Assignment 07 - Data Analytics Project.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Assignment 07 - Data Analytics Project\n",
    "\n",
    "**Due: Fri 12/13/2019**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Please fill these in before submitting, just in case I accidentally mix up file names while grading**:\n",
    "\n",
    "Name: Stephanie Duckworth\n",
    "\n",
    "CWID-5: 33881"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import urllib.request # Request Library for Downloading URLs\n",
    "import os.path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "matplotlib.rcParams['figure.figsize'] = (12, 10) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook uses the QSAR Biodegradation Data Set found at https://archive.ics.uci.edu/ml/datasets/QSAR+biodegradation.\n",
    "\n",
    "1055 chemicals are classified into two categories: Ready Biodegradable (RB) and Non-Ready Biodegradable (NRB). 41 molecular features are used to describe each instance of a chemical. The following table shows the names of the features and a brief description. Through out this journal the columns will be referenced via the index for convience. The goal of this notebook is to examine the data and develop a model that will predict whether a chemical is a of the RB or NRB class. \n",
    "\n",
    "\n",
    "|\tFeatures Number\t|\tAbbreviation\t|\tDescription\t|\n",
    "|-------------------|-------------------|---------------|\n",
    "|\t0\t|\tSpMax_L\t|\tLeading eigenvalue from Laplace matrix\t|\n",
    "|\t1\t|\tJ_Dz(e)\t|\tBalaban-like index from Barysz matrix weighted by Sanderson electronegativity\t|\n",
    "|\t2\t|\tnHM\t|\tNumber of heavy atoms\t|\n",
    "|\t3\t|\tF01[N-N]\t|\tFrequency of N-N at topological distance 1\t|\n",
    "|\t4\t|\tF04[C-N]\t|\tFrequency of C-N at topological distance 4\t|\n",
    "|\t5\t|\tNssssC\t|\tNumber of atoms of type ssssC\t|\n",
    "|\t6\t|\tnCb-\t|\tNumber of substituted benzene C(sp2)\t|\n",
    "|\t7\t|\tC%\t|\tPercentage of C atoms\t|\n",
    "|\t8\t|\tnCp\t|\tNumber of terminal primary C(sp3)\t|\n",
    "|\t9\t|\tnO\t|\tNumber of oxygen atoms\t|\n",
    "|\t10\t|\tF03[C-N]\t|\tFrequency of C-N at topological distance 3\t|\n",
    "|\t11\t|\tSdssC\t|\tSum of dssC E-states\t|\n",
    "|\t12\t|\tHyWi_B(m)\t|\tHyper-Wiener-like index (log function) from Burden matrix weighted by mass\t|\n",
    "|\t13\t|\tLOC\t|\tLopping centric index\t|\n",
    "|\t14\t|\tSM6_L\t|\tSpectral moment of order 6 from Laplace matrix\t|\n",
    "|\t15\t|\tF03[C-O]\t|\tFrequency of C - O at topological distance 3\t|\n",
    "|\t16\t|\tMe\t|\tMean atomic Sanderson electronegativity (scaled on Carbon atom)\t|\n",
    "|\t17\t|\tMi\t|\tMean first ionization potential (scaled on Carbon atom)\t|\n",
    "|\t18\t|\tnN-N\t|\tNumber of N hydrazines\t|\n",
    "|\t19\t|\tnArNO2\t|\tNumber of nitro groups (aromatic)\t|\n",
    "|\t20\t|\tnCRX3\t|\tNumber of CRX3\t|\n",
    "|\t21\t|\tSpPosA_B(p)\t|\tNormalized spectral positive sum from Burden matrix weighted by polarizability\t|\n",
    "|\t22\t|\tnCIR\t|\tNumber of circuits\t|\n",
    "|\t23\t|\tB01[C-Br]\t|\tPresence/absence of C - Br at topological distance 1\t|\n",
    "|\t24\t|\tB03[C-Cl]\t|\tPresence/absence of C - Cl at topological distance 3\t|\n",
    "|\t25\t|\tN-073\t|\tAr2NH / Ar3N / Ar2N-Al / R..N..R\t|\n",
    "|\t26\t|\tSpMax_A\t|\tLeading eigenvalue from adjacency matrix (Lovasz-Pelikan index)\t|\n",
    "|\t27\t|\tPsi_i_1d\t|\tIntrinsic state pseudoconnectivity index - type 1d\t|\n",
    "|\t28\t|\tB04[C-Br]\t|\tPresence/absence of C - Br at topological distance 4\t|\n",
    "|\t29\t|\tSdO\t|\tSum of dO E-states\t|\n",
    "|\t30\t|\tTI2_L\t|\tSecond Mohar index from Laplace matrix\t|\n",
    "|\t31\t|\tnCrt\t|\tNumber of ring tertiary C(sp3)\t|\n",
    "|\t32\t|\tC-026\t|\tR--CX--R\t|\n",
    "|\t33\t|\tF02[C-N]\t|\tFrequency of C - N at topological distance 2\t|\n",
    "|\t34\t|\tnHDon\t|\tNumber of donor atoms for H-bonds (N and O)\t|\n",
    "|\t35\t|\tSpMax_B(m)\t|\tLeading eigenvalue from Burden matrix weighted by mass\t|\n",
    "|\t36\t|\tPsi_i_A\t|\tIntrinsic state pseudoconnectivity index - type S average\t|\n",
    "|\t37\t|\tnN\t|\tNumber of Nitrogen atoms\t|\n",
    "|\t38\t|\tSM6_B(m)\t|\tSpectral moment of order 6 from Burden matrix weighted by mass\t|\n",
    "|\t39\t|\tnArCOOR\t|\tNumber of esters (aromatic)\t|\n",
    "|\t40\t|\tnX\t|\tNumber of halogen atoms\t|\n",
    "|\t41\t|\tExperimental Class\t|\tReady Biodegradable (RB) and Not Ready Biodegradable (NRB)\t|\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Exploration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def urlretrieve_reporthook(block_number, read_size, total_file_size):\n",
    "    if block_number % 100 == 0:\n",
    "        print(\"\\rReading %d / %d complete\" % (read_size * block_number, total_file_size), end=\"\")\n",
    "\n",
    "data_url = 'https://archive.ics.uci.edu/ml/machine-learning-databases/00254/biodeg.csv'\n",
    "data_file = './data/biodeg.csv'\n",
    "\n",
    "if not os.path.exists(data_file):\n",
    "    print('Beginning file download with urllib2...')\n",
    "    urllib.request.urlretrieve(data_url, data_file, reporthook=urlretrieve_reporthook)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('./data/biodeg.csv', sep=';', header=None)"
   ]
  },
  {
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>40</th>\n",
       "      <th>41</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3.919</td>\n",
       "      <td>2.6909</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>31.4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.949</td>\n",
       "      <td>1.591</td>\n",
       "      <td>0</td>\n",
       "      <td>7.253</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>RB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.170</td>\n",
       "      <td>2.1144</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>30.8</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.315</td>\n",
       "      <td>1.967</td>\n",
       "      <td>0</td>\n",
       "      <td>7.257</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>RB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.932</td>\n",
       "      <td>3.2512</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>26.7</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3.076</td>\n",
       "      <td>2.417</td>\n",
       "      <td>0</td>\n",
       "      <td>7.601</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>RB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.000</td>\n",
       "      <td>2.7098</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3.046</td>\n",
       "      <td>5.000</td>\n",
       "      <td>0</td>\n",
       "      <td>6.690</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>RB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4.236</td>\n",
       "      <td>3.3944</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>29.4</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.351</td>\n",
       "      <td>2.405</td>\n",
       "      <td>0</td>\n",
       "      <td>8.003</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>RB</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 42 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      0       1   2   3   4   5   6     7   8   9  ...  32  33  34     35  \\\n",
       "0  3.919  2.6909   0   0   0   0   0  31.4   2   0 ...   0   0   0  2.949   \n",
       "1  4.170  2.1144   0   0   0   0   0  30.8   1   1 ...   0   0   0  3.315   \n",
       "2  3.932  3.2512   0   0   0   0   0  26.7   2   4 ...   0   0   1  3.076   \n",
       "3  3.000  2.7098   0   0   0   0   0  20.0   0   2 ...   0   0   1  3.046   \n",
       "4  4.236  3.3944   0   0   0   0   0  29.4   2   4 ...   0   0   0  3.351   \n",
       "\n",
       "      36  37     38  39  40  41  \n",
       "0  1.591   0  7.253   0   0  RB  \n",
       "1  1.967   0  7.257   0   0  RB  \n",
       "2  2.417   0  7.601   0   0  RB  \n",
       "3  5.000   0  6.690   0   0  RB  \n",
       "4  2.405   0  8.003   0   0  RB  \n",
       "\n",
       "[5 rows x 42 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>40</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "      <td>1055.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.782644</td>\n",
       "      <td>3.069508</td>\n",
       "      <td>0.716588</td>\n",
       "      <td>0.042654</td>\n",
       "      <td>0.980095</td>\n",
       "      <td>0.290047</td>\n",
       "      <td>1.646445</td>\n",
       "      <td>37.055640</td>\n",
       "      <td>1.376303</td>\n",
       "      <td>1.803791</td>\n",
       "      <td>...</td>\n",
       "      <td>0.129858</td>\n",
       "      <td>0.883412</td>\n",
       "      <td>1.274882</td>\n",
       "      <td>0.961137</td>\n",
       "      <td>3.918240</td>\n",
       "      <td>2.558417</td>\n",
       "      <td>0.686256</td>\n",
       "      <td>8.629492</td>\n",
       "      <td>0.051185</td>\n",
       "      <td>0.723223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.546916</td>\n",
       "      <td>0.831308</td>\n",
       "      <td>1.462452</td>\n",
       "      <td>0.256010</td>\n",
       "      <td>2.332955</td>\n",
       "      <td>1.073771</td>\n",
       "      <td>2.224822</td>\n",
       "      <td>9.144466</td>\n",
       "      <td>1.963521</td>\n",
       "      <td>1.775435</td>\n",
       "      <td>...</td>\n",
       "      <td>0.644057</td>\n",
       "      <td>1.520467</td>\n",
       "      <td>2.273994</td>\n",
       "      <td>1.257013</td>\n",
       "      <td>0.999602</td>\n",
       "      <td>0.642765</td>\n",
       "      <td>1.090389</td>\n",
       "      <td>1.241986</td>\n",
       "      <td>0.318970</td>\n",
       "      <td>2.239286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.803900</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.267000</td>\n",
       "      <td>1.467000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.917000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>4.481000</td>\n",
       "      <td>2.502750</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>30.450000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.487500</td>\n",
       "      <td>2.103000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.991000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.828000</td>\n",
       "      <td>3.046300</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>37.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.726000</td>\n",
       "      <td>2.458000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.499000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5.125000</td>\n",
       "      <td>3.437650</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>43.400000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.987000</td>\n",
       "      <td>2.870500</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>9.020500</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6.496000</td>\n",
       "      <td>9.177500</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>60.700000</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>10.695000</td>\n",
       "      <td>5.825000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>14.700000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>27.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 41 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                0            1            2            3            4   \\\n",
       "count  1055.000000  1055.000000  1055.000000  1055.000000  1055.000000   \n",
       "mean      4.782644     3.069508     0.716588     0.042654     0.980095   \n",
       "std       0.546916     0.831308     1.462452     0.256010     2.332955   \n",
       "min       2.000000     0.803900     0.000000     0.000000     0.000000   \n",
       "25%       4.481000     2.502750     0.000000     0.000000     0.000000   \n",
       "50%       4.828000     3.046300     0.000000     0.000000     0.000000   \n",
       "75%       5.125000     3.437650     1.000000     0.000000     1.000000   \n",
       "max       6.496000     9.177500    12.000000     3.000000    36.000000   \n",
       "\n",
       "                5            6            7            8            9   \\\n",
       "count  1055.000000  1055.000000  1055.000000  1055.000000  1055.000000   \n",
       "mean      0.290047     1.646445    37.055640     1.376303     1.803791   \n",
       "std       1.073771     2.224822     9.144466     1.963521     1.775435   \n",
       "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
       "25%       0.000000     0.000000    30.450000     0.000000     0.000000   \n",
       "50%       0.000000     1.000000    37.500000     1.000000     2.000000   \n",
       "75%       0.000000     3.000000    43.400000     2.000000     3.000000   \n",
       "max      13.000000    18.000000    60.700000    24.000000    12.000000   \n",
       "\n",
       "          ...                31           32           33           34  \\\n",
       "count     ...       1055.000000  1055.000000  1055.000000  1055.000000   \n",
       "mean      ...          0.129858     0.883412     1.274882     0.961137   \n",
       "std       ...          0.644057     1.520467     2.273994     1.257013   \n",
       "min       ...          0.000000     0.000000     0.000000     0.000000   \n",
       "25%       ...          0.000000     0.000000     0.000000     0.000000   \n",
       "50%       ...          0.000000     0.000000     0.000000     1.000000   \n",
       "75%       ...          0.000000     1.000000     2.000000     2.000000   \n",
       "max       ...          8.000000    12.000000    18.000000     7.000000   \n",
       "\n",
       "                35           36           37           38           39  \\\n",
       "count  1055.000000  1055.000000  1055.000000  1055.000000  1055.000000   \n",
       "mean      3.918240     2.558417     0.686256     8.629492     0.051185   \n",
       "std       0.999602     0.642765     1.090389     1.241986     0.318970   \n",
       "min       2.267000     1.467000     0.000000     4.917000     0.000000   \n",
       "25%       3.487500     2.103000     0.000000     7.991000     0.000000   \n",
       "50%       3.726000     2.458000     0.000000     8.499000     0.000000   \n",
       "75%       3.987000     2.870500     1.000000     9.020500     0.000000   \n",
       "max      10.695000     5.825000     8.000000    14.700000     4.000000   \n",
       "\n",
       "                40  \n",
       "count  1055.000000  \n",
       "mean      0.723223  \n",
       "std       2.239286  \n",
       "min       0.000000  \n",
       "25%       0.000000  \n",
       "50%       0.000000  \n",
       "75%       0.000000  \n",
       "max      27.000000  \n",
       "\n",
       "[8 rows x 41 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1055 entries, 0 to 1054\n",
      "Data columns (total 42 columns):\n",
      "0     1055 non-null float64\n",
      "1     1055 non-null float64\n",
      "2     1055 non-null int64\n",
      "3     1055 non-null int64\n",
      "4     1055 non-null int64\n",
      "5     1055 non-null int64\n",
      "6     1055 non-null int64\n",
      "7     1055 non-null float64\n",
      "8     1055 non-null int64\n",
      "9     1055 non-null int64\n",
      "10    1055 non-null int64\n",
      "11    1055 non-null float64\n",
      "12    1055 non-null float64\n",
      "13    1055 non-null float64\n",
      "14    1055 non-null float64\n",
      "15    1055 non-null int64\n",
      "16    1055 non-null float64\n",
      "17    1055 non-null float64\n",
      "18    1055 non-null int64\n",
      "19    1055 non-null int64\n",
      "20    1055 non-null int64\n",
      "21    1055 non-null float64\n",
      "22    1055 non-null int64\n",
      "23    1055 non-null int64\n",
      "24    1055 non-null int64\n",
      "25    1055 non-null int64\n",
      "26    1055 non-null float64\n",
      "27    1055 non-null float64\n",
      "28    1055 non-null int64\n",
      "29    1055 non-null float64\n",
      "30    1055 non-null float64\n",
      "31    1055 non-null int64\n",
      "32    1055 non-null int64\n",
      "33    1055 non-null int64\n",
      "34    1055 non-null int64\n",
      "35    1055 non-null float64\n",
      "36    1055 non-null float64\n",
      "37    1055 non-null int64\n",
      "38    1055 non-null float64\n",
      "39    1055 non-null int64\n",
      "40    1055 non-null int64\n",
      "41    1055 non-null object\n",
      "dtypes: float64(17), int64(24), object(1)\n",
      "memory usage: 346.2+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().values.any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.drop(columns=[41])\n",
    "y = df[41]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "m =  1055\n",
      "n =  41\n"
     ]
    }
   ],
   "source": [
    "m = X.shape[0]\n",
    "n = X.shape[1]\n",
    "print('m = ', m)\n",
    "print('n = ', n)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = y.replace({'RB':0})\n",
    "y = y.replace({'NRB':1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = X.astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "scaler = StandardScaler()\n",
    "X = scaler.fit_transform(X)\n",
    "X = pd.DataFrame(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>40</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.579865</td>\n",
       "      <td>-0.455652</td>\n",
       "      <td>-0.490223</td>\n",
       "      <td>-0.16669</td>\n",
       "      <td>-0.420308</td>\n",
       "      <td>-0.270248</td>\n",
       "      <td>-0.740386</td>\n",
       "      <td>-0.618770</td>\n",
       "      <td>0.317793</td>\n",
       "      <td>-1.016454</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.20172</td>\n",
       "      <td>-0.581289</td>\n",
       "      <td>-0.560901</td>\n",
       "      <td>-0.764983</td>\n",
       "      <td>-0.970085</td>\n",
       "      <td>-1.505801</td>\n",
       "      <td>-0.629667</td>\n",
       "      <td>-1.108824</td>\n",
       "      <td>-0.160545</td>\n",
       "      <td>-0.323123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-1.120710</td>\n",
       "      <td>-1.149466</td>\n",
       "      <td>-0.490223</td>\n",
       "      <td>-0.16669</td>\n",
       "      <td>-0.420308</td>\n",
       "      <td>-0.270248</td>\n",
       "      <td>-0.740386</td>\n",
       "      <td>-0.684415</td>\n",
       "      <td>-0.191738</td>\n",
       "      <td>-0.452944</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.20172</td>\n",
       "      <td>-0.581289</td>\n",
       "      <td>-0.560901</td>\n",
       "      <td>-0.764983</td>\n",
       "      <td>-0.603766</td>\n",
       "      <td>-0.920550</td>\n",
       "      <td>-0.629667</td>\n",
       "      <td>-1.105602</td>\n",
       "      <td>-0.160545</td>\n",
       "      <td>-0.323123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1.556084</td>\n",
       "      <td>0.218665</td>\n",
       "      <td>-0.490223</td>\n",
       "      <td>-0.16669</td>\n",
       "      <td>-0.420308</td>\n",
       "      <td>-0.270248</td>\n",
       "      <td>-0.740386</td>\n",
       "      <td>-1.132986</td>\n",
       "      <td>0.317793</td>\n",
       "      <td>1.237584</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.20172</td>\n",
       "      <td>-0.581289</td>\n",
       "      <td>-0.560901</td>\n",
       "      <td>0.030931</td>\n",
       "      <td>-0.842975</td>\n",
       "      <td>-0.220118</td>\n",
       "      <td>-0.629667</td>\n",
       "      <td>-0.828495</td>\n",
       "      <td>-0.160545</td>\n",
       "      <td>-0.323123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-3.260993</td>\n",
       "      <td>-0.432906</td>\n",
       "      <td>-0.490223</td>\n",
       "      <td>-0.16669</td>\n",
       "      <td>-0.420308</td>\n",
       "      <td>-0.270248</td>\n",
       "      <td>-0.740386</td>\n",
       "      <td>-1.866017</td>\n",
       "      <td>-0.701269</td>\n",
       "      <td>0.110565</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.20172</td>\n",
       "      <td>-0.581289</td>\n",
       "      <td>-0.560901</td>\n",
       "      <td>0.030931</td>\n",
       "      <td>-0.873001</td>\n",
       "      <td>3.800364</td>\n",
       "      <td>-0.629667</td>\n",
       "      <td>-1.562345</td>\n",
       "      <td>-0.160545</td>\n",
       "      <td>-0.323123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.999976</td>\n",
       "      <td>0.391006</td>\n",
       "      <td>-0.490223</td>\n",
       "      <td>-0.16669</td>\n",
       "      <td>-0.420308</td>\n",
       "      <td>-0.270248</td>\n",
       "      <td>-0.740386</td>\n",
       "      <td>-0.837585</td>\n",
       "      <td>0.317793</td>\n",
       "      <td>1.237584</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.20172</td>\n",
       "      <td>-0.581289</td>\n",
       "      <td>-0.560901</td>\n",
       "      <td>-0.764983</td>\n",
       "      <td>-0.567735</td>\n",
       "      <td>-0.238796</td>\n",
       "      <td>-0.629667</td>\n",
       "      <td>-0.504667</td>\n",
       "      <td>-0.160545</td>\n",
       "      <td>-0.323123</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 41 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         0         1         2        3         4         5         6   \\\n",
       "0 -1.579865 -0.455652 -0.490223 -0.16669 -0.420308 -0.270248 -0.740386   \n",
       "1 -1.120710 -1.149466 -0.490223 -0.16669 -0.420308 -0.270248 -0.740386   \n",
       "2 -1.556084  0.218665 -0.490223 -0.16669 -0.420308 -0.270248 -0.740386   \n",
       "3 -3.260993 -0.432906 -0.490223 -0.16669 -0.420308 -0.270248 -0.740386   \n",
       "4 -0.999976  0.391006 -0.490223 -0.16669 -0.420308 -0.270248 -0.740386   \n",
       "\n",
       "         7         8         9     ...          31        32        33  \\\n",
       "0 -0.618770  0.317793 -1.016454    ...    -0.20172 -0.581289 -0.560901   \n",
       "1 -0.684415 -0.191738 -0.452944    ...    -0.20172 -0.581289 -0.560901   \n",
       "2 -1.132986  0.317793  1.237584    ...    -0.20172 -0.581289 -0.560901   \n",
       "3 -1.866017 -0.701269  0.110565    ...    -0.20172 -0.581289 -0.560901   \n",
       "4 -0.837585  0.317793  1.237584    ...    -0.20172 -0.581289 -0.560901   \n",
       "\n",
       "         34        35        36        37        38        39        40  \n",
       "0 -0.764983 -0.970085 -1.505801 -0.629667 -1.108824 -0.160545 -0.323123  \n",
       "1 -0.764983 -0.603766 -0.920550 -0.629667 -1.105602 -0.160545 -0.323123  \n",
       "2  0.030931 -0.842975 -0.220118 -0.629667 -0.828495 -0.160545 -0.323123  \n",
       "3  0.030931 -0.873001  3.800364 -0.629667 -1.562345 -0.160545 -0.323123  \n",
       "4 -0.764983 -0.567735 -0.238796 -0.629667 -0.504667 -0.160545 -0.323123  \n",
       "\n",
       "[5 rows x 41 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>40</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>...</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "      <td>1.055000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-1.138952e-15</td>\n",
       "      <td>-1.560626e-16</td>\n",
       "      <td>-6.377205e-16</td>\n",
       "      <td>3.841582e-16</td>\n",
       "      <td>-4.028899e-16</td>\n",
       "      <td>-7.881005e-16</td>\n",
       "      <td>2.327785e-16</td>\n",
       "      <td>6.082549e-16</td>\n",
       "      <td>-1.574307e-16</td>\n",
       "      <td>7.534784e-17</td>\n",
       "      <td>...</td>\n",
       "      <td>1.220719e-17</td>\n",
       "      <td>-9.599483e-16</td>\n",
       "      <td>-7.976768e-16</td>\n",
       "      <td>1.635343e-16</td>\n",
       "      <td>-5.283162e-16</td>\n",
       "      <td>-2.400397e-16</td>\n",
       "      <td>-7.009664e-16</td>\n",
       "      <td>-1.027456e-15</td>\n",
       "      <td>7.926782e-16</td>\n",
       "      <td>1.315430e-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "      <td>1.000474e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-5.090295e+00</td>\n",
       "      <td>-2.726645e+00</td>\n",
       "      <td>-4.902229e-01</td>\n",
       "      <td>-1.666895e-01</td>\n",
       "      <td>-4.203080e-01</td>\n",
       "      <td>-2.702484e-01</td>\n",
       "      <td>-7.403857e-01</td>\n",
       "      <td>-4.054169e+00</td>\n",
       "      <td>-7.012688e-01</td>\n",
       "      <td>-1.016454e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.017203e-01</td>\n",
       "      <td>-5.812892e-01</td>\n",
       "      <td>-5.609013e-01</td>\n",
       "      <td>-7.649826e-01</td>\n",
       "      <td>-1.652680e+00</td>\n",
       "      <td>-1.698809e+00</td>\n",
       "      <td>-6.296666e-01</td>\n",
       "      <td>-2.990574e+00</td>\n",
       "      <td>-1.605451e-01</td>\n",
       "      <td>-3.231234e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-5.517972e-01</td>\n",
       "      <td>-6.820895e-01</td>\n",
       "      <td>-4.902229e-01</td>\n",
       "      <td>-1.666895e-01</td>\n",
       "      <td>-4.203080e-01</td>\n",
       "      <td>-2.702484e-01</td>\n",
       "      <td>-7.403857e-01</td>\n",
       "      <td>-7.227073e-01</td>\n",
       "      <td>-7.012688e-01</td>\n",
       "      <td>-1.016454e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.017203e-01</td>\n",
       "      <td>-5.812892e-01</td>\n",
       "      <td>-5.609013e-01</td>\n",
       "      <td>-7.649826e-01</td>\n",
       "      <td>-4.311156e-01</td>\n",
       "      <td>-7.088642e-01</td>\n",
       "      <td>-6.296666e-01</td>\n",
       "      <td>-5.143332e-01</td>\n",
       "      <td>-1.605451e-01</td>\n",
       "      <td>-3.231234e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>8.297054e-02</td>\n",
       "      <td>-2.793041e-02</td>\n",
       "      <td>-4.902229e-01</td>\n",
       "      <td>-1.666895e-01</td>\n",
       "      <td>-4.203080e-01</td>\n",
       "      <td>-2.702484e-01</td>\n",
       "      <td>-2.906983e-01</td>\n",
       "      <td>4.861639e-02</td>\n",
       "      <td>-1.917381e-01</td>\n",
       "      <td>1.105654e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.017203e-01</td>\n",
       "      <td>-5.812892e-01</td>\n",
       "      <td>-5.609013e-01</td>\n",
       "      <td>3.093125e-02</td>\n",
       "      <td>-1.924075e-01</td>\n",
       "      <td>-1.563008e-01</td>\n",
       "      <td>-6.296666e-01</td>\n",
       "      <td>-1.051170e-01</td>\n",
       "      <td>-1.605451e-01</td>\n",
       "      <td>-3.231234e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.262732e-01</td>\n",
       "      <td>4.430569e-01</td>\n",
       "      <td>1.938845e-01</td>\n",
       "      <td>-1.666895e-01</td>\n",
       "      <td>8.536237e-03</td>\n",
       "      <td>-2.702484e-01</td>\n",
       "      <td>6.086763e-01</td>\n",
       "      <td>6.941214e-01</td>\n",
       "      <td>3.177926e-01</td>\n",
       "      <td>6.740749e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.017203e-01</td>\n",
       "      <td>7.671520e-02</td>\n",
       "      <td>3.190257e-01</td>\n",
       "      <td>8.268451e-01</td>\n",
       "      <td>6.882017e-02</td>\n",
       "      <td>4.857623e-01</td>\n",
       "      <td>2.878725e-01</td>\n",
       "      <td>3.149741e-01</td>\n",
       "      <td>-1.605451e-01</td>\n",
       "      <td>-3.231234e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3.134246e+00</td>\n",
       "      <td>7.350931e+00</td>\n",
       "      <td>7.719065e+00</td>\n",
       "      <td>1.155714e+01</td>\n",
       "      <td>1.501809e+01</td>\n",
       "      <td>1.184236e+01</td>\n",
       "      <td>7.353986e+00</td>\n",
       "      <td>2.586873e+00</td>\n",
       "      <td>1.152747e+01</td>\n",
       "      <td>5.745661e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>1.222543e+01</td>\n",
       "      <td>7.314763e+00</td>\n",
       "      <td>7.358442e+00</td>\n",
       "      <td>4.806415e+00</td>\n",
       "      <td>6.782672e+00</td>\n",
       "      <td>5.084491e+00</td>\n",
       "      <td>6.710646e+00</td>\n",
       "      <td>4.890059e+00</td>\n",
       "      <td>1.238576e+01</td>\n",
       "      <td>1.174001e+01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 41 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 0             1             2             3             4   \\\n",
       "count  1.055000e+03  1.055000e+03  1.055000e+03  1.055000e+03  1.055000e+03   \n",
       "mean  -1.138952e-15 -1.560626e-16 -6.377205e-16  3.841582e-16 -4.028899e-16   \n",
       "std    1.000474e+00  1.000474e+00  1.000474e+00  1.000474e+00  1.000474e+00   \n",
       "min   -5.090295e+00 -2.726645e+00 -4.902229e-01 -1.666895e-01 -4.203080e-01   \n",
       "25%   -5.517972e-01 -6.820895e-01 -4.902229e-01 -1.666895e-01 -4.203080e-01   \n",
       "50%    8.297054e-02 -2.793041e-02 -4.902229e-01 -1.666895e-01 -4.203080e-01   \n",
       "75%    6.262732e-01  4.430569e-01  1.938845e-01 -1.666895e-01  8.536237e-03   \n",
       "max    3.134246e+00  7.350931e+00  7.719065e+00  1.155714e+01  1.501809e+01   \n",
       "\n",
       "                 5             6             7             8             9   \\\n",
       "count  1.055000e+03  1.055000e+03  1.055000e+03  1.055000e+03  1.055000e+03   \n",
       "mean  -7.881005e-16  2.327785e-16  6.082549e-16 -1.574307e-16  7.534784e-17   \n",
       "std    1.000474e+00  1.000474e+00  1.000474e+00  1.000474e+00  1.000474e+00   \n",
       "min   -2.702484e-01 -7.403857e-01 -4.054169e+00 -7.012688e-01 -1.016454e+00   \n",
       "25%   -2.702484e-01 -7.403857e-01 -7.227073e-01 -7.012688e-01 -1.016454e+00   \n",
       "50%   -2.702484e-01 -2.906983e-01  4.861639e-02 -1.917381e-01  1.105654e-01   \n",
       "75%   -2.702484e-01  6.086763e-01  6.941214e-01  3.177926e-01  6.740749e-01   \n",
       "max    1.184236e+01  7.353986e+00  2.586873e+00  1.152747e+01  5.745661e+00   \n",
       "\n",
       "           ...                 31            32            33            34  \\\n",
       "count      ...       1.055000e+03  1.055000e+03  1.055000e+03  1.055000e+03   \n",
       "mean       ...       1.220719e-17 -9.599483e-16 -7.976768e-16  1.635343e-16   \n",
       "std        ...       1.000474e+00  1.000474e+00  1.000474e+00  1.000474e+00   \n",
       "min        ...      -2.017203e-01 -5.812892e-01 -5.609013e-01 -7.649826e-01   \n",
       "25%        ...      -2.017203e-01 -5.812892e-01 -5.609013e-01 -7.649826e-01   \n",
       "50%        ...      -2.017203e-01 -5.812892e-01 -5.609013e-01  3.093125e-02   \n",
       "75%        ...      -2.017203e-01  7.671520e-02  3.190257e-01  8.268451e-01   \n",
       "max        ...       1.222543e+01  7.314763e+00  7.358442e+00  4.806415e+00   \n",
       "\n",
       "                 35            36            37            38            39  \\\n",
       "count  1.055000e+03  1.055000e+03  1.055000e+03  1.055000e+03  1.055000e+03   \n",
       "mean  -5.283162e-16 -2.400397e-16 -7.009664e-16 -1.027456e-15  7.926782e-16   \n",
       "std    1.000474e+00  1.000474e+00  1.000474e+00  1.000474e+00  1.000474e+00   \n",
       "min   -1.652680e+00 -1.698809e+00 -6.296666e-01 -2.990574e+00 -1.605451e-01   \n",
       "25%   -4.311156e-01 -7.088642e-01 -6.296666e-01 -5.143332e-01 -1.605451e-01   \n",
       "50%   -1.924075e-01 -1.563008e-01 -6.296666e-01 -1.051170e-01 -1.605451e-01   \n",
       "75%    6.882017e-02  4.857623e-01  2.878725e-01  3.149741e-01 -1.605451e-01   \n",
       "max    6.782672e+00  5.084491e+00  6.710646e+00  4.890059e+00  1.238576e+01   \n",
       "\n",
       "                 40  \n",
       "count  1.055000e+03  \n",
       "mean   1.315430e-17  \n",
       "std    1.000474e+00  \n",
       "min   -3.231234e-01  \n",
       "25%   -3.231234e-01  \n",
       "50%   -3.231234e-01  \n",
       "75%   -3.231234e-01  \n",
       "max    1.174001e+01  \n",
       "\n",
       "[8 rows x 41 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Visualization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Outliers and Boxplots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x2880 with 41 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize = (25, 40))\n",
    "unique_fliers = []\n",
    "\n",
    "col = 5\n",
    "row = np.ceil(X.shape[1]/col).astype(int)\n",
    "\n",
    "\n",
    "for r in range(X.shape[1]):\n",
    "    plt.subplot(row,col,r+1)\n",
    "    dict_boxPlot = plt.boxplot(X[r])\n",
    "    _,fliers = dict_boxPlot['fliers'][0].get_data()\n",
    "    plt.title('Feature {} Fliers: {}'.format(r, fliers.size))\n",
    "\n",
    "    lower_cap = dict_boxPlot['caps'][0].get_ydata()[0] \n",
    "    upper_cap = dict_boxPlot['caps'][1].get_ydata()[0] \n",
    "\n",
    "    idx_tuple = np.where((X[r] > upper_cap) | (X[r] < lower_cap))\n",
    "    idx_fliers = idx_tuple[0]\n",
    "\n",
    "    for q in idx_fliers:    \n",
    "        unique_fliers.append(q)\n",
    "\n",
    "fig.subplots_adjust(top=0.95)\n",
    "fig.suptitle(\"All Feature Boxplots with Fliers Count\",fontsize=30)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "882\n",
      "15\n"
     ]
    }
   ],
   "source": [
    "unique_fliers,y_counts = np.unique(unique_fliers, return_counts=True)\n",
    "print(unique_fliers.size)\n",
    "print(y_counts.max()) # One instance has an outlier in this many features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Correlation and Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x180ee7bfb00>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X_corr = X.corr()\n",
    "sns.heatmap(X_corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlations greater than 0.7\n",
      "\n",
      "Feature\tFeature\tCorrelation\n",
      "0\t14\t0.9115464957533881\t\n",
      "0\t26\t0.9189280075472851\t\n",
      "4\t10\t0.8705177316992135\t\n",
      "4\t33\t0.7893575410344343\t\n",
      "4\t37\t0.7457167590848371\t\n",
      "6\t32\t0.8003257607300918\t\n",
      "9\t15\t0.8401372627371547\t\n",
      "9\t29\t0.7483794044644284\t\n",
      "10\t33\t0.8039433726719026\t\n",
      "10\t37\t0.8371897588642094\t\n",
      "12\t14\t0.8052540408637993\t\n",
      "12\t38\t0.8211443105949399\t\n",
      "13\t30\t0.7493816217978518\t\n",
      "14\t26\t0.9226494500596467\t\n",
      "16\t36\t0.8537523998606932\t\n",
      "23\t28\t0.810912269121414\t\n",
      "33\t37\t0.8027703202007902\t\n",
      "35\t38\t0.9174433469633239\t\n"
     ]
    }
   ],
   "source": [
    "threshold = 0.7\n",
    "print(\"Correlations greater than {}\\n\".format(threshold))\n",
    "print(\"Feature\\tFeature\\tCorrelation\")\n",
    "corr_col_index = []\n",
    "for i in range(X_corr.shape[0]):\n",
    "    for j in range(i+1, X_corr.shape[0]):\n",
    "        if X_corr[i][j] >= threshold:\n",
    "            print(\"{}\\t{}\\t{}\\t\".format(i, j, X_corr[i][j]))\n",
    "            corr_col_index.append([i,j])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Examining PCA Components"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtQAAAJQCAYAAACuDPM+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzs3Xl8lOW9/vHrO9nJBiELWyAsYRMEAQG3uqFVq1i12mq11WrVVrvqaW3tpj09XX5dj8fTTavW9YhVa61WS+tGFYGwStiXQFiSQBKykW3m/v0xg43KMpA8eWYmn/frNa+ZZ+ZJuKolXD58n/s255wAAAAAHJuA3wEAAACAeEahBgAAALqBQg0AAAB0A4UaAAAA6AYKNQAAANANFGoAAACgGyjUAAAAQDdQqAEAAIBuoFADAAAA3ZDsd4CjlZ+f70pKSvyOAQAAgARXVla2xzlXcKTz4q5Ql5SUaMmSJX7HAAAAQIIzs4pozmPkAwAAAOgGCjUAAADQDRRqAAAAoBso1AAAAEA3UKgBAACAbqBQAwAAAN1AoQYAAAC6gUINAAAAdAOFGgAAAOgGCjUAAADQDRRqAAAAoBso1AAAAEA3UKgBAACAbqBQAwAAAN1AoQYAAAC6gUINAAAAdAOFGgAAAOgGCjUAAADQDRRqAAAAoBs8K9Rm9gczqzazdw7xuZnZf5vZRjNbaWbTvMoCAAAAeMXLK9QPSjrvMJ+fL6k08rhR0q89zAIAAAB4wrNC7Zx7XVLtYU65WNIfXdhCSf3NbLBXeQAAAAAvJPv4aw+VtL3LcWXkvV3+xAEAAIhdwZBTRzCk9mBIHZ0hdYac2iPPHcHQe16HHy5yXkjtQafOYEidQaeOUCjyvZyCofB5nQdeh9y7v05n0KkzFP66YCj8OuQij5AUdE7Ohc8POr37+sDnIecUdE6hyGfOSU6RZyc5hd8/4D2fRz5zkgJmmv/V0/36xx4VPwu1HeQ9d5D3ZGY3KjwWouHDh3uZCQAAICqhkNP+jqBa2oNqjTy3tHdqf3v49f6OYOR1p1o6gmrt8n57Z0htnSG1dXZ9HYq8DnZ5HVJbRzBcooMHrUk9KiXJlBwIKDlgSk4yJSeFXycFTMkBUyBgCpgpyUxmUlLkOPy+lHTgdUBKCQTCn0XONUlmFnmWpH+/r8h7duC9yGtZuFDHOj8LdaWk4i7HwyTtPNiJzrnfSfqdJM2YMcP7/zcBAICE1dYZ1L79HWpq7VRTW6eaWjvVGHluags/Gls71dT273Mau3zW1Nqp5vZOtXaEjurXNZMyUpKUnpKk9OSAUpMDSktOUlpKQKlJAaWnBJSbkaLUpIDSUgJK63pO5Dkl2ZSaFFBK5JGcFD5OTjKlJAXe8zrl3efAv4ty5L3kwL+PkwLh95ICsV9cY5Wfhfo5Sbea2ROSZkna55xj3AMAAByWc04t7eFSvG9/hxoiz+8et3a+5733f97WeeQinJoUUHZ6srLSk5WVFn4Mzk1XVlqyMiPHGalJykhJUr/UJGWkJnd5HXlOOfA6OVKkA7I4uNqKo+dZoTazxyWdISnfzColfVdSiiQ5534j6QVJF0jaKKlF0nVeZQEAALGntSOo2ub295TdgxXgffs7VN/y7/cbWjuOOP6QnZ6s3IwU5WakKCc9RaMLssKvM5Ijz+H3s9L+XZqzD5Tn9GSlJSf10j8FJALPCrVz7sojfO4k3eLVrw8AAHqXc04NrZ3a29SmPU3tkefw6z2R13sjr/c2tauxrfOQ3ytgUk6kEB94DB2Q8Z7j9z9y0sPPWenJjC+gV/k58gEAAOJAa0dQNY1tqm5sU01jm2oaW1Xd2KbqhjZVN7a+W5j3NrWrPfjBcQozaUC/VA3MTFV+VpomDc1Vflaa8rNSlZeZpv793leMM1KUnZasAKUYcYJCDQBAH9XS3qmqhjbt3teq6sbWd0tzdUPru+W5urFN+/Z3fOBrAyYNzEpTYXaa8rPSNLYoW/nZqcrPTFN+dqoGZobfz89OVV6/VCUnebmXHOAvCjUAAAmmMxjSnqZ27W5ofbcs797XqqqGNlU1tKqqoVW7G1rV2PrBkYvU5IAKs8NFeXRBlmaPGhg+zklTYXa6CiKfDcxKY6wCiKBQAwAQZ0Ihp10Nrdq6p1lb9zZHnlsipblVe5raFHrfPXvJAYsU43SNLsjSKWPyVZiTpkE56SrKSVdRTpoKstKVk5HMShTAUaJQAwAQg0Ihp90NrZHC3KKte5u1ZU+zKvY2q2Jvy3uWfktLDmh4Xj8N6Z+hiYNzVJSTpqLcdBVlp2tQbrgwD8xMZSYZ8AiFGgAAnzjntLe5XZtrmrVlT5M274lcbd7Toora5vdsHJKaHNCIvH4qyc/U6WMLVJKfqZEDM1WSn6lBOemUZcBHFGoAADy2vz2oLXvCV5g314SL8+Y9zdpS06SGLnPMqUkBFedlaGR+pk4rzQ+X5vxMjRjYT4NzM5hZBmIUhRoAgB7gnNOO+v3aWN0UKc7/LtA797W+59whuekaWZCpi6cO1cj8TI0qyNSo/CwNHUBpBuIRhRoAgKMQCoWL84bqRm2oatL6qiZtrG7UxuomNbcH3z0vOz1ZowqyNGvUQI3Kz9TISGkuye+nfqn88QskEn5HAwBwEKGQU2Xdfq2vatSG6iZtiDxvrG7S/o5/F+fC7PAazJfPKFZpUZbGFGRpVEGW8rNSWS0D6CMo1ACAPs258Goaa3c3al3ksb6qUZtqmt5zU+CgnHSVFmXpEzOLNbYoW6WFWSotzFZuvxQf0wOIBRRqAECf0dDaofW7G99TntdVNb5nJ8CinPAV59mjRoRLc1G2xhRmKTeD4gzg4CjUAICE094Z0uY9TVr3vvK8o37/u+dkpSVr3KBsfeT4wRo/KFvjirI1blC2+vdL9TE5gHhEoQYAxDXnnCr2tmj59vp3H+U7G9QeDI9rJAdMowuyNH3EAF01a3i4PA/K1tD+Gcw4A+gRFGoAQFypbW7Xii7leUVlvepbwiMbGSlJmjwsV9eeUqLjhuRo3KBsjcrPUmpywOfUABIZhRoAELNaO4Iq39Wg5dvCxXn59npV7G2RJAVMGluUrQ9PHKSpw/tranF/lRZmKTmJ8gygd1GoAQAxY9/+Di3ZWquFm/dq0ZZale9qUEfQSQqvsjG1uL+unDlcU4b11+RhucpK448xAP7jJxEAwDf7Wjq0aGut3t68Vwu37NXqnQ1yLrwF99Th/XX9qaM0tTh89XlQbrrfcQHgoCjUAIBeU9/SrkVbarVwc63e3rJX5bsiBTo5oGnD++tLZ5dq1siBOmF4f6WnJPkdFwCiQqEGAHimvqVdb28Jj3As3FyrtbvDBTotOaBpwwfoy2eP1exReZpSTIEGEL8o1ACAHtPaEdTirbVasHGPFmzY8+4V6LTkgGaUDNBX5ozV7FEDNaU4V2nJFGgAiYFCDQA4ZqGQ0+qdDeECvbFGi7fWqb0zpJQk07Th4QJ90uiBOn4YBRpA4qJQAwCOyvbalkiB3qM3N+5RXWQN6HFF2bpm9gidWpqvmSV5ymQFDgB9BD/tAACHta+lQ29u2vNuiT6wDnRRTprOGl+kU0sH6pQx+SrMZhUOAH0ThRoA8AFb9zTr5fLdeml1lZZtq1PISZmpSZo9aqCuPblEp5Xma3RBFlt3A4Ao1AAASc6FZ6FfXh0u0euqGiVJEwfn6NazSnVaab6mFvdXCrsQAsAHUKgBoI8KhpwWb63VS6t36+XVVdpRv18Bk2aU5OnbF07UuROLVJzXz++YABDzKNQA0Ie0dgS1YMMevVy+W/PXVKu2uV2pyQGdNiZfXzq7VGdPKNTArDS/YwJAXKFQA0CCa2jt0Ctrq/XS6t16dV2NWtqDyk5L1lkTCnXuxEE6fVyBsliRAwCOGT9BASABBUNOCzbu0VNllXpp9W61d4ZUkJ2mj54wVB8+bpBOGjVQqcnMQwNAT6BQA0AC2VzTpKfKKvX00h3a3dCq3IwUfeLEYl08dYhOKB6gQIBVOQCgp1GoASDONbZ26K8rd2leWaXKKuoUMOn0sQX6zkUTdfaEQnYoBACPUagBIA6FQk5vbd6rp8oq9eI7u9TaEdLogkzdcf54XXrCUBXmsMkKAPQWCjUAxJGKvc36U1ml/rR0h3bU71d2erIumzZMH5s+TFOL+7PRCgD4gEINADGuua1TL6wKj3Qs2lIrM+nUMfn6+vnjde7EIqWnMNIBAH6iUANADHLOafHWOj25ZLteWLVLLe1BjczP1H98eJwunTZUg3Mz/I4IAIigUANADNlZv19PL63UU2WV2rq3RZmpSbro+CG6fMYwTR8xgJEOAIhBFGoA8FlrR1Avl1dp3pLtWrBxj5yTZo/K0xfOKtX5kwepXyo/qgEglvFTGgB84JzTysp9mle2Xc8t36mG1k4N7Z+hL5xVqo9NG6bhA/v5HREAECUKNQD0oj1NbXp22Q7NW1KpdVWNSksO6LxJg3T59GKdPHogG68AQByiUANAL1i0pVb3vbFZ/1xbrc6Q09Ti/vrBJZN04fFDlJuR4nc8AEA3UKgBwCPOOS3YuEf3/HOjFm2p1cDMVF1/6kh9bPowlRZl+x0PANBDKNQA0MOcc/rHmmrd88pGrdher0E56freRRP18ROHKyOVNaMBINFQqAGghwRDTn97Z7f+55WNWrOrQcV5GfrhpZN16bShSkumSANAoqJQA0A3dQZDem7FTt37ykZtqmnWqIJM/ezyKZo7dYhSkgJ+xwMAeIxCDQDHqK0zqKeX7tCvX92kbbUtGj8oW/9z1Qk6f9JgJbFaBwD0GRRqADhKrR1BPbFom377+mbt2teqKcNy9e0LZ+js8YUsewcAfRCFGgCi1NLeqYffqtDv39iiPU1tOrFkgH582fE6rTSfLcEBoA+jUAPAEYRCTs8u36Ef/22tqhradFppvm498wTNGjXQ72gAgBhAoQaAw1i6rU53/aVcK7bXa8qwXN171TTNKMnzOxYAIIZQqAHgIHbt268fv7hWzy7fqcLsNP3s8im65IShzEgDAD6AQg0AXbR2BPW71zfr169uUtA53XLmaH3+jDHKTOPHJQDg4PgTAgAU3t3w+ZW79KMX12pH/X5dMHmQvnH+BBXn9fM7GgAgxlGoAfR5qyr36e7nV2vx1jpNGJyjn10xRbO54RAAECUKNYA+q7qxVT99aZ3mlVUqr1+qfnjpZF0xo5hNWQAAR4VCDaDPaesM6g8LtureVzaqrTOoz542SreeNUY56Sl+RwMAxCEKNYA+ZX55lb7/13JV7G3RnAmFuvMjEzUyP9PvWACAOEahBtAn1DS26Xt/Wa2/rtyl0sIsPXz9TJ1WWuB3LABAAqBQA0hozjk9vXSHvv/XcrW0BXX7uWN10+mjlZIU8DsaACBBUKgBJKzKuhZ985l39Pr6Gk0fMUA/vmyyxhRm+x0LAJBgKNQAEk4o5PTHt7bqJy+tkyTdNfc4XTN7BLscAgA8QaEGkFA2Vjfq639apbKKOp0+tkA/uGSShg1gcxYAgHco1AASQntnSL99bZPu+edG9UtL0s+vmKJLThgqM65KAwC8RaEGEPdWbK/X1/+0Umt3N+rC4wfre3OPU35Wmt+xAAB9BIUaQNza3x7Uz/++Tvcv2KKC7DT9/lMzdM7EIr9jAQD6GAo1gLj05sY9uuPpVdpW26KrZg3XHeePZ6dDAIAvKNQA4kpTW6d+8NdyPb5ou0oG9tPjn52tk0YP9DsWAKAPo1ADiBuLttTqtnnLtaNuv2760Ch9ec5YZaQm+R0LANDHUagBxLy2zqB+/vf1+t3rm1U8oJ+evOkkzSjJ8zsWAACSKNQAYtza3Q368hPLtXZ3o66cWaxvfWSiMtP40QUAiB38qQQgJgVDTve9sVk/e3m9cjJSdP+nZ+jsCazgAQCIPRRqADFne22Lbpu3Qou21OrDxxXpvy6ZrIGsKw0AiFEUagAxwzmnp8oqdddfyiVJP718ii6bxm6HAIDYRqEGEBP2NrXpG0+v0svlVZo1Mk8/vXyKivP6+R0LAIAjolAD8N388ird8fRKNezv1J0XTND1p45UIMBVaQBAfKBQA/BNU1un/vP5cj2xeLsmDM7RozdM1bhB2X7HAgDgqFCoAfhi8dZaffXJ8CYtnztjtL48p1RpyWzSAgCIPxRqAL0qFHK695WN+sX89RrGJi0AgARAoQbQa2qb2/Xl/1uu19fX6KNTh+g/L5msLDZpAQDEOf4kA9Aryipqdetjy7S3uV3/dclkXTmzmOXwAAAJgUINwFPOOd2/YIt+9OJaDemfoac/d7ImDc31OxYAAD2GQg3AM/v2d+hrT63QS6ur9OHjivSTj01RbkaK37EAAOhRFGoAnnhnxz59/tGl2lm/X9/6SHhtaUY8AACJiEINoEc55/TYom266y/lyuuXqv+7abamj2AVDwBA4qJQA+gxzW2duvOZVXp2+U59aGyBfnHFFA3MSvM7FgAAnqJQA+gRG6oa9blHl2pTTZO+es5Y3XrmGLYPBwD0CRRqAN32zLJKffPpd5SZlqRHrp+lU8bk+x0JAIBeQ6EGcMxaO4K66y/lenzRNs0sydM9V52gopx0v2MBANCrKNQAjsn22hbd9HCZync16ObTR+v2c8cqOSngdywAAHodhRrAUVtZWa/PPLhE7Z1B3fepGZozscjvSAAA+IZCDeCo/HNtlW55dJnyMlP1xI2zNKYw2+9IAAD4ikINIGqPvb1N33p2lSYOydEfPn2iCpmXBgCAQg3gyJxz+n8vrdP/vrpJZ4wr0L1XTVNmGj8+AACQKNQAjqC9M6SvPbVCzy7fqU+cWKz//Ogkbj4EAKALCjWAQ9q3v0M3P1ymtzbv1e3njtUtZ46RGZu1AADQFYUawEHtrN+vax9YpM01zfr5FVN06bRhfkcCACAmUagBfED5zgZd9+AitbQF9dBnZrLzIQAAh0GhBvAeb2yo0eceWars9GTN+9xJGj8ox+9IAADENAo1gHfNW7Jd33h6lcYUZumB607U4NwMvyMBABDzKNQA5JzTr/6xQb+cv0GnjsnXr6+epuz0FL9jAQAQFyjUQB/XEQzpzmdW6ckllbps2jD98NLJSk1mWTwAAKJFoQb6sKa2Tn3+0aV6fX2Nvnh2qb4yp5Rl8QAAOEoUaqCPqmls03UPLtKaXY368WWT9fETh/sdCQCAuEShBvqg7bUtuub+t7W7oVX3fXqGzhxX6HckAADiFoUa6GPW7GrQp/6wSO2dIT16w2xNHzHA70gAAMQ1CjXQh7y9ea9u+OMSZaUl67GbT1JpUbbfkQAAiHsUaqCPeHn1bt36+DIVD8jQH6+fpaH9WWMaAICeQKEG+oAnF2/XHU+v1ORh/fXAtScqLzPV70gAACQMCjWQwJxz+s1rm/Xjv63VaaX5+s3V05WZxm97AAB6En+yAgkqFHL6rxfW6L4FWzR3yhD99PIpbNgCAIAHKNRAAuoIhvS1p1bqmWU7dO3JJfrOhRMVCLBhCwAAXqBQAwmmpT28++Gr62p0+7ljdcuZY9j9EAAAD3n6979mdp6ZrTOzjWZ2x0E+H25mr5jZMjNbaWYXeJkHSHT1Le365H1v6/X1NfrhpZN161lsJQ4AgNc8u0JtZkmS7pV0jqRKSYvN7DnnXHmX074l6Unn3K/NbKKkFySVeJUJSGS79u3Xp+5fpIraFv3vJ6fpvEmD/Y4EAECf4OXIx0xJG51zmyXJzJ6QdLGkroXaScqJvM6VtNPDPEDC2ljdpE/d/7YaWjv10HUzddLogX5HAgCgz/CyUA+VtL3LcaWkWe8753uSXjazL0jKlDTHwzxAQlq+vV7XPbBISQHTEzfO1qShuX5HAgCgT/Fyhvpgg5vufcdXSnrQOTdM0gWSHjazD2QysxvNbImZLampqfEgKhCfFm7eq0/+fqGy01P01M0nU6YBAPCBl4W6UlJxl+Nh+uBIx/WSnpQk59xbktIl5b//Gznnfuecm+Gcm1FQUOBRXCC+vL6+Rtc+sEiD+2do3s0nqSQ/0+9IAAD0SV4W6sWSSs1spJmlSvqEpOfed842SWdLkplNULhQcwkaOIL55VW64aElKhmYqSdunK2inHS/IwEA0Gd5Vqidc52SbpX0kqQ1Cq/msdrM7jazuZHTbpP0WTNbIelxSdc6594/FgKgi7+u3KWbHynT+MHZeuLG2crPSvM7EgAAfZqnG7s4515QeCm8ru99p8vrckmneJkBSCTPLKvUbU+u0LThA/SH605UTnqK35EAAOjz2CkRiBNPLNqmbzyzSrNHDtR9n56hzDR++wIAEAv4ExmIAw+9uVXffW61Th9boN9eM13pKUl+RwIAABEUaiDG/fa1Tfrhi2t17sQi3XPVCUpLpkwDABBLKNRAjHLO6Vf/2KBfzt+gi6YM0c+vmKKUJC8X5gEAAMeCQg3EIOecfvy3dfrNa5v0senD9OPLjldS4GB7JQEAAL9RqIEY45zTXX8p14NvbtUnZw3X9y+epABlGgCAmEWhBmJIKOR057Pv6PFF2/SZU0bq2xdOkBllGgCAWEahBmJEZzCkrz21Uk8v26Fbzhyt288dR5kGACAOUKiBGNARDOnLTyzXX1ft0m3njNUXzi71OxIAAIgShRrwWVtnULc8ukzz11Tpzgsm6LMfGuV3JAAAcBQo1ICP2jtD75bpuy8+Tp86qcTvSAAA4CixqC3gk45gSF94fCllGgCAOEehBnzQEQzpS08s00urq/S9iyZSpgEAiGMUaqCXdQZD+sr/LdcLq3brWx+ZoGtPGel3JAAA0A0UaqAXBUNOt81boedX7tI3LxivG07jBkQAAOIdhRroJcGQ0+3zVujPy3fq6+eN140fGu13JAAA0AMo1EAvCIWcvvbUSj2zbIduP3esPncGZRoAgERBoQY8Fgo53fH0Sv1paaW+Mmesbj2LTVsAAEgkFGrAQ6GQ053PrtKTSyr1xbPG6EtzKNMAACQaCjXgEeecvvPcO3p80XbdcuZofeWcsX5HAgAAHqBQAx5wzul7z63WIwu36abTR+n2c8fJzPyOBQAAPEChBnqYc053P1+uh96q0GdPG6k7zhtPmQYAIIFRqIEe5JzTD/66Rg/8a6s+c8pIffOCCZRpAAASHIUa6CHOOf3oxbW6b8EWXXtyib59IWUaAIC+gEIN9ADnnP7fS+v029c36+rZw/XdiyZSpgEA6CMo1EAPuH/BFv3vq5t05czhunvuJMo0AAB9CIUa6Kb55VX6wQtrdP6kQfrBRycpEKBMAwDQl1CogW4o39mgLz6xTJOG5OrnV0ylTAMA0AdRqIFjVN3YqhseWqyc9BTd9+kZykhN8jsSAADwQbLfAYB41NoR1Gf/WKa6lg7Nu/kkFeWk+x0JAAD4hEINHCXnnG6ft0IrK+v1m6una9LQXL8jAQAAHzHyARylX87foOdX7tLXzxuvDx83yO84AADAZxRq4Cj8efkO/eofG3T59GG66UOj/I4DAABiAIUaiFJZRZ3+46mVmjkyTz+4ZDJrTQMAAEkUaiAqlXUtuunhJRqcm67fXD1dqcn81gEAAGHclAgcQWNrh65/cInaOkN64sYTlZeZ6nckAAAQQyjUwGEEQ05femK5NtY06cHrTtSYwiy/IwEAgBjD31sDh/FfL6zRP9dW6665x+m00gK/4wAAgBhEoQYO4dG3K3T/gi267pQSXT17hN9xAABAjKJQAwfxr4179J0/r9YZ4wr0rY9M9DsOAACIYRRq4H021TTpc4+UaXRBpu658gQlBVgeDwAAHBqFGuiirrld1z+4WClJAd3/6ROVnZ7idyQAABDjWOUDiGjvDOnmR8q0s75Vj984S8V5/fyOBAAA4gCFGoi46y+r9faWWv3y41M1fUSe33EAAECcYOQDkPSnsko9+vY23fShUfroCUP9jgMAAOIIhRp93ppdDbrz2VWaPSpP//HhcX7HAQAAcSaqQm1mp5rZdZHXBWY20ttYQO/Yt79DNz9SptyMFN1z5TQlJ/HfmAAA4OgcsT2Y2XclfV3SNyJvpUh6xMtQQG8IhZxue3KFdtTt171XTVNBdprfkQAAQByK5nLcJZLmSmqWJOfcTknZXoYCesNvXt+k+Wuq9M0LJmhGCTchAgCAYxNNoW53zjlJTpLMLNPbSID33ty4Rz99aZ0uPH6wrjulxO84AAAgjkVTqJ80s99K6m9mn5U0X9LvvY0FeGfXvv36wuPLNKogSz++7HiZsRMiAAA4dkdch9o591MzO0dSg6Rxkr7jnPu758kAD7R3hvT5R5eqtSOo31w9XZlpLMUOAAC654htIrKixxsHSrSZZZhZiXNuq9fhgJ72Xy+s0bJt9br3qmkaU5jldxwAAJAAohn5mCcp1OU4GHkPiCt/Xr5DD765VdefOlIfOX6w33EAAECCiKZQJzvn2g8cRF6nehcJ6Hnrqxp1x59W6cSSAbrj/PF+xwEAAAkkmkJdY2ZzDxyY2cWS9ngXCehZja0duvnhMmWmJeveq6Yphc1bAABAD4rmjqybJT1qZv8jySRtl/QpT1MBPcQ5p689tVIVtS167IZZKsxJ9zsSAABIMNGs8rFJ0mwzy5JkzrlG72MBPeO+N7boxXd2684LJmjWqIF+xwEAAAkomlU+0iRdJqlEUvKBNXudc3d7mgzoprc379WP/rZW508apBtOG+l3HAAAkKCiGfn4s6R9ksoktXkbB+gZ1Q2tuuWxZRqR108/+RibtwAAAO9EU6iHOefO8zwJ0EM6giHd8thSNbd16rHPzlJ2eorfkQAAQAKLZrmDN81ssudJgB7y4xfXavHWOv3osskaW5TtdxwAAJDgorlCfaqka81si8IjHybJOeeO9zQZcAz+unKX7luwRdeeXKKLpw71Ow4AAOgDoinU53ueAugBW/Y06+t/WqkThvfXNy+Y4HccAADQR0SzbF6FJJlZoSQW8UVMau0I6vOPLlVykuneq6YpNZnNWwAAQO84Yusws7lmtkHSFkmvSdoq6UWPcwFH5e7ny7VmV4N+ccVUDemf4XccAADQh0RzGe/7kmZLWu+cGynpbEn/8jQVcBT+vHwMp+9yAAAgAElEQVSHHnt7m24+fbTOHF/odxwAANDHRFOoO5xzeyUFzCzgnHtF0lSPcwFR2VTTpG8+vUozRgzQ7eeO9TsOAADog6K5KbE+su3465IeNbNqSZ3exgKObH97ULc8ulRpKUm656oTlJzE3DQAAOh90TSQiyXtl/QVSX+TtEnSRV6GAqJx119Wa+3uRv38iikanMvcNAAA8Ec0q3w0dzl8yMMsQNSeWVapJxZv1y1njtYZ45ibBgAA/jlkoTazBc65U82sUZLr+pHCG7vkeJ4OOIiN1Y365tPvaObIPH1lDnPTAADAX4cs1M65UyPP7N2MmNHS3qnPP7pU/VKTdM+VzE0DAAD/HbaNmFnAzN7prTDAkXz3z6u1obpJv/zEVBXlsM8QAADw32ELtXMuJGmFmQ3vpTzAIT1VVql5ZZX6wpljdFppgd9xAAAAJEW3bN5gSavNbJGkd29QdM7N9SwV8D7rqxr1rWdX6aRRA/Ul5qYBAEAMiaZQ3+V5CuAwmtvCc9NZaSn61ZVTlRQwvyMBAAC8K5pl817rjSDAwTjn9O1n39GmmiY9cv0sFWYzNw0AAGLLEZdIMLPZZrbYzJrMrN3MgmbW0BvhgHlLKvX0sh360tmlOmVMvt9xAAAAPiCaNcf+R9KVkjZIypB0Q+Q9wFNrdzfo239+R6eOydcXzir1Ow4AAMBBRTNDLefcRjNLcs4FJT1gZm96nAt9XFNkbjonI0W/+Dhz0wAAIHZFU6hbzCxV0nIz+4mkXZIyvY2Fvsw5pzufWaWte5r16A2zVZCd5nckAACAQ4pm5OOayHm3KrxsXrGky7wMhb7ticXb9eflO/WVOWN10uiBfscBAAA4rGiuUE+T9IJzrkEsoQePba5p0veeW63TSvN1y5lj/I4DAABwRNFcoZ4rab2ZPWxmHzGzqOaugaMVDDndPm+F0lOS9LPLpyjA3DQAAIgDRyzUzrnrJI2RNE/SVZI2mdl9XgdD3/OHBVu0dFu97pp7nApzWG8aAADEh2hX+egwsxclOYWXzrtY4eXzgB6xsbpJ/+/ldTp3YpEunjrE7zgAAABRi2Zjl/PM7EFJGyV9TNJ9kgZ7nAt9yIFRj8zUJP3gkskyY9QDAADEj2iuUF8r6QlJNznn2ryNg77o929s1vLt9frvK09giTwAABB3jlionXOf6I0g6Js2VDXq5y+v1/mTBumi4/mLDwAAEH+iWeUD8ERnMKTb5q1QVnqyvv/RSYx6AACAuMQSePDNb1/frJWV+3TvVdOUn8WoBwAAiE9coYYv1u5u0C/nr9dHjh+sjzDqAQAA4tghr1Cb2SqFl8k7KOfc8Z4kQsLrCIZ0+7wVys1I0fcvnuR3HAAAgG453MjHhZHnWyLPD0eePympxbNESHi/fnWT3tnRoN9cPV15mal+xwEAAOiWQxZq51yFJJnZKc65U7p8dIeZ/UvS3V6HQ+Ip39mg//7HBs2dMkTnTRrkdxwAAIBui2aGOtPMTj1wYGYnS8r0LhISVXtneNSjf79U3TX3OL/jAAAA9IhoVvm4XtIfzCxX4ZnqfZI+42kqJKR7X9mo8l0N+t010zWAUQ8AAJAgotnYpUzSFDPLkWTOuX3ex0KieWfHPt37ykZdesJQnXscox4AACBxHHHkw8yKzOx+Sf/nnNtnZhPN7PpeyIYEcWDUIy8zVd+9iFEPAACQWKKZoX5Q0kuShkSO10v6sleBkHju+ecGrd3dqB9eOlm5/VL8jgMAANCjoinU+c65JyWFJMk51ykp6GkqJIyVlfX631c36WPTh+nsCUV+xwEAAOhx0RTqZjMbqMgmL2Y2W+EbE4HDausM6rYnV6ggK03fvnCi33EAAAA8Ec0qH1+V9Jyk0ZH1pwskfczTVEgIv5y/QRuqm/TAdScqN4NRDwAAkJiiWeVjqZmdLmmcJJO0zjnX4XkyxLXl2+v129c26eMzinXmuEK/4wAAAHgmmivUkjRTUknk/GlmJufcHz1LhbgWHvVYrkE56brzwgl+xwEAAPDUEQu1mT0sabSk5fr3zYhOEoUaB/Xc8p3aVNOs+z89QznpjHoAAIDEFs0V6hmSJjrnnNdhEP+cc7p/wRaNH5Sts8Yz6gEAABJfNKt8vCOJre0QlTc37dXa3Y36zKkjZWZ+xwEAAPBcNFeo8yWVm9kiSW0H3nTOzfUsFeLW/Qu2KD8rVXOnDDnyyQAAAAkgmkL9Pa9DIDFsrG7SP9dW68tzSpWekuR3HAAAgF4RzbJ5r/VGEMS/B/61RanJAV09e4TfUQAAAHrNIWeozWxB5LnRzBq6PBrNrCGab25m55nZOjPbaGZ3HOKcK8ys3MxWm9ljx/Y/A36ra27Xn5ZW6pKpQ5WfleZ3HAAAgF5zyCvUzrlTI8/Zx/KNzSxJ0r2SzpFUKWmxmT3nnCvvck6ppG9IOsU5V2dmLAsRpx5btE2tHSF95tSRfkcBAADoVdFu7KJI2U0/cOyc23aEL5kpaaNzbnPk65+QdLGk8i7nfFbSvc65usj3rI42D2JHe2dID725VaeV5mvcoGP67y8AAIC4dcRl88xsrpltkLRF0muStkp6MYrvPVTS9i7HlZH3uhoraayZ/cvMFprZeYfIcKOZLTGzJTU1NVH80uhNz6/cqerGNl3P1WkAANAHRbMO9fclzZa03jk3UtLZkv4VxdcdbBHi928OkyypVNIZkq6UdJ+Z9f/AFzn3O+fcDOfcjIKCgih+afSWAxu5jCnM0ulj+XcDAAD6nmgKdYdzbq+kgJkFnHOvSJoaxddVSirucjxM0s6DnPNn51yHc26LpHUKF2zEibe31Gr1zgZdz0YuAACgj4qmUNebWZak1yU9ama/ktQZxdctllRqZiPNLFXSJyQ9975znpV0piSZWb7CIyCbow0P/933xhblZabqkhPeP80DAADQN0RTqC+WtF/SVyT9TdImSRcd6Yucc52SbpX0kqQ1kp50zq02s7vN7MAuiy9J2mtm5ZJekfQfkavhiANb9zTrH2urdPWs4WzkAgAA+qxoNnZp7nL40NF8c+fcC5JeeN973+ny2kn6auSBOPPAv7YoJRDQ1SexkQsAAOi7DlmozaxR772J0CLHpnAXzvE4G2LYvpYOPbmkUhdNGaLC7PQjfwEAAECCOtzGLiwojEN6fPE27e8IslQeAADo86La2MXMpkk6VeEr1Aucc8s8TYWY1hEMb+Ry8uiBmjiEv6gAAAB9WzQbu3xH4dnpgZLyJT1oZt/yOhhi1wurdmnXvlauTgMAACi6K9RXSjrBOdcqSWb2I0lLJf2nl8EQmw5s5DIqP1Nnjiv0Ow4AAIDvolk2b6ukrnedpSm8dB76oCUVdVpZuU/XnTpSgQAbuQAAAERzhbpN0moz+7vCM9TnSFpgZv8tSc65L3qYDzHm/je2KDcjRZdNYyMXAAAAKbpC/UzkccCr3kRBrNu2t0Uvl+/WzaePVr/UqO5nBQAASHjRtKIXnXPVXd8ws3HOuXUeZUKMeuDNLQqY6VMnlfgdBQAAIGZEM0P9hpldceDAzG7Te69Yow9oaO3Qk4u366IpQzQol41cAAAADojmCvUZkn5nZpdLKpK0RtJML0Mh9jy5eLua29nIBQAA4P2OeIXaObdL0t8knSSpRNIfnXNNHudCDOkMhvTAv7Zq5sg8TRqa63ccAACAmBLNxi5/lzRL0iRJF0j6hZn91OtgiB0vra7Sjvr9uoGr0wAAAB8QzQz1vc65Tznn6p1z70g6WdI+j3Mhhty/YLNGDOynsycU+R0FAAAg5kQz8vGsmY0wszmRt1Ik/dLbWIgVS7fVaem2el13comS2MgFAADgA6IZ+fispKck/Tby1jBJz3oZCrHj/gVblJ2erMtnFPsdBQAAICZFM/Jxi6RTJDVIknNug6RCL0MhNlTWtejFVbt01czhykxjIxcAAICDiaZQtznn2g8cmFmywluQI8E99OZWmZk+fXKJ31EAAABiVjSF+jUz+6akDDM7R9I8SX/xNhb81tTWqScWbdcFkwdrSP8Mv+MAAADErGgK9R2SaiStknSTpBckfcvLUPDfU0u2q7Gtk41cAAAAjuCIg7HOuZCk30ce6AOcc3p4YYWmFPfX1OL+fscBAACIadFcoUYf89amvdpU06xrZo/wOwoAAEDMo1DjAx5eWKEB/VJ04fGD/Y4CAAAQ86Iu1GaW6WUQxIZd+/br5fIqXXFisdJTkvyOAwAAEPOi2djlZDMrl7QmcjzFzP7X82TwxeNvb1PIOV09i3EPAACAaERzhfoXkj4saa8kOedWSPqQl6Hgj/bOkB5fvF1njitUcV4/v+MAAADEhahGPpxz29/3VtCDLPDZS6t3q6axjZsRAQAAjkI0+0lvN7OTJTkzS5X0RUXGP5BYHl5YoeK8DJ0+tsDvKAAAAHEjmivUN0u6RdJQSZWSpkaOkUDW7m7Qoi21unrWCAUC5nccAACAuBHNFWpzzn3S8yTw1SMLK5SWHNAVM4r9jgIAABBXorlC/aaZvWxm15sZ2+YloMbWDj2zdIcumjJEAzJT/Y4DAAAQV45YqJ1zpZK+Jek4SUvN7Hkzu9rzZOg1Ty/doeb2IDcjAgAAHINoV/lY5Jz7qqSZkmolPeRpKvQa55weXlihKcNyNaWYv4AAAAA4WtFs7JJjZp82sxclvSlpl8LFGgngrc17tbG6SVdzdRoAAOCYRHNT4gpJz0q62zn3lsd50MseWVih/v1SdNGUIX5HAQAAiEvRFOpRzjnneRL0ut37WvXS6ipdf+pIpack+R0HAAAgLh2yUJvZL51zX5b0nJl9oFA75+Z6mgyee2zRNoWc09WzGPcAAAA4Voe7Qv1w5PmnvREEvasjGNLji7bpjLEFGj6wn99xAAAA4tYhC7Vzrizycqpz7lddPzOzL0l6zctg8NZLq3erprFN15zE1WkAAIDuiGbZvE8f5L1rezgHetnDb1WoOC9Dp48t9DsKAABAXDvcDPWVkq6SNNLMnuvyUbakvV4Hg3fW7W7U21tqdcf545UUML/jAAAAxLXDzVAfWHM6X9LPurzfKGmll6HgrUcWVig1OaArZhT7HQUAACDuHW6GukJShaSTei8OvNbY2qGnl1bqouOHKC8z1e84AAAAcS+anRJnm9liM2sys3YzC5pZQ2+EQ897ZtkONbcHuRkRAACgh0RzU+L/SLpS0gZJGZJukHSPl6HgDeecHn6rQscPy9XU4v5+xwEAAEgI0RRqOec2SkpyzgWdcw9IOtPbWPDCws212lDdpKtnc3UaAACgp0Sz9XiLmaVKWm5mP1H4RsVMb2PBC48srFBuRormThnidxQAAICEEc0V6mskJUm6VVKzpGJJl3kZCj2vqqFVL63erStmDFN6SpLfcQAAABLGEa9QR1b7kKT9ku7yNg688viibQo6x7gHAABADzvcxi6rJLlDfe6cO96TROhxHcGQHnt7m04fW6ARA5nWAQAA6EmHu0J9Ya+lgKdeXl2l6sY2/ZCr0wAAAD3uSBu7IAE8vHCrhg3I0BnjCv2OAgAAkHCi2dil0cwaIo9WNnaJL+urGrVwc60+OWuEkgLmdxwAAICEE81Nidldj83so5JmepYIPeqRhRVKTQ7o4ycW+x0FAAAgIUW1sUtXzrlnJZ3lQRb0sKa2Tj29dIcunDxYeZmpfscBAABISEe8Qm1ml3Y5DEiaocOs/oHY8cyyHWpq69Q1J3EzIgAAgFei2Snxoi6vOyVtlXSxJ2nQo55fsVPjirI1tbi/31EAAAASVjQz1Nf1RhD0rPqWdi2pqNPnzxgtM25GBAAA8Eo0Ix8jJX1BUknX851zc72Lhe56dV2NgiGnOROK/I4CAACQ0KIZ+XhW0v2S/iIp5G0c9JS/r6lSQXaaJg/N9TsKAABAQoumULc65/7b8yToMe2dIb22rkYXTRmsAGtPAwAAeCqaQv0rM/uupJcltR140zm31LNU6Ja3t+xVU1sn4x4AAAC9IJpCPVnSNQqvPX1g5MOJtahj1vzyKqWnBHTKmHy/owAAACS8aAr1JZJGOefavQ6D7nPOaf6aap1WWqD0lCS/4wAAACS8aHZKXCGJhYzjxJpdjdpRv1/nMO4BAADQK6K5Ql0kaa2ZLdZ7Z6hZNi8GzV9TJTPpzPGFfkcBAADoE6Ip1N/1PAV6zD/WVGlqcX8VZKf5HQUAAKBPiGanxNd6Iwi6r6qhVSsq9+k/PjzO7ygAAAB9RjQ7JTYqvKqHJKVKSpHU7JzL8TIYjt4/1lRLks6ZyPw0AABAb4nmCnV212Mz+6ikmZ4lwjGbv6ZKw/P6qbQwy+8oAAAAfUY0q3y8h3PuWbEGdcxpae/Ugo17NGdCkczYHREAAKC3RDPycWmXw4CkGfr3CAhixBsb9qi9M6Q5E1ndAwAAoDdFs8rHRV1ed0raKuliT9LgmM0vr1JOerJOLMnzOwoAAECfEs0M9XW9EQTHLhhy+ufaap0xrlApSUc9xQMAAIBuOGL7MrOHzKx/l+MBZvYHb2PhaCzfXq+9ze2aw+oeAAAAvS6ay5nHO+fqDxw45+okneBdJByt+WuqlBwwnT62wO8oAAAAfU40hTpgZgMOHJhZnqKbvUYvmV9epVmj8pSbkeJ3FAAAgD4nmmL8M0lvmtlTCq/ucYWkH3iaClHbuqdZG6qbdNWs4X5HAQAA6JOiuSnxj2a2ROG1p03Spc65cs+TISrz11RJkuZMYH4aAADAD1GNbkQKNCU6Bs1fU6Xxg7JVnNfP7ygAAAB9EmusxbF9LR1avLWOq9MAAAA+olDHsVfXVysYcjp7ArsjAgAA+IVCHcf+Xl6l/Kw0TRnW/8gnAwAAwBMU6jjV3hnSa+tqNGdCoQIB8zsOAABAn0WhjlOLttSqsa2T+WkAAACfUajj1Pw1VUpPCeiUMfl+RwEAAOjTKNRxyDmnv5dX6dQxBcpITfI7DgAAQJ9GoY5Da3c3akf9fp0zkdU9AAAA/EahjkPzy8O7I545nkINAADgNwp1HJq/tlpTi/urMDvd7ygAAAB9HoU6zlQ3tGrF9nqdM5HVPQAAAGIBhTrO/GNttSSxXB4AAECMoFDHmfnlVSrOy9DYoiy/owAAAEAU6rjS0t6pBRv3aM6EIpmxOyIAAEAsoFDHkQUb9qitM6RzGPcAAACIGRTqODJ/TZWy05N14sg8v6MAAAAggkIdJ0Ihp3+urdYZ4wqVksS/NgAAgFhBM4sTyyvrtaepXXMmsJkLAABALKFQx4n55VVKDpjOGEuhBgAAiCUU6jgxf02VZo7MU26/FL+jAAAAoAsKdRyo2Nus9VVNbOYCAAAQgyjUcWD+GnZHBAAAiFUU6jgwv7xKY4uyNHxgP7+jAAAA4H0o1DFuX0uHFm2t5eo0AABAjKJQx7hX11crGHKaM5FCDQAAEIso1DFu/ppq5Welauqw/n5HAQAAwEFQqGOYc05vbtyjD5UWKBAwv+MAAADgICjUMaxib4v2NrdrRkme31EAAABwCBTqGFZWUSdJmj5igM9JAAAAcCgU6hhWtq1O2enJKi3M8jsKAAAADoFCHcOWVtRp2vABzE8DAADEME8LtZmdZ2brzGyjmd1xmPM+ZmbOzGZ4mSeeNLR2aF1VI+MeAAAAMc6zQm1mSZLulXS+pImSrjSziQc5L1vSFyW97VWWeLR8W72cY34aAAAg1nl5hXqmpI3Ouc3OuXZJT0i6+CDnfV/STyS1epgl7iypqFPApCnFrD8NAAAQy7ws1EMlbe9yXBl5711mdoKkYufc8x7miEtLK+o0flCOstKS/Y4CAACAw/CyUB/sTjr37odmAUm/kHTbEb+R2Y1mtsTMltTU1PRgxNgUDDkt21bHuAcAAEAc8LJQV0oq7nI8TNLOLsfZkiZJetXMtkqaLem5g92Y6Jz7nXNuhnNuRkFBgYeRY8O63Y1qbg9SqAEAAOKAl4V6saRSMxtpZqmSPiHpuQMfOuf2OefynXMlzrkSSQslzXXOLfEwU1wo28aGLgAAAPHCs0LtnOuUdKuklyStkfSkc261md1tZnO9+nUTwdKKOhVmp2nYgAy/owAAAOAIPL3jzTn3gqQX3vfedw5x7hleZoknZRXh+WkzNnQBAACIdeyUGGOqG1u1rbaFcQ8AAIA4QaGOMUsrwvPT0yjUAAAAcYFCHWPKKuqUmhzQcUNy/I4CAACAKFCoY0xZRZ2OH5qrtOQkv6MAAAAgChTqGNLaEdQ7OxqYnwYAAIgjFOoYsnrnPrUHQxRqAACAOEKhjiFl3JAIAAAQdyjUMaSsok4lA/spPyvN7ygAAACIEoU6RjjnVFZRx9VpAACAOEOhjhHbalu0p6md+WkAAIA4Q6GOEQfmpynUAAAA8YVCHSPKKuqUnZas0sJsv6MAAADgKFCoY0RZRZ1OGDFASQHzOwoAAACOAoU6BjS2dmhdVaOmD2fcAwAAIN5QqGPA8u31co75aQAAgHhEoY4BS7bWKWDSlOJcv6MAAADgKFGoY8DSbXUaNyhH2ekpfkcBAADAUaJQ+ywYclq2rV7TR/T3OwoAAACOAYXaZ+urGtXU1sn8NAAAQJyiUPvswIYuM0bk+ZwEAAAAx4JC7bOlFXUqyE7TsAEZfkcBAADAMaBQ+6xsW52mDx8gMzZ0AQAAiEcUah/VNLapYm8L89MAAABxjELtowPz09Mo1AAAAHGLQu2jpdvqlJoU0KShOX5HAQAAwDGiUPuorKJOk4flKi05ye8oAAAAOEYUap+0dQa1qnIf89MAAABxjkLtk3d2NKg9GKJQAwAAxDkKtU+WHrghcTiFGgAAIJ5RqH1SVlGnEQP7qSA7ze8oAAAA6AYKtQ+cc1pSEd7QBQAAAPGNQu2D7bX7taepjfWnAQAAEgCF2gdl22oliRsSAQAAEgCF2gdlFXXKTkvW2KJsv6MAAACgmyjUPiirqNfU4f2VFDC/owAAAKCbKNS9rLG1Q+t2NzDuAQAAkCAo1L1sxfZ9CjnmpwEAABIFhbqXLamolZk0tbi/31EAAADQAyjUvaysok7jirKVnZ7idxQAAAD0AAp1LwqGnJZvq2fcAwAAIIFQqHvRhupG/f/27j1Yzruu4/j7k0uTtmlIS0MpSWgLFKFtkpaWCnKxVMQKDBfBgSqIyIAVEBVBQWe84NQRq4IzMiJgLToIgnIpCEMZaK3KrReaTaBWSmnMJoWkkHOS0DZpkq9/7BPYpuec5GR3z27Oeb9mOtl99tlnv/vtb85+zrO/5/x27t5roJYkSZpFDNQz6KaN2wG44LSThlyJJEmS+sVAPYNu2ridk5csYtVJxw67FEmSJPWJgXoG3bxxO+eftozEBV0kSZJmCwP1DLl7127u/N49zp+WJEmaZQzUM+TA/GkDtSRJ0uxioJ4hN2/czjHz53H2Ix4y7FIkSZLURwbqGXLTxu2cs2IpixfOH3YpkiRJ6iMD9QzYvXcfrc3jTveQJEmahQzUM+DrW3awZ+9+zvfvT0uSJM06BuoZcHNzQeITTls25EokSZLUbwbqGXDTxu088qTjeNgJi4ddiiRJkvrMQD1gVcWNG7c7f1qSJGmWMlAP2Jbx+9i2czfnPdLpHpIkSbORgXrAWpvGAFiz0kAtSZI0GxmoB6y1eZyF88PjTz1h2KVIkiRpAAzUA9Zqj/G4hy9l0QIXdJEkSZqNDNQDtH9/0WqPs3qly41LkiTNVgbqAdr4/XvYed9e1hqoJUmSZi0D9QC12l6QKEmSNNsZqAdo3aZxFi+cx5kPWzLsUiRJkjQgBuoBWr95jLMf8RAWzLfNkiRJs5VJb0D27tvPhs07WOP8aUmSpFnNQD0gt2/bxb337zNQS5IkzXIG6gFptccBL0iUJEma7QzUA9Jqj3HCogWc8dDjh12KJEmSBshAPSCt9jjnrHgI8+Zl2KVIkiRpgAzUA7B77z5uvWsHa1Y5f1qSJGm2M1APwG3f2cn9+4q1zp+WJEma9QzUA7CuuSBx9QrPUEuSJM12BuoBWN8e46Tjj2HliccOuxRJkiQNmIF6AFrtcdasfAiJFyRKkiTNdgbqPrtnz17+97s7WeN0D0mSpDnBQN1n39iyg/3lgi6SJElzhYG6z9b9cIVEz1BLkiTNBQbqPmu1x3j40sU8bOniYZciSZKkGWCg7rP1zQWJkiRJmhsM1H00fu/93HH3D1i7yvnTkiRJc4WBuo82bHZBF0mSpLnGQN1HLS9IlCRJmnMM1H3Uao9x2kOPY9lxxwy7FEmSJM0QA3UftdrjTveQJEmaYwzUfXL3rt1sHruXtS7oIkmSNKcYqPtkvfOnJUmS5iQDdZ+sa4+RwNlO+ZAkSZpTDNR9sr49zmOWL2HJogXDLkWSJEkzyEDdB1XFuvY4a5w/LUmSNOcYqPvgrvH7uHvXbudPS5IkzUEG6j5wQRdJkqS5y0DdB632GAvmhcefunTYpUiSJGmGGaj7oNUe58cefgKLF84fdimSJEmaYQbqHlUVrfaYFyRKkiTNUQbqHm383j3suG8va50/LUmSNCcZqHu0rj0GwGoDtSRJ0pxkoO7R+vY4ixbM47GnnDDsUiRJkjQEBuoetdrjnP2IpSycbyslSZLmIlNgD/btLzZscYVESZKkucxA3YNvbdvFPXv2uaCLJEnSHGag7sG6TZ0LEj1DLUmSNHcZqHvQao+zZNECHnXy8cMuRZIkSUNioO5Ba/M456xYyrx5GXYpkiRJGhID9RHas3c/t27ZwVqne0iSJM1pBuojdNt3drJn334XdJEkSZrjDNRH6MAKiZ6hliRJmtsM1EdofXucE49byMoTjx12KZIkSRoiA/URWtceY/XKZSRekChJkjSXGaiPwL179vHNrbtY6/xpSZKkOc9AfQS+cdc4+/aXC7pIkiTJQH0k1m0aB3DJcUmSJBmoj0SrPcYpSxdxytLFwy5FklHd2GMAAAxVSURBVCRJQ2agPgKtzeNO95AkSRJgoJ62Hffdzx3bfsCaFU73kCRJ0oADdZJLktyW5PYkb5ng8Tcm+UaSVpLPJzltkPX0w4Z2M396lWeoJUmSNMBAnWQ+8C7gZ4GzgEuTnHXQbl8DLqiqNcC/An8+qHr6pbW5CdSeoZYkSRKDPUN9IXB7Vd1RVXuADwHP796hqq6tqnuau18GVg6wnr5otcdYddKxnHj8McMuRZIkSSNgkIF6BbCp63672TaZVwGfmeiBJK9JcmOSG7dt29bHEqdv3SYvSJQkSdKPDDJQT7Qmd024Y/Iy4ALgioker6r3VNUFVXXB8uXL+1ji9Hxv1242j93rComSJEn6oQUDPHYbWNV1fyWw5eCdkjwT+H3gJ6tq9wDr6dmB+dOrV3iGWpIkSR2DPEN9A3BmkjOSHAO8FLi6e4ck5wF/BzyvqrYOsJa+aG0aJ4HVnqGWJElSY2CBuqr2Aq8HPgvcCny4qr6e5G1JntfsdgWwBPhIkluSXD3J4UbC+s1jPHr5EpYsGuSJfUmSJB1NBpoMq+rTwKcP2vYHXbefOcjX76eqYl17nKc95uRhlyJJkqQR4kqJh+k7O+5j287drHG6hyRJkroYqA9TyxUSJUmSNAED9WFqtcdYMC+cderSYZciSZKkEWKgPkyt9jiPPeUEFi+cP+xSJEmSNEIM1Iehqmi1x1m7yvnTkiRJeiAD9WH4v+/fw/i997ugiyRJkh7EQH0Y7t61h9Mfepx/4UOSJEkP4golh+H8007kujc/Y9hlSJIkaQR5hlqSJEnqgYFakiRJ6oGBWpIkSeqBgVqSJEnqgYFakiRJ6oGBWpIkSeqBgVqSJEnqgYFakiRJ6oGBWpIkSeqBgVqSJEnqgYFakiRJ6oGBWpIkSeqBgVqSJEnqgYFakiRJ6oGBWpIkSeqBgVqSJEnqgYFakiRJ6oGBWpIkSeqBgVqSJEnqgYFakiRJ6oGBWpIkSeqBgVqSJEnqgYFakiRJ6kGqatg1TEuSbcDGIb38ycDdQ3rto5H9mh77NT32a3rs1/TYr+mxX9Njv6ZnmP06raqWH2qnoy5QD1OSG6vqgmHXcbSwX9Njv6bHfk2P/Zoe+zU99mt67Nf0HA39csqHJEmS1AMDtSRJktQDA/X0vGfYBRxl7Nf02K/psV/TY7+mx35Nj/2aHvs1PSPfL+dQS5IkST3wDLUkSZLUAwP1YUhySZLbktye5C3DrmfUJbkzyfoktyS5cdj1jKIkVybZmmRD17aTknwuyTebf08cZo2jZJJ+/VGSzc04uyXJs4dZ46hIsirJtUluTfL1JL/RbHd8TWCKfjm+JpFkcZKvJlnX9OyPm+1nJPlKM8b+Jckxw651FEzRr6uSfLtrjJ077FpHRZL5Sb6W5FPN/ZEfWwbqQ0gyH3gX8LPAWcClSc4ablVHhWdU1bmj/mduhugq4JKDtr0F+HxVnQl8vrmvjqt4cL8A3tGMs3Or6tMzXNOo2gv8dlU9HngS8LrmZ5bja2KT9QscX5PZDVxcVWuBc4FLkjwJeDudnp0JbAdeNcQaR8lk/QJ4c9cYu2V4JY6c3wBu7bo/8mPLQH1oFwK3V9UdVbUH+BDw/CHXpKNcVV0PfP+gzc8H3t/cfj/wghktaoRN0i9NoKruqqqbm9s76XworcDxNaEp+qVJVMeu5u7C5r8CLgb+tdnuGGtM0S9NIMlK4DnA+5r74SgYWwbqQ1sBbOq638YftodSwDVJbkrymmEXcxQ5parugs6HPPCwIddzNHh9klYzJcQpDAdJcjpwHvAVHF+HdFC/wPE1qeYr+VuArcDngG8BY1W1t9nFz8ouB/erqg6MscubMfaOJIuGWOIoeSfwO8D+5v5DOQrGloH60DLBNn+znNpTquoJdKbJvC7J04ddkGalvwUeTecr1LuAvxxuOaMlyRLg34DfrKodw65n1E3QL8fXFKpqX1WdC6yk803u4yfabWarGl0H9yvJOcBbgccBTwROAn53iCWOhCTPBbZW1U3dmyfYdeTGloH60NrAqq77K4EtQ6rlqFBVW5p/twIfo/PDVof23SSnAjT/bh1yPSOtqr7bfEjtB96L4+yHkiykEw4/UFUfbTY7viYxUb8cX4enqsaA6+jMP1+WZEHzkJ+VE+jq1yXNdKOqqt3AP+AYA3gK8Lwkd9KZYnsxnTPWIz+2DNSHdgNwZnOF6THAS4Grh1zTyEpyfJITDtwGngVsmPpZalwNvKK5/QrgE0OsZeQdCIeNF+I4A3443/DvgVur6q+6HnJ8TWCyfjm+JpdkeZJlze1jgWfSmXt+LfDiZjfHWGOSfv1P1y+4oTMneM6Psap6a1WtrKrT6eStL1TVL3IUjC0XdjkMzZ9LeicwH7iyqi4fckkjK8mj6JyVBlgA/LP9erAkHwQuAk4Gvgv8IfBx4MPAI4H/A36+qrwQj0n7dRGdr+MLuBP41QNzhOeyJE8F/hNYz4/mIP4enXnBjq+DTNGvS3F8TSjJGjoXhs2nc2Luw1X1tubn/4foTF/4GvCy5uzrnDZFv74ALKczpeEW4LKuixfnvCQXAW+qquceDWPLQC1JkiT1wCkfkiRJUg8M1JIkSVIPDNSSJElSDwzUkiRJUg8M1JIkSVIPDNSSNCBJrktywQy8zhuS3JrkA4N+rWFKsizJa4ddhyQdzEAtSSOoa1Www/Fa4NnNAgiz2TI671WSRoqBWtKcluT05uzue5N8Pck1zWpmDzjDnOTkZjlckvxyko8n+WSSbyd5fZI3Jvlaki8nOanrJV6W5ItJNiS5sHn+8UmuTHJD85zndx33I0k+CVwzQa1vbI6zIclvNtveDTwKuDrJbx20//wkf5FkfZJWkl9vtv9U87rrmzoWNdvvTPKnSb6U5MYkT0jy2STfSnJZs89FSa5P8rEk30jy7iTzmscubY65Icnbu+rYleTyJOua/pzSbF+e5N+aPtyQ5CnN9j9q6rouyR1J3tAc6s+ARye5JckVSU5tarmlec2nHfFAkKQeGKglCc4E3lVVZwNjwIsO4znnAL8AXAhcDtxTVecBXwJ+qWu/46vqJ+icWb2y2fb7dJbUfSLwDOCKJMc3jz0ZeEVVXdz9YknOB14J/DjwJODVSc6rqsuALcAzquodB9X4GuAM4LyqWgN8IMli4CrgJVW1ms6Kpr/W9ZxNVfVkOqsHXkVnud8nAW/r2udC4LeB1cCjgZ9L8gjg7cDFdFYYfGKSFxzoAfDlqloLXA+8utn+18A7mj68CHhf12s8DviZ5rX+MMlC4C3At6rq3Kp6M53+f7aqzgXW0lltTpJm3HS+UpSk2erbVXUgjN0EnH4Yz7m2qnYCO5OMA59stq8H1nTt90GAqro+ydIky4BnAc9L8qZmn8V0lgQH+NwkS4I/FfhYVf0AIMlHgafRWYZ3Ms8E3l1Ve5savp9kbfN+/7fZ5/3A64B3Nvev7nofS7re431N7QBfrao7mjo+2NR2P3BdVW1rtn8AeDrwcWAP8KnmuTcBP91V31lJDtS7NMkJze1/b5YW3p1kK3DKBO/vBuDKJmx/vOv/oSTNKAO1JMHurtv7gGOb23v50Td5i6d4zv6u+/t54M/WOuh5BQR4UVXd1v1Akh8HfjBJjZlk+1Qywesf6jjd7+Pg93jgfU32niZzf1UdeM6+ruPMA55cVfc+oMBOwD74/8mDPq+aX1KeDjwH+KckV1TVP05RhyQNhFM+JGlydwLnN7dffITHeAlAkqcC41U1DnwW+PU0yTHJeYdxnOuBFyQ5rpke8kI60zKmcg1w2YELHJu53f8DnJ7kMc0+Lwf+Y5rv6cIkZzRzp18C/BfwFeAnm7nm84FLD+O41wCvP3AnybmH2H8ncOAMNklOA7ZW1XuBvweeMM33IUl94RlqSZrcXwAfTvJy4AtHeIztSb4ILAV+pdn2J3SmWLSaUH0n8NypDlJVNye5Cvhqs+l9VTXVdA/ozEl+bPM69wPvraq/SfJK4CNN0L4BePc039OX6FwguJpO0P9YVe1P8lbgWjpnqz9dVZ84xHHeALwrSYvO59H1wGWT7VxV30vy30k2AJ8BNgBvbt7bLh44d12SZkx+9C2cJElTS3IR8KaqmvIXAEmaS5zyIUmSJPXAM9SSJElSDzxDLUmSJPXAQC1JkiT1wEAtSZIk9cBALUmSJPXAQC1JkiT1wEAtSZIk9eD/ATUYHE1wXS6tAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "\n",
    "pca = PCA().fit(X)\n",
    "plt.plot(np.cumsum(pca.explained_variance_ratio_))\n",
    "plt.xlabel('number of components')\n",
    "plt.ylabel('cumulative explained variance');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\tPCA Component Variance Ratios\n",
      "Features: \t Ratio\n",
      " 5 \t 0.5578788822486604\n",
      " 10 \t 0.7484926261438216\n",
      " 15 \t 0.8627896457658564\n",
      " 20 \t 0.9329046072177467\n",
      " 25 \t 0.9688048240459584\n",
      " 30 \t 0.9874585036597339\n",
      " 35 \t 0.9968415910531903\n",
      " 40 \t 0.9999131600567793\n"
     ]
    }
   ],
   "source": [
    "ratio = pca.explained_variance_ratio_\n",
    "\n",
    "print(\"\\tPCA Component Variance Ratios\\nFeatures: \\t Ratio\")\n",
    "for i in range(5,41,5):\n",
    "    print(\" {} \\t {}\".format(i, np.sum(ratio[0:i])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1055, 2)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA(n_components=2)\n",
    "projected = pca.fit_transform(X)\n",
    "projected.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x180efac4320>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "RB_idx = np.where(y == 0)\n",
    "NRB_idx = np.where(y == 1) \n",
    "\n",
    "\n",
    "x_min, x_max = projected[:, 0].min() - 1, projected[:, 0].max() + 1\n",
    "y_min, y_max = projected[:, 1].min() - 1, projected[:, 1].max() + 1\n",
    "\n",
    "plt.scatter(projected[RB_idx, 0], projected[RB_idx, 1], c='b', edgecolor='k', alpha=0.6, label='RB')\n",
    "plt.scatter(projected[NRB_idx, 0], projected[NRB_idx, 1], c='r', edgecolor='k', alpha=0.6, label='NRB')\n",
    "plt.axis([x_min, x_max, y_min, y_max])\n",
    "plt.xlabel('Component 1')\n",
    "plt.ylabel('Component 2')\n",
    "plt.title('PCA n_Components = 2')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Pre-Processing\n",
    "\n",
    "#### Functions Used to Pre-Process Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def removeCorrelations(X, threshold):\n",
    "    \"\"\"\n",
    "    Drop Featurs with high correlations at a given threshold.\n",
    "    \n",
    "    Args\n",
    "    ----\n",
    "    X (PD DataFrame): Data with mxn, where m are instances and n is features\n",
    "    threshold (float): Threshold to drop features from data.\n",
    "    \n",
    "    Returns\n",
    "    -------\n",
    "    X_new (PD DataFrame): New data with dropped feature columns.\n",
    "    \"\"\"\n",
    "\n",
    "    X_corr = X.corr().abs()\n",
    "    traingle = X_corr.where(np.triu(np.ones(X_corr.shape), k=1).astype(np.bool))\n",
    "\n",
    "    drop_col = [column for column in traingle.columns if any(traingle[column] >= threshold)]   \n",
    "    X_new = X.drop(columns=drop_col)    \n",
    "    \n",
    "    return X_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def removeOutliers(X, y, threshold):\n",
    "    \"\"\"\n",
    "    Drop Featurs with high correlations at a given threshold.\n",
    "    \n",
    "    Args\n",
    "    ----\n",
    "    X (PD DataFrame): Data with mxn, where m are instances and n is features\n",
    "    threshold (float): Threshold to drop instances from data.\n",
    "    y (PD Series): Classification column to match instance drops.\n",
    "    \n",
    "    Returns\n",
    "    -------\n",
    "    X_new (PD DataFrame): New data with dropped istances.\n",
    "    y_new (PD Series): New classification column to math dropped instances.\n",
    "    \"\"\"\n",
    "    \n",
    "    list_fliers = []\n",
    "   \n",
    "    \n",
    "    for i in X.columns:\n",
    "        # Create figure to extract caps, but do not display figures\n",
    "        fig = plt.figure()\n",
    "        dict_boxPlot = plt.boxplot(X[i])\n",
    "        plt.close(fig)\n",
    "\n",
    "        lower_cap = dict_boxPlot['caps'][0].get_ydata()[0] \n",
    "        upper_cap = dict_boxPlot['caps'][1].get_ydata()[0] \n",
    "\n",
    "        idx_tuple = np.where((X[i] > upper_cap) | (X[i] < lower_cap))\n",
    "        idx_fliers = idx_tuple[0]\n",
    "\n",
    "        for q in idx_fliers:    \n",
    "            list_fliers.append(q)\n",
    "        i+=1\n",
    "\n",
    "    unique_fliers,y_counts = np.unique(list_fliers, return_counts=True)\n",
    "\n",
    "    drop_row_idx = np.where(y_counts >= threshold)\n",
    "    drop_row = unique_fliers[drop_row_idx]\n",
    "    \n",
    "    X_new = X.drop(drop_row)    \n",
    "    y_new = y.drop(drop_row)\n",
    "    return X_new, y_new"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Setting Threshold Parameters\n",
    "\n",
    "Threshold parameters are set to remove outliers and dropped highly correlated feature columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_orginal = X\n",
    "y_original = y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def resetThresholds(X_orginal,y_original, c_threshold, o_threshold):\n",
    "    X = X_orginal\n",
    "    y = y_original\n",
    "    \n",
    "    if c_threshold != 0:\n",
    "        X = removeCorrelations(X, c_threshold)\n",
    "    if o_threshold != 0:\n",
    "        X, y = removeOutliers(X, y_original, o_threshold)\n",
    "    return train_test_split(X, y, test_size=0.25, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Algorithm Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import VotingClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Voting classifier with a Logistic Regressor, SVM, and a Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>LogisticRegression</th>\n",
       "      <th>SVC</th>\n",
       "      <th>RandomForestClassifier</th>\n",
       "      <th>VotingClassifier</th>\n",
       "      <th>Correlation Threshold</th>\n",
       "      <th>Outlier Threshold</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.840909</td>\n",
       "      <td>0.893939</td>\n",
       "      <td>0.856061</td>\n",
       "      <td>0.867424</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.844697</td>\n",
       "      <td>0.893939</td>\n",
       "      <td>0.863636</td>\n",
       "      <td>0.871212</td>\n",
       "      <td>0.00</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.886275</td>\n",
       "      <td>0.929412</td>\n",
       "      <td>0.882353</td>\n",
       "      <td>0.921569</td>\n",
       "      <td>0.00</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.858300</td>\n",
       "      <td>0.874494</td>\n",
       "      <td>0.829960</td>\n",
       "      <td>0.866397</td>\n",
       "      <td>0.00</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.840909</td>\n",
       "      <td>0.890152</td>\n",
       "      <td>0.852273</td>\n",
       "      <td>0.867424</td>\n",
       "      <td>0.90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.893939</td>\n",
       "      <td>0.859848</td>\n",
       "      <td>0.867424</td>\n",
       "      <td>0.90</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.899225</td>\n",
       "      <td>0.934109</td>\n",
       "      <td>0.864341</td>\n",
       "      <td>0.918605</td>\n",
       "      <td>0.90</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.892000</td>\n",
       "      <td>0.908000</td>\n",
       "      <td>0.860000</td>\n",
       "      <td>0.908000</td>\n",
       "      <td>0.90</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.840909</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>0.856061</td>\n",
       "      <td>0.867424</td>\n",
       "      <td>0.85</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.840909</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>0.859848</td>\n",
       "      <td>0.867424</td>\n",
       "      <td>0.85</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.869732</td>\n",
       "      <td>0.892720</td>\n",
       "      <td>0.858238</td>\n",
       "      <td>0.881226</td>\n",
       "      <td>0.85</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.897233</td>\n",
       "      <td>0.889328</td>\n",
       "      <td>0.857708</td>\n",
       "      <td>0.897233</td>\n",
       "      <td>0.85</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.890152</td>\n",
       "      <td>0.852273</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.890152</td>\n",
       "      <td>0.856061</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.80</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.878788</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>0.856061</td>\n",
       "      <td>0.878788</td>\n",
       "      <td>0.80</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.880769</td>\n",
       "      <td>0.888462</td>\n",
       "      <td>0.830769</td>\n",
       "      <td>0.880769</td>\n",
       "      <td>0.80</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.852273</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.859848</td>\n",
       "      <td>0.871212</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.852273</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.837121</td>\n",
       "      <td>0.867424</td>\n",
       "      <td>0.75</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.871212</td>\n",
       "      <td>0.75</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.854406</td>\n",
       "      <td>0.885057</td>\n",
       "      <td>0.854406</td>\n",
       "      <td>0.865900</td>\n",
       "      <td>0.75</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    LogisticRegression       SVC  RandomForestClassifier  VotingClassifier  \\\n",
       "0             0.840909  0.893939                0.856061          0.867424   \n",
       "1             0.844697  0.893939                0.863636          0.871212   \n",
       "2             0.886275  0.929412                0.882353          0.921569   \n",
       "3             0.858300  0.874494                0.829960          0.866397   \n",
       "4             0.840909  0.890152                0.852273          0.867424   \n",
       "5             0.848485  0.893939                0.859848          0.867424   \n",
       "6             0.899225  0.934109                0.864341          0.918605   \n",
       "7             0.892000  0.908000                0.860000          0.908000   \n",
       "8             0.840909  0.886364                0.856061          0.867424   \n",
       "9             0.840909  0.886364                0.859848          0.867424   \n",
       "10            0.869732  0.892720                0.858238          0.881226   \n",
       "11            0.897233  0.889328                0.857708          0.897233   \n",
       "12            0.848485  0.890152                0.852273          0.875000   \n",
       "13            0.848485  0.890152                0.856061          0.875000   \n",
       "14            0.878788  0.886364                0.856061          0.878788   \n",
       "15            0.880769  0.888462                0.830769          0.880769   \n",
       "16            0.852273  0.875000                0.859848          0.871212   \n",
       "17            0.852273  0.875000                0.837121          0.867424   \n",
       "18            0.848485  0.875000                0.848485          0.871212   \n",
       "19            0.854406  0.885057                0.854406          0.865900   \n",
       "\n",
       "    Correlation Threshold  Outlier Threshold  \n",
       "0                    0.00                  0  \n",
       "1                    0.00                 15  \n",
       "2                    0.00                 10  \n",
       "3                    0.00                  8  \n",
       "4                    0.90                  0  \n",
       "5                    0.90                 15  \n",
       "6                    0.90                 10  \n",
       "7                    0.90                  8  \n",
       "8                    0.85                  0  \n",
       "9                    0.85                 15  \n",
       "10                   0.85                 10  \n",
       "11                   0.85                  8  \n",
       "12                   0.80                  0  \n",
       "13                   0.80                 15  \n",
       "14                   0.80                 10  \n",
       "15                   0.80                  8  \n",
       "16                   0.75                  0  \n",
       "17                   0.75                 15  \n",
       "18                   0.75                 10  \n",
       "19                   0.75                  8  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "log_clf = LogisticRegression(solver='newton-cg', C=5.0, max_iter=10000)\n",
    "svm_clf = SVC(gamma='auto',probability=True)\n",
    "rnd_clf = RandomForestClassifier(n_estimators=500, max_leaf_nodes=16, n_jobs=-1)\n",
    "voting_clf = VotingClassifier(\n",
    "    estimators=[('lr', log_clf), ('svc', svm_clf), ('rnd', rnd_clf)],\n",
    "    voting='hard')\n",
    "\n",
    "result = {}\n",
    "\n",
    "for clf in (log_clf, svm_clf, rnd_clf, voting_clf):\n",
    "    result[clf.__class__.__name__] = []\n",
    "result['Correlation Threshold'] = []\n",
    "result['Outlier Threshold'] = []\n",
    "\n",
    "\n",
    "c_list = [0, 0.9, 0.85, 0.8, 0.75]\n",
    "o_list = [0, 15, 10, 8]\n",
    "\n",
    "for c_threshold in c_list:\n",
    "    for o_threshold in o_list:\n",
    "        X_train, X_test, y_train, y_test = resetThresholds(X_orginal,y_original,c_threshold,o_threshold)\n",
    "        result['Correlation Threshold'].append(c_threshold)\n",
    "        result['Outlier Threshold'].append(o_threshold)\n",
    "\n",
    "\n",
    "        for clf in (log_clf, svm_clf, rnd_clf, voting_clf):\n",
    "            clf.fit(X_train, y_train)\n",
    "            y_pred = clf.predict(X_test)\n",
    "            result[clf.__class__.__name__].append(accuracy_score(y_test, y_pred))\n",
    "\n",
    "result_df = pd.DataFrame(result)\n",
    "display(result_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression         0.899225\n",
       "SVC                        0.934109\n",
       "RandomForestClassifier     0.882353\n",
       "VotingClassifier           0.921569\n",
       "Correlation Threshold      0.900000\n",
       "Outlier Threshold         15.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_df.max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The SVM classifier performs the best **0.934109** with a correlation threshold of **0.90** and removing instances with **10** or more outliers. The voting classifier is not an improvement from the SVM.\n",
    "\n",
    "Next the voting classifier is changed to a soft voting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>LogisticRegression</th>\n",
       "      <th>SVC</th>\n",
       "      <th>RandomForestClassifier</th>\n",
       "      <th>VotingClassifier</th>\n",
       "      <th>Correlation Threshold</th>\n",
       "      <th>Outlier Threshold</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.840909</td>\n",
       "      <td>0.893939</td>\n",
       "      <td>0.863636</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.844697</td>\n",
       "      <td>0.893939</td>\n",
       "      <td>0.867424</td>\n",
       "      <td>0.882576</td>\n",
       "      <td>0.00</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.886275</td>\n",
       "      <td>0.929412</td>\n",
       "      <td>0.878431</td>\n",
       "      <td>0.921569</td>\n",
       "      <td>0.00</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.858300</td>\n",
       "      <td>0.874494</td>\n",
       "      <td>0.821862</td>\n",
       "      <td>0.870445</td>\n",
       "      <td>0.00</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.840909</td>\n",
       "      <td>0.890152</td>\n",
       "      <td>0.863636</td>\n",
       "      <td>0.878788</td>\n",
       "      <td>0.90</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.893939</td>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.90</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.899225</td>\n",
       "      <td>0.934109</td>\n",
       "      <td>0.868217</td>\n",
       "      <td>0.914729</td>\n",
       "      <td>0.90</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.892000</td>\n",
       "      <td>0.908000</td>\n",
       "      <td>0.856000</td>\n",
       "      <td>0.900000</td>\n",
       "      <td>0.90</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.840909</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>0.859848</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.85</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.840909</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>0.856061</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.85</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.869732</td>\n",
       "      <td>0.892720</td>\n",
       "      <td>0.869732</td>\n",
       "      <td>0.885057</td>\n",
       "      <td>0.85</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.897233</td>\n",
       "      <td>0.889328</td>\n",
       "      <td>0.857708</td>\n",
       "      <td>0.901186</td>\n",
       "      <td>0.85</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.890152</td>\n",
       "      <td>0.852273</td>\n",
       "      <td>0.878788</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.890152</td>\n",
       "      <td>0.856061</td>\n",
       "      <td>0.878788</td>\n",
       "      <td>0.80</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.878788</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>0.80</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.880769</td>\n",
       "      <td>0.888462</td>\n",
       "      <td>0.842308</td>\n",
       "      <td>0.892308</td>\n",
       "      <td>0.80</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.852273</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.878788</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.852273</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.75</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.848485</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>0.856061</td>\n",
       "      <td>0.882576</td>\n",
       "      <td>0.75</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.854406</td>\n",
       "      <td>0.885057</td>\n",
       "      <td>0.846743</td>\n",
       "      <td>0.888889</td>\n",
       "      <td>0.75</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    LogisticRegression       SVC  RandomForestClassifier  VotingClassifier  \\\n",
       "0             0.840909  0.893939                0.863636          0.875000   \n",
       "1             0.844697  0.893939                0.867424          0.882576   \n",
       "2             0.886275  0.929412                0.878431          0.921569   \n",
       "3             0.858300  0.874494                0.821862          0.870445   \n",
       "4             0.840909  0.890152                0.863636          0.878788   \n",
       "5             0.848485  0.893939                0.848485          0.875000   \n",
       "6             0.899225  0.934109                0.868217          0.914729   \n",
       "7             0.892000  0.908000                0.856000          0.900000   \n",
       "8             0.840909  0.886364                0.859848          0.875000   \n",
       "9             0.840909  0.886364                0.856061          0.875000   \n",
       "10            0.869732  0.892720                0.869732          0.885057   \n",
       "11            0.897233  0.889328                0.857708          0.901186   \n",
       "12            0.848485  0.890152                0.852273          0.878788   \n",
       "13            0.848485  0.890152                0.856061          0.878788   \n",
       "14            0.878788  0.886364                0.848485          0.886364   \n",
       "15            0.880769  0.888462                0.842308          0.892308   \n",
       "16            0.852273  0.875000                0.848485          0.878788   \n",
       "17            0.852273  0.875000                0.848485          0.875000   \n",
       "18            0.848485  0.875000                0.856061          0.882576   \n",
       "19            0.854406  0.885057                0.846743          0.888889   \n",
       "\n",
       "    Correlation Threshold  Outlier Threshold  \n",
       "0                    0.00                  0  \n",
       "1                    0.00                 15  \n",
       "2                    0.00                 10  \n",
       "3                    0.00                  8  \n",
       "4                    0.90                  0  \n",
       "5                    0.90                 15  \n",
       "6                    0.90                 10  \n",
       "7                    0.90                  8  \n",
       "8                    0.85                  0  \n",
       "9                    0.85                 15  \n",
       "10                   0.85                 10  \n",
       "11                   0.85                  8  \n",
       "12                   0.80                  0  \n",
       "13                   0.80                 15  \n",
       "14                   0.80                 10  \n",
       "15                   0.80                  8  \n",
       "16                   0.75                  0  \n",
       "17                   0.75                 15  \n",
       "18                   0.75                 10  \n",
       "19                   0.75                  8  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "log_clf = LogisticRegression(solver='newton-cg', C=5.0, max_iter=10000)\n",
    "svm_clf = SVC(gamma='auto',probability=True)\n",
    "rnd_clf = RandomForestClassifier(n_estimators=500, max_leaf_nodes=16, n_jobs=-1)\n",
    "voting_clf = VotingClassifier(\n",
    "    estimators=[('lr', log_clf), ('svc', svm_clf), ('rnd', rnd_clf)],\n",
    "    voting='soft')\n",
    "\n",
    "result = {}\n",
    "\n",
    "for clf in (log_clf, svm_clf, rnd_clf, voting_clf):\n",
    "    result[clf.__class__.__name__] = []\n",
    "result['Correlation Threshold'] = []\n",
    "result['Outlier Threshold'] = []\n",
    "\n",
    "\n",
    "c_list = [0, 0.9, 0.85, 0.8, 0.75]\n",
    "o_list = [0, 15, 10, 8]\n",
    "\n",
    "for c_threshold in c_list:\n",
    "    for o_threshold in o_list:\n",
    "        X_train, X_test, y_train, y_test = resetThresholds(X_orginal,y_original,c_threshold,o_threshold)\n",
    "        result['Correlation Threshold'].append(c_threshold)\n",
    "        result['Outlier Threshold'].append(o_threshold)\n",
    "\n",
    "\n",
    "        for clf in (log_clf, svm_clf, rnd_clf, voting_clf):\n",
    "            clf.fit(X_train, y_train)\n",
    "            y_pred = clf.predict(X_test)\n",
    "            result[clf.__class__.__name__].append(accuracy_score(y_test, y_pred))\n",
    "\n",
    "result_df = pd.DataFrame(result)\n",
    "display(result_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression         0.899225\n",
       "SVC                        0.934109\n",
       "RandomForestClassifier     0.878431\n",
       "VotingClassifier           0.921569\n",
       "Correlation Threshold      0.900000\n",
       "Outlier Threshold         15.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_df.max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Not much change from soft to hard voting."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### PCA Dimension Reduction\n",
    "\n",
    "Examines algorithms as n_components increase for the various classifiers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>LogisticRegression</th>\n",
       "      <th>SVC</th>\n",
       "      <th>RandomForestClassifier</th>\n",
       "      <th>VotingClassifier</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.472868</td>\n",
       "      <td>0.554264</td>\n",
       "      <td>0.496124</td>\n",
       "      <td>0.503876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.534884</td>\n",
       "      <td>0.515504</td>\n",
       "      <td>0.519380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.496124</td>\n",
       "      <td>0.542636</td>\n",
       "      <td>0.538760</td>\n",
       "      <td>0.523256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.581395</td>\n",
       "      <td>0.674419</td>\n",
       "      <td>0.627907</td>\n",
       "      <td>0.631783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.643411</td>\n",
       "      <td>0.678295</td>\n",
       "      <td>0.639535</td>\n",
       "      <td>0.631783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.651163</td>\n",
       "      <td>0.682171</td>\n",
       "      <td>0.643411</td>\n",
       "      <td>0.631783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.651163</td>\n",
       "      <td>0.697674</td>\n",
       "      <td>0.643411</td>\n",
       "      <td>0.651163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.662791</td>\n",
       "      <td>0.689922</td>\n",
       "      <td>0.647287</td>\n",
       "      <td>0.647287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.693798</td>\n",
       "      <td>0.647287</td>\n",
       "      <td>0.651163</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   LogisticRegression       SVC  RandomForestClassifier  VotingClassifier\n",
       "0            0.472868  0.554264                0.496124          0.503876\n",
       "1            0.500000  0.534884                0.515504          0.519380\n",
       "2            0.496124  0.542636                0.538760          0.523256\n",
       "3            0.581395  0.674419                0.627907          0.631783\n",
       "4            0.643411  0.678295                0.639535          0.631783\n",
       "5            0.651163  0.682171                0.643411          0.631783\n",
       "6            0.651163  0.697674                0.643411          0.651163\n",
       "7            0.662791  0.689922                0.647287          0.647287\n",
       "8            0.666667  0.693798                0.647287          0.651163"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = resetThresholds(X_orginal,y_original,0.9,10)\n",
    "\n",
    "result = {}\n",
    "index = []\n",
    "\n",
    "voting_clf = VotingClassifier(\n",
    "    estimators=[('lr', log_clf), ('svc', svm_clf), ('rnd', rnd_clf)],\n",
    "    voting='hard')\n",
    "\n",
    "for clf in (log_clf, svm_clf, rnd_clf, voting_clf):\n",
    "    result[clf.__class__.__name__] = []\n",
    "    \n",
    "for i in range(2,11):\n",
    "    pca = PCA(n_components=i)\n",
    "    X_train_new = pca.fit_transform(X_train)\n",
    "    X_test_new = pca.fit_transform(X_test)\n",
    "    index.append(i)\n",
    "    for clf in (log_clf, svm_clf, rnd_clf, voting_clf):\n",
    "        clf.fit(X_train_new, y_train)\n",
    "        y_pred = clf.predict(X_test_new)\n",
    "        result[clf.__class__.__name__].append(accuracy_score(y_test, y_pred))\n",
    "\n",
    "result_df = pd.DataFrame(result)\n",
    "display(result_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression        0.666667\n",
       "SVC                       0.697674\n",
       "RandomForestClassifier    0.647287\n",
       "VotingClassifier          0.651163\n",
       "dtype: float64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_df.max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = resetThresholds(X_orginal,y_original,0.9,10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(771, 38)\n",
      "(258, 38)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "print(X_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A good model can be achieved with a SVM model that produces an accurcy score of **0.934109**.\n",
    "\n",
    "The results were obtained after removing correlated feature columns about a **0.9** threshold taking the data from **41** to **38** column features. After these features were removed, instances were examined for being outliers. If one row of data contained more **10** or more occurances of feature data as an outlier, the entire row was considered an outlier and removed from the data set. **26** instances were removed from the dataset leaving **1029** instances to train the model.\n",
    "\n",
    "\n",
    "Applying PCA diminsion reduction techniques did not appear to be useful with this dataset as was shown in the visualizations and models created via the reduced components. To further increase the accuracy of the model, more data can be gathered and features added that may give a better model."
   ]
  },
  {
   "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": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}