03-Classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5G UE-VBS Selection ML Model\n",
    "\n",
    "## Classification\n",
    "\n",
    "In this notebook, we will build the classification model. This model will determine if a User Equipment (UE) can become a 5G Virtual Base Station (VBS) or not.\n",
    "\n",
    "Target: **Eligibility**\n",
    "\n",
    "- <strong>Environment setup:</strong> import libraries and load the dataset\n",
    "- <strong>Data Analysis:</strong> understand the meaning and the predictive power of the variables \n",
    "- <strong>Feature Selection:</strong> evaluate and keep only the most relevant variables\n",
    "- <strong>Preprocessing:</strong> encode categorical variables and partition the dataset for training and testing\n",
    "- <strong>Model design:</strong> train and test various models\n",
    "- <strong>Performance evaluation:</strong> evaluate the models based on their metrics\n",
    "- <strong>Explainability:</strong> understand how the model produces results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Environmnet Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "jl27hA_Ucd9N"
   },
   "outputs": [],
   "source": [
    "# Pre-requisites\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import missingno as msno\n",
    "\n",
    "# Store the classifier models to save time\n",
    "import joblib\n",
    "\n",
    "# Preprocessing\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Classifiers from scikit-learn\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "\n",
    "# Performance metrics\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix\n",
    "# from sklearn.metrics import ConfusionMatrixDisplay\n",
    "from sklearn.metrics import plot_confusion_matrix\n",
    "\n",
    "from sklearn.metrics import roc_curve\n",
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "# Model Selection\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "from scipy.stats import pointbiserialr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "0HA4SPpocpEP"
   },
   "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>x</th>\n",
       "      <th>y</th>\n",
       "      <th>distance_from_base_station</th>\n",
       "      <th>battery_discharge_rate</th>\n",
       "      <th>power</th>\n",
       "      <th>memory</th>\n",
       "      <th>cpu_power</th>\n",
       "      <th>qos</th>\n",
       "      <th>qoe</th>\n",
       "      <th>data_rate</th>\n",
       "      <th>eligibility</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>134.467967</td>\n",
       "      <td>211.647510</td>\n",
       "      <td>465.575800</td>\n",
       "      <td>45.312870</td>\n",
       "      <td>9.301212</td>\n",
       "      <td>6</td>\n",
       "      <td>65.953269</td>\n",
       "      <td>0.992467</td>\n",
       "      <td>0.147388</td>\n",
       "      <td>3.237754e+04</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>583.456900</td>\n",
       "      <td>125.555800</td>\n",
       "      <td>383.632002</td>\n",
       "      <td>107.164485</td>\n",
       "      <td>10.982724</td>\n",
       "      <td>12</td>\n",
       "      <td>48.639950</td>\n",
       "      <td>0.993799</td>\n",
       "      <td>0.136998</td>\n",
       "      <td>4.545044e+05</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>367.394058</td>\n",
       "      <td>380.586430</td>\n",
       "      <td>178.448694</td>\n",
       "      <td>42.099486</td>\n",
       "      <td>17.630753</td>\n",
       "      <td>6</td>\n",
       "      <td>88.386756</td>\n",
       "      <td>0.995114</td>\n",
       "      <td>0.125348</td>\n",
       "      <td>4.774090e+04</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>263.133339</td>\n",
       "      <td>37.844211</td>\n",
       "      <td>519.320506</td>\n",
       "      <td>117.518415</td>\n",
       "      <td>8.352311</td>\n",
       "      <td>8</td>\n",
       "      <td>93.333433</td>\n",
       "      <td>0.992822</td>\n",
       "      <td>0.143667</td>\n",
       "      <td>1.416554e+06</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>61.528950</td>\n",
       "      <td>28.579757</td>\n",
       "      <td>643.812013</td>\n",
       "      <td>67.063529</td>\n",
       "      <td>6.485839</td>\n",
       "      <td>4</td>\n",
       "      <td>30.084433</td>\n",
       "      <td>0.992436</td>\n",
       "      <td>0.148444</td>\n",
       "      <td>5.775633e+04</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            x           y  distance_from_base_station  battery_discharge_rate  \\\n",
       "0  134.467967  211.647510                  465.575800               45.312870   \n",
       "1  583.456900  125.555800                  383.632002              107.164485   \n",
       "2  367.394058  380.586430                  178.448694               42.099486   \n",
       "3  263.133339   37.844211                  519.320506              117.518415   \n",
       "4   61.528950   28.579757                  643.812013               67.063529   \n",
       "\n",
       "       power  memory  cpu_power       qos       qoe     data_rate  eligibility  \n",
       "0   9.301212       6  65.953269  0.992467  0.147388  3.237754e+04        False  \n",
       "1  10.982724      12  48.639950  0.993799  0.136998  4.545044e+05        False  \n",
       "2  17.630753       6  88.386756  0.995114  0.125348  4.774090e+04        False  \n",
       "3   8.352311       8  93.333433  0.992822  0.143667  1.416554e+06        False  \n",
       "4   6.485839       4  30.084433  0.992436  0.148444  5.775633e+04        False  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Loading the dataset\n",
    "dataset_location = './Data/master_simulation_1000_2500.csv'\n",
    "df = pd.read_csv(dataset_location, index_col=0)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AxesSubplot(0.125,0.125;0.698618x0.755)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(msno.matrix(df))"
   ]
  },
  {
   "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>distance_from_base_station</th>\n",
       "      <th>battery_discharge_rate</th>\n",
       "      <th>power</th>\n",
       "      <th>memory</th>\n",
       "      <th>cpu_power</th>\n",
       "      <th>qos</th>\n",
       "      <th>qoe</th>\n",
       "      <th>data_rate</th>\n",
       "      <th>eligibility</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>465.575800</td>\n",
       "      <td>45.312870</td>\n",
       "      <td>9.301212</td>\n",
       "      <td>6</td>\n",
       "      <td>65.953269</td>\n",
       "      <td>0.992467</td>\n",
       "      <td>0.147388</td>\n",
       "      <td>3.237754e+04</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>383.632002</td>\n",
       "      <td>107.164485</td>\n",
       "      <td>10.982724</td>\n",
       "      <td>12</td>\n",
       "      <td>48.639950</td>\n",
       "      <td>0.993799</td>\n",
       "      <td>0.136998</td>\n",
       "      <td>4.545044e+05</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>178.448694</td>\n",
       "      <td>42.099486</td>\n",
       "      <td>17.630753</td>\n",
       "      <td>6</td>\n",
       "      <td>88.386756</td>\n",
       "      <td>0.995114</td>\n",
       "      <td>0.125348</td>\n",
       "      <td>4.774090e+04</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>519.320506</td>\n",
       "      <td>117.518415</td>\n",
       "      <td>8.352311</td>\n",
       "      <td>8</td>\n",
       "      <td>93.333433</td>\n",
       "      <td>0.992822</td>\n",
       "      <td>0.143667</td>\n",
       "      <td>1.416554e+06</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>643.812013</td>\n",
       "      <td>67.063529</td>\n",
       "      <td>6.485839</td>\n",
       "      <td>4</td>\n",
       "      <td>30.084433</td>\n",
       "      <td>0.992436</td>\n",
       "      <td>0.148444</td>\n",
       "      <td>5.775633e+04</td>\n",
       "      <td>False</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5245</th>\n",
       "      <td>335.201631</td>\n",
       "      <td>42.854214</td>\n",
       "      <td>11.942236</td>\n",
       "      <td>6</td>\n",
       "      <td>14.992929</td>\n",
       "      <td>0.996490</td>\n",
       "      <td>0.095631</td>\n",
       "      <td>1.126151e+05</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5246</th>\n",
       "      <td>399.576249</td>\n",
       "      <td>98.270340</td>\n",
       "      <td>10.416365</td>\n",
       "      <td>8</td>\n",
       "      <td>98.328262</td>\n",
       "      <td>0.996259</td>\n",
       "      <td>0.098850</td>\n",
       "      <td>2.371739e+06</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5247</th>\n",
       "      <td>214.765611</td>\n",
       "      <td>53.136700</td>\n",
       "      <td>15.809064</td>\n",
       "      <td>12</td>\n",
       "      <td>55.446848</td>\n",
       "      <td>0.997809</td>\n",
       "      <td>0.076410</td>\n",
       "      <td>1.069651e+05</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5248</th>\n",
       "      <td>275.915292</td>\n",
       "      <td>85.409301</td>\n",
       "      <td>13.632843</td>\n",
       "      <td>12</td>\n",
       "      <td>53.260666</td>\n",
       "      <td>0.996925</td>\n",
       "      <td>0.090078</td>\n",
       "      <td>2.350156e+05</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5249</th>\n",
       "      <td>334.955376</td>\n",
       "      <td>42.516774</td>\n",
       "      <td>11.948620</td>\n",
       "      <td>12</td>\n",
       "      <td>70.195530</td>\n",
       "      <td>0.996694</td>\n",
       "      <td>0.091915</td>\n",
       "      <td>3.766707e+06</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5250 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      distance_from_base_station  battery_discharge_rate      power  memory  \\\n",
       "0                     465.575800               45.312870   9.301212       6   \n",
       "1                     383.632002              107.164485  10.982724      12   \n",
       "2                     178.448694               42.099486  17.630753       6   \n",
       "3                     519.320506              117.518415   8.352311       8   \n",
       "4                     643.812013               67.063529   6.485839       4   \n",
       "...                          ...                     ...        ...     ...   \n",
       "5245                  335.201631               42.854214  11.942236       6   \n",
       "5246                  399.576249               98.270340  10.416365       8   \n",
       "5247                  214.765611               53.136700  15.809064      12   \n",
       "5248                  275.915292               85.409301  13.632843      12   \n",
       "5249                  334.955376               42.516774  11.948620      12   \n",
       "\n",
       "      cpu_power       qos       qoe     data_rate  eligibility  \n",
       "0     65.953269  0.992467  0.147388  3.237754e+04        False  \n",
       "1     48.639950  0.993799  0.136998  4.545044e+05        False  \n",
       "2     88.386756  0.995114  0.125348  4.774090e+04        False  \n",
       "3     93.333433  0.992822  0.143667  1.416554e+06        False  \n",
       "4     30.084433  0.992436  0.148444  5.775633e+04        False  \n",
       "...         ...       ...       ...           ...          ...  \n",
       "5245  14.992929  0.996490  0.095631  1.126151e+05        False  \n",
       "5246  98.328262  0.996259  0.098850  2.371739e+06        False  \n",
       "5247  55.446848  0.997809  0.076410  1.069651e+05        False  \n",
       "5248  53.260666  0.996925  0.090078  2.350156e+05        False  \n",
       "5249  70.195530  0.996694  0.091915  3.766707e+06        False  \n",
       "\n",
       "[5250 rows x 9 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.drop(['x', 'y'], axis=1)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 5250 entries, 0 to 5249\n",
      "Data columns (total 9 columns):\n",
      " #   Column                      Non-Null Count  Dtype  \n",
      "---  ------                      --------------  -----  \n",
      " 0   distance_from_base_station  5250 non-null   float64\n",
      " 1   battery_discharge_rate      5250 non-null   float64\n",
      " 2   power                       5250 non-null   float64\n",
      " 3   memory                      5250 non-null   int64  \n",
      " 4   cpu_power                   5250 non-null   float64\n",
      " 5   qos                         5250 non-null   float64\n",
      " 6   qoe                         5250 non-null   float64\n",
      " 7   data_rate                   5250 non-null   float64\n",
      " 8   eligibility                 5250 non-null   bool   \n",
      "dtypes: bool(1), float64(7), int64(1)\n",
      "memory usage: 374.3 KB\n",
      "None\n",
      "\n",
      "Shape: (5250, 9)\n"
     ]
    }
   ],
   "source": [
    "# Understanding the column data\n",
    "print(df.info())\n",
    "print(f\"\\nShape: {df.shape}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "False    4243\n",
      "True     1007\n",
      "Name: eligibility, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Plotting the percentage of observations that fall under each class\n",
    "ax = df[\"eligibility\"].value_counts().sort_values().plot(kind=\"barh\", color=[\"r\", \"g\"])\n",
    "totals= []\n",
    "for i in ax.patches:\n",
    "    totals.append(i.get_width())\n",
    "total = sum(totals)\n",
    "for i in ax.patches:\n",
    "     ax.text(i.get_width()+.3, i.get_y()+.20, \n",
    "     str(round((i.get_width()/total)*100, 2))+'%', \n",
    "     fontsize=10, color='black')\n",
    "plt.title(\"Eligibility\", fontsize=20)\n",
    "plt.xlabel(\"Count\", fontsize=14)\n",
    "plt.ylabel(\"Class\", fontsize=14)\n",
    "plt.show()\n",
    "print(df[\"eligibility\"].value_counts())\n",
    "fig = ax.get_figure()\n",
    "fig.savefig(\"Fig/03-01-Class_Imbalance.png\")\n",
    "fig.savefig(\"Eps/03-01-Class_Imbalance.eps\", format='eps')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vnJInJOBCwW9"
   },
   "source": [
    "### Point Biserial Correlation for Quantitative, Continuous Features with ArrDel15\n",
    "A <strong>correlation coefficient</strong> measures the extent to which two variables tend to change together. The coefficient describes both the strength and the direction of the relationship. <br>\n",
    "A <strong>point-biserial correlation (PBC)</strong> is used to measure the strength and direction of the association that exists between <strong>one continuous variable and one dichotomous variable</strong>."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>Feature</th>\n",
       "      <th>CorrCoeff</th>\n",
       "      <th>pValue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>data_rate</td>\n",
       "      <td>0.227157</td>\n",
       "      <td>2.038732e-62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>qoe</td>\n",
       "      <td>0.207547</td>\n",
       "      <td>3.511968e-52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>cpu_power</td>\n",
       "      <td>0.118286</td>\n",
       "      <td>8.057369e-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>battery_discharge_rate</td>\n",
       "      <td>0.102077</td>\n",
       "      <td>1.227105e-13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>power</td>\n",
       "      <td>0.031623</td>\n",
       "      <td>2.194416e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>distance_from_base_station</td>\n",
       "      <td>-0.026995</td>\n",
       "      <td>5.048069e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>qos</td>\n",
       "      <td>-0.184563</td>\n",
       "      <td>1.898806e-41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Feature  CorrCoeff        pValue\n",
       "0                   data_rate   0.227157  2.038732e-62\n",
       "1                         qoe   0.207547  3.511968e-52\n",
       "2                   cpu_power   0.118286  8.057369e-18\n",
       "3      battery_discharge_rate   0.102077  1.227105e-13\n",
       "4                       power   0.031623  2.194416e-02\n",
       "5  distance_from_base_station  -0.026995  5.048069e-02\n",
       "6                         qos  -0.184563  1.898806e-41"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Encoding eligibility : categorical -> numeric\n",
    "labelEncoder = LabelEncoder()\n",
    "df[\"eligibility\"] = labelEncoder.fit_transform(df[\"eligibility\"])\n",
    "df[\"eligibility\"].head()\n",
    "\n",
    "# get continuous and dichotomous data\n",
    "categorical = [\"memory\", \"eligibility\"]\n",
    "numeric = [\"distance_from_base_station\", \"battery_discharge_rate\", \"power\", \"cpu_power\", \"qos\", \"qoe\", \"data_rate\"]\n",
    "target = df[\"eligibility\"]\n",
    "\n",
    "# pbc of first question\n",
    "pbc = list()\n",
    "for col in numeric:\n",
    "    ans = pointbiserialr(df[col], target)\n",
    "    pbc.append([col, ans[0], ans[1]])\n",
    "    \n",
    "pbc_corr = pd.DataFrame(pbc, columns=[\"Feature\", \"CorrCoeff\", \"pValue\"]).sort_values(by=\"CorrCoeff\", ascending=False).reset_index(drop=True)\n",
    "pbc_corr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(7, 5))\n",
    "pbc_corr = pbc_corr.set_index(\"Feature\")\n",
    "heatmap = sns.heatmap(pbc_corr[[\"CorrCoeff\"]].sort_values(by=\"CorrCoeff\", ascending=False), vmin=-1, vmax=1, annot=True, cmap=\"BrBG\")\n",
    "heatmap.set_title(\"PBC with Eligibility\", fontdict={\"fontsize\":18}, pad=16);\n",
    "plt.savefig(\"Fig/03-02-PBC.png\")\n",
    "plt.savefig(\"Eps/03-02-PBC.eps\", format='eps')\n",
    "del pbc_corr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Spearman Correlation for Monotonic Relationships with Categorical Features\n",
    "The <strong>Spearman correlation</strong> evaluates the monotonic relationship between <strong>two continuous or ordinal variables</strong>. In a monotonic relationship, the variables tend to change together, but not necessarily at a constant rate. The Spearman correlation coefficient is based on the ranked values for each variable rather than the raw data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(7, 5))\n",
    "heatmap = sns.heatmap(df[categorical].corr(method=\"spearman\")[[\"eligibility\"]].sort_values(by=\"eligibility\", ascending=False), vmin=-1, vmax=1, annot=True, cmap=\"BrBG\")\n",
    "heatmap.set_title(\"Spearman Correlations with Eligibility\", fontdict={\"fontsize\":18}, pad=16)\n",
    "plt.savefig(\"Fig/03-03-Spearman.png\")\n",
    "plt.savefig(\"Eps/03-03-Spearman.eps\", format='eps')\n",
    "del heatmap"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing for Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['distance_from_base_station', 'battery_discharge_rate', 'power',\n",
      "       'memory', 'cpu_power', 'qos', 'qoe', 'data_rate'],\n",
      "      dtype='object')\n",
      "\n",
      "Dataset shape: (5250, 9)\n",
      "features_train shape: (3675, 8) | features_test shape: (1575, 8)\n",
      "labels_train shape: (3675,) | labels_test shape: (1575,)\n"
     ]
    }
   ],
   "source": [
    "# Encoding eligibility : categorical -> numeric\n",
    "labelEncoder = LabelEncoder()\n",
    "df[\"eligibility\"] = labelEncoder.fit_transform(df[\"eligibility\"])\n",
    "features = df.loc[:, df.columns != \"eligibility\"]\n",
    "labels = np.asarray(df[\"eligibility\"])\n",
    "print(features.columns)\n",
    "\n",
    "# Number of samples/observations/rows is greater than 100,000\n",
    "print(f\"\\nDataset shape: {df.shape}\")\n",
    "features_train, features_test, labels_train, labels_test = train_test_split(features, labels, test_size=0.30, random_state=42)\n",
    "print(f\"features_train shape: {features_train.shape} | features_test shape: {features_test.shape}\")\n",
    "print(f\"labels_train shape: {labels_train.shape} | labels_test shape: {labels_test.shape}\")\n",
    "# print(f\"{features_train.shape[1]} Features: {features_train.columns.to_list()}\")\n",
    "df.to_csv(\"./Data/encoded_master_simulation_1000_2500.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Selection\n",
    "A random forest classifier has been modeled to rank the importance of the features in the entire dataset to visualize which features provide relevant information.\n",
    "<strong>Note: Uncomment the below code if you want to re-train the model</strong>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.04462437, 0.06861646, 0.04523226, 0.01426918, 0.07306929,\n",
       "       0.11468288, 0.11442689, 0.52507867])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = features_train\n",
    "y = labels_train\n",
    "\n",
    "model = RandomForestClassifier(n_estimators=100,\n",
    "                      criterion=\"entropy\", random_state=42, n_jobs=-1)\n",
    "\n",
    "model.fit(X, y)\n",
    "joblib.dump(model, \"./Data/feature_importance_model.joblib\")\n",
    "model = joblib.load(\"./Data/feature_importance_model.joblib\")\n",
    "importances = model.feature_importances_\n",
    "importances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "importances = pd.DataFrame({\n",
    "    \"Feature\": list(X.columns),\n",
    "    \"Importance\": model.feature_importances_\n",
    "})\n",
    "importances = importances.sort_values(by=\"Importance\", ascending=False)\n",
    "importances = importances.set_index(\"Feature\")\n",
    "importances\n",
    "plt.figure(figsize=(8, 6), dpi=80)\n",
    "plt.barh(importances.index, importances.Importance)\n",
    "plt.title(\"Feature Importance Ranking obtained from Random Forest Classifier\", fontsize=12)\n",
    "plt.xlabel(\"Importances\")\n",
    "plt.ylabel(\"Features\")\n",
    "plt.savefig(\"Fig/03-04-Feature_Selection.png\")\n",
    "plt.savefig(\"Eps/03-04-Feature_Selection.eps\", format='eps')\n",
    "del importances\n",
    "del model\n",
    "del X\n",
    "del y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modeling Different Classifiers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1261   20]\n",
      " [  59  235]]\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.98      0.97      1281\n",
      "           1       0.92      0.80      0.86       294\n",
      "\n",
      "    accuracy                           0.95      1575\n",
      "   macro avg       0.94      0.89      0.91      1575\n",
      "weighted avg       0.95      0.95      0.95      1575\n",
      "\n",
      "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay object at 0x0000028AFF942DA0>\n",
      "\n",
      "Cross-Validation cv=5\n",
      "0.94 0.93 0.93 0.93 0.95 \n",
      "\n",
      "Mean: 0.94\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = RandomForestClassifier(n_jobs=-1, random_state=42)\n",
    "model.fit(features_train, labels_train)\n",
    "joblib.dump(model, \"./Classifiers/RandomForestClassifier.joblib\")\n",
    "model = joblib.load(\"./Classifiers/RandomForestClassifier.joblib\")\n",
    "model_pred = model.predict(features_test)\n",
    "print(confusion_matrix(labels_test, model_pred))\n",
    "print(classification_report(labels_test, model_pred))\n",
    "# conf_mat_plot = ConfusionMatrixDisplay.from_estimator(model, features_test, labels_test)\n",
    "conf_mat_plot = plot_confusion_matrix(model, features_test, labels_test)\n",
    "plt.title(\"Random Forest Classifier\")\n",
    "plt.savefig(\"Fig/03-05-RFC_Confusion_Matrix.png\")\n",
    "plt.savefig(\"Eps/03-05-RFC_Confusion_Matrix.eps\", format='eps')\n",
    "print(conf_mat_plot)\n",
    "print(\"\\nCross-Validation cv=5\")\n",
    "scores = cross_val_score(model, features_train, labels_train, cv=5)\n",
    "for score in scores:\n",
    "    print(\"%.2f\" %score, end=\" \")\n",
    "print(\"\\n\\nMean: %.2f\" %scores.mean())\n",
    "del model\n",
    "del model_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Gradient Boosting Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.99      0.97      1281\n",
      "           1       0.93      0.81      0.86       294\n",
      "\n",
      "    accuracy                           0.95      1575\n",
      "   macro avg       0.94      0.90      0.92      1575\n",
      "weighted avg       0.95      0.95      0.95      1575\n",
      "\n",
      "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay object at 0x0000028A83559AC8>\n",
      "\n",
      "Cross-Validation cv=5\n",
      "0.95 0.93 0.95 0.94 0.94 \n",
      "\n",
      "Mean: 0.94\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = GradientBoostingClassifier(random_state=42)\n",
    "model.fit(features_train, labels_train)\n",
    "joblib.dump(model, \"./Classifiers/GradientBoostingClassifier.joblib\")\n",
    "model = joblib.load(\"./Classifiers/GradientBoostingClassifier.joblib\")\n",
    "model_pred = model.predict(features_test)\n",
    "# print(confusion_matrix(labels_test, model_pred))\n",
    "print(classification_report(labels_test, model_pred))\n",
    "# conf_mat_plot = ConfusionMatrixDisplay.from_estimator(model, features_test, labels_test)\n",
    "conf_mat_plot = plot_confusion_matrix(model, features_test, labels_test)\n",
    "plt.title(\"Gradient Boosting Classifier\")\n",
    "plt.savefig(\"Fig/03-06-GBC_Confusion_Matrix.png\")\n",
    "plt.savefig(\"Eps/03-06-GBC_Confusion_Matrix.eps\", format='eps')\n",
    "print(conf_mat_plot)\n",
    "print(\"\\nCross-Validation cv=5\")\n",
    "scores = cross_val_score(model, features_train, labels_train, cv=5)\n",
    "for score in scores:\n",
    "    print(\"%.2f\" %score, end=\" \")\n",
    "print(\"\\n\\nMean: %.2f\" %scores.mean())\n",
    "del model\n",
    "del model_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### AdaBoost Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.99      0.97      1281\n",
      "           1       0.93      0.79      0.85       294\n",
      "\n",
      "    accuracy                           0.95      1575\n",
      "   macro avg       0.94      0.89      0.91      1575\n",
      "weighted avg       0.95      0.95      0.95      1575\n",
      "\n",
      "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay object at 0x0000028A80D4F7F0>\n",
      "\n",
      "Cross-Validation cv=5\n",
      "0.94 0.93 0.94 0.94 0.94 \n",
      "\n",
      "Mean: 0.94\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = AdaBoostClassifier(random_state=42)\n",
    "model.fit(features_train, labels_train)\n",
    "joblib.dump(model, \"./Classifiers/AdaBoostClassifier.joblib\")\n",
    "model = joblib.load(\"./Classifiers/AdaBoostClassifier.joblib\")\n",
    "model_pred = model.predict(features_test)\n",
    "# print(confusion_matrix(labels_test, model_pred))\n",
    "print(classification_report(labels_test, model_pred))\n",
    "# conf_mat_plot = ConfusionMatrixDisplay.from_estimator(model, features_test, labels_test)\n",
    "conf_mat_plot = plot_confusion_matrix(model, features_test, labels_test)\n",
    "plt.title(\"AdaBoost Classifier\")\n",
    "plt.savefig(\"Fig/03-07-AdaBoost_Confusion_Matrix.png\")\n",
    "plt.savefig(\"Eps/03-07-AdaBoost_Confusion_Matrix.eps\", format='eps')\n",
    "print(conf_mat_plot)\n",
    "print(\"\\nCross-Validation cv=5\")\n",
    "scores = cross_val_score(model, features_train, labels_train, cv=5)\n",
    "for score in scores:\n",
    "    print(\"%.2f\" %score, end=\" \")\n",
    "print(\"\\n\\nMean: %.2f\" %scores.mean())\n",
    "del model\n",
    "del model_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Performance Evaluation\n",
    "The <strong>Receiver Operator Characteristic (ROC) curve</strong> is an evaluation metric for binary classification problems. It is a probability curve that plots the TPR against FPR at various threshold values and essentially separates the ‘signal’ from the ‘noise’. The <strong>Area Under the Curve (AUC)</strong> is the measure of the ability of a classifier to distinguish between classes and is used as a summary of the ROC curve.\n",
    "<br><br>\n",
    "**The higher the AUC, the better the performance of the model at distinguishing between the positive and negative classes.**\n",
    "\n",
    "|AUC Value|Meaning|\n",
    "|:--|:--|\n",
    "|AUC = 1| The classifier is able to **perfectly distinguish** between all the Positive and the Negative class points correctly.|\n",
    "|AUC = 0| The classifier would be predicting **all Negatives as Positives, and all Positives as Negatives**|\n",
    "|0.5 < AUC < 1| There is a high chance that the classifier will be able to distinguish the classes i.e. more numbers of True positives and True negatives than False negatives and False positives|\n",
    "|AUC=0.5| The classifier is not able to distinguish between Positive and Negative class points i.e. either the classifier is predicting random class or constant class for all the data points|"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setup\n",
    "perf_df = pd.DataFrame(columns=[\"classifiers\", \"fpr\", \"tpr\", \"auc\"])\n",
    "\n",
    "classifiers = [\"AdaBoostClassifier\", \"GradientBoostingClassifier\", \"RandomForestClassifier\"]\n",
    "for clf in classifiers:\n",
    "    # Load the trained classifiers\n",
    "    model = joblib.load(f\"./Classifiers/{clf}.joblib\")\n",
    "    model_pred = model.predict(features_test)\n",
    "    fpr, tpr, _ = roc_curve(labels_test,  model_pred)\n",
    "    auc = roc_auc_score(labels_test, model_pred)\n",
    "    # Record the results\n",
    "    perf_df = perf_df.append({\"classifiers\": clf,\n",
    "                                        \"fpr\": fpr, \n",
    "                                        \"tpr\": tpr, \n",
    "                                        \"auc\": auc}, ignore_index=True)\n",
    "    del model\n",
    "    del model_pred\n",
    "\n",
    "# Set name of the classifiers as index labels\n",
    "perf_df.set_index(\"classifiers\", inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n",
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(8,6), dpi=80)\n",
    "for clf_name in perf_df.index:\n",
    "    plt.plot(perf_df.loc[clf_name][\"fpr\"], \n",
    "             perf_df.loc[clf_name][\"tpr\"], \n",
    "             label=\"{}, AUC={:.2f}\".format(clf_name, perf_df.loc[clf_name][\"auc\"]))\n",
    "    \n",
    "plt.plot([0,1], [0,1], color='red', linestyle='--')\n",
    "\n",
    "plt.xticks(np.arange(0.0, 1.1, step=0.1))\n",
    "plt.xlabel(\"False Positive Rate\", fontsize=15)\n",
    "\n",
    "plt.yticks(np.arange(0.0, 1.1, step=0.1))\n",
    "plt.ylabel(\"True Positive Rate\", fontsize=15)\n",
    "\n",
    "plt.title(\"ROC AUC Analysis\", fontweight=\"bold\", fontsize=15)\n",
    "plt.legend(prop={\"size\":13}, loc=\"lower right\")\n",
    "\n",
    "plt.savefig(\"Fig/03-08-ROCAUC_Before_SMOTE.png\")\n",
    "plt.savefig(\"Eps/03-08-ROCAUC_Before_SMOTE.eps\", format='eps')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n",
      "The PostScript backend does not support transparency; partially transparent artists will be rendered opaque.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "markers = 'oD*'\n",
    "m = 0\n",
    "fig = plt.figure(figsize=(8,6), dpi=80)\n",
    "for clf_name in perf_df.index:\n",
    "    plt.plot(perf_df.loc[clf_name][\"fpr\"], \n",
    "             perf_df.loc[clf_name][\"tpr\"], marker=markers[m], markersize='5',\n",
    "             label=\"{}, AUC={:.2f}\".format(clf_name, perf_df.loc[clf_name][\"auc\"]))\n",
    "    m += 1\n",
    "    \n",
    "plt.plot([0,1], [0,1], color='red', linestyle='--')\n",
    "\n",
    "plt.xticks(np.arange(0.0, 1.1, step=0.1))\n",
    "plt.xlabel(\"False Ineligible Rate (FIR)\", fontsize=15)\n",
    "\n",
    "plt.yticks(np.arange(0.0, 1.1, step=0.1))\n",
    "plt.ylabel(\"True Eligible Rate (TER)\", fontsize=15)\n",
    "\n",
    "plt.title(\"ROC AUC Analysis\", fontweight=\"bold\", fontsize=15)\n",
    "plt.legend(prop={\"size\":13}, loc=\"lower right\")\n",
    "\n",
    "plt.savefig(\"Fig/03-09-ROCAUC_Before_SMOTE.png\")\n",
    "plt.savefig(\"Eps/03-09-ROCAUC_Before_SMOTE.eps\", format='eps')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "02-Classification.ipynb",
   "provenance": []
  },
  "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": "3.6.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}