366/384366 [==============================] - 13s 35us/step - loss: 0.4330 - val_loss: 0.4989\n",
      "Epoch 7/10\n",
      "384366/384366 [==============================] - 13s 35us/step - loss: 0.4262 - val_loss: 0.3997\n",
      "Epoch 8/10\n",
      "384366/384366 [==============================] - 14s 35us/step - loss: 0.4219 - val_loss: 0.6562\n",
      "Epoch 9/10\n",
      "384366/384366 [==============================] - 14s 35us/step - loss: 0.4196 - val_loss: 0.5480\n",
      "Epoch 10/10\n",
      "384366/384366 [==============================] - 14s 35us/step - loss: 0.4233 - val_loss: 0.4213\n",
      "Training Layer 4 : \n",
      "Train on 384366 samples, validate on 384366 samples\n",
      "Epoch 1/10\n",
      "384366/384366 [==============================] - 11s 27us/step - loss: 0.1294 - val_loss: 0.1075\n",
      "Epoch 2/10\n",
      "384366/384366 [==============================] - 10s 27us/step - loss: 0.0995 - val_loss: 0.1038\n",
      "Epoch 3/10\n",
      "384366/384366 [==============================] - 10s 26us/step - loss: 0.0977 - val_loss: 0.1308\n",
      "Epoch 4/10\n",
      "384366/384366 [==============================] - 10s 26us/step - loss: 0.0978 - val_loss: 0.0903\n",
      "Epoch 5/10\n",
      "384366/384366 [==============================] - 10s 26us/step - loss: 0.0979 - val_loss: 0.1458\n",
      "Epoch 6/10\n",
      "384366/384366 [==============================] - 10s 26us/step - loss: 0.0978 - val_loss: 0.1334\n",
      "Epoch 7/10\n",
      "384366/384366 [==============================] - 10s 25us/step - loss: 0.0971 - val_loss: 0.0715\n",
      "Epoch 8/10\n",
      "384366/384366 [==============================] - 10s 26us/step - loss: 0.0954 - val_loss: 0.2306\n",
      "Epoch 9/10\n",
      "384366/384366 [==============================] - 10s 26us/step - loss: 0.0975 - val_loss: 0.0716\n",
      "Epoch 10/10\n",
      "384366/384366 [==============================] - 10s 27us/step - loss: 0.0966 - val_loss: 0.0703\n",
      "All Layers Done!\n"
     ]
    }
   ],
   "source": [
    "#Train Sparse Stacked Denoising Autoencoder\n",
    "layers = [100,200,400,50]\n",
    "activations=['relu','relu','relu','relu']\n",
    "rho=[0.06,0.06,0.06,0.06]\n",
    "traint0=time.time()\n",
    "stacked_layers=SDAE(XU_train,layers,activations,rho,10,'mask',0.02)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3.Training the SoftMax classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Performs a Forward Pass through the SDAE\n",
    "\n",
    "'''\n",
    "def SDAE_predict(stacked_layers,X):\n",
    "    for layer in stacked_layers:\n",
    "        X1=layer.predict(X)\n",
    "        X=X1\n",
    "    return X1    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "#forward pass through SDAE\n",
    "XL_train1=SDAE_predict(stacked_layers,XL_train)\n",
    "XL_val1=SDAE_predict(stacked_layers,XL_val)\n",
    "XL_test1=SDAE_predict(stacked_layers,XL_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Please skip this section to part 6 If only ML methods are to be tested, The following is the DL NN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Train Classifier separately\n",
    "input1 = Input(shape=(layers[-1],))\n",
    "L1 = Dense(len(yL_train.columns),activation='softmax')(input1)\n",
    "classifier=Model(input1,L1)   \n",
    "classifier.compile(optimizer=SGD(learning_rate=0.1,momentum=0.9),loss='categorical_crossentropy',metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 96292 samples, validate on 12036 samples\n",
      "Epoch 1/50\n",
      "96292/96292 [==============================] - 3s 26us/step - loss: 2.8137 - accuracy: 0.1870 - val_loss: 2.7148 - val_accuracy: 0.2038\n",
      "Epoch 2/50\n",
      "96292/96292 [==============================] - 2s 24us/step - loss: 2.6614 - accuracy: 0.2234 - val_loss: 2.6311 - val_accuracy: 0.2192\n",
      "Epoch 3/50\n",
      "96292/96292 [==============================] - 2s 24us/step - loss: 2.5980 - accuracy: 0.2376 - val_loss: 2.5866 - val_accuracy: 0.2342\n",
      "Epoch 4/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.5548 - accuracy: 0.2509 - val_loss: 2.5490 - val_accuracy: 0.2454\n",
      "Epoch 5/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.5232 - accuracy: 0.2604 - val_loss: 2.5201 - val_accuracy: 0.2492\n",
      "Epoch 6/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.4975 - accuracy: 0.2698 - val_loss: 2.5058 - val_accuracy: 0.2663\n",
      "Epoch 7/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.4749 - accuracy: 0.2777 - val_loss: 2.4773 - val_accuracy: 0.2685\n",
      "Epoch 8/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.4569 - accuracy: 0.2848 - val_loss: 2.4687 - val_accuracy: 0.2627\n",
      "Epoch 9/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.4413 - accuracy: 0.2882 - val_loss: 2.4579 - val_accuracy: 0.2826\n",
      "Epoch 10/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.4270 - accuracy: 0.2946 - val_loss: 2.4361 - val_accuracy: 0.2848\n",
      "Epoch 11/50\n",
      "96292/96292 [==============================] - 2s 23us/step - loss: 2.4146 - accuracy: 0.2980 - val_loss: 2.4281 - val_accuracy: 0.2871\n",
      "Epoch 12/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.4031 - accuracy: 0.3013 - val_loss: 2.4118 - val_accuracy: 0.2984\n",
      "Epoch 13/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3927 - accuracy: 0.3036 - val_loss: 2.4042 - val_accuracy: 0.3023\n",
      "Epoch 14/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3832 - accuracy: 0.3061 - val_loss: 2.3972 - val_accuracy: 0.2900\n",
      "Epoch 15/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3745 - accuracy: 0.3071 - val_loss: 2.3890 - val_accuracy: 0.3018\n",
      "Epoch 16/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.3664 - accuracy: 0.3096 - val_loss: 2.3777 - val_accuracy: 0.3073\n",
      "Epoch 17/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3586 - accuracy: 0.3108 - val_loss: 2.3748 - val_accuracy: 0.2949\n",
      "Epoch 18/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3517 - accuracy: 0.3132 - val_loss: 2.3686 - val_accuracy: 0.3115\n",
      "Epoch 19/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3455 - accuracy: 0.3156 - val_loss: 2.3621 - val_accuracy: 0.3127\n",
      "Epoch 20/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3393 - accuracy: 0.3156 - val_loss: 2.3561 - val_accuracy: 0.3075\n",
      "Epoch 21/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3333 - accuracy: 0.3187 - val_loss: 2.3462 - val_accuracy: 0.3105\n",
      "Epoch 22/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3284 - accuracy: 0.3195 - val_loss: 2.3457 - val_accuracy: 0.3140\n",
      "Epoch 23/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3227 - accuracy: 0.3233 - val_loss: 2.3388 - val_accuracy: 0.3217\n",
      "Epoch 24/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3180 - accuracy: 0.3234 - val_loss: 2.3371 - val_accuracy: 0.3105\n",
      "Epoch 25/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3141 - accuracy: 0.3255 - val_loss: 2.3311 - val_accuracy: 0.3189\n",
      "Epoch 26/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3094 - accuracy: 0.3253 - val_loss: 2.3230 - val_accuracy: 0.3207\n",
      "Epoch 27/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3052 - accuracy: 0.3287 - val_loss: 2.3284 - val_accuracy: 0.3179\n",
      "Epoch 28/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.3017 - accuracy: 0.3288 - val_loss: 2.3241 - val_accuracy: 0.3229\n",
      "Epoch 29/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.2977 - accuracy: 0.3307 - val_loss: 2.3190 - val_accuracy: 0.3205\n",
      "Epoch 30/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.2946 - accuracy: 0.3309 - val_loss: 2.3130 - val_accuracy: 0.3149\n",
      "Epoch 31/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.2911 - accuracy: 0.3311 - val_loss: 2.3061 - val_accuracy: 0.3284\n",
      "Epoch 32/50\n",
      "96292/96292 [==============================] - 2s 23us/step - loss: 2.2885 - accuracy: 0.3332 - val_loss: 2.3083 - val_accuracy: 0.3240\n",
      "Epoch 33/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.2849 - accuracy: 0.3332 - val_loss: 2.2983 - val_accuracy: 0.3271\n",
      "Epoch 34/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.2820 - accuracy: 0.3336 - val_loss: 2.3013 - val_accuracy: 0.3264\n",
      "Epoch 35/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.2790 - accuracy: 0.3351 - val_loss: 2.2966 - val_accuracy: 0.3226\n",
      "Epoch 36/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2763 - accuracy: 0.3362 - val_loss: 2.2944 - val_accuracy: 0.3199\n",
      "Epoch 37/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2739 - accuracy: 0.3360 - val_loss: 2.2879 - val_accuracy: 0.3277\n",
      "Epoch 38/50\n",
      "96292/96292 [==============================] - 2s 23us/step - loss: 2.2709 - accuracy: 0.3382 - val_loss: 2.2911 - val_accuracy: 0.3261\n",
      "Epoch 39/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2691 - accuracy: 0.3384 - val_loss: 2.2850 - val_accuracy: 0.3268\n",
      "Epoch 40/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2670 - accuracy: 0.3391 - val_loss: 2.2835 - val_accuracy: 0.3239\n",
      "Epoch 41/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2649 - accuracy: 0.3402 - val_loss: 2.2781 - val_accuracy: 0.3328\n",
      "Epoch 42/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2623 - accuracy: 0.3403 - val_loss: 2.2791 - val_accuracy: 0.3248\n",
      "Epoch 43/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.2602 - accuracy: 0.3403 - val_loss: 2.2787 - val_accuracy: 0.3272\n",
      "Epoch 44/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.2586 - accuracy: 0.3413 - val_loss: 2.2736 - val_accuracy: 0.3329\n",
      "Epoch 45/50\n",
      "96292/96292 [==============================] - 2s 22us/step - loss: 2.2568 - accuracy: 0.3422 - val_loss: 2.2751 - val_accuracy: 0.3309\n",
      "Epoch 46/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2545 - accuracy: 0.3436 - val_loss: 2.2760 - val_accuracy: 0.3352\n",
      "Epoch 47/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2531 - accuracy: 0.3438 - val_loss: 2.2737 - val_accuracy: 0.3285\n",
      "Epoch 48/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2510 - accuracy: 0.3440 - val_loss: 2.2675 - val_accuracy: 0.3319\n",
      "Epoch 49/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2493 - accuracy: 0.3450 - val_loss: 2.2692 - val_accuracy: 0.3396\n",
      "Epoch 50/50\n",
      "96292/96292 [==============================] - 2s 21us/step - loss: 2.2477 - accuracy: 0.3456 - val_loss: 2.2627 - val_accuracy: 0.3429\n"
     ]
    }
   ],
   "source": [
    "history=classifier.fit(XL_train1,yL_train,\n",
    "                epochs=50,\n",
    "                #batch_size=256, #default batch_size=32\n",
    "                shuffle=True,\n",
    "                validation_data=(XL_val1, yL_val))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. Fine Tuning\n",
    "#### Fine Tuning Involves removing the sparsity regularizer from all the layers , adding dropout etc and performing backpropagation to improve the overall model accuracy "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\dhruvjyoti\\appdata\\local\\programs\\python\\python37\\lib\\site-packages\\keras\\engine\\saving.py:341: UserWarning: No training configuration found in save file: the model was *not* compiled. Compile it manually.\n",
      "  warnings.warn('No training configuration found in save file: '\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[]\n",
      "[]\n",
      "[]\n",
      "[]\n",
      "[]\n"
     ]
    }
   ],
   "source": [
    "# change regularisation settings\n",
    "a=[]\n",
    "for i in range(len(stacked_layers)):\n",
    "    encoder=stacked_layers[i]\n",
    "    #encoder.layers[-2].activity_regularizer=None  \n",
    "    encoder.layers[-1].activity_regularizer=None #remove Sparse activity regularizer\n",
    "    encoder.save('temp.h5')\n",
    "    encoder=load_model('temp.h5')\n",
    "    print(encoder.losses)\n",
    "    a.append(encoder)\n",
    "    \n",
    "classifier.save('temp.h5')\n",
    "classifier=load_model('temp.h5')\n",
    "print(classifier.losses)\n",
    "a.append(classifier)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# add dropout till n-2 layers\n",
    "model=Sequential()\n",
    "for i in a:\n",
    "    model.add(i)\n",
    "    if i != classifier and i != encoder:\n",
    "        model.add(Dropout(0.02))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\dhruvjyoti\\appdata\\local\\programs\\python\\python37\\lib\\site-packages\\keras\\engine\\saving.py:341: UserWarning: No training configuration found in save file: the model was *not* compiled. Compile it manually.\n",
      "  warnings.warn('No training configuration found in save file: '\n"
     ]
    }
   ],
   "source": [
    "model.save('temp.h5')\n",
    "model=load_model('temp.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer=RMSprop(learning_rate=0.0001),loss='categorical_crossentropy',metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "#fine tune model runs for no. of epochs set below\n",
    "history=model.fit(XL_train,yL_train,\n",
    "                epochs=750,\n",
    "                batch_size=256,\n",
    "                shuffle=True,\n",
    "                validation_data=(XL_val, yL_val),\n",
    "                verbose=0)\n",
    "traint1=time.time()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train_time: 2712.742925100327\n"
     ]
    }
   ],
   "source": [
    "print(\"Train_time: \"+str(traint1-traint0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.2525774887911904, 0.951129241943359]\n",
      "Test_time: 0.374902311706543\n"
     ]
    }
   ],
   "source": [
    "test0=time.time()\n",
    "print(model.evaluate(XL_test,yL_test))\n",
    "test1=time.time()\n",
    "print(\"Test_time: \"+str(test1-test0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1fc9bd6c2e8>]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Visualise convergence with no. of epochs\n",
    "plt.plot(history.history['accuracy'])\n",
    "plt.plot(history.history['val_accuracy'])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5. Results\n",
    "#### Below are the results obtained on our NN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=model.predict(XL_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred1=y_pred.argmax(axis=1)\n",
    "y_pred1=np.append(y_pred1,np.arange(24))\n",
    "y_pred1=pd.get_dummies(y_pred1,columns=yL_train.columns)\n",
    "y_pred1.columns=yL_test.columns\n",
    "y_pred1=y_pred1.iloc[:-24]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AMAZON            500\n",
       "APPLE             474\n",
       "APPLE_ITUNES      515\n",
       "CLOUDFLARE        461\n",
       "CONTENT_FLASH     494\n",
       "DROPBOX           487\n",
       "EBAY              508\n",
       "FACEBOOK          473\n",
       "GMAIL             501\n",
       "GOOGLE            550\n",
       "HTTP              491\n",
       "HTTP_CONNECT      490\n",
       "HTTP_PROXY        533\n",
       "MICROSOFT         497\n",
       "MSN               497\n",
       "OFFICE_365        522\n",
       "SKYPE             491\n",
       "SSL               517\n",
       "TWITTER           509\n",
       "WHATSAPP          506\n",
       "WIKIPEDIA         511\n",
       "WINDOWS_UPDATE    502\n",
       "YAHOO             515\n",
       "YOUTUBE           493\n",
       "dtype: int64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yL_test.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_test1 = yL_test.columns[np.where(yL_test!=0)[1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['WINDOWS_UPDATE', 'APPLE', 'APPLE_ITUNES', 'OFFICE_365', 'MICROSOFT',\n",
       "       'WIKIPEDIA', 'CLOUDFLARE', 'HTTP_CONNECT', 'EBAY', 'MSN',\n",
       "       ...\n",
       "       'WINDOWS_UPDATE', 'YAHOO', 'HTTP_PROXY', 'GMAIL', 'HTTP', 'MSN',\n",
       "       'GMAIL', 'HTTP_CONNECT', 'WHATSAPP', 'SKYPE'],\n",
       "      dtype='object', length=12037)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred1=y_pred1.columns[np.where(y_pred1!=0)[1]]\n",
    "y_pred1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "cm=confusion_matrix(y_test1,y_pred1,labels=yL_test.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "cm_pc=pd.DataFrame((cm.T/np.sum(cm,axis=1)).T,columns=yL_test.columns,index=yL_test.columns)\n",
    "cm_pc=cm_pc.replace(np.nan,0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2520x1800 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot confusion matrix\n",
    "plt.figure(figsize=(35,25))\n",
    "df_cm=pd.DataFrame(cm,columns=yL_test.columns,index=yL_test.columns)\n",
    "sns.set(font_scale=1.0)\n",
    "sns.heatmap(cm_pc, annot=True,fmt='.1%',cmap=sns.color_palette(\"Reds\",50)) # font size\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy : 0.9503459582952563\n"
     ]
    }
   ],
   "source": [
    "print('Test Accuracy : '+str(accuracy_score(y_test1, y_pred1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "weighted F1 Score : 0.9551652225407365\n",
      "Macro F1 Score : 0.9440849230700469\n",
      "Micro F1 Score : 0.9513459582952563\n",
      "F1 Scores : \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>F1-score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AMAZON</th>\n",
       "      <td>0.958498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>APPLE</th>\n",
       "      <td>0.997895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>APPLE_ITUNES</th>\n",
       "      <td>0.999030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CLOUDFLARE</th>\n",
       "      <td>0.985043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONTENT_FLASH</th>\n",
       "      <td>0.996973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DROPBOX</th>\n",
       "      <td>0.993878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EBAY</th>\n",
       "      <td>0.998035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FACEBOOK</th>\n",
       "      <td>0.978283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GMAIL</th>\n",
       "      <td>0.959381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GOOGLE</th>\n",
       "      <td>0.648765</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HTTP</th>\n",
       "      <td>0.801205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HTTP_CONNECT</th>\n",
       "      <td>0.873767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HTTP_PROXY</th>\n",
       "      <td>0.703390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MICROSOFT</th>\n",
       "      <td>0.982213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MSN</th>\n",
       "      <td>0.987090</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OFFICE_365</th>\n",
       "      <td>0.998088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKYPE</th>\n",
       "      <td>0.979063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SSL</th>\n",
       "      <td>0.886320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TWITTER</th>\n",
       "      <td>0.978599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WHATSAPP</th>\n",
       "      <td>0.993131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WIKIPEDIA</th>\n",
       "      <td>0.998047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WINDOWS_UPDATE</th>\n",
       "      <td>0.990138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YAHOO</th>\n",
       "      <td>0.997096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YOUTUBE</th>\n",
       "      <td>0.902486</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                F1-score\n",
       "AMAZON          0.958498\n",
       "APPLE           0.997895\n",
       "APPLE_ITUNES    0.999030\n",
       "CLOUDFLARE      0.985043\n",
       "CONTENT_FLASH   0.996973\n",
       "DROPBOX         0.993878\n",
       "EBAY            0.998035\n",
       "FACEBOOK        0.978283\n",
       "GMAIL           0.959381\n",
       "GOOGLE          0.648765\n",
       "HTTP            0.801205\n",
       "HTTP_CONNECT    0.873767\n",
       "HTTP_PROXY      0.703390\n",
       "MICROSOFT       0.982213\n",
       "MSN             0.987090\n",
       "OFFICE_365      0.998088\n",
       "SKYPE           0.979063\n",
       "SSL             0.886320\n",
       "TWITTER         0.978599\n",
       "WHATSAPP        0.993131\n",
       "WIKIPEDIA       0.998047\n",
       "WINDOWS_UPDATE  0.990138\n",
       "YAHOO           0.997096\n",
       "YOUTUBE         0.902486"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('weighted F1 Score : ' + str(f1_score(y_pred1,y_test1,average = 'weighted')))\n",
    "print('Macro F1 Score : ' + str(f1_score(y_pred1,y_test1,average = 'macro')))\n",
    "print('Micro F1 Score : ' + str(f1_score(y_pred1,y_test1,average = 'micro')))\n",
    "print('F1 Scores : ' )\n",
    "pd.DataFrame(f1_score(y_pred1,y_test1,labels=yL_test.columns,average = None),columns=['F1-score'],index=yL_test.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "24"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(yL_train.columns)# No. of distinct labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Weighted Precision : 0.948367839982929\n",
      "Macro Precision : 0.9640601348949986\n",
      "Micro Precision : 0.9422459582952563\n",
      "Precision Scores : \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Precision-score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AMAZON</th>\n",
       "      <td>0.970000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>APPLE</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>APPLE_ITUNES</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CLOUDFLARE</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONTENT_FLASH</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DROPBOX</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EBAY</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FACEBOOK</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GMAIL</th>\n",
       "      <td>0.990020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GOOGLE</th>\n",
       "      <td>0.549091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HTTP</th>\n",
       "      <td>0.812627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HTTP_CONNECT</th>\n",
       "      <td>0.904082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HTTP_PROXY</th>\n",
       "      <td>0.622889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MICROSOFT</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MSN</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OFFICE_365</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKYPE</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SSL</th>\n",
       "      <td>0.889749</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TWITTER</th>\n",
       "      <td>0.988212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WHATSAPP</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WIKIPEDIA</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WINDOWS_UPDATE</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YAHOO</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YOUTUBE</th>\n",
       "      <td>0.957404</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Precision-score\n",
       "AMAZON                 0.970000\n",
       "APPLE                  1.000000\n",
       "APPLE_ITUNES           1.000000\n",
       "CLOUDFLARE             1.000000\n",
       "CONTENT_FLASH          1.000000\n",
       "DROPBOX                1.000000\n",
       "EBAY                   1.000000\n",
       "FACEBOOK               1.000000\n",
       "GMAIL                  0.990020\n",
       "GOOGLE                 0.549091\n",
       "HTTP                   0.812627\n",
       "HTTP_CONNECT           0.904082\n",
       "HTTP_PROXY             0.622889\n",
       "MICROSOFT              1.000000\n",
       "MSN                    1.000000\n",
       "OFFICE_365             1.000000\n",
       "SKYPE                  1.000000\n",
       "SSL                    0.889749\n",
       "TWITTER                0.988212\n",
       "WHATSAPP               1.000000\n",
       "WIKIPEDIA              1.000000\n",
       "WINDOWS_UPDATE         1.000000\n",
       "YAHOO                  1.000000\n",
       "YOUTUBE                0.957404"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('Weighted Precision : '+str(precision_score(y_test1, y_pred1,average=\"weighted\")))\n",
    "print('Macro Precision : '+str(precision_score(y_test1, y_pred1,average=\"macro\")))\n",
    "print('Micro Precision : '+str(precision_score(y_test1, y_pred1,average=\"micro\")))\n",
    "print('Precision Scores : ' )\n",
    "pd.DataFrame(precision_score(y_pred1,y_test1,labels=yL_test.columns,average = None),columns=['Precision-score'],index=yL_test.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Weighted Recall : 0.9499043794165593\n",
      "Macro Recall : 0.9401459582952563\n",
      "Micro Recall : 0.9357459582952563\n",
      "Recall Scores : \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>recall-score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AMAZON</th>\n",
       "      <td>0.947266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>APPLE</th>\n",
       "      <td>0.995798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>APPLE_ITUNES</th>\n",
       "      <td>0.998062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CLOUDFLARE</th>\n",
       "      <td>0.970526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONTENT_FLASH</th>\n",
       "      <td>0.993964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DROPBOX</th>\n",
       "      <td>0.987830</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EBAY</th>\n",
       "      <td>0.996078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FACEBOOK</th>\n",
       "      <td>0.957490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GMAIL</th>\n",
       "      <td>0.930582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GOOGLE</th>\n",
       "      <td>0.792651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HTTP</th>\n",
       "      <td>0.790099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HTTP_CONNECT</th>\n",
       "      <td>0.845420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HTTP_PROXY</th>\n",
       "      <td>0.807786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MICROSOFT</th>\n",
       "      <td>0.965049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MSN</th>\n",
       "      <td>0.974510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OFFICE_365</th>\n",
       "      <td>0.996183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKYPE</th>\n",
       "      <td>0.958984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SSL</th>\n",
       "      <td>0.882917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TWITTER</th>\n",
       "      <td>0.969171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WHATSAPP</th>\n",
       "      <td>0.986355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WIKIPEDIA</th>\n",
       "      <td>0.996101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WINDOWS_UPDATE</th>\n",
       "      <td>0.980469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YAHOO</th>\n",
       "      <td>0.994208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YOUTUBE</th>\n",
       "      <td>0.853526</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                recall-score\n",
       "AMAZON              0.947266\n",
       "APPLE               0.995798\n",
       "APPLE_ITUNES        0.998062\n",
       "CLOUDFLARE          0.970526\n",
       "CONTENT_FLASH       0.993964\n",
       "DROPBOX             0.987830\n",
       "EBAY                0.996078\n",
       "FACEBOOK            0.957490\n",
       "GMAIL               0.930582\n",
       "GOOGLE              0.792651\n",
       "HTTP                0.790099\n",
       "HTTP_CONNECT        0.845420\n",
       "HTTP_PROXY          0.807786\n",
       "MICROSOFT           0.965049\n",
       "MSN                 0.974510\n",
       "OFFICE_365          0.996183\n",
       "SKYPE               0.958984\n",
       "SSL                 0.882917\n",
       "TWITTER             0.969171\n",
       "WHATSAPP            0.986355\n",
       "WIKIPEDIA           0.996101\n",
       "WINDOWS_UPDATE      0.980469\n",
       "YAHOO               0.994208\n",
       "YOUTUBE             0.853526"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('Weighted Recall : '+str(recall_score(y_test1, y_pred1,average=\"weighted\")))\n",
    "print('Macro Recall : '+str(recall_score(y_test1, y_pred1,average=\"macro\")))\n",
    "print('Micro Recall : '+str(recall_score(y_test1, y_pred1,average=\"micro\")))\n",
    "print('Recall Scores : ' )\n",
    "pd.DataFrame(recall_score(y_pred1,y_test1,labels=yL_test.columns,average = None),columns=['recall-score'],index=yL_test.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Confusion Matrix with 10 chosen well-known classes\n",
    "Classes : AMAZON, FACEBOOK, GMAIL, GOOGLE, HTTP, OFFICE_365, SKYPE, TWITTER, WHATSAPP, YOUTUBE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "apps=['AMAZON','FACEBOOK','GMAIL','GOOGLE','HTTP','OFFICE_365','SKYPE','TWITTER','WHATSAPP','YOUTUBE']#reporting for these well-known Apps\n",
    "y_test1[y_test1.isin(apps)]\n",
    "cm=confusion_matrix(y_test1[y_test1.isin(apps)],y_pred1[y_test1.isin(apps)],labels=y_test1[y_test1.isin(apps)].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "cm_pc=pd.DataFrame((cm.T/np.sum(cm,axis=1)).T,columns=y_test1[y_test1.isin(apps)].unique(),index=y_test1[y_test1.isin(apps)].unique())\n",
    "cm_pc=cm_pc.replace(np.nan,0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,10))\n",
    "df_cm=pd.DataFrame(cm,columns=y_test1[y_test1.isin(apps)].unique(),index=y_test1[y_test1.isin(apps)].unique())\n",
    "sns.set(font_scale=1.1)\n",
    "sns.heatmap(cm_pc, annot=True,fmt='.1%',cmap=sns.color_palette(\"Reds\",150)) # font size\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 6. ML methods\n",
    "#### Performing ML methods on Transformed data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import sklearn libraries\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import f1_score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RandomForest\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(max_depth=60, random_state=0)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "clf=RandomForestClassifier(random_state=0, max_depth = 60)\n",
    "clf.fit(XL_train1,yL_train2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9415030324429859\n",
      "0.9007844645675832\n"
     ]
    }
   ],
   "source": [
    "print(clf.score(XL_train1,yL_train2))# train accuracy\n",
    "print(clf.score(XL_test1,yL_test2))# test accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=clf.predict(XL_test1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "weighted F1 Score : 0.8945228929535777\n",
      "weighted Precision : 0.9180610288207022\n",
      "weighted Recall : 0.9007844645675832\n"
     ]
    }
   ],
   "source": [
    "# Results\n",
    "print('weighted F1 Score : ' + str(f1_score(y_pred,yL_test2,average='weighted')))\n",
    "print('weighted Precision : ' + str(precision_score(y_pred,yL_test2,average='weighted')))\n",
    "print('weighted Recall : ' + str(recall_score(y_pred,yL_test2,average='weighted')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "macro F1 Score : 0.8811202439535777\n",
      "macro Precision : 0.9091948576626351\n",
      "macro Recall : 0.8968722819883678\n"
     ]
    }
   ],
   "source": [
    "# Results\n",
    "print('macro F1 Score : ' + str(f1_score(y_pred,yL_test2,average='macro')))\n",
    "print('macro Precision : ' + str(precision_score(y_pred,yL_test2,average='macro')))\n",
    "print('macro Recall : ' + str(recall_score(y_pred,yL_test2,average='macro')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC\n",
    "clf=SVC(random_state=0,kernel='rbf').fit(XL_train1,yL_train2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5646471316412579\n",
      "0.5619350918002824\n"
     ]
    }
   ],
   "source": [
    "print(clf.score(XL_train1,yL_train2))# train accuracy\n",
    "print(clf.score(XL_test1,yL_test2))# test accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=clf.predict(XL_test1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "weighted F1 Score : 0.5716087672478307\n",
      "weighted Precision : 0.6389943732560074\n",
      "weighted Recall : 0.5619350918002824\n"
     ]
    }
   ],
   "source": [
    "#results\n",
    "print('weighted F1 Score : ' + str(f1_score(y_pred,yL_test2,average = 'weighted')))\n",
    "print('weighted Precision : ' + str(precision_score(y_pred,yL_test2,average = 'weighted')))\n",
    "print('weighted Recall : ' + str(recall_score(y_pred,yL_test2,average = 'weighted')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "macro F1 Score : 0.5711357632472212\n",
      "macro Precision : 0.6365976626350074\n",
      "macro Recall : 0.5509540914002559\n"
     ]
    }
   ],
   "source": [
    "#results\n",
    "print('macro F1 Score : ' + str(f1_score(y_pred,yL_test2,average = 'macro')))\n",
    "print('macro' Precision :  + str(precision_score(y_pred,yL_test2,average = 'macro')))\n",
    "print('macro Recall : ' + str(recall_score(y_pred,yL_test2,average = 'macro')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "clf=DecisionTreeClassifier(max_depth=60).fit(XL_train1,yL_train2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9990030324429859"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.score(XL_train1,yL_train2) #train accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8889428844396444\n"
     ]
    }
   ],
   "source": [
    "print(clf.score(XL_test1,yL_test2))# test accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "weighted F1 Score : 0.8842415669269293\n",
      "weighted Precision : 0.9093726370501222\n",
      "weighted Recall : 0.8889428844396444\n"
     ]
    }
   ],
   "source": [
    "#results\n",
    "y_pred=clf.predict(XL_test1)\n",
    "print('weighted F1 Score : ' + str(f1_score(y_pred,yL_test2,average = 'weighted')))\n",
    "print('weighted Precision : ' + str(precision_score(y_pred,yL_test2,average = 'weighted')))\n",
    "print('weighted Recall : ' + str(recall_score(y_pred,yL_test2,average = 'weighted')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "macro F1 Score : 0.874241564456929\n",
      "macro Precision : 0.900267837461521\n",
      "macro Recall : 0.890280052745829\n"
     ]
    }
   ],
   "source": [
    "#results\n",
    "y_pred=clf.predict(XL_test1)\n",
    "print('macro F1 Score : ' + str(f1_score(y_pred,yL_test2,average = 'macro')))\n",
    "print('macro Precision : ' + str(precision_score(y_pred,yL_test2,average = 'macro')))\n",
    "print('macro Recall : ' + str(recall_score(y_pred,yL_test2,average = 'macro')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### XGBOOST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "from xgboost import XGBClassifier\n",
    "from sklearn.multiclass import OneVsRestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "train0=time.time()\n",
    "clf = OneVsRestClassifier(XGBClassifier(verbosity=0)).fit(XL_train1, yL_train2)\n",
    "train1=time.time()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train-score: 0.9510555161573488\n"
     ]
    }
   ],
   "source": [
    "print('Train-score: '+str(clf.score(XL_train1,yL_train2)))# train accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test-score: 0.9170889756583867\n"
     ]
    }
   ],
   "source": [
    "print('Test-score: '+str(clf.score(XL_test1,yL_test2)))# test accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "weighted F1 Score : 0.9205838415826192\n",
      "weighted Precision : 0.9308747161575589\n",
      "weighted Recall : 0.9170889756583867\n"
     ]
    }
   ],
   "source": [
    "#results\n",
    "y_pred=clf.predict(XL_test1)\n",
    "print('weighted F1 Score : ' + str(f1_score(y_pred,yL_test2,average = 'weighted')))\n",
    "print('weighted Precision : ' + str(precision_score(y_pred,yL_test2,average = 'weighted')))\n",
    "print('weighted Recall : ' + str(recall_score(y_pred,yL_test2,average = 'weighted')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "macro F1 Score : 0.922183877802733\n",
      "macro Precision : 0.9300789263578801\n",
      "macro Recall : 0.9111034044027819\n"
     ]
    }
   ],
   "source": [
    "#results\n",
    "y_pred=clf.predict(XL_test1)\n",
    "print('macro F1 Score : ' + str(f1_score(y_pred,yL_test2,average = 'macro')))\n",
    "print('macro Precision : ' + str(precision_score(y_pred,yL_test2,average = 'maco')))\n",
    "print('macro Recall : ' + str(recall_score(y_pred,yL_test2,average = 'macro')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}