Day_2_Answers.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "pleasant-norwegian",
   "metadata": {},
   "source": [
    "# Machine Learning Basics with Scikit-learn: Day 2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "happy-gossip",
   "metadata": {},
   "source": [
    "Today we will learn about **supervised learning.** We will cover some popular methods implemented in *scikit-learn*. If you are interested in learning more, explore the [Supervised Learning documentation](https://scikit-learn.org/stable/supervised_learning.html#supervised-learning). \n",
    "\n",
    "Supervised learning consists in learning the relationship between two datasets: the observed data `X` and an external variable `y` that we are trying to predict, usually called “target” or “labels”. Most often, `y` is a one-dimension array of length `n_samples`.\n",
    "\n",
    "To run these models easily, all supervised estimators in *scikit-learn* implement a `fit(X, y)` function to fit the model and a `predict(X)` function to predicte labels `y` given observations `X`.\n",
    "\n",
    "### Distinction between classification and regression\n",
    "Supervised learning aims to predict an outcome `y` based on information available `X`. There are two kinds of strategies: \n",
    "* If the prediction task is to classify the observations in a set of *categorical* values, in other words to “name” the objects observed, the task is said to be a **classification** task. Given the input `X` and the real values `y`, the model predicts the outcomes `y',` which can be correct or wrong. The `y` is a vector of integers (i.e., 0 or 1) or strings (e.g., `class1`,`class2`).\n",
    "* On the other hand, if the goal is to predict a *continuous* target variable, it is said to be a **regression** task. Based on the input `X`, we will obtain a numerical value `y'`, which is an approximation of the real value `y.`\n",
    "\n",
    "We will focus on classification models:\n",
    "* Nearest neighbors\n",
    "* Decision trees\n",
    "* Logistic regressions\n",
    "* Neural networks\n",
    "* Support vector machines\n",
    "\n",
    "If you're using Google Colab. Please uncomment and run the following commands. We need to use scikit-learn version 0.24.2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "union-carry",
   "metadata": {},
   "outputs": [],
   "source": [
    "# import sys\n",
    "# !{sys.executable} -m pip install scikit-learn==0.24.2\n",
    "# !{sys.executable} -m pip install scikit-learn --upgrade"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "structural-parks",
   "metadata": {},
   "source": [
    "We will start loading all the packages required to run these models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "reserved-declaration",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from matplotlib.colors import ListedColormap\n",
    "from sklearn import datasets\n",
    "from sklearn import neighbors, datasets\n",
    "from sklearn import svm\n",
    "from sklearn import tree\n",
    "\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.tree import export_text"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "blocked-courtesy",
   "metadata": {},
   "source": [
    "### Importing the dataset\n",
    "We will use the Wine dataset provided by `scikit-learn.` The data are the results of a chemical analysis of wines grown in the same region in Italy by *three* different cultivators. There are thirteen different measurements taken for different constituents found in the three types of wine:\n",
    "* Alcohol\n",
    "* Malic acid\n",
    "* Ash\n",
    "* Alcalinity of ash\n",
    "* Magnesium\n",
    "* Total phenols\n",
    "* Flavanoids\n",
    "* Nonflavanoid phenols\n",
    "* Proanthocyanins\n",
    "* Color intensity\n",
    "* Hue\n",
    "* OD280/OD315 of diluted wines\n",
    "* Proline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "complimentary-macintosh",
   "metadata": {},
   "outputs": [],
   "source": [
    "wine_dataset = datasets.load_wine()\n",
    "wine = pd.DataFrame(wine_dataset.data, columns=wine_dataset.feature_names)\n",
    "wine['target'] = pd.Series(wine_dataset.target)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "consecutive-sampling",
   "metadata": {},
   "outputs": [],
   "source": [
    "wine"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "seasonal-puppy",
   "metadata": {},
   "source": [
    "For demo purposes, we will use two dimensions of the dataset. In your exercises, you will create the models using **all** the features. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "vulnerable-angola",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = wine.iloc[:,:2]\n",
    "y = wine.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "delayed-washington",
   "metadata": {},
   "outputs": [],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "enormous-casting",
   "metadata": {},
   "source": [
    "### Datasets for your exercises\n",
    "As yesterday, we partition the dataset into training and testing datasets. We will use the 80/20 proportion."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "opposite-immune",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "comic-expense",
   "metadata": {},
   "source": [
    "We will do the same using all the features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "acquired-effects",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_all = wine.iloc[:,:-1]\n",
    "y_all = wine.target\n",
    "X_all_train, X_all_test, y_all_train, y_all_test = train_test_split(X_all, y_all, test_size = 0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fewer-ethiopia",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_all_train"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "electric-election",
   "metadata": {},
   "source": [
    "## 1. KNN (k nearest neighbors) classifier\n",
    "The simplest possible classifier is the *nearest neighbor*: given a new observation `X_test`, find in the training set the observation with the closest feature vector. The model considers *k* nearest neighbors to make a prediction.\n",
    "\n",
    "![Example of kkn](https://www.kdnuggets.com/wp-content/uploads/popular-knn-metrics-0.png)\n",
    "\n",
    "This happens for each and every test observation and that is how it finds similarities in the data. For calculating distances KNN uses a distance metric from the list of available metrics.\n",
    "\n",
    "Please see the [Nearest Neighbors section](https://scikit-learn.org/stable/modules/neighbors.html#neighbors) of the online Scikit-learn documentation for more information about this type of classifier.\n",
    "\n",
    "Let's start creating the knn model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "renewable-symposium",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn = KNeighborsClassifier()\n",
    "knn.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "blond-words",
   "metadata": {},
   "source": [
    "Now, we use this model to predict the values of the wine' class (`y_predict`) using the testing set (`X_test`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "satisfactory-riverside",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 2, 1, 0, 2, 0, 1, 0, 2, 1, 1, 0, 1, 1, 0, 2, 1, 2, 1, 1, 0, 0,\n",
       "       1, 2, 1, 1, 0, 2, 2, 1, 2, 0, 0, 0, 2, 2])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = knn.predict(X_test)\n",
    "y_predicted"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "unsigned-think",
   "metadata": {},
   "source": [
    "Let's compare how good are the predicted values with respect to the real values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "intellectual-functionality",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 2, 1, 0, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 0, 1, 1, 2, 1, 1, 2, 2,\n",
       "       1, 2, 1, 1, 2, 1, 2, 1, 0, 0, 1, 0, 2, 2])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "possible-belarus",
   "metadata": {},
   "source": [
    "We will consolidate all the results in one single table and check if the predictions are correct."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "central-raleigh",
   "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>y_real</th>\n",
       "      <th>y_predicted</th>\n",
       "      <th>is_correct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     y_real  y_predicted  is_correct\n",
       "56        0            0        True\n",
       "153       2            2        True\n",
       "127       1            1        True\n",
       "33        0            0        True\n",
       "118       1            2       False"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = pd.DataFrame({'y_real':y_test, 'y_predicted':y_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "results.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "nominated-psychiatry",
   "metadata": {},
   "source": [
    "Let's count how many are correct and how many are false"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "regional-meaning",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 27\n",
      "Wrong predictions: 9\n",
      "Percentage of correct predictions: 75.0%\n"
     ]
    }
   ],
   "source": [
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "clean-editing",
   "metadata": {},
   "source": [
    "Since we are using two dimensions, we can see visually how the model predicts.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "subsequent-conditioning",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create color maps\n",
    "h = .02  # step size in the mesh\n",
    "cmap_light = ListedColormap(['orange', 'cyan', 'cornflowerblue'])\n",
    "cmap_bold = ['darkorange', 'c', 'darkblue']\n",
    "\n",
    "# Plot the decision boundary. For that, we will assign a color to each\n",
    "# point in the mesh [x_min, x_max]x[y_min, y_max].\n",
    "x_min, x_max = X_train.iloc[:, 0].min() - 1, X_train.iloc[:, 0].max() + 1\n",
    "y_min, y_max = X_train.iloc[:, 1].min() - 1, X_train.iloc[:, 1].max() + 1\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, h),\n",
    "                     np.arange(y_min, y_max, h))\n",
    "\n",
    "# Predict the areas according to the model\n",
    "Z = knn.predict(np.c_[xx.ravel(), yy.ravel()])\n",
    "\n",
    "# Put the result into a color plot\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.contourf(xx, yy, Z, cmap=cmap_light)\n",
    "\n",
    "# Plot also the training points\n",
    "sns.scatterplot(x=X_train.iloc[:, 0], y=X_train.iloc[:, 1], hue=y_train,\n",
    "                palette=cmap_bold, alpha=1.0, edgecolor=\"black\")\n",
    "plt.xlim(xx.min(), xx.max())\n",
    "plt.ylim(yy.min(), yy.max())\n",
    "plt.title(\"3-Class classification\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "contemporary-cooling",
   "metadata": {},
   "source": [
    "### Exercise 1:\n",
    "Now, it's your turn to create the model with all the features: `X_all_train`, `y_all_train` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "cosmetic-nature",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Run a new model. Remember to use X_all_train and y_all_train\n",
    "knn = KNeighborsClassifier()\n",
    "knn.fit(X_all_train, y_all_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "developing-villa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the predicted value. Remember to use `X_all_test`\n",
    "y_all_predicted = knn.predict(X_all_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "covered-accent",
   "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>y_real</th>\n",
       "      <th>y_predicted</th>\n",
       "      <th>is_correct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>155</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>170</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>166</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     y_real  y_predicted  is_correct\n",
       "103       1            1        True\n",
       "140       2            2        True\n",
       "93        1            1        True\n",
       "8         0            0        True\n",
       "155       2            2        True\n",
       "107       1            1        True\n",
       "120       1            2       False\n",
       "73        1            0       False\n",
       "123       1            1        True\n",
       "169       2            2        True\n",
       "128       1            1        True\n",
       "52        0            0        True\n",
       "170       2            2        True\n",
       "67        1            1        True\n",
       "51        0            0        True\n",
       "127       1            1        True\n",
       "166       2            2        True\n",
       "116       1            1        True\n",
       "25        0            2       False\n",
       "66        1            1        True\n",
       "157       2            0       False\n",
       "29        0            0        True\n",
       "88        1            2       False\n",
       "16        0            0        True\n",
       "81        1            2       False\n",
       "49        0            0        True\n",
       "4         0            1       False\n",
       "14        0            0        True\n",
       "38        0            0        True\n",
       "109       1            1        True\n",
       "83        1            2       False\n",
       "168       2            0       False\n",
       "60        1            2       False\n",
       "75        1            1        True\n",
       "76        1            1        True\n",
       "35        0            0        True"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create the new result dataframe\n",
    "results = pd.DataFrame({'y_real': y_all_test, 'y_predicted': y_all_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "criminal-agreement",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 26\n",
      "Wrong predictions: 10\n",
      "Percentage of correct predictions: 72.22%\n"
     ]
    }
   ],
   "source": [
    "# Run the new results\n",
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fatal-soldier",
   "metadata": {},
   "source": [
    "Which model is doing better? The one using two variables or all the variables?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "systematic-vertex",
   "metadata": {},
   "source": [
    "## 2. Decision Trees\n",
    "Decision Trees are a non-parametric supervised learning method used for classification and regression. The goal is to create a model that predicts the value of a target variable by learning simple decision rules inferred from the data features. A tree can be seen as a piecewise constant approximation.\n",
    "\n",
    "Decision trees learn from data and set *if-then-else* decision rules. The deeper the tree, the more complex the decision rules and the fitter the model.\n",
    "\n",
    "Here is a simple example:\n",
    "\n",
    "<img src=\"https://miro.medium.com/max/2000/1*jojTznh4HOX_8cGw_04ODA.png\" alt=\"drawing\" width=\"700\"/>\n",
    "\n",
    "Some advantages of decision trees are:\n",
    "* Simple to understand and to interpret. Trees can be visualised.\n",
    "* Requires little data preparation. \n",
    "* Other techniques often require data normalisation, dummy variables need to be created and blank values to be removed. Note however that this module does not support missing values.\n",
    "* Able to handle multi-output problems.\n",
    "* Uses a white box model. If a given situation is observable in a model, the explanation for the condition is easily explained by boolean logic. By contrast, in a black box model (e.g., in an artificial neural network), results may be more difficult to interpret.\n",
    "* Possible to validate a model using statistical tests. That makes it possible to account for the reliability of the model.\n",
    "* Performs well even if its assumptions are somewhat violated by the true model from which the data were generated.\n",
    "\n",
    "The disadvantages of decision trees include:\n",
    "* Decision-tree learners can create over-complex trees that do not generalise the data well. This is called *overfitting*.\n",
    "* Decision trees can be unstable because small variations in the data might result in a completely different tree being generated. \n",
    "* There are concepts that are hard to learn because decision trees do not express them easily, such as XOR, parity or multiplexer problems.\n",
    "* Decision tree learners create biased trees if some classes dominate. It is therefore recommended to balance the dataset prior to fitting with the decision tree.\n",
    "\n",
    "For more information, check the [documentation](https://scikit-learn.org/stable/modules/tree.html).\n",
    "\n",
    "\n",
    "### Creating a decision-tree\n",
    "Let's create a decision tree classifier:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "sacred-acquisition",
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = tree.DecisionTreeClassifier()\n",
    "clf = clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "prescribed-amazon",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 25\n",
      "Wrong predictions: 11\n",
      "Percentage of correct predictions: 69.44%\n"
     ]
    }
   ],
   "source": [
    "y_predicted = clf.predict(X_test)\n",
    "results = pd.DataFrame({'y_real': y_test, 'y_predicted': y_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "\n",
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "further-venue",
   "metadata": {},
   "source": [
    "We got a high accurate model using a decision tree. We can visualize the rules created by the decision tree:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "controlling-extreme",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "|--- alcohol <= 12.75\n",
      "|   |--- malic_acid <= 2.93\n",
      "|   |   |--- alcohol <= 12.49\n",
      "|   |   |   |--- class: 1\n",
      "|   |   |--- alcohol >  12.49\n",
      "|   |   |   |--- malic_acid <= 1.31\n",
      "|   |   |   |   |--- class: 2\n",
      "|   |   |   |--- malic_acid >  1.31\n",
      "|   |   |   |   |--- malic_acid <= 2.45\n",
      "|   |   |   |   |   |--- class: 1\n",
      "|   |   |   |   |--- malic_acid >  2.45\n",
      "|   |   |   |   |   |--- class: 2\n",
      "|   |--- malic_acid >  2.93\n",
      "|   |   |--- alcohol <= 12.10\n",
      "|   |   |   |--- class: 1\n",
      "|   |   |--- alcohol >  12.10\n",
      "|   |   |   |--- alcohol <= 12.27\n",
      "|   |   |   |   |--- class: 2\n",
      "|   |   |   |--- alcohol >  12.27\n",
      "|   |   |   |   |--- alcohol <= 12.32\n",
      "|   |   |   |   |   |--- class: 1\n",
      "|   |   |   |   |--- alcohol >  12.32\n",
      "|   |   |   |   |   |--- malic_acid <= 3.85\n",
      "|   |   |   |   |   |   |--- class: 2\n",
      "|   |   |   |   |   |--- malic_acid >  3.85\n",
      "|   |   |   |   |   |   |--- malic_acid <= 4.97\n",
      "|   |   |   |   |   |   |   |--- class: 1\n",
      "|   |   |   |   |   |   |--- malic_acid >  4.97\n",
      "|   |   |   |   |   |   |   |--- class: 2\n",
      "|--- alcohol >  12.75\n",
      "|   |--- malic_acid <= 2.38\n",
      "|   |   |--- malic_acid <= 1.30\n",
      "|   |   |   |--- class: 1\n",
      "|   |   |--- malic_acid >  1.30\n",
      "|   |   |   |--- malic_acid <= 1.69\n",
      "|   |   |   |   |--- malic_acid <= 1.65\n",
      "|   |   |   |   |   |--- malic_acid <= 1.39\n",
      "|   |   |   |   |   |   |--- alcohol <= 13.36\n",
      "|   |   |   |   |   |   |   |--- class: 2\n",
      "|   |   |   |   |   |   |--- alcohol >  13.36\n",
      "|   |   |   |   |   |   |   |--- class: 0\n",
      "|   |   |   |   |   |--- malic_acid >  1.39\n",
      "|   |   |   |   |   |   |--- malic_acid <= 1.55\n",
      "|   |   |   |   |   |   |   |--- malic_acid <= 1.50\n",
      "|   |   |   |   |   |   |   |   |--- class: 0\n",
      "|   |   |   |   |   |   |   |--- malic_acid >  1.50\n",
      "|   |   |   |   |   |   |   |   |--- class: 1\n",
      "|   |   |   |   |   |   |--- malic_acid >  1.55\n",
      "|   |   |   |   |   |   |   |--- class: 0\n",
      "|   |   |   |   |--- malic_acid >  1.65\n",
      "|   |   |   |   |   |--- alcohol <= 13.53\n",
      "|   |   |   |   |   |   |--- alcohol <= 13.23\n",
      "|   |   |   |   |   |   |   |--- class: 1\n",
      "|   |   |   |   |   |   |--- alcohol >  13.23\n",
      "|   |   |   |   |   |   |   |--- malic_acid <= 1.66\n",
      "|   |   |   |   |   |   |   |   |--- class: 1\n",
      "|   |   |   |   |   |   |   |--- malic_acid >  1.66\n",
      "|   |   |   |   |   |   |   |   |--- class: 2\n",
      "|   |   |   |   |   |--- alcohol >  13.53\n",
      "|   |   |   |   |   |   |--- malic_acid <= 1.67\n",
      "|   |   |   |   |   |   |   |--- class: 0\n",
      "|   |   |   |   |   |   |--- malic_acid >  1.67\n",
      "|   |   |   |   |   |   |   |--- class: 2\n",
      "|   |   |   |--- malic_acid >  1.69\n",
      "|   |   |   |   |--- class: 0\n",
      "|   |--- malic_acid >  2.38\n",
      "|   |   |--- alcohol <= 14.19\n",
      "|   |   |   |--- malic_acid <= 5.73\n",
      "|   |   |   |   |--- malic_acid <= 3.87\n",
      "|   |   |   |   |   |--- malic_acid <= 3.75\n",
      "|   |   |   |   |   |   |--- alcohol <= 13.20\n",
      "|   |   |   |   |   |   |   |--- class: 2\n",
      "|   |   |   |   |   |   |--- alcohol >  13.20\n",
      "|   |   |   |   |   |   |   |--- alcohol <= 13.28\n",
      "|   |   |   |   |   |   |   |   |--- class: 0\n",
      "|   |   |   |   |   |   |   |--- alcohol >  13.28\n",
      "|   |   |   |   |   |   |   |   |--- alcohol <= 13.66\n",
      "|   |   |   |   |   |   |   |   |   |--- alcohol <= 13.55\n",
      "|   |   |   |   |   |   |   |   |   |   |--- class: 2\n",
      "|   |   |   |   |   |   |   |   |   |--- alcohol >  13.55\n",
      "|   |   |   |   |   |   |   |   |   |   |--- class: 0\n",
      "|   |   |   |   |   |   |   |   |--- alcohol >  13.66\n",
      "|   |   |   |   |   |   |   |   |   |--- class: 2\n",
      "|   |   |   |   |   |--- malic_acid >  3.75\n",
      "|   |   |   |   |   |   |--- class: 0\n",
      "|   |   |   |   |--- malic_acid >  3.87\n",
      "|   |   |   |   |   |--- class: 2\n",
      "|   |   |   |--- malic_acid >  5.73\n",
      "|   |   |   |   |--- class: 1\n",
      "|   |   |--- alcohol >  14.19\n",
      "|   |   |   |--- class: 0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "r = export_text(clf, feature_names=wine.columns[:2].tolist())\n",
    "print(r)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "endless-kansas",
   "metadata": {},
   "source": [
    "We can visualize the model as a decision tree. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "wrapped-vision",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(1, figsize=(10, 10))\n",
    "tree.plot_tree(clf)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "heavy-given",
   "metadata": {},
   "source": [
    "And also, we can see these regions in a 2D plane."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "helpful-somerset",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Z = clf.predict(np.c_[xx.ravel(), yy.ravel()])\n",
    "\n",
    "# Put the result into a color plot\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.figure(1, figsize=(8, 6))\n",
    "plt.pcolormesh(xx, yy, Z, shading='auto', cmap=plt.cm.Paired)\n",
    "\n",
    "# Plot also the training points\n",
    "plt.scatter(X_train.iloc[:, 0], X_train.iloc[:, 1], c=y_train, edgecolors='k', cmap=plt.cm.Paired)\n",
    "plt.xlabel(wine.columns[0])\n",
    "plt.ylabel(wine.columns[1])\n",
    "\n",
    "plt.xlim(xx.min(), xx.max())\n",
    "plt.ylim(yy.min(), yy.max())\n",
    "plt.xticks(())\n",
    "plt.yticks(())\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "medium-proxy",
   "metadata": {},
   "source": [
    "### Exercise 2:\n",
    "Now, it's your turn to create a decision tree using all the features available. How well does this model perfom? Is it better or worse?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "treated-gathering",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create your model here with X_all_train and y_all_train. Don't forget to run the fit function:\n",
    "clf = tree.DecisionTreeClassifier()\n",
    "clf = clf.fit(X_all_train, y_all_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "valuable-organizer",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the predicted value. Don't forget to use `X_all_test`\n",
    "y_all_predicted = clf.predict(X_all_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "lonely-nancy",
   "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>y_real</th>\n",
       "      <th>y_predicted</th>\n",
       "      <th>is_correct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>155</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>170</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>166</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     y_real  y_predicted  is_correct\n",
       "103       1            1        True\n",
       "140       2            2        True\n",
       "93        1            1        True\n",
       "8         0            0        True\n",
       "155       2            2        True\n",
       "107       1            1        True\n",
       "120       1            1        True\n",
       "73        1            1        True\n",
       "123       1            1        True\n",
       "169       2            2        True\n",
       "128       1            1        True\n",
       "52        0            0        True\n",
       "170       2            2        True\n",
       "67        1            1        True\n",
       "51        0            0        True\n",
       "127       1            1        True\n",
       "166       2            2        True\n",
       "116       1            1        True\n",
       "25        0            1       False\n",
       "66        1            1        True\n",
       "157       2            2        True\n",
       "29        0            0        True\n",
       "88        1            1        True\n",
       "16        0            0        True\n",
       "81        1            0       False\n",
       "49        0            0        True\n",
       "4         0            0        True\n",
       "14        0            0        True\n",
       "38        0            1       False\n",
       "109       1            1        True\n",
       "83        1            2       False\n",
       "168       2            2        True\n",
       "60        1            1        True\n",
       "75        1            1        True\n",
       "76        1            1        True\n",
       "35        0            0        True"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create the new result dataframe\n",
    "results = pd.DataFrame({'y_real': y_all_test, 'y_predicted': y_all_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "bigger-store",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Run the new results\n",
    "#print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "#print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "#print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "frank-bachelor",
   "metadata": {},
   "source": [
    "## 3. Logistic regressions\n",
    "\n",
    "Logistic regression is a statistical model that in its basic form uses a logistic function to model a binary dependent variable, although many more complex extensions exist. The logistic regression (or logit regression) estimates the parameters of a *logistic model* (a form of binary regression). \n",
    "\n",
    "<img src=\"https://scikit-learn.org/stable/_images/sphx_glr_plot_logistic_001.png\" alt=\"drawing\" width=\"400\"/>\n",
    "\n",
    "Mathematically, a binary logistic model has a dependent variable with two possible values, such as pass/fail which is represented by an indicator variable, where the two values are labeled \"0\" and \"1\".\n",
    "\n",
    "We will create a logistic regression with the current data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "experienced-sterling",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=100000.0)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log = LogisticRegression(C=1e5)\n",
    "log.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "earlier-representation",
   "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>y_real</th>\n",
       "      <th>y_predicted</th>\n",
       "      <th>is_correct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     y_real  y_predicted  is_correct\n",
       "56        0            0        True\n",
       "153       2            2        True\n",
       "127       1            1        True\n",
       "33        0            0        True\n",
       "118       1            2       False"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = log.predict(X_test)\n",
    "results = pd.DataFrame({'y_real':y_test, 'y_predicted':y_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "results.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "delayed-ethics",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 26\n",
      "Wrong predictions: 10\n",
      "Percentage of correct predictions: 72.22%\n"
     ]
    }
   ],
   "source": [
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "afraid-acquisition",
   "metadata": {},
   "source": [
    "We will plot the classification results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "fallen-electricity",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Z = log.predict(np.c_[xx.ravel(), yy.ravel()])\n",
    "\n",
    "# Put the result into a color plot\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.figure(1, figsize=(8, 6))\n",
    "plt.pcolormesh(xx, yy, Z, shading='auto', cmap=plt.cm.Paired)\n",
    "\n",
    "# Plot also the training points\n",
    "plt.scatter(X_train.iloc[:, 0], X_train.iloc[:, 1], c=y_train, edgecolors='k', cmap=plt.cm.Paired)\n",
    "plt.xlabel(wine.columns[0])\n",
    "plt.ylabel(wine.columns[1])\n",
    "\n",
    "plt.xlim(xx.min(), xx.max())\n",
    "plt.ylim(yy.min(), yy.max())\n",
    "plt.xticks(())\n",
    "plt.yticks(())\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "important-score",
   "metadata": {},
   "source": [
    "### Exercise 3:\n",
    "Now, it's your turn to create a linear regression mdoel using all the features available. How well does this model perfom? Is it better or worse?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "associate-check",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py:763: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=100000.0)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create your model here. Don't forget to run the fit function:\n",
    "log = LogisticRegression(C=1e5)\n",
    "log.fit(X_all_train, y_all_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "optical-breakdown",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the predicted value. Remember to use `X_all_test`\n",
    "y_all_predicted = log.predict(X_all_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "static-preserve",
   "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>y_real</th>\n",
       "      <th>y_predicted</th>\n",
       "      <th>is_correct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>155</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>170</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>166</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     y_real  y_predicted  is_correct\n",
       "103       1            1        True\n",
       "140       2            2        True\n",
       "93        1            1        True\n",
       "8         0            0        True\n",
       "155       2            2        True\n",
       "107       1            1        True\n",
       "120       1            1        True\n",
       "73        1            1        True\n",
       "123       1            1        True\n",
       "169       2            2        True\n",
       "128       1            1        True\n",
       "52        0            0        True\n",
       "170       2            2        True\n",
       "67        1            1        True\n",
       "51        0            0        True\n",
       "127       1            1        True\n",
       "166       2            2        True\n",
       "116       1            1        True\n",
       "25        0            1       False\n",
       "66        1            0       False\n",
       "157       2            2        True\n",
       "29        0            0        True\n",
       "88        1            1        True\n",
       "16        0            0        True\n",
       "81        1            0       False\n",
       "49        0            0        True\n",
       "4         0            1       False\n",
       "14        0            0        True\n",
       "38        0            0        True\n",
       "109       1            1        True\n",
       "83        1            2       False\n",
       "168       2            2        True\n",
       "60        1            1        True\n",
       "75        1            1        True\n",
       "76        1            1        True\n",
       "35        0            0        True"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create the new result dataframe\n",
    "results = pd.DataFrame({'y_real': y_all_test, 'y_predicted': y_all_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "seeing-maintenance",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 31\n",
      "Wrong predictions: 5\n",
      "Percentage of correct predictions: 86.11%\n"
     ]
    }
   ],
   "source": [
    "# Run the new results\n",
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "blessed-salem",
   "metadata": {},
   "source": [
    "# 4. Support vector machines (SVMs)\n",
    "\n",
    "Support Vector Machines belong to the discriminant model family. These models try to find a combination of samples to build a *threshold* maximizing the margin between two classes. After using the training examples to create this threshold, SVM maps new examples into that same space and predicts their category to belong based on which side of the gap they fall.\n",
    "\n",
    "The main features of SVMs are:\n",
    "* allowing to make a certain number of mistakes during the training process.\n",
    "* keeping margin as wide as possible so that other points can still be classified correctly.\n",
    "\n",
    "This process is known as *Soft Margin Formulation.*\n",
    "\n",
    "<img src=\"https://miro.medium.com/max/1400/1*M_3iYollNTlz0PVn5udCBQ.png\" alt=\"drawing\" width=\"500\"/>\n",
    "\n",
    "This method is highly effective to solve classification problems in non-lineable separable spaces. SVMs performs a non-linear classification using what is called the **kernel trick**, implicitly mapping their inputs into high-dimensional feature spaces.\n",
    "\n",
    "<img src=\"https://miro.medium.com/max/1400/1*mCwnu5kXot6buL7jeIafqQ.png\" alt=\"drawing\" width=\"600\"/>\n",
    "\n",
    "If you are interested in learning more, check out this [link](https://towardsdatascience.com/support-vector-machines-soft-margin-formulation-and-kernel-trick-4c9729dc8efe).\n",
    "\n",
    "\n",
    "## Creating the threshold.\n",
    "There are three main kernels used to identify these thresholds, which will separate the classes using SVM:\n",
    "* Linear: A lineal formulation will be used to separate the classes\n",
    "* Polinomial: A polinomial function (i.e., x^2, x^3, x^4, ...) will be used to separate the classes.\n",
    "* Radial Basis Function (RBF): It uses normal curves around the data points, and sums these so that the decision boundary can be defined by a type of topology condition such as curves where the sum is above a value of 0.5. \n",
    "\n",
    "We will start running the **SVC lineal kernel** in the next line:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "clinical-context",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(kernel='linear')"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc = svm.SVC(kernel='linear')\n",
    "svc.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "lovely-relaxation",
   "metadata": {},
   "source": [
    "Let's check the predicted values from the *testing* dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "musical-balance",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 2, 1, 0, 2, 0, 1, 0, 2, 1, 1, 0, 1, 1, 0, 2, 1, 2, 1, 1, 0, 0,\n",
       "       0, 2, 1, 1, 1, 2, 2, 1, 2, 0, 0, 0, 2, 0])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = svc.predict(X_test)\n",
    "y_predicted"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "radical-atlantic",
   "metadata": {},
   "source": [
    "We are going to consolidate the predicted results with the real results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "proof-header",
   "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>y_real</th>\n",
       "      <th>y_predicted</th>\n",
       "      <th>is_correct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     y_real  y_predicted  is_correct\n",
       "56        0            0        True\n",
       "153       2            2        True\n",
       "127       1            1        True\n",
       "33        0            0        True\n",
       "118       1            2       False"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = pd.DataFrame({'y_real':y_test, 'y_predicted':y_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "results.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "medical-coverage",
   "metadata": {},
   "source": [
    "Checking the results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "domestic-holmes",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 25\n",
      "Wrong predictions: 11\n",
      "Percentage of correct predictions: 69.44%\n"
     ]
    }
   ],
   "source": [
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "negative-alabama",
   "metadata": {},
   "source": [
    "Let's plot the classification model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "celtic-tyler",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdAAAAFYCAYAAADjvq21AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAACYMklEQVR4nOzdZ3iTZRfA8X9223TvRVsos+y99xSQLYigoiLiQAFxoKgoDgQHijhQFBEQXOy9996zQPee6Uyzn/dDtb4VZbZNW+7fh/e63jzNk5NacnKvc2SSJCEIgiAIwu2R2zsAQRAEQaiKRAIVBEEQhDsgEqggCIIg3AGRQAVBEAThDogEKgiCIAh3QCRQQRAEQbgDytv5YVeNUvJ1VpVXLIIgCIJQqURlGzIlSfL5t2u3lUB9nVV80jesTIISBEEQhMpu8M+X4/7rmpjCFQRBEIQ7IBKoIAiCINwBkUAFQRAE4Q6IBCoIgiAId0AkUEEQBEG4AyKBCoIgCMIdEAlUEARBEO6ASKCCIAiCcAdEAhUEQRCEOyASqCAIgiDcAZFABUEQBOEO3FYCdfd2ZeCEvuUViyAIgiBUGWIEKgiCIAh34I4SqBiFCoIgCPe6Ox6BDpzQVyRSQRAE4Z4lpnAFQRAE4Q7cdQIVo1BBEAThXlQmI1AxnSsIgiDca8QUriAIgiDcAWVZ3uz/R6HrF24py1sLgiAIQqUiRqCCIAiCcAfKLYGKNVFBEAShOivXEajYXCQIgiBUV2IKVxAEQRDuQIUkUDEKFQRBEKqbChuBiulcQRAEoToRU7iCIAiCcAcqPIGKUaggCIJQHdhlBCqmcwVBEISqTkzhCoIgCMIdsGsCFSNRQRAEoaqqFCNQkUQFQRCEqqZSJFAQo1FBEAShaqk0CfQvIokKgiAIVUGlS6CCIAiCUBVUygQqpnMFQRCEyq5SJtC/iCQqCIIgVFaVOoEKgiAIQmVV6ROomM4VBEEQKqNKn0D/IpKoIFzPapM4kpjPV8dSmX8khY1XdejNVnuHJQj3hCqTQAVBKC0pz8RzG2P4/VI2no4K/LUqzqUW8uTaKA4l5Ns7PEGo9pT2DuB2/DUKXb9wi50jEQT70putzNwdT7/aHlzIMrIhKg9HtRK1HB5u4sNXx1Jxd1DQwMfJ3qEKQrVVJUegYjpXuNftjMmltocDB5MK6d+jPimrJpD4x5MseKk3Ky5lM6ieB79fyrZ3mIJQrVXJBApic5FwbzsQn09db0dUDmreGtcOjVqJTCajf7uaPNI3Ar1F4mxqIQaLzd6hCkK1VWUT6F9EEhXuRYUmG1abRP1QD2QyWalrDWp6k2+RcFDKMZhFAhWE8lLlE6gg3Iv8XVTIZLDnTBIFRaaSxyVJYvWeK/ho5NgkCWeNwo5RCkL1Vi0SqJjOFe41fcLd2ReXT5tALb0m/8amI7EcvpjCE7O3ciUmk8wiC91quqGUy25+M0EQ7kiV2oV7M2KXrnCvaBGgZVtUDpn5Rhr4KJn++U4MFhuNvDU09XHgYEIBc3qH2jvM22a1ScTmGFHIIdRNc930tCBUFJPVxu7YvBv+TLVKoIJwr5DLZEzrEMTycxmsvpxFDTcNzgo5267l0MjXiQ96heDuWLX+eZ9JLWT+kRQ0Sjlmq4RGKePF9oGEeTjYOzThHmK02PjqWCoXM/SkFVpu+LMySZJu+cat6vlJR78ZfbfxVQgxChXuFQaLjcuZRVhtEmHuGrycVPYO6bZl6c28sCmWaR0DaeavRZIkdsXmsfRsBl8PrIVaUS1Wm4RKzmy18fiaKPKMVty9fXn6nU/54JkxJyRJavVvP1+1vqLeBjGdK9wrHJRymvlr7R3GXdkdm0eHEJeS9yGTyehR043dMbkcSyqgY4irnSMUqjubJPHzuUzyjFZcPbz4bP1BVGrNDZ9TbROoIAhVR67Bgp/2+pGzn7OKXKOo7SuUr5wiM5M3x6IzWJErlDzy8js3TZ5QTXbh3ojYnSsIlV9DXycOJuRjtf29pGSw2DieXEhDUY5QKCdGi5XdMbklyTOiVQe+3HqCDn0H3dLz74kRqJjOFYTKrVWgMxuv6pi1J5H+dd0xWSVWX8qmVaCWUPebjwQE4XatPJ/JyvOZWCWo1bApL05/j1oRTW/rHvdEAhWEOyVJEgcS8tkdm4fRYqNVoDP9arujUVb7yZsKpZDLmNElmK1RuayP1KGUyxhQ14OuYWLtUyh7V7OKWH4uE5lczvCnJjN0/AvI5bf/b/qeSqD/P50rRqPCrfj+VDrn0vQMbeCJk0rBlqgcDifm8073EFQKcUaxLKkUcgbU9WBAXQ97hyJUY/vj8lhwLBWA7kNGM3zClDu+1z2VQAXhdqQWmNgdm8fXA2uhVReXxGsZqGXGjngOJeTTRYyOBKHKsNok1kZms/hMJkgS7XoP5OEX37qre96zCXTghL5iFCrc0MWMIpr5a0uSJxQXMOgY4sr5dL1IoIJQBZitEsvOZnAiuYD4PBO1G7dg/IzZhNRpcNf3vmcTKIjNRcKNuWsUpBWar3s8vdCMm4Mo0i4Ild2B+Dy+OJqK3mzDxd2TKR/NplX3fmVWIlLshBCE/9DUX0uuwcLmazpsf1bsupihZ2dMLj1qutk5OkEQbuSDfYnMOZCM3myjXe+BfL7hMK173Fem9ZXv6RHoX8R0rvBvFHIZb3QN5qMDyfx+MQtHpYJ8k5Xn2wYQ4KK2d3iCIPyHXIOFw4kFyGRyPlixuUyma/+NSKB/EtO5wr8JdtXwab8wEvJMGC02ano4iBZhglBJWW3FxTe+PVG8y7ZDv8HlljxBJFBBuCmZTEaImzjMLwiVlSRJrL+i46czGRitEsHhdXnj1Xdp0LJ9ub6uWAP9B1H6TxAEoeqIyi5iwroovjuZjkmS0f/hCby/fHO5J08QI9B/JaZzBUEQKr9Ck5WXtsZhlaBmRBOmfrQQL/+gCnt9kUAFQRCEKscmSWyNysEqQVCtury3dEOFxyAS6A2IkaggCELl8ldFoYMJ+VzJMhAcXo/n3v/CLrGIBHoLxDEXQRAE+zuVUsAnh1LIM1pxcnFl4tvv02nA8DsqBF8WRAK9RWI0KgiCYB9Wm8TbuxM4k6YHoEn7rjz3wRc4u7rbNS6RQG+TGI0KVY0kSWy6lsO6yGwyCi3U8XJgdCNvmvhr7R2aINySo0kFnEnT4+DkzKsLfqJu01b2DgkQCVQQqr1Vl7LZG5fH5HaBhLprOJ5cwEcHk5neOYgGPk72Dk8QbuhsaiE/nUkHYMgTz1Wa5Akigd4RMZ0rVBVmq41Vl7OZ0zu0pPxgpxBX9GYbf1zK5nWRQIVKaldMDotOppNvsuETFMKL78ykZdfe9g6rFFFI4S6IogtCZZdjsKKSy66r3dvQx4n4XKOdohKE/5aUZ+K5DdHMO5xKvlmi54iHmfvr9kqXPEGMQAWhWnN3UGCy2kgrMOHn/HcSvZypJ9hVFMQXKpfLmXpe3RaPBATVrM2Uj78jMCzc3mH9JzECvUsDJ/QVI1Gh0lIp5Nxfz5O5B5OJ1Rmw2iSOJRXw05kMhjbwsnd4glBCkiT2BfZHAroNeZC5v++q1MkTxAi0zIjduUJl9UBDLxyUct7Zm0i23kItTweeaxtAI9/qs/5pthb3a1UpRKecqkaSJPbE5bEzOpczaV/jGxzKiIlT7R3WLREJVBCqOblMxuD6ngyu74lNkpCXYUNhe8vSm/nuZDrHkwuQJGgRqGV8Cz98tSp7hybcgsjMIj7cn0RWkQWNoxOPvjyL3g88jFyhsHdot0Qk0DIkducKlV11Sp5mq8QbOxPoGOLCpLa1kctkrI3M5o2d8Xx+X000SrFCVVnpzVY+PpjM8eRCAOo0acnkjxbi4e1r58huj/gLKwdiTVQQyt/RpHw8HJWMaeKDk0qBg1LOyIbeBLqoOZSQb+/whBtYcjqD48mFOLt58PL8Jby9eHWVS54gRqDlRoxGBaF8pRSYqePpcN3jdbwcSC4w2SEi4VZcySrivMwfyOGVL5YQ3rCZvUO6YyKBljOxuUgQ/lZktvH7pSwOxucjAR1quDA8whMn1e2veYW5aVhxPhNJkpD9OTUtSRLn0vQMrOtRxpELd+tYUj5fH08jU2/B3duXZ2bNq9LJE0QCFQShgtgkiXf2JODlqGRah0CQFZcZfHt3Iu/3DEEhv7312eYBWlZcyOSLo6mMiPBC/uf9isw22ga7lNO7EG5XZqGZDw8kcSXLAEDngSMY98osHLXOdo7s7okEWgHEdK4gwKmUQgwWG1M7BJZsZprSPoCXtsZxKqWQVkG394GqkMuY2a0GP5/LZPr2uJIR7Ts9QlDeZjIWysfOmFzmH0nBJoG3fxCTP/qGWhFN7R1WmREJtAKJRCrcy6J0Bpr7a0vtBJbLZDTz1xKlM9x2AgVwVit4sqUfT7b0K8tQhTLyc6IKmwQjJr7I0CdfKJlqry7ELlxBECqEr1ZFTM719Xdjcwz4iHOb1YYkSZxKKeSDfYmkJ8YTUqcB9497utolTxAJ1C7EMRfhXtQ+2IXEPCNrLmdjttowW22svZxNXI6RjjXEmmV1EJ9j4JkN0czcncCJdDOjnnuFWT+tQ6XW2Du0ciGmcO1ETOcK9xqNUs473UP46lgqy89lAlDXy4G3u4eIogdVnNFiY/7RFPbFFZ+/rVGnPlM/+g6/GqF2jqx8iQQqCEKFCXBR806PEApMVqB4DVOo+l7cEktCngkHJ2fGz/iADv2G2DukCiG+9tmZmM4V7kXOaoVIntVEXI6RQgcvlCo13+w8fc8kTxAj0EpBTOcKglDVXEgv5JvjacTnmtC6uTPxnU+q7VrnfxEJVKhWdEUWEvKM+GlVpRpIC4JQNvKNFuYcSOZsmh6Adr3v57Hp7+Hifu9VfxIJtBL5/+lcMRq9PVabxKJT6eyOySXMXUNCnonGfk680DZAbFARhDKy5nI2P57Lxmqx4OrpxXPvf0GjNp3sHZbdiAQqVAvrIrOJ0RlYOCgcZ7UCo8XGZ0dS+PF0OhNa+ds7PEGo8gwWGz9dyMVmtTJk/POMeGpqlenbWV7EV/NKSmwuuj1bo3IZ18y3ZGOKRilnfAs/dsXmYbVJdo5OEKq26GwDXx9LxWwy0qJrb0Y+89I9nzxBjEArNbG56NYVmKx4O5X+c3bTKLDYJMw26bYLlQuCAGkFJmbvTyJaZ0SuUDLosWcZMv55e4dVaYgEKlQLjfyc2BuXx9AGXiWPHU7MJ9Rdg0MVWgPNKbKQZ7IS4KxGpRBJX7APs1Xiu5NpbLmWgwT4Bocw5aNvCa0bYe/QKhWRQKsA0VP05h5q7M3rO+LJLrLQ1F/LtSwDG67qeLVTkL1DuyUFJitfHE3lXFohrholerOVR5r60LOWu71DKyFJEocTC9gXl4fFJtEmyJluNd1E55Nqpshs46l1UeQarag0GsZOfZNeIx6ulrVs75ZIoFWEmM69sWBXDR/1CWPjVR3rI3UEuKh4v2cINdyqxrm0z4+k4O6g5PvBtdEo5cTqDLyzNxEfrYomflp7hwfAolPpnE3VM6i+B2qFnM1XdRxOzOe1LsGlOqwIVVt8rgGjwoGAUD/eW74JB0cne4dUaYkEKlQbPloVjzbztXcYty1Tb+ZiehHfDwlHrSiebg7zcGBUQ282X82pFAk0Mc/I3tg8vhpYC+2fG7U61HBh2pZYTibffi9PofKJzjbw3ck0IjOLUDpqGfviWyJ53kTVWRwSALE7tzrSFVnwdlKWJM+/BLmqyS6y2Cmq0s6l6Wkd5FySPAGUchmdQl05k1Zox8iEu6U3WZm9L5EpW2K5kFFE0y69+XjVbpp36mHv0Co9MQKtgsR0bvUS4qYhU28mJd9EgMvf1ZMOJeTTwMfRjpH9zUWjIEt/fTLP1ptxdxQfI1WRJElsuKJj8ekMzDYJJ2dXJrz1EW163mfv0KoMMQIVBDvTKOU82NibmbsT2B2bS2RmEd+fTONwYj4D61aO8mitA52JzTFwID4PSSo+V3s5s4i9cfl0D3Ozc3TCnbiSZWDRqXQsEvQZ9Shf7zgtkudtEl8dqzAxEr17kiQRrTOSWmBCq1bQ0MfJLsdHBtb1xE+rZtM1HTkGKxE+jszpHYqXk6rCY/k3GqWcGV1rMPdAEivOZ6JWyMnUm3mhXQA+2soRo3Dr0gtNbPPohk36kdEvTOf+R5+2d0hVkuyvb5O3olU9P+noN6PLMRzhTokkevsupOv57mQaerONMHcNuiIL6YVmhkd4MbCuh9i2/y9sksTVLAMWm0RdL0dxVrWK0RWZmXMgmUsZRUgyGX1HjWPUpFfFZqEbeKhFjROSJLX6t2tiBFpNiNHo7bmYoefD/UkMj/AkOs9MVLYRb62a4aGubI/OpcBkZXRjH3uHWenIZTLqeVeOdVnh1lltEsvPZfBHZHEtWw8fP56f8zX1mv5rXhBukVgDrWbELt1bs+hkOsMjPFl1JYd+PRuyed4Ipo/vxPZEPd3C3FgXqSNLb7Z3mIJw11LzTTy2+hq/XcxGJpMx8pmXmL/pqEieZUCMQIV7TozOQJ7RSmyemZdGt+al0cUfJPVDPKkX4kHP53+lY4gLO2NyeaCht52jFYS7c6nrq+Sun0Dtxi146fPFuLhVjo1p1YFIoNWQmM69sbRCM6HuGqJzTAzoULPUtUY1vdGolfhoVaQXihGoPeQYLGyPyiUp30iIm4aetdxx1YjOH7crOd/Ej6fTOb1mCiq1hglvzRXJs4yJKdxqzF7TuUVmG1uu5fDmznhe2hrLJweTOZ+u53Y2rJUnZ7WC7D+LF5yLzip1LSNHT57ehN5kQ6sSH9oVLT7XyORNMaQWmmjo40RcjpHJm2NILTDZO7Qqw2S18fnhZJ5ZH83hxAJqN27JnF+3E1yrrr1Dq3bECFQoUyn5Jt7alUCIm5pmAVokCSw2GwuOplDPy5FJbQPs3lqsgbcjOQYLPWq68upXe6kb7E7Len6kZRfyxIdb6RLqyr74PF7rHGzXOO9Fi0+lM6KhFwPregLQK9ydXy5ksvRMBtM6Vo3GAPa0OyaXr46nYrBIqB0cefTlWXQfMsreYVVbYgRazQ2c0LfCRqJWm8SsPYn0CXcj3wr7UgykKjRsiMqjTbArGYVmfruYdfMblTOFXMbIhl5suppD3xAtA19aReCQb6g3ZjGKvAIKDGZqejgQ7ulg71DvKVabxOnUQnr/owNN33B3jieLcoE3YrFJ/Hwug08Pp2CwQMf+w1i484xInuVMjEDvERXREu1YcgEuGgUXs00M7x3B24+3RyaTkZ1noPvzv9AqyJlVl7IZ1sATlcK+39361fag0GRjxfks2gY74++sRm+2sS8+j9oeDrzYPtCu8VUkg8XGyvOZ7I3Lw2qTaBPkwkONvSu8RJ9cBiqFHL3Zhub/ergWmGw4qMR3/Rv5+GAyBxPyqdukFU+++SFBYrq2Qoi/SqHMHEsqoEWAlmidgRmPtC0pRODp6sDMJzpwLsuIr7OKK1kGO0dabHiEF18OrEWIW3ERBaVcxhtdgnmtSzCO98gHtiRJvLc3kUy9hZndavBh71A0Shmv74zHaLFVaCwymYxuYa4sOZOB1Va8Xm6xSfx0NoPuYa4VGktVUWC08MbOeA4nFdK4fVdmLl4lkmcFEiPQe0h578612CQUMnBxUqH+xwYcX3cnCs02HJVyzNbKsZkIwN1BydAGXvYOw24uZhSRXWTh7e41Snp6PtHCj8TdCeyPz6vwht6PNvNh9r4knl4fTR0vBy5lFhHu4cCDjcRxov8nSRKrLmez7GwGFhs4u3kwZvLr9g7rniMS6D2ovKZza7priNEZMJss7D2TRJemf2/6WLT+HBGeGrZe01HDTX2DuwgVKTbHSCNfp+saYjfxcyI2x1jh8TipFLzTI4Rr2QaS8ow8EOFFmIdYi/5/6YUmpm+LJ7PIglwuZ8DDT/LgpFdRKMXHeUUTv3GhzPSo5cbEdVmMbeLD8BlrmTi4KfVDPfltVyRnr6TRwteRxn7aSlMgXYAAZxU7YnKRJKlU7d+r2QYa+dqvPmptTwdqi01c1zmeVMCXx1LIKrJSp0kLJs9diIePn73DumfdGws9wnX+2p1bljt0XTVKxrf049eLWQwKd+PciSi+X3kUbX4+DT0d2B+fx/gWvmX2esLda+qvxWaT+PF0BoUmK0aLjbWXs7mcUUTX0Mq97miTpEpztrgiXBn+JR8cTkftG8abi37n7cVrRPK0MzECFcpUj5pueDoo+e1iFlE6A1qVnGNmG51DXZndq/K05xKKKeQyZnarwXcn0xi3+hqSBI39nHinRw206spXSEKSJPbG5bHhag5Xs4qQAY39tNxfz4NWgc72Dq9cWG0SP55OZ82KniBJ9B/7JPWbt7F3WAKinZnwf8p6XTTPaKHIbMPdQVnqWIJQOZmtEhISajsfMfovkiTxxdFUonUGuoa6km+yIZeBo1LOlqgcetR0Y1Q122x0LCmfeYdTKDDZUKrUjHruZfqPnSBa7VUg0c5MuCVlvUvXVaPEVVMmtxIqQHFvz8r7wbwnNo8YnYHGfk6svaKje5gbRktx2cjB9T3ZdFVHYz8nInyqfm9Lq01i1eUsfjqTCUCLLr14etZnaF0q97T6vUYkUEEQqoT1V3V0DnVlwxUdn/WricufBeYH1PVg6pZYBtXzYMMVXZVPoGarxIyd8VzOLCK4dj2eenMu4Y2a2zss4V9Uzrkawa5ET1GhsrHaJKKyi9vQda/pVpI8Afyc1bQKdEYuk3E5s8iOUd49o8XGq5e0XM4sYtiTk/lw5TaRPCsxMQIV/pVoiSZUJn8t+cllxQU7/slik5D9eb08SZKEBNedmy2L+26LyuW7k2kYrRJOLq70HvWoWOus5EQCFQSh0pPLZDT0cUKrUrAmMpv+dTzw0Rbv6I7VGTiVUkigi5rGvtpyef3LmUWsuZzN8eQCLDaJUDcNfWu70zvcHeVdZu2MQhNv7kogOd9cXM5w8Cgem/4eKrXYQFDZiQQq3FBFFKEXhFtxfz0PFp/OYEBdD6ZsjqFNkAtmm8TJlALGNPHm1wtZvNG1Rpm/7s7oXJaczeC+2u6EuKkx2WS4qWXsi8vjWHIB0zsF/7kB6/adTCngm+OppBZYqBFejykff4d/SFjZvgGh3Ig1UOGmKrIlmiD8l7bBLnQMcWF7VC6D6nni76wixE3NgDoe/Hohi9GNfcq8BV1agYnvT6UzsqEn66L1ZNTuBR3HsCVdhZer8581ae+sRd+2qBze3p0InjV4+bPFfPjrdpE8qxgxAhUEocoY28SHJr5ObLyqIzLLgEwGTXydeL1LMHW8HMv89TZfy6FrqAs/X8rnpa9WUvvPDT2DHn+Ot8feR1c/WBepY3gDr1tuFG+TJH67kMXPF3UAvPLFT/gFh5Z57EL5EwlUuGViOleoDJr4a2niXz5rnf90ObOI1oHOBIQGlSRPALXGgV4PTSDy54+QySBTb8bP+eZNEi6kFzL3QDI6gxW5QsGIp6eJ5FmFiQQq3JbKuju3yGxje3QOO6JzydRbcHNQ0DXMlX613XHVVPyfeVqBiePJhZitNkLdHWjqf33Hk7ImSRKXM4tIyTcT5q6hVhlPZ1psEhuu6NgXl4fFJtEm2Jkh9T1xUlW+kn9l5a//YlaL+bprFosZuQxsEjfdLStJEmsv6/j+dDoA9Vu04/kPv8Ldq3pVTrrXiAQq3JHKNBrNNVh4Y2cCfs5Kuoa5YrDY0CjlJOQambo5llk9QghwqZgWanqzlS+OpnI2VU/bYGecVHL2xKWjP27j+bYBNCynDif5Rivv7k2kwGSltqcDy89lEObhwMsdA8usNN8nh5LJN1p5pKkPGqWc9VeyeXNXAh/0DL3jTTSVXWM/J1ILTGSlJ3Lu8F4at+sCgD4/jy1LvqSvj8RlhQxvp//+KL2SWcSPZzI4n67HJzCY8TPm0Lhd54p6C0I5ErVwhbtm70T63t5E/LQqonQG9GYbjf2cuJZlIN9kpXuYK/vj8/m0XxgWGxxOzOd8uh4JqO/tSKcQlzJLMFabxJu7EvDVKnF3VLErNo98o4UWQS4083FkxflMZnarUeYbXQDmHUrGUSVnQks/ZDIZVpvEnANJ1HDTMLaJz13f/1q2gdn7EvlqYC1Uf/6+JEli2tY4Gvk68ljz6tkVJEtv5oVNsYxt4s2yS3nUa94GV28/TuzaTMdAB6Iz8uhQw5VB9T3/9fkWm8Qjm9IozMvhsenv0XPYGOSK6jtir45ELVyh2kotMHE5swhfrQpfrYoX2gWUTJUuPZvBtWwDZpvE1qhcVl7IJNBZTWM/R0DG3tg8fjydzssdg8pkZHgqpZAiiw2bTI7JxZnDCwfi7+nEki2XeOu7g9xfz4OV5zN5rUvwLd9Tb7aSmGdCIZMR4qb515GexSZxMCGfRYNrl0wlKuQyHmrsw7t7E8okgV7K0Jc03j6SmM/JDDMOCmjo48jOmFzcNEqGRXjd9etUNl5OKia19WfB0VR61XJFm3cRS+4FBoep2RuXS6i7hoH1PP71uQfj85h/NBW92Ya7tx89h49FLhcHH6oTkUCFu2bP6dwL6Xqa+Ws5nJjP291rlFpnHNbAk0dXXaNHTVcWnUzjkaY+7I/PZ12kDplMhp9WxYONvPlwfxIf9AolyPXupnl3xOTSIdiZDdF5JHw1Bgd18T+vZ4Y05WqCjoTLiZxP15NntOKqufEopMBk5aczGeyLy8PfRY3FKpFntNC/jgfDI0rv+LRJElYJHJSlk6uTSo7RUjb9Mq9mFZGQa2TO0SxSZW50HjqefF0m21f+SIS3ljWR2bQI1BLmXv2aYLcNdiHIVf3n+q8Ok1WipoeGUY28aRvsfN3adpbezLt7E4nWGQFo02sAE9/+RCTPakgkUKFM2GtzUXFZteKNHIp/fJD99f9jdAaCXdWsvpzN0AZe9KvtDhR391hyJp0etVxZF5nNxNb+dxVLjsGCXOZAs3DvkuT5l05NgvjsVBxuDgryjJYbJlC92cqMHfGEezgwqL4nF3JlqOXQ3R1OJBcQn2tkaofAkg9utUJOQx9Htkfncl+dv0dDm6/llEmPTKtN4lyaHotNwurlyazlm1Cqir9sdL5/JG881Jf7arqx+WrOXf8OK6tgVw1PtfLnqX+dyPvbpQw9Xx5NJT7PhG9QCFM++pbQehEVE6RQ4UQCraasVhvbTsRzMTYbR42S+zvUJNjHxd5hlbn63o78dCaDtkEurInM5qk/1wABNlzV0djXiTNphXQLcyNLb2FA3b8TTI9abhxPKcBZpWB71N0nUHcHJTZJ4kx0FgaTpVQS3X82CV9HBacN1pvuCl5zOZsgVzUpRjlZfvXo88R4DEWFrP/2Uxp4yriYksvJ5EJaBf2dHJ9o4cdbu+K5mmWgjpcDZ9P0RGUbeL9XyF29Jyj+YiABDYM8iRj1WEnyBAgIrUWdiMZoLbGcSdPf9WtVVZIk8cbOBM6l6/H0C+C5V1+nfd9BopZtNScSaDW082QCT87djrebAx0bB6LLN/LGooMMbF+LL6d0x8lBVW6vXdEj0WBXDaFuGtw0Co4kFTBjZwJN/Ys3EUXrDAyq58nJlEK8nJQ4qa6fQgtx1VBkkSg02+46lp413Vh+PpNGvo6MemsDHz/XlQBPLT9uucjSLZe4L9yVRr5ONxx9SlLxeu399Tw4bPJl+hfLS6b+mnXswdQBbRhSy4XNUTmlEmiou4bP7qvJjuhconUGGvk6Mamtf5kcMVHIZJhtEq5qGfqCvOuu6wvykTlIt1xIoLqRJIkNV3WcS9fj6OzCR3/sxsGxardUE26NmJSvZg5fTOGhWZt4cVQLPFwc+GrNObYdj+ep+xtjNFkYOXMjt7Pz+k5VZOm/SW0D2BOXR20vDS0DtRSZbdT1dqB9DRd+v5RFsKsaJ6WcE8mFWP+vk4dNkjiWXICzWo6/891/qWgeoMVRKUdus6HKy6fdhOV4DvyKn1adZEQ9d9ZF6hjV6Mbn/oosNvRmG+kGGW36P1Bq3czJ2YUmbTtjlSAx13jdc90cijfyPNsmgAF1PcrsfKabgwIfJxVBjjK2LVuILiO15NrJvdtJio9jZ0wuvk7KUr/fe0GMzsBT66L59kQ6MpmcERNfFMnzHiJGoNXMm98fYurIFrzz4xFmPdGB398ZSHx6PlMX7CXYR8u5mCz2nEmiW7Nb3wla2floVXzUN5T1V3Ssv6IjW1+8xtg1zJWP+oRxMqWQA/F5+DureG9fIsMbeCGXwZpIHWqFjMjMInqHu991HAq5jNe7BPHF0VS2X8uhbbAzWpWcc+l61mYX8WqnoJseYVHJ5VhtElqlREZ89HXXM5LiCFfLcFBW3HdfmUzGoPqerLmcTa8Qd14e0oWIVu3IyswiKTqSpl5KTqea0SgVPLMhmte7BBPiVv07iayLzOa7k8WFEcLqN2bKx9/gE1D2xeyFykucA61GkjIKaD5+GQ/2qIubs4ZZT3QouZavN1Hzwe95YXgzYlLz+P6VPhUWV4VvLJKkUmtPJquNGTsTCHJR4aNVcTy5EJtNomWglnyTjSuZRbzfK6RMK+r8VYnIZLURdpuViN7bm0g9LwdWR+mZPO9HGrRshyRJ7F61nDXz36WptwJfJxUjbzKaLUuSJLH4dAaHEvJpEaBlZ0wuzfyduJZtRG+2IZNBbU8HWgZoWRup49N+Ybg5VM/v5xfS9ay6lMWx5EJcPbx4fPr7tOnV395hCeVEnAO9R6RmFxLi50JkYg5THijdxd7FSU3DMC/UKgUpWRW72aOij7n8c+OGWiFnZrdgfjiVzvpIHbU9HZDLZGy6mkPrIGfe7Vm2yRPAz1nNgLp3dixmaANP5uxPYkwjbxa8OA6tmycGQxFazAwOd+LXi1l81q9mmcZ7MzKZjMea+9I22JlPD6WgVsg4k6bn2TYBdApxwWorHpFticqlqb8T26JyGdGw+p0LTS80M2NnPAq1Aw9Nfp1+o59AqSq/PQVC5SYSaDXi5+lEfFo+7SL8OXg+hX5twkquFRSZuBCbRe+WIfh7Vvwajb1r6DqpFDzbJoCHm/pyJasIJKjtFYB7JRwlRfg48XhzP745kUqbIGeCXYuQgMsZFn67VMiMLsG4O9on7vrejuQYLAxr4EWm3kyXUFcA5AoYFuHF4aQCQtw07I7Nq3YJ9HRqIZ8cTMYmQeNW7Rn4yER7hyTYWeX79BDuWLCPC81q+xDi58onv5wkzN+V0T3rkZCez5QFexjcMZyVu67w+Qvd7BajvWvoumoUZXI2srx1CXOliZ8T26NzicwqQiGT0SbYhWkdgyp0/fOfzFYJSSou9PBv65yhbmrMVolCk9UO0ZWPHIOFOfuTuJBRBECjdp157v35N31ebnYmh7asJSczHXdvX9r3HYSbpygeX52IBFrNzHysHcPeWM9b49qxYkckEz/egZebA4/2jSAho4BgX5dqtYGoOnN3VFa6UZxaIcNZLcfLUcmxpAIG1fMomTI3W22cSilkQF2PW2rtVRVE6wzMP5JCtM6Iu7cvL3z4FfWat7nhc2w2G7999RFbf1lCy6698PYPJubSWX77+hP6jHyEEU9PE1WJqgmxiaga2nY8jgkf7SDAU0uHRgHkFBhZcyCafm1C+XpqT7SOlWPNxt5F6MuLxSZxJDGfo0kFmKwSoe4aetdyw8vJfr93s1Uiu8iMm4Pyrkewy85mkKk3E59rIthVzYC6HhgtNn69kIVWrSDHYKFfbXe6hrmVUfQVr8Bk5d09iVzONuLs6s6wCZPpPfLRW0p8v339CacP7KRNz/5sXbmYwtwctG7u9Bk1jqM7NtKsYw9GTJxaAe9CKAs32kQkEmg1ZbXa2HQ0lstxOhw1Sga2r0mov6u9w7pOdUuiyfkm3tmdgJeTimb+TshlkFZo5kB8PmOa+NC/zr8XHi9PG6/qWHk+E6VcRpHZRs9abjzazBflHRY+yDdaeXlbHK0CnJDLZZxK0aNSyGgd6ExCnpE8o5W3utW44/tXBsvOZvDLhSxqNmjC9K+W4ezqfkvPK8zP5YWBHRnyxCT2rFnJs+99Tlj9RsRePs+C15+n25BRrPpuPp+tP4DWpep+wbiXiAQqVFrVKYEaLTYmbYyhT203TqXoScwz4qpRkG+0MiLCizWR2Yxv4Ufb4Iorqbg/Po9lZzOY3rn4bKauyMK8wynUdNcwrrnvHd83p8jCNyfSOJNaSH1vR8w2iahsA11CXXmsuS8aO67T3o3kfBMf7EskPtcEMhkvfrKIll173/Lzd69Zyen9O0m4dpmnZn5M3aZ/f+5eOXOcb2a+SI3a9WnWqQfdBo8qj7cglDFxjEWotOy9O7cs7YvLo4arisOJBbQKcObt7sWjsGvZBmbtSWBYA09+v5hVoQl0XaSOx5v7lWz48XBUMqmtP89vjGFME++S3p63y91RySudgsjSm4nSGZDLZNT3dsRZXXV7XW68ks3CE+lIQEBoOFM/XkhQrbq3dY98XTY+gcEc27WZWhFNS12rFdGU1IRYWnTpRX6OrgwjF+ylan5NFKqdiiz9V14OJuTT0M+JfKOVkY28SqYwa3s6cH9dT5LyTCTnm8nSm+/o/marDb3ZelulGDP1ZkLcSm/o8XZSIZfJyqT+r5eTijZBLrQKdK6yyTNGV8THB5P55kQ6DlpnnnxzDh+v2n3byRPAyz+AxKgrhNaN4MLR/aWunT+6n9C6DUmMuoKXX/XsWnOvEQlUsLvM3CJiUnIxWu7+A92eiiw2JKm4z+g/qw75u6jINVpxVsspus33eTa1kHd2J/Dgb1cZt+oaE9ZF8duFrFv6fdXxdOBYckGpxy5nFuGkkt+0J+m9YHdMLlM2x3Ew2cDQJ1/gq20n6T7kzpepWnbtS9TFM3Qb8iDfznqZYzs3kZudybGdm/hu1st0HTySqItnaNm16n9hFMQUrmBH20/E8+GyYxy/ko6Hs4bcQhOjetSlrUxv1x2rdyrIVY3BYuNatgFdkQWP/yt2cDA+n5ruDpxL0+N1G7ugt17LYcX5TAbW9SDMXYPBYsNHq+Jiup7jyQXM7F7jhrtqRzby5q1dCVhtxcXuY3QGlpzJ4OGmPrdcWrC6upJVxKIrZiTg4Zdm0nvEw3d9T42jIyOfeYkNS75h0OPPsXHpt3z77isEhdVm0OPPsWnpt4x85iU0jo53/wYEuxObiAS7WLrtMtMX7mfyiOYkpOeTpiuiYZgnWbkG1hyI4o22vvhoq1YSvZpVxIf7k+ga5srhxAIeaOiFh4OSXTG5ROkMNPHVYrTZeK5NwC3dL63AxItbYnmwkTcrL2TRo6Ybno5K9sTl4e6gQKOQ4e+s5tFmN94MFKsz8PulbKKyDfhqVQyq50GLKlBMorwUmix8ciiF48mFANRr3oZXv1hapklt16qf+eXLuQSE1sK/RhipCbGkxsfwwNPT6D5UfIZWJWIXrlCpZOUWUXfsj7w3vgMzFx/mqfsbU7eGB2sORHExNpv729dk7/5IpneuegUfvjmeyrVsA51CXDiXXkShyUqEjyMWm8TBhAI+7B2K558jU7NV4lxaIXlGK95aFRE+jqVGhT+eTsdksbE3Pp+Z3WqUdHKx2iTe2BlPy0Bn1lzOZtHg8DveDHSvScozMe9wMleyDGhd3Xl61jxadO5ZLq9lMZs4c3APuVkZuHn50LRD11LNyIWqQezCFSqVJVsu0b9dGHNXnGDFW/1LKiON6V2fJ+dux2qTiMw102JYRwK9i0dKVWWX7oSWfqyN1LEmMhutSo6TSsGmqzm0CNAyu1dISfLcfE3HivNZ+GlV+GpVJOQaKbLYeLy5b8ku3WvZBpr7awl0UZdqg6aQy+hT253DCflo1XLSCs0Eu1b/9mF3IzXfyJfH0zifaUbt4MCwp55m6BPPo1CW30egUqW+rSMwQtUjEqhQ4c5EZRIR6snB8ynXlRV87L4IJs/fQ4s6vpyPySpJoFWFTCZjcH1PBtb1IFpnwGyTCHJRl2rtteZyNluu5TAiwpNDCflcyNBT011DiwBnvjqehk2C9jVcSkajhn/ZLGSw2FApZJisEop7fC3zVnyRWYNzqTG07NaHJ177AHfvOz8DKwh/EfM+QoXTqBTYbBJ5ehNGk6XUtfScIly1agqKzGj+r8VYVTvmopDLqOPlSISPU6nkWWCysvJ8JvfVcWfVpWwG1PVgdq9Q2gW7sPJ8JqMaevL9qTSsNolm/k7E5RgwWGwciM8ruUe+0crayzrCPRxRymX4OVetteKKlF1k5pWtsZw7vBdkMh596R2RPIUyIxKoUOH6twtj09E4Wtb15f2lx0rONeryDby75Aj92oYRnZJL24jSZ+UGTuhb5RLpP+2Ly6OZvxMbruiY1iGQDjVc8dWq6B3uzrjmvhxOLMRNo+Rsmp6etdw5mapnSH1PvjuZzhs74/nkYDJPr4+mVaAzBxLyGFjX457fTftftl7T8cSaKC5nGfD0C2DWj2vwDgiyd1hCNSKmcIUKN6B9TaYvPMDA9mGs2R/NL7uvUCfYg4Pnk3m4TwO2HInl6cFNcFBXvz/PjEIzQa5qTqQUl8D7fy38tXxzPJVaHg6k5JtoHqDl5Y6BzD2QTIcaLvhqVVgliRA3Nbti82jg48iAuhVfW7c82KTiUoD5Jiu+WtVdremmF5r57UIWW6JyUGk0jJo4jYGPPHVdo3VBuFvV7xNKqPSUCjnrPhhEv5dXExHqyfCutbHaoE19P5Ztv0zr+v7MePi/W0bZu6fo3XDRKEjON6FRyEnOL06mf4nOMeCmURKfY2BFnomm/loa+2n5uG8YW6Jy2B2bh8lqI8RdwxMtfGnury23pJCSb+JatgG5DBr6OJVrA+89sbmsOJ+JXCbDy1FJQp4JP62Kcc19r/uScTPfnkhj4xUdklzOfQ89wfCJL+LkXHGlE4V7izjGIthNYZGZn3dG8suuK+QXmggPcufJgY3o0jTolhNDVUukGYVmpmyO4b46HpxMKeSljoH4O6uJ1hl4b28iRosNqwRD6nmyMzaXL/rXrNAjKpl6M18eTeVqtoGGvk5YbBIX0/W0DXbmyZZ+OKnKtnrRxqs6Vl/KZlB9D86l6UkvNFPT3YEabmr+uJTN9M5BRPg43dK9EnINTN2RisloYMbCX4lo1a5MYxXuTeIYi1ApaR1VjB/QiPEDGtk7lArjo1XRIcSVK1lFpBeamLI5Fpsk4aiU82Bjb/qEu7PyfCan0wrx1ao4lFBAl7CKaUOXY7Dw2vZ4Ooe64O+s4kRKISq5jP51PcjSm3lndyKzetQos4ReYLKy9EwGY5p489vFbEY39qamu4ZTKYWsupTNiAhPFh5P49N+YTf8QmW02FhwNJU9ccUbrRq0ak+Dlm3LJEZBuBGxiUio0m60qchosZFeaKbAZK3AiG5uQks/HFVy8o02PB0UPNTYh0WDa9OvdvGGoJENvUkrMNPMT8vhpPwKi2vN5Wya+jlxLKkQqwSvdQ7m+XYBxOoMFJhsKOSwP77s4tkTm0eLAC1rI3W81DGQPuHu1PFyZGQjb4Y18ORKlqGkNOJ/ySg089auBPbE5eHgpOWZWfN4Y+EvYr1TqBBiBCpUef9siZapN/PL+Sz2x+fhoJJTaLLR0MeREQ29bnk6sDwp5TKGNfAiKc+ETYIarmoU/9d8WiGXFa85yiRM1ltfYrkbkiSxIzqXwfU9ydBbmNjKryQJvdo5mKfXR9O/jjvbonLoXrNsGkGn5JsIdtVwOlVPg3+sdbYNdmH9FR11vR1JyTdTx6v09YRcIyvOZ3IsqQBJpaHPyEcZM/UNVGpRUEKoOCKBCtVKWoGJ6Tvi6RDswuD6nqQUWXFTyXFVy/hwfxITW/nTvob9N5UEOKvJ0lvoX8eD3bF5pWrTxucaSS8wk623EOpWMQnBaJUostjIKrLQIqD05iSlXEYTPyfMNokMveUGd7k9jio5erMVmySRY7CWKr6flG/C00lFrsGKk+rviTKz1cYXR1PZHVs8XduiSy8efXkWPoFVr+yjUPWJKVyh2hg4oS/fHE+jZ5gbpzOKMLi58ugDrandJJR11/IY3cibBUdTKCqDPph3y0WjoHWQM0arjZgcA3MOJHEoIZ81l7N5a1cCIyI82ROXT5/wshnt3YxaIUMGuKrlxORcP2Uam2NEJZfhrC67j4yONVzYF59P1zBXvjqWWjLVnlpgYvGpdDoEOxOXY6Sx39+zBmsjdeyOzUPjpOXpWfOYNu8HkTwFuxEjUKHaiEnJJU5vo327YAaE21gwtUfJtc5Ng3jh4+009HFkT1wu/Wrb//zkI019eHV7PG2DtDipFeyIzsFFo2BofU82XdMxLMITP+eKKT4ul8noGOKKRYKTKYVs/3Oq1mKT+O1iFmarRHS2gS6hd7+hyWixsTs2jx3ROeSbrOyNzaWGm4Yn10bh4aAkz2jl/noe7I/PY3B9DzR/tmvbEZ3Dsgs6ACa9/wUtuvS661gE4W6IBCpUG2ejMmkb4c/Okwkseb305qIB7WryhNlGrUAt0TqjnSIszctJxYe9Q1l2NoOdMTo8HZVc0xmJ0Rl5qLEPncsgWd2OYQ08eX1HPA828mZbdC6LTqVjk6CxryOdQ13YGpXDhFb+N7/RDeQZrby5Mx4PRwXta7jQxF/Lwfg8onUGVHIZbYKcMdsk1l/R0T3MlRERXuQbLfx2MZu1V3ORIWPwE5No1qnHzV9MEMqZSKBCtaFWKdAbzLg4qcnMLSp1rchowWCyYrVJqOSVZ4emp6OSSW0DeKKFLxmFFjRKGX5alV12kdZw0/BKpyA+OphMkIua++t5IEkSx5IKSYvL4+3uIbho7u4c6BdHUmjo60RsIWzNUNOgVXtU+pP45GXiIJnYEZ1Dn9ruzOkdirNazszdiZzPMmExm+k25EEenPQqrh5eZfSOBeHuiAQqVFkJ6fn8sOkCkfE6nB3V9G8Xxqmr6Ux7sCVvLz5Cu4gAXJzU2GwSM384RGM/LUeTChj3Lw2o9WYre2LzuJxZnHib+DnRKcS1ZPqwvDmpFIS6l22RgjvR0NeJhfeHczgxnytZRShkMsY186WJv9Nd19xNLTBxKbMIH1dH3CLa8OoHXyKXy5EkiaVz30B/fCP5xhzaBLmQWmDi3b2JWGzg4uHJ5DkLxdlOodIRCVSokj5cfoyPVp5kZLc6NK7ljcVi450fj6BRKdl5MoHGtbwIH/0DnZoEcTE2C5XFTFNvR1JyDTTxL32U5WRKAZ8eSqGhjyO1PR2QgP3xefx0NpPpnYKod5vl5Ko6lUJG51DXMp9CPpump2WAloOJel5//2Xk8uIvJzKZjKETp/Fs72XcV8uFU6mF7Ml3w2KDniPG8tgr7yJX2P/LhSD8k0igQpXzw6YLLNlyiekPtWL2smO4qBVk6010axpMt2ZBLN58idQsPZNHNEehkNG+oT97zyRxLEHHa52DSo2k4nKMfHoohbFNvFkbqeNatgGrBB6OSkZGePHe3kQ+7huGj7b8W4alF5rZci2Hs2mFSBLU83bkvjru1aZZtk2SUMhlWGw21BqHUtdUKg02m4TVJrE9KocMfSYOWmcenPSqSJ5CpSUSqFCl2GwSs5cd47mhzfhgyWHe7BRAmLsDRouNH85kcE2SqBXoxtDO4Zy5lklMSi7uzhpG9qjLQz3rsXPJzlL3W3M5m7613Vl2NpOJrf1oH+yCBOyIzuXn85l0DHFh01Udj/zLtG9ZicsxsuFKNrtj82gb7Fx8TlWCXKOV17bHM7qxN/fVsf+u4btVz8uRXy9k0SrIgy3Lv+WhqW+VXNv264/4umnZdDUHSS6nfZ/7efLNuTg4ae0YsSDcmEigQpVyNjoTmUzGzhNxjKjvQZh78UhGo5TzWFMfJmyIYcqolkQl5bJyZv/rnv/PqkUHEvIZ3sCT5gFaOtQonrKUAb3D3dkfn4+3k4qtUTnlkkAzCs3MO5xCQq4RvdlGLQ8NhxMLSM43o1XJidIZeLCRN79ezCLETUNDX/tXUbobNT0c8NWq8HewsWvDSpKvXaJB++5EnjjE2SMHMBkNqDUOzPppPTVq17N3uIJwUyKBClVKvt6Er7sjKZmFtA0pvTapUcrx1qpRKuTk6U03vM/ACX1Z+81mjBYbRRYbQS7Xn7cMclFhskrlUnghz2jh9R3xdAtzJd9owWSViMkx0jJAy3NtA3BWK4jVGZixM57B9T1ZE5ld5RMowPNtA3h9ZzxtAp3xKYpk86KTpOcVIZPJcXDS8sa3v4rkKVQZohKRUKXUCnDjUryOdo0COJpcUOpacr6JLL2ZNJ2eusHuN72XXCbD31mFq0bB4aQCrLa/686arTaOJxfioJQR+C/J9W6tv6KjkZ8jhxMLiPBxYn7/mszvXxNnjYKZuxOw2iTCPBzoHOqK2SZxIrmA22k9WFkFuKiZ2zsUjVLOLxeySM8r3vXc5f4RfLBiCzUbNLZzhIJw60QCFaqUIB9n2jcMwMvNkWNpBr4/ncGlDD27YnJ5d38yLz7YkuXbI3m8f8Ob3mvghL5MHtOWyxl6nJRyPtiXyOmUQo4nFzBzdyLhng4cTSqgT7h7mb+PHdG5BLtocNEoeKqVHz5aFf7Oap5t7Y9NgtOphQC4ahSYrRI2Cap++izmrJbjqJJjsUqAjMGPP8dTMz/GLzjU3qEJwm0RU7hClTNnYid6TP6dpwc3IV1XyIqzSfh5apk2pg0/bLrIU4MaEx7kfkv3enpwE37ZdQU/sxEvrZIV5zNRyKGpn5aYHANKuaxc+nFmF1koMFlp7OdUqmiCTCajsa8TcTlGGvk6sTcuj77h7oS6ae76HKa9GSw2PjucwsmUAgwWiXZ97uehyTPw9g+0d2iCcEfECFSocuqHeLLj0+EcupjCL7uv4e7mRFK2nk9/O8WEQY15+7F2WK02thyN46s1Z/l5RyR5hf9evk/rqGLbx8OQy2F9pA6ZDMw2WBOZjZeTihldglGWQ+UiN40CrVrOlayi665dSNeTUmDihU0x1PNy5EhifpXfhbs7JpeH/7jKwYR8ZA7OvLpgKc/P/lIkT6FKEyNQoUpqEOrJxg+HEJeax7WkHJwd1bSq54tCIWftgWhemL8bPw8nmtX2ISW7kOfm7eK5oU15a1w75P9IiG7OGjZ9Nw5dvoEzUZkcWX+MWh4OOKrK7/tltzA3MgstJOWZWHk+k/vreWCT4LeLWWTozRSZrfg7q4jNMWC2Qb7RQrTOQC0Ph5vfvJKx2iTmn8jCYpXo9cAjPDLtLZSqiimSLwjlSXY7GxNa1fOTjn4zuhzDEYS7s/FwDE/O3c6MR9qwen80Ry+lEuLrwsN9GrBm/zVaN/Dnk2e73vAefx1xKU9ZejMvbo1jUD0PIjOLOJ5cCEhoVQqKzFbMtuLjNO4OCjqHumCV4HBiAf7Oal7sEIinY9X47ltgsjJ7XxLn0vX4BoUyb91+e4ckCLfloRY1TkiS1OrfrokpXKHakCSJV77ez6tjWvP24iM81LMe15aNY/4L3Vm27TJ92oSxdOtlYlPzbnifgRP6lpwXLS9eTire6V6DXTG5JOWZuL+uOwPqeKBVy/HRqnBSyRjX3Idang6sv5LDtqhcmvppqeul4c2d8Rgs9u9peiM5Bgs/n83g6V06zmUU0aR9V2YtXXfH98tITiT+yiUK83PLMEpBuDtiBCpUG0cvpTLug600rOlJ92Y1eGZo05Jr15Jy6PjsL4zsXgdfdyfeePTWCpOX92hUkiTOpeu5nFEEMmjs68S2qBzcHZXsjsljcH1P+tV2p8hiY/m5TOJzjGjVcloHOVeKnqb/ZLVJrDifye8Xs7BKEN6oOY9Pf++Oj6ec2LOV1Yvmk5mShIu7B9npabTo3JMHnpmGT2CNMo5eEK4nRqDCPSEps4C6NTy4EJNNt+bBpa7VDnLHyUGJn6cTiRkF/3GHiieTyWjip2VkI29GNvSmrpcjBxIK0MjlNPJ1YnB9TzRKOe4OSp5u5Ue+yUpjXyd2xdx4FG0PVpvEsxui+eVCFsiVjJj4Im8vXn3HyXPH78v48cM36dB3MC3ad8bHy5M+D4zFw9eftx8fRlpiXBm/A0G4PSKBCtWGj7sjcal5hAe5cfRyWqlriRn55OtN6PKM+HrceneVipjO/X8Giw0ZoDNYrusCI5PJqOvliMkqkWOwVFhMt2p3bC4pBWacnF35cvtJhk2YXNJx5XblZmfy8+cfMPCRp1j3zUfUSDzAfZoEcvet5OjmP+gyaBQ/fTSzbN+AINwmkUCF22azSWw+GsvsZcf45JeTXIjJsndIAHRoGEh+kYl+bUKZ8d1BthyNQ5IkriToGPvuZp7o35Cfd0QytneD2753WSdRmySVqnz0FwelHJmsuNH2hXR9qWtWm8TFDD0qhQyvSrSJyGKTWHM5m6+PF39pGfH0i7i43d308t61v9KqWx/WfTeP6e29Gd7Ag7bBLkxu5U0LNwvG/FyunDlOVmpyWbwFQbgjledfoVAlHLqQwiPvb8Fdq6FHi2AKDWb6TDtB09o+LHmtL95u9uudKZfLeOfx9sz47iCvjW3NK9/sY8iMHNyd1Tw5sDGnrqbTp3Uo9ULst3Z4OqWQtZHZnE4tRAJC3TTcV8eDXrXcUMhlKOQyuoa5UmCycjXbwPJzGfSv7VGyBuqnVXEqpYCetdzs9h7+YrbaWHgijcOJ+eQZbTTt2J2Hp75JYM3ad33vhKhIAsPCidxvof4/RuLdQ7R8dWwvYfUbkxRzFS9xllSwE5FAhVt2PiaToTPWMePhtmw6EsO8307jqFHyUK96qBRy+r+8mn3zH0Cjtt+f1UO96qM3Wpi+8AAdGgZwf4dapGXr+WrNWYZ1rs38F7rd8b3/2cnldq25nM36K9n0q+1BuKcDZpuEu0bBrpgcTqUU8FLHIBRyGcMaePHytjjur+NBbK6BZzZEo1LI6BTiWlx8IbOozJtd3669cXl8eTSVIosNtYMjk+fOo3WP+0pVVbobao0DVquVQoMJo8WGRvn3ZFmm3oyzqxeF+bnX9RUVhIokpnCFW/b+0mM8PbgJ7y89yoB2NcnZ8DQXf3yYwiIz15JycHFS8/vea/YOk/EDGhG74nGGdqmNQi4jIsyTk98+xDfTeqJW3X1z5juZzo3WGVh1KYuBdT1YfTkbg8WGg0LO5ms5BLlqyDNa2XhVB4CvVsWs7jU4kJhPZKaBzqGutAhwZm9cHll6CzO7h6BW2O+fbrTOwMcHkymy2OjQbzDf7DhDm579yyx5ArTs1odTe7dRt1krll7QlUx364os/BxZSJPuA8hOT6V24+Zl9pqCcLvEMRbhlhQWmQkc/i2ThjcjK7eIr6b2LLlmsdqoM2Yxzwxuws5TiWyaM8R+gVawWx2NLjiagqtGwaarOczpE0qwqwYo3jT04pZYetdyY0tULl8OqFmSiCRJ4mq2gatZhpL6vAF32BnGYpO4mK6nwGzFT6umlocGmUxGQq6RxDwTGoWMhr5OpUZ6/8ZgsfHunkTOpevpPHAET7/z6R3FczM2q5Xpo/vRusd9XD6ym7SYqwS4ORKTmU+vkY9y8fQJWnbtzaBxz5TL6wvCX250jEVM4Qq3RFdgwFWrJj4tn54tSp+/UyrktKrnh9Umka7T/8cd7m2RmQa6hrnS0NepJHlC8aahPuHuJOcbyTFYKDDZcNEUj5L/2nVb1+vO15UlSWLTtRx+vZCFl5MST0clcTlGFDIZKjnkmmzU9XSgwGTlk0MpDKjrzsiG3ij+Ue6w0GRhd2w+qy5lkaG3UKdJS8ZOffOO47oZuULBS58tZvazY/Hw8aP3o5NABrULC9m/4XeadurOwEcmltvrC8KtEAlUuCWeLg7k683U9Hdl39kkHu0XUXLNbLFy+GIKbRr4EeCltWOUt89gsrBmfzTRyTm4O2sY2qU2/p63/h7+fzr3RqNRuay4NJ/Jev2Mj9kqoZDLsEkSZV23/veL2eyJy2V4A0+OJBWQkGsk2FXN+TQ9EtAkyJ2EQhsuGg3DI5w5lpRPVpGF59oEAMUJeG2kjiXnc7CYTQSH1+PN6e9Rv8WtFaK4G94BQXywYjPHdm7iyLYN6Avy8Q+pyeSPFlIrokm5v74g3IxIoMItcXJQMbhjLWySxJZjcXy4/BjjBzQiK8/Aa98eoE19f9YdiGbS8KqzJrV022WmfbmXZrV9aFbbm8iEHGYsOsRDverxybNdUCnvfr30L038nNAZigvCX8rQ08DHCYCcIgubr+kYWNeDIBc1WnXZvaauyMIfl7IY2dCLNZE6Hm3mQ6ibhmPJBVzMKEKuVOPXbRQjBg4nMzWZX+e9Qws/OXujs4jKNuDtpOSlrXGkFZqRyeX0GzOeMS+8jkJZcR8bKrWGDv2G0KHfkAp7TUG4VWINVLhlVxJ0dH3hV14e3YrDF1LYdCQOFycVo3vWI09v4lJsNjvnDS/TxFNeft9zlakL9vLqQ61YvPkiJ66kE+jlzPiBDTl0IYUgb2e+e7n3Hd3730aiKfkmXt4Wx0ONvVl2NoOGvk5oVQqOJhVwXx13jiUVMLi+J91rlt3xlD8uZZGQa+RMqp4ZXYKp5fn3jtVVl7LYnuXAnDUHSx7LzcrgxUEd6F/TiUKTDXOzgWz/bSkBobV4/esVePoFlFlslV1qfAx71/1KVloKrh6edOw/jLB6N2/SLlQ/opSfUCbq1vBg44dD+HHzRc7HZPFovwb0bRPKki2XyC0wsX724CqRPG02ide/O8jLD7Zk1pKjvDa2DcZtk9jw4WC2HouneR1fNhyOITJed0f3/7ddugEuaia09OPnc5n0CXenjpcD/s4qhtT3YH98Ho18nehWxo270wvM+DurkaBU8gRoHeRMUWHpkoZuXj7UCK2Jo1LOhQw9+zauAuDx196/Z5KnzWbjxzlvMvOxoRj0egJCayFXKPl4yuPMe+kpTIbr+7cK9y4xhSvcluZ1fDn13RgOnE/m1JUMVCo5bz7ajjB/+55LvB1HLqWiVirYcTKBd55oz+BO4QA0ruXNz2/eR/Pxy3ikTwRLt11i1hMd/vUeNpvE4YsppGbr8fd0ol1EwHV9Rv+pc6groe4aNl7VsT8+H4skUdvDgQkt/Wnm71Smx0AAtGoFerMNo8VGTpEF9/+rXhSXY8TBsfTmJLPJSGJiAgnGQvRmiXrNmjJm6hvUblR1puXv1u/ffELMpXP0GjqarSu/x8VBRZ7eSKf+w9Dl5PDtrFd49r3P7R2mUEmIBCrcNplMRqfGQXRqHGTvUO5Imk5PrUA3IhN01yXIGr4uuDqp8XJ1IC793wu2/7H3Gq9/dwClXE7tYHeuJeZgsdl4b3xHhnUprsLzX0UXQtw0TGzlXw7v6nqdQlx4Z08i3cLcmH80hUltAnB3VBKjM/D9qXQKbUouHDtIRKv2FBXks3jOG1gsVowW6PfQeB5+8c0yT+qVWVFhAVtXLmbwY89ycPkCPunhj5+zmhyDhU+ObKZG1yHs3biGtIRY/GqE2TtcoRIQCVS45wR4armaqCMizJO9Z5NoWNOr5Fpsah4FRWYycvU4aZRMX7ifX3ZdJafASHigG03CvdlyLI5XH2rNlsMxnL+WQf1QD/q1q8Xk+Xswmq2M7lmv5H4DJ/StkAbd/6amhwN1vRzIM1rw06p5ZkM0Dko5ZpuETbLhqJD48qUnsCCnyGBALpfj6umDv4srY6fMuKeSJ8CZg7up06QlB9Ys58nGrvg5F5+5dXdQ8mxzd6auXkm7/sM5smOjOH8qAGINVLgHtWngB0DXpsG8vfgwK3ZGUlhk5uilVEa+tYGnBzdhyZbLLN8eSaHBzFODGvP88GYM71abpVsvoVEqmLnoIGGSkaktvamJiZmLiuvvvrhgLyaz9T9f2/xnJxVjBTXEntI+EJNN4nBiPr3D3elb2502QVrMVsg32qjvIoFMgcVkwlHrgqePH68uWIpcUfnXsstaUUE+bp7e6LIyCfq/s7pQXB3KbDbh7OaOPr/ytZIT7EPswq3G8vUmZDJwdryz6jXV2bqD0Tz9yU5eH9uGX3Zd4ejlVIJ9nHnsvoZsORrLiSsZfDChI3NXHCcizIuIUE/WHoimyGRBslh5pJEnHWr8ve57MCGPnWlmXJw1TBzchOFd65R6vYT0fJ6e8Qd7Y/NQyGWYrDZaBzozPMKLmh7lX881RmdgX1wehWYbvloVHWu4cDZNz09nM8gzWnH19GLSB18S0ar9PTfy/MuFYwf56aOZeHt50tZ6rVTD8tMphXwbBX7hDWjRpTfdhzxox0iFiiQqEd1jopJyeO6z3ew/lwRAl6ZBfPFCd2oG2L+DR2Vxf4daGM1WJs/fQ3igG08ObExyVgFzfj5OpyaBtGngx4odkbzyUGueHlx8aH/2U50Y8Mpqdp5MoHWgc6n7tQ504aODV3hhRHNiUkqPUK4l5dBj8u880L0Ong4JpBWa8XRUopDBm7sSeLljII39yq8ARXS2gW3ROWQUmnHRKMg1WJi0MRqTDfyCQ3nq5Xdo3qlHub1+VdGgZTuMhiKadO3H0gWz0Zslmvk5FXfFuZzPkGen8+uXc3n2XbGJSCgmpnCrGb3BTO8X/6BXyxpkrHmK9NVP0bVpMH2mrcJgqnxNmO1pRNc6xKx4jKkjW1AzwJX72oZxddk4gryd6d68BpEJOp4c2Kjk55UKOa+NbYOTWkF8rqnUveJzjQR6OpGcWYCbc+npvwlztzPh/kas3nOVmBwj9bwc/ywen8OYxt58fCgF879UKLpbVpvE/CMpvLcvESeVnGBXNceSC1gTqcNkgw79hvDhr9tF8vyTXC7nsVff5Y+F8xjy1ItE+bRi3mWJY+p6DH76FdYv/oqHJs/AwalqVdsSyo8YgVYzv++9RkSYFy+Oalny2MujW7HzZAKr90fxYI96N3j2vUelVJQcY/mLJBX/j1wu45+TmQqFDDcXBz47ksLrnYPwc1aTVmBi4ekMnry/MZ/8eorPnu9W8vMXYrK4mpRDLX8XmntpGNfMp+RaXS8HtkXlEuSi4khSPp1CyvYo0PJzmaQVmOhc04tN0bm4+wdhlBlw9XQiolUH8nOyRTuwf2jSviuTPljA8s/eIz9HR2BYOElJSSSuXMzoF16jfd9B9g5RqEREAq1mYlJyaVHX97rHm9fxITZFbH64FR0aBbJiZyShfi4s3Xa5pO6vzSYx9+cTqJVy0vItvLwjAQeVHKNFYuLgJuw+nci4fhGlmoqfuJJO16bBbD4ax1sdSxcj6BbmxjfH0xhc34OobEOZJlC92crmazqGNfBif6Erc9dsxN3bF2NREQvfmYaDk5ZrZ08Sc+kcNRs0LrPXrQ4atunIu0s3kHDtMtnpqbi4e1Arouk9uzYs/DcxhVvNNKvty/bj8fz/5jCbTWL78QSa1va5wTOFv4zsXofTVzN4vH9DXvv2AKNmbuTtxYdp9/QKcguNOKhVWG0wtm8ETw9vzti+Dfhu4wVqB7vz4cROpe6lUsopMllwVCspMJXenWuw2FAo5FhsoCzjKvLn0vSEeziwOcHEgy++g7t38ZcqjaMjj778Doe3radt74Ec27mpTF+3upDJZITUaUCzjt0Jb9hMJE/hX4kEWs30bxeGJEk8/uE2zsdkci46k3Gzt+KgVtC3dai9w6sSHNRKVs7sz1s/HGZs7/o0q+2NyWxlVI+6eLs5olEr2Df/ATxcNCSkFeDurGH/FyP5ampPlP9odN2jeTC7TyUysmc9Vl7SYbYWH1+RJImVF7Po3y6Mc/k2WgSU7bpatM7ApcwicovM+AQGl7rm4u6JXC7HwckJoyhNJwh3TEzhVjNKhZzNc4fy/tJjDHl9HTKZjBFda7NgcveblpoT/ta5SRD75j/AZ7+d5odNF8ktNFEr0JXxAxqxeHofnB3VtKznd9P7+HlqGdIpnISMAoJDvHh6UyyN/LREZxvw9Xamb5gXcekF1Pe+856f/7TweCobruYA4OnqzqGt6xjyxKSS6+cO78Xdy5e4K5do2qFbmb2uINxrxDlQQShnhUVmBr22FqPZyqCOtbDZJJQKOfvOJnElUcfWj4ZRw9cFuHFP0VthtUk8tC4Rg74Q/5Ba9Bj2EBuXLqTjfUNp0qErcZEXWL/kG4Y/NYVfFszh8w2HcdQ63/zGgnCPutE5UJFABaECmC1W1hyI5vuNF0hML8DL1YHRveoxpld9tI6q637+dhOpxSZxPl3Pj6fTidYZadCyHWOmzODD5x6h30NPkJedScK1SLwDggitG8GGn75h1HOv0nng8LJ6i4JQLYlCCoJgZyqlghFd6zDiHxWKysKhhDzmH0ml0GzD09ef52e/SdveA5HJZEz/ajkrv/iQqPOn8A0O5fyRfSRcu8zjr31Aiy69yjwWQbiXiBGoIFRSNxuFZhSamb0/iWvZBgBade/HM7Pm/etB/+z0FLLTUtG6uRMQUrNc4hWE6kiMQIUq5eD5ZN5YdIh9Z5Pw99Ty9ODGvDy6FQrFvbVp/L9aogGYrTae2xiNwSLhExjMlLnfEtag0XU/9xdP3wA8fe+Npti3QpIkrp07xekDO7FYLCRFXeH80X0oFEra9RnE6OdfxdnN4+Y3Eu5p99YnklDpnYvOZOiM9TzevyF5m55h05whbD+RwEtf77N3aHaVkm9iZ0wOpj+PwRxLLsCm1ODp68+8dQdvmDyF0iRJYvHsGXzx+iRsVisFOdlcOnmYR6bN5ONVe1AoFcx+7mFstorpmCNUXWIKV6hUnvhwG/VDPHhp9N8zJpm5RdQds5hryx/D0/XeKz1XUGTCc8BXWCVQyKCWhwNXsw34h9Ri/IzZRLRqb+8Qq5Rzh/fy45y3mPXTupIdyAnXLvPO+BF8tu4gjs4uvD6mP6Oee0Uc8xFuOIUrRqBCpXI5PpsOjQKB4i4mGw7FkK83EebvSmzqvVmKsMho4a9a81YJonLNjH7+Neb8uo2IVu3Jzcrg4ObV7F33KwnXLts32Crg6M5NdB82utTxnRq161O3SSvOHt6LTCajXrPWJMdcs2OUQlUg1kCF61yIyeLElXTC/F3p3CSwQsuY1Q/xZM/pBD7//TR7zybSvI4vxyPTMJttBHrdm10w5DIZrp5e5GVnMf2r5dSs3whnNw8MRXoWvTedY7u20LBVe9QOjqz84kP8gkN58s25BITWsnfolZJcrsBmub7pucViRqFQIEkSkaeP0Ux0qRFuQoxA71HrDkbT7YVfCRz2LX2m/cGuUwmYLVbGzNpE35dWsfVYHJM+20Wbp1aQml1YYXFNfqA5H608icVqI3bF42z8cAhxK5+gbYQ/X605W2FxVCZebo488fpsoLgMn7ObB1aLhY8nP47ZbKbPqEdIjL7CpROHaNqhO43bd2XWkw+QkZxo58grp/Z97mf7bz+Rp8sueezq2ZPEXDpLjToN+GH268gVChq362LHKIWqQIxA70G/7LrCy1/v49PnutK2gT+7TiUyZtZmBnWoSWaegWvLx+GgViJJEq9/d5BnPt3FH7MGVkhsjWp6IUnw6XNd0KiL/zwdNUo+fa4rfab9wduPi/U+gBN7tmIs0mOzmIm5eI5n3/scrYsbO/9Yzp61K+nQbzBrf1jAE69/YO9QK536LdrSacAwXh7Rg9Y97qMgL4fT+3ZgNpt5dVRvWnTpzatf/IRcLsYXwo2JBHqPkSSJtxcfZvH0vnRrVlxkfEzv+qiUcp7/fDe/vzMQhz8Tl0wm4/WxbQgc/i25BcbrGkWXT3xQaDDj5+FU6nF/TydyCozl/vqVUU6Bkd+//qTUY7vXrKR1z35sWraIz9YfRKUu/m8z+oXXyEhOwMnZlV2rV/DISzNLrt1M/NVL7FmzksyUJLRu7nToO5iGbTpWy04kDzw9jU79h3H6wC7OH9mPxklL74EjsJjNHNi0ikNb19H7gUfsHaZQyYkEeo8xmq1EJefStWlQqcd7twrBYLTg5qwu9biDWoFSIcdovn7NqDzI5TK6NQ/mp22XGT/g76MZP225RO9W92Y3GbPFSvzVS6Uey0pNwma1Ub9F2+sSZON2Xbhy5jgqlRp9fh5uXjduY2ezWvnhwxmc2L2VTgOGUathUyxmM0s+egsXd0+mfvIdWhe3Mn9f9hYQWgtjkZ4NS75mzq/bcfXwAqDfQ4/z+pj+tOzaW5ydFW5IzFHcYzQqBT7ujpyPySr1+Mkr6ahVChauPVfq8V92X6VOsDu+/xgRlqcPn+rEjO8OMu3Lvfy+5yqT5+9h9vLjvPuEmL79i4u7JwqlitjL5687rxh7+Xxx82xDEQ5ONy8U/8fCeSRGXaHroJHs/ONnjmzfyOafFxFSNwL/kJrMf/XZ8nobdnds12Y6DRhekjwBfINCaNm1Nyf2bLNjZEJVIBLoPUYmkzH1gRY8Pnsbl+OLN1GcuprO+Dnbefux9uw4mcDQGev4eu1ZJszdznPzdvLZpG4VGmPzOr4c+epB1CoFy7ZH4uKk4ug3DxIR5nXzJ98jOt43lIvHD+Hq6c1PH82kMD8Xq8XC3nW/cmTHRpRqNc079UDjeOM2aYYiPVt/WUyLLr05vX8nc37dxvvLNzJ/4xFsVisyZCTFXCXm0rkb3qeqksnkWC2W6x63WCzI5Qo7RCRUJWIK9x40+YHmWG0SPSb/jsFkQa1SUD/Ek6eHNOHhvvVZti2SE1fSSMoqYHjXOrSN8K/wGEP9XXn/yY4V/rpVRYd+Q9jw00Ja97yPjOQEnuvXBiSJsPqNGPHUVH796iOmf7X8pve5cPQAofUacWT7eka/8FrJlKWDk5Zxr8xiyuDO9Bj2EMd2bqJmg8bl/bYqXLveA3hv4mj6PfQEXn7F7z0x+gqn9+/i4alv3vT5kiRxYOMqNq/4Hl16KuGNmjHsycmE1ReVoe4FIoHeg2QyGdMebMmUB5qTW2giKimHUW9vxGS24uyo5qlBjZGkRvSY8jsd/yxqINiPQi7HOyCIzJSkksc0jo689vVy5r00kfycbPqMfBSlWs3lk0f449t5PD/7S8LqNbzpvY1Fhbi4e5CWEItvUEipay7unigUCjSOTujz7FPE4vjuLaz69jPir17GP6Qm9z86kS73P1DqZ5Jjozi6YyNIEq169CO4Vt1bvn9weD3uf/Rppo/uS+vu/TCbTJzat51xr8y66doxwMZl37J79UrGTn2D4Fp1Obl3Gx88M4YZC1dSo3b9236/QtUiEug9TKGQ4+nqgKerP63r+3P/a2uZPqY1jmol8347xckr6Ry+kMIPmy7w8TNdaFnPz94h35M8XR14eNpMPn3xydKP+wbw9uLVXD17kjMHd2Exm+n9wCO06t4XpUr9H3crLSA0nKjzp6nXrDWHt61nyBOTSq5dPHYQF3dPEqOu0LB1hzJ9T/9FkiR0GakolCqizp9m8ewZjH9jDhGt2hN1/jTfvfsKVouF7kOLS4puWr6INYvm06HfEGRyGe899SD3jRnPoHHP3PJr9h/7JK179OPEnm0olEoeeuE13L19b/o8k9HA2u8XMHPx6pION71HPorRUMS6xV/x6CvvsHHpd5w7vAe5QknXQQ/QecAIlKrr+78KVZNIoAIAS2f05Ys/zvDK1/uIT8unhq8LZxaNIcjHmWXbIxnw6hqOLxxNsI+LvUOtUqKTc4lM0KF1UNEuwh+1qmzX1WQyGXWbtqRu05Z39Pyw+o1wdnMnrEEj1v/4NQZ9IU07dCXuyiXWfP8FQ598gV8WzGXCG3PKNO5/E33xLIvee5XMlCSsVisqtZoRE6fSrGN3ABq0bMcz737G5688Q7chD5KeFM+q7z7jg+Wb8fIvnikZ+MhEpo/uR8suvQmqdeu9V30Ca9Bv9OO3FW9WWgqOzs7XtYdr0q4ru1YtZ9qw7tRu3JxO/Ydx6cRhln78Djt+W8qb3/2G2uHGa9NC1SASqAAUN3yeMrIFXZsF88BbGzj05aiS9mHj+kVw6ko6izZc4K1x7Soknktx2cz/4zRXEnREhHkxaVhT6gRXnfZSF2OzmLJgL2euZdC8ji/ZeUUkZhQwdWQLpo5scVtnK3X5BlZ8Xj4FEWQyGY+/9j5zJj3KgIcnkJWWwi8L5uIdEMygx55l3eIvGTtlBk4uruXy+lCciNYt/pK9a3/l8dfep8N9Q7BZLWz46VvW//g1XQeNKhm1hTdsRk5mOiaDgRO7t9K254CS5Ang4eNHh76DObZr820l0Dvh7uVDQW4uudmZuHl6lzweG3kei9lMn1HjGDr+eQD6jBrH6kXz2bXqZ3b8voz7xowv19iEiiESqFBKVHIuzev4XNd7s2U9P7Ydj6uQGA5dSGHI6+uYNKwpw7rUZt+ZJDpP+pVNc4bQvM7Np9bsLTJeR6+pf/DskCbUDXbnwLlkfDycGN2zPit2RpKcWcjHz956mTiL1UZKXPR/Xk9LiOXsoT1YLBZq1m9EveZt/jVBW8xmDm1Zy65Vy0lNiMVR60zb3gPpNWIsry5Yyor5s4mLvIB/SE0iTx8jMSqSR156m9bd+wGQm53Jyb3bMRQW4B9Skybtu6JQFlesOrR1LVtX/Eh6cjyS1YaTiwudBgxn8OPPXVfRJycznfVLvubCsYNoHB1JirpCcO0GtOzeh04DhgEgl6sZ/PizHN+9hTMHd9Oya2+geJTq5umN2sEBmVyO1Xr9DlqbzYpcUf47aB21znQZOIKv3pjM+Bkf4uUfyOWTh1n5xYfk67LpO2pcqZ/vM/JR/lj4KSf3bhcJtJoQCVQopUktb174fDcGk6WkIhHAjhPxNA2/+aaKsvDawgN8/GwXxvYu3oTRq2UIgd7OzFh0kA2zh1RIDHfj1YX7eer+xixcf54HutXhm2k9iUvL541FBxndsx5frznHEwMa3vWxnIK8HL59+yUunzpKy669UTs4sv23n1AqlTw967NSm4hMhiLmTn4cq9lEi259MBUVoVJryEhO4PUxA3jpsx+Y/uUyMpITyUpLxtnNnaCadZDJZFjMJn76+B0ObFpN0w5dcfX05tDWdXz/wWs8Pv19Yi6d48DmVejz83FyccVQWECN2vXZ9suP7N/wO7OWrCsZwebpsnnrsaG07NKbJ17/AF16Kivmz0aSrITUibjuPQbVDOfc4b0079SDqAtnWPj2NAY9/iwymYzW3fux6tt5DBr3DP5/TqNmJCdycMsa3lm85q5+t7fqocmv8+uXc3n1wb4gSbh6evHIS2+z8J2XKNIXlhq5F+kLUarUqDX3Xku+6kr0AxWuM2bWJnIKjbz7RAe8XB34fuMFftxyiRMLHyr3fpw2m4Sm93z0W55Fpfx7FJFbYCT4ge/I31S5D/WnZhfSaNxPdGociL+nlq9f7FlyLT4tj0bjfuKJ/g2Ry+W3PArNyNHjP/RbAN7/eTNh9RpiMZt454kRhNSNIE+XyYWjB7GYTYTUjaBVtz5sWr6It77/o2R97se5b5GTkYZMJiPqwhmaduxGSlwMaQmxDHh4AusWf8mna/ejUmsoyNVR+GcFIwdHJ76c8QL5uTrqNG7OnrW/kpOVTt0mrWjT8z5++/oTLGYTPYaN4cKxg/gEBjPpgwUoVSpsNhvfvDUVgKdnzQPgj2/nkZmcyIS3Pip5fzmZ6Uwd2pUa4XWZ+cPqktGzxWxi8qBOKFVqMpLi8Q0OZdC4Z+gyaCRmowGNoxO716xk+bx3adWtL3KFgmM7NzFi4ov0+cfo727YrFbOHNzNtfOn8PILpF2f+3FyLr0XwGI2YdAXonV1RyaT8cPsGRTm5fLMu58hl8ux2WwsfHsa5w7vZczUN+jQd3CZxSeUrxv1AxUjUOE6P7zahzk/H2fUzI3kF5no1yaM3fNGVEgza5kMPJw1xKflEx7kXvJ4XFoe3m4333hhMltJySrEx90RJ4eK3+0Yk5KHr7sThy6ksOyN+0pdC/FzJdTflQ2HY+569Hlk+wYkJI7v2kydpi15f/lGtK5ubP9tKau++5xmnbqzZtF8Jr79CUWFBexb/zsDHn6SC0cP8tEfu0rK/237dQn71v1GUK06rPruc6IvnOHquZNoXdzQ5+fRoGU7oi6cpkO/wZw9tJcpHy3EP6QmR3duYtkns2jXewBXz54kMSqS7PQUnn33s5L1SrlczkNTZvDCwPYMe2oqXn4BXDt3ip7Dx5R6L+7evoTVa0huVgZfvTGZvqMfw2QwsPr7Lwip04Cn3voIV09vJEli3eIvea5fa4oK8/HyC2TExBeZvXIbx3dtRpIkho5/Ae+AIKwWC9npqbi4e+DgdOdt8AxFeuZMegRjkZ7mnXpy9tAefvv6Y15dsJSQOg1Kfk6pUuPs9vfO51HPvcLsZ8cyqX9b6jVrzdWzJygqyKd9v8G07zPojuO5FTmZ6cjk8lLrskL5EAlUuI5apWDGI22Z8UjbCn9tmUzGhPsb8/z83fz8xn24ajXkFBh5ccFeJgz878PpkiTx+e+n+WDZMTQqBYUGMxPub8ysx9tft55bnty0amJSc2lW24dLsVn0avn32UqjyUJyZiEqhYwi4/Vrd/9FqZATEBZOSmxUyWO7Vq8gNT4WpUrF5DnfoFAW/1Me8sQkkqKvcnz3FiRJ4vHX3if64llqhNfl7ME9DBn/fKnauT2HjeGPhfNo2LoDG35ayIinphBUqw669FQCa9Xh1N7tFBUWsHv1Cj5etafkbGSXgSPIzcrk2rkTZCQnElqvISZD0XWbjRy1zlhMJmY+NgSAwNBwkmOiaNn175+xWiykxsfw4ieLOLpzIx9PeQLJZsPJxZXc7EymDe9OUM06hNVvRPTFs7zx7a8EhNbi8skjfPnGCzw+/X36PvhYqd/Nr1/ORS6XY9Dr6TxwOGOmzLjloz3/b+NPC3H18OL5hb+UrOXuWvUzX705hfeXb7purdlmsxFz6Sz5OTqefONDCvJyObZjI536D6Pz/Q8QWI49WhOuXea7d18lOfYakgShdRswfsaHoi9sORKl/IRK581H2xLk7Uyt0T/Q6blfCB/9A/VDPXlp9L/OogCwZMslvlt/nt2fjSDulyc4vWgMhy+k8N7SoxUYOaiUcmw2CbkM3lt6jGOXUwEoKDIx+Ys9SJKESqngNlZO8HBx4MFJrwJw+sAuPnhmDJdPHsVsMlIroklJ8vxLROsOBIfXw2a1os/PQ7LZUCiVWC2W65KITC5HoVBybNcWQutGsH7JN6jUGlp260NafCzpSQnUadwCZzeP6woLNGjRlqy0FFQaDQZ9IQqFkp1/lK5+tGv1Cuq3bMdX204y/ctlJMdFsX7J11w5cxwAY1ERP338NkZDET/OeZPdq1cS3qg5CpWKPqMe4+vtp/l6+2k6DRjG7jUrGT9jNoFh4chkMhq0bMfYqW+y4aeFJa93Ys9WVi+az8vzl/DF5mN8vHoP6UkJLPv03Vv/hf+fYzs30X/sk6U2QnUdNJK0hFheHtGT+Ct/F/k/tGUtLw3vzldvTGHTsu+Y+/w4ls97l9Y97mPUc6+Ua/LUF+Qz+9mxdB008s/f2Sla97iPD54Zg8loKLfXvdeJEahwHUmSOHQhhdjUPFrW9aNeSMUeH1GrFCyc1ou3H2tPdHLuLRWzn//HaeZN6kr9EE8Agn1cWPhSLzo++wuvj21TYaPQzFwDoX6uRCXnMX5AI4a/sQG1So4u30ivliEEezujKzTioLm9XaIGfXFT86PbN9K+7yAunziCl38gURfOYLOW3nV6+cRhFEolFrMJhVJFaL2GxEZepO+ocWxd8QMNWrYrSQjHdm5CrpDjoHUmKzWJCW/NpWXXPkBxvd0lc2cSeeYYusx08nRZpYquXzlzHFcPL6IvnuXckX0YCgtY/f18rp0/RZse93Hx+CHOH9nHa1//DEBInQY8/tr7LJnzJl9Mfw6ZXE5hfh4Rrdrx3rKN5GZl4OUXyOWTR7CazfR9cFzJa/Ua8TBHd2wk+uLZUlOntSKakpoQU/L/N//8A6Ofn16ygcrN05sJb83lxSFdGTXpVRwcb68pguzP9cv/J0kSSpWKHsPHMOf5R/hk9V4ObFrNqu8+Z8DDE7h4cCfpCTE0aNqc0IYt+OzliUz6YAEN25Rfacoj29ZTu3ELegx7qOSxfqMf59S+HRzfvUWsuZYTkUCFUtJ1eoa8vo7cQiONa3kz7ct99G0Tyrcv9UJZgVOhAAFeWgK8bm39Kj4tn0Y1S68rhge6UWgwU2Sy4Ox4+9N3dyLIW0ue3kREqCcWq43oFY8RlZSLp6sGmUxGzVHf8+TARrfVHi47z8Cid4tHoM5Oanb//CW9WwZyPDIFmUXGgjdeYMzk13FycWPXH8s5tX8nPkHBqNRqbFYLbl4+tOrWh9zsTPJyspk5bggtu/UhJS6K0/t34eUfSOalczg4aWnRpXep1+4+9EH2rvsV74AgPp02gfGvz8avRhjHd21mzQ9fYDYakcvlKGwWPJ01WG02ok/uJ+7CSfIL9Hyx+ShyuYKfPprJib3bkCGjIC+Xr7adRJeRhpOLa8lanX+NMACiL50lsGbt634P4Q2bE33hDG17DeDE7q0UFeajz88j9P9272alJlEjvF6p57l5euPg5ERBju62E2jbXgNY/+NX1GncomSkv+P3pQTVrMN9Dz3B2YO7ObhlLcs/e48Hnp7G6i9nM6a+C+H1HTiddpRVB3Yx9OmX+WH2DOb+vrPceqtmpib9a+nAkDr1yUpJLpfXFEQCFf7hmU930rlJELOfKm6kXGS0cP/0Ncz//TRTRrawd3j/qVV9PzYcjuXx/n8f3dh1KpFQPxe0/7eZ6K/R9Y4TCbg7axjVo26ZtmoL8XOlSbg3vVqG8MkvJwnw0jKqR12iknOZMn8P4wc2YvX+KH55e0DJc3T5BpZtu8yluGycHdUM6RxOuwj/kg/bqKQcTEYDKqWcOi4FnFg+FqVCjtVqY+x7WzgWdYmXhvfAZDTQpH03atSuT/SFMyhVarSuxX08x774Ju9NGIV/SE3CGzUnNzsDvxphtO5xHzt+X0po/YakJ8TzeKf6xecb73+AYRMmk6fLRuPoSEGOjtA6Ebw38UFyszKo06QljVp34vD29TipVXhqJNo3Deblh1oD8N7So2w+GkdSzFV+/PBN6jZrxbRPv8dkNPLLgjnMf/VZpn32w3UJRZIkTu7ZhlKl5qEXXisZWdtsNk7s3kJOdiaHtq6lXrPWOLu5c3THZlr36IskSchkMmo1bMqp/TtKFVGIu3IRSZLw8CldijLm0jlO7d+BxtGJ9n3u/9fen/eNGc/lk0eYNrwbLbr0JjnmGknRV3llwVIAfINDOXtwNxGtOrB16ddMbeVJU//iL33hng5oFDJOHdyBTC4n8tRR6rcon30FtSKasuq7zxg2YXLJ7ILNauX0gd2MmTKD5fPeY9+G3zEW6WnWsTujJr2KX/C92V+3LIljLEKJ3AIjISMXkfz7k2gd/046+84mMXn+Hk58+9ANnl0xDCYLv+y6yrHLqYT4uvBw3wb4e2o5fDGFwa+t461H29KjZQ2OR6YzfeF+5j3XleFdiz9MbTaJJ+Zs48C5ZEZ0q0NKViHrD8WwbEY/+rQuuw+TY5dTGfjqWl5/uDV7zySx50wSvu6OjOlVnwPnk/FwcWDpjOLiBIs2nOflr/fTt3UIEWFeFBkt/LbnKkHezvzy9gC83RyZu+I4r35zAAe1gjPfj6X2/+1OTsoooPaYH3nxsyVoXd3Y8dtSDm9bT+N2nfHw8ePRl98p+dmiwgK2/fIjO/5YTmZKImq1BjcvH4Jr1ePC8QMMffIFegwdTX6Ojp8/fx8kyEpLxq9GGMd2bUbj4Ejr7v1wcfck+uJZrpw5jlKtxgELob5OHP9uLHK5rOR33XDcT8RlGglv2IzXv1lRkiytFgvThnXjmXc/p06T0l/KYiMvMG/aBHyCauCkdWHAIxORy+WsXfwl186fwmIyMmn2lzRu2xmA/Bwdbz46iMdfe5/GbTuTGBXJuxNGMfDRiTTv1JOEqEh+/vwDhjwxie5DHgSKk/TSj9/m6M5NtO8ziML8PI7t3MgTr8+mba8B/JMkSaz9YQFbVy5m9POv0abnfagdHDEZDUwb1p36zdvgGxzCph+/5OehNUt9KcjUm5m6K4PGnXvTsHUHug4aWQZ/YdezWa3MevIBPHz86D92AjablXU/foXJUFR89tTBkVHPvoyTixs7/1jG9t9+4oOft+DiXnWqe9mLOMYi3BKj2YpSIcdRU/rPwsNFQ0GRyU5R/U2Xb6DX1D/wcnNgYLuanI/NotkTy1j7/iDaRQSw4cPBzPn5OJ//cZrwIDcWv9qHnv+3C/a3PVc5H53Fme/HlrzHfWeTGDVzI7ErHy+zOrWt6/vz+6wBTPx4JzJgSKdwsvIMfLTyBI/2i2DOxE4ArN4fxawlR3hvfAfm/naebWcyMBmNDOpUGy+tnEHT17L/i5GYzcVrcFarhIezptRrebhosFptLJjxPJJko16zNvQe9SgHN67i7R9Xl/pZR60zgx57lkGPPYvVYsFkNPBcv9bUqFsfrZtbSQF2ZzcPJn2wgKd7tSCwZm0uHjtIn5GP0n/sBI7v2oJBX0DvkY+gUCq4eOwQTo4q+rWrWZI8AeRyGYM71eKTX8/QvHOPUklFoVTSsE0nYiPPX5dAjUV6tK5uvDTvBzYsXcj3709HkiSCw+vi5umDxsGhJHkCuLh70PfBxzi0ZS2N23YmOLweMxauZM33C9i16mc8/QJ5ZNrMkkpGUNzC7dT+nXy4clvJruE+ox7l3QkjadyuS6kznlaLhSPb15MYdQWVWsOWlT/g4u6BxWJm/Y9fU7dZKwLDwsnNzkSuUJKpt+Cj/fvLZ1KeCU9vb3KzMnDUll8dablCwStf/MT6Jd/w7ayXkMnktOs9kIZtOvL5q88wb+2BUju1k2Oj2LN2JQMfmVhuMd0LRAIVSvi4O1IrwI3f9lxlZPe/W0J9u+4897WteYNnVoy5K07QrLYP373cq+QDuXvzSJ79dBdHv3mQVvX8+GXm9SOIv6zaF8UzQ5uU+oLQuUkQYf6uHDifTPfmNcokTrPFSmGRheljWqHLN6JUynFxVPPdS71KztJKksSsH48w5YEWvLHkFM9+uJD6LdqhL8hj2dw3yM+8hMVqY8uxOJqEF68RqlRyFq6/wPQxf38ZXrjuPFpHNSZDETXC63H51BEsZhNvLvodD5//7uOqUCopSM/BycWVjKR4mnXsUeq6Sq2hTtOWxFw6i8ViZvTz01Gq1KU29sRcOkfkmRMUmi0c/3O38f87FpmJJElcOXOCAQ+XvhZ35QKtuve97jm1IpqQlZZCYvRVho5/gaHjX8Bms/HR5Meo3agZqfEx1z1HpVJjMZtL/n9weD2efe/z/3zvR3dupOfwMaWO3ITWjaBOk5acO7y3ZBRqs1r5dNqT5Ofo6DZ4FKH1GrJp2bcsnvMm3gHBdB44gm6DR5GaEMs740fQZdADLDiwjqmtPXHVKEkrMLHoXB4dxzzHmh++YMpHC/8rpDLh4KRlxMSpjJg4teSxA5tWUadJy+t2ajdo2Y7IUxW7Q706EglUKCGTyfj8hW4MnbGOg+dTaFzLi81H4zgXncmez0bYOzzWHYhm8fQ+pUYzo7rXZfL8PSRnFhLk43zD58tkxVOL/2T7c/2sLHyz9hyzlhyhpr8rNfxcuBSbjd5oZu7TXUoVoriamENGThHHr+kYMG4SDVq2B0Dr4sa41z5kcv9WTB3WkBU7Ipk3qSsqtQa9wcgnqy5yKSGPbk382H8hg/VH4un/+GS2rPiBAY88RVj9Rtf19fwvzm7uFObl4uUfxNWzJ0rq0EJx3dzoC2do2a0Puoy0fz1D2XP4GDYt/46aEc04dOkkc1ccZ9KwZkDxruiTVzMJCKnFhWMH2PnHcrrc/0DxyG3xVxiL9DRpd30lJpVaw7hXZjFn0iN0Hzoa74BgDm1Zg9Vi4YGZLzJteHfirlwktG7xxiGT0cCWlYtp0qErJkPRLXU5kSH7978Dq7XU38Hx3VvJycpg5verSopDdLxvMC+N6MVbi34vaXkWGBZORKv26LKz8W7Vm6c2rcHdSUOewUzf0eM5tnsLfR98/K4KOtypgNBa/7pT++rZEwSGhVd4PNWNOAcqlNK+YQDHvhmNh4uG/eeS6dYsmKNfP4ifZ8X/4/8nB7WSQkPpAgQmixWzxYpGffPp1xFd67Bg1RkKi/4erew4Ec+1pBze/uEQYaMWMeT1tXzxx2kWrDrDzpMJ//pB+1/m/XqSz347xfQxrVCrFBw8n0yInwvPD2/O05/sYPX+vwsh6PKNBHg5EZeuJ/gfuyfVDo4EBAWjUsrRFRjxcHFg5HMvA9C4Ux+KQnuy/JKW/KCudBn6CBt+Wkj7voP4/ZtPeGl4Dyb2as7iD9/419Ha/3PUOtOkQzfUDg4c3bmJbb8uwWQ0kJ2ewtczpxLeqBmpcdF0GfjvX558AmswZPwLXDlznHptuvHFhhg87/8Gr0EL+WztNRQOLhTk6njhw6/Zv+F3JnRvzNO9mhN35QKvLlj6nwXf2/UeyIxvf8FqsXDt3Em6DhrFa1//jKunN4+/9gHvPfUgP3zwOr8v/JRpw4tbncVfucik/u3Y9uuSm/53att7IDt+/4mCXF3JY9EXzxB98QxN2v9d4eHsoT10HjC8VP9ODx9/GrXpyIVjB0rd86m3PsZsMnLq8AF6PzieTg8+RadBD7L1t6XUimjCsAmTbxpXeajZoAl+waEsfOcldBlpGIuK2Pzz95zev5Oug0fZJabqRIxAy8GBc8nM/OEQRy6lEeit5bmhTXl2aNNy28Je1mr4ulRI2zJJkvjijzMsWH2G5MxC2jX05+3H2tO+4fW7IQEe6lWPd386wroGg9D8Wej+k19O0jYi4JbK/A3tHM7WY3E0HLeEYV1qk5KlZ9ORGNy0Gp4b1oyW9fzYeiyOl7/aR6+WIXy7/hzOjmrWzx6M+z/WHv8pp8DIrCVHmTmuHXNXnChpQL79RDwzvjvIm4+2ZdqXe7m/fU0UCjkhfi5EJecytm8NzuzfRuP/G43pMlJJjIslp4kTNf2Lpxn/GlXm52RzdPsGVBoHTIbtNO/cEw9vP5Jjo+ly/wMU5uUV75rNy2HmY0OZPPebG+78HDb+Bd6b+CDDn5rCyT3b+HHOm2gcnejUfyhaF1eiLpyhdY9+//07fWISWhdXln/6LmazqbjKkUxGkQVUKhVT5/9IaN0IGrfrTEFeDgqFEkftjWcKzCYj8VcvoXF0onbj5rTo0rskibXrPZDYy+fZs3YlNWrXRy2HjMQYtBo5w54qLmAgk8noNeLh/7x/g5btaNtrIC8/0Iu2vQZQmJ/HqX3beWrmx6VGiY5aLfm67Ouen6fLum49U+PoyIufLCL64hkObFpNdkYqnr7+vL14dUmhe3uQyWRM+WghKz7/gBeHdsVsMtKkXVde/3qFKPVXBsQu3DJ28ko6/V9ZzdynOzOoQy0ux+t4/vNdDGxfizcerfjSeBXJaLKQkVuEr7vTLW3ImfnDIbYcjePzF7pRN9iDNQeieemrfWz9aChNa1/f+cVssfLI+1s5eD6ZXi1DuBCbRWGRmY1zhlDD99Y3aJy8ks72E/G4atXM+fk4y9+4j3YRfyftRRvOs3p/FGvfH8TEj3eiUMj4ckqPG9wRvl57lj2nEzkRmc4Pr/ahY+O/e1Qu3nyRX3ZdIUNXxAdPdSwp7zfotbW0qe/Hl+su0XHII7TrN4SM5ER+/3wWI9p6sWzLBdZ9MIhgH2fCH/uNfF027/+8mcCwcArzctG6uPLj3JlYrRZ06ankZKbToksvUuKiuXLmOMOefIHfvvmUz9cfvOHUZuTpY3z5xmScXd1o2LojhiI9x3ZuolaDJjw961Oc3W6+U9NmtXJk+wYunzyCSuNARKv2NOvY/bbbimWnp/DeUw/i6RdInSYtuHziMIX5ebz+zQpcPbzQF+Tz/ID2DH9qCruXzefbqZ1pFxHAvrNJTPh0H33GTeWPhZ8yf+Phm07nJkZFcmr/LhwcHWnTawAqtYYDG/8gOTaKoFp1CK0TwSfTnuTN734rKYd3bNdmFs+ewbx1B0qVRKwKJElCkqTr2ssJN3ajXbgigZaxh2Zton1EAJOGNyt5LD4tjxZPLif+lyfsUuC8vNlsEu/9dJTP/ziNRqXAYrXx8uhWTHmg+X+OuguKTISN+p4zi8aWWrv89JeTnLqWwZLXrt9g8pfT1zLYfToRtVLOmF71cHO+syL32XkGaj/0A9nrny71eFJGAW0m/kzS70+SlFFA48d+uu5n/mn6wv2olXLm/XaanA1Pl3rfKVmFtHhyGfe1DaNT46CSs6oXYrLoOfV3pjzQgsjkAvacScHH3ZGRnWqw5kAUEWFefDW1J6euptNqQnE1n7+6sUBxO7PJ93ei3+jHiLl4jhc/XVSSsPau/43Ny7/Hzcubtr0G0O0m03U2q5Wzh/cSF3kBpVpNsw7dy70h9b/5/JWn8Q+txchnXgKKP/SXfDQTs9HI+Bmz2b1mJaf27SAt+gI/Tm5Dl6ZBJc/deDiGF5dcQePmS/eho2nXe+Atv25aYhzvTRhFeOPm1G3SksjTx4i5dI6eIx5m7Q8LCG/YFH1BHjmZ6Uye+w3hDZuV9VsXKqkbJVDxVaSMXYzNovP//aOG4sP1Xq4OJKQX2Cmq8vXxyhNsOhLL0a8fJPG38ez57AGWbLnIoo0X/vM58Wn5+Lo7Xbfxp0vTIC7EZP3n84wmCwtWnWHWj0dYuO4c4Q8t5oOlR7mdL4J/cdWqUSrkRCfnlnr89LUMwv6cOnXUKDFZbP/29FI8XRxIzynCQa0k6h/3OxOVQaifK4kZBXi4/D1qaVjTiy1zh7LhUDTbj0TRrKYLjnILs5cfp1fLUBZM7sGa/VG0nbii+DX8Akuq9QDEXDxLWL0Izh7cQ/+HJ5Qa7XXqP4yczDTqNWt9S7st5QoFzTp2Z/DjzzFg7AS7JE+bzcbx3VvpPuRBLhw9QGL0FWQyGQMfmciR7esByMlIwz+kJvGx8XRuEljq+V2bBhMXFU1gWDi69Ot3Bd/Iz5+9T88RY3nhw6+4b8x4Js/9hi73P0Bc5Hk+X3+QPqMeZdSzrzBv7QGRPIUSIoGWsTrBHhy+mFLqsZSsQjJzDQT52H8jTlmTJInPfj/Nty/1omZAcdWbeiEefDG5O5/9euo/nxfs40yaTk9admGpx49cTKVujf+eMnzlmwNk5Rn+1959x1VZ/n8cfx0OewsooiggTkBFVBDIiXtvXLmtTLPUflZmNswyK1dlzkzNrbjKvcDUzL1woKiITAFlz3P//qAw8ph6vihH/Tz/43Cf+3xuH3je5zr3dX0uIlcN4fTiAZxc2I+1+yP4eUf4A8dGxtxj7f4rHLkQqzVgDdUGjOpWl6Ff7eJGXCoApyISGPt9KO/0KlyfOGfDKboEProJeO/m1VkfepXXO3kxdNr98525msg734XSN6gGpyMSafOvhg11q5Yl7LvebJ/elX5BNZjQrwE31w7jo0F+GBio2HzoGhpUVPXyZsbGA8VncqpUhdelUqGtO72iKKDw3Nx7h8JaP+jblg0LZjJt1ACmjOhFWkpS0TVY2zmQeDuKis4VOHqxeEj+ER6Ls0tlEqKjsLF//Pt7iqJwMmwPrXoPKvZ46+DBnAwtXCtav2lrvPxeeWA5iHi5yV9DCRsf7EO3SVtxsrOgo78bV6Lv8uaMfbzWyeuZ9WN9lvILNMSnZFLLxa7Y47XdHLgZn/bQ51lbmDC0vSf9puxg7tgWVK1ow9Yj15my/Cibp3ZGo1HYf+oWV26l4OFqT5O6FcnJK2DZznDClw3E5q9JPZUdrZk+8hUmLTrMkHaFX20WFGgYPXs/IWHXaFynAuE3krGxMGbj1E6U/9ds4kmv+qIoCg1fX4WKwmYSVSvaEhWfSrdJWzkXeYf9s/57CU9qRg57TkTh6mhFSNhVWjWsjO/rq1CpVJgYqRnToy7Ldl5kQt8GD/0K39PNHg9XO7Jy8jH9x4xiF0drjI1NKOfs8sA9vSoedbh55SLt+g3l12XzqOnjV/QGH7Z1HfaOTlw+/Sf+bZ7u/pMl5UToLmwdyjJ5cQj2jk5oCgpY8/1XzH5vJH5/fR3rG9SOlbO/oNuwUQyavohlE5rjW8uRQ+djGD4jjKb9RrNx4Wze+mruE722oZEROVmZxZoo5GRlYmTyfN3nFM+WBGgJa+ThxNIP2vDR4sP0/Pg3ytqYMbp7Xd77j624nmdGhmpqu9mz6/hN2vq6Fj2+7egNGtZ0fPgTgWmvBfLlimM0e2cdd+5lU796OZZ/2Bb3CjYEjlpDdm4BjTzLM3fTWextTFnyfuEaUMd/9a6t5lyG2KTMop8XbD3HhetJXFs1GEszYxRFYdKiw7z2zV62fFE8TNRqAz4Z4s/EAb6kpGVjaqxm7YGrXLqZTAd/N5Z/2EbrB58bcams2H2JncducuJyPE3qViTIpxIhB6/y3YYz1HZ3oEtgFaIS0vhyxXHe6l6X8cHaewnfSkjj2zUnWb7rItm5+RgbqRnQqibvBtfn48GNWLA/QevzLKxs8G/diYToKFCpeD+4NfUaBxF78xqR4WfpNnwMIQtnP/UF/CVl/8ZVBI9+D3vHwgldBmo1PUeOZ/f65bTsORAo7JLUtu9QDu/cSkC34fT8YglxsXFUdHamed9R7N+0hi5DRz9R03iVSkVA265smD+TYR9+iUqlQqPRsH7etwS07fo0LlW8IGQS0VOUX6BBbaB6rr5C08WWQ5G8OXMfX74WSCOP8hw4fZuPFh9mzcftaert/MjnK4pCgUYp2u1l+PTdmJkYMmdMs7/ezBRGfL0HSzMjdh+P4vt3mtPC537XoBlrT3LsUjyrJrcDwH/kaqYMDyi2mXVmdh4Vey4iYsXgx1ry8jAFBRrG/RDGyr2X8XS148KNZGq5lOHstSQys/NwLGNOcIvqhIRFUMbKjP6tatKvZY0HRr5/u3gzmVbjQ+jdrBplbc2IvZNOeQdLklOzWbX3Mju/6UabTw5TxaMOo7/4/oHnZ2WkM23UAMwsLPHya0xWRhrGpmakJt3h0PaNjP12ITW8G+p8vc/S5EFd6DvmA2rVL76E6t3uzXhr2tyi5gmKohCyYBbbVizC0dkFtVpNQUE+8dFRdBk6ik6DRj7x/7mMtHt8PWYwGWmpVK9bn8unjmFtZ8+7s5YUG5WKl4/0wi0lz3r7r9LSObAKlmZGfLPmBJ8tPYqnqz0hUzoS4FXh0U+mcARgqC58w1MUhTX7rxC5akjRm6CBgYqPBvnh98ZqFv5fS16duoNJA/3wrlqWXcduMm/LWfZ826PofOlZedhZFZ+Za2ZiiJmxIZnZefA/BOjEhYc5cTmeDo1cWR96FWNDAy5HpbB5aiea1K3Iucg7DPtqD0PbezFr/Sl6N6/20PBUFIWBU3cypoc3P248TSULQ2qUMWHX2VtEpuYyrrcPrcaFcDdLoYpHHa3nMLOw5MP5qwndspZ9IStJiL6JmYUlfq068tmyLY/dlUgfeDYM4ND2TcUC9PrFc2Smp1HR7f6kJpVKxSsdurN/4yocnCpQs54fF44dIiP1HoFtu+r0gdXCyoaPfwrh0smjxFy/SuMOPahRz/eF//Ar/jcSoKJEtPCpVGxUqCtFgbx8DabGxf80zYzV5OTl0zmwCmVtzZiz4TTLdoTjXa0sYXN6Uc35/sSjtn6uLP7tPD7V76/d/PXIdRxszJ5ovei/RcbcY/aGUxipCzsEnVrUj+HT9zC2t0/RSLuOe1mWfdiGoLEb6N6kKmv2XWF8cH2t5zt6MY57GTmcuBRHYHkz+njdn/iyLjyJQ2eiyckrIC8vX+vz/2ZsYkqrXgNp1WugztemD9r1H8EnQ7ry4+SxNGzRlvhbN/ht+QIGjPuoWDcggBUzp9C6z+CiBvjtB4xg7dyvWfP9V7z5+WydXl+lUlGrfqMHRsBCPIwEqNArBgYqnOwt+C7kNBMH+BY9PmfD6aJ7n/6eTg/tVgQwoW99mo5ZT8/Jv9IxoAoXriexbOdF1nzSXucRRUZWHkHjNlDRwZLUjByMVDBk6k4u3b6L77/u9dZysSMrN58K9hYkpGQ99Jx/hsfRxteFn7ddYH7H4t1qOlSzZeDGa1R1tiE16q5ONT8PFEXh6rlT3Lh8HkMjY96ePo9TB/eyL2QltvZlGTdjEVW96hV7jkaj4dTBfbw5pXhQtgkezLhuTRHiWZEAFXpFo1GITkhjwdbznIpIxN/TibAztzkXeYf4lMxHnwAoa2vO0Xl9WLbzIvtP3qJSOSv++DG4aJnN47iXnsOCredYsj2cW4lpGBuqURuAubEhzuZqerhZUqAoRMaqOHA6mn4t7/ezPXklARsLY27EpeJTrdxDX0OtNiA3T4ORoQE5+Qr8Y8JnTr6CsaHBE/Xifd5ERVxk3sfjyEpPw9P3FXKyMlk563Pq+Ddl9BffP/Teo0qlwtDQkJzsrGLLenKyszDS0vReiKfl5bhJJ54bKhVYmhuzY3pXgupXIjoxjY4BbmyY0gEbi8d/c7QyN2ZUt7osndiGz4cHPFF4xidnEDh6LSciEhjewYuxvXwwNFBhZ2VGQU4unzSrhLeTBfUrWDKyngOjZu1n3YEI7qbnsPdEFMGfbmN0N2+2HIqkT1CNh75OC59KbD0cSe9m1VgTnozmrwl9iqKwJjyJnk2rFjbfeAHvw8VFXefLN/sT0LYL1T08OLZ7ExEnDhLUtTdGJqZ8PWZQsS3K/kmlUuHfpjMh82cWre/VaDSELJiJf9suz/IyxEtORqBCr6hUKl5tXYupv/zJkvdbY6g2IDevgH5TtjOwTa0SeQ1FUUhIycTCzEjrEpUR3+ylnZ8LoaducT48Bk8HU+yNIC4lg/buNqj/sXF0PSdLatne5fVv9pCVm4+ZsSET+zdk9b7LjOnhTbkyD19OUcvFjrpVy2JpYUK6kTFjd0dRy96MS8nZONhZUNvegvJ25sTcKyiR69YnmxZ/R9Muvdm9cgGjOlZj9c99SU7N5oPFYeQZV0Kj0XAidFfR3pz/1u+dD5k+ZjDvB7emam0frpwunDU7fuZPz/hKxMtMAlTonS9GBNDn0+24911Cw5qO/BEeh59HeSYP0j654156DlEJabg4WmFt8d8L3/edvMXY70O5fSedvHwNnQOr8N3bzYt2W4mMucfR8Dhcy1lhpxTwUXPnwvumdcryeWg0UfdyHjhnam4BagMVefkaala2Ztqq47wbXJ/3+j167e/P77ei9bsbqexoRdcm1cjJy6eZsZpD52LZevg6Yd/1psG4vY/xr/b8yM3O4ti+7bTrP5zW9crz0auFy2zK21mwbnIrXPv9QvN+IwndsvahAWppU4ZPf95E+PEjxN68RuMO3WXWrHjmJECF3vl7C7Gz1xK5fOsunw31x8PV/oHjCgo0vL/gED9tu0AFewtikjJ4o3NtpgwLwMDgwTfSS1HJ9P1sOz+MbUZZG3OM1AYs2xVO38+2s316VwB+PxdDy/qV2HTwKhP9HYu9Ib/RwJGRv0VyKCqVgEpWKMCea3e5cTcHVCq83Ox5v19DOvi7PnbXKUc7Cw7/EMzy3RdZtuMisckZlC9jzqutaz20icPzLj31LiZm5iRHX+NV7/LFfmdspMbPswKKRvPIfrYqlQrPhgF4Ngx4muUK8VASoEJv1XEvSx33B7c1+9tXq45z4nI8F5cNpFwZc2KTMuj18W/MWn+Kcb0f7Pozb/M5AjzLM3LGfqpVtCU+JRN7a1OiE9O5eDO5WDtCbQ1GrEzUKMDP55NZdCoRlQqsLU0Y0bk2q/deYcsXnansaP3E12lhZsQbnevwRucH13p+sOB3kuJjH7oO9HlkbmVDdmYGZcq7cOTiIQb/Y7vRggINJ6/E06SeIdZ2D35oEkKfyCQioZcUReHA6WjmbznH7+duaw20uZvO8v07zYvuMzrZWzBnTDPmbjqj9ZynriZw9GI8YXN6cXhuMBErBtO7eXU0isL12MIdVAK9nNhz4hbdmlRj0+W7xV5365UUTAwNyM4roHOTqozt24AaLvas3R/BxqmddArPR4lPyaQgX/tkmueVqZk53q+0QGVoRMjvN5i35SzZufnEJ2cw7NsDOFSuxpnDobzSvsejTyZEKZIAFXonJS2bxm+tY8zsA5y4Es/r3+wlaOwG0jJzi47RaBTikjOo5mxb7Lk1KpUh5k4G2uTkFjC2d72ikaaBgYrxwT4Yqg2KJrq6V7TFt5YjpqaGJKHmgwPR/HI2kc8PxRAak8mE/r6YmRiRm6/hctRdejWvztWVQ4ptyC0ereuwt9i9dimdRoxj0eFMbNrPw33AL8Sa1MSpam3S7ybj10r7/U8h9IV8hSv0zoR5v1PH3YEf3mle1At32PTdfLT4CLPeKlwob2CgopGHE5t+j6RXs/tt3kIOXiXQq3iYhZ25zcSFh7gUlYLLv0aJKpWKCvYW8I8B7qL/a0mLsRuo7e5Ag+o1ScvMxcfMiJtxqfy4+Sy7v+2Op5t8vfi/qFytFuNmLOKHD8dgY+9A+4Gvk5udzZ97t1HBrRrvz12BsYluG6UL8axIgAq9oigKq/dd5trK4r1wJw/yo9HINUUBCvD58AD6fLqN2Dvp+Hs5cfBsDF+tPM7GzzsWHXP2WiK9P/mN795uzterjrN810V6NatWdO6o+FTCbyb/Mz+LJvbM33qOxdvOE52Yjr21Kf1b1uTPeX2o4FB8E3ChmxreDZm15XdOH9rPjUvnMbe0ZsKcZVSuXjLLlYR42iRAhV75uxeuuWnxP00LUyNy8oqvh2zm7cxv07owc90plu+6hIerHbu+6UbdqvcnHs1ef5rxvX3o1awaq/Ze4vz1JHp/so0BrWoSm5zBN6tP4FjGDGMjdbFz21iaMKFvAyaU8jZ0dd3LsvbgrVKt4WkyUKvxadISnyYtS7sUIZ6Y3AMVesXAQEX7Rm7M3XS22OM/bDxDpwC3B46vX8ORXya15diCviyd2KZYeAJERKfg99f9yW6Nq+JgbYpfLUfmbz3HwTO3GR9cn/jkLOq6Ozxwbn3wds962Ng/fCayEKL0yAhU6J2v33iFoHEbOHklobAX7tnbnLmayP5ZPZ/4XB6u9oSdiaZJ3Yr0DapBSNhVVu65TN+gGsQkZfDxT0dwdbLm9NVEWjVweQpXI4R4UckIVOgd94q2nFk8gCZ1KhIZc4/WDVw4tai/TluRvdOrHnM2nGbhr+dJz8pjYv+GZOXksz40gjJWJhyZG4zaQKW3DQvG/RDKnbjbpV2GEEILGYEKvWRjacKb3er+z+cxUKko0Cj8uOkM438Ixc7aFLWBiub1nJk8qBHLd10kN78Av1rlH32yUnA3PQdNwYvXC1eIF4EEqHihzVp/ird7ehf10c3LL+CnbRd4b/7vrNxzGUWBbdO7am399zyIj75JUlwMlja2VKpaU3rBCvEMSYCKUnUqIoEFW88Rk5SBv4cTr3WqjZ11ya3/Ox95hz4tAgHIyc2n0wdbyMkrYNrrrxB7J4OFvxbuO1q7in5OInqYaxdOs2r2l9y+HoFTZTeSEuIwNTOjx+vj8Q1qV9rlCfFSkHugokRl5+az488bbD96g6yc/P88NiTsKh3e24ybkw2D23pw+VYKjUauJuExN85+HO4VbTl2qbAp+S+7L6EoCvtm9uCNznX4dKg/e2Z0Z/zcMDKynp92eVfOnGD6W4OoG9gMv2ZBqPIyqOnlRfNufVn+7Scc2LS6tEsU4qUgASpKzI4/b+Aa/BNfrTjO9JXHcQ3+iW1/XNd6bH6BhnE/hLJhSkcm9G1At8ZVWfJ+a1o3dGHG2pMlVtOYHt58vfoEm36/xvajNxjS3hO1+v6fvYerPZ6udhy+EFNir1mSfGuVx9TMouhnRVFYMu1Dug57ix3L5tLA/AbzX/OgR41Mti2eQefBo/hl5udkpqeVYtVCvBzkK1xRIhLvZvLq1J1s+rwTgbUrAHDkQiydJ27h/JIBONpZFDv+6u27GBmq8fcs3nZvQOuavDXrQInVVa9aOVZMalvUyq+pt3Ox3yuKQkpaDhamRiX2miXpjc51mLLhRtHPkeFnyM7M4Pq5P3mvtxf/16c+AIG1K+BRuQzDv5+PZ8MAjuzYTFDPAaVUtRAvBxmBihKx/kAE7fxci8ITwN/TiY7+bqw7EPHA8baWJqSkZZOdW/xr3tuJ6ZSx/u9NsZ9UUP3KHJ3Xl1WT2zF7/Sniku83m1+x5zI5eQXPTTP42BuRuHt5c+X0MXo2rVrsdy18KpFyJ5FKVWsSGxVZShUK8fKQEagoEamZuZS1MXvg8XK25tzLyH3g8fJ2FgR4VWDyT0f4ckQgarUB8ckZfLb0KB8MaPhUamzfyI0zVxPxHLScpt7OxCVlEJecweYvOuvtLNw3Z+4jMSa6aD9QE3Nz0lKSKGPvwLWYe7g52RQdG5+SiUZRkZ2Zjqm59OsV4mmTEagoEa0auLA+NILUjJyix9Kzcll34AptfLV3+Fk8oSUnLifg3m8JLcdtwGPQcro1die4efWnVucHA3y5sPRVgptX55MhjbiyYrBez8DNzs1HUTRFP9f2a8z1i+fwad2dcfP+IDqx8F5nakYOI2cf5JV2XTmycwu+LWQmrhBPm4xARYnwqV6OzoFVCBy1lje71gGVinmbz9LG14UGNRy1PqesrTl7Z/bgwvUkYpLSqVetHA5aRrElrbydBcEtnl5IP02m5ha07j2Yk2F78GzaHa9hP+NczobbCXdp0LQ1+bn5uNasLTuaCPEMSIC+QOKTM9h5LApTYzXtG7k+8/Z0c8Y0Y9sfN1gfGoGiwNQRAXRo9GAD+H/zdLOX/TWfQPfX3uFeyh32bV5Hk64DsLItQ62cbI7s2opdOSfGfrOgtEsU4qUgAfqC+HHzWSYtOkzL+pXJyM5j9Kz9/DKpLa0bPrsG6SqVig7+bnTwf3RoCt0ZqNUMm/glbYIHE7plLdcvncfS2pbhH06jpo+fdCMS4hmRAH0BnIu8w5SlRzmxsB+u5a0BOHQuhm6TtnJt1RCszPWzUbp4tCZ1nVl/WPsaVWf3GvQf+9EzrkgI8TeZRPQCWL33MkPaeRSFJxSuC2zk6cSvR7Q3MhDPh8FtPbCytSvtMoQQWkiAvgCycwu0jjKtzIweWGcphBCiZEiAvgA6+LuxdGd4sX6utxLS2PHnTdo8w3ugouQN/WoXCbejSrsMIYQWcg/0BdC8njON61TE941VDG7rQUZ2Hot+u8DHg/yo4CAL6oUQ4mmQAH0BqFQq5o8PYt/JW2w5FImpsSG/TuuCd9WypV2aEEK8sCRAXxAqlYqg+pUJql+5tEsRQoiXgtwDFUIIIXQgASqEHmvd0AVzS6vSLkMIoYUEqBB6rE+LGljalCntMoQQWkiACqHHsnLy0Wg0jz5QCPHMSYAKocdGzdrHndjo0i5DCKGFBKgQQgihAwlQIYQQQgcSoEIIIYQOJECFEEIIHUiACqHHOge6Y2FlU9plCCG0kAAVQo91fcUdC2sJUCH0kQSoEHosJS2bgoKC0i5DCKGFBKgQemz83DCS4m6XdhlCCC0kQIUQQggdSIAKIYQQOpAAFUIIIXQgASqEEELoQAJUCD3Wu3l1LG1sS7sMIYQWEqBC6LG2vq6YW1qXdhlCCC0kQIXQY3HJGeTn55V2GUIILSRAhdBjExceIjk+trTLEEJoIQEqhBBC6EACVAghhNCBBKgQQgihAwlQIYQQQgcSoELosYFtPLCytSvtMoQQWkiACqHHmnk7Y2ZhWdplCCG0kAAVQo/diEslLze3tMsQQmghASqEHvts6R+kJMaVdhlCCC0kQIUQQggdSIAKIYQQOpAAFUIIIXQgASqEEELoQAJUCD32WqfaWJexL+0yhBBaSIAKoccaeThham5R2mUIIbSQABVCj12KSiY3J7u0yxBCaCEBKoQem77qOHfvJJR2GUIILSRAhRBCCB1IgAohhBA6kAAVQgghdCABKoQQQuhApSjK4x+sUiUCN59eOUIIIYRecVEUpay2XzxRgAohhBCikHyFK4QQQuhAAlQIIYTQgQSoEEIIoQMJUCGEEEIHEqBCCCGEDiRAhRBCCB1IgAohhBA6kAAVQgghdCABKoQQQujg/wGuHZLkZQRrlAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the line, the points, and the nearest vectors to the plane\n",
    "fignum = 1\n",
    "plt.figure(fignum, figsize=(8, 6))\n",
    "plt.clf()\n",
    "\n",
    "plt.scatter(svc.support_vectors_[:, 0], svc.support_vectors_[:, 1], s=80,\n",
    "            facecolors='none', zorder=10, edgecolors='k')\n",
    "plt.scatter(X_train.iloc[:, 0], X_train.iloc[:, 1], c=y_train, zorder=10, cmap=plt.cm.Paired,\n",
    "            edgecolors='k')\n",
    "plt.axis('tight')\n",
    "\n",
    "Z = svc.predict(np.c_[xx.ravel(), yy.ravel()])\n",
    "\n",
    "# Put the result into a color plot\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.figure(fignum, figsize=(8, 6))\n",
    "plt.pcolormesh(xx, yy, Z, cmap=plt.cm.Paired, shading='auto')\n",
    "plt.contour(xx, yy, Z, colors=['k', 'k', 'k'], linestyles=['--', '-', '--'],\n",
    "            levels=[-.5, 0, .5])\n",
    "plt.xlabel(wine.columns[0])\n",
    "plt.ylabel(wine.columns[1])\n",
    "plt.xlim(x_min, x_max)\n",
    "plt.ylim(y_min, y_max)\n",
    "\n",
    "plt.xticks(())\n",
    "plt.yticks(())\n",
    "fignum = fignum + 1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "outer-garlic",
   "metadata": {},
   "source": [
    "### Exercise 4:\n",
    "Now, it's your turn to create a SVM model with a linear kernel using all the features available. How well does this model perfom? Is it better or worse?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "proper-double",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(kernel='linear')"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create your model here. Don't forget to run the fit function:\n",
    "svc = svm.SVC(kernel='linear')\n",
    "svc.fit(X_all_train, y_all_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "baking-keeping",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the predicted value. Don't forget to use X_all_test\n",
    "y_all_predicted = svc.predict(X_all_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "urban-adapter",
   "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>y_real</th>\n",
       "      <th>y_predicted</th>\n",
       "      <th>is_correct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>155</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>170</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>166</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     y_real  y_predicted  is_correct\n",
       "103       1            1        True\n",
       "140       2            2        True\n",
       "93        1            1        True\n",
       "8         0            0        True\n",
       "155       2            2        True\n",
       "107       1            1        True\n",
       "120       1            1        True\n",
       "73        1            1        True\n",
       "123       1            1        True\n",
       "169       2            2        True\n",
       "128       1            1        True\n",
       "52        0            0        True\n",
       "170       2            2        True\n",
       "67        1            1        True\n",
       "51        0            0        True\n",
       "127       1            1        True\n",
       "166       2            2        True\n",
       "116       1            1        True\n",
       "25        0            1       False\n",
       "66        1            1        True\n",
       "157       2            2        True\n",
       "29        0            0        True\n",
       "88        1            1        True\n",
       "16        0            0        True\n",
       "81        1            1        True\n",
       "49        0            0        True\n",
       "4         0            1       False\n",
       "14        0            0        True\n",
       "38        0            0        True\n",
       "109       1            1        True\n",
       "83        1            2       False\n",
       "168       2            2        True\n",
       "60        1            1        True\n",
       "75        1            1        True\n",
       "76        1            1        True\n",
       "35        0            0        True"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create the new result dataframe\n",
    "results = pd.DataFrame({'y_real': y_all_test, 'y_predicted': y_all_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "descending-spice",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 33\n",
      "Wrong predictions: 3\n",
      "Percentage of correct predictions: 91.67%\n"
     ]
    }
   ],
   "source": [
    "# Run the new results\n",
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "suspended-horse",
   "metadata": {},
   "source": [
    "### Polynominal kernel\n",
    "Now, we will use a polynomial kernel with degree 2. This will include a quadratic function in the formula. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "false-making",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(gamma=2, kernel='poly')"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc = svm.SVC(kernel='poly', gamma=2)\n",
    "svc.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "suited-prompt",
   "metadata": {},
   "source": [
    "Now, we calculate the predicted values from the testing dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "endless-server",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 27\n",
      "Wrong predictions: 9\n",
      "Percentage of correct predictions: 75.0%\n"
     ]
    }
   ],
   "source": [
    "y_predicted = svc.predict(X_test)\n",
    "results = pd.DataFrame({'y_real':y_test, 'y_predicted':y_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "\n",
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "configured-geneva",
   "metadata": {},
   "source": [
    "We will plot the plane to see the new thresholds "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "familiar-fitness",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the line, the points, and the nearest vectors to the plane\n",
    "fignum = 1\n",
    "plt.figure(fignum, figsize=(8, 6))\n",
    "plt.clf()\n",
    "\n",
    "plt.scatter(svc.support_vectors_[:, 0], svc.support_vectors_[:, 1], s=80,\n",
    "            facecolors='none', zorder=10, edgecolors='k')\n",
    "plt.scatter(X_train.iloc[:, 0], X_train.iloc[:, 1], c=y_train, zorder=10, cmap=plt.cm.Paired,\n",
    "            edgecolors='k')\n",
    "\n",
    "plt.axis('tight')\n",
    "x_min, x_max = X_train.iloc[:, 0].min() - 1, X_train.iloc[:, 0].max() + 1\n",
    "y_min, y_max = X_train.iloc[:, 1].min() - 1, X_train.iloc[:, 1].max() + 1\n",
    "\n",
    "Z = svc.predict(np.c_[xx.ravel(), yy.ravel()])\n",
    "\n",
    "# Put the result into a color plot\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.figure(fignum, figsize=(4, 3))\n",
    "plt.pcolormesh(xx, yy, Z, cmap=plt.cm.Paired, shading='auto')\n",
    "plt.contour(xx, yy, Z, colors=['k', 'k', 'k'], linestyles=['--', '-', '--'],\n",
    "            levels=[-.5, 0, .5])\n",
    "plt.xlabel(wine.columns[0])\n",
    "plt.ylabel(wine.columns[1])\n",
    "plt.xlim(x_min, x_max)\n",
    "plt.ylim(y_min, y_max)\n",
    "\n",
    "plt.xticks(())\n",
    "plt.yticks(())\n",
    "fignum = fignum + 1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "intensive-brunswick",
   "metadata": {},
   "source": [
    "### RBF kernel\n",
    "Lastly, let's use the Radial Basis Function kernel to detect other thresholds."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "uniform-fireplace",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(gamma=2)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc = svm.SVC(kernel='rbf', gamma=2)\n",
    "svc.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "harmful-compression",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 32\n",
      "Wrong predictions: 4\n",
      "Percentage of correct predictions: 88.89%\n"
     ]
    }
   ],
   "source": [
    "y_predicted = svc.predict(X_test)\n",
    "results = pd.DataFrame({'y_real':y_test, 'y_predicted':y_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "\n",
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "three-sleeping",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the line, the points, and the nearest vectors to the plane\n",
    "fignum = 1\n",
    "plt.figure(fignum, figsize=(8, 6))\n",
    "plt.clf()\n",
    "\n",
    "plt.scatter(svc.support_vectors_[:, 0], svc.support_vectors_[:, 1], s=80,\n",
    "            facecolors='none', zorder=10, edgecolors='k')\n",
    "plt.scatter(X_train.iloc[:, 0], X_train.iloc[:, 1], c=y_train, zorder=10, cmap=plt.cm.Paired,\n",
    "            edgecolors='k')\n",
    "\n",
    "plt.axis('tight')\n",
    "Z = svc.predict(np.c_[xx.ravel(), yy.ravel()])\n",
    "\n",
    "# Put the result into a color plot\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.figure(fignum, figsize=(4, 3))\n",
    "plt.pcolormesh(xx, yy, Z, cmap=plt.cm.Paired, shading='auto')\n",
    "plt.contour(xx, yy, Z, colors=['k', 'k', 'k'], linestyles=['--', '-', '--'],\n",
    "            levels=[-.5, 0, .5])\n",
    "plt.xlabel(wine.columns[0])\n",
    "plt.ylabel(wine.columns[1])\n",
    "plt.xlim(x_min, x_max)\n",
    "plt.ylim(y_min, y_max)\n",
    "\n",
    "plt.xticks(())\n",
    "plt.yticks(())\n",
    "fignum = fignum + 1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "weighted-disposal",
   "metadata": {},
   "source": [
    "## 5. Neural Networks: Muli-layer perceptron\n",
    "\n",
    "Multi-layer perceptron (MLP) is an artificial neural network. The model learns from the dataset through three types of layers:\n",
    "* *The input layer*: The input layer receives the features to be processed.\n",
    "* *The output layer*: The prediction and classification is performed by the output layer.\n",
    "* *The hidden layers*: An arbitrary number of hidden layers that are placed in between the input and output layer perform the computation. Each layer has nodes (also known as neurons) that transforms the values from the previous layer into an activation funcition. The output layer receives the activated values from the last hidden layer and transforms them into output values.\n",
    "\n",
    "<img src=\"https://static.javatpoint.com/tutorial/tensorflow/images/multi-layer-perceptron-in-tensorflow.png\" alt=\"drawing\" width=\"400\"/>\n",
    "\n",
    "Multi-layer perceptrons are networks of nodes that pass a linear combination of their inputs from one layer to another. As they do this, the nodes decide how to modify their inputs, utilizing a given activation function. The activation function of a neuron is the key here. By selecting non-linear activation functions, such as the logistic function \\sigma(x) shown below, the neural network can embed non-linearity in its operation:\n",
    "\n",
    "<img src=\"https://www.baeldung.com/wp-content/uploads/sites/4/2020/06/Copy-of-Copy-of-Blank-Diagram1.svg\" alt=\"drawing\" width=\"400\"/>\n",
    "\n",
    "Neural networks are good for non-linear classes of problems. The first advantage of neural networks is, therefore, their flexibility in addressing problems with non-linear shapes. MLPs are implemented in diverse fields such as speech recognition, image recognition, and machine translation software. From artificial neural networks, **deep learning** has emerged as a sub-field of machine learning.\n",
    "\n",
    "Let's create a simple perceptron with 5 nodes in two layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "buried-corpus",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MLPClassifier(alpha=1e-05, hidden_layer_sizes=(5, 2), max_iter=2000,\n",
       "              random_state=1, solver='lbfgs')"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf = MLPClassifier(solver='lbfgs', alpha=1e-5, hidden_layer_sizes=(5, 2), random_state=1, max_iter=10000)\n",
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "assured-philip",
   "metadata": {},
   "source": [
    "We will use the trained perpcetron to predict the wines' classes and compare with the real values. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "spanish-painting",
   "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>y_real</th>\n",
       "      <th>y_predicted</th>\n",
       "      <th>is_correct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     y_real  y_predicted  is_correct\n",
       "56        0            0        True\n",
       "153       2            2        True\n",
       "127       1            0       False\n",
       "33        0            0        True\n",
       "118       1            2       False"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = clf.predict(X_test)\n",
    "results = pd.DataFrame({'y_real':y_test, 'y_predicted':y_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "results.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "loving-restoration",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 15\n",
      "Wrong predictions: 21\n",
      "Percentage of correct predictions: 41.67%\n"
     ]
    }
   ],
   "source": [
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "common-infrastructure",
   "metadata": {},
   "source": [
    "Let's see visually how the neural network performed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "religious-microwave",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Z = clf.predict(np.c_[xx.ravel(), yy.ravel()])\n",
    "\n",
    "# Put the result into a color plot\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.figure(1, figsize=(8, 6))\n",
    "plt.pcolormesh(xx, yy, Z, shading='auto', cmap=plt.cm.Paired)\n",
    "\n",
    "# Plot also the training points\n",
    "plt.scatter(X_train.iloc[:, 0], X_train.iloc[:, 1], c=y_train, edgecolors='k', cmap=plt.cm.Paired)\n",
    "plt.xlabel(wine.columns[0])\n",
    "plt.ylabel(wine.columns[1])\n",
    "\n",
    "plt.xlim(xx.min(), xx.max())\n",
    "plt.ylim(yy.min(), yy.max())\n",
    "plt.xticks(())\n",
    "plt.yticks(())\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "gentle-government",
   "metadata": {},
   "source": [
    "### Exercise 5:\n",
    "Now, it's your turn to create a neural network using all the features available from the wine dataset. Try with 20 nodes and 10 layers => (20,10). How well does this model perfom? Is it better or worse?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "imperial-tobacco",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MLPClassifier(alpha=1e-05, hidden_layer_sizes=(20, 15), max_iter=10000,\n",
       "              random_state=1, solver='lbfgs')"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create your model here. Don't forget to run the fit function, and use the X_all_train and y_all_train dataframes.\n",
    "clf = MLPClassifier(solver='lbfgs', alpha=1e-5, hidden_layer_sizes=(20, 15), random_state=1, max_iter=10000)\n",
    "clf.fit(X_all_train, y_all_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "after-final",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the predicted value. Don't forget to use `X_all_test`\n",
    "y_all_predicted = clf.predict(X_all_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "welcome-oasis",
   "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>y_real</th>\n",
       "      <th>y_predicted</th>\n",
       "      <th>is_correct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>155</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>170</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>166</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     y_real  y_predicted  is_correct\n",
       "103       1            1        True\n",
       "140       2            2        True\n",
       "93        1            1        True\n",
       "8         0            0        True\n",
       "155       2            2        True\n",
       "107       1            1        True\n",
       "120       1            1        True\n",
       "73        1            1        True\n",
       "123       1            1        True\n",
       "169       2            2        True\n",
       "128       1            1        True\n",
       "52        0            0        True\n",
       "170       2            2        True\n",
       "67        1            1        True\n",
       "51        0            0        True\n",
       "127       1            1        True\n",
       "166       2            2        True\n",
       "116       1            1        True\n",
       "25        0            1       False\n",
       "66        1            1        True\n",
       "157       2            2        True\n",
       "29        0            0        True\n",
       "88        1            1        True\n",
       "16        0            0        True\n",
       "81        1            1        True\n",
       "49        0            0        True\n",
       "4         0            1       False\n",
       "14        0            0        True\n",
       "38        0            0        True\n",
       "109       1            1        True\n",
       "83        1            2       False\n",
       "168       2            2        True\n",
       "60        1            1        True\n",
       "75        1            1        True\n",
       "76        1            1        True\n",
       "35        0            0        True"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create the new result dataframe\n",
    "results = pd.DataFrame({'y_real': y_all_test, 'y_predicted': y_all_predicted})\n",
    "results[\"is_correct\"] = [i['y_real'] == i['y_predicted'] for k,i in results.iterrows()]\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "retired-plaintiff",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct predictions: 33\n",
      "Wrong predictions: 3\n",
      "Percentage of correct predictions: 91.67%\n"
     ]
    }
   ],
   "source": [
    "# Run the new results\n",
    "print(\"Correct predictions: \" + str(len(results[results.is_correct == True])))\n",
    "print(\"Wrong predictions: \" + str(len(results[results.is_correct == False])))\n",
    "print(\"Percentage of correct predictions: \" + str(round(100*len(results[results.is_correct == True])/len(results),2)) + \"%\")"
   ]
  }
 ],
 "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": "3.9.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}