1. Supervised Learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "toc": true
   },
   "source": [
    "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Supervised-Learning\" data-toc-modified-id=\"Supervised-Learning-1\">Supervised Learning</a></span><ul class=\"toc-item\"><li><span><a href=\"#Part-1:-Regression-on-California-Test-Scores\" data-toc-modified-id=\"Part-1:-Regression-on-California-Test-Scores-1.1\">Part 1: Regression on California Test Scores</a></span><ul class=\"toc-item\"><li><span><a href=\"#1.1-Visualize-the-univariate-distribution\" data-toc-modified-id=\"1.1-Visualize-the-univariate-distribution-1.1.1\">1.1 Visualize the univariate distribution</a></span></li><li><span><a href=\"#1.2-Visualize-the-dependency-of-the-target-on-each-feature-from-1.1.\" data-toc-modified-id=\"1.2-Visualize-the-dependency-of-the-target-on-each-feature-from-1.1.-1.1.2\">1.2 Visualize the dependency of the target on each feature from 1.1.</a></span></li><li><span><a href=\"#1.3-Modelling-to-evaluate-relationship\" data-toc-modified-id=\"1.3-Modelling-to-evaluate-relationship-1.1.3\">1.3 Modelling to evaluate relationship</a></span><ul class=\"toc-item\"><li><span><a href=\"#1.3.1-Train-test-split\" data-toc-modified-id=\"1.3.1-Train-test-split-1.1.3.1\">1.3.1 Train test split</a></span></li><li><span><a href=\"#1.3.2-Build-Models\" data-toc-modified-id=\"1.3.2-Build-Models-1.1.3.2\">1.3.2 Build Models</a></span></li><li><span><a href=\"#1.3.3-Evaluate-Models-Using-Cross-Validation\" data-toc-modified-id=\"1.3.3-Evaluate-Models-Using-Cross-Validation-1.1.3.3\">1.3.3 Evaluate Models Using Cross-Validation</a></span></li><li><span><a href=\"#1.3.4-Does-scaling-the-data-with-the-StandardScaler-help?\" data-toc-modified-id=\"1.3.4-Does-scaling-the-data-with-the-StandardScaler-help?-1.1.3.4\">1.3.4 Does scaling the data with the StandardScaler help?</a></span></li></ul></li><li><span><a href=\"#1.4-Tune-the-parameters-of-the-models-where-possible-using-GridSearchCV.\" data-toc-modified-id=\"1.4-Tune-the-parameters-of-the-models-where-possible-using-GridSearchCV.-1.1.4\">1.4 Tune the parameters of the models where possible using GridSearchCV.</a></span><ul class=\"toc-item\"><li><span><a href=\"#KNN-for-regression\" data-toc-modified-id=\"KNN-for-regression-1.1.4.1\">KNN for regression</a></span></li><li><span><a href=\"#Ridge-Regression\" data-toc-modified-id=\"Ridge-Regression-1.1.4.2\">Ridge Regression</a></span></li><li><span><a href=\"#Lasso-Regression\" data-toc-modified-id=\"Lasso-Regression-1.1.4.3\">Lasso Regression</a></span></li></ul></li><li><span><a href=\"#1.5-Compare-the-coefficients-of-your-two-best-linear-models-(not-knn)\" data-toc-modified-id=\"1.5-Compare-the-coefficients-of-your-two-best-linear-models-(not-knn)-1.1.5\">1.5 Compare the coefficients of your two best linear models (not knn)</a></span></li><li><span><a href=\"#1.6-Discuss-which-final-model-you-would-choose-to-predict-new-data\" data-toc-modified-id=\"1.6-Discuss-which-final-model-you-would-choose-to-predict-new-data-1.1.6\">1.6 Discuss which final model you would choose to predict new data</a></span></li></ul></li><li><span><a href=\"#Part-2:-Classification-on-red-and-white-wine-characteristics\" data-toc-modified-id=\"Part-2:-Classification-on-red-and-white-wine-characteristics-1.2\">Part 2: Classification on red and white wine characteristics</a></span><ul class=\"toc-item\"><li><span><a href=\"#2.1-Visualize-the-univariate-distribution\" data-toc-modified-id=\"2.1-Visualize-the-univariate-distribution-1.2.1\">2.1 Visualize the univariate distribution</a></span></li><li><span><a href=\"#Visualize-the-dependency-of-the-target-on-each-feature-from-2.1.\" data-toc-modified-id=\"Visualize-the-dependency-of-the-target-on-each-feature-from-2.1.-1.2.2\">Visualize the dependency of the target on each feature from 2.1.</a></span></li><li><span><a href=\"#2.2-Modelling-on-Relationships\" data-toc-modified-id=\"2.2-Modelling-on-Relationships-1.2.3\">2.2 Modelling on Relationships</a></span><ul class=\"toc-item\"><li><span><a href=\"#Split-Training-and-Test\" data-toc-modified-id=\"Split-Training-and-Test-1.2.3.1\">Split Training and Test</a></span></li><li><span><a href=\"#Scaling\" data-toc-modified-id=\"Scaling-1.2.3.2\">Scaling</a></span></li><li><span><a href=\"#Build-Models-and-Evaluate\" data-toc-modified-id=\"Build-Models-and-Evaluate-1.2.3.3\">Build Models and Evaluate</a></span></li><li><span><a href=\"#Logistic-Regression\" data-toc-modified-id=\"Logistic-Regression-1.2.3.4\">Logistic Regression</a></span></li><li><span><a href=\"#Penalized-Logistic-Regression\" data-toc-modified-id=\"Penalized-Logistic-Regression-1.2.3.5\">Penalized Logistic Regression</a></span></li></ul></li><li><span><a href=\"#2.3-Tune-the-parameters-where-possible-using-GridSearchCV.\" data-toc-modified-id=\"2.3-Tune-the-parameters-where-possible-using-GridSearchCV.-1.2.4\">2.3 Tune the parameters where possible using GridSearchCV.</a></span><ul class=\"toc-item\"><li><span><a href=\"#KNN\" data-toc-modified-id=\"KNN-1.2.4.1\">KNN</a></span></li><li><span><a href=\"#Penalized-Logistic-Regression\" data-toc-modified-id=\"Penalized-Logistic-Regression-1.2.4.2\">Penalized Logistic Regression</a></span></li></ul></li><li><span><a href=\"#2.4-Change-the-cross-validation-strategy\" data-toc-modified-id=\"2.4-Change-the-cross-validation-strategy-1.2.5\">2.4 Change the cross-validation strategy</a></span></li><li><span><a href=\"#2.5-Compare-the-coefficients\" data-toc-modified-id=\"2.5-Compare-the-coefficients-1.2.6\">2.5 Compare the coefficients</a></span></li></ul></li></ul></li></ul></div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Supervised Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# import modules\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "# print plots if needed\n",
    "%matplotlib inline\n",
    "\n",
    "# set number of rows pandas display for data frame\n",
    "pd.options.display.max_rows = 10"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 1: Regression on California Test Scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>distcod</th>\n",
       "      <th>county</th>\n",
       "      <th>district</th>\n",
       "      <th>grspan</th>\n",
       "      <th>enrltot</th>\n",
       "      <th>teachers</th>\n",
       "      <th>calwpct</th>\n",
       "      <th>mealpct</th>\n",
       "      <th>computer</th>\n",
       "      <th>testscr</th>\n",
       "      <th>compstu</th>\n",
       "      <th>expnstu</th>\n",
       "      <th>str</th>\n",
       "      <th>avginc</th>\n",
       "      <th>elpct</th>\n",
       "      <th>readscr</th>\n",
       "      <th>mathscr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>75119</td>\n",
       "      <td>Alameda</td>\n",
       "      <td>Sunol Glen Unified</td>\n",
       "      <td>KK-08</td>\n",
       "      <td>195</td>\n",
       "      <td>10.900000</td>\n",
       "      <td>0.510200</td>\n",
       "      <td>2.040800</td>\n",
       "      <td>67</td>\n",
       "      <td>690.799988</td>\n",
       "      <td>0.343590</td>\n",
       "      <td>6384.911133</td>\n",
       "      <td>17.889910</td>\n",
       "      <td>22.690001</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>691.599976</td>\n",
       "      <td>690.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>61499</td>\n",
       "      <td>Butte</td>\n",
       "      <td>Manzanita Elementary</td>\n",
       "      <td>KK-08</td>\n",
       "      <td>240</td>\n",
       "      <td>11.150000</td>\n",
       "      <td>15.416700</td>\n",
       "      <td>47.916698</td>\n",
       "      <td>101</td>\n",
       "      <td>661.200012</td>\n",
       "      <td>0.420833</td>\n",
       "      <td>5099.380859</td>\n",
       "      <td>21.524664</td>\n",
       "      <td>9.824000</td>\n",
       "      <td>4.583333</td>\n",
       "      <td>660.500000</td>\n",
       "      <td>661.900024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>61549</td>\n",
       "      <td>Butte</td>\n",
       "      <td>Thermalito Union Elementary</td>\n",
       "      <td>KK-08</td>\n",
       "      <td>1550</td>\n",
       "      <td>82.900002</td>\n",
       "      <td>55.032299</td>\n",
       "      <td>76.322601</td>\n",
       "      <td>169</td>\n",
       "      <td>643.599976</td>\n",
       "      <td>0.109032</td>\n",
       "      <td>5501.954590</td>\n",
       "      <td>18.697226</td>\n",
       "      <td>8.978000</td>\n",
       "      <td>30.000002</td>\n",
       "      <td>636.299988</td>\n",
       "      <td>650.900024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>61457</td>\n",
       "      <td>Butte</td>\n",
       "      <td>Golden Feather Union Elementary</td>\n",
       "      <td>KK-08</td>\n",
       "      <td>243</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>36.475399</td>\n",
       "      <td>77.049202</td>\n",
       "      <td>85</td>\n",
       "      <td>647.700012</td>\n",
       "      <td>0.349794</td>\n",
       "      <td>7101.831055</td>\n",
       "      <td>17.357143</td>\n",
       "      <td>8.978000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>651.900024</td>\n",
       "      <td>643.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>61523</td>\n",
       "      <td>Butte</td>\n",
       "      <td>Palermo Union Elementary</td>\n",
       "      <td>KK-08</td>\n",
       "      <td>1335</td>\n",
       "      <td>71.500000</td>\n",
       "      <td>33.108601</td>\n",
       "      <td>78.427002</td>\n",
       "      <td>171</td>\n",
       "      <td>640.849976</td>\n",
       "      <td>0.128090</td>\n",
       "      <td>5235.987793</td>\n",
       "      <td>18.671329</td>\n",
       "      <td>9.080333</td>\n",
       "      <td>13.857677</td>\n",
       "      <td>641.799988</td>\n",
       "      <td>639.900024</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   distcod   county                         district grspan  enrltot  \\\n",
       "1    75119  Alameda               Sunol Glen Unified  KK-08      195   \n",
       "2    61499    Butte             Manzanita Elementary  KK-08      240   \n",
       "3    61549    Butte      Thermalito Union Elementary  KK-08     1550   \n",
       "4    61457    Butte  Golden Feather Union Elementary  KK-08      243   \n",
       "5    61523    Butte         Palermo Union Elementary  KK-08     1335   \n",
       "\n",
       "    teachers    calwpct    mealpct  computer     testscr   compstu  \\\n",
       "1  10.900000   0.510200   2.040800        67  690.799988  0.343590   \n",
       "2  11.150000  15.416700  47.916698       101  661.200012  0.420833   \n",
       "3  82.900002  55.032299  76.322601       169  643.599976  0.109032   \n",
       "4  14.000000  36.475399  77.049202        85  647.700012  0.349794   \n",
       "5  71.500000  33.108601  78.427002       171  640.849976  0.128090   \n",
       "\n",
       "       expnstu        str     avginc      elpct     readscr     mathscr  \n",
       "1  6384.911133  17.889910  22.690001   0.000000  691.599976  690.000000  \n",
       "2  5099.380859  21.524664   9.824000   4.583333  660.500000  661.900024  \n",
       "3  5501.954590  18.697226   8.978000  30.000002  636.299988  650.900024  \n",
       "4  7101.831055  17.357143   8.978000   0.000000  651.900024  643.500000  \n",
       "5  5235.987793  18.671329   9.080333  13.857677  641.799988  639.900024  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"https://vincentarelbundock.github.io/Rdatasets/csv/Ecdat/Caschool.csv Links to an external site.\",\n",
    "                  index_col=0)\n",
    "data.head()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y = data['testscr'] \n",
    "\n",
    "X = data.loc[:, data.columns.difference(['testscr', 'readscr', 'mathscr','distcod', 'county', 'district', 'grspan'])]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.1 Visualize the univariate distribution \n",
    "Visualize the univariate distribution of the target feature and each of the three continuous explanatory variables that you think are likely to have a relationship with the target feature.   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x103e37c88>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y.hist(grid = False, bins = 20, color='#208EA5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1168ff828>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X['str'].hist(grid = False, bins = 20, color='#208EA5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x116b96710>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X['avginc'].hist(grid = False, bins = 20, color='#208EA5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x116c5dfd0>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X['expnstu'].hist(grid = False, bins = 20, color='#208EA5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2 Visualize the dependency of the target on each feature from 1.1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.scatter(data['str'], data['testscr'],  marker='o',color='#208EA5')\n",
    "\n",
    "# add a title\n",
    "plt.title(\"Average test score by Student teacher ratio\")\n",
    "\n",
    "# to display the chart\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.scatter(data['avginc'], data['testscr'],  marker='o',color='#208EA5')\n",
    "\n",
    "# add a title\n",
    "plt.title(\"Average test score by district average income\")\n",
    "\n",
    "# to display the chart\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.scatter(data['expnstu'], data['testscr'],  marker='o',color='#208EA5')\n",
    "\n",
    "# add a title\n",
    "plt.title(\"Average test score by expenditure per student\")\n",
    "\n",
    "# to display the chart\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.3 Modelling to evaluate relationship\n",
    "Split data in training and test set. Build models that evaluate the relationship between all available X variables in the California test dataset and the target variable. Evaluate KNN for regression, Linear Regression (OLS), Ridge, and Lasso using cross-validation with the default parameters. Does scaling the data with the StandardScaler help?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 1.3.1 Train test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda/envs/ml/lib/python3.7/site-packages/sklearn/utils/__init__.py:4: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working\n",
      "  from collections import Sequence\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "# shuffling the dataset\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>avginc</th>\n",
       "      <th>calwpct</th>\n",
       "      <th>compstu</th>\n",
       "      <th>computer</th>\n",
       "      <th>elpct</th>\n",
       "      <th>enrltot</th>\n",
       "      <th>expnstu</th>\n",
       "      <th>mealpct</th>\n",
       "      <th>str</th>\n",
       "      <th>teachers</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>316</th>\n",
       "      <td>11.826000</td>\n",
       "      <td>13.375800</td>\n",
       "      <td>0.237500</td>\n",
       "      <td>38</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>160</td>\n",
       "      <td>6870.346191</td>\n",
       "      <td>53.503201</td>\n",
       "      <td>14.705882</td>\n",
       "      <td>10.880000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>8.279000</td>\n",
       "      <td>24.609400</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>20.533880</td>\n",
       "      <td>487</td>\n",
       "      <td>4818.612793</td>\n",
       "      <td>77.148399</td>\n",
       "      <td>20.008217</td>\n",
       "      <td>24.340000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>355</th>\n",
       "      <td>12.640000</td>\n",
       "      <td>2.451000</td>\n",
       "      <td>0.121212</td>\n",
       "      <td>24</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>198</td>\n",
       "      <td>4593.196777</td>\n",
       "      <td>12.745100</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>9.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62</th>\n",
       "      <td>10.905643</td>\n",
       "      <td>58.752201</td>\n",
       "      <td>0.108992</td>\n",
       "      <td>560</td>\n",
       "      <td>32.989491</td>\n",
       "      <td>5138</td>\n",
       "      <td>5606.781738</td>\n",
       "      <td>84.999001</td>\n",
       "      <td>17.670021</td>\n",
       "      <td>290.774994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>172</th>\n",
       "      <td>12.749143</td>\n",
       "      <td>41.773399</td>\n",
       "      <td>0.152223</td>\n",
       "      <td>606</td>\n",
       "      <td>8.842000</td>\n",
       "      <td>3981</td>\n",
       "      <td>5358.517090</td>\n",
       "      <td>51.770901</td>\n",
       "      <td>18.957144</td>\n",
       "      <td>210.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        avginc    calwpct   compstu  computer      elpct  enrltot  \\\n",
       "316  11.826000  13.375800  0.237500        38   2.500000      160   \n",
       "14    8.279000  24.609400  0.000000         0  20.533880      487   \n",
       "355  12.640000   2.451000  0.121212        24   0.000000      198   \n",
       "62   10.905643  58.752201  0.108992       560  32.989491     5138   \n",
       "172  12.749143  41.773399  0.152223       606   8.842000     3981   \n",
       "\n",
       "         expnstu    mealpct        str    teachers  \n",
       "316  6870.346191  53.503201  14.705882   10.880000  \n",
       "14   4818.612793  77.148399  20.008217   24.340000  \n",
       "355  4593.196777  12.745100  22.000000    9.000000  \n",
       "62   5606.781738  84.999001  17.670021  290.774994  \n",
       "172  5358.517090  51.770901  18.957144  210.000000  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 1.3.2 Build Models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### KNN for regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: 0.424\n",
      "Test set score: 0.029\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsRegressor \n",
    "\n",
    "knn = KNeighborsRegressor(n_neighbors=3)\n",
    "\n",
    "knn.fit(X_train, y_train)\n",
    "\n",
    "print(\"Training set score: {:.3f}\".format(knn.score(X_train, y_train)))\n",
    "print(\"Test set score: {:.3f}\".format(knn.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Linear Regression (OLS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lr.coef_: [ 5.44223554e-01 -8.83598155e-02  5.80342195e+00  3.30043134e-03\n",
      " -2.14288402e-01  1.23265852e-03  2.00663899e-03 -3.81931221e-01\n",
      " -2.59144037e-01 -3.24527985e-02]\n",
      "lr.intercept_: 661.1394147432457\n",
      "Training set score: 0.832\n",
      "Test set score: 0.746\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "lr = LinearRegression()\n",
    "lr.fit(X_train, y_train)\n",
    "print(\"lr.coef_: {}\".format(lr.coef_)) \n",
    "print(\"lr.intercept_: {}\".format(lr.intercept_))\n",
    "\n",
    "print(\"Training set score: {:.3f}\".format(lr.score(X_train, y_train))) \n",
    "print(\"Test set score: {:.3f}\".format(lr.score(X_test, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Ridge Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: 0.832\n",
      "Test set score: 0.745\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import Ridge\n",
    "ridge = Ridge()\n",
    "ridge.fit(X_train, y_train)\n",
    "print(\"Training set score: {:.3f}\".format(ridge.score(X_train, y_train))) \n",
    "print(\"Test set score: {:.3f}\".format(ridge.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Increasing alpha forces coefficients to move more toward zero"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: 0.832\n",
      "Test set score: 0.745\n"
     ]
    }
   ],
   "source": [
    "ridge10 = Ridge(alpha=10).fit(X_train, y_train)\n",
    "print(\"Training set score: {:.3f}\".format(ridge10.score(X_train, y_train))) \n",
    "print(\"Test set score: {:.3f}\".format(ridge10.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Decreasing alpha means approach to OLS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: 0.832\n",
      "Test set score: 0.746\n"
     ]
    }
   ],
   "source": [
    "ridge01 = Ridge(alpha=0.1).fit(X_train, y_train)\n",
    "print(\"Training set score: {:.3f}\".format(ridge01.score(X_train, y_train))) \n",
    "print(\"Test set score: {:.3f}\".format(ridge01.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Lasso Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: 0.831\n",
      "Test set score: 0.744\n",
      "Number of features used: 8\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import Lasso\n",
    "lasso = Lasso(max_iter=10000)\n",
    "lasso.fit(X_train, y_train)\n",
    "print(\"Training set score: {:.3f}\".format(lasso.score(X_train, y_train))) \n",
    "print(\"Test set score: {:.3f}\".format(lasso.score(X_test, y_test))) \n",
    "print(\"Number of features used: {}\".format(np.sum(lasso.coef_ != 0)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Increasing alpha forces more coefficients to zero"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: 0.828\n",
      "Test set score: 0.737\n",
      "Number of features used: 7\n"
     ]
    }
   ],
   "source": [
    "lasso5 = Lasso(alpha=5, max_iter=100000).fit(X_train, y_train) \n",
    "print(\"Training set score: {:.3f}\".format(lasso5.score(X_train, y_train))) \n",
    "print(\"Test set score: {:.3f}\".format(lasso5.score(X_test, y_test))) \n",
    "print(\"Number of features used: {}\".format(np.sum(lasso5.coef_ != 0)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Decreasing alpha means approach to OLS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: 0.832\n",
      "Test set score: 0.745\n",
      "Number of features used: 10\n"
     ]
    }
   ],
   "source": [
    "lasso001 = Lasso(alpha=0.01, max_iter=100000).fit(X_train, y_train) \n",
    "print(\"Training set score: {:.3f}\".format(lasso001.score(X_train, y_train))) \n",
    "print(\"Test set score: {:.3f}\".format(lasso001.score(X_test, y_test))) \n",
    "print(\"Number of features used: {}\".format(np.sum(lasso001.coef_ != 0)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 1.3.3 Evaluate Models Using Cross-Validation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### KNN for Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score: -0.088\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "print(\"Average cross-validation score: {:.3f}\".format(np.mean(cross_val_score(KNeighborsRegressor(), \n",
    "                                                                              X_train, y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Linear Regression (OLS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score: 0.807\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score: {:.3f}\".format(np.mean(cross_val_score(LinearRegression(), \n",
    "                                                                              X_train, y_train, cv=10))))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Ridge Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score: 0.808\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score: {:.3f}\".format(np.mean(cross_val_score(Ridge(), \n",
    "                                                                              X_train, y_train, cv=10))))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Lasso Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score: 0.810\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score: {:.3f}\".format(np.mean(cross_val_score(Lasso(max_iter=100000), \n",
    "                                                                              X_train, y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 1.3.4 Does scaling the data with the StandardScaler help?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn import preprocessing\n",
    "# Use transformed data with a model:\n",
    "scaler = preprocessing.StandardScaler()\n",
    "scaler.fit(X_train)\n",
    "X_train_scaled = scaler.transform(X_train)\n",
    "X_test_scaled = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### KNN for Regression\n",
    "\n",
    "Scaling help improve KNN for regression a lot, since KNN model is larged determined by the scale of variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training without scale: 0.424\n",
      "Test without scale: 0.029\n",
      "Training with scale: 0.878\n",
      "Test with scale: 0.692\n"
     ]
    }
   ],
   "source": [
    "knn.fit(X_train, y_train)\n",
    "print(\"Training without scale: {:.3f}\".format(knn.score(X_train, y_train)))\n",
    "print(\"Test without scale: {:.3f}\".format(knn.score(X_test, y_test)))\n",
    "\n",
    "knn.fit(X_train_scaled, y_train)\n",
    "print(\"Training with scale: {:.3f}\".format(knn.score(X_train_scaled, y_train)))\n",
    "print(\"Test with scale: {:.3f}\".format(knn.score(X_test_scaled, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score without scale: -0.088\n",
      "Average cross-validation score with scale: 0.731\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score without scale: {:.3f}\".format(np.mean(cross_val_score(KNeighborsRegressor(), \n",
    "                                                                              X_train, \n",
    "                                                                              y_train, cv=10))))\n",
    "\n",
    "print(\"Average cross-validation score with scale: {:.3f}\".format(np.mean(cross_val_score(KNeighborsRegressor(), \n",
    "                                                                              X_train_scaled, \n",
    "                                                                              y_train, cv=10))))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Linear Regression (OLS)\n",
    "\n",
    "Scaling does not affect OLS model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training without scale: 0.832\n",
      "Test without scale: 0.746\n",
      "Training with scale: 0.832\n",
      "Test with scale: 0.746\n"
     ]
    }
   ],
   "source": [
    "lr.fit(X_train, y_train)\n",
    "print(\"Training without scale: {:.3f}\".format(lr.score(X_train, y_train)))\n",
    "print(\"Test without scale: {:.3f}\".format(lr.score(X_test, y_test)))\n",
    "\n",
    "lr.fit(X_train_scaled, y_train)\n",
    "print(\"Training with scale: {:.3f}\".format(lr.score(X_train_scaled, y_train)))\n",
    "print(\"Test with scale: {:.3f}\".format(lr.score(X_test_scaled, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score without scale: 0.807\n",
      "Average cross-validation score with scale: 0.807\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score without scale: {:.3f}\".format(np.mean(cross_val_score(LinearRegression(), \n",
    "                                                                              X_train, \n",
    "                                                                              y_train, cv=10))))\n",
    "\n",
    "print(\"Average cross-validation score with scale: {:.3f}\".format(np.mean(cross_val_score(LinearRegression(), \n",
    "                                                                              X_train_scaled, \n",
    "                                                                              y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Ridge Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training without scale: 0.832\n",
      "Test without scale: 0.745\n",
      "Training with scale: 0.832\n",
      "Test with scale: 0.748\n"
     ]
    }
   ],
   "source": [
    "ridge.fit(X_train, y_train)\n",
    "print(\"Training without scale: {:.3f}\".format(ridge.score(X_train, y_train)))\n",
    "print(\"Test without scale: {:.3f}\".format(ridge.score(X_test, y_test)))\n",
    "\n",
    "ridge.fit(X_train_scaled, y_train)\n",
    "print(\"Training with scale: {:.3f}\".format(ridge.score(X_train_scaled, y_train)))\n",
    "print(\"Test with scale: {:.3f}\".format(ridge.score(X_test_scaled, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score without scale: 0.808\n",
      "Average cross-validation score with scale: 0.807\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score without scale: {:.3f}\".format(np.mean(cross_val_score(Ridge(), \n",
    "                                                                              X_train, \n",
    "                                                                              y_train, cv=10))))\n",
    "\n",
    "print(\"Average cross-validation score with scale: {:.3f}\".format(np.mean(cross_val_score(Ridge(), \n",
    "                                                                              X_train_scaled, \n",
    "                                                                              y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Lasso Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training without scale: 0.831\n",
      "Test without scale: 0.744\n",
      "Training with scale: 0.823\n",
      "Test with scale: 0.742\n"
     ]
    }
   ],
   "source": [
    "lasso.fit(X_train, y_train)\n",
    "print(\"Training without scale: {:.3f}\".format(lasso.score(X_train, y_train)))\n",
    "print(\"Test without scale: {:.3f}\".format(lasso.score(X_test, y_test)))\n",
    "\n",
    "lasso.fit(X_train_scaled, y_train)\n",
    "print(\"Training with scale: {:.3f}\".format(lasso.score(X_train_scaled, y_train)))\n",
    "print(\"Test with scale: {:.3f}\".format(lasso.score(X_test_scaled, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score without scale: 0.810\n",
      "Average cross-validation score with scale: 0.805\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score without scale: {:.3f}\".format(np.mean(cross_val_score(Lasso(max_iter=100000), \n",
    "                                                                              X_train, \n",
    "                                                                              y_train, cv=10))))\n",
    "\n",
    "print(\"Average cross-validation score with scale: {:.3f}\".format(np.mean(cross_val_score(Lasso(max_iter=100000), \n",
    "                                                                              X_train_scaled, \n",
    "                                                                              y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.4 Tune the parameters of the models where possible using GridSearchCV."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using GridsearchCV to tune the parameter improves our models' perfermance on predicting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### KNN for regression "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best mean cross-validation score: 0.765\n",
      "best parameters: {'n_neighbors': 13}\n",
      "test-set score: 0.706\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "param_grid = {'n_neighbors': np.arange(1, 15, 1)} \n",
    "\n",
    "grid = GridSearchCV(KNeighborsRegressor(), param_grid=param_grid, cv=10)\n",
    "\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"best mean cross-validation score: {:.3f}\".format(grid.best_score_))\n",
    "print(\"best parameters: {}\".format(grid.best_params_))\n",
    "print(\"test-set score: {:.3f}\".format(grid.score(X_test_scaled, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Ridge Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best mean cross-validation score: 0.808\n",
      "best parameters: {'alpha': 10}\n",
      "test-set score: 0.749\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "param_grid = {'alpha': np.arange(0, 30, 1)} #np.arange creates sequence of numbers for each k value\n",
    "\n",
    "grid = GridSearchCV(Ridge(), param_grid=param_grid, cv=10)\n",
    "\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "\n",
    "#extract best score and parameter by calling objects \"best_score_\" and \"best_params_\"\n",
    "print(\"best mean cross-validation score: {:.3f}\".format(grid.best_score_))\n",
    "print(\"best parameters: {}\".format(grid.best_params_))\n",
    "print(\"test-set score: {:.3f}\".format(grid.score(X_test_scaled, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Lasso Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best mean cross-validation score: 0.809\n",
      "best parameters: {'alpha': 0.21000000000000002}\n",
      "test-set score: 0.749\n"
     ]
    }
   ],
   "source": [
    "param_grid = {'alpha': np.arange(0.01, 20, 0.1)} #np.arange creates sequence of numbers for each k value\n",
    "\n",
    "grid = GridSearchCV(Lasso(max_iter=100000), param_grid=param_grid, cv=10)\n",
    "\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "\n",
    "#extract best score and parameter by calling objects \"best_score_\" and \"best_params_\"\n",
    "print(\"best mean cross-validation score: {:.3f}\".format(grid.best_score_))\n",
    "print(\"best parameters: {}\".format(grid.best_params_))\n",
    "print(\"test-set score: {:.3f}\".format(grid.score(X_test_scaled, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.5 Compare the coefficients of your two best linear models (not knn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I choose `Ridge(alpha=10)` and Lasso with `Lasso(alpha=0.21000000000000002)` to compare based on the `GridSearchCV` results.**\n",
    "**Two models agreed on the most important features is  `mealpct` (percent qualifying for reduced-price lunch), `elpct` (percent of English learners)  and `avginc` (district average income)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ridge10.coef_: [ 3.93263723 -1.47292229  0.47246612  0.9048331  -4.29797339 -0.14706884\n",
      "  1.18568776 -9.27650008 -0.33469443 -0.65639799]\n",
      "ridge10.intercept_: 653.5317460317458\n"
     ]
    }
   ],
   "source": [
    "ridge10 = Ridge(alpha=10).fit(X_train_scaled, y_train)\n",
    "print(\"ridge10.coef_: {}\".format(ridge10.coef_)) \n",
    "print(\"ridge10.intercept_: {}\".format(ridge10.intercept_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lasso02.coef_: [  3.70158456  -0.7126675    0.52301986   0.          -3.78905691\n",
      "  -0.           1.06009605 -10.46987237  -0.20298529  -0.        ]\n",
      "lasso02.intercept_: 653.5317460317458\n"
     ]
    }
   ],
   "source": [
    "lasso02 = Lasso(alpha=0.21000000000000002).fit(X_train_scaled, y_train)\n",
    "print(\"lasso02.coef_: {}\".format(lasso02.coef_)) \n",
    "print(\"lasso02.intercept_: {}\".format(lasso02.intercept_))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.6 Discuss which final model you would choose to predict new data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I would choose `Ridge(alpha=10)` as my final model. Firstly, it has the best test-set score; Secondly, though `Lasso(alpha=0.21000000000000002)` has the same test-set score, using L2 regularization would not exclude some features in the new data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 2: Classification on red and white wine characteristics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "white = pd.read_csv(\"winequality-white.csv\", sep = ';')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "red = pd.read_csv(\"winequality-red.csv\", sep = ';')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Add a new column to each data frame called \"winetype\".  For the white wine dataset label the values in this column with a 0, indicating white wine.  For the red wine dataset, label values with a 1, indicating red wine.  Combine both datasets into a single dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "white['winetype'] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "red['winetype'] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "wine = pd.concat([white,red], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "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>fixed acidity</th>\n",
       "      <th>volatile acidity</th>\n",
       "      <th>citric acid</th>\n",
       "      <th>residual sugar</th>\n",
       "      <th>chlorides</th>\n",
       "      <th>free sulfur dioxide</th>\n",
       "      <th>total sulfur dioxide</th>\n",
       "      <th>density</th>\n",
       "      <th>pH</th>\n",
       "      <th>sulphates</th>\n",
       "      <th>alcohol</th>\n",
       "      <th>quality</th>\n",
       "      <th>winetype</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7.0</td>\n",
       "      <td>0.27</td>\n",
       "      <td>0.36</td>\n",
       "      <td>20.7</td>\n",
       "      <td>0.045</td>\n",
       "      <td>45.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>1.0010</td>\n",
       "      <td>3.00</td>\n",
       "      <td>0.45</td>\n",
       "      <td>8.8</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6.3</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.34</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.049</td>\n",
       "      <td>14.0</td>\n",
       "      <td>132.0</td>\n",
       "      <td>0.9940</td>\n",
       "      <td>3.30</td>\n",
       "      <td>0.49</td>\n",
       "      <td>9.5</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8.1</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.40</td>\n",
       "      <td>6.9</td>\n",
       "      <td>0.050</td>\n",
       "      <td>30.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>0.9951</td>\n",
       "      <td>3.26</td>\n",
       "      <td>0.44</td>\n",
       "      <td>10.1</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7.2</td>\n",
       "      <td>0.23</td>\n",
       "      <td>0.32</td>\n",
       "      <td>8.5</td>\n",
       "      <td>0.058</td>\n",
       "      <td>47.0</td>\n",
       "      <td>186.0</td>\n",
       "      <td>0.9956</td>\n",
       "      <td>3.19</td>\n",
       "      <td>0.40</td>\n",
       "      <td>9.9</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7.2</td>\n",
       "      <td>0.23</td>\n",
       "      <td>0.32</td>\n",
       "      <td>8.5</td>\n",
       "      <td>0.058</td>\n",
       "      <td>47.0</td>\n",
       "      <td>186.0</td>\n",
       "      <td>0.9956</td>\n",
       "      <td>3.19</td>\n",
       "      <td>0.40</td>\n",
       "      <td>9.9</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   fixed acidity  volatile acidity  citric acid  residual sugar  chlorides  \\\n",
       "0            7.0              0.27         0.36            20.7      0.045   \n",
       "1            6.3              0.30         0.34             1.6      0.049   \n",
       "2            8.1              0.28         0.40             6.9      0.050   \n",
       "3            7.2              0.23         0.32             8.5      0.058   \n",
       "4            7.2              0.23         0.32             8.5      0.058   \n",
       "\n",
       "   free sulfur dioxide  total sulfur dioxide  density    pH  sulphates  \\\n",
       "0                 45.0                 170.0   1.0010  3.00       0.45   \n",
       "1                 14.0                 132.0   0.9940  3.30       0.49   \n",
       "2                 30.0                  97.0   0.9951  3.26       0.44   \n",
       "3                 47.0                 186.0   0.9956  3.19       0.40   \n",
       "4                 47.0                 186.0   0.9956  3.19       0.40   \n",
       "\n",
       "   alcohol  quality  winetype  \n",
       "0      8.8        6         0  \n",
       "1      9.5        6         0  \n",
       "2     10.1        6         0  \n",
       "3      9.9        6         0  \n",
       "4      9.9        6         0  "
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wine.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>fixed acidity</th>\n",
       "      <th>volatile acidity</th>\n",
       "      <th>citric acid</th>\n",
       "      <th>residual sugar</th>\n",
       "      <th>chlorides</th>\n",
       "      <th>free sulfur dioxide</th>\n",
       "      <th>total sulfur dioxide</th>\n",
       "      <th>density</th>\n",
       "      <th>pH</th>\n",
       "      <th>sulphates</th>\n",
       "      <th>alcohol</th>\n",
       "      <th>quality</th>\n",
       "      <th>winetype</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1594</th>\n",
       "      <td>6.2</td>\n",
       "      <td>0.600</td>\n",
       "      <td>0.08</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.090</td>\n",
       "      <td>32.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>0.99490</td>\n",
       "      <td>3.45</td>\n",
       "      <td>0.58</td>\n",
       "      <td>10.5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1595</th>\n",
       "      <td>5.9</td>\n",
       "      <td>0.550</td>\n",
       "      <td>0.10</td>\n",
       "      <td>2.2</td>\n",
       "      <td>0.062</td>\n",
       "      <td>39.0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>0.99512</td>\n",
       "      <td>3.52</td>\n",
       "      <td>0.76</td>\n",
       "      <td>11.2</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1596</th>\n",
       "      <td>6.3</td>\n",
       "      <td>0.510</td>\n",
       "      <td>0.13</td>\n",
       "      <td>2.3</td>\n",
       "      <td>0.076</td>\n",
       "      <td>29.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>0.99574</td>\n",
       "      <td>3.42</td>\n",
       "      <td>0.75</td>\n",
       "      <td>11.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1597</th>\n",
       "      <td>5.9</td>\n",
       "      <td>0.645</td>\n",
       "      <td>0.12</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.075</td>\n",
       "      <td>32.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>0.99547</td>\n",
       "      <td>3.57</td>\n",
       "      <td>0.71</td>\n",
       "      <td>10.2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1598</th>\n",
       "      <td>6.0</td>\n",
       "      <td>0.310</td>\n",
       "      <td>0.47</td>\n",
       "      <td>3.6</td>\n",
       "      <td>0.067</td>\n",
       "      <td>18.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>0.99549</td>\n",
       "      <td>3.39</td>\n",
       "      <td>0.66</td>\n",
       "      <td>11.0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      fixed acidity  volatile acidity  citric acid  residual sugar  chlorides  \\\n",
       "1594            6.2             0.600         0.08             2.0      0.090   \n",
       "1595            5.9             0.550         0.10             2.2      0.062   \n",
       "1596            6.3             0.510         0.13             2.3      0.076   \n",
       "1597            5.9             0.645         0.12             2.0      0.075   \n",
       "1598            6.0             0.310         0.47             3.6      0.067   \n",
       "\n",
       "      free sulfur dioxide  total sulfur dioxide  density    pH  sulphates  \\\n",
       "1594                 32.0                  44.0  0.99490  3.45       0.58   \n",
       "1595                 39.0                  51.0  0.99512  3.52       0.76   \n",
       "1596                 29.0                  40.0  0.99574  3.42       0.75   \n",
       "1597                 32.0                  44.0  0.99547  3.57       0.71   \n",
       "1598                 18.0                  42.0  0.99549  3.39       0.66   \n",
       "\n",
       "      alcohol  quality  winetype  \n",
       "1594     10.5        5         1  \n",
       "1595     11.2        6         1  \n",
       "1596     11.0        6         1  \n",
       "1597     10.2        5         1  \n",
       "1598     11.0        6         1  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wine.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.1 Visualize the univariate distribution\n",
    "\n",
    "Visualize the univariate distribution of the target feature and each of the three explanatory variables that you think are likely to have a relationship with the target feature.   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1abc6dd8>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.countplot(x=\"winetype\", data=wine, palette=\"Set2\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda/envs/ml/lib/python3.7/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1ad00b00>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEKCAYAAADpfBXhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzt3Xl8VfWd//HXJ/u+JwSyEmQRVFAji7QWl3ZordDpTytoVaz+mM7U6Txmpu200/6c1s5M27HLdB7aaa27U3VcW7Sote4KAhFZZAlrICFAAllYkpDt+/sjASMGcknuzbn35P18PHjknnNPzv3kAu/7zfec7/drzjlERMRforwuQEREgk/hLiLiQwp3EREfUriLiPiQwl1ExIcU7iIiPqRwFxHxIYW7iIgPKdxFRHwoxqsXzsnJcaWlpV69vIhIRHrvvfcOOOdyBzrOs3AvLS2loqLCq5cXEYlIZrYrkOPULSMi4kMKdxERH1K4i4j4kMJdRMSHFO4iIj6kcBcR8SGFu4iIDyncRUR8SOEuIuJDno1QldB7dMXuUz533YziYaxERIabWu4iIj6kcBcR8SGFu4iIDyncRUR8SOEuIuJDCncRER9SuIuI+JDCXUTEhxTuIiI+pHAXEfEhhbuIiA8p3EVEfEjhLiLiQwp3EREfUriLiPiQwl1ExIcU7iIiPqRwFxHxoYDC3czmmlmlmW0zs2/38/wiM6s3szW9f24NfqkiIhKoAddQNbNo4G7g00ANsMrMljjnNp506P86524LQY0iInKGAmm5Twe2Oed2OOfagceB+aEtS0REhiKQcC8Aqvts1/TuO9n/MbN1ZvaUmRUFpToRERmUQMLd+tnnTtp+Dih1zp0H/Bl4qN8TmS02swozq6ivrz+zSkVEJGCBhHsN0LclXgjU9j3AOXfQOXesd/O3wIX9ncg5d49zrtw5V56bmzuYekVEJACBhPsqYLyZjTWzOGABsKTvAWY2us/mPGBT8EoUEZEzNeDdMs65TjO7DXgJiAbud85tMLM7gArn3BLg62Y2D+gEGoBFIaxZREQGMGC4AzjnlgJLT9p3e5/H3wG+E9zSRERksDRCVUTEhxTuIiI+pHAXEfEhhbuIiA8p3EVEfEjhLiLiQwp3EREfUriLiPiQwl1ExIcU7iIiPqRwFxHxIYW7iIgPKdxFRHxI4S4i4kMKdxERH1K4i4j4kMJdRMSHFO4iIj6kcBcR8SGFu4iIDyncRUR8SOEuIuJDCncRER9SuIuI+JDCXUTEhxTuIiI+pHAXEfEhhbuIiA8FFO5mNtfMKs1sm5l9+zTHXW1mzszKg1eiiIicqQHD3cyigbuBzwKTgYVmNrmf41KBrwMrgl2kiIicmUBa7tOBbc65Hc65duBxYH4/x/0Q+A+gLYj1iYjIIAQS7gVAdZ/tmt59J5jZ+UCRc+75INYmIiKDFEi4Wz/73IknzaKAXwD/OOCJzBabWYWZVdTX1wdepYiInJFAwr0GKOqzXQjU9tlOBc4BXjezKmAmsKS/i6rOuXucc+XOufLc3NzBVy0iIqcVSLivAsab2VgziwMWAEuOP+mca3bO5TjnSp1zpcC7wDznXEVIKhYRkQENGO7OuU7gNuAlYBPwhHNug5ndYWbzQl2gDN2mvYdYvuMgh1o7vC5FRIZJTCAHOeeWAktP2nf7KY6dM/SyJBicc7yyuY5XN9cB8PzaWsblprBwerHHlYlIqGmEqk91dzueWb2HVzfXcWFJJn93+Xgum5THjgNH+MPaPV6XJyIhFlDLXSLPc+tqeW93I5dOzOWKs0dhZoxKS8DM+POm/fxhzR7mTysY+EQiEpHUcveh7m7H3a9tIy81nst7g/24ORNzKclK4nvPfkBNY4uHVYpIKCncfeilDfvYsv8Il07MI8o+Okwhyoxryovoco47X6r0qEIRCTV1y0S4R1fs/si2c467XttGTkoc5xam9/s9WclxXD+jmPvfqeKbfzGRwsyk4ShVRIaRWu4+s3nfYfY2tzFnwsdb7X3dPHssBtz/dtWw1SYiw0fh7jMrdzaQnhjL1KKM0x43JiORq6aO4fFVu2lu0f3vIn6jcPeR1vYuttUd4dyCdKKjTt1qP+7/frKMlvYu/mfFrmGoTkSGk8LdRzbvO0SXc5xT0H9f+8kmj0njk+NzeHBZFR1d3SGuTkSGk8LdR9bvaSY9MZaizMSAv2fRxaXUHz7GK5v2h7AyERluCnefaOvoYmvdEc4Zk/aR+9oH8qkJuYxOT+DRldUDHywiEUPh7hOb9h6iqzvwLpnjYqKj+FJ5EW9trae6QYOaRPxC4e4TH+xpJi0hhqKsM79n/UsXFWHA/65S613ELxTuPtDR1c3WuiNMKUg/7b3tp1KQkciciXk8UVGtC6siPqFw94Gqg0fp7HZMHJU66HMsnF5M3eFjJ6YHFpHIpnD3gW11R4g2ozQ7edDnuHRiLqPS4nls5e6BDxaRsKdw94HtdUcozk4iLmbwf50x0VFcW17EG1vqNVukiA8o3CPckWOd1Da3cVZeypDP9aWLetZBf0IXVkUinsI9wu2oPwLAWblDD/fCzCQ+NSGXJypq6NSFVZGIpnCPcNvqjpAQG0XBGYxKPZ2F04vZd6iN1yvrg3I+EfGGwj2COefYVneEspyUQd0C2Z/LJuWRlxrPI+9qMjGRSKZwj2BVB1toau0ISn/7cbHRUdwws4Q3ttSzed+hoJ1XRIaXwj2CLdt+AAhOf3tfN8wqITE2mnve3BHU84rI8FG4R7CKqkZS4mPITokL6nkzkuK49qIilqypZW9za1DPLSLDQ+EewVbubKAkO+mMZoEM1C2fGIsD7n97Z9DPLSKhp3CPULVNrexpah3SqNTTKcpK4spzR/PYymoaj7aH5DVEJHRivC5ABqdiVyNAyMId4GuXnsXz62r55Stb+f68KSf2P7ri1FMUXDejOGT1iEjg1HKPUKt2NpAcF01+ekLIXmNifioLphfzP+/uYnvvYCkRiQwBhbuZzTWzSjPbZmbf7uf5r5rZejNbY2Zvm9nk4Jcqfa2qauCCksyAFsIeir+/YgIJsdH8+x83hfR1RCS4BuyWMbNo4G7g00ANsMrMljjnNvY57FHn3K97j58H/ByYG4J6R6STu0Fa27uo3HeYwiCNSj2d3NR4vnbpWfzkxc28uaWeSybkhvw1RWToAmm5Twe2Oed2OOfagceB+X0PcM71He2SDLjglSgn291wFAeUhLC/va+bZ5dSmp3Et59eR3Nrx7C8pogMTSAXVAuAvtME1gAzTj7IzL4G/AMQB1wWlOqkX1UHW4gyKMo88yX1AnXybwufPWc0v3lzO1++dwULLioK+u2XukgrElyBtNz7+1/8sZa5c+5u59w44J+A7/V7IrPFZlZhZhX19ZqYarB2HTxKQUbikOZvP1NFWUlccfYo1u9pZk1107C9rogMTiDpUAMU9dkuBGpPc/zjwBf6e8I5d49zrtw5V56bq77bwejqduxpaqV4EAthD9UlE3IpzU7m92v2aEEPkTAXSLivAsab2VgziwMWAEv6HmBm4/tsXglsDV6J0te+5jY6uhxFHoR7lBkLpxeREh/DQ8t30aDBTSJha8A+d+dcp5ndBrwERAP3O+c2mNkdQIVzbglwm5ldAXQAjcBNoSx6JNvd22L2ouUOkJoQy00Xl/KbN3bw4LKdLL5kHCnx3o2FU1+9SP8C+l/pnFsKLD1p3+19Hv9dkOuSU6huaCE1IYb0xFjPashLTeCGmSXc/85OfvvWDm6ZPZY0D+sRkY/TCNUIs7uhheKs0EwWdiZKc5JZNLuU5tYO7nlrh7poRMKM5paJIEeOddJwtJ3ppVlDPtfpujMCVZaTwi2zx/Lgsir++43tXDdd3SAi4UIt9whS09DT3+7FxdRTKcpK4q8+VUZibDT3vb2DB97ZiXMawybiNYV7BNnd0DN4qSAj9NMOnIm81AT+Zs44Jo5K5QfPbWThb9/VRGMiHlO4R5DdjS2MTh/ewUuBSoiN5vqZJfzoi+eysfYQn/3Pt/j+kg3sOnjU69JERiT1uUeIbueoaWzlguIMr0s5pZ774Iu54uxR/MeLm/ndil08tLyKT03I5fJJecyZmBdWXUoifqZwjxB1h47R3tkd0vlkgiU3NZ47r5nKN/9iIg8v38WStbW8XrkB2MCY9AQuLM3iwuIMykuzmJSfSkx0+P0mIhLpFO4R4vhw/0hq+ealJfCNv5jIP35mAjsOHOWtLfWs2tXIqp0NPLe2ZwaL1IQY5k7JJzUhlrLcZKI8vsVTxC8U7hGiurGVhNgospPjvC7ljJkZ43JTGJebwqLZYwHY09RKRVUDb245wAsf7OPIsU5yUuK5ZHwO04oziIlSa15kKPQ/KELUNLZQlOn94KVgKchIZP60An72palUfO8Kri0vIjbaeOb9PfzsT1tYubOBrm7dUikyWAr3CNDe2c3+Q20UDMPKS15IiI1malEGt116FosuLiUtIYbfr9nDL/68hdW7Guns6va6RJGIo26ZCLC3uZVuF9rFOcKBmTFhVCrj81Ko3H+YP2/cz1Ora1hd3chff2ocV00dQ0JstNdlikQEhXsEqGlsBYj4lnugUx6YGZPy05g4KpUNtYd4b1cj33xqHf++dBNXX1jI/GkFTBmT5psuKpFQULhHgOrGFtITY0lLGFkzL5oZ5xSk829/eQ7LdxzkkeW7eOCdKn771k7G5iRz1XmjMTNGpSV4XapI2FG4R4CaxlYKI7zVPhRmxsXjcrh4XA6NR9t5ccM+nl9Xy12vbaPbwej0BGaVZTO1KINY3TMvAijcw17j0fagzQTpB5nJcSycXszC6cXUHW7jh89tZFVVI8+8v4cXN+zjskl5zBibTXSUumxkZFO4h7m1NT2LUY/klvvp+upnjcthZlk2Ow8e5fXKep5ft5eVOxv4wrSCYaxQJPwo3MPc2upmjPCbCTKcmBllOSmMzU5m097D/HF9Lfe+vYPslDgWX1KmC68yIincw9zamiZyU+OJ1y2AAzIzJo9Joyw3mWdW1/CjFzazencjv1xw/sduodTaq+J3uvoUxpxzrK1u8v397cGWEBvNwunFfO/Ks/nTxv0semAlR451el2WyLBSuIexmsZWDh5tpzBLXTJnysy49ZNl/Oe101hV1cgN962gubXD67JEho3CPYx9eDFVLffBmj+tgLuvu4AP9jSz+OEKjnV2eV2SyLBQuIextdVNxMVEka9BOkMy95x8fnrNVFbsbOAbT66jWxOSyQigC6phbG1NM1PGpOme7SCYP62A2qY2fvLiZgozE3UdQ3xPLfcw1dnVzfqaZqYWhu+yepHmq58q47oZxfz369vZUNvsdTkiIaVwD1Pb6o/Q2tHFtCKFe7CYGf9y1WSmFqbz9OoaGo62e12SSMgo3MPU2uqei6lTFe5BFR8TzV3XXQDAoyt3aa548S2Fe5haU91MWkIMpdnqGw62oqwkrrmwiNqmNl7ZXOd1OSIhEdAFVTObC/wSiAbudc79+KTn/wG4FegE6oGvOOd2BbnWEWVtdRNTizI0dH4ITjcK9ezRaZSXZPLmlnomj06LqIXHRQIxYMvdzKKBu4HPApOBhWY2+aTD3gfKnXPnAU8B/xHsQkeS1vYuKvcfVn97iH3u3NGkJcby9OoaOtQ9Iz4TSLfMdGCbc26Hc64deByY3/cA59xrzrmW3s13gcLgljmyrN/TTFe3U7iHWEJsNH95fgF1h4/xmrpnxGcCCfcCoLrPdk3vvlO5BXihvyfMbLGZVZhZRX19feBVjjBrqhsBFO7DYMKoVM4vyuCtbQc4cOSY1+WIBE0g4d5fp2+/Q/zM7MtAOXBnf8875+5xzpU758pzc3MDr3KEWVPdRFFWItkp8V6XMiLMPSefmCjjj+v2el2KSNAEEu41QFGf7UKg9uSDzOwK4LvAPOecmkBD8P7uJqYVZXpdxoiRmhDLZZPyqNx/mM37DnldjkhQBBLuq4DxZjbWzOKABcCSvgeY2fnAb+gJdnVeDsH+Q23sbW7jfHXJDKtZ47LJSYnnj+v2anIx8YUBw9051wncBrwEbAKecM5tMLM7zGxe72F3AinAk2a2xsyWnOJ0MoD3d/cMXppWrHAfTjFRUVx13mgOHm3nvrd3el2OyJAFdJ+7c24psPSkfbf3eXxFkOsasdZUNxEbbUweneZ1KSPO+FGpTB6dxl2vbuOL5xeSn67ZOCVyaYRqmFlT3cjk0WkfWxZOhsfnzh1NZ7fjRy9s8roUkSFRuIeRrm7Huppm3QLpoazkOL56SRl/WFPLyp0NXpcjMmgK9zCyZf9hWtq71N/usb+ecxZj0hP4/pINdGlhD4lQCvcwsqZ3JkjdBumtxLhovnvlZDbuPcRjK089P41IOFO4h5E1u5vISIrVTJBh4HPn5jOzLIuf/qmSRs37LhFI4R5G1lQ3MU0zQYYFM+P786ZwuK2Tn71c6XU5ImdM4R4mDrd1sKVOM0GGk0n5adw4q4TfrdjNqipdXJXIonAPE+trmnFOk4WFm298ZiKFmYl866l1tHVo5KpEjoAGMUnoPbisCoBtdUeobWrzthg5ITk+hh9/8Tyuv3cFP395C//8ubO9LkkkIAr3MFHd2Ep2chxJcforCTezz8ph4fRi7n1rB3Mm5nLxuJxTHnu61Z+um1EcivJE+qVumTDgnKOmoUVLvYWxf/7cJMpyU/ib361m18GjXpcjMiCFexiobW7j8LFOhXsYS02I5d4by3EObn2ogsNtHV6XJHJaCvcw8P7unpWXijITPa5ETqc0J5n/vv4Cdhw4ylceXEWD7n+XMKYO3jCwZncTMVGmWQjDxOn6zQH+a8H5/P0Ta/jLX73DfTddxFl5KcNUmUjgFO5h4P3qJsZkJBITFfm/SA0UjH5w5XmjGZ2RwOKHK5h319vcPLuUxZ8cR3pSrNeliZygcPdYW0cX62uamTE2y+tS5AxcUJzJ7782mx+9sJm7X9vOw8t2cfnZeURHGUWZSWQmxxEbHfkf1hK5FO4eW7+nmfaubkqyk70uRQLU97eT2eNyKMtJ5q2tB3h5436Otn840Ck1PobM5Dgyk2LJTI4jOgqKMpMoykpidHoCMQp/CSGFu8eOD2sv1mRhEWt0eiJfKi/COcf+w8fY19xKw9EOGlvaaTzazu6GFtbvaeb1yvoT3xMTZZxbmM7F47KZPS6HC0oyT7lAi+6dl8FQuHusoqqRstxkUuL1VxHpzIz8tATy0z5+Ybyr2zFnYi7VDS1UN7aw48BRVu1s4Ndv7ODu17YTFxNFeUkmV5w9is9MGUVhpj7sZWiUKB7q7na8t6uRuVPyvS5FQiw6ynhr64ET2yVZyZRkJXPVeWOoOniU2Ogo3txazx3Pb+SO5zdyfnEGV19YyOfPG+Nh1RLJFO4e2lZ/hObWDspLM+no0oo/I1F8bDQT89NOdK/sPHCUFz/Yx7Pv1/DdZz/gX5/fxLTiDD5xVg6ZSXEeVyuRROHuoeP97eWlWSzfftDjasRLffvV0xNjuWlWKXuaWnl3x0FW9P6ZUZbNFZNGkRinxdNlYLpc76H3qhrJSYnTykvyMWZGYWYSV19YxDc+M5Hy0ize3X6Qn79cyZrqRq/LkwigcPfQql0NlJdkaeUlOa2MpDi+MK2Ar116Ftkp8TxRUcPT79XQ3tntdWkSxhTuHtnX3EZ1QyvlpVoMWwIzJiORxZeUcdmkPFbvbuRXr2+juqHF67IkTCncPfLujp4+9hljsz2uRCJJlBlXnD2KRbNLOdTWwdW/XsaW/Ye9LkvCkMLdI8u2HyAtIYbJY9K8LkUi0Pi8VBZ/chzOwTW/Xn5iZlGR4wIKdzOba2aVZrbNzL7dz/OXmNlqM+s0s6uDX6b/LO+9+yE6Sv3tMjj56Qk8/dcXk54Yy6IHVqkFLx8xYLibWTRwN/BZYDKw0Mwmn3TYbmAR8GiwC/Sj6oYWqhtauXicumRkaIqykvjdrTOIj4nihvtWqA9eTgik5T4d2Oac2+GcawceB+b3PcA5V+WcWwfo8n0Alvf2t89SuEsQFGUl8cgtM2ht7+Km+1fS3KpVoiSwcC8Aqvts1/Tuk0F6d/tBspPjmJCX6nUp4hMT81O596aLqG5s4e8ef5+ubo14HukCCff+OoUH9S/HzBabWYWZVdTX1w/8DT7knGP5joPMLMsmSv3tEkTTx2bx/XlTeL2ynp/9qdLrcsRjgUw/UAMU9dkuBGoH82LOuXuAewDKy8tHZNOi6mALe5vbmKkuGQmB62eU8MGeQ/zq9e00tXRwTkF6v8dpqmD/C6TlvgoYb2ZjzSwOWAAsCW1Z/nV8DhldTJVQ+cG8KUwtTOeZ92tobNEi3iPVgOHunOsEbgNeAjYBTzjnNpjZHWY2D8DMLjKzGuAa4DdmtiGURUeyN7fUk5+WQFmOVl6S0IiLieK/Fp6Pc/DEqmr1v49QAc0K6ZxbCiw9ad/tfR6voqe7Rk6jvbObt7cd4KqpYzSfjIRUSXYy86cV8ERFNa9uruPTk0d5XZIMM41QHUYVuxo4cqyTSyfmel2KjADTijK4oDiD1yvr2K3730cczec+jH712naio4w9Ta2nXRdTJFg+f94Ydhw4ypMV1fztZeOJi1F7bqTQ3/Qwqtx/mLE5ycTHaLEFGR4JsdFcfUEhB4+28+KGfV6XI8NI4T5MqhtaqD98jImjNHBJhldZbgqzx2Xz7o6DbK3T/DMjhcJ9mLxWWQegcBdPfGZKPrkp8Tz9Xg2t7V1elyPDQOE+TF7bXEd2chw5qfFelyIjUGx0FNeUF3LkWCfPrRvUGESJMAr3YXC4rYNl2w8yMV+tdvFOYWYSl07MY011E0vX7/W6HAkxhfsweHnjfo51dnPuKYaCiwyXORPzKMhI5LvPrqfucJvX5UgIKdyHwXNraynISKQoK8nrUmSEi44yrrmwkKPtXXzn6fU4p9GrfqVwD7HGo+28tfUAn586miiNSpUwkJeWwD/NncQrm+t4oqJ64G+QiKRwD7EXN+yjs9tx1XljvC5F5ISbLy5lVlk2P3huI9vqjnhdjoSAwj3EnltbS1lOMlO0ELaEkago4+fXTiU+Joqv/W61bo/0IU0/EEJ1h9pYvuMgf3vZeE0UJiExlGksRqcn8otrp3Hzg6u4/Q8fcOc1U4NYmXhN4R5CT62uwTmYN1VdMhJe+n4ozJmQy5Pv1XCss5uZZdlayMMnFO4h0tnVzf8s38XF47I5Ky/F63JETunys0dR29TGc2tryUiM9bocCRL1uYfIyxv3U9vcxqKLS70uReS0osxYML2IMRmJPLZqN+tqmrwuSYJA4R4iDyyrojAzkcvP1iIJEv7iY6K5cVYJKfEx3HDfSlbvbvS6JBkihXsIbKw9xMqdDdw4q4ToKF1IlciQmhDLrZ8oIyMpli/fu4J3th3wuiQZAoV7CNz/zk4SY6O5tlwXpiSyZCbH8eRfzaIoM4mbH1jFQ8uqNIo1Qincg6xy32GeWV3DwunFpCfp4pREnry0BP73r2byifE5/MuSDSx+5D0ajrZ7XZacIYV7kP3b0k2kJsTy9cvP8roUkUHLSIrjvpvK+d6VZ/N6ZR1z7nyN+97eSXtnt9elSYB0K2QQvV5Zx5tb6vnelWeTkRTndTkiQ2Jm3PrJMi6ZkMsPn9/ID5/fyEPLqnoX3s7sdz1W3SMfPtRyD5KOrm7+fekmSrKTuHFWqdfliATNhFGpPPyV6Tyw6CKykuNYsraWn7y4mefX1bK3udXr8uQU1HIPkh8t3cyW/Uf47Y3lWmFefMfMuHRSHnMm5vLjFzbzzvaDrNjZwLLtBxmdnsAFxZlMLcrwukzpQ+EeBM+vq+X+d3ay6OJSPj1Z97VLZBtovpqS7GRKspNpae9kXU0zq3c38sf1e3nhg70s236Aq84bw2emjCI1QTcUeEnhPkSV+w7zrafWcWFJJv/8ubO9Lkdk2CTFxTCzLJuZZdnsP9TG+7ub2Lr/CP/45Frin43iskl5XHneaC6ZkEuagn7YKdyHoKKqgVsfriA5Poa7r7tA3TEyYo1KS2DuOfk456huaGFtTTNvbT3ACx/sI8qgNDuZ62YUc+mkPMpykjVL6jAwrwYolJeXu4qKCk9ee6icczy3bi/feHIthRmJPHjzdIqze5bQG8oUrCJ+0tXdE/Sb9x2mcv8h9h86BsCY9ATOL87k/OIMLijJZMqYNOJjoj2uNnKY2XvOufKBjguo5W5mc4FfAtHAvc65H5/0fDzwMHAhcBC41jlXdaZFR4LKfYf51z9u5K2tB7iwJJN7bywnM1m3PYqcLDrKKM1JpjQnmbnn5HPJhBxe21zHyqpGVu/q6acHiIuOIicljlFpCeSnJ5CflsCo9ARS42MwM91eOUgDhruZRQN3A58GaoBVZrbEObexz2G3AI3OubPMbAHwE+DaUBTshbrDbby2uY7n1vZcMEqJj+H/fX4yN8wsUVeMSIAKM5O4YVYpN/TeKlx3qI3Vu5t4v7qRVzfVsb3+CO9XfzgjZVJcNPlpCWzZf5hJ+alMyE+lNDuZzKRYdesEIJCW+3Rgm3NuB4CZPQ7MB/qG+3zg+72PnwLuMjNzw9zn45zDOeh2ju7er8e3HT2/JnZ0dXOss5v2zm6OdXZxrKOb9q5ujnV0c+RYB82tHRw82s7epjb2NLWysfYQ+w61AVCancRtl57FzbPHqrUucoZO1WVZkpXMzbPHAtByrJN9h9p6/jS3sf9QG09UVNPSZxnA1PgYirOTKMlOYkx6IlkpcWQlxZGV3PMnOT6G+Jgo4mOjiYuOIi4miugo4/jHgRkYRt/Ph5P3GT23f1rvc/RuR5JAwr0A6LtEeg0w41THOOc6zawZyAaCPq3cfW/v5KcvVX4kuI+HeTAlxEaRkRhHfnoCF5RkMjY7mTEZCZiZgl0kRJLiYyjLTaEs98MFbhZcVER1Ywtb9x9hV0MLuw8eZVdDC5v3HubVzXW0dQzvlAg9HwTHH3/4AXDi4+OjX/p8YNiJ7ds/P5kF00Pb3RRIuPf3cXVylAZyDGa2GFjcu3nEzCoDeH2v5HCKD6frh7mQMHfK90luUjgnAAAEsElEQVQ+Qu9TYD72Pvnx/9vCH8LCwX97SSAHBRLuNUBRn+1CoPYUx9SYWQyQDjScfCLn3D3APYEU5jUzqwjkivRIp/cpMHqfAqP3KXgCuRq4ChhvZmPNLA5YACw56ZglwE29j68GXh3u/nYREfnQgC333j7024CX6LkV8n7n3AYzuwOocM4tAe4DHjGzbfS02BeEsmgRETm9gO5zd84tBZaetO/2Po/bgGuCW5rnIqL7KAzofQqM3qfA6H0KEs9GqIqISOhoBI6IiA8p3E9iZn9vZhvM7AMze8zMEryuKVyY2f1mVmdmH/TZl2VmL5vZ1t6vmV7WGA5O8T7daWabzWydmT1rZiN+8vP+3qc+z33DzJyZ5XhRmx8o3PswswLg60C5c+4cei4g6+Lwhx4E5p6079vAK8658cArvdsj3YN8/H16GTjHOXcesAX4znAXFYYe5OPvE2ZWRM90J5qFbwgU7h8XAyT23q+fxMfv6R+xnHNv8vHxC/OBh3ofPwR8YViLCkP9vU/OuT855zp7N9+lZ7zIiHaKf08AvwC+RT8DISVwCvc+nHN7gJ/S02LYCzQ75/7kbVVhb5Rzbi9A79c8j+uJBF8BXvC6iHBkZvOAPc65tV7XEukU7n309hfPB8YCY4BkM/uyt1WJn5jZd4FO4Hde1xJuzCwJ+C5w+0DHysAU7h91BbDTOVfvnOsAngEu9rimcLffzEYD9H6t87iesGVmNwGfB67XCO5+jaOnYbXWzKro6bpabWb5nlYVoRTuH7UbmGlmSdYzv+flwCaPawp3faeeuAn4g4e1hK3eBW/+CZjnnGvxup5w5Jxb75zLc86VOudK6Zmz6gLn3D6PS4tICvc+nHMr6JmPfjWwnp73RyPmepnZY8ByYKKZ1ZjZLcCPgU+b2VZ67nD48enOMRKc4n26C0gFXjazNWb2a0+LDAOneJ8kSDRCVUTEh9RyFxHxIYW7iIgPKdxFRHxI4S4i4kMKdxERH1K4y4hgZlWDmWHQzB40s6vP4PjS/mY5FBluCncRER9SuIvvmNnvzey93nn5F/fz/I2986qvNbNHeveVmNkrvftfMbPiPt9yiZktM7Mdx1vx1uPO3nn/15vZtcP044kEJKA1VEUizFeccw1mlgisMrOnjz9hZlPomZxqtnPugJll9T51F/Cwc+4hM/sK8F98OH3xaOATwCR6plt4CvgiMA2YCuT0vs6bw/CziQRELXfxo6+b2Vp65k0vAsb3ee4y4Cnn3AEA59zx+cRnAY/2Pn6EnjA/7vfOuW7n3EZgVO++TwCPOee6nHP7gTeAi0Ly04gMglru4itmNoee2T1nOedazOx1oO9SiUZgi0D0PebYSd/f96tIWFLLXfwmHWjsDfZJwMyTnn8F+JKZZUPPGrC9+5fx4ZKK1wNvD/A6bwLXmlm0meUClwArg/EDiASDWu7iNy8CXzWzdUAlPV0zJzjnNpjZvwFvmFkX8D6wiJ61c+83s28C9cDNA7zOs/R05aylp5X/LefcPjMrDd6PIjJ4mhVSRMSH1C0jIuJDCncRER9SuIuI+JDCXUTEhxTuIiI+pHAXEfEhhbuIiA8p3EVEfOj/A/W9EZD/Qq3LAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(wine['alcohol'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda/envs/ml/lib/python3.7/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1aceb7b8>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(wine['fixed acidity'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda/envs/ml/lib/python3.7/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1b1bf588>"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAEKCAYAAAAyx7/DAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzt3Xl4XNd93vHvb7BjsC8kARALNy3URomUZEmWZcmOrViJZSeurdiJ5a2q7NhNm2at+/RJnTRN0qxu1LiKo9hWLcu7ayuObVmbrY0SJVEkRZEUSJEECZDY93Uwp3/MHWoIAsQAmDszF/N+ngcPBnfuzP3havTy4NxzzjXnHCIiEhyhTBcgIiJLo+AWEQkYBbeISMAouEVEAkbBLSISMApuEZGAUXCLiASMgltEJGAU3CIiAZPvx5vW1dW5trY2P95aRGRVeuGFF3qdc/XJ7OtLcLe1tbFr1y4/3lpEZFUys2PJ7quuEhGRgFFwi4gETFJdJWZ2FBgBZoGIc26Hn0WJiMjCltLHfbNzrte3SkREJCnqKhERCZhkg9sBPzGzF8zsLj8LEhGR80u2q+QG51ynma0BHjazA865nyXu4AX6XQAtLS0pLlNEROKSanE75zq9793Ad4Fr5tnnXufcDufcjvr6pMaQi4jIMiwa3GYWNrPy+GPgHcA+vwsTEZH5JdNVshb4rpnF93/AOfcjX6vKMQ/sPD7v9g9eqy4nETnXosHtnDsCXJGGWkREJAkaDigiEjAKbhGRgFFwi4gEjIJbRCRgFNwiIgGj4BYRCRgFt4hIwCi4RUQCRsEtIhIwCm4RkYBRcIuIBIyCW0QkYBTcIiIBo+AWEQkYBbeISMAouEVEAkbBLSISMApuEZGAUXCLiASMgltEJGAU3CIiAaPgFhEJGAW3iEjAKLhFRAJGwS0iEjAKbhGRgFFwi4gEjIJbRCRgFNwiIgGj4BYRCZj8TBeQq5xz/PNTR7mypSrTpYhIwCi4M+SljkE+99B+8kLG2y5aw1suqCdklumyRCQA1FWSId958QTFBSHeeclafrL/NN/c1ZHpkkQkIJIObjPLM7OXzOwhPwvKBdORKA/t6eIdW9dxzwev4toNNew9OcRUZDbTpYlIACylxf1bwKt+FZJLHjvYzeD4DO+9qgkzY2tjBVEHx/rGM12aiARAUsFtZuuB24Av+ltObvjuiyepKyvixs11ALTWhAkZvN47luHKRCQIkm1x/y3we0DUx1pywuD4NI8e6ObdVzSSnxc7/YX5IdZXl3KkZzTD1YlIECwa3Gb2S0C3c+6FRfa7y8x2mdmunp6elBW42vzkldNMz0b5lauaztq+oS7MycEJ9XOLyKKSaXHfALzbzI4CDwK3mNn/nbuTc+5e59wO59yO+vr6FJe5euzvGqasKJ9LGivO2r6xLkzUwXH1c4vIIhYNbufcHzrn1jvn2oA7gEedc7/ue2WrVHv3KJvqw9icMdsttaWEDI6on1tEFqFx3Gl2uGeUTfVl52wvys9jfXWpLlCKyKKWFNzOucedc7/kVzGr3ehUhK6hSTatOTe4IdbPfWJgXP3cInJeanGnUXzUyHwtbogFd9TBiYGJdJYlIgGj4E6jw15wb14Tnvf5NeVFAPSNTqetJhEJHgV3Gh3uHiM/ZLTWzh/cFSUF5IeMvtGpNFcmIkGi4E6j9u5RWmpLKcib/7SHzKgJF9I3pha3iCxMwZ1Gh3tG2bxA/3ZcbbiQXrW4ReQ8FNxpEpmNcrRvbMERJXG1ZUX0j00TdS5NlYlI0Ci40+R4/zgzs27BESVxtWWFRKKO4YmZNFUmIkGj4E6Twz2xiTWbF2txh72RJernFpEFKLjTpL07NhRwY/38I0riassKAQ0JFJGFKbjT5HDPKGsriqgoLjjvfpXxIYFjukApIvNTcKfJQmuUzBUyozpcqBa3iCxId3lPk47+CdpqS3lg5/FF960NF6rFLSILUos7DaYjUXpHp6gsPX83SVxdWRF9o9NEoxoSKCLnUnCnwenhSQAqF+nfjqsJx4YEnh6Z9LMsEQkoBXcanIoHd0nyLW7QzYNFZH4K7jToGlpacNeGY0MCj+k2ZiIyDwV3GnQNxtbXTja4K0sLyAsZR9XiFpF5KLjToGtokvKifIoK8pLaP2RGTWkhR/sU3CJyLgV3GpwammRdZfGSXlMdLqCjX3fCEZFzKbjToGt46cFdEy6ko38cp1UCRWQOBXcadA1O0FhZsqTXVJcWMjIVYUirBIrIHApun83MRukZnVpWixtQd4mInEPB7bPukSmcg4al9nGXxoL7eL+GBIrI2RTcPjs1FGsxL7vFPaDgFpGzKbh9Fp9807DEPu7igjyqSgvoUItbROZQcPusa9AL7qqltbgBWmpK1VUiIudQcPusa2iScGEe5UVLX0G3ubqUEwO6OCkiZ1Nw++zU8ATrKosxsyW/trmmlBMD48xqeVcRSaDg9lnX0OSS+7fjmmtKmJl1Z5aFFREBBbfvugYnlzwUMK6lphTQkEAROZuC20eR2SjdI8sP7ubqWHBrZImIJFJw+6hndIqog3XL7CpprCohZApuETmbgttHb4zhXl6LuzA/RENlCR0aWSIiCRYNbjMrNrPnzOxlM3vFzP5bOgpbDU55wb3UWZOJmmtK1MctImdJpsU9BdzinLsC2AbcamZv8res1aHTu/PNUlcGTNRcXaquEhE5y6LB7WJGvR8LvC8NLE7CqaFJSgryqChZ+uSbuJaaUrpHppiYnk1hZSISZEklipnlAS8Am4F7nHM7fa0q4B7YeRyAna/3Ey7K42vPdSz7vdrqwkDsju9bGytSUp+IBFtSFyedc7POuW3AeuAaM7t07j5mdpeZ7TKzXT09PamuM5CGJmaoSPIGwQvZVF8GwJHe0UX2FJFcsaRRJc65QeBx4NZ5nrvXObfDObejvr4+ReUF29DEDJXFKwvuDV6L+0iPbhwsIjHJjCqpN7Mq73EJ8HbggN+FBV3UOUYmZ6gsXVlwlxTm0VRVwpEetbhFJCaZPu4G4MteP3cI+IZz7iF/ywq+0ckIUQeVK+wqAdhYH+ZIr1rcIhKzaHA75/YAV6ahllUlfpPflXaVAGysC/PtF0/inFvWKoMisrpo5qRP4sG90ouTABvryxiditA9MrXi9xKR4FNw+yQe3FUp6ioBOKx+bhFBwe2b4YkZ8kNGSWHeit9rY3xIoEaWiAgKbt8MTc5QWVKQkj7phopiSgryFNwiAii4fZOKyTdxoZCxoS6sSTgiAiQ55V2Wbmhihrba8IreIz51HiAvZOw5MbTSskRkFVCL2wdR5xiemEnJGO64+vIiBsammZzRYlMiuU7B7YOxqdRNvomrKyvCAcf6tMSrSK5TcPvgzOSbVLa4y4oANPVdRBTcfhhO4eSbuLryQgBe61Zwi+Q6BbcPBn1ocRfl51EbLuTVruGUvaeIBJOC2wfDEzPkhYxwCibfJGqoLGa/glsk5ym4fTA0kbrJN4nWVZZwrG+c0alISt9XRIJFwe2DoYkIFSlYFXCuRu9u8QfU6hbJaQpuHwxNTFO5ghsEL6ShKna3eHWXiOQ2BXeKRaOO4ckIlSWFKX/viuJ8qkoLdIFSJMcpuFOsf3ya2ajzpcVtZmxtqGB/p4JbJJcpuFPs1NAkkNqhgIm2NlRw4NQIkdmoL+8vItlPwZ1iXV5wp3LyTaKtjRVMRaIc7dMSryK5SsGdYl1DE4B/Le6LGyoAeEXdJSI5S8GdYl1Dk+SZES7yZ8XcTfVlFOaFNLJEJIcpuFPs1NAkFSX5hHy6G3thfogta8t4tWvEl/cXkeyn4E6xrqEJ3/q34y5trGTPiUGcc74eR0Syk4I7xbqGJn3r3467qrWKwfEZjvTqAqVILlJwp5BzLhbcPkx3T7S9tRqAF44N+HocEclOCu4UGhifYToSpbLU3+DeWFdGVWkBLyq4RXKSgjuF4kMB/VhgKlEoZFzVUq0Wt0iOUnCnkN+zJhNtb63mte5RBsenfT+WiGQXBXcKdaYxuK9qifVzv3R80PdjiUh2UXCn0KmhCfJDRlmxP5NvEl3RXEleyNRdIpKDFNwp1DU0ydqKYt8m3yQqLcxna0OFglskBym4U+jU0CTrvLvUpMP21mp2dwxqpUCRHKPgTqGuDAT3xMwsB05p+rtILlk0uM2s2cweM7NXzewVM/utdBQWNLHJNxM0VKQ3uEETcURyTTIt7gjwn5xzFwNvAn7TzLb6W1bwDIzPMDkTpdG7L2Q6NFaV0FBZzC4Ft0hOWTS4nXNdzrkXvccjwKtAk9+FBc3x/nEAWmpK03rcq1qrNYNSJMcsqY/bzNqAK4GdfhQTZGeCuza9wb29pZqTgxNnZm2KyOqX9IBjMysDvg38B+fcOav4m9ldwF0ALS0tKSsw2z2w8zgAjx/sBuDp9j4K89N3zTfez/3isUFuuzx93TQikjlJJYyZFRAL7a86574z3z7OuXudczucczvq6+tTWWMg9I9NU1aUn9bQhtg9KIsLQrpAKZJDFm1xm5kB/wS86pz7a/9LCqb+sWlqwoW+Hyfewk/UUFnCT/afYvOasnOe++C1ufPXj0iuSKZ5eAPwG8AtZrbb+3qXz3UFTv94eoJ7Pi01pXQOTjCjiTgiOWHRFrdz7knA/zncATYbdQyNz1DdnJngbq0p5QkHJwYm2FAXzkgNIpI+mjmZAoPj0zjIWIu72RuCeLxPtzITyQX+L2OXA/q9NbGrw/4v5zqfcFE+dWWFZ4YkLma+fnJQf7hIUKjFnQL9Y7HgrinNTIsboLm6lBMDE7rzu0gOUHCnwMDYNHkhoyINN1BYSFN1CSNTEYYnIxmrQUTSQ8GdAv1j01SXFqRlHe6FNFfH+rk7kuwuEZHgUnCnQCaHAsatqywmz4yTg5r6LrLaKbhTYGBshuoM9m8DFOSFWFdZTMeAWtwiq52Ce4UmpmeZmJnNeIsbYH11CScHJojqAqXIqqbgXqEzQwEz3OIGWF9dylQkSu/IVKZLEREfKbhX6MxQwCxpcUNsBqWIrF4K7hXqG421brMhuOvLiyjKD3FiUP3cIquZgnuFTg9PUllSQHFBXqZLIWRGU1XJklrco1MRDupmwyKBouBeoe6RKdZWFGW6jDPWV5fQNTiZ9EqBP3i5k688c5TRKU3cEQkKBfcKRGaj9IxMsbY8fXd2X0xLTZhZ5+hMYjx3/9g0+04O4YBjWqBKJDAU3CtwrH+cSNSxpiKLgtu75+XRvsX7uZ9s7yVkRn7IONqr4BYJCq0OuAKvnY71DWdTV0mZt1JgrAW98C3kxqYivHCsn23NVfSPTycV9CKSHdTiXoFDp0cBWJNFXSUArbVhjvWNn3cizrOv9zEz63jzljraakvpGppQP7dIQCi4V+Dg6RFqwoVpv0HwYlprSpmYmV1wIo5zjude7+fCteWsrSimrTZM1MGLuuGwSCBkV+IEzGunR1hTnj3dJHFttbHblx1bYKXA08NTjExGuLSpAojds9KA54/2p6tEEVkBBfcyTUeiHOkZY20WXZiMqy0rJFyYt+BIkfaeWBfPpvrYXeGLCvJorCrhudcV3CJBoOBepqN9Y0SiLqsuTMaZ2Zl+7vm0d49QV1ZEVcL6Km21pezuGGQqMpuuMkVkmRTcy3TIG1GSbRcm41prS+kbm6ZnTj/3dCTK671jbF5z9t3g2+rCTEWi7D0xlM4yRWQZFNzLdOj0KCGLrQ+SjVq9O7/P7bd+6fgAM7OOzV43SVyLt/8eBbdI1lNwL9OhUyO01YYpyMvOU9hYXUK4KJ9vv3DirO1PtvdiwMY5wV1WlE9ZUX7Sd4oXkcwJ7AScB3YeP2fbB69tSdvxD5wa5qJ1FWk73lLlh0Jc01bDowe7Odo7RltdrGvkyfZe1leXnLMolpnRXFOq4BYJgOxsLma5QW+m4eXNlZku5byu3VhDfsj48jNHARienOHljkE2rymbd/9WBbdIICi4l2F3xyAA25qrMlzJ+VUUF3DbZQ18c9cJ+kan+M/f2UvUwQVry+fdv6W2lI7+caJR3fpMJJspuJdhd8cgZnD5+uwOboCP3rCB0akI7/ibn/HQni5+/9aLaK0Nz7tvc03s1mfduvWZSFZTcC/Dyx2DbFlTRllR9l8iuKK5ih2t1YxORfjfH7qKT75104L7xkeiqLtEJLtlf/JkGeccuzsG+YWtazNdStLu/fAOxqcjrK8uPe9+8SGBx/rGuGZDTTpKE5FlUHAv0fH+cQbGZ9jWXJ3pUpJWEy5M6p6YjVUlhAw61OIWyWrqKlmi+IXJK7J8RMlyFOaHaKwqUVeJSJZTcC/R7o5BSgryuHCBkRlB11JTuuCqgiKSHRYNbjO7z8y6zWxfOgrKdrs7BrmsqZL8LJ0xuVKt3pBAEcleyaTPl4Bbfa4jEKYjUV7pHF6V3SRxzTWl9I5OM6a74YhkrUUvTjrnfmZmbf6Xkv1e6RxiOhIN1IXJ+ZYGOJ+WhCGBFzdk75R+kVy2Ov/e98nPX+vFDN60cfUOlWutiU3O0QVKkeyVsuA2s7vMbJeZ7erp6UnV22aVJw71cHlTJbVl2bmUayqcaXHrru8iWStlwe2cu9c5t8M5t6O+vj5Vb5s1hsZneOn4ADddsPp+t0SVpQVUFGt5V5Fspq6SJD3Z3kvUwU0Xru7gBmitDSu4RbJYMsMBvwY8A1xoZifM7OP+l5V9njjUTUVxPlcEYGGplWqrC9PePZrpMkRkAcmMKvm1dBSSzZxzPHGohxu31K/a8duJtjZU8IOXOxkan6GytCDT5YjIHKs/hVLg4OkRTg9Prfr+7bhLGmPDAPd3DWe4EhGZT6AXmRqfjlBckEfIzNfjPH4wNkqmb2x6yeOigyg+fvuVziGu21Sb4WpEZK7AtrhPDIzz5z86wEN7unw9zmzU8fXnO1hfXUJlSW50G9SXF7GmvEgtbpEsFcjg7hyc4P5njhGZdTz3eh/9Y9O+HetH+07xeu8YN27JjW6SuK2NFezvVHCLZKPABff4dIRPfHkX07NRPnrDBkJmPPLqaV+O5Zzjnsfa2VgfPtPvmysuaaygvXuUqchspksRkTkCF9zfe6mT/V3DfGBHM5vXlHHdxlp2dwzSPTyZ8mM9fqiH/V3D3H3TJt/70bPBAzuPn/nqH5shEnV8/pH2TJclInMELrj3nBikqrSAC9fF1sN+ywX1FOSH+GmKW93RqOOeR9tprCzmPduaUvreQdBQWQxA1+BEhisRkbkCF9x7Tw5xWVMl5rWAw0X53LCpln2dwxw6PZKSYzjn+ON/2c+uYwN85m1bKMwP3GlasZpwIYX5IbqGUv+XjIisTKCGA05FZjl0eoRP3LjxrO03bKrjqfY+7nmsnb+748qk32++oX0fvLaFv3+0nX9+6igfu2EDd1zdvOK6gyhkRkNFMZ1DanGLZJtANSUPnRplZtZxaePZNzIoLcrn2g01/ODlTo72ji37/aPO8dcPH+KvHj7Er1zVxH+57eIzLftc1FBVzKmhSaJRl+lSRCRBoIJ778khAC5rOvcONDdsqSM/L8QXnji8rPcen47wlWeO8vlHXuN929fz5796OaFQ7oY2QENlCVORKK/3Lf8fQxFJvUB1lew9OURlSQHNNSXnPFdRXMAdVzfzteeO88m3bqK1Nrzg+0SjjuHJGUanIoxNRXjx2AC7jg0wHYly+7ZGrmyu4pu7Tvj5qwTC5jVlGPD/XjrJb7/jwkyXIyKeQAX3vpNDXNpUsWD3xd03beK7L53kN/7pOR686000Vr0R8N3Dk3zp6aM8f7Sf/Z3DjE2/MT45ZLGFlW66YA1N1ef+o5CrqksLuWBtOQ8+38Fn3raFghxYYEskCAIT3NORKAdPjfDRG9oW3KexqoT7P34tv/HFndxx77P88XsuZWwqwvNH+3lg53EiUccV6yt53/b1tNSG2X18gPxQiAvXlVORI9PZl+raDTV85dlj/HT/aX7xsoZMlyMiBCi4D50eYXo2yqXz9G8n2tZcxf2fiIX3nfc9B0BeyHjvlU18+ubNtNW90YXywM48X2teDS5YV05TVQlf3XlcwS2SJQIT3PvOc2Fyrm3NVfz4P76FIz1j1JYV8lR7L6WF+Tx9uI+nD/f5XeqqEjLj165p5i9/cojXe8fYULfwtQMRSY/AdFruPTlEeXE+rbWlSe3fWFXCm7fUcXFDBaWFgfn3KSu9/+pm8kPG3zx8COc0NFAk0wIT3Ps6h7m0sTKnx1VnypryYv7927bw/Zc7+YsfH8x0OSI5LxBN0ZnZKK92DXPnda3n3S8XbnKQKZ+5ZTOnhyf5h8cPUxsuPGf2qoikTyCC+7XTo0xHFr8wKf4xMz53+6X0j03zJ//yKr2j0/zuOy8kL8cnKYlkQiCCe19n8hcmJfUS/5K5flMdfaPTfOGJw7R3j/B3d1xJuCgQHyORVSMQfdz7Tg5RVpRP23lmQ0p65IWM27c18ke/vJVHD3TzoS/uZHDcvzsQici5AhHce08OsbWxIufXDskWZsZHbtjAP/z6dvZ3DnPHvc/SPaLlX0XSJeuDO+JdmFQ3SfZ55yXruO8jV3Osb5z3f+EZTgyMZ7okkZyQ9cF9uGeMyZmogjvLxG9xdrx/nDuva+XU8CS3ff5J2rtHM12ayKqX9cEdX8pVI0qyV0ttmH9740YiUcf7/88zfGNXB7Naw1vEN1k/HGDfySFKC/M01TrLNVSW8O9u3Mg3Xujg9761h7/88UEuaawgXJRPyIym6hJODU0SiTryDJqqS7jlorVsa67SkEKRJcr64N57cohLGiv0P3cA1JUX8cmbNrGvc5hHXj3N4wd7iLe788yoKImFuAMGx6e557HD1Huv+dCbWijKn3/Rr1FvzfSvPnuMqUiUt1xQT3HBG/t+8NoW/385kSyS1cE9G3WxUQvX5OZ9H4PIzLisqZLLmiqJOsf49CyzUUd5cSy04yamZ6krL+Trz3fwuYf2c99Tr/PpmzfzniubzoRyNOr41osn+LN/PUD/WGzIoQGHukf46PUbNH5cclZWf/L3nhxiYmZWFyYDKmRG2QLhWlKYx+3bmnj3FY082d7LX/zoIH/wnb381cOHeOcla5mYjnLw9DD7Tg6zvbWav/3ANg6dHqGjf5yv7jzOP/78CB978wYqirWOuuSerA7u+585RmlhHm/fujbTpYgPEmdk3nF1M1e31fDz13r43kudVJYUUBMu5C9+9XLet309oZBxYmCCC9dVcOf1bdz/zDHuf+YYd71Fa6ZI7sna4O4bneIHezr5wI5mtapygJmxeU0Zm9eUnbU9EnU8+HzHWds21Zfxgaubuf/ZY3zvpZN8+LpWrRopOSVrhwM++HwH05Eod15//hUBJTdd3FDB2y5ew0sdg/zjz49kuhyRtEoquM3sVjM7aGbtZvYHfhc1Mxvl/meOceOWOjavKff7cBJQN1+4hksaK/jTHx7g7vtf4PSwpt1Lbli0q8TM8oB7gF8ATgDPm9n3nXP7/SrqG7s6ODU8yZ+851K/DiGrQMiMO65uYWRqhr/76Ws8+Ve9XLeplksaK7hwbTltdWFaa0t1ByRZdZL5RF8DtDvnjgCY2YPA7UDKg/to7xh/8i/7+emr3VzWVMnNF61J9SFklckLGZ9662Zuu6yBzz/Szu6OAX766mkS77C2pryIttpYiLfVhWmsKqaypICK4oIzC5fFe8jNLOGx9x0787goP0RxQR5FBbHvBaEQEzOzjE1FODk4wfG+cY72jXGsf5yTAxNMRaLMRqNUlhTQWhumzauhtSZMXXkhlSUFlBTkpaSP3jmHc+C8x1EHDnfmXLiEn+P7AGfG2ofMyA8ZeSEjzyxnF3VzzjEbdUSijpnZKJHZ2OOQxT5v8a/E85XuayzJBHcTkHh16ARwbaoLGZqY4bbP/xyA37/1Ij725jZNupGkxEenbG+tZntrNdORKL2jU/SNTdPnfe8emeTAqWGGJyO+15MfMtZXl7C+upSpmSkK8kL0jExz8NTIgscPWewfjfh3IxakZm8EbtQB84SvX6sLmPFGMHH2/4uOcw863+1I5y1t3v2W/37z3Qd1/v3mKyY1zGKTzNZWFPPUH9zi34E8yQT3fOl5zikws7uAu7wfR81s2Tcn/NQfw6cW360O6F3uMdIoKHWCak2Zw2f/mNW1zqFaV+AwYH8471PJ1Jr0SIxkgvsEkDh1cT3QOXcn59y9wL3JHnilzGyXc25Huo63XEGpE1SrX1SrP3K51mRGlTwPbDGzDWZWCNwBfD9VBYiIyNIs2uJ2zkXM7NPAj4E84D7n3Cu+VyYiIvNKapyUc+6HwA99rmWp0tYts0JBqRNUq19Uqz9ytlab74qsiIhkr6yd8i4iIvPLuuBebHq9mRWZ2de953eaWVvCc3/obT9oZu/Mglp/28z2m9keM3vEzFoTnps1s93el+8Xe5Oo9SNm1pNQ0ycSnrvTzF7zvu7Mglr/JqHOQ2Y2mPBc2s6rmd1nZt1mtm+B583MPu/9HnvM7KqE59J9Ther9UNejXvM7GkzuyLhuaNmttc7p7uyoNa3mtlQwn/n/5rwXFqX50ii1t9NqHOf9/ms8Z5b/nmNzbbKji9iFz8PAxuBQuBlYOucfT4FfMF7fAfwde/xVm//ImCD9z55Ga71ZqDUe/zJeK3ez6NZdl4/Avz9PK+tAY5436u9x9WZrHXO/p8hdsE8E+f1LcBVwL4Fnn8X8K/E5kK8CdiZiXOaZK3Xx2sAfjFeq/fzUaAui87rW4GHVvrZSUetc/b9ZeDRVJzXbGtxn5le75ybBuLT6xPdDnzZe/wt4G0Wm296O/Cgc27KOfc60O69X8Zqdc495pwb9358ltgY+ExI5rwu5J3Aw865fufcAPAwcKtPdcLSa/014Gs+1rMg59zPgP7z7HI78BUX8yxQZWYNpP+cLlqrc+5prxbI7Gc1mfO6kJV8zpdlibWm7LOabcE93/T6poX2cc5FgCGgNsnXptJSj/dxYq2vuGIz22Vmz5rZe/woMEGytf6q96fyt8wsPukqa8+r1/W0AXg0YXM6z+tiFvpd0n1Ol2ruZ9UBPzGzFyw2QzobXGdmL5vZv5q6dXAkAAAEm0lEQVTZJd62rD2vZlZK7B/nbydsXvZ5zbZl05KZXr/QPklNzU+hpI9nZr8O7ABuStjc4pzrNLONwKNmttc5d3i+16dAMrX+APiac27KzO4m9lfNLUm+NpWWcrw7gG8552YTtqXzvC4mWz6rSTOzm4kF95sTNt/gndM1wMNmdsBraWbKi0Crc27UzN4FfA/YQhafV2LdJE855xJb58s+r9nW4k5mev2ZfcwsH6gk9qdKUlPzUyip45nZ24HPAu92zk3FtzvnOr3vR4DHgSszWatzri+hvn8Etif72hRbyvHuYM6fnmk+r4tZ6HdJ9zlNipldDnwRuN051xffnnBOu4Hv4m8X5KKcc8POuVHv8Q+BAjOrI0vPq+d8n9Wln1c/O+6X0dGfT+xCzQbeuLhwyZx9fpOzL05+w3t8CWdfnDyCvxcnk6n1SmIXS7bM2V4NFHmP64DX8PEiSpK1NiQ8fi/wrPe4Bnjdq7nae1yTyVq9/S4kdnHHMnVeveO0sfBFtNs4++Lkc5k4p0nW2kLsutD1c7aHgfKEx08Dt2a41nXx/+7Ewu64d46T+uyks1bv+XjjMpyq8+rrL7TMk/Au4JAXeJ/1tn2OWIsVoBj4pvchew7YmPDaz3qvOwj8YhbU+lPgNLDb+/q+t/16YK/3wdoLfDwLav0fwCteTY8BFyW89mPe+W4HPprpWr2f/wj4szmvS+t5JdaC6gJmiLX2Pg7cDdztPW/EbkJy2KtnRwbP6WK1fhEYSPis7vK2b/TO58ve5+OzWVDrpxM+q8+S8I/NfJ+dTNbq7fMRYgMnEl+3ovOqmZMiIgGTbX3cIiKyCAW3iEjAKLhFRAJGwS0iEjAKbhGRgFFwSyCY2d1m9mHv8UfMrPE8+37Om/jkax1ztrcttEKcSKppOKAEjpk9DvyOc+6cpTDNLM+dPQU+XTW1EVux7tJ0H1tyj1rcknXM7MPeYlcvm9n93rY/MrPfMbP3EVv35aveOsYl3rrG/9XMngT+jZl9ydsPM7vaW1/6ZTN7zszK5xyrzGJrpb/orY18ezJ1eI+3e889Q2xGr0haZNsiU5LjvJXePktsAZ7e+KLzcc65b1ns5tVnWtyxVX2ZdM692fv5Vu97IfB14APOuefNrAKYmHPISeC9zrlhb72LZy12A4at56vD88/AZ5xzT5jZ/0zNGRBZnFrckm1uIbbiXy+AO3s1tfP5+jzbLgS6nHPPe+817GJLAScy4E/NbA+xJQqagLWL1WFmlUCVc+4Jb9P9SdYpsmJqcUu2MZa3FOfYMt/rQ0A9sN05N2NmR4mth7PYa5dbp8iKqcUt2eYR4P1mVguwQBfFCFA+z/a5DgCNZna1917l3lLAiSqBbi+0bwbi9wU9bx3OuUFgyMzi61Z/KIl6RFJCLW7JKs65V8zsvwNPmNks8BKx1dUSfQn4gplNANed572mzewDwP8ysxJi/dtvB0YTdvsq8APvZq27iYV9snV8FLjPzMaBHy/j1xVZFg0HFBEJGHWViIgEjIJbRCRgFNwiIgGj4BYRCRgFt4hIwCi4RUQCRsEtIhIwCm4RkYD5/0rMn3BUXAtTAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(wine['citric acid'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize the dependency of the target on each feature from 2.1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1a621860>"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(y=\"winetype\", x=\"citric acid\", data=wine, orient=\"h\", palette=\"Set2\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1b4c7b38>"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(y=\"winetype\", x=\"alcohol\", data=wine, orient=\"h\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1b590eb8>"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAEKCAYAAAARnO4WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAGIZJREFUeJzt3X2QXXV9x/H3Nxshy0N4CizJ4pDEiUFBDLC7KlZbJBmhTdkO4AibdbU4TTE+jWPS6ug4UzvtONjWMrVLixFw2V1QQ+pSLdYEU7EQ3IdA1BjTmKw83IXNA8KGsEnI5ts/7oN3s3ez5ybn5Oze3+f1T+4953d+53vmnvPJb3/33nPN3RERkco3Le0CRETk5FDgi4gEQoEvIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+CIigZiedgHFZs2a5XPnzk27DBGRKaOvr2+Pu58fpe2kCvy5c+fS29ubdhkiIlOGmT0Tta2mdEREAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUBMqi9eSfm6urrIZDKR2+/ZsweAWbNmlbWf2tpaGhsby9pGRCYXBf4Ul8lkeOa5DKedfUGk9q8NHwTA970eeR+vvbzruGoTkclFgV8BTjv7Ai69pilS2y0bOgEity/eRkSmNs3hi4gEQoEvIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+BF1dXXR1dWVdhlSRK+JSHn0I+YRZTKZtEuQo+g1ESmPRvgiIoFQ4IuIBEKBLyISCAW+iEggFPgiIoFQ4IuIBEKBLyISCAW+iEggFPgiIoFQ4IuIBEKBLyISCAW+iEggFPgiIoFQ4IuIBEKBLyISCAW+iEggFPgiIoFI9BevzOw64E6gCljt7l9JYj9DQ0O0t7fT2NhIV1cXzc3NzJw5c0y7TCbDXXfdxYoVK5gzZw4A27ZtY/Xq1TQ3N/P444/T3NwMQHt7OwsXLuSRRx7h5ptvTqJsickdd9zBrl27mD17NocOHWLv3r1UV1czPDzM0qVL2bBhA/v372fGjBkcPHiQ5cuX89hjj7F161bmz5/PwMAAK1as4Le//S1r166loaGB7u5uPvShD7Fjxw6eeOIJrrjiCrZu3cqKFSvYt28fq1evZvny5SxYsAD4/Tl47bXXcv/99486x0opdS4mKV/feNdGpQnteKNKbIRvZlXAvwLXA28FbjWztyaxr3Xr1tHf309nZyf9/f2sX7++ZLvOzk4OHDhAR0dHYVl7ezvuPmrbfH+PPPIIAGvWrEmibInJrl27AHjhhRfYu3cvAMPDwwB8//vfZ//+/QAcOHAAd6etrY2tW7cCsHPnzsI5sXbtWgC6u7uB7PnyxBNPAPDUU08V2uXPmba2tkIN+XOmvb19zDlWSqlzMUn5+sa7NipNaMcbVZJTOg3Ab9x9p7sfAh4EGuPeydDQED09Pbg7g4ODuDs9PT0MDQ2NapfJZBgcHARgcHCQgYEBtm3bVgiGkZER3J3u7m66u7tx91Hbv/LKK3GXLjF45plnyt4m/5oXy58bxUZGRkq2y28/PDzM9u3bR52D+XX5c6yUUudikorrK3VtVJrQjrccSU7p1ALPFT1/HnhH3DtZt27dmHA+cuQI69ev58Ybbyws6+zsHNWmo6Oj5ImQD/6j7d27l/3799Pa2hpT5fEYGBhgxJL9LfoDr/6OgX2HJ+exlwjlk6mtrY1FixaVPGc6OjpYtWrVmOWlzsVS7eJSfI2UujYqTWjHW44kR/hWYtmYq8LMlptZr5n17t69u+ydbNq0acxFPzIyQl9f36hlR4/gikdqowosceGKjGd4eLjkOQil/2ootXy8dnEprq/UtVFpQjveciQ5NHweeGPR84uAMX+7uvvdwN0AdXV1ZaftlVdeSXd396gLrqqqiquuumpUu5qamlEXVk1NDUNDQ2NC38zGDf05c+awYsWKcktMVGtrK7v3vZ7oPmaccQ7nn/mGSXnsO3fuTLWG6upqFi1aNOYchOw5VkqpczFJxddIqWuj0oR2vOVIcoTfAywws3lmdgpwC/Bw3DtZsmQJZqP/mJg2bRqLFy8etaypqWnU82XLlhU+kVOsqqqKqqqqMcvPO++8GKqVuJV6rU6mlpaWkucgZM+xUkqdi0kqrq/UtVFpQjveciQW+O5+GPgE8N/AVuA77r4l7v3MnDmT+vp6zIyamhrMjPr6+jEfxaqtrS2MpGpqapgzZw4LFy6kuroayAaHmdHQ0EBDQ8OYC/iss86Ku3SJwcUXX1z2NvnXvFipUXap/0xqamoK21dXV7NgwYJR52B+Xf4cK6XUuZik4vpKXRuVJrTjLUeiX7xy9/9y9ze7+5vc/e+S2s+SJUuYN28eTU1NzJs3b9z/0ZuampgxY8aoEVVzczNmNmrbfH/XX389gD6HP8ldcMEFAMyePbvwl1g+eJcuXcrpp58OwIwZMzAzWlpaeMtb3gLA/PnzC+dE/o29hoYGIHu+XH311QBcccUVhXb5c6alpaVQQ/6caW5uHnOOlVLqXExSvr5QRruhHW9UNpnepKyrq/Pe3t60yygp/wmVyTiPvXvf61x6TdPEjYEtG7KfEInaPr/NZJ3Dh8n3moicTGbW5+51Udrq1goiIoFQ4IuIBEKBLyISCAW+iEggFPgiIoFQ4IuIBEKBLyISCAW+iEggFPgiIoFQ4IuIBEKBLyISCAW+iEggFPgiIoFQ4IuIBEKBLyISiCR/07ai1NbWpl2CHEWviUh5FPgRNTY2pl2CHEWviUh5NKUjIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCDKCnwzOz2pQkREJFmRAt/MrjazXwFbc8/fbmatiVYmIiKxijrC/xrwfmAvgLtvBt6bVFEiIhK/yFM67v7cUYtGYq5FREQSFPX2yM+Z2dWAm9kpwKfITe+IiMjUEHWEfzvwcaAWyACLcs9FRGSKiDTCd/c9wLKEaxERkQRF/ZTOfDP7TzPbbWa7zKzLzOYnXZyIiMQn6pROJ/AdYDYwB/gu8EBSRYmISPyivmlr7n5/0fN2M/tEEgVJ+V57eRdbNnRGbgtEbl/Y5kz9YLjIVBc18DeY2eeABwEHPgj8wMzOBXD3lxKqTyZQW1teEO85eCoAs858Q/SNzqwtez8iMvmYu0/cyKz/GKvd3WOZz6+rq/Pe3t44uhIRCYKZ9bl7XZS2UT+lM+/EShIRkbRF/ZROr5mtMLOzky5IRESSEfVTOreQ/dJVr5k9aGbvNzNLsC4REYlZpMB399+4+xeAN5P9iOY9wLNm9jf5N25FRGRyi3zzNDO7HPhH4KvAQ8DNwBDw42RKExGROEV609bM+oCXgW8Cn3P3g7lVPzOzdydVnIiIxCfq5/A/4O47ixeY2Tx373f3GxOoS0REYhZ1SmdNxGUiIjJJHXOEb2aXAJcCZ5lZ8Uh+JjAjycJERCReE03pLASWAmcDf1q0fB/wF0kVJSIi8Ttm4Lt7F9BlZu9y940nqSYREUlA1Dn8vWb2qJn9ErIf0TSzLyZYl4iIxCxq4H8D+DzwOoC7/5zst29FRGSKiBr4p7l791HLDsddjIiIJCdq4O8xszeRvRc+ZnYz8EJiVYmISOyifvHq48DdwCVmlgH6gebEqhIRkdhFvR/+TmCxmZ0OTHP3fcmWJSIicYt6L51TgZuAucD0/J2R3f3LiVUmIiKxijql0wW8AvQBBydoKyIik1DUwL/I3a9LtBKZFLq6ushkMrH1t2fPHgBmzZoVW5/Ho7a2lsbGxlRrEElb1MB/wsze5u6/SLQaSV0mkyHzbD81Z5wSS38H9x8C4PC09P4wHHz1UGr7FplMogb+HwAfMbN+slM6Bri7X55YZZKamjNOoaXuwlj6aut9ESC2/k6kBpHQRQ386xOtQkREEjfR7ZFnuvsQ2btjiojIFDbRCL+T7O2R+8h+y9aK1jkwP6G6REQkZhPdHnlp7uH/Ao8BP3X3XydelYiIxC7qvXTuBWYD/2JmO8xsjZl9OsG6REQkZlFvrfBjM/sJUA9cA9wOXAbcmWBtIiISo6i3VngUOB3YCPwUqHf3XUkWJiIi8Yo6pfNz4BDZUf3lwGVmVp1YVSIiEruoUzqfATCzM4A/JzunfyFwanKliYhInKJO6XwCeA9wFfAMcA/ZqR0REZkion7Tthr4J6DP3fXThiIiU1DUKZ2vJl2IiIgkK+qbtiIiMsUp8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhAK/hK6uLrq6utIuQyRVug4qT9Rv2gYlk8mkXYJI6nQdVB6N8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EXkmJ566ilWrlzJ5s2b+cEPfsDKlSu59957C8vy67/97W+zcuVKnnzySTKZDF/84hfZuHEjq1atYvv27YX+tm3bVliWf/z000/T2trK0NAQQ0NDhcd5+f62b98+Zl2p9sUmWh9VvoaBgYGy2x9r27jqiyKxwDeze8xsl5n9Mql9iEjyHnzwQQA6OzvZsGEDAFu2bCksy6/v6ekBYM2aNXR2dnLgwAEeeugh3J22trZCf+3t7YVl+ccPPPAA/f39rF+/nnXr1hUe5+X7a2trG7OuVPtiE62PKl9DR0dH2e2PtW1c9UWR5Aj/PuC6BPsXkYTt27ePkZERgMK/xUZGRkouHxwcHPV8eHi4MKIfHh4uLMs/HhkZwd3p7u6mp6cHd6enp4ehoSEymUyhv+Hh4VHrhoaGxrQvNtH6qIprGBwcnHCUf3T78baNq76oEvtNW3d/zMzmJtV/kvbu3cvBgwdpbW1Nu5STbmBggDccOZx2GbH63WuHef3AQJCv54kYGBjgwIEDsfVXPMofz+HDhzEzAI4cOcL69evZsWPHmHb5de6Ou49aduONNxbarVu37pjro+rs7Bz1vKOjg1WrVkVuP962cdUXVepz+Ga23Mx6zax39+7daZcjIgkpHtEfSz4AR0ZG6OvrG/PXQvG6TZs2jfoLpK+vb1S7idZHdXQNpWqKur54XVz1RZXYCD8qd78buBugrq7OUy4HgPPOOw+AFStWpFzJydfa2srhlzJplxGrc06bzvRz5wT5ep6I1tZWdu7cGVt/1dXVABOGvpnh7lRVVXHVVVexY8eOMQGaX5efBhoZGSksK3bllVcec31UNTU1o2qoqakpq/3R6+KuL6rUR/giMnmdf/75sfXV0tJCc3PzMdtMnz6dqqoqAKZNm8bixYtpamoa0y6/bsmSJYUpoPyyYhOtj+roGpYtW1ZW+/G2jau+qBT4IjKuM888sxDA+X+LVVVVlVx+9Ai4urqaBQsWsHDhwsJIv7q6uvC4qqoKM6OhoYH6+nrMjPr6embOnEltbW2hv+rq6lHrZs6cOaZ9sYnWR1VcQ01NDXPmzCmr/XjbxlVfVEl+LPMBYCOw0MyeN7OPJrUvEUnOLbfcAmRHrddccw0Al156aWFZfn19fT0AN998M01NTcyYMYObbroJM6OlpaXQX3Nzc2FZ/vGtt97KvHnzCqP2/OO8fH8tLS1j1pVqX2yi9VHla5hodF+q/bG2jau+KCz/BslkUFdX5729vWmXUfg0R4hzvvk5/Ja6C2Ppr633RYDY+jveGqafWxvk63kiQr4OphIz63P3uihtNaUjIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+CIigVDgi4gEIvUfMZ+Mamtr0y5BJHW6DiqPAr+ExsbGtEsQSZ2ug8qjKR0RkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhH7TVsYYfPUQbb0vxtPXvkMAsfV3XDW8eojac1PbvcikocCXUWpra2Pt79QjewCYfu6sWPstR+258R+XyFSkwJdRGhsb0y5BRBKiOXwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUCYu6ddQ4GZ7QaeibHLWcCeGPtLi45j8qmUY9FxTD7lHsvF7n5+lIaTKvDjZma97l6Xdh0nSscx+VTKseg4Jp8kj0VTOiIigVDgi4gEotID/+60C4iJjmPyqZRj0XFMPokdS0XP4YuIyO9V+ghfRERyKjbwzazKzJ4ys++nXcuJMLOzzWyNmf3azLaa2bvSrul4mNlnzGyLmf3SzB4wsxlp1xSVmd1jZrvM7JdFy841s3Vmtj337zlp1hjFOMfx1dy59XMz+w8zOzvNGqModRxF61aamZtZej+xFtF4x2FmnzSzbbnr5Y4491mxgQ98GtiadhExuBP4obtfArydKXhMZlYLfAqoc/fLgCrglnSrKst9wHVHLfsc8Ki7LwAezT2f7O5j7HGsAy5z98uB/wM+f7KLOg73MfY4MLM3AkuAZ092QcfpPo46DjO7BmgELnf3S4F/iHOHFRn4ZnYR8CfA6rRrORFmNhN4L/BNAHc/5O4vp1vVcZsOVJvZdOA0YCDleiJz98eAl45a3Ah8K/f4W8CfndSijkOp43D3H7n74dzTJ4GLTnphZRrn9QD4GvBXwJR4Y3Kc4/gY8BV3P5hrsyvOfVZk4AP/TPaFP5J2ISdoPrAbuDc3PbXazE5Pu6hyuXuG7EjlWeAF4BV3/1G6VZ2wGnd/ASD37wUp1xOH24BH0i7ieJjZDUDG3TenXcsJejPwHjP7mZn9xMzq4+y84gLfzJYCu9y9L+1aYjAduBK4y92vAPYzNaYORsnNbzcC84A5wOlm1pxuVVLMzL4AHAY60q6lXGZ2GvAF4Etp1xKD6cA5wDuBVcB3zMzi6rziAh94N3CDmf0WeBB4n5m1p1vScXseeN7df5Z7vobsfwBTzWKg3913u/vrwFrg6pRrOlGDZjYbIPdvrH96n0xm9mFgKbDMp+bntN9EdjCxOXfdXwRsMrMLU63q+DwPrPWsbrKzFLG9AV1xge/un3f3i9x9Ltk3Bn/s7lNyNOnuLwLPmdnC3KJrgV+lWNLxehZ4p5mdlhutXMsUfPP5KA8DH849/jDQlWItx83MrgP+GrjB3V9Lu57j4e6/cPcL3H1u7rp/Hrgyd/1MNd8D3gdgZm8GTiHGm8JVXOBXoE8CHWb2c2AR8Pcp11O23F8oa4BNwC/InndT5puRZvYAsBFYaGbPm9lHga8AS8xsO9lPhnwlzRqjGOc4vg6cCawzs6fN7N9SLTKCcY5jyhnnOO4B5uc+qvkg8OE4/+rSN21FRAKhEb6ISCAU+CIigVDgi4gEQoEvIhIIBb6ISCAU+DKlmNmncncN7TCzG8wslm8em9mrMfQxbj35/s1sjpmtyT1eZGZ/fKL7FYlKH8uUKcXMfg1c7+79Mff7qrufEWefE/VvZh8hewfRTyS1X5FiGuHLlJH7UtB84OHc/fU/YmZfz63rMrOW3OO/NLOO3OM3mdkPzazPzH5qZpfkls8zs41m1mNmf3uMfX4vt+0WM1tetPw6M9tkZpvN7NHcsuJ6SvZvZnMt+5sApwBfBj6Y+8LTBy17b/3zc+2mmdlvpsJ93WXqmJ52ASJRufvtuVsBXOPue3Ij5LzlwONm1g98luzNpyD7jd7b3X27mb0DaCX71fU7yd6Urs3MPn6M3d7m7i+ZWTXQY2YPkR0ofQN4r7v3m9m5JbY7Zv/ufsjMvkTRCD/3n9Eysnd7XQxsdvfYvlYvohG+VAR3HyR7t8QNwGdzIX0G2Zu0fdfMngb+HZid2+TdwAO5x/cfo+tPmdlmsveKfyOwgOx/Jo/lp5XcvdS92aP2X+weoCX3+Dbg3ojbiUSiEb5UkrcBe8neghmyA5qX3X3ROO2P+QaWmf0R2ZH2u9z9NTP7H2AGYBNtG6X/MY3dnzOzQTN7H/AOsqN9kdhohC8VwcwagOuBK4CVZjbP3YeAfjP7QK6Nmdnbc5s8zu9/ZnG8YD0L+F0u7C/h99NEG4E/NLN5uX5LTelE6X8f2RuXFVsNtAPfcfeRcbYTOS4KfJnyzOxUsnPqt7n7ANk5/Htyt2JeBnw0Ny2zhewPsUD2N48/bmY9ZIO9lB8C03N3Kv1bstM6uPtusu8ZrM31++0S20bpfwPw1vybtrllDwNnoOkcSYA+likyiZhZHfA1d39P2rVI5dEcvsgkkfvS1sfQ3L0kRCN8EZFAaA5fRCQQCnwRkUAo8EVEAqHAFxEJhAJfRCQQCnwRkUD8PynD9DftuhtdAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(y=\"winetype\", x=\"fixed acidity\", data=wine, orient=\"h\", palette = \"pastel\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2 Modelling on Relationships\n",
    "Split data into training and test set. Build models that evaluate the relationship between all available X variables in the dataset and the target variable. Evaluate Logistic Regression, Penalized Logistic Regression, and KNN for classification using cross-validation. How different are the results? How does scaling the data with StandardScaler influence the results?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split Training and Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y = wine['winetype'] \n",
    "\n",
    "X = wine.loc[:, wine.columns != 'winetype']\n",
    "\n",
    "# shuffling the dataset\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "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>fixed acidity</th>\n",
       "      <th>volatile acidity</th>\n",
       "      <th>citric acid</th>\n",
       "      <th>residual sugar</th>\n",
       "      <th>chlorides</th>\n",
       "      <th>free sulfur dioxide</th>\n",
       "      <th>total sulfur dioxide</th>\n",
       "      <th>density</th>\n",
       "      <th>pH</th>\n",
       "      <th>sulphates</th>\n",
       "      <th>alcohol</th>\n",
       "      <th>quality</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1444</th>\n",
       "      <td>7.3</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.02</td>\n",
       "      <td>2.2</td>\n",
       "      <td>0.072</td>\n",
       "      <td>31.0</td>\n",
       "      <td>92.0</td>\n",
       "      <td>0.99566</td>\n",
       "      <td>3.32</td>\n",
       "      <td>0.68</td>\n",
       "      <td>11.1</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>7.5</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.11</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.079</td>\n",
       "      <td>11.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>0.99680</td>\n",
       "      <td>3.42</td>\n",
       "      <td>0.58</td>\n",
       "      <td>9.6</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>961</th>\n",
       "      <td>7.1</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.14</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.078</td>\n",
       "      <td>7.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>0.99592</td>\n",
       "      <td>3.27</td>\n",
       "      <td>0.62</td>\n",
       "      <td>9.3</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2491</th>\n",
       "      <td>6.0</td>\n",
       "      <td>0.20</td>\n",
       "      <td>0.24</td>\n",
       "      <td>5.3</td>\n",
       "      <td>0.075</td>\n",
       "      <td>49.0</td>\n",
       "      <td>201.0</td>\n",
       "      <td>0.99466</td>\n",
       "      <td>3.21</td>\n",
       "      <td>0.43</td>\n",
       "      <td>9.5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2493</th>\n",
       "      <td>7.6</td>\n",
       "      <td>0.31</td>\n",
       "      <td>0.23</td>\n",
       "      <td>12.7</td>\n",
       "      <td>0.054</td>\n",
       "      <td>20.0</td>\n",
       "      <td>139.0</td>\n",
       "      <td>0.99836</td>\n",
       "      <td>3.16</td>\n",
       "      <td>0.50</td>\n",
       "      <td>9.7</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      fixed acidity  volatile acidity  citric acid  residual sugar  chlorides  \\\n",
       "1444            7.3              0.67         0.02             2.2      0.072   \n",
       "66              7.5              0.52         0.11             1.5      0.079   \n",
       "961             7.1              0.56         0.14             1.6      0.078   \n",
       "2491            6.0              0.20         0.24             5.3      0.075   \n",
       "2493            7.6              0.31         0.23            12.7      0.054   \n",
       "\n",
       "      free sulfur dioxide  total sulfur dioxide  density    pH  sulphates  \\\n",
       "1444                 31.0                  92.0  0.99566  3.32       0.68   \n",
       "66                   11.0                  39.0  0.99680  3.42       0.58   \n",
       "961                   7.0                  18.0  0.99592  3.27       0.62   \n",
       "2491                 49.0                 201.0  0.99466  3.21       0.43   \n",
       "2493                 20.0                 139.0  0.99836  3.16       0.50   \n",
       "\n",
       "      alcohol  quality  \n",
       "1444     11.1        6  \n",
       "66        9.6        5  \n",
       "961       9.3        5  \n",
       "2491      9.5        5  \n",
       "2493      9.7        4  "
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn import preprocessing\n",
    "# Use transformed data with a model:\n",
    "scaler = preprocessing.StandardScaler()\n",
    "scaler.fit(X_train)\n",
    "X_train_scaled = scaler.transform(X_train)\n",
    "X_test_scaled = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Build Models and Evaluate"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### KNN"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Build Model**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set score: 0.960\n",
      "Test set score: 0.938\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "knn = KNeighborsClassifier(n_neighbors=5)\n",
    "knn.fit(X_train, y_train)\n",
    "\n",
    "print(\"Training set score: {:.3f}\".format(knn.score(X_train, y_train)))\n",
    "print(\"Test set score: {:.3f}\".format(knn.score(X_test, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Training with scaled data get improved result**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training with scale set score: 0.994\n",
      "Test with scale set score: 0.994\n"
     ]
    }
   ],
   "source": [
    "knn.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"Training with scale set score: {:.3f}\".format(knn.score(X_train_scaled, y_train)))\n",
    "print(\"Test with scale set score: {:.3f}\".format(knn.score(X_test_scaled, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Cross Validation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score: 0.943\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "print(\"Average cross-validation score: {:.3f}\".format(np.mean(cross_val_score(KNeighborsClassifier(), \n",
    "                                                                              X_train, y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With scaling, we get a higher predict score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score with scaling:: 0.993\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score with scaling:: {:.3f}\".format(np.mean(cross_val_score(KNeighborsClassifier(), \n",
    "                                                                              X_train_scaled, y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Logistic Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Build Model**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "logreg .coef_: [[  1.37933076  12.791699    -0.38755615  -0.1514902   17.96551218\n",
      "    0.05839286  -0.06355232 -17.26589021   7.39287084   9.34780803\n",
      "   -0.53050872   0.04826096]]\n",
      "Training set score: 0.989\n",
      "Test set score: 0.991\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "logreg = LogisticRegression(C=1e90)\n",
    "\n",
    "logreg.fit(X_train, y_train)\n",
    "\n",
    "print(\"logreg .coef_: {}\".format(logreg .coef_))\n",
    "\n",
    "print(\"Training set score: {:.3f}\".format(logreg.score(X_train, y_train)))\n",
    "print(\"Test set score: {:.3f}\".format(logreg.score(X_test, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Training with scaled data get improved result**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "With scaling logreg .coef_: [[-0.16363364  1.23106878 -0.34129951 -4.60336162  0.76912171  1.14992851\n",
      "  -2.83637555  4.99557407 -0.17447084  0.34125188  1.89156906  0.36139468]]\n",
      "With scaling Training set score: 0.994\n",
      "With scaling Test set score: 0.996\n"
     ]
    }
   ],
   "source": [
    "logreg.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"With scaling logreg .coef_: {}\".format(logreg .coef_))\n",
    "\n",
    "print(\"With scaling Training set score: {:.3f}\".format(logreg.score(X_train_scaled, y_train)))\n",
    "print(\"With scaling Test set score: {:.3f}\".format(logreg.score(X_test_scaled, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Cross Validation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score: 0.987\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score: {:.3f}\".format(np.mean(cross_val_score(LogisticRegression(C=1e90), \n",
    "                                                                              X_train, y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With scaling, we get a higher predict score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score with scaling: 0.993\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score with scaling: {:.3f}\".format(np.mean(cross_val_score(LogisticRegression(C=1e90), \n",
    "                                                                              X_train_scaled, y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Penalized Logistic Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Build Model**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "logreg .coef_: [[ 0.69351965  7.86591635 -0.54155001 -0.16787247  2.50505967  0.04909166\n",
      "  -0.06130386 -3.1711382   2.25828752  6.09757349 -0.73479254 -0.10142225]]\n",
      "Training set score: 0.981\n",
      "Test set score: 0.985\n"
     ]
    }
   ],
   "source": [
    "logreg = LogisticRegression(C=1)\n",
    "\n",
    "logreg.fit(X_train, y_train)\n",
    "\n",
    "print(\"logreg .coef_: {}\".format(logreg .coef_))\n",
    "\n",
    "print(\"Training set score: {:.3f}\".format(logreg.score(X_train, y_train)))\n",
    "print(\"Test set score: {:.3f}\".format(logreg.score(X_test, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Training with scaled data get improved result**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "With scaling logreg .coef_: [[ 0.35479618  1.22439923 -0.32048344 -3.06843166  0.8213788   0.8044157\n",
      "  -2.56715057  3.42382927  0.24012907  0.48639375  1.15616032  0.30774051]]\n",
      "With scaling Training set score: 0.993\n",
      "With scaling Test set score: 0.996\n"
     ]
    }
   ],
   "source": [
    "logreg.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"With scaling logreg .coef_: {}\".format(logreg .coef_))\n",
    "\n",
    "print(\"With scaling Training set score: {:.3f}\".format(logreg.score(X_train_scaled, y_train)))\n",
    "print(\"With scaling Test set score: {:.3f}\".format(logreg.score(X_test_scaled, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Cross Validation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score: 0.980\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score: {:.3f}\".format(np.mean(cross_val_score(LogisticRegression(C=1), \n",
    "                                                                              X_train, y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With scaling, we get a higher predict score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average cross-validation score with scaling: 0.993\n"
     ]
    }
   ],
   "source": [
    "print(\"Average cross-validation score with scaling: {:.3f}\".format(np.mean(cross_val_score(LogisticRegression(C=1), \n",
    "                                                                              X_train_scaled, y_train, cv=10))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.3 Tune the parameters where possible using GridSearchCV."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tuning the parameters improve the results."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best mean cross-validation score: 0.993\n",
      "best parameters: {'n_neighbors': 16}\n",
      "test-set score: 0.994\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "param_grid = {'n_neighbors': np.arange(1, 20, 1)} \n",
    "\n",
    "grid = GridSearchCV(KNeighborsClassifier(), param_grid=param_grid, cv=10)\n",
    "\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"best mean cross-validation score: {:.3f}\".format(grid.best_score_))\n",
    "print(\"best parameters: {}\".format(grid.best_params_))\n",
    "print(\"test-set score: {:.3f}\".format(grid.score(X_test_scaled, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Penalized Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best mean cross-validation score: 0.993\n",
      "best parameters: {'C': 11}\n",
      "test-set score: 0.996\n"
     ]
    }
   ],
   "source": [
    "param_grid = {'C': np.arange(1, 1000, 10)} \n",
    "\n",
    "grid = GridSearchCV(LogisticRegression(), param_grid=param_grid, cv=10)\n",
    "\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"best mean cross-validation score: {:.3f}\".format(grid.best_score_))\n",
    "print(\"best parameters: {}\".format(grid.best_params_))\n",
    "print(\"test-set score: {:.3f}\".format(grid.score(X_test_scaled, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.4 Change the cross-validation strategy\n",
    "\n",
    "Change the cross-validation strategy in GridSearchCV from ‘stratified k-fold’ to ‘kfold’ with shuffling. Do the parameters for models that can be tuned change? Do they change if you change the random seed of the shuffling? Or if you change the random state of the split into training and test data?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import KFold"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**The parameters for models change**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best mean cross-validation score: 0.994\n",
      "best parameters: {'n_neighbors': 1}\n",
      "test-set score: 0.996\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=5, shuffle = True, random_state = 2)\n",
    "\n",
    "param_grid = {'n_neighbors': np.arange(1, 20, 1)} \n",
    "\n",
    "grid = GridSearchCV(KNeighborsClassifier(), param_grid=param_grid, cv=kfold)\n",
    "\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"best mean cross-validation score: {:.3f}\".format(grid.best_score_))\n",
    "print(\"best parameters: {}\".format(grid.best_params_))\n",
    "print(\"test-set score: {:.3f}\".format(grid.score(X_test_scaled, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best mean cross-validation score: 0.993\n",
      "best parameters: {'C': 1}\n",
      "test-set score: 0.996\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=5, shuffle = True, random_state = 2)\n",
    "param_grid = {'C': np.arange(1, 1000, 10)} \n",
    "\n",
    "grid = GridSearchCV(LogisticRegression(), param_grid=param_grid, cv=kfold)\n",
    "\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"best mean cross-validation score: {:.3f}\".format(grid.best_score_))\n",
    "print(\"best parameters: {}\".format(grid.best_params_))\n",
    "print(\"test-set score: {:.3f}\".format(grid.score(X_test_scaled, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*If I change the random state of the split into training and test data, the parameters change*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best mean cross-validation score: 0.993\n",
      "best parameters: {'n_neighbors': 8}\n",
      "test-set score: 0.993\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=5, shuffle = True, random_state = 99)\n",
    "\n",
    "param_grid = {'n_neighbors': np.arange(1, 20, 1)} \n",
    "\n",
    "grid = GridSearchCV(KNeighborsClassifier(), param_grid=param_grid, cv=kfold)\n",
    "\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"best mean cross-validation score: {:.3f}\".format(grid.best_score_))\n",
    "print(\"best parameters: {}\".format(grid.best_params_))\n",
    "print(\"test-set score: {:.3f}\".format(grid.score(X_test_scaled, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best mean cross-validation score: 0.993\n",
      "best parameters: {'C': 11}\n",
      "test-set score: 0.996\n"
     ]
    }
   ],
   "source": [
    "kfold = KFold(n_splits=5, shuffle = True, random_state = 99)\n",
    "param_grid = {'C': np.arange(1, 1000, 10)} \n",
    "\n",
    "grid = GridSearchCV(LogisticRegression(), param_grid=param_grid, cv=kfold)\n",
    "\n",
    "grid.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"best mean cross-validation score: {:.3f}\".format(grid.best_score_))\n",
    "print(\"best parameters: {}\".format(grid.best_params_))\n",
    "print(\"test-set score: {:.3f}\".format(grid.score(X_test_scaled, y_test)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.5 Compare the coefficients \n",
    "\n",
    "Lastly, compare the coefficients for Logistic Regression and Penalized Logistic Regression and discuss which final model you would choose to predict new data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*I would use penalized logistic regression. According to the test score and cross-validation score below, penalized logistic regression model performs as well as logistic regression but with smaller coefficients, which deminish the risk of overfitting.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "logreg .coef_: [[-0.16363364  1.23106878 -0.34129951 -4.60336162  0.76912171  1.14992851\n",
      "  -2.83637555  4.99557407 -0.17447084  0.34125188  1.89156906  0.36139468]]\n",
      "With scaling Training set score: 0.994\n",
      "With scaling Test set score: 0.996\n",
      "Average cross-validation score with scaling: 0.993\n"
     ]
    }
   ],
   "source": [
    "logreg = LogisticRegression(C=1e90)\n",
    "\n",
    "logreg.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"logreg .coef_: {}\".format(logreg .coef_))\n",
    "\n",
    "print(\"With scaling Training set score: {:.3f}\".format(logreg.score(X_train_scaled, y_train)))\n",
    "print(\"With scaling Test set score: {:.3f}\".format(logreg.score(X_test_scaled, y_test)))\n",
    "\n",
    "print(\"Average cross-validation score with scaling: {:.3f}\".format(np.mean(cross_val_score(LogisticRegression(C=1e90), \n",
    "                                                                              X_train_scaled, y_train, cv=10))))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "logreg .coef_: [[-0.07340647  1.23583272 -0.33553696 -4.34794168  0.77829167  1.09446719\n",
      "  -2.793611    4.72114342 -0.10089774  0.36678909  1.76192181  0.35275525]]\n",
      "With scaling Training set score: 0.994\n",
      "With scaling Test set score: 0.996\n",
      "Average cross-validation score with scaling: 0.993\n"
     ]
    }
   ],
   "source": [
    "logreg = LogisticRegression(C=11)\n",
    "\n",
    "logreg.fit(X_train_scaled, y_train)\n",
    "\n",
    "print(\"logreg .coef_: {}\".format(logreg .coef_))\n",
    "\n",
    "print(\"With scaling Training set score: {:.3f}\".format(logreg.score(X_train_scaled, y_train)))\n",
    "print(\"With scaling Test set score: {:.3f}\".format(logreg.score(X_test_scaled, y_test)))\n",
    "\n",
    "print(\"Average cross-validation score with scaling: {:.3f}\".format(np.mean(cross_val_score(LogisticRegression(C=11), \n",
    "                                                                              X_train_scaled, y_train, cv=10))))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.8.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {
    "height": "595px",
    "width": "859px"
   },
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": true,
   "toc_position": {
    "height": "669px",
    "left": "164px",
    "top": "216px",
    "width": "165px"
   },
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}