FinalSubmissionNotebook.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "WorkonLater.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Ultan-Kearns/LYIT-Machine-Learning-Project/blob/main/FinalSubmissionNotebook.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SaYS6bFNmWuN"
      },
      "source": [
        "# Loading files and importing libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 72
        },
        "id": "8ySUe3n2zXfN",
        "outputId": "d9253769-fc0f-4964-bffa-dbe64eb1c1fc"
      },
      "source": [
        "#import libraries\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import sklearn as sk\n",
        "import matplotlib \n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn import tree\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import classification_report,confusion_matrix,accuracy_score\n",
        "import seaborn as sns # for the correlation matrix plot\n",
        "import pydot\n",
        "import pydotplus\n",
        "import io\n",
        "import cv2\n",
        "from google.colab import files\n",
        "\n",
        "uploaded = files.upload()\n",
        "wdbcwh = pd.read_csv(io.BytesIO(uploaded['wdbcwh.csv']), header = 0)"
      ],
      "execution_count": 1482,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-70bf1440-6a62-403b-befc-1577aa539f32\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-70bf1440-6a62-403b-befc-1577aa539f32\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving wdbcwh.csv to wdbcwh (27).csv\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N963plximUXS"
      },
      "source": [
        "# Prepping and analyzing the data\n",
        "In this section we prep and analyze the data we do this by dividing the data into two groups training and testing.  We will train the models with the training data so that we can predict the values within the testing dataset.  We randomize the data to produce a more accurate model and to show us the variation in performance of our models as the data is different everytime. We could also use a random seed to produce the same \"random\" set of rows everytime, although we wanted to avoid this to reduce our models bias towards the data when training and testing our models."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D6Qw95tszgIg"
      },
      "source": [
        "# split data into training and test datasets\n",
        "features = list(wdbcwh.columns[2:32])\n",
        "\n",
        "X = wdbcwh[features]\n",
        "y = wdbcwh['Diagnosis']\n",
        "# could use random state 42 for same results every time\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3)"
      ],
      "execution_count": 1483,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "65JKC0E6WSeX"
      },
      "source": [
        "Here we are showing some of the values in our training sets and our test sets, by doing this we can see that we have a decent amount of randomness in our data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dPHeQ96EpAo8",
        "outputId": "6f96c613-1531-4907-a777-e71beeb3866b"
      },
      "source": [
        "print('X_train ',X_train)\n",
        "print('y_train ',y_train)\n",
        "print('X_test ',X_test)\n",
        "print('y_test ',y_test)"
      ],
      "execution_count": 1484,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X_train       Mean_Radius  Mean_Texture  ...  Worst_Symmetry  Worst_Fractal_Dimension\n",
            "475       12.830         15.73  ...          0.3006                  0.07802\n",
            "93        13.450         18.30  ...          0.2678                  0.06603\n",
            "71         8.888         14.64  ...          0.2254                  0.10840\n",
            "394       12.100         17.72  ...          0.3049                  0.07081\n",
            "123       14.500         10.89  ...          0.2889                  0.08006\n",
            "..           ...           ...  ...             ...                      ...\n",
            "31        11.840         18.70  ...          0.4761                  0.14020\n",
            "28        15.300         25.27  ...          0.4027                  0.09876\n",
            "294       12.720         13.78  ...          0.2369                  0.06922\n",
            "113       10.510         20.19  ...          0.2383                  0.09026\n",
            "245       10.480         19.86  ...          0.2883                  0.07748\n",
            "\n",
            "[398 rows x 30 columns]\n",
            "y_train  475    B\n",
            "93     B\n",
            "71     B\n",
            "394    B\n",
            "123    B\n",
            "      ..\n",
            "31     M\n",
            "28     M\n",
            "294    B\n",
            "113    B\n",
            "245    B\n",
            "Name: Diagnosis, Length: 398, dtype: object\n",
            "X_test       Mean_Radius  Mean_Texture  ...  Worst_Symmetry  Worst_Fractal_Dimension\n",
            "517       19.890         20.26  ...          0.2549                  0.09136\n",
            "314        8.597         18.60  ...          0.3142                  0.08116\n",
            "116        8.950         15.76  ...          0.1652                  0.07722\n",
            "474       10.880         15.62  ...          0.2581                  0.10800\n",
            "288       11.260         19.96  ...          0.2955                  0.07009\n",
            "..           ...           ...  ...             ...                      ...\n",
            "377       13.460         28.21  ...          0.2694                  0.07061\n",
            "60        10.170         14.88  ...          0.3557                  0.08020\n",
            "300       19.530         18.90  ...          0.2968                  0.09929\n",
            "543       13.210         28.06  ...          0.2473                  0.06443\n",
            "479       16.250         19.51  ...          0.3318                  0.09136\n",
            "\n",
            "[171 rows x 30 columns]\n",
            "y_test  517    M\n",
            "314    B\n",
            "116    B\n",
            "474    B\n",
            "288    B\n",
            "      ..\n",
            "377    B\n",
            "60     B\n",
            "300    M\n",
            "543    B\n",
            "479    M\n",
            "Name: Diagnosis, Length: 171, dtype: object\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lHiGaozoahwT"
      },
      "source": [
        "We show the shape of our test and training sets here, when using different split values we will get differing results.  It is important to know the shape of our sets as we can reshape them if needed to fit our models"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "razAblX4qtPl",
        "outputId": "f9aba3e6-d93c-4d63-e22f-5c930704ce74"
      },
      "source": [
        "# print test sets shape will be same as training sets shape\n",
        "print('X_test set shape: ',X_test.shape)\n",
        "print('y_test set shape: ',y_test.shape)\n",
        "print('X_train shape: ', X_train.shape)\n",
        "print('y_train shape: ', y_train.shape)"
      ],
      "execution_count": 1485,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X_test set shape:  (171, 30)\n",
            "y_test set shape:  (171,)\n",
            "X_train shape:  (398, 30)\n",
            "y_train shape:  (398,)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6m3wXanraw-v"
      },
      "source": [
        "Here we are replacing diagnosis values with a 1 for malignant and a 0 for benign, we do this because it makes it easier to perform numerical calculations on the sets. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7hCqV87N0k7x"
      },
      "source": [
        "# Disabled pandas warning due to false positive\n",
        "pd.options.mode.chained_assignment = None  # default='warn'\n",
        "\n",
        "\n",
        "# Replace malignant with 1 and benign with 0\n",
        "y_train = y_train.replace({'M':1,'B':0})\n",
        "y_test = y_test.replace({'M':1,'B':0})\n",
        "y = y.replace({'M':1,'B':0})\n",
        "# resetting pandas warning\n",
        "pd.options.mode.chained_assignment = 'warn'  # default='warn'\n",
        "X_trainNB, y_trainNB = X_train, y_train\n"
      ],
      "execution_count": 1486,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "esjXl_W26F7C"
      },
      "source": [
        "## Show standard deviation and variance between sets\n",
        "\n",
        "Here we are showing both the standard deviation and variance of the sets, these are good methods for detecting outliers in our set, as we can see both sets have a somewhat high level of standard deviation.  We may try dropping columns from our dataframe to try and reduce this.  We can also see from these variables the amount of variation in certain features or throughout the overall dataset, these are useful metrics when training the models as we can ensure the models are trained with a large variety of different values.\n",
        "\n",
        "Standard deviation is calculated by calculating the sum of each element in our data set then subtracting the population mean from the element and squaring we then divide by n and get the square root\n",
        "\n",
        "Variance is calculated by essentially the same way but without getting the square root of our end result"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3BVX2LRv51ZB",
        "outputId": "f46c4229-7872-4dfa-914c-4dbc89e352fa"
      },
      "source": [
        "print('Standard deviation of values in x training set ',X_train.std())\n",
        "print('Variance of diagnosis values in x training set: ', X_train.var())"
      ],
      "execution_count": 1487,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Standard deviation of values in x training set  Mean_Radius                  3.481296\n",
            "Mean_Texture                 4.222748\n",
            "Mean_Perimeter              23.815839\n",
            "Mean_Area                  340.095674\n",
            "Mean_Smoothness              0.013770\n",
            "Mean_Compactness             0.050612\n",
            "Mean_Concavity               0.074402\n",
            "Mean_Concave_Points          0.036789\n",
            "Mean_Symmetry                0.026499\n",
            "Mean_Fractal_Dimension       0.007415\n",
            "Radius_SE                    0.241815\n",
            "Texture_SE                   0.499631\n",
            "Perimeter_SE                 1.735896\n",
            "Area_SE                     35.367067\n",
            "Smoothness_SE                0.002893\n",
            "Compactness_SE               0.016731\n",
            "Concavity_SE                 0.031862\n",
            "Concave_Points_SE            0.006136\n",
            "Symmetry_SE                  0.007288\n",
            "Fractal_Dimension_SE         0.002664\n",
            "Worst_Radius                 4.842164\n",
            "Worst_Texture                6.086432\n",
            "Worst_Perimeter             33.341683\n",
            "Worst_Area                 565.517578\n",
            "Worst_Smoothness             0.022049\n",
            "Worst_Compactness            0.155266\n",
            "Worst_Concavity              0.204543\n",
            "Worst_Concave_Points         0.064660\n",
            "Worst_Symmetry               0.060943\n",
            "Worst_Fractal_Dimension      0.018919\n",
            "dtype: float64\n",
            "Variance of diagnosis values in x training set:  Mean_Radius                    12.119420\n",
            "Mean_Texture                   17.831597\n",
            "Mean_Perimeter                567.194182\n",
            "Mean_Area                  115665.067713\n",
            "Mean_Smoothness                 0.000190\n",
            "Mean_Compactness                0.002562\n",
            "Mean_Concavity                  0.005536\n",
            "Mean_Concave_Points             0.001353\n",
            "Mean_Symmetry                   0.000702\n",
            "Mean_Fractal_Dimension          0.000055\n",
            "Radius_SE                       0.058475\n",
            "Texture_SE                      0.249631\n",
            "Perimeter_SE                    3.013334\n",
            "Area_SE                      1250.829439\n",
            "Smoothness_SE                   0.000008\n",
            "Compactness_SE                  0.000280\n",
            "Concavity_SE                    0.001015\n",
            "Concave_Points_SE               0.000038\n",
            "Symmetry_SE                     0.000053\n",
            "Fractal_Dimension_SE            0.000007\n",
            "Worst_Radius                   23.446550\n",
            "Worst_Texture                  37.044656\n",
            "Worst_Perimeter              1111.667792\n",
            "Worst_Area                 319810.131016\n",
            "Worst_Smoothness                0.000486\n",
            "Worst_Compactness               0.024108\n",
            "Worst_Concavity                 0.041838\n",
            "Worst_Concave_Points            0.004181\n",
            "Worst_Symmetry                  0.003714\n",
            "Worst_Fractal_Dimension         0.000358\n",
            "dtype: float64\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L63BMNQJZFWg",
        "outputId": "3fe6ee64-7156-4d50-fa1d-e02d5cb71d9b"
      },
      "source": [
        "\n",
        "print('Standard deviation of values in y training set ',y_train.std())\n",
        "print('Variance of values in y training set: ', y_train.var())"
      ],
      "execution_count": 1488,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Standard deviation of values in y training set  0.4870706585410926\n",
            "Variance of values in y training set:  0.2372378264116536\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "POHC2Qe-51pY",
        "outputId": "543cc330-1944-40f4-8877-bb4e1a57e21a"
      },
      "source": [
        "print('Standard deviation of values in x testing set ',X_test.std())\n",
        "print('Variance of values in x testing set: ', X_test.var())\n"
      ],
      "execution_count": 1489,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Standard deviation of values in x testing set  Mean_Radius                  3.629898\n",
            "Mean_Texture                 4.484972\n",
            "Mean_Perimeter              25.451116\n",
            "Mean_Area                  378.945944\n",
            "Mean_Smoothness              0.014744\n",
            "Mean_Compactness             0.057739\n",
            "Mean_Concavity               0.091109\n",
            "Mean_Concave_Points          0.043244\n",
            "Mean_Symmetry                0.029512\n",
            "Mean_Fractal_Dimension       0.006174\n",
            "Radius_SE                    0.346711\n",
            "Texture_SE                   0.656698\n",
            "Perimeter_SE                 2.570971\n",
            "Area_SE                     63.136715\n",
            "Smoothness_SE                0.003245\n",
            "Compactness_SE               0.020421\n",
            "Concavity_SE                 0.025953\n",
            "Concave_Points_SE            0.006264\n",
            "Symmetry_SE                  0.010124\n",
            "Fractal_Dimension_SE         0.002611\n",
            "Worst_Radius                 4.823623\n",
            "Worst_Texture                6.293949\n",
            "Worst_Perimeter             34.291128\n",
            "Worst_Area                 579.498822\n",
            "Worst_Smoothness             0.024569\n",
            "Worst_Compactness            0.162488\n",
            "Worst_Concavity              0.218417\n",
            "Worst_Concave_Points         0.068355\n",
            "Worst_Symmetry               0.064027\n",
            "Worst_Fractal_Dimension      0.015936\n",
            "dtype: float64\n",
            "Variance of values in x testing set:  Mean_Radius                    13.176157\n",
            "Mean_Texture                   20.114976\n",
            "Mean_Perimeter                647.759324\n",
            "Mean_Area                  143600.028269\n",
            "Mean_Smoothness                 0.000217\n",
            "Mean_Compactness                0.003334\n",
            "Mean_Concavity                  0.008301\n",
            "Mean_Concave_Points             0.001870\n",
            "Mean_Symmetry                   0.000871\n",
            "Mean_Fractal_Dimension          0.000038\n",
            "Radius_SE                       0.120208\n",
            "Texture_SE                      0.431252\n",
            "Perimeter_SE                    6.609893\n",
            "Area_SE                      3986.244718\n",
            "Smoothness_SE                   0.000011\n",
            "Compactness_SE                  0.000417\n",
            "Concavity_SE                    0.000674\n",
            "Concave_Points_SE               0.000039\n",
            "Symmetry_SE                     0.000103\n",
            "Fractal_Dimension_SE            0.000007\n",
            "Worst_Radius                   23.267337\n",
            "Worst_Texture                  39.613791\n",
            "Worst_Perimeter              1175.881434\n",
            "Worst_Area                 335818.884628\n",
            "Worst_Smoothness                0.000604\n",
            "Worst_Compactness               0.026402\n",
            "Worst_Concavity                 0.047706\n",
            "Worst_Concave_Points            0.004672\n",
            "Worst_Symmetry                  0.004099\n",
            "Worst_Fractal_Dimension         0.000254\n",
            "dtype: float64\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FOwCIZ2yZMhv",
        "outputId": "cd799596-6782-4a17-8692-39c53b9101fb"
      },
      "source": [
        "print('Standard deviation of values in y testing set ',y_test.std())\n",
        "print('Variance of values in y testing set: ', y_test.var())"
      ],
      "execution_count": 1490,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Standard deviation of values in y testing set  0.4767739312220712\n",
            "Variance of values in y testing set:  0.22731338149294825\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4zV46iL03M8x"
      },
      "source": [
        "Lets look at a histogram of diagnosis values in both our training and test sets to get a visualized view of our sets"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 438
        },
        "id": "EWXUxovm5ynb",
        "outputId": "224501f1-62c3-4de1-8ab1-f2936d3289a8"
      },
      "source": [
        "plt.hist(X_train)\n"
      ],
      "execution_count": 1491,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([[28.,  2.,  0., ...,  0.,  0.,  0.],\n",
              "        [28.,  1.,  1., ...,  0.,  0.,  0.],\n",
              "        [30.,  0.,  0., ...,  0.,  0.,  0.],\n",
              "        ...,\n",
              "        [28.,  2.,  0., ...,  0.,  0.,  0.],\n",
              "        [29.,  1.,  0., ...,  0.,  0.,  0.],\n",
              "        [29.,  1.,  0., ...,  0.,  0.,  0.]]),\n",
              " array([   0. ,  343.2,  686.4, 1029.6, 1372.8, 1716. , 2059.2, 2402.4,\n",
              "        2745.6, 3088.8, 3432. ]),\n",
              " <a list of 398 Lists of Patches objects>)"
            ]
          },
          "metadata": {},
          "execution_count": 1491
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 438
        },
        "id": "le3ockCb5y4z",
        "outputId": "9d34e741-3492-4d70-fe82-0168e793137e"
      },
      "source": [
        "plt.hist(X_test)\n"
      ],
      "execution_count": 1492,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([[28.,  0.,  1., ...,  0.,  0.,  0.],\n",
              "        [30.,  0.,  0., ...,  0.,  0.,  0.],\n",
              "        [30.,  0.,  0., ...,  0.,  0.,  0.],\n",
              "        ...,\n",
              "        [28.,  0.,  1., ...,  0.,  0.,  0.],\n",
              "        [28.,  2.,  0., ...,  0.,  0.,  0.],\n",
              "        [28.,  1.,  1., ...,  0.,  0.,  0.]]),\n",
              " array([   0. ,  425.4,  850.8, 1276.2, 1701.6, 2127. , 2552.4, 2977.8,\n",
              "        3403.2, 3828.6, 4254. ]),\n",
              " <a list of 171 Lists of Patches objects>)"
            ]
          },
          "metadata": {},
          "execution_count": 1492
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAD4CAYAAAD1jb0+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAObElEQVR4nO3df4xldXnH8fenLP5IMbLIZN0A6SAlAgl1IROK0RiLVZGSggmpkMZuWprVVhJNTZpVk3ZM+oc2RdsmRLMW4jax/qg/AlFbu93dRExa7KArLmwoC8V0ycqORRT/sQWf/nG/swyzM3tnZ+6d2e/u+5Xc3HO+55x7nvNs5rN3zj3nTqoKSVJ/fmm9C5AkrYwBLkmdMsAlqVMGuCR1ygCXpE5tWMudnXvuuTU5ObmWu5Sk7t1///0/qqqJheNrGuCTk5PMzMys5S4lqXtJfrDYuKdQJKlTBrgkdcoAl6ROGeCS1CkDXJI6ZYBLUqeGBniSlyT5dpLvJXkwyYfb+IVJ7ktyMMnnk7xo/OVKkuYs5x34z4Frquo1wBbg2iRXAx8FPl5Vvwr8GLh1fGVKkhYaGuA18LM2e2Z7FHAN8MU2vhO4cSwVSpIWtaxz4EnOSLIPOALsAh4Fnq6qZ9sqh4Dzlth2W5KZJDOzs7MrLvTynZeveNvFtn/l3n0vmJ+enl7V60vSWltWgFfVc1W1BTgfuAq4ZLk7qKodVTVVVVMTE8fcyi9JWqETugqlqp4G9gKvBc5OMvddKucDT4y4NknScSznKpSJJGe36ZcCbwYOMAjym9pqW4G7x1WkJOlYy/k2ws3AziRnMAj8L1TVV5M8BHwuyV8A3wXuHGOdkqQFhgZ4VT0AXLHI+GMMzodLktaBd2JKUqdOiQCf3P61Y8buePceAHbvuWjJ7Q5tv3dsNUnSuJ0SAS5JpyMDXJI6ZYBLUqcMcEnqlAEuSZ0ywCWpU10F+Cv37uPQ9nu5fOflR6cPXHIpMPg2wTvevQemX/78NwtOvxyA299x/dHXmL8NwIFLLl30MkRJOtl1FeCSpOcZ4JLUKQNckjplgEtSpwxwSerUaRXgC/8OpiT17LQKcEk6lRjgktQpA1ySOmWAS1KnDHBJ6pQBLkmdMsAlqVMGuCR1ygCXpE4Z4JLUqaEBnuSCJHuTPJTkwSTvbePTSZ5Isq89rht/uZKkORuWsc6zwPur6jtJXgbcn2RXW/bxqvqr8ZUnSVrK0ACvqsPA4Tb9TJIDwHnjLkySdHwndA48ySRwBXBfG7otyQNJ7kqycYlttiWZSTIzOzu7qmIlSc9bdoAnOQv4EvC+qvop8AngImALg3foty+2XVXtqKqpqpqamJgYQcmSJFhmgCc5k0F4f6aqvgxQVU9W1XNV9QvgU8BV4ytTkrTQcq5CCXAncKCqPjZvfPO81d4O7B99eZKkpSznKpTXAe8Evp9k7k/afBC4JckWoIDHgXeNpUJJ0qKWcxXKt4Assujroy9HkrRc3okpSZ0ywCWpUwa4JHXKAJekThngktQpA1ySOmWAS1KnDHBJ6pQBLkmdMsAlqVMGuCR1ygCXpE4Z4JLUKQNckjplgEtSpwxwSeqUAS5JnTLAJalTBrgkdcoAl6ROGeCS1CkDXJI6ZYBLUqcMcEnqlAEuSZ0ywCWpU0MDPMkFSfYmeSjJg0ne28bPSbIrySPteeP4y5UkzVnOO/BngfdX1WXA1cB7klwGbAd2V9XFwO42L0laI0MDvKoOV9V32vQzwAHgPOAGYGdbbSdw47iKlCQd64TOgSeZBK4A7gM2VdXhtuiHwKYlttmWZCbJzOzs7CpKlSTNt+wAT3IW8CXgfVX10/nLqqqAWmy7qtpRVVNVNTUxMbGqYiVJz1tWgCc5k0F4f6aqvtyGn0yyuS3fDBwZT4mSpMUs5yqUAHcCB6rqY/MW3QNsbdNbgbtHX54kaSkblrHO64B3At9Psq+NfRD4CPCFJLcCPwB+ZzwlSpIWMzTAq+pbQJZY/KbRliNJWi7vxJSkThngktQpA1ySOmWAS1KnDHBJ6pQBLkmdMsAlqVMGuCR1ygCXpE4Z4JLUKQNckjplgEtSpwxwSeqUAS5JnTLAJalTBrgkdcoAl6ROGeCS1CkDXJI6ZYBLUqcMcEnqlAEuSZ0ywCWpUwa4JHXKAJekTg0N8CR3JTmSZP+8sekkTyTZ1x7XjbdMSdJCy3kH/mng2kXGP15VW9rj66MtS5I0zNAAr6pvAk+tQS2SpBOwmnPgtyV5oJ1i2bjUSkm2JZlJMjM7O7uK3UmS5ltpgH8CuAjYAhwGbl9qxaraUVVTVTU1MTGxwt1JkhZaUYBX1ZNV9VxV/QL4FHDVaMuSJA2zogBPsnne7NuB/UutK0kajw3DVkjyWeCNwLlJDgF/DrwxyRaggMeBd42xRknSIoYGeFXdssjwnWOoRZJ0ArwTU5I6ZYBLUqcMcEnqlAEuSZ0ywCWpUwa4JHXKAJekThngktQpA1ySOmWAS1KnDHBJ6pQBLkmdMsAlqVMGuCR1ygCXpE4Z4JLUKQNckjplgEtSpwxwSeqUAS5JnTLAJalTBrgkdcoAl6ROGeCS1CkDXJI6ZYBLUqeGBniSu5IcSbJ/3tg5SXYleaQ9bxxvmZKkhZbzDvzTwLULxrYDu6vqYmB3m5ckraGhAV5V3wSeWjB8A7CzTe8EbhxxXZKkIVZ6DnxTVR1u0z8ENi21YpJtSWaSzMzOzq5wd5KkhVb9IWZVFVDHWb6jqqaqampiYmK1u5MkNSsN8CeTbAZoz0dGV5IkaTlWGuD3AFvb9Fbg7tGUI0laruVcRvhZ4N+AVyc5lORW4CPAm5M8Avxmm5ckraENw1aoqluWWPSmEdciSToB3okpSZ0ywCWpUwa4JHXKAJekThngktQpA1ySOmWAS1KnDHBJ6pQBLkmdMsAlqVMGuCR1ygCXpE4Z4JLUKQNckjplgEtSpwxwSeqUAS5JnTLAJalTBrgkdcoAl6ROGeCS1CkDXJI6ZYBLUqcMcEnqlAEuSZ3asJqNkzwOPAM8BzxbVVOjKEqSNNyqArz5jar60QheR5J0AjyFIkmdWm2AF/AvSe5Psm2xFZJsSzKTZGZ2dnaVu5MkzVltgL++qq4E3ga8J8kbFq5QVTuqaqqqpiYmJla5O0nSnFUFeFU90Z6PAF8BrhpFUZKk4VYc4El+OcnL5qaBtwD7R1WYJOn4VnMVyibgK0nmXucfquqfR1KVJGmoFQd4VT0GvGaEtUiSToCXEUpSpwxwSeqUAS5JnTLAJalTBrgkdcoAl6ROGeCS1CkDXJI6ZYBLUqcMcEnqlAEuSZ0ywCWpUwa4JHXKAJekThngktQpA3yFDm2/lwOXXHp0/sAllzK5/WtMT08vuv7uPRcdMza3zXJdvvPyE65Tx1rs30LqkQEuSZ0ywCWpUwa4JHXKAJekThngktQpA1ySOmWAL9PuPRdx+zuuP2Z8cvvXOLT93heM3fHuPUe3WWjuUsCF2wzb91KWep3p6emjdQDc/o7rX3AZ4npekjjKfb9y774XzC91Gec4eXmn5juRn+3VMsAlqVMGuCR1ygCXpE6tKsCTXJvk4SQHk2wfVVGSpOFWHOBJzgDuAN4GXAbckuSyURUmSTq+1bwDvwo4WFWPVdX/Ap8DbhhNWZKkYVJVK9swuQm4tqr+sM2/E/j1qrptwXrbgG1t9tXAwyewm3OBH62owNOHPRrOHg1nj4Zbzx79SlVNLBzcMO69VtUOYMdKtk0yU1VTIy7plGKPhrNHw9mj4U7GHq3mFMoTwAXz5s9vY5KkNbCaAP8P4OIkFyZ5EXAzcM9oypIkDbPiUyhV9WyS24BvAGcAd1XVgyOrbGBFp15OM/ZoOHs0nD0a7qTr0Yo/xJQkrS/vxJSkThngktSpkzbAT+fb9JPcleRIkv3zxs5JsivJI+15YxtPkr9tfXogyZXzttna1n8kydb1OJZxSHJBkr1JHkryYJL3tnF71CR5SZJvJ/le69GH2/iFSe5rvfh8uwCBJC9u8wfb8sl5r/WBNv5wkreuzxGNT5Izknw3yVfbfD89qqqT7sHgQ9FHgVcBLwK+B1y23nWt4fG/AbgS2D9v7C+B7W16O/DRNn0d8E9AgKuB+9r4OcBj7Xljm9643sc2ov5sBq5s0y8D/pPB1znYo+d7FOCsNn0mcF879i8AN7fxTwJ/1Kb/GPhkm74Z+Hybvqz9/L0YuLD9XJ6x3sc34l79CfAPwFfbfDc9OlnfgZ/Wt+lX1TeBpxYM3wDsbNM7gRvnjf99Dfw7cHaSzcBbgV1V9VRV/RjYBVw7/urHr6oOV9V32vQzwAHgPOzRUe1Yf9Zmz2yPAq4BvtjGF/ZorndfBN6UJG38c1X186r6L+Agg5/PU0KS84HfAv6uzYeOenSyBvh5wH/Pmz/Uxk5nm6rqcJv+IbCpTS/Vq9Oih+3X2CsYvMO0R/O0UwP7gCMM/nN6FHi6qp5tq8w/3qO9aMt/AryCU7xHwF8Dfwr8os2/go56dLIGuI6jBr+3nfbXfyY5C/gS8L6q+un8ZfYIquq5qtrC4C7pq4BL1rmkk0qS64EjVXX/eteyUidrgHub/rGebL/2056PtPGlenVK9zDJmQzC+zNV9eU2bI8WUVVPA3uB1zI4fTR3A9/84z3ai7b85cD/cGr36HXAbyd5nMFp2muAv6GjHp2sAe5t+se6B5i7SmIrcPe88d9rV1pcDfyknUb4BvCWJBvb1RhvaWPda+cd7wQOVNXH5i2yR02SiSRnt+mXAm9m8FnBXuCmttrCHs317iZgT/st5h7g5nYFxoXAxcC31+YoxquqPlBV51fVJIOM2VNVv0tPPVrvT4CP88nwdQyuLngU+NB617PGx/5Z4DDwfwzOp93K4FzbbuAR4F+Bc9q6YfCHNR4Fvg9MzXudP2DwgcpB4PfX+7hG2J/XMzg98gCwrz2us0cv6NGvAd9tPdoP/FkbfxWDcDkI/CPw4jb+kjZ/sC1/1bzX+lDr3cPA29b72MbUrzfy/FUo3fTIW+klqVMn6ykUSdIQBrgkdcoAl6ROGeCS1CkDXJI6ZYBLUqcMcEnq1P8D/M+3xTS2SXAAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 317
        },
        "id": "IfBhS8Jq3SUq",
        "outputId": "4c1cd7ae-1ef5-41e7-a834-7be5848506a3"
      },
      "source": [
        "plt.hist(y_train)\n"
      ],
      "execution_count": 1493,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([245.,   0.,   0.,   0.,   0.,   0.,   0.,   0.,   0., 153.]),\n",
              " array([0. , 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1. ]),\n",
              " <a list of 10 Patch objects>)"
            ]
          },
          "metadata": {},
          "execution_count": 1493
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 317
        },
        "id": "RDoBDTSb3VLj",
        "outputId": "82bdbc01-3d65-4840-e41c-2caf71e81596"
      },
      "source": [
        "plt.hist(y_test)\n"
      ],
      "execution_count": 1494,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([112.,   0.,   0.,   0.,   0.,   0.,   0.,   0.,   0.,  59.]),\n",
              " array([0. , 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1. ]),\n",
              " <a list of 10 Patch objects>)"
            ]
          },
          "metadata": {},
          "execution_count": 1494
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD4CAYAAAAXUaZHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAANcUlEQVR4nO3cfYxl9V3H8fenjIjUtjzsSNZddGi6VTcYA5kgDUmt3cZQMCyJhECsXZuNm9Zaq5jY1f6B0X8g0dY2IdVNwW5NRRAb2UjV4BZCbGR1KMijlS3lYXFhpxbwodFC+vWPe2wm6wxzZ859YH77fiWbuffcc+/5np3Z95459yFVhSSpLa+b9gCSpNEz7pLUIOMuSQ0y7pLUIOMuSQ2amfYAAJs2baq5ublpjyFJG8p999339aqaXe6210Tc5+bmWFhYmPYYkrShJHlqpds8LSNJDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDXpNvEO1j7m9d0xt209ed+nUti1Jr8Yjd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAatGvckNyU5luThJcvOSHJnkse7r6d3y5Pkk0kOJ3kwyfnjHF6StLxhjtw/A1x83LK9wMGq2gYc7K4DvBvY1v3ZA3xqNGNKktZi1bhX1T3AN45bvBPY313eD1y+ZPlna+Be4LQkm0c1rCRpOOs9535WVR3tLj8HnNVd3gI8s2S9I92y/yfJniQLSRYWFxfXOYYkaTm9n1CtqgJqHffbV1XzVTU/OzvbdwxJ0hLrjfvz/3e6pft6rFv+LHD2kvW2dsskSRO03rgfAHZ1l3cBty9Z/t7uVTMXAi8tOX0jSZqQmdVWSHIz8A5gU5IjwLXAdcCtSXYDTwFXdqt/AbgEOAx8E3jfGGaWJK1i1bhX1dUr3LRjmXUL+GDfoSRJ/fgOVUlqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAb1inuSX03ySJKHk9yc5JQk5yQ5lORwkluSnDyqYSVJw1l33JNsAX4ZmK+qc4GTgKuA64GPV9VbgBeA3aMYVJI0vL6nZWaA70kyA5wKHAXeCdzW3b4fuLznNiRJa7TuuFfVs8DvAk8ziPpLwH3Ai1X1SrfaEWDLcvdPsifJQpKFxcXF9Y4hSVpGn9MypwM7gXOA7wdeD1w87P2ral9VzVfV/Ozs7HrHkCQto89pmXcBX6uqxap6Gfg8cBFwWneaBmAr8GzPGSVJa9Qn7k8DFyY5NUmAHcCjwF3AFd06u4Db+40oSVqrPufcDzF44vTLwEPdY+0DPgJck+QwcCZw4wjmlCStwczqq6ysqq4Frj1u8RPABX0eV5LUj+9QlaQGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJapBxl6QGGXdJalCvuCc5LcltSf45yWNJ3pbkjCR3Jnm8+3r6qIaVJA2n75H7J4C/rqofBn4MeAzYCxysqm3Awe66JGmC1h33JG8C3g7cCFBV36qqF4GdwP5utf3A5X2HlCStTZ8j93OAReCPktyf5NNJXg+cVVVHu3WeA85a7s5J9iRZSLKwuLjYYwxJ0vH6xH0GOB/4VFWdB/wXx52CqaoCark7V9W+qpqvqvnZ2dkeY0iSjjfT475HgCNVdai7fhuDuD+fZHNVHU2yGTjWd0hJGqe5vXdMbdtPXnfpWB533UfuVfUc8EySH+oW7QAeBQ4Au7plu4Dbe00oSVqzPkfuAB8CPpfkZOAJ4H0M/sO4Nclu4Cngyp7bkCStUa+4V9UDwPwyN+3o87iSpH58h6okNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDesc9yUlJ7k/yl931c5IcSnI4yS1JTu4/piRpLUZx5P5h4LEl168HPl5VbwFeAHaPYBuSpDXoFfckW4FLgU931wO8E7itW2U/cHmfbUiS1q7vkfvvA78OfLu7fibwYlW90l0/AmxZ7o5J9iRZSLKwuLjYcwxJ0lLrjnuSnwaOVdV967l/Ve2rqvmqmp+dnV3vGJKkZcz0uO9FwGVJLgFOAd4IfAI4LclMd/S+FXi2/5iSpLVY95F7Vf1GVW2tqjngKuCLVfWzwF3AFd1qu4Dbe08pSVqTcbzO/SPANUkOMzgHf+MYtiFJehV9Tst8R1XdDdzdXX4CuGAUjytJWh/foSpJDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktSgdcc9ydlJ7kryaJJHkny4W35GkjuTPN59PX1040qShtHnyP0V4NeqajtwIfDBJNuBvcDBqtoGHOyuS5ImaN1xr6qjVfXl7vJ/AI8BW4CdwP5utf3A5X2HlCStzUjOuSeZA84DDgFnVdXR7qbngLNWuM+eJAtJFhYXF0cxhiSp0zvuSb4X+HPgV6rq35feVlUF1HL3q6p9VTVfVfOzs7N9x5AkLdEr7km+i0HYP1dVn+8WP59kc3f7ZuBYvxElSWvV59UyAW4EHquqjy256QCwq7u8C7h9/eNJktZjpsd9LwJ+DngoyQPdst8ErgNuTbIbeAq4st+IkqS1Wnfcq+rvgKxw8471Pq4kqT/foSpJDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktQg4y5JDTLuktSgscQ9ycVJvpLkcJK949iGJGllI497kpOAG4B3A9uBq5NsH/V2JEkrG8eR+wXA4ap6oqq+BfwpsHMM25EkrWBmDI+5BXhmyfUjwI8fv1KSPcCe7up/JvnKOre3Cfj6Ou/bS66fxlaBKe7zFLnPJ4YTbp9zfa99/sGVbhhH3IdSVfuAfX0fJ8lCVc2PYKQNw30+MbjPJ4Zx7fM4Tss8C5y95PrWbpkkaULGEfd/BLYlOSfJycBVwIExbEeStIKRn5apqleS/BLwN8BJwE1V9ciot7NE71M7G5D7fGJwn08MY9nnVNU4HleSNEW+Q1WSGmTcJalBGybuq32kQZLvTnJLd/uhJHOTn3K0htjna5I8muTBJAeTrPia141i2I+uSPIzSSrJhn/Z3DD7nOTK7nv9SJI/mfSMozbEz/YPJLkryf3dz/cl05hzVJLclORYkodXuD1JPtn9fTyY5PzeG62q1/wfBk/MfhV4M3Ay8E/A9uPW+UXgD7rLVwG3THvuCezzTwKndpc/cCLsc7feG4B7gHuB+WnPPYHv8zbgfuD07vr3TXvuCezzPuAD3eXtwJPTnrvnPr8dOB94eIXbLwH+CghwIXCo7zY3ypH7MB9psBPY312+DdiRJBOccdRW3eeququqvtldvZfBewo2smE/uuJ3gOuB/57kcGMyzD7/AnBDVb0AUFXHJjzjqA2zzwW8sbv8JuBfJzjfyFXVPcA3XmWVncBna+Be4LQkm/tsc6PEfbmPNNiy0jpV9QrwEnDmRKYbj2H2eandDP7n38hW3efu19Wzq+qOSQ42RsN8n98KvDXJl5Lcm+TiiU03HsPs828B70lyBPgC8KHJjDY1a/33vqqpffyARifJe4B54CemPcs4JXkd8DHg56c8yqTNMDg18w4Gv53dk+RHq+rFqU41XlcDn6mq30vyNuCPk5xbVd+e9mAbxUY5ch/mIw2+s06SGQa/yv3bRKYbj6E+xiHJu4CPApdV1f9MaLZxWW2f3wCcC9yd5EkG5yYPbPAnVYf5Ph8BDlTVy1X1NeBfGMR+oxpmn3cDtwJU1d8DpzD4ULFWjfxjWzZK3If5SIMDwK7u8hXAF6t7pmKDWnWfk5wH/CGDsG/087Cwyj5X1UtVtamq5qpqjsHzDJdV1cJ0xh2JYX62/4LBUTtJNjE4TfPEJIccsWH2+WlgB0CSH2EQ98WJTjlZB4D3dq+auRB4qaqO9nrEaT+LvIZnmy9hcMTyVeCj3bLfZvCPGwbf/D8DDgP/ALx52jNPYJ//FngeeKD7c2DaM497n49b9242+Ktlhvw+h8HpqEeBh4Crpj3zBPZ5O/AlBq+keQD4qWnP3HN/bwaOAi8z+E1sN/B+4P1Lvsc3dH8fD43i59qPH5CkBm2U0zKSpDUw7pLUIOMuSQ0y7pLUIOMuSQ0y7pLUIOMuSQ36X7ryi8l1uEYjAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tTKjzh8miMId"
      },
      "source": [
        "## Show correlation\n",
        "Here we are showing the correlation between each feature within the dataset.  Correlation is determined by analyzing the values of different features and comparing them with each other.  A 1 means that the values of the feature match exactly a value of -1 means that the values have very little in common.  It is important that we use a wide variety of features which have low correlation as it will yield more accurate results.  I have shown correlations of features for both the X and Y triaining and test sets."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GanMBBm4ZXul"
      },
      "source": [
        "### Training Set Correlation values"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "ftfJEYaBeJUe",
        "outputId": "0e538ffb-ec5e-4663-b72a-caec3f7a79d0"
      },
      "source": [
        "# Show correlation of our training set\n",
        "X_train.corr(method='pearson')"
      ],
      "execution_count": 1495,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Mean_Radius</th>\n",
              "      <th>Mean_Texture</th>\n",
              "      <th>Mean_Perimeter</th>\n",
              "      <th>Mean_Area</th>\n",
              "      <th>Mean_Smoothness</th>\n",
              "      <th>Mean_Compactness</th>\n",
              "      <th>Mean_Concavity</th>\n",
              "      <th>Mean_Concave_Points</th>\n",
              "      <th>Mean_Symmetry</th>\n",
              "      <th>Mean_Fractal_Dimension</th>\n",
              "      <th>Radius_SE</th>\n",
              "      <th>Texture_SE</th>\n",
              "      <th>Perimeter_SE</th>\n",
              "      <th>Area_SE</th>\n",
              "      <th>Smoothness_SE</th>\n",
              "      <th>Compactness_SE</th>\n",
              "      <th>Concavity_SE</th>\n",
              "      <th>Concave_Points_SE</th>\n",
              "      <th>Symmetry_SE</th>\n",
              "      <th>Fractal_Dimension_SE</th>\n",
              "      <th>Worst_Radius</th>\n",
              "      <th>Worst_Texture</th>\n",
              "      <th>Worst_Perimeter</th>\n",
              "      <th>Worst_Area</th>\n",
              "      <th>Worst_Smoothness</th>\n",
              "      <th>Worst_Compactness</th>\n",
              "      <th>Worst_Concavity</th>\n",
              "      <th>Worst_Concave_Points</th>\n",
              "      <th>Worst_Symmetry</th>\n",
              "      <th>Worst_Fractal_Dimension</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Mean_Radius</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.342784</td>\n",
              "      <td>0.998029</td>\n",
              "      <td>0.990185</td>\n",
              "      <td>0.102811</td>\n",
              "      <td>0.457384</td>\n",
              "      <td>0.617161</td>\n",
              "      <td>0.804515</td>\n",
              "      <td>0.096083</td>\n",
              "      <td>-0.351260</td>\n",
              "      <td>0.664788</td>\n",
              "      <td>-0.115480</td>\n",
              "      <td>0.658927</td>\n",
              "      <td>0.785774</td>\n",
              "      <td>-0.288878</td>\n",
              "      <td>0.159214</td>\n",
              "      <td>0.113084</td>\n",
              "      <td>0.354568</td>\n",
              "      <td>-0.175036</td>\n",
              "      <td>-0.092924</td>\n",
              "      <td>0.970345</td>\n",
              "      <td>0.324832</td>\n",
              "      <td>0.964967</td>\n",
              "      <td>0.940037</td>\n",
              "      <td>0.091686</td>\n",
              "      <td>0.391240</td>\n",
              "      <td>0.478547</td>\n",
              "      <td>0.736711</td>\n",
              "      <td>0.151488</td>\n",
              "      <td>-0.017289</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Texture</th>\n",
              "      <td>0.342784</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.350775</td>\n",
              "      <td>0.352883</td>\n",
              "      <td>-0.049680</td>\n",
              "      <td>0.250413</td>\n",
              "      <td>0.324380</td>\n",
              "      <td>0.312368</td>\n",
              "      <td>0.091119</td>\n",
              "      <td>-0.081113</td>\n",
              "      <td>0.338179</td>\n",
              "      <td>0.410560</td>\n",
              "      <td>0.352487</td>\n",
              "      <td>0.347251</td>\n",
              "      <td>0.019562</td>\n",
              "      <td>0.212215</td>\n",
              "      <td>0.136717</td>\n",
              "      <td>0.171747</td>\n",
              "      <td>0.044948</td>\n",
              "      <td>0.054149</td>\n",
              "      <td>0.374630</td>\n",
              "      <td>0.912376</td>\n",
              "      <td>0.384671</td>\n",
              "      <td>0.375565</td>\n",
              "      <td>0.077433</td>\n",
              "      <td>0.302354</td>\n",
              "      <td>0.318807</td>\n",
              "      <td>0.310821</td>\n",
              "      <td>0.137712</td>\n",
              "      <td>0.133442</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Perimeter</th>\n",
              "      <td>0.998029</td>\n",
              "      <td>0.350775</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.988509</td>\n",
              "      <td>0.136273</td>\n",
              "      <td>0.508723</td>\n",
              "      <td>0.656625</td>\n",
              "      <td>0.833393</td>\n",
              "      <td>0.129675</td>\n",
              "      <td>-0.303624</td>\n",
              "      <td>0.676031</td>\n",
              "      <td>-0.107550</td>\n",
              "      <td>0.676952</td>\n",
              "      <td>0.792317</td>\n",
              "      <td>-0.275531</td>\n",
              "      <td>0.204502</td>\n",
              "      <td>0.143392</td>\n",
              "      <td>0.384304</td>\n",
              "      <td>-0.155744</td>\n",
              "      <td>-0.055037</td>\n",
              "      <td>0.971154</td>\n",
              "      <td>0.333187</td>\n",
              "      <td>0.970609</td>\n",
              "      <td>0.940207</td>\n",
              "      <td>0.120301</td>\n",
              "      <td>0.435940</td>\n",
              "      <td>0.516874</td>\n",
              "      <td>0.765647</td>\n",
              "      <td>0.177322</td>\n",
              "      <td>0.029087</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Area</th>\n",
              "      <td>0.990185</td>\n",
              "      <td>0.352883</td>\n",
              "      <td>0.988509</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.099790</td>\n",
              "      <td>0.450671</td>\n",
              "      <td>0.623074</td>\n",
              "      <td>0.806317</td>\n",
              "      <td>0.096768</td>\n",
              "      <td>-0.327142</td>\n",
              "      <td>0.698584</td>\n",
              "      <td>-0.091465</td>\n",
              "      <td>0.689057</td>\n",
              "      <td>0.821103</td>\n",
              "      <td>-0.245199</td>\n",
              "      <td>0.164741</td>\n",
              "      <td>0.124254</td>\n",
              "      <td>0.352521</td>\n",
              "      <td>-0.156043</td>\n",
              "      <td>-0.070391</td>\n",
              "      <td>0.971221</td>\n",
              "      <td>0.330651</td>\n",
              "      <td>0.965227</td>\n",
              "      <td>0.960805</td>\n",
              "      <td>0.099242</td>\n",
              "      <td>0.374792</td>\n",
              "      <td>0.470091</td>\n",
              "      <td>0.723748</td>\n",
              "      <td>0.138784</td>\n",
              "      <td>-0.014537</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Smoothness</th>\n",
              "      <td>0.102811</td>\n",
              "      <td>-0.049680</td>\n",
              "      <td>0.136273</td>\n",
              "      <td>0.099790</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.621257</td>\n",
              "      <td>0.462496</td>\n",
              "      <td>0.505894</td>\n",
              "      <td>0.568329</td>\n",
              "      <td>0.570408</td>\n",
              "      <td>0.251055</td>\n",
              "      <td>0.062629</td>\n",
              "      <td>0.230254</td>\n",
              "      <td>0.191509</td>\n",
              "      <td>0.315067</td>\n",
              "      <td>0.283035</td>\n",
              "      <td>0.186777</td>\n",
              "      <td>0.351371</td>\n",
              "      <td>0.151525</td>\n",
              "      <td>0.291836</td>\n",
              "      <td>0.148347</td>\n",
              "      <td>-0.000243</td>\n",
              "      <td>0.166861</td>\n",
              "      <td>0.135721</td>\n",
              "      <td>0.796408</td>\n",
              "      <td>0.430580</td>\n",
              "      <td>0.378667</td>\n",
              "      <td>0.453335</td>\n",
              "      <td>0.366940</td>\n",
              "      <td>0.498688</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Compactness</th>\n",
              "      <td>0.457384</td>\n",
              "      <td>0.250413</td>\n",
              "      <td>0.508723</td>\n",
              "      <td>0.450671</td>\n",
              "      <td>0.621257</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.864799</td>\n",
              "      <td>0.813006</td>\n",
              "      <td>0.587692</td>\n",
              "      <td>0.576419</td>\n",
              "      <td>0.480127</td>\n",
              "      <td>0.027176</td>\n",
              "      <td>0.541446</td>\n",
              "      <td>0.466100</td>\n",
              "      <td>0.077014</td>\n",
              "      <td>0.741623</td>\n",
              "      <td>0.503673</td>\n",
              "      <td>0.614237</td>\n",
              "      <td>0.185882</td>\n",
              "      <td>0.529800</td>\n",
              "      <td>0.494415</td>\n",
              "      <td>0.261283</td>\n",
              "      <td>0.547876</td>\n",
              "      <td>0.466019</td>\n",
              "      <td>0.548217</td>\n",
              "      <td>0.873754</td>\n",
              "      <td>0.800407</td>\n",
              "      <td>0.806002</td>\n",
              "      <td>0.510341</td>\n",
              "      <td>0.727751</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Concavity</th>\n",
              "      <td>0.617161</td>\n",
              "      <td>0.324380</td>\n",
              "      <td>0.656625</td>\n",
              "      <td>0.623074</td>\n",
              "      <td>0.462496</td>\n",
              "      <td>0.864799</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.900771</td>\n",
              "      <td>0.466296</td>\n",
              "      <td>0.349558</td>\n",
              "      <td>0.583680</td>\n",
              "      <td>0.066258</td>\n",
              "      <td>0.612064</td>\n",
              "      <td>0.596462</td>\n",
              "      <td>0.027889</td>\n",
              "      <td>0.687677</td>\n",
              "      <td>0.677454</td>\n",
              "      <td>0.690478</td>\n",
              "      <td>0.134503</td>\n",
              "      <td>0.492497</td>\n",
              "      <td>0.635617</td>\n",
              "      <td>0.326433</td>\n",
              "      <td>0.674534</td>\n",
              "      <td>0.616527</td>\n",
              "      <td>0.419895</td>\n",
              "      <td>0.759300</td>\n",
              "      <td>0.892644</td>\n",
              "      <td>0.855938</td>\n",
              "      <td>0.407975</td>\n",
              "      <td>0.551995</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Concave_Points</th>\n",
              "      <td>0.804515</td>\n",
              "      <td>0.312368</td>\n",
              "      <td>0.833393</td>\n",
              "      <td>0.806317</td>\n",
              "      <td>0.505894</td>\n",
              "      <td>0.813006</td>\n",
              "      <td>0.900771</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.431457</td>\n",
              "      <td>0.151364</td>\n",
              "      <td>0.695170</td>\n",
              "      <td>-0.002931</td>\n",
              "      <td>0.707419</td>\n",
              "      <td>0.737823</td>\n",
              "      <td>-0.039266</td>\n",
              "      <td>0.471819</td>\n",
              "      <td>0.372932</td>\n",
              "      <td>0.600347</td>\n",
              "      <td>0.042623</td>\n",
              "      <td>0.250851</td>\n",
              "      <td>0.815428</td>\n",
              "      <td>0.315599</td>\n",
              "      <td>0.839102</td>\n",
              "      <td>0.790571</td>\n",
              "      <td>0.437221</td>\n",
              "      <td>0.669451</td>\n",
              "      <td>0.735465</td>\n",
              "      <td>0.914296</td>\n",
              "      <td>0.374413</td>\n",
              "      <td>0.386695</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Symmetry</th>\n",
              "      <td>0.096083</td>\n",
              "      <td>0.091119</td>\n",
              "      <td>0.129675</td>\n",
              "      <td>0.096768</td>\n",
              "      <td>0.568329</td>\n",
              "      <td>0.587692</td>\n",
              "      <td>0.466296</td>\n",
              "      <td>0.431457</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.502564</td>\n",
              "      <td>0.308992</td>\n",
              "      <td>0.117461</td>\n",
              "      <td>0.318942</td>\n",
              "      <td>0.224227</td>\n",
              "      <td>0.159431</td>\n",
              "      <td>0.398990</td>\n",
              "      <td>0.291854</td>\n",
              "      <td>0.355331</td>\n",
              "      <td>0.429499</td>\n",
              "      <td>0.360228</td>\n",
              "      <td>0.142589</td>\n",
              "      <td>0.098356</td>\n",
              "      <td>0.170618</td>\n",
              "      <td>0.128180</td>\n",
              "      <td>0.443890</td>\n",
              "      <td>0.452706</td>\n",
              "      <td>0.397617</td>\n",
              "      <td>0.402276</td>\n",
              "      <td>0.682648</td>\n",
              "      <td>0.467627</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Fractal_Dimension</th>\n",
              "      <td>-0.351260</td>\n",
              "      <td>-0.081113</td>\n",
              "      <td>-0.303624</td>\n",
              "      <td>-0.327142</td>\n",
              "      <td>0.570408</td>\n",
              "      <td>0.576419</td>\n",
              "      <td>0.349558</td>\n",
              "      <td>0.151364</td>\n",
              "      <td>0.502564</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.016853</td>\n",
              "      <td>0.136238</td>\n",
              "      <td>0.021898</td>\n",
              "      <td>-0.128846</td>\n",
              "      <td>0.384186</td>\n",
              "      <td>0.573566</td>\n",
              "      <td>0.440490</td>\n",
              "      <td>0.331586</td>\n",
              "      <td>0.321876</td>\n",
              "      <td>0.722503</td>\n",
              "      <td>-0.287248</td>\n",
              "      <td>-0.070033</td>\n",
              "      <td>-0.243438</td>\n",
              "      <td>-0.267521</td>\n",
              "      <td>0.486748</td>\n",
              "      <td>0.449830</td>\n",
              "      <td>0.347428</td>\n",
              "      <td>0.160840</td>\n",
              "      <td>0.328546</td>\n",
              "      <td>0.782754</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Radius_SE</th>\n",
              "      <td>0.664788</td>\n",
              "      <td>0.338179</td>\n",
              "      <td>0.676031</td>\n",
              "      <td>0.698584</td>\n",
              "      <td>0.251055</td>\n",
              "      <td>0.480127</td>\n",
              "      <td>0.583680</td>\n",
              "      <td>0.695170</td>\n",
              "      <td>0.308992</td>\n",
              "      <td>-0.016853</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.218173</td>\n",
              "      <td>0.966179</td>\n",
              "      <td>0.953630</td>\n",
              "      <td>0.097730</td>\n",
              "      <td>0.351825</td>\n",
              "      <td>0.267896</td>\n",
              "      <td>0.533028</td>\n",
              "      <td>0.172927</td>\n",
              "      <td>0.225358</td>\n",
              "      <td>0.734633</td>\n",
              "      <td>0.257195</td>\n",
              "      <td>0.737344</td>\n",
              "      <td>0.756334</td>\n",
              "      <td>0.149532</td>\n",
              "      <td>0.306757</td>\n",
              "      <td>0.363172</td>\n",
              "      <td>0.556075</td>\n",
              "      <td>0.125339</td>\n",
              "      <td>0.080889</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Texture_SE</th>\n",
              "      <td>-0.115480</td>\n",
              "      <td>0.410560</td>\n",
              "      <td>-0.107550</td>\n",
              "      <td>-0.091465</td>\n",
              "      <td>0.062629</td>\n",
              "      <td>0.027176</td>\n",
              "      <td>0.066258</td>\n",
              "      <td>-0.002931</td>\n",
              "      <td>0.117461</td>\n",
              "      <td>0.136238</td>\n",
              "      <td>0.218173</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.245384</td>\n",
              "      <td>0.095883</td>\n",
              "      <td>0.367638</td>\n",
              "      <td>0.211091</td>\n",
              "      <td>0.202243</td>\n",
              "      <td>0.263575</td>\n",
              "      <td>0.356962</td>\n",
              "      <td>0.260974</td>\n",
              "      <td>-0.116451</td>\n",
              "      <td>0.441371</td>\n",
              "      <td>-0.107391</td>\n",
              "      <td>-0.096073</td>\n",
              "      <td>-0.050520</td>\n",
              "      <td>-0.085576</td>\n",
              "      <td>-0.054827</td>\n",
              "      <td>-0.118234</td>\n",
              "      <td>-0.123741</td>\n",
              "      <td>-0.022580</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Perimeter_SE</th>\n",
              "      <td>0.658927</td>\n",
              "      <td>0.352487</td>\n",
              "      <td>0.676952</td>\n",
              "      <td>0.689057</td>\n",
              "      <td>0.230254</td>\n",
              "      <td>0.541446</td>\n",
              "      <td>0.612064</td>\n",
              "      <td>0.707419</td>\n",
              "      <td>0.318942</td>\n",
              "      <td>0.021898</td>\n",
              "      <td>0.966179</td>\n",
              "      <td>0.245384</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.924946</td>\n",
              "      <td>0.075491</td>\n",
              "      <td>0.432708</td>\n",
              "      <td>0.297668</td>\n",
              "      <td>0.594032</td>\n",
              "      <td>0.210053</td>\n",
              "      <td>0.251712</td>\n",
              "      <td>0.716133</td>\n",
              "      <td>0.270151</td>\n",
              "      <td>0.740018</td>\n",
              "      <td>0.732008</td>\n",
              "      <td>0.122622</td>\n",
              "      <td>0.370477</td>\n",
              "      <td>0.405817</td>\n",
              "      <td>0.586121</td>\n",
              "      <td>0.139924</td>\n",
              "      <td>0.119487</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Area_SE</th>\n",
              "      <td>0.785774</td>\n",
              "      <td>0.347251</td>\n",
              "      <td>0.792317</td>\n",
              "      <td>0.821103</td>\n",
              "      <td>0.191509</td>\n",
              "      <td>0.466100</td>\n",
              "      <td>0.596462</td>\n",
              "      <td>0.737823</td>\n",
              "      <td>0.224227</td>\n",
              "      <td>-0.128846</td>\n",
              "      <td>0.953630</td>\n",
              "      <td>0.095883</td>\n",
              "      <td>0.924946</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.028281</td>\n",
              "      <td>0.276891</td>\n",
              "      <td>0.204468</td>\n",
              "      <td>0.456445</td>\n",
              "      <td>0.038718</td>\n",
              "      <td>0.109097</td>\n",
              "      <td>0.852403</td>\n",
              "      <td>0.295088</td>\n",
              "      <td>0.851589</td>\n",
              "      <td>0.884797</td>\n",
              "      <td>0.143761</td>\n",
              "      <td>0.329161</td>\n",
              "      <td>0.399481</td>\n",
              "      <td>0.620649</td>\n",
              "      <td>0.123026</td>\n",
              "      <td>0.050910</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Smoothness_SE</th>\n",
              "      <td>-0.288878</td>\n",
              "      <td>0.019562</td>\n",
              "      <td>-0.275531</td>\n",
              "      <td>-0.245199</td>\n",
              "      <td>0.315067</td>\n",
              "      <td>0.077014</td>\n",
              "      <td>0.027889</td>\n",
              "      <td>-0.039266</td>\n",
              "      <td>0.159431</td>\n",
              "      <td>0.384186</td>\n",
              "      <td>0.097730</td>\n",
              "      <td>0.367638</td>\n",
              "      <td>0.075491</td>\n",
              "      <td>-0.028281</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.296741</td>\n",
              "      <td>0.221427</td>\n",
              "      <td>0.292568</td>\n",
              "      <td>0.335873</td>\n",
              "      <td>0.429959</td>\n",
              "      <td>-0.276354</td>\n",
              "      <td>-0.069574</td>\n",
              "      <td>-0.270142</td>\n",
              "      <td>-0.235627</td>\n",
              "      <td>0.346370</td>\n",
              "      <td>-0.083128</td>\n",
              "      <td>-0.092540</td>\n",
              "      <td>-0.138672</td>\n",
              "      <td>-0.130308</td>\n",
              "      <td>0.117333</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Compactness_SE</th>\n",
              "      <td>0.159214</td>\n",
              "      <td>0.212215</td>\n",
              "      <td>0.204502</td>\n",
              "      <td>0.164741</td>\n",
              "      <td>0.283035</td>\n",
              "      <td>0.741623</td>\n",
              "      <td>0.687677</td>\n",
              "      <td>0.471819</td>\n",
              "      <td>0.398990</td>\n",
              "      <td>0.573566</td>\n",
              "      <td>0.351825</td>\n",
              "      <td>0.211091</td>\n",
              "      <td>0.432708</td>\n",
              "      <td>0.276891</td>\n",
              "      <td>0.296741</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.778496</td>\n",
              "      <td>0.743105</td>\n",
              "      <td>0.359843</td>\n",
              "      <td>0.809284</td>\n",
              "      <td>0.172324</td>\n",
              "      <td>0.156691</td>\n",
              "      <td>0.227882</td>\n",
              "      <td>0.161273</td>\n",
              "      <td>0.217038</td>\n",
              "      <td>0.688768</td>\n",
              "      <td>0.661867</td>\n",
              "      <td>0.486585</td>\n",
              "      <td>0.274858</td>\n",
              "      <td>0.623452</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Concavity_SE</th>\n",
              "      <td>0.113084</td>\n",
              "      <td>0.136717</td>\n",
              "      <td>0.143392</td>\n",
              "      <td>0.124254</td>\n",
              "      <td>0.186777</td>\n",
              "      <td>0.503673</td>\n",
              "      <td>0.677454</td>\n",
              "      <td>0.372932</td>\n",
              "      <td>0.291854</td>\n",
              "      <td>0.440490</td>\n",
              "      <td>0.267896</td>\n",
              "      <td>0.202243</td>\n",
              "      <td>0.297668</td>\n",
              "      <td>0.204468</td>\n",
              "      <td>0.221427</td>\n",
              "      <td>0.778496</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.757885</td>\n",
              "      <td>0.282085</td>\n",
              "      <td>0.744891</td>\n",
              "      <td>0.112143</td>\n",
              "      <td>0.088711</td>\n",
              "      <td>0.146781</td>\n",
              "      <td>0.111633</td>\n",
              "      <td>0.121901</td>\n",
              "      <td>0.429624</td>\n",
              "      <td>0.649241</td>\n",
              "      <td>0.388847</td>\n",
              "      <td>0.163697</td>\n",
              "      <td>0.425365</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Concave_Points_SE</th>\n",
              "      <td>0.354568</td>\n",
              "      <td>0.171747</td>\n",
              "      <td>0.384304</td>\n",
              "      <td>0.352521</td>\n",
              "      <td>0.351371</td>\n",
              "      <td>0.614237</td>\n",
              "      <td>0.690478</td>\n",
              "      <td>0.600347</td>\n",
              "      <td>0.355331</td>\n",
              "      <td>0.331586</td>\n",
              "      <td>0.533028</td>\n",
              "      <td>0.263575</td>\n",
              "      <td>0.594032</td>\n",
              "      <td>0.456445</td>\n",
              "      <td>0.292568</td>\n",
              "      <td>0.743105</td>\n",
              "      <td>0.757885</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.308025</td>\n",
              "      <td>0.628327</td>\n",
              "      <td>0.332525</td>\n",
              "      <td>0.082556</td>\n",
              "      <td>0.369590</td>\n",
              "      <td>0.312726</td>\n",
              "      <td>0.190546</td>\n",
              "      <td>0.432359</td>\n",
              "      <td>0.545222</td>\n",
              "      <td>0.590001</td>\n",
              "      <td>0.116274</td>\n",
              "      <td>0.314526</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Symmetry_SE</th>\n",
              "      <td>-0.175036</td>\n",
              "      <td>0.044948</td>\n",
              "      <td>-0.155744</td>\n",
              "      <td>-0.156043</td>\n",
              "      <td>0.151525</td>\n",
              "      <td>0.185882</td>\n",
              "      <td>0.134503</td>\n",
              "      <td>0.042623</td>\n",
              "      <td>0.429499</td>\n",
              "      <td>0.321876</td>\n",
              "      <td>0.172927</td>\n",
              "      <td>0.356962</td>\n",
              "      <td>0.210053</td>\n",
              "      <td>0.038718</td>\n",
              "      <td>0.335873</td>\n",
              "      <td>0.359843</td>\n",
              "      <td>0.282085</td>\n",
              "      <td>0.308025</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.323507</td>\n",
              "      <td>-0.174954</td>\n",
              "      <td>-0.055603</td>\n",
              "      <td>-0.151618</td>\n",
              "      <td>-0.166653</td>\n",
              "      <td>-0.020366</td>\n",
              "      <td>0.057354</td>\n",
              "      <td>0.035767</td>\n",
              "      <td>-0.039922</td>\n",
              "      <td>0.409618</td>\n",
              "      <td>0.099998</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Fractal_Dimension_SE</th>\n",
              "      <td>-0.092924</td>\n",
              "      <td>0.054149</td>\n",
              "      <td>-0.055037</td>\n",
              "      <td>-0.070391</td>\n",
              "      <td>0.291836</td>\n",
              "      <td>0.529800</td>\n",
              "      <td>0.492497</td>\n",
              "      <td>0.250851</td>\n",
              "      <td>0.360228</td>\n",
              "      <td>0.722503</td>\n",
              "      <td>0.225358</td>\n",
              "      <td>0.260974</td>\n",
              "      <td>0.251712</td>\n",
              "      <td>0.109097</td>\n",
              "      <td>0.429959</td>\n",
              "      <td>0.809284</td>\n",
              "      <td>0.744891</td>\n",
              "      <td>0.628327</td>\n",
              "      <td>0.323507</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.071879</td>\n",
              "      <td>-0.004476</td>\n",
              "      <td>-0.034697</td>\n",
              "      <td>-0.057999</td>\n",
              "      <td>0.197866</td>\n",
              "      <td>0.404105</td>\n",
              "      <td>0.413000</td>\n",
              "      <td>0.223259</td>\n",
              "      <td>0.121409</td>\n",
              "      <td>0.627468</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Radius</th>\n",
              "      <td>0.970345</td>\n",
              "      <td>0.374630</td>\n",
              "      <td>0.971154</td>\n",
              "      <td>0.971221</td>\n",
              "      <td>0.148347</td>\n",
              "      <td>0.494415</td>\n",
              "      <td>0.635617</td>\n",
              "      <td>0.815428</td>\n",
              "      <td>0.142589</td>\n",
              "      <td>-0.287248</td>\n",
              "      <td>0.734633</td>\n",
              "      <td>-0.116451</td>\n",
              "      <td>0.716133</td>\n",
              "      <td>0.852403</td>\n",
              "      <td>-0.276354</td>\n",
              "      <td>0.172324</td>\n",
              "      <td>0.112143</td>\n",
              "      <td>0.332525</td>\n",
              "      <td>-0.174954</td>\n",
              "      <td>-0.071879</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.389386</td>\n",
              "      <td>0.994048</td>\n",
              "      <td>0.986414</td>\n",
              "      <td>0.190447</td>\n",
              "      <td>0.457645</td>\n",
              "      <td>0.524263</td>\n",
              "      <td>0.777076</td>\n",
              "      <td>0.234110</td>\n",
              "      <td>0.071938</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Texture</th>\n",
              "      <td>0.324832</td>\n",
              "      <td>0.912376</td>\n",
              "      <td>0.333187</td>\n",
              "      <td>0.330651</td>\n",
              "      <td>-0.000243</td>\n",
              "      <td>0.261283</td>\n",
              "      <td>0.326433</td>\n",
              "      <td>0.315599</td>\n",
              "      <td>0.098356</td>\n",
              "      <td>-0.070033</td>\n",
              "      <td>0.257195</td>\n",
              "      <td>0.441371</td>\n",
              "      <td>0.270151</td>\n",
              "      <td>0.295088</td>\n",
              "      <td>-0.069574</td>\n",
              "      <td>0.156691</td>\n",
              "      <td>0.088711</td>\n",
              "      <td>0.082556</td>\n",
              "      <td>-0.055603</td>\n",
              "      <td>-0.004476</td>\n",
              "      <td>0.389386</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.399371</td>\n",
              "      <td>0.387599</td>\n",
              "      <td>0.213777</td>\n",
              "      <td>0.385015</td>\n",
              "      <td>0.385807</td>\n",
              "      <td>0.376103</td>\n",
              "      <td>0.258020</td>\n",
              "      <td>0.228657</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Perimeter</th>\n",
              "      <td>0.964967</td>\n",
              "      <td>0.384671</td>\n",
              "      <td>0.970609</td>\n",
              "      <td>0.965227</td>\n",
              "      <td>0.166861</td>\n",
              "      <td>0.547876</td>\n",
              "      <td>0.674534</td>\n",
              "      <td>0.839102</td>\n",
              "      <td>0.170618</td>\n",
              "      <td>-0.243438</td>\n",
              "      <td>0.737344</td>\n",
              "      <td>-0.107391</td>\n",
              "      <td>0.740018</td>\n",
              "      <td>0.851589</td>\n",
              "      <td>-0.270142</td>\n",
              "      <td>0.227882</td>\n",
              "      <td>0.146781</td>\n",
              "      <td>0.369590</td>\n",
              "      <td>-0.151618</td>\n",
              "      <td>-0.034697</td>\n",
              "      <td>0.994048</td>\n",
              "      <td>0.399371</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.978649</td>\n",
              "      <td>0.205154</td>\n",
              "      <td>0.511641</td>\n",
              "      <td>0.568750</td>\n",
              "      <td>0.807557</td>\n",
              "      <td>0.259310</td>\n",
              "      <td>0.117633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Area</th>\n",
              "      <td>0.940037</td>\n",
              "      <td>0.375565</td>\n",
              "      <td>0.940207</td>\n",
              "      <td>0.960805</td>\n",
              "      <td>0.135721</td>\n",
              "      <td>0.466019</td>\n",
              "      <td>0.616527</td>\n",
              "      <td>0.790571</td>\n",
              "      <td>0.128180</td>\n",
              "      <td>-0.267521</td>\n",
              "      <td>0.756334</td>\n",
              "      <td>-0.096073</td>\n",
              "      <td>0.732008</td>\n",
              "      <td>0.884797</td>\n",
              "      <td>-0.235627</td>\n",
              "      <td>0.161273</td>\n",
              "      <td>0.111633</td>\n",
              "      <td>0.312726</td>\n",
              "      <td>-0.166653</td>\n",
              "      <td>-0.057999</td>\n",
              "      <td>0.986414</td>\n",
              "      <td>0.387599</td>\n",
              "      <td>0.978649</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.186941</td>\n",
              "      <td>0.421864</td>\n",
              "      <td>0.495458</td>\n",
              "      <td>0.740198</td>\n",
              "      <td>0.202392</td>\n",
              "      <td>0.062320</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Smoothness</th>\n",
              "      <td>0.091686</td>\n",
              "      <td>0.077433</td>\n",
              "      <td>0.120301</td>\n",
              "      <td>0.099242</td>\n",
              "      <td>0.796408</td>\n",
              "      <td>0.548217</td>\n",
              "      <td>0.419895</td>\n",
              "      <td>0.437221</td>\n",
              "      <td>0.443890</td>\n",
              "      <td>0.486748</td>\n",
              "      <td>0.149532</td>\n",
              "      <td>-0.050520</td>\n",
              "      <td>0.122622</td>\n",
              "      <td>0.143761</td>\n",
              "      <td>0.346370</td>\n",
              "      <td>0.217038</td>\n",
              "      <td>0.121901</td>\n",
              "      <td>0.190546</td>\n",
              "      <td>-0.020366</td>\n",
              "      <td>0.197866</td>\n",
              "      <td>0.190447</td>\n",
              "      <td>0.213777</td>\n",
              "      <td>0.205154</td>\n",
              "      <td>0.186941</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.536802</td>\n",
              "      <td>0.471253</td>\n",
              "      <td>0.518835</td>\n",
              "      <td>0.479026</td>\n",
              "      <td>0.607300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Compactness</th>\n",
              "      <td>0.391240</td>\n",
              "      <td>0.302354</td>\n",
              "      <td>0.435940</td>\n",
              "      <td>0.374792</td>\n",
              "      <td>0.430580</td>\n",
              "      <td>0.873754</td>\n",
              "      <td>0.759300</td>\n",
              "      <td>0.669451</td>\n",
              "      <td>0.452706</td>\n",
              "      <td>0.449830</td>\n",
              "      <td>0.306757</td>\n",
              "      <td>-0.085576</td>\n",
              "      <td>0.370477</td>\n",
              "      <td>0.329161</td>\n",
              "      <td>-0.083128</td>\n",
              "      <td>0.688768</td>\n",
              "      <td>0.429624</td>\n",
              "      <td>0.432359</td>\n",
              "      <td>0.057354</td>\n",
              "      <td>0.404105</td>\n",
              "      <td>0.457645</td>\n",
              "      <td>0.385015</td>\n",
              "      <td>0.511641</td>\n",
              "      <td>0.421864</td>\n",
              "      <td>0.536802</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.879920</td>\n",
              "      <td>0.797435</td>\n",
              "      <td>0.620852</td>\n",
              "      <td>0.813844</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Concavity</th>\n",
              "      <td>0.478547</td>\n",
              "      <td>0.318807</td>\n",
              "      <td>0.516874</td>\n",
              "      <td>0.470091</td>\n",
              "      <td>0.378667</td>\n",
              "      <td>0.800407</td>\n",
              "      <td>0.892644</td>\n",
              "      <td>0.735465</td>\n",
              "      <td>0.397617</td>\n",
              "      <td>0.347428</td>\n",
              "      <td>0.363172</td>\n",
              "      <td>-0.054827</td>\n",
              "      <td>0.405817</td>\n",
              "      <td>0.399481</td>\n",
              "      <td>-0.092540</td>\n",
              "      <td>0.661867</td>\n",
              "      <td>0.649241</td>\n",
              "      <td>0.545222</td>\n",
              "      <td>0.035767</td>\n",
              "      <td>0.413000</td>\n",
              "      <td>0.524263</td>\n",
              "      <td>0.385807</td>\n",
              "      <td>0.568750</td>\n",
              "      <td>0.495458</td>\n",
              "      <td>0.471253</td>\n",
              "      <td>0.879920</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.836521</td>\n",
              "      <td>0.525866</td>\n",
              "      <td>0.686121</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Concave_Points</th>\n",
              "      <td>0.736711</td>\n",
              "      <td>0.310821</td>\n",
              "      <td>0.765647</td>\n",
              "      <td>0.723748</td>\n",
              "      <td>0.453335</td>\n",
              "      <td>0.806002</td>\n",
              "      <td>0.855938</td>\n",
              "      <td>0.914296</td>\n",
              "      <td>0.402276</td>\n",
              "      <td>0.160840</td>\n",
              "      <td>0.556075</td>\n",
              "      <td>-0.118234</td>\n",
              "      <td>0.586121</td>\n",
              "      <td>0.620649</td>\n",
              "      <td>-0.138672</td>\n",
              "      <td>0.486585</td>\n",
              "      <td>0.388847</td>\n",
              "      <td>0.590001</td>\n",
              "      <td>-0.039922</td>\n",
              "      <td>0.223259</td>\n",
              "      <td>0.777076</td>\n",
              "      <td>0.376103</td>\n",
              "      <td>0.807557</td>\n",
              "      <td>0.740198</td>\n",
              "      <td>0.518835</td>\n",
              "      <td>0.797435</td>\n",
              "      <td>0.836521</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.501731</td>\n",
              "      <td>0.512531</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Symmetry</th>\n",
              "      <td>0.151488</td>\n",
              "      <td>0.137712</td>\n",
              "      <td>0.177322</td>\n",
              "      <td>0.138784</td>\n",
              "      <td>0.366940</td>\n",
              "      <td>0.510341</td>\n",
              "      <td>0.407975</td>\n",
              "      <td>0.374413</td>\n",
              "      <td>0.682648</td>\n",
              "      <td>0.328546</td>\n",
              "      <td>0.125339</td>\n",
              "      <td>-0.123741</td>\n",
              "      <td>0.139924</td>\n",
              "      <td>0.123026</td>\n",
              "      <td>-0.130308</td>\n",
              "      <td>0.274858</td>\n",
              "      <td>0.163697</td>\n",
              "      <td>0.116274</td>\n",
              "      <td>0.409618</td>\n",
              "      <td>0.121409</td>\n",
              "      <td>0.234110</td>\n",
              "      <td>0.258020</td>\n",
              "      <td>0.259310</td>\n",
              "      <td>0.202392</td>\n",
              "      <td>0.479026</td>\n",
              "      <td>0.620852</td>\n",
              "      <td>0.525866</td>\n",
              "      <td>0.501731</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.548598</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Fractal_Dimension</th>\n",
              "      <td>-0.017289</td>\n",
              "      <td>0.133442</td>\n",
              "      <td>0.029087</td>\n",
              "      <td>-0.014537</td>\n",
              "      <td>0.498688</td>\n",
              "      <td>0.727751</td>\n",
              "      <td>0.551995</td>\n",
              "      <td>0.386695</td>\n",
              "      <td>0.467627</td>\n",
              "      <td>0.782754</td>\n",
              "      <td>0.080889</td>\n",
              "      <td>-0.022580</td>\n",
              "      <td>0.119487</td>\n",
              "      <td>0.050910</td>\n",
              "      <td>0.117333</td>\n",
              "      <td>0.623452</td>\n",
              "      <td>0.425365</td>\n",
              "      <td>0.314526</td>\n",
              "      <td>0.099998</td>\n",
              "      <td>0.627468</td>\n",
              "      <td>0.071938</td>\n",
              "      <td>0.228657</td>\n",
              "      <td>0.117633</td>\n",
              "      <td>0.062320</td>\n",
              "      <td>0.607300</td>\n",
              "      <td>0.813844</td>\n",
              "      <td>0.686121</td>\n",
              "      <td>0.512531</td>\n",
              "      <td>0.548598</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                         Mean_Radius  ...  Worst_Fractal_Dimension\n",
              "Mean_Radius                 1.000000  ...                -0.017289\n",
              "Mean_Texture                0.342784  ...                 0.133442\n",
              "Mean_Perimeter              0.998029  ...                 0.029087\n",
              "Mean_Area                   0.990185  ...                -0.014537\n",
              "Mean_Smoothness             0.102811  ...                 0.498688\n",
              "Mean_Compactness            0.457384  ...                 0.727751\n",
              "Mean_Concavity              0.617161  ...                 0.551995\n",
              "Mean_Concave_Points         0.804515  ...                 0.386695\n",
              "Mean_Symmetry               0.096083  ...                 0.467627\n",
              "Mean_Fractal_Dimension     -0.351260  ...                 0.782754\n",
              "Radius_SE                   0.664788  ...                 0.080889\n",
              "Texture_SE                 -0.115480  ...                -0.022580\n",
              "Perimeter_SE                0.658927  ...                 0.119487\n",
              "Area_SE                     0.785774  ...                 0.050910\n",
              "Smoothness_SE              -0.288878  ...                 0.117333\n",
              "Compactness_SE              0.159214  ...                 0.623452\n",
              "Concavity_SE                0.113084  ...                 0.425365\n",
              "Concave_Points_SE           0.354568  ...                 0.314526\n",
              "Symmetry_SE                -0.175036  ...                 0.099998\n",
              "Fractal_Dimension_SE       -0.092924  ...                 0.627468\n",
              "Worst_Radius                0.970345  ...                 0.071938\n",
              "Worst_Texture               0.324832  ...                 0.228657\n",
              "Worst_Perimeter             0.964967  ...                 0.117633\n",
              "Worst_Area                  0.940037  ...                 0.062320\n",
              "Worst_Smoothness            0.091686  ...                 0.607300\n",
              "Worst_Compactness           0.391240  ...                 0.813844\n",
              "Worst_Concavity             0.478547  ...                 0.686121\n",
              "Worst_Concave_Points        0.736711  ...                 0.512531\n",
              "Worst_Symmetry              0.151488  ...                 0.548598\n",
              "Worst_Fractal_Dimension    -0.017289  ...                 1.000000\n",
              "\n",
              "[30 rows x 30 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 1495
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZYDI3zYOZec8"
      },
      "source": [
        "### Test set correlation values"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "kEhxeH7seAAl",
        "outputId": "7935fae0-40b2-4f0a-c356-f6011de90ab0"
      },
      "source": [
        "X_test.corr(method='pearson')"
      ],
      "execution_count": 1496,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Mean_Radius</th>\n",
              "      <th>Mean_Texture</th>\n",
              "      <th>Mean_Perimeter</th>\n",
              "      <th>Mean_Area</th>\n",
              "      <th>Mean_Smoothness</th>\n",
              "      <th>Mean_Compactness</th>\n",
              "      <th>Mean_Concavity</th>\n",
              "      <th>Mean_Concave_Points</th>\n",
              "      <th>Mean_Symmetry</th>\n",
              "      <th>Mean_Fractal_Dimension</th>\n",
              "      <th>Radius_SE</th>\n",
              "      <th>Texture_SE</th>\n",
              "      <th>Perimeter_SE</th>\n",
              "      <th>Area_SE</th>\n",
              "      <th>Smoothness_SE</th>\n",
              "      <th>Compactness_SE</th>\n",
              "      <th>Concavity_SE</th>\n",
              "      <th>Concave_Points_SE</th>\n",
              "      <th>Symmetry_SE</th>\n",
              "      <th>Fractal_Dimension_SE</th>\n",
              "      <th>Worst_Radius</th>\n",
              "      <th>Worst_Texture</th>\n",
              "      <th>Worst_Perimeter</th>\n",
              "      <th>Worst_Area</th>\n",
              "      <th>Worst_Smoothness</th>\n",
              "      <th>Worst_Compactness</th>\n",
              "      <th>Worst_Concavity</th>\n",
              "      <th>Worst_Concave_Points</th>\n",
              "      <th>Worst_Symmetry</th>\n",
              "      <th>Worst_Fractal_Dimension</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Mean_Radius</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.286064</td>\n",
              "      <td>0.997722</td>\n",
              "      <td>0.983170</td>\n",
              "      <td>0.314733</td>\n",
              "      <td>0.604780</td>\n",
              "      <td>0.792942</td>\n",
              "      <td>0.861399</td>\n",
              "      <td>0.252547</td>\n",
              "      <td>-0.210241</td>\n",
              "      <td>0.724747</td>\n",
              "      <td>-0.065776</td>\n",
              "      <td>0.724744</td>\n",
              "      <td>0.735436</td>\n",
              "      <td>-0.088994</td>\n",
              "      <td>0.295285</td>\n",
              "      <td>0.421286</td>\n",
              "      <td>0.422812</td>\n",
              "      <td>0.010380</td>\n",
              "      <td>0.072217</td>\n",
              "      <td>0.968388</td>\n",
              "      <td>0.238931</td>\n",
              "      <td>0.965590</td>\n",
              "      <td>0.943394</td>\n",
              "      <td>0.178633</td>\n",
              "      <td>0.461976</td>\n",
              "      <td>0.629344</td>\n",
              "      <td>0.760385</td>\n",
              "      <td>0.193055</td>\n",
              "      <td>0.071255</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Texture</th>\n",
              "      <td>0.286064</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.287677</td>\n",
              "      <td>0.260184</td>\n",
              "      <td>0.027621</td>\n",
              "      <td>0.209498</td>\n",
              "      <td>0.264090</td>\n",
              "      <td>0.259124</td>\n",
              "      <td>0.031737</td>\n",
              "      <td>-0.067123</td>\n",
              "      <td>0.186435</td>\n",
              "      <td>0.349223</td>\n",
              "      <td>0.184364</td>\n",
              "      <td>0.170072</td>\n",
              "      <td>-0.022268</td>\n",
              "      <td>0.154959</td>\n",
              "      <td>0.165669</td>\n",
              "      <td>0.149186</td>\n",
              "      <td>-0.053894</td>\n",
              "      <td>0.055306</td>\n",
              "      <td>0.306834</td>\n",
              "      <td>0.911348</td>\n",
              "      <td>0.303001</td>\n",
              "      <td>0.277462</td>\n",
              "      <td>0.074516</td>\n",
              "      <td>0.225781</td>\n",
              "      <td>0.264051</td>\n",
              "      <td>0.263770</td>\n",
              "      <td>0.033771</td>\n",
              "      <td>0.085589</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Perimeter</th>\n",
              "      <td>0.997722</td>\n",
              "      <td>0.287677</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.983148</td>\n",
              "      <td>0.354120</td>\n",
              "      <td>0.651361</td>\n",
              "      <td>0.827772</td>\n",
              "      <td>0.886632</td>\n",
              "      <td>0.288364</td>\n",
              "      <td>-0.156298</td>\n",
              "      <td>0.735652</td>\n",
              "      <td>-0.051667</td>\n",
              "      <td>0.741078</td>\n",
              "      <td>0.743482</td>\n",
              "      <td>-0.059729</td>\n",
              "      <td>0.336407</td>\n",
              "      <td>0.461183</td>\n",
              "      <td>0.455990</td>\n",
              "      <td>0.035819</td>\n",
              "      <td>0.104886</td>\n",
              "      <td>0.967440</td>\n",
              "      <td>0.241168</td>\n",
              "      <td>0.970385</td>\n",
              "      <td>0.944960</td>\n",
              "      <td>0.212170</td>\n",
              "      <td>0.497986</td>\n",
              "      <td>0.660736</td>\n",
              "      <td>0.782950</td>\n",
              "      <td>0.215712</td>\n",
              "      <td>0.108458</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Area</th>\n",
              "      <td>0.983170</td>\n",
              "      <td>0.260184</td>\n",
              "      <td>0.983148</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.329931</td>\n",
              "      <td>0.587537</td>\n",
              "      <td>0.796583</td>\n",
              "      <td>0.854603</td>\n",
              "      <td>0.254357</td>\n",
              "      <td>-0.180501</td>\n",
              "      <td>0.797738</td>\n",
              "      <td>-0.025194</td>\n",
              "      <td>0.798894</td>\n",
              "      <td>0.825634</td>\n",
              "      <td>-0.019163</td>\n",
              "      <td>0.296507</td>\n",
              "      <td>0.429408</td>\n",
              "      <td>0.413432</td>\n",
              "      <td>0.054688</td>\n",
              "      <td>0.088022</td>\n",
              "      <td>0.949036</td>\n",
              "      <td>0.202027</td>\n",
              "      <td>0.948717</td>\n",
              "      <td>0.958253</td>\n",
              "      <td>0.171047</td>\n",
              "      <td>0.422787</td>\n",
              "      <td>0.596895</td>\n",
              "      <td>0.719307</td>\n",
              "      <td>0.154969</td>\n",
              "      <td>0.048862</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Smoothness</th>\n",
              "      <td>0.314733</td>\n",
              "      <td>0.027621</td>\n",
              "      <td>0.354120</td>\n",
              "      <td>0.329931</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.731966</td>\n",
              "      <td>0.630427</td>\n",
              "      <td>0.644475</td>\n",
              "      <td>0.537370</td>\n",
              "      <td>0.635765</td>\n",
              "      <td>0.384375</td>\n",
              "      <td>0.074770</td>\n",
              "      <td>0.400663</td>\n",
              "      <td>0.329788</td>\n",
              "      <td>0.363593</td>\n",
              "      <td>0.383196</td>\n",
              "      <td>0.421501</td>\n",
              "      <td>0.446709</td>\n",
              "      <td>0.276982</td>\n",
              "      <td>0.266518</td>\n",
              "      <td>0.357846</td>\n",
              "      <td>0.109970</td>\n",
              "      <td>0.393514</td>\n",
              "      <td>0.360520</td>\n",
              "      <td>0.822491</td>\n",
              "      <td>0.559649</td>\n",
              "      <td>0.549722</td>\n",
              "      <td>0.606880</td>\n",
              "      <td>0.449005</td>\n",
              "      <td>0.513581</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Compactness</th>\n",
              "      <td>0.604780</td>\n",
              "      <td>0.209498</td>\n",
              "      <td>0.651361</td>\n",
              "      <td>0.587537</td>\n",
              "      <td>0.731966</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.915044</td>\n",
              "      <td>0.863165</td>\n",
              "      <td>0.630073</td>\n",
              "      <td>0.557897</td>\n",
              "      <td>0.529315</td>\n",
              "      <td>0.073870</td>\n",
              "      <td>0.569558</td>\n",
              "      <td>0.467545</td>\n",
              "      <td>0.239853</td>\n",
              "      <td>0.734621</td>\n",
              "      <td>0.762592</td>\n",
              "      <td>0.703079</td>\n",
              "      <td>0.295808</td>\n",
              "      <td>0.464574</td>\n",
              "      <td>0.624370</td>\n",
              "      <td>0.221771</td>\n",
              "      <td>0.677805</td>\n",
              "      <td>0.600330</td>\n",
              "      <td>0.596906</td>\n",
              "      <td>0.852408</td>\n",
              "      <td>0.847897</td>\n",
              "      <td>0.836307</td>\n",
              "      <td>0.510383</td>\n",
              "      <td>0.612921</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Concavity</th>\n",
              "      <td>0.792942</td>\n",
              "      <td>0.264090</td>\n",
              "      <td>0.827772</td>\n",
              "      <td>0.796583</td>\n",
              "      <td>0.630427</td>\n",
              "      <td>0.915044</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.955500</td>\n",
              "      <td>0.560739</td>\n",
              "      <td>0.324621</td>\n",
              "      <td>0.700284</td>\n",
              "      <td>0.089016</td>\n",
              "      <td>0.729092</td>\n",
              "      <td>0.662279</td>\n",
              "      <td>0.217359</td>\n",
              "      <td>0.642814</td>\n",
              "      <td>0.764821</td>\n",
              "      <td>0.678860</td>\n",
              "      <td>0.236803</td>\n",
              "      <td>0.372778</td>\n",
              "      <td>0.800181</td>\n",
              "      <td>0.252841</td>\n",
              "      <td>0.839927</td>\n",
              "      <td>0.795222</td>\n",
              "      <td>0.498944</td>\n",
              "      <td>0.753009</td>\n",
              "      <td>0.874190</td>\n",
              "      <td>0.877538</td>\n",
              "      <td>0.414505</td>\n",
              "      <td>0.456969</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Concave_Points</th>\n",
              "      <td>0.861399</td>\n",
              "      <td>0.259124</td>\n",
              "      <td>0.886632</td>\n",
              "      <td>0.854603</td>\n",
              "      <td>0.644475</td>\n",
              "      <td>0.863165</td>\n",
              "      <td>0.955500</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.518269</td>\n",
              "      <td>0.210933</td>\n",
              "      <td>0.710404</td>\n",
              "      <td>0.058041</td>\n",
              "      <td>0.726321</td>\n",
              "      <td>0.669774</td>\n",
              "      <td>0.145802</td>\n",
              "      <td>0.521126</td>\n",
              "      <td>0.621894</td>\n",
              "      <td>0.650093</td>\n",
              "      <td>0.171623</td>\n",
              "      <td>0.274089</td>\n",
              "      <td>0.868290</td>\n",
              "      <td>0.250457</td>\n",
              "      <td>0.893844</td>\n",
              "      <td>0.852331</td>\n",
              "      <td>0.481472</td>\n",
              "      <td>0.666621</td>\n",
              "      <td>0.786167</td>\n",
              "      <td>0.905794</td>\n",
              "      <td>0.380102</td>\n",
              "      <td>0.340750</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Symmetry</th>\n",
              "      <td>0.252547</td>\n",
              "      <td>0.031737</td>\n",
              "      <td>0.288364</td>\n",
              "      <td>0.254357</td>\n",
              "      <td>0.537370</td>\n",
              "      <td>0.630073</td>\n",
              "      <td>0.560739</td>\n",
              "      <td>0.518269</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.437242</td>\n",
              "      <td>0.300552</td>\n",
              "      <td>0.145115</td>\n",
              "      <td>0.313863</td>\n",
              "      <td>0.236742</td>\n",
              "      <td>0.238712</td>\n",
              "      <td>0.461442</td>\n",
              "      <td>0.485785</td>\n",
              "      <td>0.473291</td>\n",
              "      <td>0.487560</td>\n",
              "      <td>0.272823</td>\n",
              "      <td>0.278071</td>\n",
              "      <td>0.074453</td>\n",
              "      <td>0.319137</td>\n",
              "      <td>0.278669</td>\n",
              "      <td>0.394146</td>\n",
              "      <td>0.514701</td>\n",
              "      <td>0.504720</td>\n",
              "      <td>0.486453</td>\n",
              "      <td>0.735222</td>\n",
              "      <td>0.378432</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mean_Fractal_Dimension</th>\n",
              "      <td>-0.210241</td>\n",
              "      <td>-0.067123</td>\n",
              "      <td>-0.156298</td>\n",
              "      <td>-0.180501</td>\n",
              "      <td>0.635765</td>\n",
              "      <td>0.557897</td>\n",
              "      <td>0.324621</td>\n",
              "      <td>0.210933</td>\n",
              "      <td>0.437242</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.032344</td>\n",
              "      <td>0.234290</td>\n",
              "      <td>0.077172</td>\n",
              "      <td>-0.046841</td>\n",
              "      <td>0.458789</td>\n",
              "      <td>0.556675</td>\n",
              "      <td>0.468383</td>\n",
              "      <td>0.374131</td>\n",
              "      <td>0.423150</td>\n",
              "      <td>0.596712</td>\n",
              "      <td>-0.161603</td>\n",
              "      <td>-0.002553</td>\n",
              "      <td>-0.103627</td>\n",
              "      <td>-0.137245</td>\n",
              "      <td>0.566009</td>\n",
              "      <td>0.491327</td>\n",
              "      <td>0.350627</td>\n",
              "      <td>0.217596</td>\n",
              "      <td>0.354516</td>\n",
              "      <td>0.716770</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Radius_SE</th>\n",
              "      <td>0.724747</td>\n",
              "      <td>0.186435</td>\n",
              "      <td>0.735652</td>\n",
              "      <td>0.797738</td>\n",
              "      <td>0.384375</td>\n",
              "      <td>0.529315</td>\n",
              "      <td>0.700284</td>\n",
              "      <td>0.710404</td>\n",
              "      <td>0.300552</td>\n",
              "      <td>0.032344</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.205374</td>\n",
              "      <td>0.980299</td>\n",
              "      <td>0.961058</td>\n",
              "      <td>0.262182</td>\n",
              "      <td>0.363292</td>\n",
              "      <td>0.497916</td>\n",
              "      <td>0.502456</td>\n",
              "      <td>0.317067</td>\n",
              "      <td>0.242253</td>\n",
              "      <td>0.714971</td>\n",
              "      <td>0.101379</td>\n",
              "      <td>0.717885</td>\n",
              "      <td>0.772510</td>\n",
              "      <td>0.131216</td>\n",
              "      <td>0.264773</td>\n",
              "      <td>0.416926</td>\n",
              "      <td>0.508245</td>\n",
              "      <td>0.047413</td>\n",
              "      <td>-0.005516</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Texture_SE</th>\n",
              "      <td>-0.065776</td>\n",
              "      <td>0.349223</td>\n",
              "      <td>-0.051667</td>\n",
              "      <td>-0.025194</td>\n",
              "      <td>0.074770</td>\n",
              "      <td>0.073870</td>\n",
              "      <td>0.089016</td>\n",
              "      <td>0.058041</td>\n",
              "      <td>0.145115</td>\n",
              "      <td>0.234290</td>\n",
              "      <td>0.205374</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.195008</td>\n",
              "      <td>0.126952</td>\n",
              "      <td>0.443966</td>\n",
              "      <td>0.259695</td>\n",
              "      <td>0.196492</td>\n",
              "      <td>0.180903</td>\n",
              "      <td>0.476602</td>\n",
              "      <td>0.322444</td>\n",
              "      <td>-0.103700</td>\n",
              "      <td>0.358911</td>\n",
              "      <td>-0.093833</td>\n",
              "      <td>-0.060078</td>\n",
              "      <td>-0.116872</td>\n",
              "      <td>-0.107383</td>\n",
              "      <td>-0.094828</td>\n",
              "      <td>-0.123700</td>\n",
              "      <td>-0.143244</td>\n",
              "      <td>-0.096030</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Perimeter_SE</th>\n",
              "      <td>0.724744</td>\n",
              "      <td>0.184364</td>\n",
              "      <td>0.741078</td>\n",
              "      <td>0.798894</td>\n",
              "      <td>0.400663</td>\n",
              "      <td>0.569558</td>\n",
              "      <td>0.729092</td>\n",
              "      <td>0.726321</td>\n",
              "      <td>0.313863</td>\n",
              "      <td>0.077172</td>\n",
              "      <td>0.980299</td>\n",
              "      <td>0.195008</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.956455</td>\n",
              "      <td>0.258267</td>\n",
              "      <td>0.398510</td>\n",
              "      <td>0.532153</td>\n",
              "      <td>0.523900</td>\n",
              "      <td>0.327818</td>\n",
              "      <td>0.245097</td>\n",
              "      <td>0.703534</td>\n",
              "      <td>0.100183</td>\n",
              "      <td>0.722973</td>\n",
              "      <td>0.761519</td>\n",
              "      <td>0.141189</td>\n",
              "      <td>0.310953</td>\n",
              "      <td>0.451451</td>\n",
              "      <td>0.528121</td>\n",
              "      <td>0.066517</td>\n",
              "      <td>0.031595</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Area_SE</th>\n",
              "      <td>0.735436</td>\n",
              "      <td>0.170072</td>\n",
              "      <td>0.743482</td>\n",
              "      <td>0.825634</td>\n",
              "      <td>0.329788</td>\n",
              "      <td>0.467545</td>\n",
              "      <td>0.662279</td>\n",
              "      <td>0.669774</td>\n",
              "      <td>0.236742</td>\n",
              "      <td>-0.046841</td>\n",
              "      <td>0.961058</td>\n",
              "      <td>0.126952</td>\n",
              "      <td>0.956455</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.197422</td>\n",
              "      <td>0.302764</td>\n",
              "      <td>0.430723</td>\n",
              "      <td>0.402597</td>\n",
              "      <td>0.221619</td>\n",
              "      <td>0.164357</td>\n",
              "      <td>0.709029</td>\n",
              "      <td>0.086602</td>\n",
              "      <td>0.713564</td>\n",
              "      <td>0.791154</td>\n",
              "      <td>0.108858</td>\n",
              "      <td>0.248040</td>\n",
              "      <td>0.395474</td>\n",
              "      <td>0.478103</td>\n",
              "      <td>0.016636</td>\n",
              "      <td>-0.030354</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Smoothness_SE</th>\n",
              "      <td>-0.088994</td>\n",
              "      <td>-0.022268</td>\n",
              "      <td>-0.059729</td>\n",
              "      <td>-0.019163</td>\n",
              "      <td>0.363593</td>\n",
              "      <td>0.239853</td>\n",
              "      <td>0.217359</td>\n",
              "      <td>0.145802</td>\n",
              "      <td>0.238712</td>\n",
              "      <td>0.458789</td>\n",
              "      <td>0.262182</td>\n",
              "      <td>0.443966</td>\n",
              "      <td>0.258267</td>\n",
              "      <td>0.197422</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.403696</td>\n",
              "      <td>0.401212</td>\n",
              "      <td>0.406295</td>\n",
              "      <td>0.531234</td>\n",
              "      <td>0.425641</td>\n",
              "      <td>-0.134658</td>\n",
              "      <td>-0.089008</td>\n",
              "      <td>-0.109858</td>\n",
              "      <td>-0.072231</td>\n",
              "      <td>0.251538</td>\n",
              "      <td>-0.002280</td>\n",
              "      <td>0.007165</td>\n",
              "      <td>-0.029751</td>\n",
              "      <td>-0.066825</td>\n",
              "      <td>0.067030</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Compactness_SE</th>\n",
              "      <td>0.295285</td>\n",
              "      <td>0.154959</td>\n",
              "      <td>0.336407</td>\n",
              "      <td>0.296507</td>\n",
              "      <td>0.383196</td>\n",
              "      <td>0.734621</td>\n",
              "      <td>0.642814</td>\n",
              "      <td>0.521126</td>\n",
              "      <td>0.461442</td>\n",
              "      <td>0.556675</td>\n",
              "      <td>0.363292</td>\n",
              "      <td>0.259695</td>\n",
              "      <td>0.398510</td>\n",
              "      <td>0.302764</td>\n",
              "      <td>0.403696</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.902586</td>\n",
              "      <td>0.756550</td>\n",
              "      <td>0.443026</td>\n",
              "      <td>0.805155</td>\n",
              "      <td>0.271224</td>\n",
              "      <td>0.117162</td>\n",
              "      <td>0.325073</td>\n",
              "      <td>0.274352</td>\n",
              "      <td>0.243549</td>\n",
              "      <td>0.665705</td>\n",
              "      <td>0.602148</td>\n",
              "      <td>0.481079</td>\n",
              "      <td>0.283498</td>\n",
              "      <td>0.547199</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Concavity_SE</th>\n",
              "      <td>0.421286</td>\n",
              "      <td>0.165669</td>\n",
              "      <td>0.461183</td>\n",
              "      <td>0.429408</td>\n",
              "      <td>0.421501</td>\n",
              "      <td>0.762592</td>\n",
              "      <td>0.764821</td>\n",
              "      <td>0.621894</td>\n",
              "      <td>0.485785</td>\n",
              "      <td>0.468383</td>\n",
              "      <td>0.497916</td>\n",
              "      <td>0.196492</td>\n",
              "      <td>0.532153</td>\n",
              "      <td>0.430723</td>\n",
              "      <td>0.401212</td>\n",
              "      <td>0.902586</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.825624</td>\n",
              "      <td>0.401472</td>\n",
              "      <td>0.685475</td>\n",
              "      <td>0.404545</td>\n",
              "      <td>0.135136</td>\n",
              "      <td>0.453721</td>\n",
              "      <td>0.406328</td>\n",
              "      <td>0.295776</td>\n",
              "      <td>0.647759</td>\n",
              "      <td>0.716441</td>\n",
              "      <td>0.591405</td>\n",
              "      <td>0.296858</td>\n",
              "      <td>0.486902</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Concave_Points_SE</th>\n",
              "      <td>0.422812</td>\n",
              "      <td>0.149186</td>\n",
              "      <td>0.455990</td>\n",
              "      <td>0.413432</td>\n",
              "      <td>0.446709</td>\n",
              "      <td>0.703079</td>\n",
              "      <td>0.678860</td>\n",
              "      <td>0.650093</td>\n",
              "      <td>0.473291</td>\n",
              "      <td>0.374131</td>\n",
              "      <td>0.502456</td>\n",
              "      <td>0.180903</td>\n",
              "      <td>0.523900</td>\n",
              "      <td>0.402597</td>\n",
              "      <td>0.406295</td>\n",
              "      <td>0.756550</td>\n",
              "      <td>0.825624</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.338340</td>\n",
              "      <td>0.571545</td>\n",
              "      <td>0.416136</td>\n",
              "      <td>0.098028</td>\n",
              "      <td>0.451010</td>\n",
              "      <td>0.407333</td>\n",
              "      <td>0.270142</td>\n",
              "      <td>0.498981</td>\n",
              "      <td>0.560381</td>\n",
              "      <td>0.629839</td>\n",
              "      <td>0.204355</td>\n",
              "      <td>0.303881</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Symmetry_SE</th>\n",
              "      <td>0.010380</td>\n",
              "      <td>-0.053894</td>\n",
              "      <td>0.035819</td>\n",
              "      <td>0.054688</td>\n",
              "      <td>0.276982</td>\n",
              "      <td>0.295808</td>\n",
              "      <td>0.236803</td>\n",
              "      <td>0.171623</td>\n",
              "      <td>0.487560</td>\n",
              "      <td>0.423150</td>\n",
              "      <td>0.317067</td>\n",
              "      <td>0.476602</td>\n",
              "      <td>0.327818</td>\n",
              "      <td>0.221619</td>\n",
              "      <td>0.531234</td>\n",
              "      <td>0.443026</td>\n",
              "      <td>0.401472</td>\n",
              "      <td>0.338340</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.465355</td>\n",
              "      <td>-0.050356</td>\n",
              "      <td>-0.122405</td>\n",
              "      <td>-0.026391</td>\n",
              "      <td>-0.017646</td>\n",
              "      <td>-0.010428</td>\n",
              "      <td>0.064563</td>\n",
              "      <td>0.038381</td>\n",
              "      <td>-0.015472</td>\n",
              "      <td>0.362878</td>\n",
              "      <td>0.044517</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Fractal_Dimension_SE</th>\n",
              "      <td>0.072217</td>\n",
              "      <td>0.055306</td>\n",
              "      <td>0.104886</td>\n",
              "      <td>0.088022</td>\n",
              "      <td>0.266518</td>\n",
              "      <td>0.464574</td>\n",
              "      <td>0.372778</td>\n",
              "      <td>0.274089</td>\n",
              "      <td>0.272823</td>\n",
              "      <td>0.596712</td>\n",
              "      <td>0.242253</td>\n",
              "      <td>0.322444</td>\n",
              "      <td>0.245097</td>\n",
              "      <td>0.164357</td>\n",
              "      <td>0.425641</td>\n",
              "      <td>0.805155</td>\n",
              "      <td>0.685475</td>\n",
              "      <td>0.571545</td>\n",
              "      <td>0.465355</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.044779</td>\n",
              "      <td>-0.000287</td>\n",
              "      <td>0.077090</td>\n",
              "      <td>0.059260</td>\n",
              "      <td>0.113541</td>\n",
              "      <td>0.359045</td>\n",
              "      <td>0.307279</td>\n",
              "      <td>0.197483</td>\n",
              "      <td>0.088062</td>\n",
              "      <td>0.494224</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Radius</th>\n",
              "      <td>0.968388</td>\n",
              "      <td>0.306834</td>\n",
              "      <td>0.967440</td>\n",
              "      <td>0.949036</td>\n",
              "      <td>0.357846</td>\n",
              "      <td>0.624370</td>\n",
              "      <td>0.800181</td>\n",
              "      <td>0.868290</td>\n",
              "      <td>0.278071</td>\n",
              "      <td>-0.161603</td>\n",
              "      <td>0.714971</td>\n",
              "      <td>-0.103700</td>\n",
              "      <td>0.703534</td>\n",
              "      <td>0.709029</td>\n",
              "      <td>-0.134658</td>\n",
              "      <td>0.271224</td>\n",
              "      <td>0.404545</td>\n",
              "      <td>0.416136</td>\n",
              "      <td>-0.050356</td>\n",
              "      <td>0.044779</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.295662</td>\n",
              "      <td>0.993335</td>\n",
              "      <td>0.978779</td>\n",
              "      <td>0.275557</td>\n",
              "      <td>0.517963</td>\n",
              "      <td>0.685014</td>\n",
              "      <td>0.811589</td>\n",
              "      <td>0.267599</td>\n",
              "      <td>0.153812</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Texture</th>\n",
              "      <td>0.238931</td>\n",
              "      <td>0.911348</td>\n",
              "      <td>0.241168</td>\n",
              "      <td>0.202027</td>\n",
              "      <td>0.109970</td>\n",
              "      <td>0.221771</td>\n",
              "      <td>0.252841</td>\n",
              "      <td>0.250457</td>\n",
              "      <td>0.074453</td>\n",
              "      <td>-0.002553</td>\n",
              "      <td>0.101379</td>\n",
              "      <td>0.358911</td>\n",
              "      <td>0.100183</td>\n",
              "      <td>0.086602</td>\n",
              "      <td>-0.089008</td>\n",
              "      <td>0.117162</td>\n",
              "      <td>0.135136</td>\n",
              "      <td>0.098028</td>\n",
              "      <td>-0.122405</td>\n",
              "      <td>-0.000287</td>\n",
              "      <td>0.295662</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.291550</td>\n",
              "      <td>0.256206</td>\n",
              "      <td>0.246929</td>\n",
              "      <td>0.308239</td>\n",
              "      <td>0.331327</td>\n",
              "      <td>0.325625</td>\n",
              "      <td>0.177027</td>\n",
              "      <td>0.198254</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Perimeter</th>\n",
              "      <td>0.965590</td>\n",
              "      <td>0.303001</td>\n",
              "      <td>0.970385</td>\n",
              "      <td>0.948717</td>\n",
              "      <td>0.393514</td>\n",
              "      <td>0.677805</td>\n",
              "      <td>0.839927</td>\n",
              "      <td>0.893844</td>\n",
              "      <td>0.319137</td>\n",
              "      <td>-0.103627</td>\n",
              "      <td>0.717885</td>\n",
              "      <td>-0.093833</td>\n",
              "      <td>0.722973</td>\n",
              "      <td>0.713564</td>\n",
              "      <td>-0.109858</td>\n",
              "      <td>0.325073</td>\n",
              "      <td>0.453721</td>\n",
              "      <td>0.451010</td>\n",
              "      <td>-0.026391</td>\n",
              "      <td>0.077090</td>\n",
              "      <td>0.993335</td>\n",
              "      <td>0.291550</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.975259</td>\n",
              "      <td>0.304021</td>\n",
              "      <td>0.568704</td>\n",
              "      <td>0.724715</td>\n",
              "      <td>0.835454</td>\n",
              "      <td>0.293633</td>\n",
              "      <td>0.198044</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Area</th>\n",
              "      <td>0.943394</td>\n",
              "      <td>0.277462</td>\n",
              "      <td>0.944960</td>\n",
              "      <td>0.958253</td>\n",
              "      <td>0.360520</td>\n",
              "      <td>0.600330</td>\n",
              "      <td>0.795222</td>\n",
              "      <td>0.852331</td>\n",
              "      <td>0.278669</td>\n",
              "      <td>-0.137245</td>\n",
              "      <td>0.772510</td>\n",
              "      <td>-0.060078</td>\n",
              "      <td>0.761519</td>\n",
              "      <td>0.791154</td>\n",
              "      <td>-0.072231</td>\n",
              "      <td>0.274352</td>\n",
              "      <td>0.406328</td>\n",
              "      <td>0.407333</td>\n",
              "      <td>-0.017646</td>\n",
              "      <td>0.059260</td>\n",
              "      <td>0.978779</td>\n",
              "      <td>0.256206</td>\n",
              "      <td>0.975259</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.258167</td>\n",
              "      <td>0.475176</td>\n",
              "      <td>0.646774</td>\n",
              "      <td>0.763491</td>\n",
              "      <td>0.226609</td>\n",
              "      <td>0.126989</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Smoothness</th>\n",
              "      <td>0.178633</td>\n",
              "      <td>0.074516</td>\n",
              "      <td>0.212170</td>\n",
              "      <td>0.171047</td>\n",
              "      <td>0.822491</td>\n",
              "      <td>0.596906</td>\n",
              "      <td>0.498944</td>\n",
              "      <td>0.481472</td>\n",
              "      <td>0.394146</td>\n",
              "      <td>0.566009</td>\n",
              "      <td>0.131216</td>\n",
              "      <td>-0.116872</td>\n",
              "      <td>0.141189</td>\n",
              "      <td>0.108858</td>\n",
              "      <td>0.251538</td>\n",
              "      <td>0.243549</td>\n",
              "      <td>0.295776</td>\n",
              "      <td>0.270142</td>\n",
              "      <td>-0.010428</td>\n",
              "      <td>0.113541</td>\n",
              "      <td>0.275557</td>\n",
              "      <td>0.246929</td>\n",
              "      <td>0.304021</td>\n",
              "      <td>0.258167</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.631990</td>\n",
              "      <td>0.611867</td>\n",
              "      <td>0.607189</td>\n",
              "      <td>0.521811</td>\n",
              "      <td>0.663891</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Compactness</th>\n",
              "      <td>0.461976</td>\n",
              "      <td>0.225781</td>\n",
              "      <td>0.497986</td>\n",
              "      <td>0.422787</td>\n",
              "      <td>0.559649</td>\n",
              "      <td>0.852408</td>\n",
              "      <td>0.753009</td>\n",
              "      <td>0.666621</td>\n",
              "      <td>0.514701</td>\n",
              "      <td>0.491327</td>\n",
              "      <td>0.264773</td>\n",
              "      <td>-0.107383</td>\n",
              "      <td>0.310953</td>\n",
              "      <td>0.248040</td>\n",
              "      <td>-0.002280</td>\n",
              "      <td>0.665705</td>\n",
              "      <td>0.647759</td>\n",
              "      <td>0.498981</td>\n",
              "      <td>0.064563</td>\n",
              "      <td>0.359045</td>\n",
              "      <td>0.517963</td>\n",
              "      <td>0.308239</td>\n",
              "      <td>0.568704</td>\n",
              "      <td>0.475176</td>\n",
              "      <td>0.631990</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.918145</td>\n",
              "      <td>0.809077</td>\n",
              "      <td>0.600954</td>\n",
              "      <td>0.816049</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Concavity</th>\n",
              "      <td>0.629344</td>\n",
              "      <td>0.264051</td>\n",
              "      <td>0.660736</td>\n",
              "      <td>0.596895</td>\n",
              "      <td>0.549722</td>\n",
              "      <td>0.847897</td>\n",
              "      <td>0.874190</td>\n",
              "      <td>0.786167</td>\n",
              "      <td>0.504720</td>\n",
              "      <td>0.350627</td>\n",
              "      <td>0.416926</td>\n",
              "      <td>-0.094828</td>\n",
              "      <td>0.451451</td>\n",
              "      <td>0.395474</td>\n",
              "      <td>0.007165</td>\n",
              "      <td>0.602148</td>\n",
              "      <td>0.716441</td>\n",
              "      <td>0.560381</td>\n",
              "      <td>0.038381</td>\n",
              "      <td>0.307279</td>\n",
              "      <td>0.685014</td>\n",
              "      <td>0.331327</td>\n",
              "      <td>0.724715</td>\n",
              "      <td>0.646774</td>\n",
              "      <td>0.611867</td>\n",
              "      <td>0.918145</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.894806</td>\n",
              "      <td>0.546537</td>\n",
              "      <td>0.702022</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Concave_Points</th>\n",
              "      <td>0.760385</td>\n",
              "      <td>0.263770</td>\n",
              "      <td>0.782950</td>\n",
              "      <td>0.719307</td>\n",
              "      <td>0.606880</td>\n",
              "      <td>0.836307</td>\n",
              "      <td>0.877538</td>\n",
              "      <td>0.905794</td>\n",
              "      <td>0.486453</td>\n",
              "      <td>0.217596</td>\n",
              "      <td>0.508245</td>\n",
              "      <td>-0.123700</td>\n",
              "      <td>0.528121</td>\n",
              "      <td>0.478103</td>\n",
              "      <td>-0.029751</td>\n",
              "      <td>0.481079</td>\n",
              "      <td>0.591405</td>\n",
              "      <td>0.629839</td>\n",
              "      <td>-0.015472</td>\n",
              "      <td>0.197483</td>\n",
              "      <td>0.811589</td>\n",
              "      <td>0.325625</td>\n",
              "      <td>0.835454</td>\n",
              "      <td>0.763491</td>\n",
              "      <td>0.607189</td>\n",
              "      <td>0.809077</td>\n",
              "      <td>0.894806</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.505580</td>\n",
              "      <td>0.516934</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Symmetry</th>\n",
              "      <td>0.193055</td>\n",
              "      <td>0.033771</td>\n",
              "      <td>0.215712</td>\n",
              "      <td>0.154969</td>\n",
              "      <td>0.449005</td>\n",
              "      <td>0.510383</td>\n",
              "      <td>0.414505</td>\n",
              "      <td>0.380102</td>\n",
              "      <td>0.735222</td>\n",
              "      <td>0.354516</td>\n",
              "      <td>0.047413</td>\n",
              "      <td>-0.143244</td>\n",
              "      <td>0.066517</td>\n",
              "      <td>0.016636</td>\n",
              "      <td>-0.066825</td>\n",
              "      <td>0.283498</td>\n",
              "      <td>0.296858</td>\n",
              "      <td>0.204355</td>\n",
              "      <td>0.362878</td>\n",
              "      <td>0.088062</td>\n",
              "      <td>0.267599</td>\n",
              "      <td>0.177027</td>\n",
              "      <td>0.293633</td>\n",
              "      <td>0.226609</td>\n",
              "      <td>0.521811</td>\n",
              "      <td>0.600954</td>\n",
              "      <td>0.546537</td>\n",
              "      <td>0.505580</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.521593</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Worst_Fractal_Dimension</th>\n",
              "      <td>0.071255</td>\n",
              "      <td>0.085589</td>\n",
              "      <td>0.108458</td>\n",
              "      <td>0.048862</td>\n",
              "      <td>0.513581</td>\n",
              "      <td>0.612921</td>\n",
              "      <td>0.456969</td>\n",
              "      <td>0.340750</td>\n",
              "      <td>0.378432</td>\n",
              "      <td>0.716770</td>\n",
              "      <td>-0.005516</td>\n",
              "      <td>-0.096030</td>\n",
              "      <td>0.031595</td>\n",
              "      <td>-0.030354</td>\n",
              "      <td>0.067030</td>\n",
              "      <td>0.547199</td>\n",
              "      <td>0.486902</td>\n",
              "      <td>0.303881</td>\n",
              "      <td>0.044517</td>\n",
              "      <td>0.494224</td>\n",
              "      <td>0.153812</td>\n",
              "      <td>0.198254</td>\n",
              "      <td>0.198044</td>\n",
              "      <td>0.126989</td>\n",
              "      <td>0.663891</td>\n",
              "      <td>0.816049</td>\n",
              "      <td>0.702022</td>\n",
              "      <td>0.516934</td>\n",
              "      <td>0.521593</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                         Mean_Radius  ...  Worst_Fractal_Dimension\n",
              "Mean_Radius                 1.000000  ...                 0.071255\n",
              "Mean_Texture                0.286064  ...                 0.085589\n",
              "Mean_Perimeter              0.997722  ...                 0.108458\n",
              "Mean_Area                   0.983170  ...                 0.048862\n",
              "Mean_Smoothness             0.314733  ...                 0.513581\n",
              "Mean_Compactness            0.604780  ...                 0.612921\n",
              "Mean_Concavity              0.792942  ...                 0.456969\n",
              "Mean_Concave_Points         0.861399  ...                 0.340750\n",
              "Mean_Symmetry               0.252547  ...                 0.378432\n",
              "Mean_Fractal_Dimension     -0.210241  ...                 0.716770\n",
              "Radius_SE                   0.724747  ...                -0.005516\n",
              "Texture_SE                 -0.065776  ...                -0.096030\n",
              "Perimeter_SE                0.724744  ...                 0.031595\n",
              "Area_SE                     0.735436  ...                -0.030354\n",
              "Smoothness_SE              -0.088994  ...                 0.067030\n",
              "Compactness_SE              0.295285  ...                 0.547199\n",
              "Concavity_SE                0.421286  ...                 0.486902\n",
              "Concave_Points_SE           0.422812  ...                 0.303881\n",
              "Symmetry_SE                 0.010380  ...                 0.044517\n",
              "Fractal_Dimension_SE        0.072217  ...                 0.494224\n",
              "Worst_Radius                0.968388  ...                 0.153812\n",
              "Worst_Texture               0.238931  ...                 0.198254\n",
              "Worst_Perimeter             0.965590  ...                 0.198044\n",
              "Worst_Area                  0.943394  ...                 0.126989\n",
              "Worst_Smoothness            0.178633  ...                 0.663891\n",
              "Worst_Compactness           0.461976  ...                 0.816049\n",
              "Worst_Concavity             0.629344  ...                 0.702022\n",
              "Worst_Concave_Points        0.760385  ...                 0.516934\n",
              "Worst_Symmetry              0.193055  ...                 0.521593\n",
              "Worst_Fractal_Dimension     0.071255  ...                 1.000000\n",
              "\n",
              "[30 rows x 30 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 1496
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KVfD1mErlnKw"
      },
      "source": [
        "## Heatmap of correlation in our dataset\n",
        "As you can see from the legend the lighter the square the more highly correlated the feature data and the darker the square the less correlated."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JUoi-ZQ8ll4Y",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 354
        },
        "outputId": "02bbe56b-d1ce-4589-87ad-5299306dfcf6"
      },
      "source": [
        "# it computes the Pearson’s correlation coefficient.\n",
        "corr_matrix = X.corr()\n",
        "#The parameter ‘ annot=True ‘ displays the values of the correlation coefficient\n",
        "#in each cell\n",
        "sns.heatmap(corr_matrix, annot = False)\n",
        "plt.show()\n"
      ],
      "execution_count": 1497,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nJzjhwOkmmzq"
      },
      "source": [
        "# Training the linear regression model\n",
        "\n",
        "Here we are using linear regression to compare features in our model to see the correlation between them.  We are training this model to show us how dependent the variables are on one another."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "x4MVGZmc5iY7"
      },
      "source": [
        "### Reference\n",
        "SKLearn documentation on Linear Regression: https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BC07qA925yK9"
      },
      "source": [
        "# Creating linear regression model for Mean_Radius and Mean_Fractal_Dimension\n",
        "Here I have trained a model to show us the negative correlation between two features in our dataset, we can clearly see from this model that the Mean_Radius has a negative impact on the Mean_Fractal_Dimension.  When the Mean_Fractal_Dimension value increases then on average the Mean_Radius value decreases and vice versa."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 382
        },
        "id": "3Kz8R71_g0zy",
        "outputId": "ef7adfb0-31e0-49b7-a843-5871b79ac388"
      },
      "source": [
        "# Here we train a model to perform linear regression data set to predict the diagnosis\n",
        "from sklearn.linear_model import LinearRegression\n",
        "from sklearn.metrics import mean_squared_error, r2_score\n",
        "from math import sqrt\n",
        "from scipy import stats\n",
        "# create the linear regression model by fitting training set to the test set\n",
        "fractalDimensionArrayTest = np.array(X_test['Mean_Fractal_Dimension']).reshape(-1,1)\n",
        "fractalDimensionArrayTrain = np.array(X_train['Mean_Fractal_Dimension']).reshape(-1,1)\n",
        "meanRadiusArrayTrain = np.array(X_train['Mean_Radius']).reshape(-1,1)\n",
        "regression_model = LinearRegression().fit(meanRadiusArrayTrain, fractalDimensionArrayTrain)\n",
        "# Score the model and retrive r^2 value\n",
        "y_pred = regression_model.predict(fractalDimensionArrayTest)\n",
        "rSquared = regression_model.score(fractalDimensionArrayTest, y_pred)\n",
        "print('y test r^2 value: ', rSquared)\n",
        "print('mean_squared_error value: ', mean_squared_error(fractalDimensionArrayTest, y_pred))\n",
        "print('Coefficient: ',regression_model.coef_)\n",
        "print('Intercept: ',regression_model.intercept_)\n",
        "\n",
        "plt.xlabel(\"Mean_Radius\")\n",
        "plt.ylabel(\"Mean_Fractal_Dimension\")\n",
        "plt.title('Mean_Fractal_Dimension and Mean_Radius')\n",
        "plt.scatter(fractalDimensionArrayTest,y_pred)\n",
        "plt.plot(fractalDimensionArrayTest,y_pred,c='r')\n",
        "print('predict ',regression_model.predict([[153.4]]))\n"
      ],
      "execution_count": 1498,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "y test r^2 value:  1.0\n",
            "mean_squared_error value:  0.00014555891227921536\n",
            "Coefficient:  [[-0.00074819]]\n",
            "Intercept:  [0.07334571]\n",
            "predict  [[-0.04142696]]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qFls_Z_FdLOf"
      },
      "source": [
        "I will demonstrate with some predictions below that the values are negatively skewed.  We will see that whenever we pass in a high fractal dimension the value will be lower and vice versa.  We can also see there is a low mean square error and a negative coefficient.  Since our coefficient is negative we will see that negative values will produce higher values for our dependent variable where as higher values will produce low values for our dependent variable"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "v-YeLvDgWm1Z",
        "outputId": "43977103-c4df-4413-dd02-b00b67231d4e"
      },
      "source": [
        "# As we can see not too much of a difference between the values here\n",
        "print('this value will be low: ', regression_model.predict([[0.5]]))\n",
        "print('this value will be high: ', regression_model.predict([[0.1]]))\n",
        "\n",
        "# just to make our example clearer let's use ridiculously inflated values  \n",
        "print('let\\'s make our model predict some grossly inflated / deflated values to show that these features do indeed have a negative impact on one another')\n",
        "print('this value will be low: ', regression_model.predict([[100]]))\n",
        "# I use a negative value here to demonstrate the radius the higher the mean fractal dimension\n",
        "print('this value will be high: ', regression_model.predict([[-100]]))\n"
      ],
      "execution_count": 1499,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "this value will be low:  [[0.07297161]]\n",
            "this value will be high:  [[0.07327089]]\n",
            "let's make our model predict some grossly inflated / deflated values to show that these features do indeed have a negative impact on one another\n",
            "this value will be low:  [[-0.0014735]]\n",
            "this value will be high:  [[0.14816492]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6iBpsTa-er-E"
      },
      "source": [
        "As we can see above that whenever I make the value of the mean fractal dimension higher then the value of our dependent variable will decrease, whenever I make it lower it will increase.  We can also see that the values don't have too much of an effect on each other this is because the low value of the negative correlation"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "D7End3_e5JQP"
      },
      "source": [
        "# Positive correlation\n",
        "Here we compare the concave points severity with the concavity severity, we see that there is a positive linear correlation between the two variables, we also see that the R^2 value is one which indicates a possible overfitting of our model.  We can see clearly from this graph that the two variables have a high correlation and that if the value increases in one of the variables then we can predict that the other variable also increases"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 382
        },
        "id": "NUm5oqU4YyxT",
        "outputId": "92b03ef9-e91c-4f69-b8fa-3f059e389e0f"
      },
      "source": [
        "# create the linear regression model by fitting training set to the test set\n",
        "concavePointsSeverityArrayTest = np.array(X_test['Concave_Points_SE']).reshape(-1,1)\n",
        "concavePointsSeverityArrayTrain = np.array(X_train['Concave_Points_SE']).reshape(-1,1)\n",
        "concaveSeverityTrain = np.array(X_train['Concavity_SE']).reshape(-1,1)\n",
        "# here we need to reshape the data using numpy so that the training data matches the testdata\n",
        "regression_model = LinearRegression().fit(concaveSeverityTrain, concavePointsSeverityArrayTrain)\n",
        "# Score the model and retrive r^2 value\n",
        "y_pred = regression_model.predict(concavePointsSeverityArrayTest)\n",
        "rSquared = regression_model.score(concavePointsSeverityArrayTest, y_pred)\n",
        "print('y test r^2 value: ', rSquared)\n",
        "print('mean_squared_error value: ', mean_squared_error(concavePointsSeverityArrayTest, y_pred))\n",
        "print('Coefficient: ',regression_model.coef_)\n",
        "print('Intercept: ',regression_model.intercept_)\n",
        "plt.xlabel(\"Concavity_SE\")\n",
        "plt.ylabel(\"Concave_Points_SE\")\n",
        "plt.title('Linear regression using Convatiy Severity and Concave Points Severity')\n",
        "plt.scatter(concavePointsSeverityArrayTest,y_pred)\n",
        "plt.plot(concavePointsSeverityArrayTest,y_pred,c='r')"
      ],
      "execution_count": 1500,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "y test r^2 value:  1.0\n",
            "mean_squared_error value:  3.5719696254071577e-05\n",
            "Coefficient:  [[0.14594559]]\n",
            "Intercept:  [0.00720811]\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f5f7925cf50>]"
            ]
          },
          "metadata": {},
          "execution_count": 1500
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GjuuHSSoMpwD"
      },
      "source": [
        "We can see here there is a positive coefficient which is the amount we expect the value to increase by when making a prediction, we can also see that there is a low mean squared error value this indicates the performance of the model and the distance away the points are from our line"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XZ9SNcrxiE0I"
      },
      "source": [
        "Below we see the second models features have on one another, we can clearly see that the data has a positive correlation from our graph but let's show some concrete examples and perform some predictions using it."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ofv2lg_kajVZ",
        "outputId": "54611fa1-c685-438d-d533-3547c22edbd2"
      },
      "source": [
        "print('this value will be high: ', regression_model.predict([[0.5]]))\n",
        "print('this value will be low: ', regression_model.predict([[0.1]]))\n",
        "\n",
        "# just to show how high the impact will be with grossly inflated / deflated values values\n",
        "\n",
        "print('let\\'s make our model predict some grossly inflated / deflated values to show severity of impact')\n",
        "print('this value will be high: ', regression_model.predict([[100]]))\n",
        "print('this value will be low: ', regression_model.predict([[0.0000000001]]))\n"
      ],
      "execution_count": 1501,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "this value will be high:  [[0.08018091]]\n",
            "this value will be low:  [[0.02180267]]\n",
            "let's make our model predict some grossly inflated / deflated values to show severity of impact\n",
            "this value will be high:  [[14.6017676]]\n",
            "this value will be low:  [[0.00720811]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "znQ9wWrKfenz"
      },
      "source": [
        "Since the above values are highly correlated we see a greater rise and decrease in the predicted values when predicting than we did in the first linear regression model.  This is due to the fact that these two variables have a positive correlation of 0.776207, the features we used in our first model have a lower negative correlation which is why we don't see as much of an impact on them.  If they had a negative correlation of 0.776207 we would see the same rise and decrease in the first linear regression models predicted valules as well"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nGr0Ctt4eZ-b"
      },
      "source": [
        "# Examples of low and high correlation in our dataset"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-czUs2SDbPeP"
      },
      "source": [
        "Additionally we can see here the Mean Perimeter has a positive correlation with our Mean Radius feature.  Below this image we can see an example of low negative correlation, these variables don't really impact each other much"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "Zsy1970XoOZO",
        "outputId": "b97db231-d5f9-4ab3-b6f4-168ad2a9a3c1"
      },
      "source": [
        "# Example of positive Linear correlation between radius and perimeter\n",
        "# since these correlate highly we can remove one of them from set\n",
        "\n",
        "plt.scatter(X_train['Mean_Perimeter'],X_train['Mean_Radius'])\n"
      ],
      "execution_count": 1502,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.collections.PathCollection at 0x7f5f7922b890>"
            ]
          },
          "metadata": {},
          "execution_count": 1502
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "hhwzfmxSoTtU",
        "outputId": "9a111320-6915-449d-c317-2a6a061a5b34"
      },
      "source": [
        "# example of low negative correlation\n",
        "plt.scatter(X_train['Mean_Fractal_Dimension'],X_train['Mean_Radius'])\n"
      ],
      "execution_count": 1503,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.collections.PathCollection at 0x7f5f79190a90>"
            ]
          },
          "metadata": {},
          "execution_count": 1503
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO2dfZAc5Xngf8+uBtgVPlaKZQ6tkYV9jjhjBclsgIucHMI2ENlgBWyDgn325VJy7uKqiHLJli+uSBBXoZxi41QldQ6xSZwKwYoR7AlDLIhFzjEJxCt2ZViDzjaWgYEY8bHYSAua3X3uj5le9c722x/T3TM9vc+vaksz/fl2j/p5n34+RVUxDMMwyktPpwdgGIZh5IsJesMwjJJjgt4wDKPkmKA3DMMoOSboDcMwSs6iTg8giNe//vW6cuXKTg/DMAyjazhw4MDzqrosaF0hBf3KlSsZGRnp9DAMwzC6BhH5iWudmW4MwzBKjgl6wzCMkhNpuhGRM4G/Bk4HFLhZVf9ERHYDqxqbDQATqromYP/DwM+BaWBKVYcyGrthGIYRgzg2+ingk6r6sIi8DjggIvep6tXeBiLyeeDlkGOsV9XnU47VMAzDaIFIQa+qzwLPNj7/XEQeAwaB7wOIiAAfAi7OcZyGYRhGiySKuhGRlcBa4CHf4l8FfqqqP3DspsC9IqLAn6vqzY5jbwY2A6xYsSLJsErD8GiVXfsO8czEJMsH+th66So2rh3s9LAMw+hyYgt6ETkV2ANsUdWf+VZtAm4L2fWdqloVkTcA94nI46r67eaNGhPAzQBDQ0MLrqTm8GiVz9zxCJO1aQCqE5N85o5HAEzYG4aRilhRNyJSoS7kb1XVO3zLFwFXArtd+6pqtfHvc8CdwPlpBlxWdu07NCvkPSZr0+zad6hDIzIMoyxECvqGDf4rwGOq+oWm1e8GHlfVpx37Lm44cBGRxcAlwKPphlxOnpmYTLTcMAwjLnE0+nXAR4CLRWSs8behse4amsw2IrJcRO5pfD0d+I6IHAT+FbhbVb+Z0dhLxfKBvkTLDcMw4hIn6uY7gDjWfSxg2TPAhsbnJ4Bz0w1xYbD10lVzbPQAfZVetl66KmQvwzCMaApZ62Yh4jlcLerGMIysMUFfIDauHTTBbhhG5litG8MwjJJjgt4wDKPkmKA3DMMoOSboDcMwSo4JesMwjJJjgt4wDKPkmKA3DMMoOSboDcMwSo4JesMwjJJjgt4wDKPkmKA3DMMoOSboDcMwSo4VNTMKhfXNNYzsMUFvFAbrm2sY+WCmG6MwWN9cw8iHOD1jzxSR+0Xk+yIyLiK/11i+Q0SqAe0Fm/e/TEQOicgPRWRb1hdglAfrm2sY+RDHdDMFfFJVH240+j4gIvc11t2kqn/s2lFEeoE/A94DPA18V0T2qur30w7cKB/LB/qoBgh165trGOmI1OhV9VlVfbjx+efAY0Bcg+n5wA9V9QlVPQ58DXh/q4M1ys3WS1fRV+mds8z65hpGehLZ6EVkJbAWeKix6BMi8j0RuUVElgTsMgg85fv+NI5JQkQ2i8iIiIwcOXIkybCMEuBF20zWpumVei/6wYE+brxytTliDSMlsQW9iJwK7AG2qOrPgP8NvAVYAzwLfD7NQFT1ZlUdUtWhZcuWpTmU0WV40Tae2WZadVaTNyFvGOmJJehFpEJdyN+qqncAqOpPVXVaVWeAv6BupmmmCpzp+/7GxjLDmMWibQwjX+JE3QjwFeAxVf2Cb/kZvs1+A3g0YPfvAm8VkbNE5CTgGmBvuiEbZcOibQwjX+JE3awDPgI8IiJjjWX/E9gkImsABQ4DHwcQkeXAl1V1g6pOicgngH1AL3CLqo5nfA1Gl2PRNoaRL5GCXlW/A0jAqnsc2z8DbPB9v8e1rWFAPdrGnxELFm1jGFliJRCMjuM5XK3GjWHkgwl6oxBsXDtogt0wcsJq3RiGYZQcE/SGYRglxwS9YRhGyTFBbxiGUXLMGZsR1hnJMIyiYoI+A6wzkmEYRcZMNxngqtWyZfcY63buZ3jUyvsYhtE5TKPPgLCaLN2q3ZspyjDKg2n0GTDQXwld322VGP1lg5UTk5W9mRhGd2KCPiXDo1VeeXUqcrtuqsRoZYMNo1yYoE/Jrn2HqM1o5HbdVInRygYbRrkwQZ+SOMKv2yoxuialbpqsDMM4gQn6lLiEX68IQnf2PbUm3YZRLizqJiWuWurdJtz9WNlgwygXkYJeRM4E/ho4nXo3qZtV9U9EZBdwOXAc+BHwX1V1ImD/w8DPgWlgSlWHsht+5ymrULSywYZRHkQ13JHY6A17hqo+LCKvAw4AG6k3+t7faBf4RwCq+umA/Q8DQ6r6fNxBDQ0N6cjISPyrMAzDWOCIyAGXIh1po1fVZ1X14cbnnwOPAYOqeq+qenGFD1IX/IZhGEbBSOSMFZGVwFrgoaZVvwX8vWM3Be4VkQMisjnk2JtFZERERo4cOZJkWIZhGEYIsQW9iJwK7AG2qOrPfMt/H5gCbnXs+k5VfQfw68DvisivBW2kqjer6pCqDi1btiz2BRiGYRjhxBL0IlKhLuRvVdU7fMs/BrwPuFYdxn5VrTb+fQ64Ezg/5ZgNwzCMBEQKehER4CvAY6r6Bd/yy4BPAVeo6jHHvosbDlxEZDFwCfBoFgM3DMMw4hFHo18HfAS4WETGGn8bgD8FXgfc11j2JQARWS4i9zT2PR34jogcBP4VuFtVv5n9ZRiGYRguIuPoVfU7gASsuidgGar6DLCh8fkJ4Nw0A2wVK7Mbjd0jw1gYlDIz1jo+RdPOe2QTimF0llLWuilimd3h0Srrdu7nrG13F6LrVLvukdW2N4zOU0pBX7Qyu0UUdtU23aMiTrqGsdAopaAvWpndogm74dFqoNMFsr9HRZt0DWMhUkpBX7Qyu0UTdrv2HSIo6UEg83tUtEnXMBYipXTGpqkomYfjcPlAX6CppFPCzjXBKNk7Yl1lnLu1tr05lo1upJSCHlors5tXJErRhJ1r4hnMYeIpUxlni+YyupXSCvpWCLOlp3mQsxR2WWiU7Z54sqpt32ltOq//H4aRNybofeRpS89C2GWlUXajll0EbbpovhbDiIsJeh/tsqW3qplmqVF2WwepImjTRfO1GEZcSifo07zet2rSSHLONJppHhplp80hcSmCNl00X4thxKVUgj6uEHUJt1ZMGlHnbD7XseNTLWumWWuUQWPfsnuM6+8aZ/vl5xRK4BdBm+5Gk5dhQIyesZ2g1Z6x63bud0aTPLDtYmC+cIO6VnbjlatbemBd5xzoqwAwMVmLfazDO98bur5dY0973DzI+toNo2yk6hnbTcR5vc86S9V1zonJWiIh3yuuXNUTbFw7yI1XrmZwoA+hPoGlEXRhZo+ilSnI+toNYyFRKtNNnNf7rG29rnMmZTrmm1WYEzWpvT1q7EWLJuk2B7JhFIVSafRBpQ8qvcLR16Zmq0YO9FcC923V1ht0zlZIm6wUVDjtut1jrAyplhk1dosmiaZoVUkNI4g4rQTPFJH7ReT7IjIuIr/XWL5URO4TkR80/l3i2P+jjW1+ICIfzfoC/DS/3i/pr4DWzSie8Hvl1SkqvXPNJGkiJ/znhOAOLVFkUWMmyCTlvSO4qmV6Y/f8CX4smiSaIlYlNYwgIp2xInIGcIaqPtzo/3oA2Ah8DHhRVXeKyDZgiap+umnfpcAIMERd7hwAzlPVl8LO2aoztpkwR+nikxdlHjkR5tyMIsoRG8VZ2+4OLFTmx++UbqZbwiyLRBznv2G0izBnbJxWgs8CzzY+/1xEHgMGgfcDFzU2+yrwj8Cnm3a/FLhPVV9sDOQ+4DLgtsRX0QIuG/PLkzXGtl+S+HhRwrBVm3YWNWbi+AqqE5Os27k/UIib/Ts5RYjtN4w4JLLRi8hKYC3wEHB6YxIA+DfqjcCbGQSe8n1/urEs6NibRWREREaOHDmSZFhOsiyRG+c1Peq4vT1CpSc7s5Gf9Wcvi2U2MvNCdlgJZqNbiC3oReRUYA+wRVV/5l+ndftPqoB8Vb1ZVYdUdWjZsmVpDjVLlnXp44RlRjk3p2eUU09ZNCdE8KrzBtm171AqZ97waJU9B6qxf4CihU52K0Xre2AYLmKFV4pIhbqQv1VV72gs/qmInKGqzzbs+M8F7FrlhHkH4I3UTTxtIcu69C6ziH+5/3yu7SeO1Rj9g0tmz5GkHILLdBQ0CUE9Nt8Vtuk3L5h9vjUsU9boFuI4Y4W6Df5FVd3iW74LeMHnjF2qqp9q2ncpdQfsOxqLHqbujH0x7JxZOWNbJSgL00WvCD+6ccO85S5H3ZL+Cv0n1R3BPQ5BHOTMGx6tsvXrB6nNnNi+0iPs+uC5XLd7zNkxyjVJDfRVGNt+iWWcGkZJSJsZuw74CHCxiIw1/jYAO4H3iMgPgHc3viMiQyLyZYCGQP9D4LuNvxuihHwRcGnIQbg0ZldM/yuvTs3a+eNo2x479o7PEfIAtRllx95xZ27AQH+FrZeumucXADh6fGpWky9SP1vDMLInTtTNd3CHh78rYPsR4Ld9328Bbml1gJ0gSdSEK2Im6LX+6GtTscoiBDnzXPtNTNZwVU9QrY/j+rvGeenY3P1r0zo7tiCqE5MMj1ZNqzeMElCqEghxibJJu8wdwlyPc5DjLezYZ227O3JsrTjzXNa3lxuTw8Sx4Ekiyv9gbfI6h/lNjCwpVQkEj7C09Dhhkq5oimsvXBFaVCvq2K6wu16RyEJdSxzmmTC884WFAYZFCk3Wprn+rnFL8W8zlnFrZE3pNPqoSJY4nYqSRFP4Na8g56r/2K7GFXEcn9svP4ettx+kNh0viNL/ZhDWMMM775bdY4HHeelYbdbsY82w20MRumkZ5aI0gt4TuEFmCP9DEjebMU6maPOkEuVcTROOl8Tm3ysyZ/KIOq83AcYp32ACJ38s49bImlII+jjhkN5DkmWnorjROf5jpyk10LxvVGhkEjtvkNbvwgROvhShm5ZRLkoh6OMIXO8hybLvZxyBl2empEtTB1h7w71zIm2CzC7Do1V27B2ffSvor/SwpL/CxLFa6BuDCZx8sd60RtaUQtBHCVz/Q5JlNqNL8+oVYUa1LdEScbR8D7/ZJSgB61hthtqMctPVa2a3MYHTfizj1siaUgj6sBDBwYCHJKtKjWmcq36yDKW7/q7xWCasXfsOzUvAghPx9a02S0+ChRC6sWqiRpaUQtBnJXCTkoUgTFrvxr9fkMmmOTGqGc/sEvYW5F+Xl8Bp9boNw0hOKQR9J191WxGESUIyXfsHCcmTF4WnRfjNLmFvQcsH+nLXti2E0DDaRykEPXTPq27SkMwgXEIyzGQz0FdhxxXnzN6jrZeummejh3o9nvVnL8td27YQQsNoH6XMjM2SrJs/xw3JVHCerxVhuPjkuXP6xrWD7PrguXP6xS7pr7DrA+dy/+NHci10NjxapcdRoMciegwje0qj0edBHnbkJELadT6X2WVJf4VXazOBE0l1YpKttx+ccyzXW9B1jizZLLRt754GvclYRI8bc1wbaTCNvkGQ5p5HCd+wejdBBJ3PVYtn++XncOOVq50VNWvTyvV3jbc8xiy07bAmKVYDPxirfWOkxQQ97gfJ5axMo9m6hPTnP3SusxZ00Pn8jtcl/ZVZIblx7eC8piV+oqJywsaYhbbtunczqibkHVjPACMtJuhxP0guLTuNZrtx7eCs1t1csTKOJu1NSv6M1VdrM4nGEOV3CBtjWqyhdnLMcW2kJdJGLyK3AO8DnlPVtzeW7QY89W4AmFDVNQH7HgZ+DkwDU642V53G9cBMq9JX6c08M9RlG4+T+h43LHGgrxJYvuCkXpnTetDlB8griqlb0/s7aSO32jdGWuJo9H8FXOZfoKpXq+qahnDfA9wRtGOD9Y1tCynkwf3AeJpsHpptEGGatKeFxzUn7bjinHktBHsEjk/rvP6yccwAWUUf5fm2kBedtpHnaUozFgaRzcEBRGQl8A1Po/ctF+BJ4GJV/UHAfoeBIVV9Psmg2t0cPKimS6VHOPWURbMFvjoZ5RCnOqeroXjcVoYC/Hjne2OffyE1EHdNsEH3PC8s6saIIqw5eNrwyl8Ffhok5BsocK+IKPDnqnpzyvPlQnNm7Wl9FY4enwpsuOHfbvlAH+vPXsb9jx8JfQDTPqRRsfcu7a7Z/BLWyjDMDLDQs1iLYCPvloRAo5ikFfSbgNtC1r9TVasi8gbgPhF5XFW/HbShiGwGNgOsWLEi5bCS43+Q1u3cP0/znaxNs2PvOK9NzcyJq/+bB5+c3abZ3t1cBjhomziECZSgom0uwnrhhpkBogRd2bVN133zktrKdr1G+Wg56kZEFgFXArtd26hqtfHvc8CdwPkh296sqkOqOrRs2bJWh5UJLsE2MVmLzGr1NN2g6JjmbeIS5kN4YNvFsYVMkK1XgGsvXDGnRn2zLd51/tP6Kh23X7eDsL66Zbxeo3ykCa98N/C4qj4dtFJEFovI67zPwCXAoynOlwtJBFtcnpmYjDS3JHntz8oZF+QIvenqNXxu42rA7XRcf/ayeY5dgKPHp9ixd35Z5LLFePvvWxBlu16jfMQJr7wNuAh4vYg8DWxX1a8A19BkthGR5cCXVXUDcDpwZ91fyyLgb1X1m9kOPx2uEgdXnTfIngPVec7HUyo9sRKOlg/0RQryJJNJltU5w2y9Llv8/Y8f4dRTFs279tq0Op27ZYvx9u7bWdvunhe1BOW7XqNcRAp6Vd3kWP6xgGXPABsan58Azk05vlwJE2w3Xrk6sN57VPSLp2mHNdsW6pNKEvtuO5xxWTodyxrjbTHtRjeyoDNjwwTbxrWDbL101ax27r2aN5s+PnzhisCY8DC7bnOyUlHsu2FZq651S/orCyrG22LajW5kQVevDNPOXGadG69cHSt2OsjcEhTHXqQwxais1aB12y8/B1g4/U2tn6vRjcRKmGo37UqYCksEcplekibJ+EMPXXc6LFmp3fjHe1pfBRFmk8bi5AwYhtEZ8kyY6mrCtLMsarLHyWiFztp3g2LgH9h2ceAbzZ4D1QWTDWsYZWJBC3pwOzmzcLrF6SZV6ZGO2XeDhPnWrx/k+rvGA6OL4piZ4iZPlT3JyjCKxIIX9C6yqLIYS/t3FaFvojnLdkl/he2Xn5NKOAZNRLUZDQ0hDbumzw4/wq0PPjnH2Xzd7jG27B6bk8GbR+euPLFJyeh2FnTUTRhByUVXnTfIrn2HYldwjKP916Y1VuXIrV8/OMeR+9KxGltvP5gqYifLsMnh0eocIe8RFGHUTY00FkLmr1F+TNCH4HVr+vHO97L10lXsOVBN9MCHhVj68WLqXcfate8QtZn5rlxvkmi1hHBS30DYG82ufYeczmYPT5gXoUhYXLppUjIMFyboI/CE6JbdY4kf+Oa3AlfHKgifOMIEoL/tYVKNM+5EBNF14+MKac/8EUQRk466aVIyDBdmow8hTtRM1APvd/ZGHc/l7HQ5hqE+eYRNQGG2Ze/zFkeEkcdAXyXSLh02xubtuqnLVBZOebPxG53GNPoQ4kTNBD3wLlNKVHEsCJ44tl66KrCoWKVXmHbkQcTV9DeuHQwdD9SrdmZhpvKEeTd1mUqbCWs2fqMImKAPIUpbD3rgox5sz+7vEq5BE8fGtYPs+uC5DPRVZpct6a+w6wPnOo8Tpen72XrpKiq94eE/Sc1UYeUhvO09/0eSUsvtJu2kZDZ+owiY6SaEMHOEq+FH3G5MSc0XYUXNgo7jehMJmrw2rh2c1yAl7r5xx+jhMmNEmTdaMX9kZTJJU1DObPxGETBBH4JLGLfilGxenlXNFNdxXCUcBvor85YBvBwh5CG9s9QVPz/ykxfnlIUO6tSVNO6+KLH6Vu3SKAIm6ENoRRgP9FcCE45cJpkshI7rOFtvP0hteq4N/5VXpxgerSZy+EKyDF6XJu1627ntoafm+Rr8b0Gt9KwtSp/b9Wcvm5dfUFTHs1FeTNBHkEQYD49WeeXVqXnLeyS8J2seuMwxtRkNFHZBby9zSJDB69KkXW87Loeyt30r5g/XOi9noR0RMMOjVfYcqM4R8gJcdZ41+jbaiwn6AFq17boSmwIWtSXkzmWOcdnpAafJx0vOah5j83UcOz7l1KRdbw29Ehw95L0FtWL+CGuE7i3P25wT9FahwP2PH8n8XIYRRmTUjYjcIiLPicijvmU7RKQqImONvw2OfS8TkUMi8kMR2ZblwPNgeLTKmuvvZcvuscThcMOj1VDThxdlMTxaZe0Nyc7RHK752eFHYmXCJk1M8iJhXMp78wQRFGHkqpPzzMSkM1Rx0wVnhoYwBkUFVXrDTUmuRujN00meETDmiDWKQpzwyr8CLgtYfpOqrmn83dO8UkR6gT8Dfh14G7BJRN6WZrB54gmtoMiTKGHg7RvGMxOTs9u5KkNet3tsnvAOEqZ/8+CTsSaJVmPA404QcfIM/Pt6oYpLfA7hkxf1MPSmpdx45eo54aOnVJr+azZJ6Nq0cv1d485JLigs0lWiIS/B200ZwEa5iRT0qvpt4MUWjn0+8ENVfUJVjwNfA97fwnHaQpTQCqtHEzexKmo7bfxFFQBrxjURtRoDHneCiCsgm/d9tTYz+9lLxhr5yYu8NnVi+UvHanPuQZBJzL9NEM2x+klyF7LA2g4aRSGNjf4TIvJfgBHgk6r6UtP6QeAp3/engQtcBxORzcBmgBUrVqQYVmvEEVoum27cxCpXM5MgogqANeParpXInrjRRi47+EBfhcUnLwrct5XIm7B7kCSSJovSC0l8K9Z20CgKsVoJishK4Buq+vbG99OB56kroH8InKGqv9W0zweAy1T1txvfPwJcoKqfiDpfu1oJ+ll7w72hddj99Fd6WLL45DkOSNe+/sSqJOeAuk05bg2Zgb4KY9svidwuqRM4bPuwVoyuY5617e7IKpfNDMa4B4djtmJM4wRv5XoNo11k3kpQVX/qO/hfAN8I2KwKnOn7/sbGssLhCot0caw2wzFf5EalR6j0ypyY9WYBMDxa5eUEQh7cBcCCCCmMOXv+5nDLqKiTwA5Utx9kx95xXp6s95G96rzBRH1k405cHr0ikfdAGmONI2zT5C4UJTbfMJLSUq0bETnD9/U3gEcDNvsu8FYROUtETgKuAfa2cr68cdmAo4SnR21GWdQjobbwHXvHmXEfIpD1Zy+bZ2d3MREyibTqaA7sQDWtTEzW5jiGJ44dj3M5gNtu7WJadfYe+J21fhRyi5zxY1E0RrcSqdGLyG3ARcDrReRpYDtwkYisof6MHQY+3th2OfBlVd2gqlMi8glgH9AL3KKq47lcRUpcD2oMq9Ysk7WZUG02qo5MEF68tV8LXbdzf6BG3CPCWdvuDtSqoxy6aQXY0ePxywwkLdngOVC9e7By292pxpqGdpQzsJLGRh5ECnpV3RSw+CuObZ8BNvi+3wPMC70sGq4HeHCgj6OvTcUW0lm/wrtKFgeZMTxHZnViki27x9ixd5wdV9R7ykYJwbAwwCRmFohnynCZT+I4Sl32+nYI27zr6BelPo9RPqxMMeFhcDuuOCdu9n+oQF3iKCYWxmkBpoq4Xav8NeT7T3KbRsIEVZIOVH5a0a7jhoLmHbIYVmY67zr6VtLYyItYUTftphNRN2GvzC5zQTODA308sO1i5/GDioyFUekVrv7lM+c4O9efvWzO9yiNe3Cgj2caQiuIL169JlbUTRLNfkl/hf6TgsMrsyBP84bLNBb222aFKyJJgB/HjCoyFi6ZR92UkbBojDjhfQDHjgdXhvSODyds0z2O+i5+atM6p/Kh5/zE9z0ord9PmJD3jytsfVCpYBeVXuGVV0+Em7rMD0W1RXfS4WoljY28MNNNDILMBZUeob8pTT9upuZNV6+JFPIeUVtFrV8+0Oc074Q1K28mThtEgEU9Mi+CabI2zfV3nfDDp2mvl0drPn8toR7HPWmHsLVMWiMvTKOPQVCkyPqzl3HbQ0/N29bvjAzSWgG2fv1g5mPsr/RwrDY3gLOv0sv6s5dxx4GnOVabPyWcvEhix5/DXO3+ut1jgZPMZC04iPSlY7XZcyWJR09SHbOVN4LmN5WgCdglbNMmX7n2LeKbjtHdmI2+BeKYMQb6Khw9PjUviWpqehqHLJxHlFnGT68Imy6Yb8/3d24KotXMzrh+Cz+enTuuLTquuSho37i4bPK9IsyoOoVtmixZy7A18sBs9CkI0rziFBpzJSeF8eELV8wT1H6bfBjTquw5UJ0jLNbecG/sgmhJBMzwaNVZQ36Jo8MW1M0sb/nMPc7JK211zFZw2d5nVEMnjjRZspZha7Qbs9GH4LIHJ40tj8s3Dj7L1ktXzVZbHHrT0kT7+0PxhkersevqJHE0evfEZeLYfvk5zgxWcHeTyrI6ZhLCcgia+wD4/QBpnLaWYWu0G9PoHQyPVvnk3x0MrKjo0mbTMjFZm2O/j6pxH4QnLJLEXifRhl1adq/InLeJuCYXmFv4zf8G5YpMCquOmRRXEtT6s5eFJi+liZCx6Bqj3ZigDyBMa4W6VtpX6Z0jHJLY08PwerpCtKknCE9Y5KUNh5k6PGEb5FQMewvy4tPjOka9jN8sCCvJEGZeSZMlm3eGrWE0Y4I+gCjb8KBPGCRxfMYlSkh7k0rz5OIXFi7h2lxiOak2HFcbbc5LeMtn7gkU3P4Qz7j3PWs7dlAOxRZH7wDvt0kTIeNt468mOq+jlmFkiAl65jtcw7RPT5g2P+h3f+9ZstHpTwhN1ziuvXAFn9u4OjREL0hrFODK897I5zaubnlsYaaOdTv3O4XepgvODHQsb7rgRCXrsAmu+b5njf9eBpWe8GguHpcmWzaooxZYXRsjexa8oA8qJOUyw/jt0M37JWkoEkV1YpKBvsq8Gvcetz74JH/z4JOzvoIgTXfj2kFGfvLinMxaBfYcqDL0pqVz7OHVicnQYzVz8qKe2ete0l/hvb90xpy3maBsWG9y8TpJeeGg/kknbJLNMyql+bcMK2LnLx6XRjBb5I3RThZ8HL0rjjrILOJ3Nrr2y5JKjzClGqtcclAcdtgYl/RXeOXVqcA6/K6Yblf8N2hgolTS+jBRcfN51UPFlXoAABNBSURBVHxJ81u2WgPH6toYWWNx9CE4a9FzoiBYkCmiHaFwQULYRZA2GDbGsDeQydo0W3aPsWX3GAN9lVnnp0sLdfHMxGSitwbve1C0E+QXlZLmt2x1X4u8MdrJghf0YbXowzS1Vmq15011YpJ1O/fPCtEsxugP+Uwq1E7rqwRG0YSZPVzhmXlGpcS5Ty5zXquC2SJvjHYS6eoXkVtE5DkRedS3bJeIPC4i3xORO0VkwLHvYRF5RETGRKRQNQ28ZBjPJu8nzgMXVas9frmwcJb0VxLVhPcX+cpKaHghn0mFmohb4w+rs5533fdmou7T4EAf1164ItOCY+2+RmNhE2mjF5FfA14B/lpV395Ydgmwv9Eu8I8AVPXTAfseBoZU9fkkg8rbRh9kC/Y0tiQhfP5IjYH+CqrMNs1e+Qt9/POPXpyjBSaNtRfgpqvXACSuCe/VasnKA+ONxXXfmlnSX2HiWC30/O2yR8cpPrb2hnsDzVn+N7u8ipgVgaKPz4gmlY1eVb8tIiublt3r+/og8IE0A2w3QbZmT8gncay5ath7E0mzkP+Vtyzl4Sdfjh1r/ytvWTovCemzw4/Eqn+Tdebu8oG+2Ugef+TMhW9eMu+avFII/jhx1zFbJa5gitueb/vl50SaUsJ6FkSNtcgtAos+PiM9WWRp/Bbw9451CtwrIgdEZHPYQURks4iMiMjIkSNHMhiWm7xrjbgmkgefeIl3rDgtdh34wy/MH8/Qm5bS25OVYSgelR5h66WrGB6tsudAdXYSmVbl4Sdf5qrzBgNNEFGXuf7sZYnG4ZnbVm67m+t2j8WqSR+3PV+eppSitwgs+viM9KRyxorI7wNTwK2OTd6pqlUReQNwn4g8rqrfDtpQVW8Gboa66SbNuKLIO+LBNWFMq/LAj15MdZzr7xpnOkE0Tlr6Kj3ceOUvsXHtIOt27g8UCPc/fiTwTWgiIrfAH9MfRbPW2XwHXDHoSSb1VjX2KIpexKzo4zPS07JGLyIfA94HXKsOQ7+qVhv/PgfcCZzf6vmyJO9OPllNGEHHyTIxKx4n1PKkAiHqPjR3nnLhFZiLMnkFjSOsOmW7KMIYwij6+Iz0tCToReQy4FPAFap6zLHNYhF5nfcZuAR4NGjbdpN3xMP6s5eljrppnniGR6usveHekD3ywS+MkwqEOKYZr/MUzG3p55UFjiowFzWOIrTnSzOGsFLJRRif0R1Emm5E5DbgIuD1IvI0sB34DHAydXMMwIOq+jsishz4sqpuAE4H7mysXwT8rap+M5eraIG8XtM9O3Ya40pz5E+STkt54AnjJLHf3n2Ig2cLDnII+ssthOEaR1bt+dJEpSQdgz/JzB/VlJeT1FoYlp8FXwIha9KWRkhayqBdeBFJzcW/ROq2eK+Cp9chy1VL3sVAXyU0QieIVkJiW8E10fqzhvM+l59Wyy4Y5cZKILSRMAfWYAxhGORU7LSQDxqDUs8Z8Gub/rDPpOGdSYV8rwif/9C5mQvZIK3WVT55YjL7ipNx2ieak9RIign6jElSUuEsR4Nt/4Och022FXpFIiNf8kCAU5qavOTRSDssljxMsGZdcTKOEDcnqZEU63aQMUkcW3Gcm0WJZZ5WjRX5kjUKzjj9LAmLJY8SrFlq2FHnMidpOcnb6W4afcbEcWy5nG0w90EeHq2mMttk1d7QO1ZW2bZJx+WK08+SsNDRoNIPfpJq2K00jGmHL8LoDO3ITDZBnwNhET1B5o+gB3l4tMrW2w8GHiOMvkrPbG34HoGAviUtkaWZ5toLV8xru1jpEWdZ5nbYpAf6K4E5Cl7pB6gnqzVvk1TDjnqoLQJm4dGOJjQm6NtM3Do71981HthdKozFJ/Xil5VZCfksGRzo43MbVzP0pqXzhJmrNk7eNunh0SqvvDo1b3mlV2aFuCeE0xb/ivNQ5xX6axSTdmQmm6BvM3F/1FYyYCu9PYmjV9LgtQP0ooj84ZYDAR2s/NqvS5hFxennUWVx175DgW8Ti09aFFgvP8358niorfJkd9OOJjQm6NtMXj/qQF+Fl9so5ONEvjTH3NemZwI7V3lEmS3ysmW6hGwe9zPr398qT3Y/7WhCY1E3bSZuVM5AX8V5jErv3AILlR5BpD3hjh6euSEsOmDj2kEe2HYxN129hqPHpzh6/MR/ZK9zVfP+3j4/3vleHth28RxhlVeVxXbWesm63IBVnux+2tGExjT6NuP9eH579CmV+fPtjivOYevXD84xKVR6hF0fPBdgjqZ89PhUB4qdxdced+07FOhv8DpXRf2H9kcpBZHWltnOtn5ZO1ut8mQ5yNsvY4K+Q7w2NTP7+aVj8zMsowSC9++6nftj2+V7pG5Xb7abn1LpaXmi8GuPrrGGCZ0ogRSnJEBazbvdkS5ZPtTWZNyIg9W66QCu2jWt1DBZ6ciuDTq2p6E2CzSY7wRNSl9T9mqlV1h80iJenqyF1r2JuuaoOj95ZMl2kqSO1aCJsF33xJzAxcJq3RSMLF+3eyOKhwU99K6H0f/QHn1tKlEET/MkUZvW2f1d4/M6V4URVTuoTMKlFcdqp+LuzQncXZig7wBZvm6HCfkoQRimkQVpipXeusfXldiUBH/nqjCS1A7qdlpNnOlE3H07knyM7LComw6QZeTFoGNy8ARhVIauq+9qUCTArg+cy64Pnhu7520YSxefPJuAFFbjYyE1xegmx2o3jdUwQd8RsgynalUQRoXlDY9Wuf6u8dmJ4OhrU7Njn8nAr/PMxGTkZOOdL+/Qs6LQTS39ummsRkxnrIjcQr0/7HOq+vbGsqXAbmAlcBj4kKq+FLDvR4HPNr5+TlW/GnW+sjtjs6bZBOOvee+y2Z617e7AuHsBbrp6DVtvPzgvJLIH+MLVa0JDHePivYksFLNMHDrpWE1KN411oRDmjI2r0f8VcFnTsm3At1T1rcC3Gt+bT7yUeuvBC6g3Bt8uIktintOIiT/JaOulq9hzoBqqJUO4RuaKe5+hHv8f1Qu2r9LLhy9cMSvMmw093huHvf7PpZveXrpprEaC8EoRWQl8w6fRHwIuUtVnReQM4B9VdVXTPpsa23y88f3PG9vdFnYu0+hbJ27opksju+q8wTmdooIYdDhIvXWusszNbxhJw0wtnM8w3OQVXnm6qj7b+Pxv1JuBNzMIPOX7/nRjmZEBQYIvrpYcFJa3/uxlsRp6u4S8QKCAdkWFJG027s8Urk5MsvXrB+dci7edZ1ryQk/LFoZp5EsZFYpMwitVVUUklYdORDYDmwFWrFiRxbBKjSuOOayuejPNpXejNHk4YYYJ+rGTOuKSxIDv2Ds+L6yzNqPs2DvuDAn1Qk8txtuIS5b5AUWaMNII+p+KyBk+081zAdtUgYt8398I/GPQwVT1ZuBmqJtuUoxrQeCKmjl5Uc+8LNWwKJw4JQb8uH4YgZZCHuPGgLuSt/zLwxprT9am+eTfHeS63WMdf+iM4pJVfkDREsrShFfuBT7a+PxR4P8EbLMPuEREljScsJc0lhkpCSute+OVq1nSf6L65cmL3D9zmHBMgtJ5bTnKiTutGuqgNoysAgSKVlU0lqAXkduAfwFWicjTIvLfgJ3Ae0TkB8C7G98RkSER+TKAqr4I/CHw3cbfDY1lRkqi4phfrZ0omjYxWXMKtqwiXFyJWxDd+DhOY2T/xNXMZ4frmlIS05GV8jWCyCo/oGgRZbEEvapuUtUzVLWiqm9U1a+o6guq+i5VfauqvtsT4Ko6oqq/7dv3FlX9D42/v8zrQhYaYYlSSbSJuP+Bo3Jho0xDrnDPOElTANsvP8d57tseemp2DM33JIyFGsZpuEmTie1XWHoc2eOdSiizzNguJSyOOYk24fqP7cXBe8e+6eo1Tq19oK8SWigtbNKJOymFmYU8p6v/ngCzpRpcJRssi9NoptX8gGaFJagGVSdLd1hRsy7G5chMUjQtafXDoHDIHVe4te2oSSfJpOSq1OkX5EH3xJUzUMZ6OVCsaI+4FGnMrRSJc/m6ekWYUe34NZmgLyFJOybF/Y/dSkncqEknar1fAPRVejhWmy/oN11wZubj7laKFu0Rh24cczMuhWVGlR/vfG+bRzMfE/QlJE/BllTbCZp0BGbLKIRNSs0C4Fhthp5Gb1zVura06YIz+dzG1ZmPu1vpxvLBacdchLeBonf6MkFfUooi2DauHWTkJy9y64NPzsbgK7DnQJWhNy0NnZTW7dw/TwDMaPKiZ0UQBO0aR9GiPeKQZsxFeRtoZ9/hVjBnrJE79z9+ZF6iVZzwxiyEVtyonrxp1zi6sXxwmjEXJV696EXeTKM3cidMYAdpZFt2j3H9XeOJyjm4cAmC6+8ab6uW3y6TStE1yyDSjLlIbzBFeYsOwgS9kTth9ktXtMJLx2pUeoRKr8wpmZxUaLke+JeO1WYnkXa87rdLIHWj4znNmItuGy8KJuiN3AnT2K7bPebcrzajDPRVWHzyopaFlksQNJO3w7KdAqnImqWLVsfcjW8wncBs9EbuhNkvowTdy5O12aYqYT1wXSTJls3zdX8h9b5tJ0W3jRcF0+iNTIiKKElSk95PWo03yCxw9LWpwGqYeb7ud4NJpSjRSUnpxjeYdmOC3khNmhA3b/2OvePzhG9WGm+zIOhUpmyRBVJRwhSNfDDTjZGatCFuG9cOMrb9Er7YqKeT9yu4ve7PpyhhikY+mEZvpCariJJ2arxF1q47QZHCFI3sMUHfRRTVhmohbt2P/Yblxkw3XUJRMjyDKFtESZxGKGU4p5+y/YbGXEyj7xKKXKyqGyJK4tIJp2QRHKFl+g2N+bQs6EVkFbDbt+jNwB+o6hd921xEvZfsjxuL7lDVG1o950Km6DbUsti8OzGhFmUSL8tvaMynZUGvqoeANQAi0gtUgTsDNv0nVX1fq+cx6pgNtT10YkIt+iRudD9Z2ejfBfxIVX+S0fGMJsyG2h46Uf2xGytOGt1FVoL+GuA2x7r/JCIHReTvRcTZc05ENovIiIiMHDlyJKNhlQeL/W4PnZhQbRI38kY0oAdnogOInAQ8A5yjqj9tWvfvgBlVfUVENgB/oqpvjTrm0NCQjoyMpBqXYbRKJ8JYixo6a3QPInJAVYcC12Ug6N8P/K6qXhJj28PAkKo+H7adCXrDMIxkhAn6LEw3m3CYbUTk34uIND6f3zjfCxmc0zAMw4hJqjh6EVkMvAf4uG/Z7wCo6peADwD/XUSmgEngGk37CmEYhmEkIpWgV9WjwC80LfuS7/OfAn+a5hyGYRhGOqwEgmEYRskxQW8YhlFyUkfd5IGIHAFaTb56PRAa1VNy7Prt+u36FyZvUtVlQSsKKejTICIjrhCjhYBdv12/Xf/CvX4XZroxDMMoOSboDcMwSk4ZBf3NnR5Ah7HrX9jY9RvzKJ2N3jAMw5hLGTV6wzAMw4cJesMwjJLTVYJeRC4TkUMi8kMR2Raw/mQR2d1Y/5CIrGwsXykikyIy1vj7UvO+3UCr199Y90si8i8iMi4ij4jIKe0cexak+P2v9f32YyIyIyJr2j3+tKS4/oqIfLXxuz8mIp9p99izIMX1nyQif9m4/oONFqcLC1Xtij+gF/gR9d60JwEHgbc1bfM/gC81Pl8D7G58Xgk82ulr6OD1LwK+B5zb+P4LQG+nr6ld19+0zWrq3dA6fk1t/P1/E/ha43M/cBhY2elrauP1/y7wl43PbwAOAD2dvqZ2/nWTRn8+8ENVfUJVjwNfA97ftM37ga82Pt8OvMsrk1wC0lz/JcD3VPUggKq+oKrTdBdZ/f6bGvt2G2muX4HFIrII6AOOAz9rz7AzI831vw3YD6CqzwETwIJKquomQT8IPOX7/nRjWeA2qjoFvMyJ6ppnicioiPxfEfnVvAebA2mu/xcBFZF9IvKwiHyqDePNmrS/v8fVuNteFpk01387cBR4FngS+GNVfTHvAWdMmus/CFwhIotE5CzgPODM3EdcIFKVKe4ingVWqOoLInIeMCwi56hqt2k1rbIIeCfwy8Ax4FuNbjTf6uyw2ouIXAAcU9VHOz2WNnM+MA0sB5YA/yQi/6CqT3R2WG3jFuA/AiPUa2j9M/X7sWDoJo2+ytxZ+I2NZYHbNF5TTwNeUNXXVPUFAFU9QN3W94u5jzhbWr5+6trPt1X1eVU9BtwDvCP3EWdLmuv3CGtiX3TSXP9vAt9U1VrDdPEA3We6SPP8T6nqdaq6RlXfDwwA/68NYy4M3STovwu8VUTOajQkvwbY27TNXuCjjc8fAParqorIMhHpBRCRNwNvBbpNm2n5+oF9wGoR6W88AP8Z+H6bxp0Vaa4fEekBPkR32uch3fU/CVwMs13hLgQeb8uosyPN89/fuG5E5D3AlKp22///dHTaG5zkD9hAfSb+EfD7jWU3AFc0Pp8CfB34IfCvwJsby68CxoEx4GHg8k5fSzuvv7Huw4178Cjwvzp9LR24/ouABzt9DZ24fuDUxvJx6hP81k5fS5uvfyVwCHgM+Afq5Xw7fj3t/LMSCIZhGCWnm0w3hmEYRguYoDcMwyg5JugNwzBKjgl6wzCMkmOC3jAMo+SYoDcMwyg5JugNwzBKzv8HEikFfzczNpkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Q-dT4pgBS4-8"
      },
      "source": [
        "def predictValues(realValues,predictedValues):\n",
        "  correct = 0;\n",
        "  count = 0;\n",
        "  actualValues = np.array(realValues).reshape(-1,1)\n",
        "  for i in realValues:\n",
        "    print('Actual: ', realValues[count], ' Predicted: ', predictedValues[count])\n",
        "    if predictedValues[count] == realValues[count]:\n",
        "      correct+=1\n",
        "    count+=1\n",
        "  print('Total values: ',count, ' correctly predicted values: ',correct)"
      ],
      "execution_count": 1504,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5L63W60GpGVH"
      },
      "source": [
        "def predictKnnValues(realValues,predictedValues):\n",
        "  correct = 0;\n",
        "  count = 0;\n",
        "  actualValues = np.array(realValues).reshape(-1,1)\n",
        "  for i in realValues:\n",
        "    print('Actual: ', realValues[count], ' Predicted: ', predictedValues[count])\n",
        "    if predictedValues[count] == realValues[count]:\n",
        "      correct+=1\n",
        "    count+=1\n",
        "  print('Total values: ',count, ' correctly predicted values: ',correct)"
      ],
      "execution_count": 1505,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sEhh-MN3TxmC"
      },
      "source": [
        "# KNN Predictions\n",
        "\n",
        "Here we are using KNN to make predictions based on a number of features fed into the model.  As we can see it's score is 0.96 which isn't perfect but it's better to avoid overfitting the model so that it can adapt to new data. This algorithm works by clustering data with similar features and classifying them."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RKUgWqgQ5R0H"
      },
      "source": [
        "### references\n",
        "\n",
        "SKLearn documentation: https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsClassifier.html"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ndm4DL8G1lG-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d57134a9-0d5a-4bd5-a00e-9393feea973c"
      },
      "source": [
        "# K Nearest Neighbour\n",
        "\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "knn = KNeighborsClassifier(n_neighbors=55)\n",
        "knn.fit(X_train,y_train)\n",
        "print('Performance of model: ',knn.score(X,y))\n"
      ],
      "execution_count": 1506,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Performance of model:  0.9156414762741653\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uuntf9ZpWj4M",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d049bb03-48cf-4ce0-ad31-e28774822240"
      },
      "source": [
        "# Make prediction on testSet\n",
        "predictValues(y,knn.predict(X))"
      ],
      "execution_count": 1507,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  0  Predicted:  0\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  1  Predicted:  1\n",
            "Actual:  0  Predicted:  0\n",
            "Total values:  569  correctly predicted values:  521\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kZglfe8VT1ZP"
      },
      "source": [
        "# Naive Bayes prediction of diagnosis\n",
        "\n",
        "Here we use naive bayes on both sets to perform  the diagnosis based on features from the dataset, I chose to input a large number of features into this model as generally this algorithm performs better when fitted with more features, in Naive Bayes the weights of the features have no bearing on the output.  The model below appears to be overfitted\n",
        "\n",
        "## References\n",
        "\n",
        "Sklearn documentation: https://scikit-learn.org/stable/modules/naive_bayes.html"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xkv0-QPOmtPV"
      },
      "source": [
        "You may also notice we are using a particular kind of Naive Bayes called Gaussian Naive Bayes for this model.  We chose to include this version of Naive Bayes due to it's ability to handle continuous data. The algorithm also has the advantage of speed and is frequently used in email spam classifiers.  The algorithm works by assuming that the continuous values across the features are normally distributed in a Gaussian fashion(think of a bell curve!)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Tt8iXCKZNIvw",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7eba6328-6624-492d-ff38-cefc28e6a0d3"
      },
      "source": [
        "\n",
        "from sklearn.naive_bayes import GaussianNB\n",
        "yTestArray = np.array(y_test).reshape(-1,1)\n",
        "y_trainArray = np.array(y_train).reshape(-1,1)\n",
        "probability = np.array(y_test.iloc[0:2]).reshape(-1,1)\n",
        "\n",
        "# here we smooth the data to improve accuracy\n",
        "naive_bayes = GaussianNB(priors=None)\n",
        "y_train_pred = naive_bayes.fit(X_train,y_train).predict(y_trainArray)\n",
        "print('Score of model on training set: ', naive_bayes.score(X_train,y_train))\n",
        "y_pred = naive_bayes.fit(X_train,y_train).predict(yTestArray)\n",
        "print('Score of model on test sets: ', naive_bayes.score(X_test,y_test))\n",
        "print('Probability of diagnosis values occurring: ',naive_bayes.predict_proba(probability))\n"
      ],
      "execution_count": 1508,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score of model on training set:  0.9623115577889447\n",
            "Score of model on test sets:  0.9122807017543859\n",
            "Probability of diagnosis values occurring:  [[0.         1.        ]\n",
            " [0.92010396 0.07989604]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3LlAxnEfQ8fC",
        "outputId": "f2a1ae93-653b-4a61-9072-2d416f1d4c8d"
      },
      "source": [
        "predictValues(yTestArray,y_pred)\n",
        "predictValues(y_trainArray,y_train_pred)\n"
      ],
      "execution_count": 1509,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Total values:  171  correctly predicted values:  171\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [1]  Predicted:  1\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Total values:  398  correctly predicted values:  398\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3Ii6cjIA7OOF"
      },
      "source": [
        "## Second Naive Bayes Model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WintGUKo8jer"
      },
      "source": [
        "Demonstrating the zero frequency phenomenon, we can see below that when we train the model using our NB train sets we experience a phenomenon known as zero frequency.  This occurs when the model takes all probabilities of our features and performs calculations which result in a total score of zero.  Here we induced zero frequency by removing the smoothing of the data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C7uxefjX-OiQ",
        "outputId": "2f948931-34a3-4aab-bddb-9f87da838aea"
      },
      "source": [
        "naive_bayes = GaussianNB(priors=None,var_smoothing=1)\n",
        "\n",
        "\n",
        "y_train_pred = naive_bayes.fit(X_train,y_train).predict(y_trainArray)\n",
        "y_pred = naive_bayes.fit(X_train,y_train).predict(yTestArray)\n",
        "print('Score of model: ', naive_bayes.score(X_test,y_test))\n",
        "\n"
      ],
      "execution_count": 1510,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score of model:  0.8421052631578947\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "loGauTUK-h63",
        "outputId": "bc021a15-d250-4210-ea7b-afc0c92241ed"
      },
      "source": [
        "predictValues(y_trainArray,y_train_pred)\n",
        "predictValues(yTestArray,y_pred) "
      ],
      "execution_count": 1511,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Total values:  398  correctly predicted values:  245\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Actual:  [0]  Predicted:  0\n",
            "Actual:  [1]  Predicted:  0\n",
            "Total values:  171  correctly predicted values:  112\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5a4h9R6FmTOr"
      },
      "source": [
        "# Decision Trees"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fIYuKECtrdqG"
      },
      "source": [
        "from sklearn import metrics #Import scikit-learn metrics module for accuracy calculation\n"
      ],
      "execution_count": 1512,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Lhl5NMMpmY1w"
      },
      "source": [
        "Training the first default tree model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2UbsvOVbmUcb"
      },
      "source": [
        "clf = tree.DecisionTreeClassifier()\n",
        "\n",
        "# Train Decision Tree Classifer\n",
        "clf = clf.fit(X_train,y_train)\n",
        "#clf = clf.fit (X,y)\n"
      ],
      "execution_count": 1513,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XzUk3uZBrHr1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "917d8b7c-50e2-4c5a-9bd2-62e90a3dd184"
      },
      "source": [
        "print(y_test.shape,y_train.shape,X_train.shape,X_test.shape)"
      ],
      "execution_count": 1514,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(171,) (398,) (398, 30) (171, 30)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JYAyAyCQmdnh"
      },
      "source": [
        "Here we display the tree"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BXPMuy4QmW3M",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 930
        },
        "outputId": "be856f9f-e0a6-43b0-cb3f-680256c0dd18"
      },
      "source": [
        "from IPython.display import Image\n",
        "from six import StringIO\n",
        "import pydot\n",
        "dot_data = StringIO()\n",
        "tree.export_graphviz(clf, out_file=dot_data, feature_names=features)\n",
        "graph, = pydot.graph_from_dot_data(dot_data.getvalue())\n",
        "Image(graph.create_png())"
      ],
      "execution_count": 1515,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAB6UAAAORCAIAAAA1ceAoAAAABmJLR0QA/wD/AP+gvaeTAAAgAElEQVR4nOzdeVyUVf//8TOsM2wDKAqyKLjhguWS4VaguadpgJD6TStNQ2/3LUHTTBI1oVT0m5p136WiYpqaG+KumN2aKaYRxiYIigrCDDLA/P64vvf8uNlEFkfx9fyL61znOtf7DIMP58PhXDKtVisAAAAAAAAAAHjOGeg7AAAAAAAAAAAAtYB6NwAAAAAAAACgPqDeDQAAAAAAAACoD4z0HQAAAAD1yrlz51JSUvSdAniOjRgxQt8RAAAAnlcynlcJAACAWuTn57dz5059pwCeY3xGAwAAqDbWdwMAAKCW+fr67tixQ98pgOfP9u3b/f399Z0CAADgOcb+3QAAAAAAAACA+oB6NwAAAAAAAACgPqDeDQAAAAAAAACoD6h3AwAAAAAAAADqA+rdAAAAAAAAAID6gHo3AAAAAAAAAKA+oN4NAACApy0qKsrNzU0mk8lksgULFpTbZ9WqVTKZzMDAwN3d/eTJk085oeTGjRv/+Mc/2rVrZ2lpaWRkpFQqW7VqNXjw4HPnzuklT+0q+V2QyWTGxsaOjo6jRo36448/qjjCzz//rFQq9+7dW6c5a9fp06d79OhhZmbm4OAwd+7cR48e1bxzfn6+u7t7cHCwrkWj0SxcuNDNzc3ExMTR0XHWrFlqtbqWZwIAAIDyUO8GAADA0+bj43Pz5s3mzZsLITZu3KjRaEp1KCoq+uqrr4QQvXv3vn79+muvvfb0Q27atMnDw+P3339ftWpVSkpKXl7epUuXlixZ8uDBgytXrjz9PLVO911QKpVarfbBgwfr168/ffp0165db9y4UZURtFptXYesXXFxcf369evTp8+dO3d27dr1zTfffPTRRzXvHBQUVOoVmzZt2ooVK0JCQrKysn744YcNGzaMGzeulicDAACA8lDvBgAAgN507tz59u3bu3fvLtUeFRXl6Oiol0iS2NjYCRMm9OrV6+jRo/3797e2tjY1NXVzc/P391+4cGFBQYEes9URc3PzIUOGfPnll7m5uatXr67KJYMHD87Ozh4yZEhVOqvV6u7du9cso9BoNJs3bx4zZkz1Ll+yZIm9vf3ixYvNzc27des2d+7cb7/99vr16zXpfPbs2atXr5ZsuXnz5vr16999992AgABLS0svL68pU6Zs2bKl6gvnAQAAUG3UuwEAAKA3gYGBQoh169aVal+1atXMmTP1kej/fPbZZ0VFRZ9//rmRkVGpU/379588ebJeUj0FXbt2FUKUKuDWik2bNmVmZlb78kePHq1bt65Tp05xcXHLli2rxgiFhYX79+9//fXXZTKZ1DJw4ECtVrtnz55qd1ar1bNnzw4PDy/ZeOHCheLi4ldffVXXMmDAACHEoUOHqhEbAAAAT4R6NwAAAPSmd+/ebdq0OXbsWMntIM6cOaNSqfr161eqc1FR0cKFC11cXBQKRYcOHSIjI6X2U6dOtW3bVqlUyuVyDw8PqaoYERFhbm5uZma2Z8+egQMHWllZOTk5bd26tSqpCgoKjh492qBBA6n4WwmtVrtq1ao2bdqYmpra2NgMGzZMt/63KgH+9a9/denSRS6Xm5ubN2vWbMmSJRVNp02bNtJu5p07d1apVEKIOXPmSH2+/fbbSl6cJ1VYWCiEMDU1fewET58+7eLiIpPJ1qxZ89j5Tps2bebMmQkJCTKZrEWLFkKIEydOdO3a1czMzMrKysPDIycnp6JIeXl5q1at6tKlS2Zm5okTJ1auXOng4CCEyM3NlVXA09Oz7Dg3b97Mzc11cXHRtUg76vz+++/V7hwUFDRp0iQ7O7uSjQYGBkIIhUKha2nZsqUQgvXdAAAATwH1bgAAAOjTxIkThRDr16/XtXzxxRczZswo23PevHnLly8PCwtLT08fMmTIyJEjf/31VyFERkaGv79/YmJiWlqahYXFqFGjhBCBgYHTp09Xq9WWlpaRkZEJCQlubm7jx48vu1d4WUlJSfn5+VKNsnKLFi36+OOPg4KCMjMzT548mZKS0qtXr4yMjKoECA8Pf/fdd319fdPS0lJTU+fPny8V/cudztWrV5s1a+bs7PzLL7+YmZkJIZYvX/7BBx8sW7Zs7Nixlbw4T0p6NOhLL7302An27Nnz7Nmzugsrn294ePiQIUOaN2+u1Wr/+uuvvLy8oUOH+vr63rt3Lz4+vlWrVuVuEZOdnb106dLu3btrtdrY2NhPPvnE1tZWd9bCwkJbgdjY2LKj3b59WwhhaWmpa5HL5QqFQppONTqfOXMmISFh5MiRpa51d3cX/13dbtCggRDizp07ZW8EAACA2kW9GwAAAPo0ZswYc3Pz7777Tq1WCyFu3rx54cKFsjXE/Pz8iIiI4cOH+/j4WFtbBwcHGxsbb968WQjh6+v7ySef2NjY2NraDh06NCsrq2RhsXv37lZWVnZ2dgEBAXl5ecnJyY+NJK01trCwqLybWq1etWrV22+/PXr0aKVS6eHhsX79+rt373799dclu5UbQKPRLF682Nvbe968eba2tjY2Nh988MErr7xS0XQMDQ2nTp2anJy8a9cuaViVShUVFfX+++9X/uJUXV5eXlRU1KxZsxo1ajR16tSqT7CUqrzgiYmJOTk57dq1k8vljRs3joqKatiwYak+Go3mpZdeOnz48OnTp2fOnGlubv5E0ynr0aNHQghDQ8OSjcbGxtIb70k7q9XqadOmRURElL3Ww8NjwIABa9eujYmJyc/Pv3379q5du2QyWVV+1wIAAIAaot4NAAAAfVIqlSNHjrx///62bduEEGFhYYGBgSYmJqW63bhxQ6VStW/fXjpUKBT29vZlHx5obGwshCgqKip7I2nMqtQcpUq3tG1IJeLi4nJzc7t06aJreeWVV0xMTM6fP19u/5IBfv/99wcPHvTv3193VqpoVzKdcePGKZVK3VbR33///bBhw6ysrESVX5yKZGdny2QypVI5derUQYMG/fLLL9LDQp90gpXMtxQ3N7dGjRqNHj160aJFiYmJ5V5ubGx8+fLlvn379urVKywsLC8vr4rTqYhcLhf/2bBFp6CgoOTGI1XvPH/+/A8//LCip6pu27bNz8/v3XfftbW17dGjx48//qjVaqVV3gAAAKhT1LsBAACgZ9JTK9evX//gwYMdO3ZIO5yUIpU7g4ODdXs0JyUlSSXp/fv3e3l52dnZmZqazpkzp+Z5mjVrJpfL//zzz8q7PXjwQJRZBm5tbf3w4cPH3kJaQm5tbV32VEXTsbCw+PDDD8+ePfvLL78IIdatWzdlyhTpVCUvTlUolUqtVltYWJiamvrNN980bdq05hOsnEKhiImJ6dmz59KlS93c3AICAspdZK1UKoODg8+cOVNcXOzp6fnpp5/ev39fd/ZJ9++2t7cX/3nlJSqVKj8/X9oN/Ik6nz59+sqVK+PGjatogkqlcv369ampqSqVKiEh4YsvvhBCNGnSpAqvDQAAAGqEejcAAAD07OWXX/b09Pzll18mTJjg5+dnY2NTto/0SMCwsLCS2zSfO3cuOTl5+PDh9vb258+fz87ODg0NrXkeU1PT/v37371798yZM2XP3rt3Typ0StXqUsXfBw8eODk5PfYWUunz7t27pdorn86UKVOMjY3DwsJOnjzp7OwsPUFRVPziVH3K5arJBB+rXbt2e/fuTUtLmzt3bmRk5MqVKyvqaW5uPnPmzAsXLjRs2PC1116bPXt2enq6ePL9u11dXS0tLZOSknQtf/31lxCiQ4cOT9p506ZNR48eNTAwkMrr0uu/dOlSmUxW7rbpFy5cEEJ4e3tX9dUBAABAdVHvBgAAgP5JS7x37tw5ffr0cjs4OzvL5fLffvutVPuVK1c0Gk1gYKCbm5tcLpfJZLWSZ9GiRaampjNmzCi77vjq1atGRkZCiPbt21tYWJSsb54/f76goKBz586PHb9Zs2a2traHDx8u1V75dJycnEaMGLFz584FCxZMmzZN117Ri1NDNZlg5dLS0q5duyaEsLOz+/zzzzt16iQdVkIulwcGBl68eLFNmzbz5s2rxk2NjIwGDRp08uTJ4uJiqeXAgQMymWzo0KFP2nnz5s0ly+vSfvFBQUFarbbk9i86GzZscHV1ff3116sRGwAAAE+EejcAAAD0b8SIEQ0bNhw+fLibm1u5HeRy+Xvvvbd169aIiIicnJyioqLU1NT09HQXFxchRHR0dH5+fnx8fBW3ln6sl19++Ycffrh69WqvXr1+/vnn7OxsjUbz999/b9iw4YMPPpC21ZbL5TNnzty1a9f333+fk5Nz5cqVjz76yMHBYcKECY8d39TUdP78+SdPnpwyZcqtW7eKi4sfPnx47dq1x05n5syZhYWF9+/f792792NfnBq+CDWZYFm2trZpaWmJiYkPHz5MSkqaOHHi9evXCwoKLl26lJSUVO4OJGUZGxu///773333XTUCCCEWLFiQkZHxySef5OXlnTt3bsWKFWPHjm3durV0duHChUqlUvdLiMo7V65r165JSUmFhYWJiYmzZs2Kjo7etGlT2V3pAQAAUPsq+htAAAAAoBp8fX19fX0r77Nr1y5pL46GDRtOnjxZapwzZ87Zs2elr4ODg6UNlA0MDNq2bXvq1CmtVvvo0aO5c+e6uLgYGRnZ2dn5+PjExcVptdq5c+fa2tpaW1v7+fmtWbNGCNG8efN58+aZmZkJIVq2bJmQkPD1119Lj3Zs2rTpn3/+WcW5JCcnz5o1y8PDw8LCwtDQ0NraumPHjh988MGZM2ekDsXFxStWrGjZsqWxsbGNjc3w4cNv3LghnVq7du1jA6xZs8bDw0Mul8vl8o4dO65du7ai6SQnJ+tSeXt7b9y4sVTUil6cSpw5c6ZVq1bShwIHBwc/P7+yfSqZ4OrVq6XvkZmZ2dChQx8734sXLzZt2lShUPTs2fP8+fPdu3e3sbExNDRs0qRJUFBQYWFhFb8pNXTixImuXbuampo6ODjMnj07Pz9fd2rBggWWlpaHDh2qSueSSq7vlvTt29fa2trIyMjGxmbw4MEXLlyoYrzIyEg+owEAANSETKvVPvUaOwAAAOotPz8/IcSOHTv0HQR4/mzfvt3f35/PaAAAANXGfiYAAAAAAAAAgPqAejcAAABeLNevX5dVLCAgQN8Ba8cLMk0AAACgJCN9BwAAAACeKnd39xdhv4gXZJoAAABASazvBgAAAAAAAADUB9S7AQAAAAAAAAD1AfVuAAAAAAAAAEB9QL0bAAAA0Keff/5ZqVTu3bu31jtXxenTp3v06GFmZubg4DB37txHjx5V5ar8/Hx3d/fg4GBdi5eXV9mnYlpYWEhnQ0ND3d3dFQqFubm5u7v7ggULcnJyamsKAAAAgA71bgAAAECfnuipkrX7CMq4uLh+/fr16dPnzp07u3bt+uabbz766KOqXBgUFHTjxo3HduvZs6f0xalTp8aPH5+cnJyRkbFkyZLQ0FBfX98aRQcAAADKY6TvAAAAAMALbfDgwdnZ2XXR+bGWLFlib2+/ePFimUzWrVu3uXPnzpkzZ86cOe7u7pVcdfbs2atXr5ZqlMvlOTk5lpaWupaJEyeOGDFC+trExGTSpElyuVwI4efnt2PHjh07dqSnpzs4ONTWXAAAAADB+m4AAADgKdNqtTt27Pj666/1G6OwsHD//v2vv/66TCaTWgYOHKjVavfs2VPJVWq1evbs2eHh4aXaDx48WLLYnZKScvXq1d69e0uHu3btkordEkdHRyFEbm5urUwEAAAA0KHeDQAAANStoqKikJCQ1q1bKxSKhg0burq6hoSESGufT58+7eLiIpPJ1qxZI4SIiIgwNzc3MzPbs2fPwIEDraysnJyctm7dKo1TqnMpubm5ZXfQlnh6epbtf/PmzdzcXBcXF11L8+bNhRC///57JXMJCgqaNGmSnZ1d5VNetmzZ1KlTKzobHx9vbW3dtGnTygcBAAAAnhT1bgAAAKBuhYaGLly4cMWKFffu3Tt8+HB+fr61tbW1tbUQomfPnmfPntX1DAwMnD59ulqttrS0jIyMTEhIcHNzGz9+vEajKdu5FAsLC20FYmNjy/a/ffu2EKLkomy5XK5QKDIyMiq6xZkzZxISEkaOHFn5fG/dunX8+HEfH59S7RqN5tatW2vWrImOjl69erWJiUnl4wAAAABPiv27AQAAgLq1e/fuzp07Dx06VAjRqVOnt956a+PGjQUFBZUUfLt37y5tABIQEHDq1Knk5GRp8XUtevTokRDC0NCwZKOxsbFarS63v1qtnjZt2u7dux878rJly/7xj38YGJReW+Ps7JyRkdGgQYPly5f7+/tXNzgAAABQIdZ3AwAAAHUrPz9fq9XqDouKioyNjUsVmisi1cSl9d21S6qnFxYWlmwsKChQKBTl9p8/f/6HH34obb1dibS0tJ9++mns2LFlT6WkpGRmZm7ZsuW7777r2LFjZmZmNaMDAAAAFaDeDQAAANStQYMG/fvf/96zZ49arf71119379795ptvVrHeXXVPun+3vb29ECInJ0fXolKp8vPzHRwcynY+ffr0lStXxo0b99gYoaGh48ePL/l0Sh1jY2M7O7t+/fpt27YtLi4uJCTkCaYHAAAAVAH7mQAAAAB1a9GiRf/+97/Hjh2bm5vr4OAwYsSIpUuX1vpdpP27q97f1dXV0tIyKSlJ1/LXX38JITp06FC286ZNm44ePVpqi5KlS5cuXbr0woULXbp0kVpu3769ZcuWGzduVH7rFi1aGBoaxsXFVT0tAAAAUBWs7wYAAADqVlxcXEJCwp07dzQaTXJyckREhI2Njb5DCSMjo0GDBp08ebK4uFhqOXDggEwmk/YZL2Xz5s0lH4B5584dIURQUJBWq9UVu4UQoaGho0ePtrW1LXltVlZWqUdcxsfHFxUVOTs71/6sAAAA8GKj3g0AAADUrcmTJ7u4uOTm5uo7SGkLFizIyMj45JNP8vLyzp07t2LFirFjx7Zu3Vo6u3DhQqVSefjw4SqOlpGR8c0330yfPr1Uu7m5+eHDh2NiYnJycjQazaVLl8aMGWNubj5jxozanAwAAABAvRsAAACoayEhIVevXrWxsZF20zYxMWnbtu2uXbuEEGvWrHnllVeEEHPnzn3rrbciIiLCwsKEEB06dLh58+aGDRtmzpwphBgwYEB8fHypzjUP1q5du0OHDh0+fLhBgwY+Pj7vv//+unXrSnZ4og1Sli9fPnToUBcXl1Ltcrm8R48e48aNc3R0tLS09PPza9asWWxsbPv27Ws+BQAAAKAk2RP9FxYAAAConJ+fnxBix44d+g7yDImIiIiPj5cK2UKIgoKCefPmRURE3L9/X6FQ6Dcbninbt2/39/fnMxoAAEC18bxKAAAAoA7dvn17ypQpv/32m67FxMTExcVFo9FoNBrq3QAAAEAtYj8TAAAAoA4pFApjY+NNmzZlZGRoNJq0tLSNGzcuXLgwICDAyspK3+kAAACAeoV6NwAAAFCHpEc+Xr16tVWrVgqFom3btps3b162bNl3332n72gAAABAfcN+JgAAAEDd6tWr15EjR/SdAgAAAKj/WN8NAAAAAAAAAKgPqHcDAAAAAAAAAOoD6t0AAADA823lypWNGjWSyWTr169/+ncPDQ11d3dXKBTm5ubu7u4LFizIycnRndVoNAsXLnRzczMxMXF0dJw1a5ZarS55+ZYtW1555RVLS8umTZu+9957t2/ffuozAAAAQP1BvRsAAAB4vs2aNevs2bP6uvupU6fGjx+fnJyckZGxZMmS0NBQX19f3dlp06atWLEiJCQkKyvrhx9+2LBhw7hx43RnIyMjR40a5efnl5qaumfPnpMnTw4cOLCwsFAf8wAAAEB9QL0bAAAAQPWZmJhMmjTJzs7OwsLCz89v2LBhR44cSU9PF0LcvHlz/fr17777bkBAgKWlpZeX15QpU7Zs2fLHH39I1/7v//5vkyZNZs+erVQqX3755RkzZvz222/nz5/X64QAAADwHKPeDQAAAKD6du3aJZfLdYeOjo5CiNzcXCHEhQsXiouLX331Vd3ZAQMGCCEOHTokHaakpDg4OMhkMunQ2dlZCJGUlPS0sgMAAKC+od4NAAAA1JoTJ0507drVzMzMysrKw8ND2sn61KlTbdu2VSqVcrncw8NDqvaGh4ebm5sbGBh07ty5cePGxsbG5ubmnTp16tWrl7Ozs1wut7a2njNnjjTsV199JZfLGzVqNHHiRAcHB7lc3r1790rWQRcVFS1cuNDFxUWhUHTo0CEyMrKSeCXl5ubKKuDp6VmVVyA+Pt7a2rpp06ZCCAMDAyGEQqHQnW3ZsqUQQre+283NLTMzU3dW2rzbzc2tKjcCAAAAyqLeDQAAANSOvLy8oUOH+vr63rt3Lz4+vlWrVgUFBUKIjIwMf3//xMTEtLQ0CwuLUaNGCSGmTZs2e/ZsrVa7bt26v//++/bt26+99tqlS5c+/vjjS5cu3bt3b8yYMStWrLh8+bIQYsqUKWPHjlWpVFOnTk1MTLx48WJhYWHfvn1TUlLKTTJv3rzly5eHhYWlp6cPGTJk5MiRv/76a0XxSrKwsNBWIDY2tpK5azSaW7durVmzJjo6evXq1SYmJkIId3d3UaK6LYRo0KCBEOLOnTvS4fz582/fvr169eqHDx/GxcWFh4f379+/ioV1AAAAoCzq3QAAAEDtSExMzMnJadeunVwub9y4cVRUVMOGDYUQvr6+n3zyiY2Nja2t7dChQ7OysnQFXyFE27ZtzczMGjRo8M477wghXFxcGjZsaGZmNnr0aCHE9evXdT2NjIzatGljamratm3biIiIhw8fbt68uWyM/Pz8iIiI4cOH+/j4WFtbBwcHGxsbb968uaJ4tcLZ2dnJyWnRokXLly/39/eXGj08PAYMGLB27dqYmJj8/Pzbt2/v2rVLJpNpNBqpw+uvvz537twpU6ZYWVm1b9/+4cOHGzdurK1IAAAAeAFR7wYAAABqh5ubW6NGjUaPHr1o0aLExMRy+xgbGwshioqKyp6S1kQXFhaW7KkrDZfSpUsXMzOzktVwnRs3bqhUqvbt20uHCoXC3t7++vXrVYlXbSkpKZmZmVu2bPnuu+86duyo26Vk27Ztfn5+7777rq2tbY8ePX788UetViut8hZCBAUFff3110ePHs3Nzb1582b37t27detW0aJ1AAAA4LGodwMAAAC1Q6FQxMTE9OzZc+nSpW5ubgEBAWq1Wgixf/9+Ly8vOzs7U1NT3ZbcNWdqalpynbhOXl6eECI4OFi39XZSUpJKpaooXknV3r/b2NjYzs6uX79+27Zti4uLCwkJkdqVSuX69etTU1NVKlVCQsIXX3whhGjSpIkQIj09PTQ09MMPP+zdu7e5ubmrq+uGDRvS0tJWrFhRK68PAAAAXkDUuwEAAIBa065du71796alpc2dOzcyMnLlypXJycnDhw+3t7c/f/58dnZ2aGhordxIo9E8ePDAycmp7Ck7OzshRFhYWMndt8+dO1duvFLXVnv/bp0WLVoYGhrGxcWVe/bChQtCCG9vbyFEfHx8UVGRVPuWWFlZ2draVnQtAAAA8FjUuwEAAIDakZaWdu3aNSGEnZ3d559/3qlTp2vXrl25ckWj0QQGBrq5ucnlcplMViv3On78uFarLXfNtbOzs1wu/+2336oSr4YxsrKyRo4cWbJFqmI7OzuX23/Dhg2urq6vv/66EEIq1qenp+vOPnz48N69exVdCwAAADwW9W4AAACgdqSlpU2cOPH69esFBQWXLl1KSkry9PR0cXERQkRHR+fn58fHx58/f77a4xcXF9+/f7+wsPD333+fNm2ai4vL2LFjy3aTy+Xvvffe1q1bIyIicnJyioqKUlNT09PTy41X7TASc3Pzw4cPx8TE5OTkaDSaS5cujRkzxtzcfMaMGVKHrl27JiUlFRYWJiYmzpo1Kzo6etOmTdJO5a6urt7e3hs2bDh58qRarU5JSZkwYYIQ4oMPPqhhKgAAALywqHcDAAAAtcPOzq6oqKh79+5mZmZvvvnmxIkTJ0+e7OHhMXfu3LVr1zo4OAQFBXl5eQkhevbsOWvWLGmjag8Pj9OnT4eGhk6cOFEIMWDAgC1btkRGRg4YMEAIMWXKlG3btknj5+fne3h4KBSKXr16tWrV6tixY6ampkKIVatW9ezZUwgxa9YsHx8fIUR4ePj06dNDQ0MbNGjg4OAwbdq0+/fvlxuvhlOWy+U9evQYN26co6OjpaWln59fs2bNYmNjdU/LtLa2fvnllxUKRadOna5fv37q1ClpMxMhhEwm27FjR0BAwAcffGBjY9O2bdvk5OSoqKhevXrVMBUAAABeWDKtVqvvDAAAAKg//Pz8hBA7duzQd5D6ZuLEiTt27MjKytJ3ENSh7du3+/v78xkNAACg2ljfDQAAADwfioqK9B0BAAAAeKZR7wYAAAAAAAAA1AfUuwEAAIBn3fz58zdv3pydne3q6rpz5059xwEAAACeUUb6DgAAAADgMUJCQkJCQvSdAgAAAHjWsb4bAAAAAAAAAFAfUO8GAAAAAAAAANQH1LsBAACAWrNy5cpGjRrJZLL169frMcbEiRNl/zF69GghxKefftq2bVsrKytTU9MWLVrMmTMnNze33Gvz8/Pd3d2Dg4N1LZ999pnsv7Vv3/6J8hQXF4eFhXXv3r1Ue+Ujh4aGuru7KxQKc3Nzd3f3BQsW5OTkSKd2796tu6Rhw4ZPFAYAAAD1GPVuAAAAoNbMmjXr7Nmz+k4hhBC2trYHDhy4cePGpk2bhBAxMTGTJ09OTEy8e/duSEhIeHi4n59fuRcGBQXduHGjFpPEx8e/9tprM2bMUKlUT3ThqVOnxo8fn5ycnJGRsWTJktDQUF9fX+nUW2+9lZqaevLkyUGDBtViVAAAADzvqHcDAAAA9ZBCoRgwYECrVq1MTU2FEBYWFhMmTE+j/p4AACAASURBVLC1tbW0tBwxYsTw4cMPHjyYkpJS6qqzZ89evXq17Gj/+te/tCWU26dcly9fnjdv3kcfffTyyy+X26GSkU1MTCZNmmRnZ2dhYeHn5zds2LAjR46kp6cLIWQymaOjY69evVq2bFnFJAAAAHgRUO8GAAAA6r99+/YZGhrqDqU9QEotuFar1bNnzw4PD6/F+7700ktRUVGjRo2Syu5PZNeuXXK5XHfo6OgohKhoGxYAAABAUO8GAAAAytWmTRuZTGZgYNC5c2epLjxnzhylUimXy7/99lshxKlTp9q2bSu1eHh4HDp0qOwgU6ZMMTExsbe3lw4nTZpkbm4uk8nu3r0rtRQVFS1cuNDFxUWhUHTo0CEyMrLsILm5ubIKeHp6Vm92t27dUigUrq6uJRuDgoKk9dTVG7OuxcfHW1tbN23aVN9BAAAA8Oyi3g0AAACU4+rVq82aNXN2dv7ll1/MzMyEEMuXL//ggw+WLVs2duxYIURGRoa/v39iYmJaWpqFhcWoUaPKDvLVV1+NGDFCd7h27drFixeX7DBv3rzly5eHhYWlp6cPGTJk5MiRv/76a6lBLCwstBWIjY2txtRUKlVMTMz48eNNTEx0jWfOnElISBg5cmS5l3z88cc2NjYmJiaurq7Dhg27cOFCNe5bvZE1Gs2tW7fWrFkTHR29evXqkpkBAACAUqh3AwAAAOUwNDScOnVqcnLyrl27pBaVShUVFfX+++9Lh76+vp988omNjY2tre3QoUOzsrLu3LnzRLfIz8+PiIgYPny4j4+PtbV1cHCwsbHx5s2ba3kmZYSEhDg4OHz22We6FrVaPW3atIiIiHL7jxkz5qeffkpJScnNzd26dWtycvLrr78eFxdX8yRVGdnZ2dnJyWnRokXLly/39/ev+U0BAABQj1HvBgAAAMo3btw4pVKp28/6+++/HzZsmJWVVdmexsbGQoiioqInGv/GjRsqlap9+/bSoUKhsLe3v379es1SP8auXbu2b99+6NAhS0tLXeP8+fM//PBDaYPsspydnTt27GhhYWFiYuLp6bl582a1Wr127dqah6nKyCkpKZmZmVu2bPnuu+86duyYmZlZ8/sCAACgvqLeDQAAAJTPwsLiww8/PHv27C+//CKEWLdu3ZQpU3Rn9+/f7+XlZWdnZ2pqOmfOnGqMn5eXJ4QIDg7W7cedlJRU6hmSolb37962bduyZcuOHz/erFkzXePp06evXLkybty4Kg7i4eFhaGj4559/PtGtqz2ysbGxnZ1dv379tm3bFhcXFxISUuv3BQAAQL1BvRsAAACo0JQpU4yNjcPCwk6ePOns7Ny8eXOpPTk5efjw4fb29ufPn8/Ozg4NDa3G4NKTIcPCwkpuyX3u3LlS3Wpr/+7Vq1d///33MTExTZo0Kdm+adOmo0ePGhgYSDV0KdXSpUtlMlnZzcSFEMXFxcXFxaampk884cepfOQWLVoYGhrWyj4qAAAAqK+odwMAAAAVcnJyGjFixM6dOxcsWDBt2jRd+5UrVzQaTWBgoJubm1wul8lkFY1gZGSk0WjKPeXs7CyXy3/77bfaz/3ftFrt3Llzr1y5snv3bgsLi1JnN2/eXLKGLu1CHhQUpNVqu3TpIoTo379/yf4XLlzQarXdunWrebBKRs7Kyir18Mz4+PiioiJnZ+ea3xcAAAD1FfVuAAAAoDIzZ84sLCy8f/9+7969dY0uLi5CiOjo6Pz8/Pj4+PPnz1d0eYsWLe7du7d7926NRnPnzp2kpCTdKblc/t57723dujUiIiInJ6eoqCg1NTU9Pb3Wp3Dt2rXly5dv2LDB2Ni45HYoK1eurMrlt27d2rZt24MHDzQazblz58aNG+fi4vLRRx9JZwMCAho3bnzx4sVqBKtkZHNz88OHD8fExOTk5Gg0mkuXLo0ZM8bc3HzGjBnVuBEAAABeENS7AQAAgMp07NjR29t76tSpJRs9PDzmzp27du1aBweHoKAgLy8vIUTPnj2nT5/es2dPIcSsWbN8fHyEEIGBgd7e3u+8807r1q2XLFmiUCiEEN26dUtJSRFChIeHT58+PTQ0tEGDBg4ODtOmTbt//36tT0Gr1dbk8gEDBgQHBzs5OZmZmY0YMaJHjx6xsbENGjSQzhYUFGRmZu7Zs6fca2NjY3v27NmkSZPz589fvnzZwcGhR48eJ0+efOzIcrm8R48e48aNc3R0tLS09PPza9asWWxsrO7xngAAAEBZshr+3xcAAAAoyc/PTwixY8cOfQd5oU2cOHHfvn2pqalP4V7FxcVeXl5jx459//33n8LtSpk2bdr3339/9+7dp3/rurB9+3Z/f38+owEAAFQb67sBAACAekitVh86dCg+Pr6goKDu7lJUVLR79+6HDx8GBATU3V3K0mq1aWlpp0+f/uuvv57mfQEAAPCMM9J3AAAAAAC17969ewMGDBBCjBo16vvvv6+juxw/fjwqKurAgQNmZmZ1dIty7dmzZ/jw4dLXup1VAAAAAPYzAQAAQG1iPxOg2tjPBAAAoIbYzwQAAAAAAAAAUB9Q7wYAAAAAAAAA1AfUuwEAAAAAAAAA9QH1bgAAAAAAAABAfUC9GwAAAAAAAABQHxjpOwAAAADqm9jYWD8/P32neOY8fPjQ0tJS3ymeFYWFhQUFBWZmZvoO8mxJTU3VdwQAAIDnG+u7AQAAUJu6devm6emp7xTPnMzMzOjo6OTkZH0HeVZcuXLlxIkTKpVK30GeLU5OTr6+vvpOAQAA8ByTabVafWcAAAAA6rMzZ84MGDBgwIABW7duNTLiLyyFEOLBgwdvvPHGnTt3jh8/7urqqu84AAAAqCeodwMAAAB16Ny5c/379+/bt++2bduMjY31HecZcv/+/b59+969e/f48ePNmjXTdxwAAADUB9S7AQAAgLpy8eLFN954w9PT88cffzQ1NdV3nGfOnTt3+vTpo1arjx8/7ujoqO84AAAAeO5R7wYAAADqxKVLl954442uXbvu3r2bYndF7ty507t370ePHh07doySNwAAAGqIejcAAABQ+y5fvtynT5/OnTvv2bNHLpfrO84zLTMzs3fv3gUFBcePH2/SpIm+4wAAAOA5ZqDvAAAAAECVBAQEyCq1b98+fWf8P7///vsbb7zRsWPH3bt3U+x+rEaNGsXExJiYmHh7e6elpVV7nKioKDc3t3LfG5XvD75y5cpGjRrJZLL169dX++4AAAB4FlDvBgAAwHPj8OHDDx480Gg06enpQoihQ4cWFBTk5eVlZmaOHz9e3+n+z40bN/r37+/u7v7jjz8qFAp9x3k+NGrU6OjRo0ZGRr1795a+udXg4+Nz8+bN5s2bK5VKrVar1WoLCwtVKlVGRoaZmVklF86aNevs2bPVuykAAACeKdS7AQAA8HyQyWQ9evRQKpVGRka6FmNjYzMzMzs7u86dO+s3nuTPP//09vZ2c3M7cOCAhYWFvuM8Txo3bhwTE2NgYODt7V3tkncphoaGCoWiUaNGrVq1qpUBAQAA8Iyj3g0AAIDnw9atWytZpTthwoQ333zzaeYpKz4+3tvbu1mzZgcPHqTYXQ1SyVsmk/Xu3fv27du1OPLu3btrcTQAAAA8s6h3AwAAoJ5Yvny5mZmZpaVlZmbmzJkzHR0d+/fvb2JiYm9vL3WYNGmSubm5TCa7e/eu1FJUVLRw4UIXFxeFQtGhQ4fIyMhq3z0pKalv37729vb79++3tLSshfm8kOzt7Y8cOaLRaPr376/7NtWiU6dOtW3bVqlUyuVyDw+PQ4cOldvtxIkTXbt2NTMzs7Ky8vDwyMnJEbX6bgEAAEAdod4NAACAemLOnDkzZszIzc0NCQlxdXX19PT88ssvR4wYoeuwdu3axYsXl7xk3rx5y5cvDwsLS09PHzJkyMiRI3/99ddq3Do5Odnb29vGxubIkSM2NjY1ncmLzcnJ6dixY3l5eX369Kl5yTsmJmblypW6w4yMDH9//8TExLS0NAsLi1GjRpW9JC8vb+jQob6+vvfu3YuPj2/VqlVBQYGovXcLAAAA6g71bgAAANQ3y5Ytmzx5clRUlLu7eyXd8vPzIyIihg8f7uPjY21tHRwcbGxsvHnz5ie9XUpKire3t5WVVXR0tK2tbQ2C4/84OzsfO3YsNzf3jTfeyMrKetLLs7OzZf/Rp0+fkqd8fX0/+eQTGxsbW1vboUOHZmVl3blzp9TliYmJOTk57dq1k8vljRs3joqKatiwYW29WwAAAFCnqHcDAADgBXXjxg2VStW+fXvpUKFQ2NvbX79+/YkGSU1N9fb2Njc3j46ObtCgQR3EfEFJJe+cnJxqlLyVSqX2P44dO1ZRN2NjYyFEUVFRqXY3N7dGjRqNHj160aJFiYmJUmOtvFsAAABQ16h3AwAA4AWVl5cnhAgODtatBU5KSlKpVFUfISMjo1+/fkZGRocPH27YsGGdJX1Bubi4HD9+PDs7u2/fvvfu3aveIF5eXrNmzdId7t+/38vLy87OztTUdM6cOeVeolAoYmJievbsuXTpUjc3t4CAALVaXfN3CwAAAJ4C6t0AAAB4QdnZ2QkhwsLCtCWcO3euipdnZmb27t27sLAwJiZG90hM1C4XF5djx47dv3//jTfeqHbJWyc5OXn48OH29vbnz5/Pzs4ODQ2tqGe7du327t2blpY2d+7cyMjIlStX1vDdAgAAgKeDejcAAADqMyMjI41GU+4pZ2dnuVz+22+/VWPYO3fu9OnTp6Cg4NixY02aNKlZRlSmadOmx48fv3fvXt++fe/fv1+Toa5cuaLRaAIDA93c3ORyuUwmK7dbWlratWvXhBB2dnaff/55p06drl27VpN3CwAAAJ4a6t0AAACoz1q0aHHv3r3du3drNJo7d+4kJSXpTsnl8vfee2/r1q0RERE5OTlFRUWpqanp6emPHfPBgwcDBgx4+PDhkSNHHB0d6zI+hBCiadOmR44cycjIGDRoUE5OTrXHcXFxEUJER0fn5+fHx8efP3++3G5paWkTJ068fv16QUHBpUuXkpKSPD09q/1uAQAAwNNEvRsAAADPk4cPH77++uvt2rUTQuzdu7dly5YhISHSqeXLl69atUoI0apVq++//15qDAwM9Pb2fuedd1q3br1kyRKFQiGE6NatW0pKihAiPDx8+vTpoaGhDRo0cHBwmDZt2mNXED948KBv37537949duxYs2bN6myi+C8tW7Y8duxYcnKy9JuGcvucPXu2devWCQkJ2dnZDg4Ob7zxRqkOHh4ec+fOXbt2rYODQ1BQkJeXlxCiZ8+e06dP79mzpxBi1qxZPj4+dnZ2RUVF3bt3NzMze/PNNydOnDh58mRRrXcLAAAAnjKZVqvVdwYAAADg+ZCdnd2vX7+0tLQTJ064ubnpO84L588///T29m7WrNnBgwctLS31HQcAAADPHNZ3AwAAAFWSk5PTv3//pKSkI0eOUOzWi1atWsXExPz9998DBw6saJU3AAAAXmTUuwEAAIDHU6lUQ4cO/fvvv2NiYtzd3fUd58XVunXrY8eO3bx5c9CgQbm5ufqOAwAAgGcL9W4AAADgMVQq1Ztvvnnt2rWYmJi2bdvqO86LrnXr1jExMX/99dfAgQMpeQMAAKAk6t0AAABAZdRq9ZAhQ65cuXL06FHpOZnQO3d392PHjsXHxw8aNCgvL0/fcQAAAPCsoN4NAAAAVKigoMDPz+/ixYsHDx708PDQdxz8f+7u7ocPH/7jjz+GDRumVqv1HQcAAADPBOrdAAAAQPkKCgp8fHxOnz59+PDhzp076zsOSuvQoUN0dPSlS5feeustSt4AAAAQ1LsBAACAcmk0Gj8/v5MnTx4+fPiVV17RdxyU76WXXoqOjr548eKwYcPy8/P1HQcAAAB6Rr0bAAAAKK2oqOh//ud/jh49unfv3q5du+o7Dirz8ssvR0dH//rrr5S8AQAAQL0bAAAA+C9SsXvv3r379u177bXX9B0HjyeVvC9cuDB8+HBK3gAAAC8y6t0AAADA/1dUVDRmzJg9e/bs27fPy8tL33FQVR07djxy5Mj58+fffvvtR48e6TsOAAAA9IN6NwAAAPB/iouL33vvvV27dv3000/e3t76joMn06lTpyNHjsTGxg4fPpySNwAAwIuJejcAAAAghBBarTYwMDAyMnLHjh19+vTRdxxUR+fOnffv33/mzJl33nlHo9HoOw4AAACeNurdAAAAgNBqtZMmTdq8efPOnTsHDx6s7ziovm7duh04cCA6OjogIICSNwAAwIuGejcAAABeLMXFxaVatFrtP/7xj40bN27fvn3IkCF6SYVa1L1794MHDx45cuSdd94pLCws20Gr1T79VAAAAHgKqHcDAADgBfL777+/9dZbpTZ3njdv3vr16//5z3++9dZb+gqG2tW9e/cDBw4cOnSobMl70aJFUVFR+goGAACAOkW9GwAAAC+QkJCQffv2vf3227qS98cff/zFF1/885//DAgI0G821K4ePXocOHDg4MGDI0eO1JW8FyxYsHjx4k8//VS/2QAAAFBHZPwpHwAAAF4QN2/ebNmyZXFxsZGRUe/evffs2fPZZ58tW7bs22+/HT16tL7ToU6cOnVq0KBBgwYN+uGHHxYuXLhs2TLpE9CBAwcGDBig73QAAACoZdS7AQAA8KL46KOPNm3aJD3D0MjIyMnJKTk5+ZtvvhkzZoy+o6EORUdHDx06tHXr1pcvX5Y+/hgaGr766qtnzpzRdzQAAADUMurdAAAAeCFkZGS4uLgUFBToWgwNDV1dXS9dumRhYaHHYHgK3nnnncjIyFKffU6dOtWzZ099RQIAAEBdYP9uAAAAvBDCw8NLlTuLioqSkpL69u378OFDfaXCUzB//vyyxW4jI6PPPvtMX5EAAABQR1jfDQAAgPovJyfH0dExNze37CljY+NOnTodPnzYysrq6QdDndJqtTNmzPjyyy8r+tTz73//u1OnTk85FQAAAOoO67sBAABQ/61bty4/P7/cUxqN5vz588OHDy8uLn7KqVDX5s+fX3Zdv46xsfGyZcueciQAAADUKdZ3AwAAoJ7Lz893cnLKysoqe8rIyMjQ0HDy5MmzZ89u3Ljx08+GOnXv3r0vv/xy1apVjx49kp5TWoqBgcH169dbtmz59LMBAACgLrC+GwAAAPXct99+e//+/VKNxsbGcrk8MDDw77//XrlyJcXuesnW1nbx4sVpaWkrVqxo0KCBkZFRqQ6GhoahoaF6yQYAAIC6wPpuAAAA1GdFRUUtWrRISkrS/b/X2NjY0NDwww8//Pjjj+3t7fUbD09NXl7exo0bly5dmpWVVXLvGkNDw5s3b7q4uOgxGwAAAGoL67sBAABQn+3YsUNX7DYyMlIoFDNmzLh169aXX35JsfuFYm5uPnXq1JSUlHXr1jVu3NjAwEAmkwkhDAwMwsPD9Z0OAAAAtYP13QAAAKi3tFqth4dHXFycoaGhQqGYNWvW1KlTra2t9Z0Levbo0aNvvvlm6dKl6enpxcXFcrk8JSWlYcOG+s4FAACAmqLeDQAA8ATOnTu3atUqfadAVd2+ffv06dPGxsatWrVq0aKFsbGxvhOVr1u3bjNmzNB3itrk5+en7wiPV1xcnJycfO3aNZVK1aZNm3bt2uk7Eappx44d+o4AAACeFaUf2AIAAIBKpKSk7Ny509fXV99BUCV///13+/btW7RoUfZBhc+O2NhYfUeofTt37vT09HRyctJ3kMoYGBg0a9asadOmKSkpCQkJrVu3fpbfJyhXampqvfwJAgAA1cb/5wAAAJ4YawmfC0VFRWq12sLCQt9BHuO5WApdDdOnTx8xYoS+U1RVcXGxSqV69t8tKGX79u3+/v76TgEAAJ4hPK8SAAAA9ZOhoSHlS1SRgYEB7xYAAIB6gHo3AAAAAAAAAKA+oN4NAAAAAAAAAKgPqHcDAAAAAAAAAOoD6t0AAAAAAAAAgPqAejcAAEAti4qKcnNzk8lkMplswYIF5fZZtWqVTCYzMDBwd3c/efLkU05Y0pYtW2QyWffu3fWYoYY0Gk1ISEiLFi1MTEysra3bt2+fmJioO7Vw4UI3NzcTExNHR8dZs2ap1eoqXlvKp59+2rZtWysrK1NT0xYtWsyZMyc3N1d39rPPPpP9t/bt29fVhOuFTp06SS+Ug4PD1KlTK+r2559/vvLKKxYWFgYGBgMGDHiaCcV//yxL5HK5q6vr+++///fffz/paOPGjbO0tJTJZL/99pvU8vPPPyuVyr1799Z28FpWXFwcFhZW+b8S+fn57u7uwcHBlfTZsmXLK6+8Ymlp2bRp0/fee+/27du6U15eXrIyeIIoAACoBurdAAAAtczHx+fmzZvNmzcXQmzcuFGj0ZTqUFRU9NVXXwkhevfuff369ddee00PKf9jy5YtzZs3P3fu3F9//aXHGDXh7+//z3/+84cfflCpVH/88Ufz5s11lehp06atWLEiJCQkKyvrhx9+2LBhw7hx46p4bSkxMTGTJ09OTEy8e/duSEhIeHi4n59fnc+t/rp48WK/fv1kMtmlS5e+/PJLXXtRUVHv3r11h61atbpw4UJISMjIkSMPHjz4lEPqfpaVSqVWqy0qKkpOTv70008jIyM9PT2zsrKeaLSNGzdu2LChZItWq63VvHUiPj7+tddemzFjhkqlqqRbUFDQjRs3KukQGRk5atQoPz+/1NTUPXv2nDx5cuDAgYWFhZVc0rNnz2qGBgAALzDq3QAAAHWlc+fOt2/f3r17d6n2qKgoR0dHvUQqJSsr69q1a4sXLxZC/POf/9RLBo1Gs3nz5jFjxlTv8m3btu3evXvHjh2vvvqqkZGRg4PDnj17pLXVN2/eXL9+/bvvvhsQEGBpaenl5TVlypQtW7b88ccfj722LAsLiwkTJtja2lpaWo4YMWL48OEHDx5MSUnRdfjXv/6lLeHq1avVm9GLIyAgQKvV7tu3r2TjsWPHjh07du3atZKNP//8s7+/f23dV61WV+8PGgwMDBo1avQ///M/kydPzszMjI6OrmGSwYMHZ2dnDxkypIbjVKKGP1+XL1+eN2/eRx999PLLL1fS7ezZs499w//v//5vkyZNZs+erVQqX3755RkzZvz222/nz5+Xzsrl8pycnJI/QRMmTJgzZ071YgMAgBcZ9W4AAIC6EhgYKIRYt25dqfZVq1bNnDlTH4lK2759++DBg4cOHSqXy6Vy7dO8+6NHj9atW9epU6e4uLhly5ZVbxBpBA8Pj7KnLly4UFxc/Oqrr+papA0xDh069Nhry9q3b5+hoaHusGHDhkKIyle8onJvv/22iYnJTz/9VLLx8OHDTZo02blzp65FrVZfvny5f//+tXXfTZs2ZWZm1mSEFi1aCCFKbsdRRTKZrCb3fSK18vP10ksvRUVFjRo1ytTUtKI+arV69uzZ4eHhlQ+VkpLi4OCgewWcnZ2FEElJSdLhwYMHLS0tS3a+evVqyZX+AAAAVUS9GwAAoK707t27TZs2x44dK/ln/mfOnFGpVP369SvVuaioaOHChS4uLgqFokOHDpGRkVL7qVOn2rZtq1Qq5XK5h4eHVKuNiIgwNzc3MzPbs2fPwIEDraysnJyctm7d+qQJt2zZ8vbbb1taWvbr1y8xMfHUqVMlzy5fvtzMzMzS0jIzM3PmzJmOjo43btx4opwVycvLW7VqVZcuXTIzM0+cOLFy5UoHBwchRG5ubtk9fCWenp5lxykoKIiNja1o5amBgYEQQqFQ6FpatmwphJDWd1d+7WPdunVLoVC4urpW73IIIZRKZf/+/aOjo3Wbqkub//j6+pasdx89erR///4mJiZCCK1Wu2rVqjZt2piamtrY2AwbNuz69etSt3LfridOnOjatauZmZmVlZWHh0dOTs60adNmzpyZkJAgk8mksnU1xMfHCyFeeuklXUtF73+tVrtixYrWrVubmpoqlcrZs2frLjl9+rSLi4tMJluzZo0QYsqUKSYmJvb29tLZSZMmmZuby2Syu3fvSi1l51JRvNr6+aqioKCgSZMm2dnZVd7Nzc2t5K8ZpN8WuLm5ldt52bJllezqDgAAUAnq3QAAAHVo4sSJQoj169frWr744osZM2aU7Tlv3rzly5eHhYWlp6cPGTJk5MiRv/76qxAiIyPD398/MTExLS3NwsJi1KhRQojAwMDp06er1WpLS8vIyMiEhAQ3N7fx48eX3Su8EsnJyTdu3JB2D5e2oi61pcmcOXNmzJiRm5sbEhLi6urq6emp1WqfKGdZ2dnZS5cu7d69u1arjY2N/eSTT2xtbXVnLSwstBWIjY0tO1paWlpBQcG///1vb29vBwcHuVzepk2btWvXSgvV3d3dxX+q25IGDRoIIe7cufPYayunUqliYmLGjx8vFWElH3/8sY2NjYmJiaur67Bhwy5cuPDYcRAQEKBWq3Ubgxw5cuSNN97w8/O7cuXKn3/+KTWW3Mxk0aJFH3/8cVBQUGZm5smTJ1NSUnr16pWRkSHKe7vm5uYOHTrU19f33r178fHxrVq1KigoCA8PHzJkSPPmzbVabTX2rH/w4MF33323du3awYMHe3l56dorev8vWLBg7ty5EyZMyMjIuH379rx583SX9OzZ8+zZs7rDr776asSIEbrDtWvXShsNSfLy8srOpWy82v35qoozZ84kJCSMHDnysT3nz59/+/bt1atXP3z4MC4uLjw8vH///uXW2W/dunX8+HEfH5/qRQIAAC+6iv7HAwAAgLKk5cxV6dm8efO///77wYMH5ubmNjY2KpVKq9UmJCQ4OTk9evTo4cOHQog+ffpIndVqtZmZmbSdsVarValUpqamgYGBpcYMCQkRQmRmZmq12qCgICGEWq2WTq1du1YI8ddff1V9Lp9//vl7770nfZ2dnW1qamplZSXl1Cl1l2rkLKmgoKBpcgOcmgAAIABJREFU06avvfZaqY16q+3KlStCiL59+545cyYrK+vBgwdSPfH777+XOgwYMMDW1vbo0aNqtTo9PX379u0ymezNN9+syrWVCAoKatWqVclZJCcnX7x48eHDh48ePTp37lzHjh0VCsXVq1erMgtfX19fX99qvQDPLiFEZGTkY7s9fPhQoVCMGzdOOpw5c6ZGoykuLm7SpMlnn30mNXbq1KmgoECr1apUKgsLC93bT6vV/vLLL0KITz/9VDos9XaVdpTet29fqZv6+PhI9e4qkp49qyOTyT777DMpUrl073+VSmVmZta3b1/dKemPMC5duiQdSvu/r169WjocNWpU48aNdZ1XrFghhLhz504lcymp1n++Snr11VdfeumlUo0qlapLly6pqalarVb6NVJQUFAlgwQHB+teRicnp5SUlHK7TZ48ed26dVUMVvV/kwEAwAuC9d0AAAB1SKlUjhw58v79+9u2bRNChIWFBQYGllwULLlx44ZKpdI9KVGhUNjb2+s2atAxNjYWQhQVFZW9kTTmE63vljYzkb62srLq169fTk7Onj17KrmkhjmNjY0vX77ct2/fXr16hYWF5eXlVT1tuaQ9hdu1a9e9e3dbW1ulUrl48WKlUvn1119LHbZt2+bn5/fuu+/a2tr26NHjxx9/1Gq10irvx15bkV27dm3fvv3QoUMltxt2dnbu2LGjhYWFiYmJp6fn5s2b1Wq19EsIVMLCwmLw4MFSGffRo0dGRkZGRkYymczHx0fa0uTatWsdO3aU3lFxcXG5ubldunTRXf7KK6+YmJjonnlYipubW6NGjUaPHr1o0aLExMSa5FQqldLHp9mzZ2u1WqVSKUUql+79/9dff6lUqj59+tTk1pKqzKXWf74ea/78+R9++GEVn74bFBT09ddfHz16NDc39+bNm927d+/WrVvJJ75K0tLSfvrpp7Fjx9Z+XAAA8GKg3g0AAFC3pKdWrl+//sGDBzt27JB2OClFqkwFBwfrttNNSkqSnoW4f/9+Ly8vOzs7U1PTOXPm1Faqq1evXrlyZciQIbo77t27V5TZ0qTWcyqVyuDg4DNnzhQXF3t6en766af379/XnX3S/YWlXYl1GxwLIUxMTJo2bZqQkKC73fr161NTU1Wq/8fefQc0dfX/Az8XCCQkTEEJskELKO6JaN2r1pbKqrauIjhxixW11kFFfQTFUVdtrVKGUmzVFkVUpIKjoiIURCxDmYKIEJAQ8vvjPt/88jBCgEBCeL/+Muec3Pu5l/PJNR8u5/IyMzP/85//EEKMjY2leW+jQkNDd+/effPmTQsLCwnDHBwcVFVVRStygAQeHh4FBQX379+/cuUK/UBRQoiLi8ujR48yMzPFFzMpKysjhHA4HPG36+rq0n8t0RCLxYqNjXVyctq1a5eVlRW9dkobo92yZYuRkdGmTZvqFWobnf8vX74khDS7sLU0pDwW2eaXZPHx8cnJyZ6entIMzs/PDwgI8PLyGj9+PJvNtrS0PHHiRF5eHn0Pu7iAgIBFixYxmcyWxgMAAABAQ70bAAAAoH0NGDBgxIgR9+7d8/b2dnV11dPTaziGrogFBgaK/yFeQkJCTk6Os7OzkZHR3bt33759GxAQIKuozp079/nnn4vvrrS0lMViXb16lX6OXKNkFSebzV67du39+/cNDAzGjBmzfv36/Px80vL1hTkcTq9evVJTU8Uba2trdXR0Gt0vvab2uHHjWvFeQkhwcPDZs2djY2PpirkEdXV1dXV19C3kINlHH32kpaX122+/xcXF0avJE0KcnJy4XO758+cfPHhA/7wIIbq6uoSQetXtsrIyExOTpjbep0+f33//PS8vz9fXNywsbN++fW2MVktLa/fu3e/evaN/j0Vrav7TRdv379+3cac06Y9FVvkl2alTp65fv66iokJXzOkPh127dlEURa/pLy4jI0MgEIgnjra2tr6+fkpKiviwgoKCkJAQ8XMLAAAA0FKodwMAAAC0O7p8c/78+dWrVzc6wNTUlMlkPnr0qF57cnIyn89funSplZUVk8mkKEom8QiFwtDQ0GXLlok36unpubq6CgSCkJCQpt4o2ziZTObSpUsfPnxoZ2cn/hy/FnF3d09KSnrx4gX9ksfjZWdnOzg4NDr4xIkTlpaWH374YUvfKxQKfX19k5OTo6Ki6t1fTJsyZYr4y/v37wuFwpEjR7buoLoUJpM5c+bM8+fPs1gsFZX/fj1RUVH57LPPzpw50717dzU1Nbqxb9++HA5HvJZ69+7dmpqawYMHN7rlvLw8+vcZhoaG33333aBBg+r9eqN15s6dO3z48EuXLoWHh9MtTc3/vn37qqio3Lp1S8otq6mpNbUkUSuORSb5JcHp06fFK+bi63eLrzlDo38nQZfdae/evSstLTU1NRUfFhAQ8MUXX4g/YxMAAACgpVDvBgAAAGh3bm5uBgYGzs7OVlZWjQ5gMpkLFiz45Zdfjhw5Ul5eLhAIXr58mZ+fb2ZmRgiJiYmprq7OyMhoap3ilrpz5462tvaoUaPqtS9ZsoRIXNKkPeJkMBgLFy786aefWnUoZM2aNebm5vPnz8/JySkpKfH19a2qqhJV94YNG5adnV1bW5uVlbVu3bqYmJhTp06J1k+X/N6tW7fq6OhcvXqVEJKamrpnz54TJ04wGAzxVSBE99i+evUqNDS0rKyMz+cnJCR4enqamZnR5xOa5eHhkZ6ePmPGDPFGV1fX1NRUZ2dnUQuTyVy7dm1kZOTZs2fLy8uTk5OXLFnC5XK9vb0b3WxeXt7ixYvT0tJqamqSkpKys7PpVTv09fXz8vKysrLevXvXoiXvaRRFHTx4kKIoHx8ferWQpua/oaEhvRD5qVOnysvLnzx5Inl1eBsbm9LS0qioKD6fX1xcnJ2d3eyxNKuN+dUW4hlkaWk5bty4EydOxMXFVVVV5ebm0j+1r776SjS+sLDwhx9+aOqXggAAAADSav6RlgAAAADwf8LCwpr9H1RkZKS1tTUhxMDAYPny5XTjhg0b7ty5Q/978+bNRkZGhBAVFRV7e/vbt28LhcL379/7+vqamZmpqanRZbKUlBShUOjr66uvr6+rq+vq6nro0CFCiLW19caNGzU1NQkhvXr1yszMPH78uLa2NiHE3Nz82bNnksP76quv2Gy2mppa//79Hz58KGrfsWMHvaQ1IaRnz56HDx8OCAhgsViEEFNT059//pke1qI4c3JyWnOWWyg3N/fzzz/X09PT0NAYNmzYH3/8IeqaNGmSrq6umpqanp7eRx99RN92LeV7t2zZoqWlFR0dLRQKk5OTG/2/9N69e+nBa9eutba2pk+siYnJokWL8vLypIzfxcXFxcWlTadA8RBCwsLCpBxcU1PTv3//uro68UaBQNC/f3+BQCDeWFdXt3fv3l69ejEYDD09PWdn5/T0dLqr4XTNyspydHTU09NTVVU1Njb28/Orra0VCoUPHz40NzdnsVhOTk4FBQUSAvvrr7969+5N/6yNjY0XL14s6qIfqKirq/vdd98Jm57/79698/T07NatG4fDcXJy2rp1KyHExMTk8ePHwcHB9OeApqbmzJkzhUJhSUnJuHHjmEympaXlihUr1q9fTwixsbHJyclp6ljaW0JCwqhRo0SfDEZGRo6Ojrdu3Wo4Uvz+bpp4BgmFwtevX69atcrGxkZDQ4PD4YieHyuyZs2aL774oqURSvOZDAAAAF0KJRQKZV5DBwAAAFBW4eHh7u7u+B8UyJCrqyshJCIiQt6ByBJFUWFhYW5ubvIOBJQcPpMBAACgHqxnAgAAAAAAAAAAAADKAPVuAAAAAKWSlpZGNc3Dw0PeAQIoCiQLAAAAgPJRk3cAAAAAACBLtra2+NN+AGkgWQAAAACUD+7vBgAAAAAAAAAAAABlgHo3AAAAAAAAAAAAACgD1LsBAAAAAAAAAAAAQBmg3g0AAADQVVy5ckVHR+f333+X+WBpxMfHjxo1SlNTk8vl+vr6vn//vtWDQ0JChg4dqqWlZW5uvmDBgoKCAlHX9u3b7e3ttbW1NTQ0bGxsNmzYUFFRIatDgK6ms6SM5GkvuTcgIMDW1pbFYrHZbFtb2y1btpSXl8vqEAAAAAA6HurdAAAAAF1Fix7NJ9vn+KWkpEyePHnChAnFxcWRkZE//PDDkiVLWjc4LCxszpw5rq6uL1++vHjxYlxc3LRp02pra+ne2NjY5cuXZ2VlvX792t/fPygoyNXVVYYHAl1KZ0kZydNecu/t27cXLVqUk5NTWFi4Y8eOgIAAFxcXGR4IAAAAQAej8ERyAAAAAOmFh4e7u7vjf1At5eHhce/evczMTIqiCCH79u3bsGFDamqqra1tSwePHz8+PT395cuXdO/hw4eXL19O3wlLCJkxY8bFixdVVVXpTbm7u4eHh+fk5JiamnbYwbYUXXyMiIiQdyCyRFFUWFiYm5ubvAPprFqUMpKnveTezz77LCQkhMlk0r1ubm4RERF5eXlcLrcDDrPt8JkMAAAA9eD+bgAAAAClJRQKIyIijh8/Lt8wamtrL1++/OGHH9KVO0LItGnThELhxYsXWzE4NzeXy+WKeumaXXZ2Nv3y0qVLoroeIcTAwIAQwuPx2uXAQOl0xpQhzU17yb2RkZGiYjchpGfPnoQQrAIEAAAAnRfq3QAAAADKQyAQ+Pv7f/DBBywWy8DAwNLS0t/fn77HNj4+3szMjKKoQ4cOEUKOHDnCZrM1NTUvXrw4bdo0bW1tExOTX375hd5OvcH1VFRUUE0YMWJEw/EvXryoqKgwMzMTtVhbWxNCnjx50orBVlZWRUVFol568W4rK6tGT8irV69YLJalpaXE0wZdlxKkTEOSp73k3oyMDF1dXXNzc2l2BAAAAKCAUO8GAAAAUB4BAQFbt27du3dvaWnp1atXq6urdXV1dXV1CSFOTk537twRjVy6dOnq1aurqqq0tLTCwsIyMzOtrKwWLVrE5/MbDq6Hw+EIm5CYmNhwPF2S1tLSErUwmUwWi1VYWNiKwZs2bSooKAgODn737l1KSkpQUNCUKVMaLRryeLzY2NhFixapq6s3c+Kgq1KClKlH8rRvqpfP57969erQoUMxMTHBwcFIGQAAAOi8UO8GAAAAUB5RUVGDBw+eOXMmi8UaNGjQJ598EhcXV1NTI+Etjo6O2trahoaGHh4elZWVOTk5Mo/q/fv3hBDxFRUIIQwGo6qqqhWDP/zwQ19fXx8fH21t7b59+7579+7kyZON7tff35/L5e7cuVMmRwFKSQlSph7J076pXlNTUxMTk23btu3Zs8fd3b21gQMAAADIH+rdAAAAAMqjurpa/LltAoGAwWDUq5o1hb6jk75ZVbbo1YFra2vFG2tqalgsVisG+/n5HT9+/Pr16xUVFS9evHB0dBw5cmRubm697URGRoaHh0dHR4vfJAtQjxKkjDjJ015Cb25ublFRUUhIyE8//TRw4EDxJYMAAAAAOhfUuwEAAACUx/Tp0//++++LFy9WVVU9ePAgKipqxowZUhbvpNfSxYiNjIwIIeXl5aIWHo9XXV3N5XJbOjg/Pz8gIMDLy2v8+PFsNtvS0vLEiRN5eXl79+4V30hoaOju3btv3rxpYWEhm2MGJaUEKSMiedpL7mUwGIaGhpMnTw4NDU1JSfH392/hIQIAAAAoCjV5BwAAAAAAMrNt27a///57/vz5FRUVXC7Xzc1t165dMt8LvRix9OMtLS21tLSys7NFLc+fPyeE9OvXr6WDMzIyBAKBsbGxqFdbW1tfXz8lJUXUEhwcHB0dHRsby+FwWnBU0CUpQcrQJE976ZPCxsZGVVVVPKEAAAAAOhfUuwEAAACUR0pKSmZmZnFxsZqaAv03T01Nbfr06XFxcXV1dSoqKoSQP/74g6KomTNntnSwiYkJISQ/P180/t27d6WlpaampoQQoVC4cePGN2/eREVFKdQZAIWlBCkjedpL7i0pKVmxYkVISIiohf6VEp1QAAAAAJ0R1jMBAAAAUB7Lly83MzOrqKiQdyD1bdmypbCw8JtvvqmsrExISNi7d+/8+fM/+OADunfr1q06OjpXr15tdrClpeW4ceNOnDgRFxdXVVWVm5vr7e1NCPnqq68IIampqXv27Dlx4gSDwRBfMmLfvn1yOm5QdEqQMpKnveReNpt99erV2NjY8vJyPp+flJQ0b948Npu9Zs0aOR47AAAAQFug3g0AAACgPPz9/Z8+faqnp0eXtNTV1e3t7SMjIwkhhw4dGjp0KCHE19f3k08+OXLkSGBgICGkX79+L168OHHixNq1awkhU6dOzcjIqDe47YH16dMnOjr66tWr3bp1mzVr1sKFC48ePSo+QHy1BwmDKYqKiIjw8PD46quv9PT07O3tc3JyLly4MHr06HobAZCGEqSM5GkvuZfJZI4aNcrT07Nnz55aWlqurq4WFhaJiYl9+/Zt+yEAAAAAyAWFbwUAAAAA0gsPD3d3d1fY/0EdOXIkIyODrsoRQmpqajZu3HjkyJE3b96wWCz5xgZNcXV1JYRERETIOxBZoigqLCzMzc1N3oE0AynT2Sn4ZzIAAAB0PAVapQ4AAAAA2qKgoMDHx+fRo0eiFnV1dTMzMz6fz+fzUbwDqAcpAwAAAKB8sJ4JAAAAgJJgsVgMBuPUqVOFhYV8Pj8vL+/kyZNbt2718PDQ1taWd3QACgcpAwAAAKB8UO8GAAAAUBL08+uePn3au3dvFotlb29/+vTp3bt3//TTT/IODUARIWUAAAAAlA/WMwEAAABQHqNHj7527Zq8owDoNJAyAAAAAEoG93cDAAAAAAAAAAAAgDJAvRsAAAAAAAAAAAAAlAHq3QAAAAAgrX379nXv3p2iqO+//77j9759+3Z7e3ttbW0NDQ0bG5sNGzZUVFSIegMCAmxtbVksFpvNtrW13bJlS3l5eaPbqa6utrW13bx5c0cFDl2XfFOGVldXFxgY6Ojo2LCLz+f7+/vb2Nioq6vr6ur27ds3Kyur4TCkDAAAAHQiqHcDAAAAgLTWrVt3584dee09NjZ2+fLlWVlZr1+/9vf3DwoKcnV1FfXevn170aJFOTk5hYWFO3bsCAgIcHFxaXQ7fn5+6enpHRU1dGnyTRlCSEZGxpgxY9asWcPj8Rr2uru7nzlz5ty5czwe759//rG2thb/HZIIUgYAAAA6ETyvEgAAAAA6Bw6H4+3traqqSghxc3O7cOFCeHh4bm6uqakpIURdXX3ZsmVMJpMQ4urqGhERERERkZ+fz+VyxTdy586dp0+fyiV+gA72+PHj7du3L1mypLKyUigU1usNDQ2Niop6/Pixg4MDIYTL5V68eLHhRpAyAAAA0Lng/m4AAAAA6BwuXbpEF7tpBgYGhBDRXauRkZF0sZvWs2dPQki9m1WrqqrWr18fFBTUEeECyFv//v0vXLgwZ84cDQ2Nhr1Hjx4dNGgQXexuClIGAAAAOh3UuwEAAAA6pVu3bg0bNkxTU1NbW9vBwYFeq/r27dv29vY6OjpMJtPBwSE6OpoQEhQUxGazVVRUBg8e3KNHDwaDwWazBw0aNHr0aFNTUyaTqauru2HDBnqzBw8eZDKZ3bt3X7x4MZfLZTKZjo6Od+/ebSoMgUCwdetWMzMzFovVr1+/sLAwCeGJq6iooJowYsQIac7Aq1evWCyWpaVlo70ZGRm6urrm5ubijX5+fsuWLTM0NJRm+6BkkDLiampqEhMTBwwYIHkYUgYAAAA6HdS7AQAAADqfysrKmTNnuri4lJaWZmRk9O7du6amhhBSWFjo7u6elZWVl5fH4XDmzJlDCFm1atX69euFQuHRo0f//fffgoKCMWPGJCUlff3110lJSaWlpfPmzdu7d+/jx48JIT4+PvPnz+fxeCtXrszKynr48GFtbe2kSZNyc3MbjWTjxo179uwJDAzMz8//+OOPZ8+e/eDBg6bCE8fhcIRNSExMbPYM8Hi82NjYRYsWqauri7fz+fxXr14dOnQoJiYmODhYvPevv/7KzMycPXt2C082KAOkTD15eXk1NTV///33uHHj6DK9nZ3d4cOHxZc9QcoAAABAZ4R6NwAAAEDnk5WVVV5e3qdPHyaT2aNHjwsXLtCLe7i4uHzzzTd6enr6+vozZ84sKSkpLi4Wvcve3l5TU7Nbt26ff/45IcTMzMzAwEBTU/OLL74ghKSlpYlGqqmp2dnZaWho2NvbHzly5N27d6dPn24YRnV19ZEjR5ydnWfNmqWrq7t582YGg3H69OmmwpMhf39/Lpe7c+fOeu2mpqYmJibbtm3bs2ePu7u7qL2qqmrVqlVHjhyRbRjQWSBl6qGX+jE0NNy1a1dKSkphYeGnn366fPnykJAQegBSBgAAADop1LsBAAAAOh8rK6vu3bt/8cUX27Zty8rKanQMg8EghAgEgoZd9F3PtbW14iP5fH6j2xkyZIimpqZ4aU8kPT2dx+P17duXfslisYyMjNLS0qQJry0iIyPDw8Ojo6O1tLTqdeXm5hYVFYWEhPz0008DBw4sKiqi2zdt2uTl5UUv6g1dUBdPmYboFb379Onj6Oior6+vo6Pz7bff6ujoHD9+nB6AlAEAAIBOCvVuAAAAgM6HxWLFxsY6OTnt2rXLysrKw8OjqqqKEHL58uWxY8caGhpqaGiI1hduOw0NDfGbXkUqKysJIZs3bxatI5ydnc3j8ZoKT1yrFyMODQ3dvXv3zZs3LSwsGvYyGAxDQ8PJkyeHhoampKT4+/sTQuLj45OTkz09PVt9BqCz68op0ygul0sIef36tahFXV3d3Nw8MzOTIGUAAACgM0O9GwAAAKBT6tOnz++//56Xl+fr6xsWFrZv376cnBxnZ2cjI6O7d+++ffs2ICBAJjvi8/llZWUmJiYNu+in2AUGBoovJZyQkNBoePXe27rFiIODg8+ePRsbG2tsbCw5bBsbG1VV1ZSUFELIqVOnrl+/rqKiQhcH6bB37dpFUdSDBw9aeD6gs+qaKdMUDofTq1ev1NRU8cba2lodHR2ClAEAAIDODPVuAAAAgM4nLy+PLlQZGhp+9913gwYNSk1NTU5O5vP5S5cutbKyYjKZFEXJZF83b94UCoWN3kBqamrKZDIfPXokTXhtDEMoFPr6+iYnJ0dFRXE4nHq9JSUl9Z6ql5GRIRAITE1NCSGnT58WLw7SN976+fkJhcIhQ4a0MTDoFLpgyjTL3d09KSnpxYsX9Esej5edne3g4ECQMgAAANCZod4NAAAA0Pnk5eUtXrw4LS2tpqYmKSkpOzt7xIgRZmZmhJCYmJjq6uqMjIy7d++2evt1dXVv3rypra198uTJqlWrzMzM5s+f33AYk8lcsGDBL7/8cuTIkfLycoFA8PLly/z8/EbDa3UwtNTU1D179pw4cYLBYIiv5EDfBstms69evRobG1teXs7n85OSkubNm8dms9esWdPG/YJy6IIp06w1a9aYm5vPnz8/JyenpKTE19e3qqpq48aN7b1fAAAAgHaFejcAAABA52NoaCgQCBwdHTU1NWfMmLF48eLly5c7ODj4+voePnyYy+X6+fmNHTuWEOLk5LRu3bq9e/cSQhwcHOLj4wMCAhYvXkwImTp1akhISFhY2NSpUwkhPj4+oaGh9Parq6sdHBxYLNbo0aN79+5948YN+ul2+/fvd3JyIoSsW7du1qxZhJCgoKDVq1cHBAR069aNy+WuWrXqzZs3jYbXxkMWCoUSeplM5qhRozw9PXv27KmlpeXq6mphYZGYmCh6MCB0cV0wZQghiYmJTk5OxsbGd+/effz4MZfLHTVqVFxcHN2rp6d3+/ZtExOTAQMG9OzZ8969e5cvXx4wYEDb9wsAAAAgR5Tkbw4AAAAAIC48PNzd3V25/we1ePHiiIiIkpISeQfSVbi6uhJCIiIi5B2ILFEUFRYW5ubmJu9AOgJSRo66wmcyAAAAtAju7wYAAACA+gQCgbxDAOhMkDIAAAAACgL1bgAAAAAAAAAAAABQBqh3AwAAAMD/t2nTptOnT799+9bS0vL8+fPyDgdA0SFlAAAAABSKmrwDAAAAAAAF4u/v7+/vL+8oADoNpAwAAACAQsH93QAAAAAAAAAAAACgDFDvBgAAAAAAAAAAAABlgHo3AAAAgDLYt29f9+7dKYr6/vvv5RjG4sWLqf/zxRdfEEK2b99ub2+vra2toaFhY2OzYcOGiooK0XjJveKqq6ttbW03b97conjq6uoCAwMdHR0bdsXHx48aNUpTU5PL5fr6+r5//55uj4qKEh2CgYFBi3YHnUsnzRqahLndLGne2zDjAgICbG1tWSwWm822tbXdsmVLeXk53YWsAQAAAMWBejcAAACAMli3bt2dO3fkHQUhhOjr6//xxx/p6emnTp0ihMTGxi5fvjwrK+v169f+/v5BQUGurq6iwZJ7xfn5+aWnp7cokoyMjDFjxqxZs4bH49XrSklJmTx58oQJE4qLiyMjI3/44YclS5bQXZ988snLly/j4uKmT5/eot1Bp9NJs4ZInNvNkvK9DTPu9u3bixYtysnJKSws3LFjR0BAgIuLC92FrAEAAADFgXo3AAAAAMgSi8WaOnVq7969NTQ0CCEcDsfb21tfX19LS8vNzc3Z2fnPP//Mzc2lB0vuFblz587Tp09bFMbjx483bty4ZMmSAQMGNOzdsWOHkZHRt99+y2azR44c6evr++OPP6alpRFCKIrq2bPn6NGje/Xq1ZrjB2i5FmWN5LktmZTvbTTj1NXVly1bZmhoyOFwXF1dP/3002vXruXn5xNkDQAAACgS1LsBAAAAoB1dunRJVVVV9JJe60B0Y6nkXlpVVdX69euDgoJatN/+/ftfuHBhzpw5dAFRXG1t7eXLlz/88EOKouiWadOmCYXCixcvtmgXAO1Ecl5ImNvNkua9TWVcZGQkk8kUvezF4g6jAAAgAElEQVTZsychpKkFiAAAAADkBfVuAAAAAMViZ2dHUZSKisrgwYPpCteGDRt0dHSYTOaPP/5ICLl9+7a9vT3d4uDgEB0d3XAjPj4+6urqRkZG9Mtly5ax2WyKol6/fk23CASCrVu3mpmZsVisfv36hYWFNdxIRUUF1YQRI0a07uhevXrFYrEsLS2l7/Xz86PvKm3dHht68eJFRUWFmZmZqMXa2poQ8uTJE1ntAjpYV84amZMy4zIyMnR1dc3NzTsmKgAAAAApqck7AAAAAAD4H0+fPrWxsamrq7t37x59j+eePXtqa2vNzMzmz59PCCksLHR3d/fx8REKhR999NGcOXNE9TiRgwcPlpaWxsTE0C8PHz5saWm5fv160YCNGzcGBwefO3duwoQJe/funT17trW19ZAhQ8Q3wuFwhEKhDA+Nx+PFxsYuWrRIXV1dyt6//vorMzNz//79DY+x1QoKCgghWlpaohYmk8lisQoLC2W1C+hgXTZrZK7ZjOPz+UVFRb/++mtMTMzJkyc7JioAAAAA6eH+bgAAAADFoqqqunLlypycnMjISLqFx+NduHBh4cKF9EsXF5dvvvlGT09PX19/5syZJSUlxcXFLdpFdXX1kSNHnJ2dZ82apauru3nzZgaDcfr0aRkfSQP+/v5cLnfnzp1S9lZVVa1aterIkSOyDeP9+/eEEPH1IgghDAajqqpKtjuCDtNls0a2pMk4U1NTExOTbdu27dmzx93dvQOiAgAAAGgR1LsBAAAAFI6np6eOjo5o/dyzZ89++umn2traDUcyGAxCiEAgaNH209PTeTxe37596ZcsFsvIyIh+WmP7iYyMDA8Pj46OFr+xWnLvpk2bvLy86GWCZYheg7i2tla8saamhsViyXZH0JG6YNbInDQZl5ubW1RUFBIS8tNPPw0cOLCoqKgDAgMAAACQHurdAAAAAAqHw+F4eXnduXPn3r17hJCjR4/6+PiIei9fvjx27FhDQ0MNDY0NGza0YvuVlZWEkM2bN4tWFs7Ozq73lEgi05WIQ0NDd+/effPmTQsLCyl74+Pjk5OTPT09W3p0zaIXaC4vLxe18Hi86upqLpcr831Bh+lqWSNzUmYcg8EwNDScPHlyaGhoSkqKv79/B8QGAAAAID3UuwEAAAAUkY+PD4PBCAwMjIuLMzU1pR+oSAjJyclxdnY2MjK6e/fu27dvAwICWrFx+kl0gYGBQjEJCQn1htErETcqMTFR+t0FBwefPXs2NjbW2NhY+t5Tp05dv35dRUWFrhXSMe/atYuiqAcPHrT4mMVYWlpqaWllZ2eLWp4/f04I6devX1s2C3LXdbKmPbQ042xsbFRVVVNSUjomPAAAAAApod4NAAAAoIhMTEzc3NzOnz+/ZcuWVatWidqTk5P5fP7SpUutrKyYTCZFUU1tQU1Njc/nN9plamrKZDIfPXok+7j/l1Ao9PX1TU5OjoqK4nA4Leo9ffq0eK2QXm3Zz89PKBTWe0JgS6mpqU2fPj0uLq6uro5u+eOPPyiKmjlzZls2C3LXFbKm/UjOuJKSktmzZ4uPz8jIEAgEpqamHRYhAAAAgDRQ7wYAAABQUGvXrq2trX3z5s348eNFjWZmZoSQmJiY6urqjIyMu3fvNvV2Gxub0tLSqKgoPp9fXFwsfjszk8lcsGDBL7/8cuTIkfLycoFA8PLly/z8fJkfQmpq6p49e06cOMFgMMQXdti3b1+zvc3y8PDo0aPHw4cPWxHYli1bCgsLv/nmm8rKyoSEhL17986fP/+DDz5oxaZAoSh91jSrLXkhAZvNvnr1amxsbHl5OZ/PT0pKmjdvHpvNXrNmjWx3BAAAANBGqHcDAAAAKKiBAweOGzdu5cqV4o0ODg6+vr6HDx/mcrl+fn5jx44lhDg5Oa1evdrJyYkQsm7dulmzZhFCli5dOm7cuM8///yDDz7YsWMH/TDGkSNH5ubmEkKCgoJWr14dEBDQrVs3Lpe7atWqN2/eyPwQhEJhq3ubVVNTU1RUdPHixUZ7ExMTnZycjI2N7969+/jxYy6XO2rUqLi4OLq3T58+0dHRV69e7dat26xZsxYuXHj06NG2BAMKQumzhjQ3t9uSFxIwmcxRo0Z5enr27NlTS0vL1dXVwsIiMTFR9ABPAAAAAAVBtfFrBgAAAECXEh4e7u7ujv9BNWXx4sWXLl16+fJlB+yrrq5u7Nix8+fPX7hwYXtsf9WqVWfPnn39+nV7bFycq6srISQiIqK9d9SRKIoKCwtzc3OTdyCdgGyzpr3zQrIOyxoRfCYDAABAPbi/GwAAAABkqaqqKjo6OiMjo6ampv32IhAIoqKi3r175+HhIdstC4XCvLy8+Ph4+iGWAB1AVlnTfnkhGbIGAAAAFIeavAMAAAAAAKVSWlo6depUQsicOXPOnj3bTnu5efPmhQsX/vjjD01NTdlu+eLFi87OzvS/u3XrJtuNAzRKVlnTfnkhGbIGAAAAFAfWMwEAAABoAfztPMgc1jMBaDV8JgMAAEA9WM8EAAAAAAAAAAAAAJQB6t0AAAAAAAAAAAAAoAxQ7wYAAAAAAAAAAAAAZYB6NwAAAAAAAAAAAAAoA9S7AQAAAAAAAAAAAEAZoN4NAAAAAAAAAAAAAMpATd4BAAAAAHQ+FEXJOwRQKi4uLvIOQfbc3d3d3d3lHQUAAAAAdC2odwMAAAC0gKOjY1hYmOQxfD4/Ojr6119/pShq4cKFjo6OHRObonF3d1+1atXIkSPlHUgnYGpqKu8QZKzZNIGGAgMDCSGrV6+WdyAKobi4ODg4OD09fcCAAa6urjY2NvKOCAAAADoHSigUyjsGAAAAACVRV1d34cIFX1/fwsLCFStWbNy4UVdXV95ByQ1FUWFhYW5ubvIOBKBzoJMlPDxc3oEokPj4+C1btty8eXPixIn+/v5Dhw6Vd0QAAACg6LB+NwAAAIBsxMTEDB482MPDw8nJKSMjY/fu3V252A0A0HZOTk43bty4du3au3fvhg0bNmnSpAcPHsg7KAAAAFBoqHcDAAAAtNW9e/fGjRs3adIkAwODR48enTlzxtjYWN5BAQAoiYkTJyYmJl67dq28vHzo0KGTJk36+++/5R0UAAAAKCjUuwEAAABaLz093c3NbcSIEe/fv4+Li7t27ZqDg4O8gwIAUEITJ068e/futWvX3r59O3To0I8//vjhw4fyDgoAAAAUDurdAAAAAK2Rl5fn7e3dt2/flJSUsLCwO3fujB49Wt5BAQAouYkTJ967d+/q1av5+flDhgz5+OOPk5KS5B0UAAAAKBDUuwEAAABapqKiIiAgwNbW9o8//jh8+PCTJ09cXV3lHRQAQBcyceLE+/fvX7x4MS8vj656P378WN5BAQAAgEJAvRsAAABAWjU1NcePH7e2tt6zZ4+fn9+zZ8+8vLxUVVXlHRcAQJdDUdTHH3/84MGDqKioV69eDRo0yM3NLS0tTd5xAQAAgJyh3g0AAADQvLq6uoiICDs7u9WrVy9YsCAzM9PX15fJZMo7LgCALo2uev/9999RUVEZGRl9+vRxc3NLT0+Xd1wAAAAgN6h3AwAAADQjJiZm8ODBHh4egwcPTk1N3b17t66urryDAgCA/xJVvUNDQ5OTk+3t7d3c3J49eybvuAAAAEAOUO8GAAAAaNK9e/fGjx8/adIkAwODR48ehYeHm5ubyzsoAABohIqKiqura0pKSmho6JMnT+zs7Nzc3DIyMuQdFwAAAHQo1LsBAAAAGpGenu7m5jZixIiqqqpbt25du3bNwcFB3kEBAEAz6Kp3ampqaGjo48eP+/TpM3fu3OfPn8s7LgAAAOggqHcDAAAA/I/i4uKVK1f27dv36dOnYWFhCQkJY8aMkXdQAADQAnTV+59//jl37lxiYqK9vf3cuXMzMzPlHRcAAAC0O9S7AQAAAP6roqIiICDA2to6MjLy8OHDycnJrq6u8g4KAABaSbTCycmTJxMSEuzs7ObOnfvixQt5xwUAAADtCPVuAAAAAFJTU3P8+HEbG5s9e/b4+fk9e/bMy8tLVVVV3nEBAEBbMRiMuXPnpqamnjx58s6dO3Z2dt7e3q9evZJ3XAAAANAuUO8GAACALq2uri4iIsLOzm7FihXu7u6ZmZm+vr4sFkvecQEAgCzRVe9//vnnxIkTMTExVlZWqHoDAAAoJdS7AQAAoOuKiYkZMmSIh4fH4MGD09PTDxw4oKurK++gAACgvYiq3sHBwVeuXKGr3nl5efKOCwAAAGQG9W4AAADoiu7fvz9hwoRJkyZ169YtKSkpPDzcwsJC3kEBAEBHUFdX9/LyyszMDA4Ovnz5Ml31zs/Pl3dcAAAAIAOodwMAAEDX8uzZMzc3t+HDh/N4vFu3bl27dq1fv37yDgoAADoaXfV+8eLFwYMHL1261KtXr5UrVxYUFMg7LgAAAGgT1LsBAACgqyguLt64caODg8PTp0/DwsISEhLGjBkj76AAAECeRFXv/fv3R0RE2NjYrFy5srCwUN5xAQAAQCuh3g0AAADKr6KiIiAgwNra+ty5c8HBwcnJya6urvIOCgAAFIWGhoaXl9fz58937doVHh5OV72LiorkHRcAAAC0GOrdAAAAoMz4fP7x48dtbGwCAgL8/PyePXvm5eWlqqoq77gAAEDhaGpqrly58vnz5zt37gwPD7e2tt64ceObN2/kHRcAAAC0AOrdAAAAoJyEQmFERISdnd2KFSvc3d0zMzN9fX1ZLJa84wIAAIXGZrNFVe8ff/zR3NwcVW8AAIBOBPVuAAAAUEIxMTGDBw/28PAYNGhQenr6gQMH9PT05B0UAAB0GqKqt5+f34kTJ+iqd1lZmbzjAgAAgGag3g0AAABK5f79+xMmTJg0aVK3bt2SkpLCw8MtLCzkHRQAAHRKHA7H19c3Ozvbz8/v+PHjdNX77du38o4LAAAAmoR6NwAAACiJ7OzsuXPnDh8+vLKy8ubNm9euXevXr5+8gwIAgE5PVPXetGnTsWPHrK2tt23bhqo3AACAYkK9GwAAADq9169fb9y48YMPPrh//35YWFhCQsKHH34o76AAAECpaGlp+fr65uTkrF+//sCBA3TVu7y8XN5xAQAAwP9AvRsAAAA6scrKyoCAAGtr67Nnzx48eDA5OdnV1ZWiKHnHBQAAyomuemdmZi5fvjwoKMja2jogIIDH48k7LgAAAPgv1LsBAACgU+Lz+cePH6cLDZs2bcrIyPDy8lJTU5N3XAAAoPz09fW3bduWmZm5bNkyf39/CwsLVL0BAAAUBOrdAAAA0MkIhcKIiAg7O7sVK1Z88skn6enpvr6+LBZL3nEBAEDX0q1bN7rqvXTpUlHVu6qqSt5xAQAAdGmodwMAAEBnEhMTM2TIEA8Pj0GDBqWlpR07dszQ0FDeQQEAQNdlYGBAV70XLly4fft2VL0BAADkC/VuAAAA6BwePHgwceLESZMm6evrP3z4MDw83NLSUt5BAQAAEEKIgYHB7t27s7KyFixY8O233/bu3fvAgQPV1dXyjgsAAKDLQb0bAAAAFF12dra3t/fw4cMrKipu3rx57dq1/v37yzsoAACA+gwNDemq95w5c77++mtUvQEAADoeJRQK5R0DAAAAQONev369b9++oKAgS0vL7du3u7i4UBQl76CgSfv27Xv9+rXo5bFjxyZOnGhtbS1qWblyJZfLlUdoAIooOjr6xo0bopfXrl0jhEyaNEnUMm7cuClTpsghMpCFoqKi/fv3HzhwoHv37mvWrFm8eLGGhoa8gwIAAFB+qHcDAACAIqqsrDx06JC/v7+WltbWrVsXLlyopqYm76CgGcuWLTt69Ki6unrDrtraWj09vYKCAlVV1Y4PDEAx3bhxY/z48QwGQ0Wl/t/d1tXV8fn82NjYcePGySU2kJXc3Nx9+/YdP368R48emzZtwuUMAACgvaHeDQAAAB1NKBRKuE2bz+efPn36m2++qaioWLt2ra+vL4vF6sjwoNXi4+NHjx7daJe6uvrSpUsDAwM7OCQARVZXV8flcouKihrtNTQ0zM/Px6+IlENOTs5//vOfY8eOcbncr7/+GlVvAACA9oP1uwEAAKBDZWRkuLi4NNolFAojIiLs7e1XrFgxc+bMzMzMbdu2odjdiYwaNcrY2LjRrpqams8//7yD4wFQcCoqKl988UWjfxKhrq4+d+5cFLuVhpmZ2YEDB549ezZz5kwfH59evXodP368tra2qfHBwcFlZWUdGSEAAIDSQL0bAAAAOk5ubu6YMWMiIyN///33el0xMTFDhgzx8PAYOHBgWlrasWPHunfvLpcgodUoivryyy8ZDEbDLlNT06FDh3Z8SAAK7vPPP6+pqWnYjl8RKSW66p2enj558uRly5b17t37+PHjAoGg3rCsrKw1a9ZMnjz53bt3cokTAACgU0O9GwAAADpISUnJhAkTSkpKVFRU1q1bJ/qG//Tp048//njSpEn6+voPHz4MDw+3tLSUb6jQap9//jmfz6/XqK6uPm/ePDxrFKChIUOGNPqJZ25uPnjw4I6PBzqAubn5sWPHMjIyJk2atGzZMgcHhzNnzohXvXfs2EFRVFJS0rRp03g8nhxDBQAA6IxQ7wYAAICOwOPxpk+fnpWVxefz6+rqnj9/fubMmezsbG9v7/79+xcVFd24cePatWv9+/eXd6TQJv379+/Vq1e9xpqaGg8PD7nEA6D45s6dW++vItTV1efPny+ncKCDWFhYHDt27NmzZ6NHj/7qq6/69etHV72zsrLOnDnD5/Nra2vv3bs3Y8aM6upqeQcLAADQmeB5lQAAANDu+Hz+jBkzbty4Ibrzl6IoDodTXV1tbW3t7+/v7Ows3whBhnbu3Ll9+3bxu7zt7OxSU1PlGBKAInv+/HnD3xKlpaV98MEHcokHOl5aWtr27dvDwsL69OljYGAQHx8v+ghVU1MbN27c77//rqGhId8gAQAAOgvc3w0AAADtSygUenp6Xr9+XbwAKhQKeTzezJkzk5OTUexWMrNnzxZ/CBuDwZg3b54c4wFQcDY2Ng4ODqIFfyiK6tevH4rdXYqtrW1ISMjTp0+tra1v3rwpfrmsra2NjY11d3eX8HBLAAAAEId6NwAAALSvdevWnT17tuHzuAQCQXR0dHl5uVyigvZjZWU1cOBAFZX//j+ztrbW3d1dviEBKLi5c+eqqqrS/1ZTU5s7d6584wG5sLOz43A4ampq9doFAsGlS5fmzZtXV1cnl8AAAAA6F9S7AQAAoB35+/sHBgY29RX9/fv3e/bs6eCQoAPMnTuXrndTFDVs2DALCwt5RwSg0GbPni36nMSviLqsjIyMkJCQho/8JYQIBIKwsLCFCxdiPVIAAIBmod4NAAAA7eXkyZObN2+W8OWcz+cHBga+evWqI6OCDuDu7k4X71RUVHCnKkCzjI2NR44cqaKioqKi4ujoaGJiIu+IQA6++eYb0W3+DQkEgrNnz/r4+HRkSAAAAJ0R6t0AAADQLqKiory9vRstdqupqamrq9P/FgqFP/74Y4dGBu3PyMhozJgx9HrELi4u8g4HoBP48ssvKYqiKOrLL7+UdywgB2lpaWFhYXV1dQwGo6kxAoHgyJEjGzdu7MjAAAAAOh0Kfw8FAAAAMnfr1q3JkyfX1NQQQuilSOkHbWlqavbq1WvAgAF2dna2trb29vaWlpYN1yoFJfDDDz989dVXEydOvHbtmrxjAegE3rx50717d6FQWFhY2K1bN3mHAx0tPz8/MTHxxYsX//777/Pnz58/f56Tk0OvbaKqqspgMGpra0WPrPz222+3bt0q13gBAAAUF+rdAABdhaur6/nz5+UdBUD7CgsLc3Nza+NG6LuSARQK5jaAgmj7N+jw8HAs0Q4Ky8XFJSIiQt5RAAC0CW6nAgDoQkaMGLF69Wp5RwHK79GjRxwOx9jYWFNTsyP3K8PywapVq0aOHCmrrXVZBw8e9PT07OBpoJQwt7uI+Ph4iqJGjRol70CgcQkJCUFBQbLaWlhYWOveyOfzi4qKioqKbG1tWSyWrOIBIIQEBgbKOwQAABlAvRsAoAsxMTFp++2BAM2S1zSTYU1w5MiRSJa2+/DDD3v06CHvKJQB5nYXMX36dEIIh8ORdyDQJBnWu5GJoIBwZzcAKAfUuwEAAACgXaDYDdAiqHQDAAAAtJ2KvAMAAAAAAAAAAAAAAJAB1LsBAAAAAAAAAAAAQBmg3g0AAAAAAAAAAAAAygD1bgAAAAAAAAAAAABQBqh3AwDA/zdo0CCKoiiK4nK5K1eubGrYs2fPhg4dyuFwVFRUpk6d2pERijx+/NjDw8PS0lJDQ8PAwKB///47d+6USyQiCxYsYDKZFEVVV1fLN5L2Ex8fP2rUKE1NTS6X6+vr+/79e8nj6+rqAgMDHR0d67WPHTuWakD8QW0hISFDhw7V0tIyNzdfsGBBQUGB7A+mDS5cuGBlZUWHvWXLlkbH7N+/n6IoFRUVW1vbuLi4jgzPw8Oj4ekVd+nSpY6MR0E0NRv5fP7WrVutrKzU1dV79uy5bt26qqqqpjayfft2e3t7bW1tDQ0NGxubDRs2VFRUNDqyurra1tZ28+bNopadO3fW+0H07dtXJocmQ53oKkAISU9PX7FiRZ8+fbS0tNTU1HR0dHr37v3RRx8lJCTIKyQZEv+coSiKwWD07Nlzzpw5//zzj5RbuHLlio6Ozu+//96uccpWx1xlpE9kOeoUyYjLTT3NTq0WzXA+n+/v729jY6Ourq6rq9u3b9+srKyGwzrp5QYAoF2h3g0AAP/fw4cPJ0+eTFFUUlLSgQMHRO0CgWD8+PGil717975//76/v//s2bP//PPPjo8zOTnZ0dHRyMjoxo0bb9++vXPnztSpU2/evNnxkYg7ffr0unXr5BtDu0pJSZk8efKECROKi4sjIyN/+OGHJUuWSBifkZExZsyYNWvW8Hg8abbv5ORE/yMsLGzOnDmurq4vX768ePFiXFzctGnTamtrZXAMMjJr1qwXL15YW1sTQk6ePMnn8+sNEAgEBw8eJISMHz8+LS1tzJgxHRzh1atXy8rK+Hx+fn4+IWTmzJk1NTWVlZVFRUWLFi3q4GAUgYTZuGrVqr179/r7+5eUlJw7d+7EiROenp5NbSc2Nnb58uVZWVmvX7/29/cPCgpydXVtdKSfn196erosj6FDdJarACHk1KlTDg4OT5482b9/f25ubmVlZVJS0o4dO8rKypKTk+USkmyJPmd0dHSEQmFZWdn3338fHx8/bNgwKaeWUChs7yBlq8OuMtInshx1lmTE5Uac5KnV0hnu7u5+5syZc+fO8Xi8f/75x9rautFfzHTSyw0AQLtCvRsAAP6Hh4eHUCisdz/OjRs3bty4kZqaKt545coVd3d3We23qqqq4f1ZTdm3b5+urm5QUJCFhQWTyezdu/eOHTtYLJasgpFei8KWOz6ff/r06Xnz5rXu7Tt27DAyMvr222/ZbPbIkSN9fX1//PHHtLS0Rgc/fvx448aNS5YsGTBgQMNeJpNZXl4uFOPt7b1hwwa699ixY8bGxuvXr9fR0RkwYMCaNWsePXp09+7d1oXdrgYPHlxQUBAVFVWv/cKFCz179pRLSIQQiqJGjRqlo6OjpqYmamEwGJqamoaGhoMHD271luU14ds4dSXMxhcvXnz//fdz58718PDQ0tIaO3asj49PSEhIU7fQcjgcb29vfX19LS0tNzc3Z2fnP//8Mzc3t96wO3fuPH36tOHbf/75Z/Fp3+gYuesUV4HExERvb+/Ro0dfv359ypQpurq6GhoaVlZW7u7uW7durampkVVUioPNZn/88ccHDhyoqKgIDg6W5i0fffTR27dvP/74Y2kGyyS7O8tVRspEljvFT0ZcbuqRPLVaNMNDQ0OjoqIiIiKGDx+upqbG5XIvXrzY8DbtTn25AQBoP6h3AwDA//jss8/U1dV/++038carV68aGxufP39e1FJVVfX48eMpU6bIar+nTp0qKiqScnBJScnbt29LS0tFLerq6nL5k+1Gw6YoquMjkez9+/dHjx4dNGhQSkrK7t27W7GF2tray5cvf/jhh6KjmzZtmlAovHjxYqPj+/fvf+HChTlz5mhoaDTs/fPPP7W0tEQvc3Nznz59KrphLTc3l8vlinZkampKCMnOzm5F2O1t6dKlhJCjR4/Wa9+/f//atWvlEREhhPzyyy+amppN9Xp7e8+YMaN1W25RnspE26cukTgb79+/X1dXN3z4cFELvSBAdHR0o5u6dOmSqqqq6KWBgQEhpN69pVVVVevXrw8KCmpdtHLXKa4CO3fuFAgE3333najKJjJlypTly5fLKipFM2zYMEJIe5Su2pjdnesqI00iKwLFT0ZcbuqRMLVaOsPpYBwcHCTsrrNfbgAA2g/q3QAA8D90dHSmTJkSExMjWsGWXqvBxcVF/MsVfUuduro6IUQoFO7fv9/Ozk5DQ0NPT+/TTz8V3auyZ88eTU1NLS2toqKitWvX9uzZMz09/datW8OGDdPU1NTW1nZwcCgvL1+1atXatWszMzMpirKxsWk2yKFDh1ZWVo4fP/6vv/5q2BsUFMRms1VUVAYPHtyjRw8Gg8FmswcNGjR69GhTU1Mmk6mrqyu6yUty/JJ7Gw1bRUXl8uXL06ZN09HR4XK5P/zwA91+5MgRNputqal58eLFadOmaWtrm5iY/PLLL6IdCQSCrVu3mpmZsVisfv36hYWF0e0NT1dTjY2qrKzcv3//kCFDioqKbt26tW/fPi6XSwipqKhoarXNESNGNNzOixcvKioqzMzMRC30ah5PnjyR+LOSyu7du8UXJ7WyshL/lksv3m1lZdX2Hcnc+PHj7ezsbty4If6nxH/99RePx5s8eXK9wU39iG/fvm1vb6+jo8NkMh0cHOhKa7MTpo0aDebHH3/kcJVzBdkAACAASURBVDgURenp6UVFRT148MDc3FxVVXX27NmkwYT38fFRV1c3MjKiN7hs2TI2m01R1OvXr0kTud/UGWiUrKauZCoqKoQQ8b8O6dWrFyFEyiWSX716xWKxLC0txRv9/PyWLVtmaGjY0mAUhOJfBWpqaq5fv96tWze6+CuBhMCkSbGff/55yJAhTCaTzWZbWFjs2LGDNJGwdnZ2FEXR1x26trVhwwZ6zI8//kiaTv+Wold2ElV4JRxgfHy8mZkZRVGHDh1q9ngbnn/lvsrU02giKwLFT0ZpdOXLjfjUatEMr6mpSUxMbPRvF8R19ssNAEA7EgIAQNfg4uLi4uIizchz584RQn777Tf65eXLl6Ojo2/fvk0ISU9PpxuXLFly5coV+t9bt25VV1f/+eefy8rKnjx5MmjQIAMDg4KCArrXz8+PELJy5crg4ODPPvvswYMH2traAQEBVVVVBQUFn332WXFxsVAonDVrlrW1tZTHwuPxhgwZQl/I7O3tAwICSkpKxAd88803hJC7d+9WVla+fv2avmHz8uXLxcXFlZWVPj4+hJBHjx5JE7/k3nph0wd7/fr1srKy0tLS6dOna2hoVFZW1ut9+/ZtUVHR6NGj2Wx2TU0N3btu3ToNDY3z58+/efNm06ZNKioq9+/fr6ioaHi6Gm1seJbKysp27tzZr1+/ffv2VVRUSHlum3Lr1i1CyN69e8UbWSzWhAkTJL9x+PDh/fv3lzDg5cuX9vb2AoFA1HLz5k0Gg3Hw4MHy8vKnT5/a2dlNmTJFmiAJIWFhYdKMlMl2rK2t//33X3pZ1VWrVonanZ2dT58+/e7dO0KI+Plp9EcsFAojIiK2bdtWWlpaUlIyYsSIbt260eMlTxgp0QuqfvLJJ/XamwomNTVVU1Nz3rx59LCvv/765MmTonfVm/Bz5szp0aOH6OXevXsJIaLZWC/3//nnn6Z2Wo9sp664hrORLjRs2bJF1ELXE52dnZvdWmVlpZaWlo+Pj3hjfHz8zJkzhUJhcXExIcTPz0/UtWPHDhMTE11dXQaDYWFh8cknn9y7d0/KyDt4biv4VeDZs2eEkBEjRjQ7UprAmkqxwMBAQsh3331XUlJSWlp67NixOXPmCJtI2NraWgsLCzMzs9raWtHeV69eHRgYSP9bysnfkGj9btrPP/9MCFm/fr00B0ivohAcHCzN8Yqff6W/yohrNJGbQpdNpRkpq+0oeDKKw+WmnnpTq0Uz/N9//yWEDBgwYOzYsUZGRhoaGra2tocOHaqrqxONaafLjfTfFwAAFBnq3QAAXYX0/3999+4di8Xy9PSkX65du5bP59fV1RkbG+/cuZNuHDRoEP0lmcfjcTgceolJ2r179wgh27dvp1/SX0Kqqqrol/RfYV+6dKneTlv65aqmpubAgQO2trZ01bt79+43b94U9dL17nfv3tEvf/rpJ0JIcnKyeIShoaHNxt/s0TVa7xYd7JkzZwghT58+bbT38OHDhJDnz58LhcKqqipNTU3Rjng8noaGxtKlSxs9XU2dw3rnx9zcfMyYMfUWMG21q1evEkL2798v3qitre3o6Cj5jc1WIpYvX3706NF6jZs3bxb9bt7ExCQ3N1eaIOVS7y4rK2Oz2Xp6ejweTygUZmZmmpiYvH//vl69u6kfcb1t+vv7E0KKioqEEieM9BotQEgO5tixY4SQs2fPhoSErFmzRvyNrShAiOKX8gzIfOqKa3Q2Tp06VV9f//r161VVVfn5+eHh4RRFzZgxo9mt+fn59e7dWzxO+ldxL1++FDZWgMjJyXn48OG7d+/ev3+fkJAwcOBAFosl+nyQrIPntoJfBR48eEAImThxouRhLQ1MPMVqamp0dXXHjRsnem9tbW1QUFC9XYgnLF0fDw8Pp7sqKyvNzMzevn0rlHryN0pU766oqDh//nyPHj26d+9Oz7FmD7DRendTHyni578rXGVEGiayBB1f71bwZBSHy0099aZWi2Y4/dDdSZMm/fXXXyUlJWVlZRs3bqTPFT2g/S43qHcDgHLAeiYAAFAfh8P56KOP6O8/79+/V1NTU1NToyhq1qxZ9N/PpqamDhw4kMFgEEJSUlIqKipEd1sTQoYOHaqurt7U0wWtrKy6d+/+xRdfbNu2LSsrq9VBMhgMHx+ff/75JzEx8dNPPy0qKnJ1dX3z5k2jg+k/8qVv26TfS/7vj4Ilx9/So2sYpGhHTUVF96anp/N4PNFjiFgslpGRUVpaWqOnS5pzyGAwHj9+PGnSpNGjRwcGBlZWVkoTsARMJpOInUNaTU1NGx8TmpeX99tvv82fP1+80c/P7/jx49evX6+oqHjx4oWjo+PIkSMV8EliNB0dndmzZ7958yY0NJQQEhgYuHTpUvqHK66pH3G9YfScEQgEDXckPmHaSHIwXl5eLi4uixcvDg8P37NnT9t3J81ORWQ+dZsVGhrq6uo6d+5cfX39UaNG/frrr0KhsFu3bpLfFRkZGR4eHh0dLb5G8KZNm7y8vJp6VKmpqenAgQM5HI66uvqIESNOnz5dVVVF1xwVjYJfBTgcDpFiteWWBiaeYk+ePCkrKxNfEFlVVbXhghjiCevp6amjoyNaSPfs2bOffvqptrY2kXryN+Xt27cUReno6KxcuXL69On37t2j51gbr1ASPlKU/ioj0mgiKxQFT8ZmddnLTcOp1aIZTq9Z1KdPH0dHR319fR0dnW+//VZHR+f48eP0AKW53AAAtBPUuwEAoBEeHh4FBQX379+/cuUKvRgIIcTFxeXRo0eZmZlXrlxxd3enG8vKysj/VR9EdHV16ZtbG2KxWLGxsU5OTrt27bKysvLw8BCtStk6w4cP//XXX5csWVJcXHzjxo2Wvl1y/C09ulajv2Vt3rxZtDRkdnY2j8dr9HRJeQ51dHQ2b978119/1dXVjRgxYvv27eK/D2jpqpT0upniS7jyeLzq6mp6jctWCwgIWLRoEf0lkJafnx8QEODl5TV+/Hg2m21paXnixIm8vDz6Zi7FRD+18vvvvy8rK4uIiFi8eHHDMU39iAkhly9fHjt2rKGhoYaGhvjK8u1HQjC0Xbt2VVRUyPZZYc3uVES2U7dZOjo633///cuXL3k8XmZm5n/+8x9CiLGxsYS3hIaG7t69++bNmxYWFqLG+Pj45ORkT09PKffr4OCgqqpKL82hgBT5KmBhYcFkMps9dW35AKc/63R1dRt2NZWwHA7Hy8vrzp079B21R48epdfOIi2Z/I2i7++ura19+fLlDz/8YG5u3vYDlEy5rzIijSayAlLkZGxW17zcNDq1WjTD6UZ6gXKaurq6ubl5ZmYmUbrLDQBAe0C9GwAAGvHRRx9paWn99ttvcXFxY8aMoRudnJy4XO758+cfPHgwbtw4upEuB9T7KlVWVmZiYtLUxvv06fP777/n5eX5+vqGhYXt27evpeHNmjWr3g0yX375JZHidr+GJMffiqNrHfpZQ6KVXmkJCQmkidMl/Tlks9lr1669f/++gYHBmDFj1q9fT//FMYfDaeqPvxITExtux9LSUktLKzs7W9Ty/PlzQki/fv1afdQFBQUhISF0sVgkIyNDIBCIVxu1tbX19fVTUlJavaP2NmDAgBEjRty7d8/b29vV1VVPT6/hmKZ+xDk5Oc7OzkZGRnfv3n379m1AQEAHBCxhvhFC+Hz+ypUr9+/fn5CQsHPnzo7ZaUOymrotdf/+fUKI6COuoeDg4LNnz8bGxtariZ86der69esqKip0NYQ+3l27dlEURa+/UU9dXV1dXZ3owYOKRpGvAhoaGlOmTHn9+nWjjywuLS2ly0Bt+QCnf7jixSba/2PvvuOiuPa/gZ+tLE2KoiJFEGwo9ooVoyJGUYQFokYxoqiJxoJAbDEWLsVcNaIJEjXJVaSIscaGREVEYsMgKCKGrogVEZBddp8/5vfs3UtZFliYLZ/3H3llZ2bPfAbmy+wcz56RXbArVqzgcDg7d+68du2ahYUF9TA60vSTX06teoVS46sMpaFCVkLKXIyN0sDLTUOnVpPOcD09ve7du2dmZkovFAqFBgYGRO0uNwAArQH93QAAUA8ej+fi4nLs2DFtbW0m8/8uFkwmc9asWb/99lvHjh3ZbDa1sG/fvnp6etIfr1NTU6urqwcPHlxvy8XFxdTHdxMTk3/961+DBg2q9WleHh8/fqz1rqysLNKsu2LZ+Zt6dM1mYWHB4/HS0tJqLa/3x9WMnyGPx1u2bNndu3d79+5NTQHZVGw2e+rUqdeuXROJRNSSc+fOMRgMFxeXZrRGCQkJmTt3rrGxsfRC6racutWkvH///vXr1xYWFs3eURug+lOOHTu2atWqejdo6Fecnp4uEAiWLVvWrVs3Ho/HYDBaPWvDYSjLly9ftGjRqlWrVq9evXXrVhl9BGw2W/75VWTvtCEtP3WbKjIy0traety4cXVXicXigICA9PT0EydO1BpBSQg5dOiQdG+I9ISq1PQC0pNjEEKoh6eNHDmy1Q6lRZT8KrB582YtLa3Vq1fXHY764MEDKltL/oBbWVkZGxtT8+1Kk12w5ubmHh4ex44d27hx48qVKyXLm3fyN6r1rlDqfZWRXchKSMmLUTaNutzIPrWaeoZ7enreu3fv6dOn1MuKioq8vDx7e3uidpcbAIDWgP5uAACon5eXV1ZW1rRp06QX8vn8zMxMV1dXyRIej7dmzZrjx48fPny4rKwsPT196dKlpqamvr6+9TZbXFy8ZMmSR48eVVdX37t3Ly8vj/pOqLGxcXFxcW5u7vv37+W5pXF1dY2NjX379u27d+9OnjwZGBg4Y8aMZvR3y87f6NE1NbaMGAsWLDh69Oi+ffvKyspqamoKCwufPXtW74+roZ9hozgczhdffEE9vbMZNm7cWFJS8u2333748CElJSUsLMzb27tnz57U2k2bNhkYGNTtHmpISUnJwYMH6/YOW1tbOzo6RkZGXrt2rbKysqCggPppL1y4sHmx24aHh0eHDh1cXV27detW7wYN/YotLS0JIQkJCVVVVdnZ2XJOvNtCDYUhhOzdu9fMzGzWrFmEkKCgIDs7uzlz5ki+f13rhLe1tX39+vWJEycEAkFpaan0sLUm7bRRLTx1ZRs2bFheXp5QKMzNzfXz80tISDhw4IBkBnbpEzszMzM0NDQyMpLD4Uh/sV3OEZFFRUXR0dFv374VCAQpKSk+Pj6WlpZLly5tjYNSCGW+CgwYMODIkSMPHjwYM2bMH3/88e7dO4FA8M8//0RGRi5cuJCay7ipwaRpaWmtW7fu2rVrK1asKCoqEolE79+/z8zMbLRg16xZIxQK37x5M2HCBOkfUbNPfhlacoB1Sf/88/Ly1Pgq08JCpoUyF6NsGnW5afTUatIZvnr16q5du3p7e+fn57969SogIKCyslLOjniVu9wAACheQ9/QAQAANdPU561XV1f3799fJBJJL6ypqenfv39NTY30QpFIFBYW1r17dw6HY2Rk5OrqmpWVRa0KCQmhnsNjYWHxn//8RywW5+bmOjg4GBkZsVisLl26rF+/XigUisXiu3fvdu3aVVtbe/To0c+fP5ed7eLFi56enjY2NlpaWlwut2fPnps3b66qqqLW7tq1S0dHhxBiZWWVlJQUHBxMff2zU6dOR44ciY6O7tSpEyHEyMjo6NGjsvM3ulY69urVq6mD7d69e05OzuHDh6l5LczNzR88eLB3714qFbV2//791HPMunbt+vjxY7FY/PHjx4CAAEtLSzabbWJi4ubmlpGRUe+Pq6GfYRu4evXqsGHDtLS0TE1N165dK/mZi8XijRs36uvrX7hwgXqZkpIyatQoyayUnTt3dnBwuHr1qmT71atXz507t969vHz5cuXKlba2tlpaWnp6epJHCDaKEBITE9OC45O3nePHj1MzFXTo0OGrr76iFvr7+9+4cYP6/w0bNlAzdTKZTDs7u6SkJHEDv2KxWBwQEGBsbGxoaMjn88PDwwkhNjY2gYGBsk+YRpWVlY0dO5Ya2MhkMm1tbbdv3y5ZW2+Y6dOnMxgMY2Nj6kBWrVpFjSU0MDC4ffu2uE6dvnr1ytHRkcfjWVtbL1++fO3atYQQW1vb/Pz8urUv4yfQ2mSfjZMmTTI0NGSz2UZGRp9++ik1Dk5C+sROT0+v9+N0WFhY3Z1KD7ijrFmzxsbGRldXl81mm5ubL1q0qLi4WM5DaLNzW5oyXwUo+fn5fn5+9vb2enp6LBbL0NBw4MCBCxcuTE5ObjRYo3+TxWJxeHi4vb09j8fj8XgDBw7cu3evuIGCzc/Pl6RydHT8+eefa0VtxsmfnJzco0cP6hwzNTXl8/l1t5FxgHv27KH+Cuno6Li4uDR6vNI//9TUVDW+yjSpkGuJiYlRyB10U9tR8mLE5YYiz6kl/xkuFosLCgo+++wzIyMjLS2tYcOGnTt3rt79KvZy09T7BQAA5cQQi8X1/lEGAAA1w+fzCSFxcXF0BwFoLQwGIyYmxsPDQ0naAVAUnNsAyiA2NtbT07Pld9CKagdA4XC/AADqAfOZAAAAAAAAAAAAAIA6QH83AAAol0ePHjEa5uXlRXdAAJqhRkC9acgZriGHCSoNZykAAKgoNt0BAAAA/kevXr3wDV8AGVAjoN405AzXkMMElYazFAAAVBTGdwMAAAAAAAAAAACAOkB/NwAAAAAAAAAAAACoA/R3AwAAAAAAAAAAAIA6QH83AABokD/++MPAwOD06dMK31ge169fHzVqlI6OjqmpaUBAwMePH5u38ZYtW+zs7Nq1a6elpWVra+vv719eXi5ZO378+LoPldLT01PUUYBKU6ESkKiqqurVq9eGDRskSwQCwaZNm7p168blcs3MzPz8/CorKyVrUQKgQKpVMiKRaOfOnQ4ODjK2qVtQsq8pAM2jNrXT0CoUDgCAMkN/NwAAaJAmPXZJsc9oysjImDx58ieffFJaWnr8+PGDBw8uXbq0eRsnJiZ+9dVXubm5L1++DAoK2rVrF5/Pl7330aNHK+xIQJWpSglIW79+fVZWlvSSlStXhoWFBQUFvXr16siRI5GRkT4+PrIbQQlA86hQyWRnZ48dO3b16tUVFRUyNqtbUM24pgA0Sj1qR8YqFA4AgFITAwCAZnB3d3d3d6c7heby9PS0trYWiUTUy7CwMAaD8fDhw2Zs/OmnnwqFQsnGHh4ehJD8/HzqpZOTU1lZmXRrvr6+ly9fVuzhKCdCSExMjPK0A9KaVAISycnJkydPJoSsX7+eWpKTk8NkMhcvXizZhhqpmpmZSb1UyxLAua2BmlQyaWlps2bNOnz48IABA/r3799Qm3ULStzYNQWkxcTEKOQOWlHtQL0UVTuyy0pdCwf3CwCgHjC+GwAA1JlYLI6Li9u/fz+9MYRC4dmzZ8eNG8dgMKglzs7OYrH45MmTzdj4zJkzLBZLsn2HDh0IIZKRR+fPn9fX15esLSgoePDgwYQJE1rhsEAFqGIJSFRWVq5du3bXrl3SC2/duiUSiYYPHy5ZMmXKFELIhQsXqJcoAWgJFS2Z/v37x8fHz5kzR0tLq6E26y0o0tg1BUBO6lc7sssKhQMAoMzQ3w0AAGqlpqYmKCioZ8+e2traHTp0sLa2DgoKogbdXL9+3dLSksFghIeHE0L27dunq6uro6Nz8uRJZ2fndu3amZubHz16lGqn1sa1lJeX150gmDJixIi62z99+rS8vNzS0lKyxMbGhhDy999/t3BjQkhRUZG2tra1tXW9a4ODg7/++uv6f1igjtSgBCTWr1//5ZdfmpiYSC9kMpmEEG1tbcmS7t27E0IePnxYbyMoAZBNnUpGtnoLqi7Z1xQACc2pHXmgcAAAlAr6uwEAQK2EhIRs2rQpLCzs9evXFy9erKqqMjQ0NDQ0JISMHj36xo0bki2XLVu2atWqyspKfX39mJiYnJycbt26LVq0SCAQ1N24Fj09vYa+OXXz5s262z9//pwQIj3mlMfjaWtrl5SUtHDjioqKxMTERYsWcbncumuLioquXLni5ubW0IGA+lGDEqAkJyfn5OTMnj271vJevXqR/+3dbt++PSGktLS0biMoAWiU2pSMbA0VVC2yrykA0jSkduSBwgEAUDbo7wYAALVy4sSJwYMHu7i4aGtrDxo0aMaMGdeuXauurpbxFgcHh3bt2pmYmHh5eX348CE/P1/hqT5+/EgIkf7eKyGEw+FUVla2cOOgoCBTU9Nt27bVu9/g4ODly5dT42FBQ6hBCRBCKisrV65cuW/fvrqr7O3tp0yZsnfv3sTExKqqqufPnx8/fpzBYFD9JrWgBKBR6lEysskoqFpkX1MApGlC7cgJhQMAoGzw6R8AANRKVVWVWCyWvKypqeFwOLVuexpCjcqpt9eshXg8HiFEKBRKL6yurpaek6EZGx8/fjw2NvbChQvS45gkiouLT5065e3t3bLsoGLUoAQIIevWrVu8eLGZmVm9a6Ojo/l8/rx584yNjUeNGvX777+LxWJqlLc0lADIQz1KRjbZBSUh+5oCUIsm1I48UDgAAEoI/d0AAKBWpk6deufOnZMnT1ZWVt6+ffvEiRPTpk2T8+5Lfk2dTbJz586EkLKyMsmSioqKqqoqU1PTZm8cHR0dHBx85coVKyurekOGhIQsWrSIuvEDzaEGJXD9+vX09HQfH5+G9m5gYPDTTz8VFhZWVFTk5OR8//33hJAuXbrU2gwlAPJQg5KRrdGCojR6TQGoRe1rRx4oHAAA5cSmOwAAAIAibd68+c6dO97e3uXl5aamph4eHtu3b1f4XqjZJOXf3traWl9fPy8vT7LkyZMnhJB+/fo1b+M9e/ZcuHAhMTFRT0+v3j0+f/48KioqKytL/pCgHtSgBA4cOHD58uVak5Bs3759+/btt27dGjJkSK3tb926RQhxdHSUXogSADmpQcnIJk9BNXpNAahL7WunUSgcAAClhf5uAABQKxkZGTk5OaWlpWy2El3j2Gz21KlTr127JhKJqE6Hc+fOMRgMFxeXpm4sFosDAwPfvHlz4sQJGccYEhIyd+5cY2PjVjsmUFJqUAKHDh06dOiQ5OXLly9NTEzWr1/f0NSokZGR1tbW48aNk16IEgA5qUHJyCa7oOS8pgDUpfa1IwMKBwBAyWE+EwAAUCtfffWVpaVleXk53UFq27hxY0lJybfffvvhw4eUlJSwsDBvb++ePXtSazdt2mRgYHDx4sVGN87MzAwNDY2MjORwONLf6t2xY4dkXyUlJQcPHly1alXbHybQTj1KQLZhw4bl5eUJhcLc3Fw/P7+EhIQDBw5Qs8FSUAIgP00oGRnkuaYA1EuTaweFAwCg5NDfDQAAaiUoKOjBgwdGRkbUjQeXy7Wzszt+/DghJDw8fOjQoYSQgICAGTNm7Nu3b+fOnYSQfv36PX36NDIycs2aNYSQKVOmZGdn19q45cH69Olz4cKFixcvtm/f3s3N7Ysvvvjxxx+lN5D+uq6MjeX5Vm9oaKiLi4ulpWXLY4PKUY8SkM3Q0HDAgAHa2tqDBg169OhRUlJSrclMUAIgP/UomZs3b44ePbpLly6pqan37983NTUdNWrUtWvXGt1Lk2aKAJCm9rUjYxUKBwBAyTHwlxoAQEPw+XxCSFxcHN1BWte+ffuys7Op2ypCSHV1dWBg4L59+968eaOtrU1vNmhtDAYjJibGw8NDSdqhBUpALeHcbj0oGZBfbGysp6dny++gFdUOvVA7aklD7hcAQO1hqikAAFAfz58/X7FiRVpammQJl8u1tLQUCAQCgQB3X6D2UAIATYKSAWge1A4AACgzzGcCAADqQ1tbm8PhHDhwoKSkRCAQFBcX//zzz5s2bfLy8mrXrh3d6QBaHUoAoElQMgDNg9oBAABlhv5uAABQH9QDiB48eNCjRw9tbW07O7tDhw4FBwf/+uuvdEcDaAsoAYAmQckANA9qBwAAlBnmMwEAALUyZsyYS5cu0Z0CgDYoAYAmQckANA9qBwAAlBbGdwMAAAAAAAAAAACAOkB/NwAAAAAAAAAAAACoA/R3AwAAyGvHjh0dO3ZkMBg//fRT2+99y5YtdnZ27dq109LSsrW19ff3Ly8vl3MtRSQS7dy508HBoW2DgzqjtygoDZ3YISEhvXr10tbW1tXV7dWr18aNG8vKymhJCCCBkgGQhzJXyrZt2xj/q2/fvnK+FwAA2gb6uwEAAOTl5+d348YNuvaemJj41Vdf5ebmvnz5MigoaNeuXXw+X861hJDs7OyxY8euXr26oqKizbOD2qK3KIjMEzspKWnRokX5+fklJSVbt24NCQlxd3enJSSABEoGQB7KXCmt+l4AAFAIPK8SAABANejp6fn6+rJYLEKIh4dHfHx8bGxsQUGBhYVFo2vv37+/ZcuWpUuXfvjwQSwW03sgAIoi+8Tmcrlffvklj8cjhPD5/Li4uLi4uGfPnpmamtIRFoB+KBkAeTT6qek///nP3Llzm/deAABoAxjfDQAAoBrOnDlDdWdTOnToQAiRDB2SvbZ///7x8fFz5szR0tJqu8QArUz2iX38+HGq545iZmZGCKk7zw+A5kDJAMijJZ+a8IkLAEAZoL8bAABU0tWrV4cNG6ajo9OuXTt7e3tqjtGkpCQ7OzsDAwMej2dvb3/hwgVCyK5du3R1dZlM5uDBgzt16sThcHR1dQcNGjRmzBgLCwsej2doaOjv7081+8MPP/B4vI4dOy5ZssTU1JTH4zk4OKSmpjYUo6amZtOmTZaWltra2v369YuJiZERT1p5eTmjASNGjJDnJ1BUVKStrW1tbd2MtaCWUBSyZWdnGxoadu3ateVNgXpAyciGkgEKKgUAAFQO+rsBAED1fPjwwcXFxd3d/fXr19nZ2T169KiuriaElJSUeHp65ubmFhcX6+npzZkzhxCycuXKtWvXisXiH3/88Z9//nn+/PnYsWPvZe8XpwAAIABJREFU3bv3zTff3Lt37/Xr1/Pnzw8LC7t//z4hZMWKFd7e3hUVFV9//XVubu7du3eFQuGkSZMKCgrqTRIYGBgaGrpz585nz55Nnz599uzZt2/fbiieND09PXEDbt682ehPoKKiIjExcdGiRVwut6lrQS2hKBoiEAiKiorCw8MTEhL27NmDogAKSqYhKBmQhkqp1zfffGNkZMTlcq2trWfOnHnr1q3mtQMAAK0E/d0AAKB6cnNzy8rK+vTpw+PxOnXqFB8fT03f4e7u/u233xoZGRkbG7u4uLx69aq0tFTyLjs7Ox0dnfbt23/22WeEEEtLyw4dOujo6FAzMD569EiyJZvN7t27t5aWlp2d3b59+96/f3/o0KG6Maqqqvbt2+fq6urm5mZoaLhhwwYOh3Po0KGG4ilQUFCQqanptm3bmrEW1BKKoiEWFhbm5uabN28ODQ319PRsm52C8kPJNAQlA9JQKXXNnz//1KlTBQUF5eXlR48ezc/PHzduXEZGRmvvFwAA5If+bgAAUD3dunXr2LHj3LlzN2/enJubW+82HA6HEFJTU1N3FTVaTSgUSm8pEAjqbWfIkCE6OjrS92YSWVlZFRUVffv2pV5qa2t37tz50aNH8sRriePHj8fGxl64cEFfX7+pa0FdaXhRyFBQUPDixYuoqKhff/114MCBL168aMu9g9JCyTQEJQPSUCl1WVhYDBw4UE9Pj8vljhgx4tChQ5WVlXv37m2bvQMAgDzQ3w0AAKpHW1s7MTFx9OjR27dv79atm5eXV2VlJSHk7Nmz48ePNzEx0dLSkkwQ2XJaWlrSo5YkPnz4QAjZsGGDZCLIvLy8ioqKhuJJa/ZsktHR0cHBwVeuXLGysmrqWlBjmlwUsnE4HBMTk8mTJ0dHR2dkZAQFBTW7KVAnKJmGoGRAGiqlUfb29iwW6/Hjxy1vCgAAFAX93QAAoJL69Olz+vTp4uLigICAmJiYHTt25Ofnu7q6du7cOTU19d27dyEhIQrZkUAgePv2rbm5ed1VJiYmhJCdO3dKzwWZkpJSb7xa723ebJJ79uw5fPhwYmJily5dmroW1J5mFoX8bG1tWSwWvnIOEigZ2VAyQEGlyCYSiUQikZaWVsubAgAARUF/NwAAqJ7i4uLMzExCiImJyb/+9a9BgwZlZmamp6cLBIJly5Z169aNx+MxGAyF7OvKlStisbjeEUAWFhY8Hi8tLU2eeC2MIRaLAwIC0tPTT5w4oaen16S1oAk0sChke/Xq1ezZs6WXZGdn19TUWFhYtOp+QVWgZGpByUC9UCl1OTk5Sb+8deuWWCweOXJka+8XAADkh/5uAABQPcXFxUuWLHn06FF1dfW9e/fy8vJGjBhhaWlJCElISKiqqsrOzk5NTW12+yKR6M2bN0Kh8O+//165cqWlpaW3t3fdzXg83oIFC44ePbpv376ysrKamprCwsJnz57VG6/ZYSiZmZmhoaGRkZEcDkf6q7jUOCbZa0ETaGBRyKarq3vx4sXExMSysjKBQHDv3r358+fr6uquXr26VfcLqgIlUwtKBuqFSqmrqKgoOjr67du3AoEgJSXFx8fH0tJy6dKlrb1fAABogoa+2gMAAGrG3d3d3d2d7hSKkZub6+DgYGRkxGKxunTpsn79eqFQKBaLAwICjI2NDQ0N+Xx+eHg4IcTGxmbNmjU6OjqEECsrq6SkpODgYAMDA0JIp06djhw5Eh0d3alTJ0KIkZHR0aNHxWKxr68vh8MxMzNjs9nt2rWbOXNmTk4Otd/vv/+e2lhXV3fWrFlisfjjx48BAQGWlpZsNtvExMTNzS0jI6OheC2Rnp5e73U8LCys0bVisTglJWXUqFGmpqbU8s6dOzs4OFy9erWFqZQNISQmJkZ52mlLGlgU4sZObBcXF2traz09PS0tLRsbGy8vr/T09JbvlBaafG63EpSMepdMK4mJiVHIHbSi2mkDqJS6lbJmzRobGxtdXV02m21ubr5o0aLi4mI536v81Ol+AQA0GUMsFjetgxwAAFQTn88nhMTFxdEdRNktWbIkLi7u1atXdAeBJmMwGDExMR4eHkrSjtpAUdAO57ZqQcmoq9jYWE9Pz5bfQSuqHVWHSlFCuF8AAPWA+UwAAABqq6mpoTsCgHJBUQA0CUoGQB6oFAAAaA3o7wYAAAAAAAAAAAAAdYD+bgAAgP9at27doUOH3r17Z21tfezYMbrjANAPRQHQJCgZAHmgUgAAoPWw6Q4AAACgRIKCgoKCguhOAaBEUBQATYKSAZAHKgUAAFoPxncDAAAAAAAAAAAAgDpAfzcAAAAAAAAAAAAAqAP0dwMAgHrasWNHx44dGQzGTz/9RGOMJUuWMP6/uXPnEkK2bNliZ2fXrl07LS0tW1tbf3//8vLyWu8SiUQ7d+50cHCo22BUVNTQoUP19fW7du26YMGC58+fy5kkJCSkV69e2traurq6vXr12rhxY1lZGbXqxIkTkpAdOnRoweGCylDaAiGECASCoKAgW1tbLpdraGjYt2/f3NxceVobP348ow49PT2CkxyaRZnLhDR8pZDnKiNbvS2jiKBeSlsmAoFg06ZN3bp143K5ZmZmfn5+lZWVTWpTxocxiaqqql69em3YsEGyREYBoogAANoM+rsBAEA9+fn53bhxg+4UhBBibGx87ty5rKysAwcOEEISExO/+uqr3Nzcly9fBgUF7dq1i8/nS2+fnZ09duzY1atXV1RU1GoqJiZmzpw5fD6/sLDw5MmT165dc3Z2FgqF8sRISkpatGhRfn5+SUnJ1q1bQ0JC3N3dqVUzZswoLCy8du3a1KlTFXHEoAKUtkAIIZ6enr/99tuRI0cqKioePnxoY2PT1N46aaNHjyY4yaFZlLlMZFwpGr3KyNZQyygiqJfSlsnKlSvDwsKCgoJevXp15MiRyMhIHx8f+VuTUWLS1q9fn5WVJb1ERgGiiAAA2gz6uwEAAFqXtrb2lClTevTooaWlRQjR09Pz9fU1NjbW19f38PBwdXU9f/58QUEBtfH9+/cDAwOXLl06YMCAuk1FRER06dJl7dq1BgYGAwYMWL16dVpaWmpqqjwxuFzul19+aWJioqenx+fzZ86ceenSpWfPnhFCGAyGmZnZmDFjunfvrrjjBpBLrQKJjo4+ceJEXFzc8OHD2Wy2qanpyZMn+/btK09TPB6vrKxMLMXX19ff35/gJAcVV6tMZF8pZF9lZJPRMooIlJx0mTx9+vSnn36aN2+el5eXvr7++PHjV6xYERUV9fDhQ3makl1iEjdu3Hjw4EGthTIKEEUEANBm0N8NAADQps6cOcNisSQvqS+0SkYP9e/fPz4+fs6cOVSnRi0FBQWmpqYMBoN6aWFhQQjJy8uTZ7/Hjx/n8XiSl2ZmZoSQlgybBWgNP/7446BBg+zt7Zvx3vPnz+vr60teFhQUPHjwYMKECYpLB6AUZF8pZF9lWtIygKq4deuWSCQaPny4ZMmUKVMIIRcuXJDn7fIUQmVl5dq1a3ft2lVreUsKEAAAFAX93QAAoOx69+7NYDCYTObgwYOpGwZ/f38DAwMej/fLL78QQpKSkuzs7Kgl9vb29d7MrFixgsvldu7cmXr55Zdf6urqMhiMly9fUktqamo2bdpkaWmpra3dr1+/mJiYuo2Ul5fXnR2YMmLEiOYdXVFRkba2trW1tTwbd+vW7cWLF5KX1OTd3bp1a8Z+s7OzDQ0Nu3bt2oz3glJRpwKprq6+efOm7PF08gsODv76668V0hSoOnUqk6Zq0lUGNJk6lQmTySSEaGtrS5ZQQ6rlHN8tj/Xr11Nfm5O9GQoQAIAW6O8GAABl9+DBAysrKwsLi7/++ktHR4cQEhoaunDhwuDgYG9vb0JISUmJp6dnbm5ucXGxnp7enDlz6jbyww8/eHh4SF7u3bv3u+++k94gMDAwNDR0586dz549mz59+uzZs2/fvl2rET09PXEDbt682YxDq6ioSExMXLRoEZfLlWf7devWPX/+fM+ePe/fv8/IyNi1a5eTk1OTukgEAkFRUVF4eHhCQsKePXvk3C8oM3UqkOLi4urq6jt37jg6OpqamvJ4vN69e+/du1csFjfxp0KKioquXLni5ubW1DeCWlKnMmmSpl5lQJOpU5n06tWL/G/vdvv27QkhpaWl8v0wGpGcnJyTkzN79mzZm6EAAQDogv5uAABQdiwW6+uvv87Pzz9+/Di1pKKiIj4+/osvvqBeuru7f/vtt0ZGRsbGxi4uLq9evWrq/UxVVdW+fftcXV3d3NwMDQ03bNjA4XAOHTqk4COpIygoyNTUdNu2bXJuP27cuICAgBUrVrRr165v377v37//+eefm7RHCwsLc3PzzZs3h4aGenp6Nj0yKB11KhBqgh0TE5Pt27dnZGSUlJTMnDnzq6++ioqKampTwcHBy5cvp4b4AahTmTRJU68yoMnUqUzs7e2nTJmyd+/exMTEqqqq58+fHz9+nMFgCASCljdeWVm5cuXKffv2NbolChAAgC64BwAAABXg4+NjYGAgmSTx8OHDM2fObNeuXd0tORwOIaSmpqZJ7WdlZVVUVEieiaetrd25c+dHjx61LHUjjh8/Hhsbe+HCBekZh2Vbv379/v37L1++XF5e/vTpUwcHh5EjR8r5FDJKQUHBixcvoqKifv3114EDB0rPjgKqS20KhJoptU+fPg4ODsbGxgYGBt99952BgcH+/fub1E5xcfGpU6eoAYkAFLUpE/k14yoDGk6dyiQ6OprP58+bN8/Y2HjUqFG///67WCymRnm30Lp16xYvXkw9B0UGFCAAAI3Q3w0AACpAT09v8eLFN27c+OuvvwghP/7444oVKyRrz549O378eBMTEy0tLX9//2a0/+HDB0LIhg0bJBNE5uXl1X24kALnXY2Ojg4ODr5y5YqVlZWcb3n27FlISMjixYsnTJigq6trbW0dGRlZXFwcFhYm/345HI6JicnkyZOjo6MzMjKCgoKaFBuUk9oUiKmpKSFEMs0rIYTL5Xbt2jUnJ6dJgUNCQhYtWiT9gFYAtSkTOTXjKgOgTmViYGDw008/FRYWVlRU5OTkfP/994SQLl26NCO2tOvXr6enp/v4+MjeDAUIAEAv9HcDAIBqWLFiBYfD2blz57Vr1ywsLGxsbKjl+fn5rq6unTt3Tk1NfffuXUhISDMapx43tHPnTuk5IlNSUmptpqh5V/fs2XP48OHExMQm3XdlZ2fX1NRIv6Vdu3bGxsYZGRnyNyJha2vLYrGa915QQupRIHp6et27d8/MzJReKBQKDQwM5E/7/PnzqKioZcuWyf8W0BDqUSbyaN5VBoCob5ncunWLEOLo6Ni8t0scOHDg8uXLTCaT6n+njmj79u0MBkMyETkKEACAdmy6AwAAAMjF3Nzcw8MjJiamuLj422+/lSxPT08XCATLli3r1q0bIYTBYDTUApvNbmjeRgsLCx6Pl5aWpvDYtYjF4sDAwDdv3pw4cYLNbtpV2NzcnBDy7NkzyZL379+/fv3awsKi0fe+evVq+fLl0pMgU73n8rwXVIJ6FAghxNPTc/v27U+fPqUCV1RU5OXlTZs2Tf4WQkJC5s6da2xs3GoZQVWpTZnI0JKrDABR3zKJjIy0trYeN25cC9s5dOiQ9ITjL1++NDExWb9+PTVJNwoQAEBJYHw3AACojDVr1giFwjdv3kyYMEGy0NLSkhCSkJBQVVWVnZ2dmpra0NttbW1fv3594sQJgUBQWlqal5cnWcXj8RYsWHD06NF9+/aVlZXV1NQUFhZK9ywrSmZmZmhoaGRkJIfDkf5+7o4dOxp9r7W1taOjY2Rk5LVr1yorKwsKCnx9fQkhCxcupDbw8vLq1KnT3bt3675XV1f34sWLiYmJZWVlAoHg3r178+fP19XVXb16tWIPEGikBgVCCFm9enXXrl29vb3z8/NfvXoVEBBQWVkZGBhIrZVxklNKSkoOHjy4atWq1sgGakA9ykSGRq8yjRYRgHqUybBhw/Ly8oRCYW5urp+fX0JCwoEDB7hcLrW2lQqhJR/zAABAgdDfDQAAKmPgwIGOjo5ff/219EJ7e/uAgIC9e/eampquX79+/PjxhJDRo0evWrVq9OjRhBA/Pz83NzdCyLJlyxwdHT/77LOePXtu3bpVW1ubECJ53uOuXbtWrVoVEhLSvn17U1PTlStXvnnzRuGHIBaLZW9w8+bN0aNHd+nSJTU19f79+6ampqNGjbp27RohhMFgxMXFeXl5LVy40MjIyM7OLj8/Pz4+fsyYMdR7q6urX7x4cfLkybrN8ni8UaNG+fj4mJmZ6evr8/l8KyurmzdvSp4ZBWpADQqEEGJkZJSUlGRubj5gwAAzM7O//vrr7NmzAwYMoNbKOMkpoaGhLi4uVL8MQF3qUSYyrhSNXmVkF5GMlkFzqEeZGBoaDhgwQFtbe9CgQY8ePUpKSpKezKSVCqHRAgQAgLbBwF9kAAANwefzCSFxcXF0B9EsS5YsOXPmTGFhYRvsSyQSjR8/3tvb+4svvmheCytXrjx8+LD00wJVC4PBiImJ8fDwUJJ2oFFNLRCNPclxbmsyxV5HNLaIFCI2NtbT07Pld9CKagck2v5q0hLKXES4XwAA9YDx3QAAAK2rsrLywoUL2dnZ1dXVrbeXmpqaEydOvH//3svLq6nvFYvFxcXF169ff/LkSWtkA5BB/gLBSQ4aS1HXERQRqLG2uZq0BIoIAKDN4BEKAAAArev169dTpkwhhMyZM+fw4cOttJcrV67Ex8efO3dOR0enqe89efKkq6sr9f/t27dXdDQAWeQvEJzkoLEUdR1BEYEaa5urSUugiAAA2gzmMwEA0BT4fiKoPcz5AOoK5zaAMsB8JqD2cL8AAOoB85kAAAAAAAAAAAAAgDpAfzcAAAAAAAAAAAAAqAP0dwMAAAAAAAAAAACAOkB/NwAAAAAAAAAAAACoA/R3AwAAAAAAAAAAAIA6YNMdAAAA2k5hYWFsbCzdKVqkpqaGxWLRnQLUX0pKCt0RAFoFzm2AZlNs+aj6RzJQHiKRqKamhsPhtLypwsJCc3PzlrcDAEAv9HcDAGiQmzdvenp60p0CQAXs2rVr165ddKcAUDyc2wBKAh/JQDm5u7vTHQEAoKUYYrGY7gwAAACNKCsr8/b2Pn369LZt2wICAuiOIy8LC4tVq1atXr2a7iAA/1VYWDhw4MBJkyZFRUXRneV//Pzzz8uXLy8pKWnXrh3dWQCa5t27d9bW1itXrty0aRPdWeoxbdq03Nzce/fuKWT4J4CyYTAYMTExHh4edAehTVFRUWhoaEREhLW19ZYtW9zd3RkMBt2hAADohPm7AQBA2T169GjEiBGpqalXr15Voc5uQgiTyRSJRHSnAPgvoVDo5eVlbGwcERFBd5ba3NzcRCLR6dOn6Q4C0GQ7d+4UiUQrVqygO0j9wsPDc3Nzd+zYQXcQAGgVZmZmu3fvfvTo0dixYz/77LP+/fvHxcXRHQoAgE7o7wYAAKUWHR09ZMiQ9u3b375928HBge44TcNisWpqauhOAfBf69atu3PnTmxsrL6+Pt1ZajMyMpowYQJu0UHlvHv37ocffli9erWhoSHdWepnZWW1bt26rVu35uTk0J0FAFqLlZVVRETE/fv3e/Xq5eHhMXr06GvXrtEdCgCAHujvBgAAJSUUCgMDAz/77LM5c+YkJiaamprSnajJML4blMoff/yxY8eOH3/8sX///nRnqZ+Hh8f58+ffvXtHdxCAJlDywd0Uf3//Hj16LFu2jO4gANC6+vTpExsbe+PGDQ6HM27cuEmTJqWlpdEdCgCgraG/GwAAlNHLly+nTJmye/fugwcPRkREqOiUoxjfDcqjsLBw/vz5Xl5e3t7edGdpkKurq1gsPnXqFN1BAOSl/IO7KWw2OyIiIiEh4ejRo3RnAYBWN3LkyD///PPSpUuvX78ePHiwh4cHvt4BABoF/d0AAKB07ty5M2TIkOzs7KSkpAULFtAdp/kwvhuUhDJP2y3N0NBw0qRJmNIEVIhKDO6mDB8+3MfHZ9WqVW/evKE7CwC0hYkTJ96+fTs6OjotLa13796+vr7Pnj2jOxQAQFtAfzcAACiX3377bcyYMXZ2dmlpaUOGDKE7TouwWCz0d4MyUOZpu2vh8/kXLlx4/fo13UEAGqcqg7slQkNDWSzWN998Q3cQAGgjDAaDz+dnZGSEh4efPXvW1tY2MDDw7du3dOcCAGhd6O8GAABl8fHjx0WLFnl7e69YseLMmTNGRkZ0J2opJpOJ+UyAdso/bbc0V1dXJpOJKU1AJajQ4G6KgYFBWFhYZGRkcnIy3VkAoO1wOJzFixc/ffp0586dhw4dsrGx2bx58/v37+nOBQDQWtDfDQAASqGwsHDcuHGxsbHx8fHBwcFMpjpcoTC+G2inEtN2S2vXrh2mNAGVoHKDuymzZ892dnb29fUVCAR0ZwGANsXlchcvXpyTk+Pv779z504bG5uQkJCPHz/SnQsAQPHUoTcBAABU3dWrV4cMGfLu3bubN2+6urrSHUdhML4b6KUq03bX4uHhQT1ii+4gALKo3OBuifDw8Nzc3B07dtAdBABooKenFxAQkJOT88UXX2zevLlnz5779+/H51UAUDPo7wYAADqJxeLdu3dPnDjR0dHx9u3bvXv3pjuRImF8N9BLhabtljZjxgw2m33ixAm6gwA0SEUHd1OsrKzWrVu3devWnJwcurMAAD06dOgQHBz8+PFjJyenZcuW9evXLy4uTiwW050LAEAx0N8NAAC0KS8v9/T09PPz27ZtW1RUlK6uLt2JFAzju4FGqjVttzR9fX0nJydMaQLKTHUHd1P8/f179OixbNkyuoMAAJ0sLCwiIiLS09P79Onj6enp4ODw559/0h0KAEAB0N8NAAD0ePz48YgRI/7888/z588HBAQwGAy6EykexncDXVRu2u5a+Hx+QkJCaWkp3UEA6qHSg7spbDY7IiIiISHh6NGjdGcBAJr17t07Njb25s2burq6EyZMmDRp0p07d+gOBQDQIujvBgAAGpw+fXr48OE8Hu/27duffPIJ3XFaC8Z3Ay1UdNpuaS4uLhwOB1OagHJS9cHdlOHDh/v4+KxaterNmzd0ZwEA+g0bNiwhISEpKamqqmro0KEeHh6PHz+mOxQAQDOhvxsAANqUWCwOCQmZOXPm9OnTk5KSunbtSneiVoTx3UALFZ22W5qenp6zszOmNAElpAaDuyVCQ0NZLNY333xDdxAAUBajR49OSkq6ePFiVlZW7969PTw8/vnnH7pDAQA0Gfq7AQCg7bx69WrKlCnffvttRETEb7/9pq2tTXei1oXx3dD2VHfa7lo8PDwSExNfvHhBdxCA/6Eeg7spBgYGYWFhkZGRycnJdGcBACUyceLEe/fuRUdH37lzp1evXr6+viUlJXSHAgBoAvR3AwBAG0lLSxs6dGhmZubVq1d9fHzojtMWML4b2piqT9stbfr06Twe7/fff6c7CMB/qdPgbsrs2bOdnZ19fX0FAgHdWQBAiTCZTD6f//Dhwz179pw6dcrW1jYwMPDdu3d05wIAkAv6uwEAoC0cOXJk1KhRlpaWt2/fHj58ON1x2giTyUR/N7QZNZi2W5qOjs7UqVNjY2PpDgLwX+o0uFsiPDw8Nzd3x44ddAcBAKXD5XIXL1785MmTDRs2RERE2NjYhISEVFVV0Z0LAKAR6O8GAIDWJRQKAwMDP//8cx8fn0uXLnXq1InuRG0H85lAW1KDabtr4fP5V65cefbsGd1BAAhRx8HdFCsrq3Xr1m3dujUnJ4fuLACgjHR1dQMCAnJycnx8fL777rsePXrs379fKBTSnQsAoEHo7wYAgFZUXFw8bty4vXv3xsTE7N69m8Ph0J2oTWE+E2gzajNtt7RPP/1UR0cHU5qAklDLwd0Uf3//Hj16LFu2jO4gAKC8jI2Ng4ODs7OzXV1dly9fbm9vHxcXJxaL6c4FAFAP9HcDAEBruX79+pAhQ0pLS1NSUvh8Pt1xaIDx3dA21Gnabmk6OjrTpk2Li4ujOwiA2g7uprDZ7IiIiISEhKNHj9KdBQCUmpmZ2e7du7OyssaOHevl5TV8+PDTp0/THQoAoDb0dwMAQKvYv3//hAkTBg8e/Ndff/Xt25fuOPTA+G5oA2o2bXctfD7/2rVrxcXFdAcBTafGg7spw4cP9/HxWbVq1Zs3b+jOAgDKzsrKKiIi4v79+1ZWVi4uLqNHj05KSqI7FADAf6G/GwAAFKyqqmrBggVLlixZvXr1yZMn1XIonJwwvhvagPpN2y1t6tSpenp6x48fpzsIaDT1HtwtERoaymKxvvnmG7qDAIBq6Nu3b2xs7I0bN9hs9tixYydNmnT//n26QwEAEIL+bgAAUKz8/PwxY8acPn363LlzwcHBTKZGX2gwvhtam1pO2y2Nx+NNmzYtNjaW7iCg0dR+cDfFwMAgLCwsMjIyOTmZ7iwAoDJGjhx55cqVS5cuvXr1atCgQR4eHnj4LQDQTqO7IQAAQLHOnTs3YMAAoVB469YtJycnuuPQD+O7oVWp67TdtfD5/OTk5KKiIrqDgIbSkMHdlNmzZzs7O/v6+goEArqzAIAqmThx4p07d6Kjo9PS0nr37u3r6/vs2TO6QwGA5kJ/NwAAKIBYLA4JCZk2bdrUqVOTk5Otra3pTqQUML4bWo96T9stzdnZWV9f/9ixY3QHAQ2lIYO7JcLDw3Nzc3fs2EF3EABQMQwGg8/nZ2RkhIeHnzlzpnv37oGBgW/fvqU7FwBoIvR3AwBAS5WVlbm5uW3YsCEoKOjw4cM6Ojp0J1IWGN8NrUe9p+2WpqWl5eLiEhcXR3cQ0EQaNbibYmVltW7duq1bt2JGAgBoBg6Hs3jx4uzs7O3btx86dMjGxiYkJKSyspLuXACgWdBPdFegAAAgAElEQVTfDQAALfLo0aMRI0akpqZevXo1ICCA7jjKBeO7oZWo/bTdtfD5/Bs3buTn59MdBDSOpg3upvj7+/fo0WPZsmV0BwEAVaWjo/P111/n5OT4+/sHBQV17do1JCTk48ePdOcCAE2B/m4AAGi+6OjoIUOGtG/f/vbt2w4ODnTHUToY3w2tQUOm7Zbm5ORkaGgYHx9PdxDQLBo4uJvCZrMjIiISEhKOHj1KdxYAUGF6enoBAQE5OTlffPHF5s2be/bsuX//fnw2BoA2gP5uAABoDqFQGBgYOHv27Dlz5ly+fNnU1JTuRMoI47tB4TRn2m5pXC7XxcUlNjaW7iCgWTRzcDdl+PDhPj4+q1atevPmDd1ZAEC1dejQITg4OCsry8nJadmyZf3798ccZQDQ2tDfDQAATfby5cspU6bs3r37wIEDERERXC6X7kRKCuO7QeE0Z9ruWjw8PFJTU3Nzc+kOAppCYwd3S4SGhrJYrHXr1tEdBADUgaWlZURERHp6up2dnaen58iRI//880+6QwGA2kJ/NwAANM2dO3eGDBmSnZ2dlJS0YMECuuMoNYzvBsXStGm7pU2aNMnIyOjYsWN0BwFNocmDuykGBgZhYWH79+9PTk6mOwsAqInevXvHxsbevHlTV1d3woQJkyZNunv3Lt2hAEANob8bAACa4LfffhszZoydnV1aWtqQIUPojqPsML4bFEgDp+2WxuFwZs6ciW9AQ9vA4G7K7NmznZ2dlyxZIhAI6M4CAOpj2LBhCQkJly5devv27ZAhQzw8PB4/fkx3KABQK+jvBgAAuXz8+HHRokXe3t4rVqw4c+aMkZER3YlUAMZ3g6Jo5rTdtfD5/L/++isnJ4fuIKD+MLhbIjw8/J9//vn+++/pDgLwXzNnzjSXoq+vv3z5cslLa2vrwsJCujNC4yZOnHjr1q2LFy9mZWX16dNn3rx5mLUMABQF/d0AAPA/NmzY8Pr161oLCwsLx40bFxsbGx8fHxwczGTi8iEXjO8GRdHYabulTZw40cTEJD4+nu4goOYwuFualZXVunXrtmzZgn9qAuUxYsSIIinv379/8eKF5GW7du3Mzc3pzgjymjhx4r1796KiopKTk3v27Onr6/vixQu6QwGAykOHBQAA/Nf58+e3b9/u5eUlPSr56tWrQ4YMeffu3c2bN11dXWmMp3KYTCbGd0PLafK03dLYbPaMGTNiY2PpDgJqDoO7a/H39+/Ro8eXX35JdxCA/zN79mwGg1HvKg6HM3/+/DbOAy3EZDL5fP7Dhw/37Nlz6tQpGxubwMDAsrIyGW95+fJlm8UDAFWE/m4AAPg/Hz588PHxYTKZly9f3rRpEyFELBbv3r174sSJjo6Ot2/f7t27N90ZVQzmM4GW0/Bpu2vx8PC4c+fOkydP6A4CaguDu+tis9kRERGXLl2Kjo6mOwsAIYRYWloOHTq03q8bCoVCDw+Pto8ELcflchcvXvzkyZMNGzZERETY2NiEhIRUVVXV3bKoqKhPnz540CUAyID+bgAA+D8bN24sKSkRiUQikSgoKOjIkSOenp5+fn7btm2LiorS1dWlO6DqwXwm0EKYtruWCRMmdOzYEU+thNaDwd31Gj58uI+Pz8qVK9+8eUN3FgBCCJk3b17dId5MJtPBwQGTmag0XV3dgICAnJychQsXfvfddz169Ni/f79QKJTeZsuWLS9evPjkk0+ysrLoygkASg793QAAQAghaWlpu3fvlv40+cUXX1y+fPnSpUsBAQENfWkUZMP4bmghTNtdC4vFcnV1RX83tBIM7pYhNDSUxWKtW7eO7iAAhBBS7yBuBoMxb968tg8DCmdsbBwcHPz48WNnZ+cvv/zS3t4+Li5OLBYTQrKzsw8cOEAIKS8vHzduXF5eHt1hAUAZob8bAACIUCicN2+e9NdCxWKxSCTS19cfNGgQjcFUHcZ3Q0tg2u568fn8e/fuPXr0iO4goIYwuFsGAwODsLCw/fv3Jycn050FgJiYmDg6OrJYrFrL3dzcaMkDrcHc3DwiIiI7O3vs2LFeXl7Dhw8/ffr0+vXrqXsWoVD4+vXr8ePHl5SU0J0UAJQO+rsBAIB8//33GRkZtb4qKBQKi4qK5s6dSw2mgGbA+G5oNkzb3ZDx48d36tTp2LFjdAcBdYPB3Y2aPXu2s7PzkiVLBAIB3VkAyOeffy79GZXFYjk5ObVv357GSNAarKysIiIibt261b59+xkzZhw7dkzyJ0ggEBQVFX3yySdv376lNyQAKBv0dwMAaLqcnJxvv/223m5ZoVB45syZ77//vu1TqQeM74bmwbTdMrBYrFmzZmFKE1A4DO6WR3h4+D///IMPBqAMXF1d2Wy25KVYLJ47dy6NeaBVDRo06Ny5cyNHjpT+pRNCBALB48ePnZycPnz4QFc2AFBC6O8GANBoYrHYx8dHxhhkBoMRGBh47dq1tkylNphMJsZ3Q6PqPv8N03bL5uHh8ffffz98+JB6ef/+/fXr19+9e5feVKBazp49e/r0acngUAzulpOVldW6deu2bNmSk5MjvfyPP/6gKxJoLH19/enTp3M4HOolh8OZPn06vZGgVV2/fv3GjRt1v18iEAju3r3L5/Px1RMAkGDgW+oAAJrs119/XbBgQd1rAZvNFgqFenp6M2fO9PDwcHZ2rjWYAup19+7d4OBgQkhVVVVZWVlhYWFpaenAgQOptQKBYMyYMdQGAJTc3Nzhw4cfOXJk4sSJ1JI//vhj2rRpBw8exEwmDRGJRBYWFjNnzmzfvv3hw4f/+ecfQsjly5cnTJhAdzRQGWvWrPn3v//dr1+/bdu2TZs27bvvvtu1a1dubi76uxslFAqHDBnSuXPn8+fPE0Ly8vKWLl167ty5vLw8S0tLutOBZvn999/d3NzEYjGbzXZ3dz969CjdiaAVjR49OjU1tdYEjBIsFovP5x85ckT6iUQAoLHQeQEAoLlevny5cuVK6SUcDkcoFHK53GnTps2fP9/JyYnL5dIVTxX17NnzzJkzlZWV0guvXr0q+f8lS5a0eShQar///ntpaenkyZM3bty4adOmZ8+eYdpu2TIzM2NjY8Vi8b59+7hcbnV1NbW87lPLAGTIyclhMBgPHjxwcXHp06dPbm7u2rVr0dktDzabHRER4eDgcOTIkefPn2/YsKGmpobJZN64cQP93dDGpk6dqqurW15eXlNTM2fOHLrjQCs6e/as7Ifl1tTUxMbGdujQYc+ePW2WCgCUFvq7AQA01/Lly9+/fy8Wi6nHKnK53OnTp1NPo+LxeHSnU0m6urpubm4xMTH1fqGSw+G4uLi0fSpQZrGxsYQQsVi8ffv2hIQEgUCAabvr9fLly3379kVFRWVlZUm6uSWd3QT93dBEjx8/FovF1NebHj58KBKJoqKi7Ozs3N3dGQwG3emU3fDhw93c3NauXVtSUkJN28XhcFJSUry8vOiOBppFS0vL3d39l19+0dfXnzx5Mt1xoBWlp6cPGDDgyZMn5eXlhBAWi8Vms6urq6W/pSoSifbu3du+ffvNmzfTFhQAlAPmMwGAlkpJSSkoKKA7BTRZWlrav/71L0IIi8Xq16/fmDFjBg8erNhubgcHB3NzcwU2qBIuXrzo5ORUdzmbzZ46derJkyfbPhIorRcvXpiamkomeWez2QwGIzw8fPHixfQGU0IfP34cPXr0nTt3GvrsmpKSMmLEiDZOBapLX1+f6jSRoJ64YG9vv23btunTp6PXuyEfPnzYunVrWFgYk8mUnligX79+9+/fpzGYmqH+NRQa9ffff2/fvv2TTz7BpVNOCvl8TuP5+e7du2fPnhUXF1P/LSgoePnyJfV8eDabXVNTIxaLvb29nZ2d6UoI0CgLC4uRI0fSnULNob8bAFqKz+cfO3aM7hSgjGJiYjw8POhO0dZEIlHnzp1LS0trLWcwGEePHvX09KQlFSinyMjIpUuXUjdpFCaTKRaL/f39t2/fjgHLtRQUFPTv3//du3f1Pgb2r7/+Gjp0aNunAlX09u1bIyOjelcxGAwGg3Hq1KlPP/20jVOphPj4+GXLlr1+/bruFLosFqusrExHR4eWYOoH/+ICrUQhn89xfgK0hLu7e1xcHN0p1Bwm8gcABXB3dxeDSikoKCgtLW3VXdB9VtKGyWR+/vnnHA6n1nIul4veE6glPj6+1hKRSCQWi3fs2OHo6Pj8+XNaUiktCwsLGf+8imfqgvyoZ5zWi8FgHDx4EH+u63XhwgUPD4/S0tJ6nxdXU1Nz69attk+lxmJiYlr1o5raCA4OFgqFdKdQDZpzflZVVdEdAaB+7u7uCqxEaAj6uwEANJG5uXmHDh3oTqG25s6dW2v+bjabPW3aND09PboigRIqKytLTEyUHtwtUVNTk5SUNH/+/HoHMmuyCRMmbN26lcms5xMshsOD/Ort72YwGEwm85dffpk/f37bR1IJTk5OZ8+e1dPTq/tvuoQQLpd748aNtk8F4Ofnh0sA1KKlpUV3BACgE/q7AQAAFGzgwIE9e/aUXlJTU/PZZ5/RlQeU09mzZ+sdI8lms5lMZkBAwKlTp+rt2NVw33zzzfTp0+t2t6GzA+SXm5vL5XKllzAYDBaLFRsb+/nnn9OVSiVMmTLl7t273bp1q/uNCqFQeP36dVpSgYbD338AAKgFN1EAAACK5+3tLd0XoKWlNWXKFBrzgBKKj4+ve4vOYrG6deuWmpoaHByMoUn1YjAY//nPfywtLWt1t6G/A+RXa3w31dkdFxfn5uZGVyQVYmtre/v2bRcXl1r/ICcSia5fvy7W4AnNAAAAQEmgvxsAAEDxPv/8c8k8FWw2e+bMmbq6uvRGAqXy8ePHP/74Q3p8NzWs28/P7++//x4yZAiN2ZSfvr7+77//zmKxpJ+Xhf5ukN/Tp0+rq6up/2cymVwu9+zZszNnzqQ3lQrR09M7duxYUFAQNQmMZHlZWVl2djaNwQAAAAAI+rsBAABag5mZ2ahRo6gOuJqaGi8vL7oTgXK5ePFiZWWl5CWbzba2tr558yaGdcvJ3t7+4MGD0iNJ0d8N8nv8+DH1P0wmk8PhnD17dvLkyfRGUjkMBiMgIODs2bO6urqSL1uwWCxM4Q0AAAC0Q383AABAq/D29qY647S1tZ2cnOiOA8rl+PHj1AzU1LDuNWvWpKenDx06lO5cqmT27NlLliyRdHOjvxvkV1RURAhhsVhaWlrnzp375JNP6E6kqpydnaWn82YymSkpKXSHAgAAAE2H/m4AoEFUVBSDwXBwcKArwP379728vKytrbW0tDp06NC/f/9t27ZRq+Lj47t168aoj5WVFS1pq6qqevXqtWHDBumFAoEgKCjI1taWy+UaGhr27ds3Nze33reHhIT06tVLW1tbV1e3V69eGzduLCsrq7WNSCTauXNn3d/Itm3bav0Q+vbtq7gjU3Pu7u7U/f+sWbN4PB7dcUCJ1NTUnDhxQiAQsFisrl273rhxA8O6m2f37t0DBgygplNAfzfIqbS0tLKyksFg6OjoXLlyxdHRke5Eqo2aztvZ2ZnJZAoEgitXrtCdSFMMGjSI+mxmamr69ddfN7TZ48ePhw4dqqenx2QyaX+UCO23AC0kEAg2bdrUrVs3LpdrZmbm5+cn/VUtQkhUVNTQoUP19fW7du26YMGC58+f12qhoY/ctYwfP77unYienh61ViU+nyvPLdWOHTs6duzIYDB++umnRjeuG5vL5Xbs2HH8+PFhYWFv3ryR3viPP/4wMDA4ffp0q2WXS1vGUImb6OvXr48aNUpHR8fU1DQgIODjx48t2bihmpXnFhvoVfux2gAAbSAqKsrGxiYlJeXJkye2trZtvPf09HQHB4dFixb9+eefnTt3zs/PP3DgwJUrV6gOZTc3Nzc3N1tb25cvX759+5YQUlNTU11d/f79+/Hjx7dxVMr69euzsrJqLfT09MzMzDxy5MjgwYNLS0uXLFlSXl5e79uTkpIWLVo0b948bW3tc+fOzZkzJzU19eLFi5INsrOzFyxYkJyc3L9//1Y8jOYqLCxU3S9HDxw4MDU11czMLDY2lu4szWRhYTFy5MiWt6PSv0eFy8zMfPv2LYPBmDp1qqenZ15eXl5eXqvuUSG/x5SUlIKCAoXkUaCFCxeuXbv2w4cPZ8+eNTIyojtOK3JwcDA3N1dIU6r7F0khnjx5QgjR1tbesGFDbm5uQ/9aTAuF/JZp+Xs7Z84cHo937Nix7OzsX375RUdHp40DKA9FXTcbdffuXScnp0uXLt27d69z586S5TU1NZMmTUpMTKRe9ujR49atWz/88MNff/11+PDhNggmA723AC23cuXKgwcPHjp06NNPP71z586MGTOePXt25MgRam1MTMycOXNCQkJ8fX3/+ecfNzc3Z2fnW7duSSb8aeFH7tGjRyvsSFqf8txS+fn5zZw5s3v37vJsXCu2WCx+9+5dWlrar7/+GhAQsHPnzlOnTkkesqIkj+dtsxgqcROdkZExefJkPz+/ixcv/v333y4uLqWlpQcPHmzexjJqttFbbKCfGACgZdzd3d3d3eXf/uXLl9bW1tQH7o0bN7ZesIbMmzevS5cu0ks+fvw4bdo06SU2NjYGBga13jhjxgz591JdXX3w4MF58+Y1OyclOTmZmlR0/fr1koVHjx5lMBh///23PC24urpWVlZKXvL5fEJIcXEx9TItLW3WrFmHDx8eMGBA//79a71369at//nPf5qXnBASExPTvPdKi4mJofdCqeGaVN34PSothfwe3d3d6T4OjaaQv6gUug8FGoTrphpQ1HWTyHE+UP0ykZGR0gsvXbpECMnIyJBe6OTkdOrUKYUEE4vFFRUVI0eObOq7aL8FELfs83lOTg6TyVy8eLFkCdXNl5mZSb10dHTs0qWLSCSiXoaHhxNCrl+/Tr2U/ZG7Ficnp7KyMuklvr6+ly9fpv5fGT6fy9lOy2+pZJPzVKSeo/vjjz/K2Wy9sePi4phMZseOHal+cA2kEjfRnp6e1tbWkjIMCwtjMBgPHz5sxsaya1b2LbZsTe0/gebBfCYA0NZiY2M//fRTFxcXHo9HfVZr4wCvXr169+7d69evJUu4XK48XwE7ceKEPO1//Pjxxx9/HDRoUEZGRnBwcPODElJZWbl27dpdu3bVWk61b29vL08jx48fl55Mw8zMjBAiGQzev3//+Pj4OXPmKPlcCnRfLpupurra39+f7hTNp/D+TboPSFns2rWroqKizXanwN+j0n46DwoKevXqFd0pWpGifoMSCuw9VzmHDh168OAB3SnqodhfMV1H8ejRo6NHj9K1d9q18b8Lzpo1i8vlnjp1SnrhxYsXu3TpcuzYMcmSysrK+/fvK/BRIgcOHHjx4kVT30XvLUDLP5/funVLJBINHz5csoSaH+bChQvUy4KCAlNTUwaDQb20sLAghEi+v9Wkj9znz5/X19eXvCwoKHjw4MGECROaEVsJyXlLJY/mnYrN4+7u7u3t/eLFC3nmRVFLyn8TLRQKz549O27cOEkZOjs7i8XikydPNmNj2TUr+xYblAH6uwGgrUVFRc2aNUtfX3/y5Mm5ublJSUnSa0NDQ3V0dPT19V+8eLFmzRozM7OsrKyamppNmzZZWlpqa2v369dPMnApKSnJzs7OwMCAx+PZ29tLPm7KNnTo0A8fPkyYMCE5OVmxh/bhw4d///vfQ4YMefHixdWrV3f8P/buPKCJa20Y+AkQkpAQFkWIshjADcW9qIBU6nVvqQgILm/FKi7UIqICFbdapSy20CLoFZXbXmSxSsFWuVjkVVzAuqAiiFIsBAhEBJUlLCHJ98d8X77cACEkgSTD8/uLzJw58yTPnOTMYebM8eMMBgMh1Nra2utcZgQCYe7cuVIqDAsL++KLL0xMTMQXdnV1FRYWTp8+Xb4gy8vLDQ0Nrays5NscDAiRSPz6669VHQVQOzt37qRQKKqOAldCQ0MNDAxUHQXQDBs2bJg8ebKqo8CtCRMm+Pj4qDqK4cLAwGDJkiW5ubmiWaR5PB5CyNPTU3y8+/r160uWLNHV1UUICYXC77//ftKkSSQSycjIaOXKlWVlZVixXjvhN2/edHBw0NPTo9Pp9vb2zc3NgYGBu3fvrqioIBAIA5qTRFWnAMrqn2PPihD/+camyHj+/Dn20traWnzsFZu829raWvaPqC8RERFSpmjXUL1mXEpm//3vf8+ePZtMJlOp1LFjx37zzTc9D0X5zg1l5+vrixDKzs5GCN2+fdvS0pJAIGAX8sfGxlKpVC0trVmzZpmamhKJRCqVOnPmzPnz51tYWJDJZENDw+DgYFFVvR7bCQkJVCpVT08vKytr2bJldDrd3Nw8NTVVtFXP9igRBpLaxvutXzr1P4l+9epVa2urpaWlaImNjQ1C6OnTpwoW7hecYqshGO8GAAwpFov14sULFxcXhBB218/PP/8sXiA4ODgoKKi1tTU8PJzJZM6dO1coFIaGhkZFRcXExNTV1X3yySdr16598OABQojD4Xh7e1dWVrLZbBqNtm7dOlliCA4Onj179pMnT5ydnSdPnhwVFSX+b+peBQYGPnv2TEqB9+/fHzt2zNHRUSgUFhYWHjp0yNjYWLSWRqP1dR1QYWFhX3XeuXOnoqJi7dq1EsvZbHZXV9fDhw9dXV0ZDAaZTJ40aVJ8fLz0y2R4PF5tbe2JEydyc3Pj4uKwcx5ZfPXVV0ZGRrq6ukwmc+XKlffv35dxQ4CBJ1UCMAQIBAI8rxLISHQlFwA44OPj097enpubi738448//vGPf3h5eRUXF798+RJbePXqVW9vb+zvw4cPf/XVV2FhYa9fv87Pz6+urp4/fz6Hw0G9dcJbW1vd3Nw8PT2bmprKy8vHjx/f1dUVGxv7ySef2NjYCIVCbDZ8WajkFEC5/fOJEycisdFthNCIESMQQg0NDdjLffv21dfXx8XFtbS0lJSUxMbGLlmyRPqlLbKora29ceOGh4eH+EIN7Z/n5eUdP34c+7vXjPeV2djY2M8++8zT05PNZtfU1Ozbt+/Fixc9D0X5zg1lh11v9OrVK4SQs7Oz+JMSAgMD9+7dKxQKT548+ffff9fX17u4uBQVFX311VdFRUVNTU0bNmyIjo5+8uQJVr7XY9vf33/Xrl3t7e36+vrp6ekVFRXW1tZ+fn7Y/7Ha2tp6tkeJMJDUNi69/n6p/0k09k8m8XsjyGQyhULB3r4ihfsi9yk2GAoDvkcLAAD+24Dmn/r22283btyI/f3+/XsSiUSn0yXu6w8LC0MIiebDam9v19PT8/HxwV5yuVwSieTv7y9Rc3h4OELo9evXsoTR1dX1ww8/YN1WhNCoUaNu3LghXgD776644uJiKbVZWVm5uLhITLSnCC6XO3v27JqaGqFQiHWjRfN3FxcXI4QWLVp0586dxsbGd+/ehYaGIoSSk5OlVGhqaooQGjFixA8//NDV1dWzwJw5c3pOTMZisR49etTS0tLZ2VlQUDBjxgwKhSLjbeBIqfOQKl4PkIMSZ5eDPKqQsvIIsw2qkLK+UQejNqAs8LuJA0r8npTxeGhpaaFQKJs3b8Ze7t69m8fjCQSC0aNHHz16FFs4c+ZMrO/H5XJpNJqoUy0UCv/880+E0JEjR7CXEp1wbKDq999/l9iph4cHNsgou6E/BRiM/vnSpUuNjY2vX7/e3t5eV1d34cIFAoEgPn8xNqM3xtzcvLq6umclvXa5pdixY4fExNPq0D+XsZ6ep1TR0dGitRIZlyDKbFdXl6Ghoaurq2hVd3d3bGysUOqhKH5gKGX+bgyBQDA0NMT+xh7iHRcXh708dOgQQqilpQV7+dNPP4mfQmJtLS0tTSj12Jb4TOLj4xFCf/31l7Dv9igexkDbuHj9slDzk2jscZHff/+9+EI6nY6Np8tdWEqb7fcUu1fQox4acH03AGBIYXcyYn/T6fTFixc3Nzf3OqOWyIsXL7hc7pQpU7CXFArFzMxMdFuWCJFIRAjx+XxZwiASiQEBAc+fPy8sLFy5cuXr16+9vLzevn0rXka8lyP9FkIikfjkyZNFixbNnz8/Jiamra1Nlhik27dv35YtW7CJwCRgM4hNnjzZ0dHR2NjYwMDg66+/NjAwOH36tJQKq6urX79+nZKS8tNPP82YMUPGqe4sLCxmzJhBo9F0dXXnzp2blJTU3t6OdYwAAAAAAIBq0Wi0FStWYENgnZ2dOjo6Ojo6BALBw8MDm9KktLR0xowZWD+5pKSktbV19uzZos0/+OADXV3de/fu9Vq5tbX1qFGj1q9ff/jw4crKSkXiHPpTgMHon6elpXl5eX322WfGxsZOTk6//vqrUCjErvJGCIWFhZ0+ffr69eutra2vXr1ydHScN28eNhYpNzabffnyZWwaDRHN6p+Ln1L97//+r+wbijL79OnTd+/eiU9Ar62t3e8ELwM6N5RRW1ubUCik0+myFMYu9e3u7haPB7uSWsZjW1QJtpUs7XGgbVy8flmo+Uk0dlut6DPHdHV19TqL4IAK90W+U2wwNGC8GwAwdJ49e1ZcXPzJJ5+IJt7CHnAhcT+jBOyXb//+/aKtqqqquFwuQujKlSsLFiwwMTEhkUjiE6LJbs6cOb/++uv27dsbGhqk9MBiY2NFPZJeGRgY7N+//86dOwKBYO7cuUeOHBH/4R/o1GO3b98uLi7evHlzr/vCpjN78+aNaImurq6VlVVFRYWUCIlEoomJyeLFi9PS0kpKSrBLHgbK3t5eW1tbdHssAAAAAABQLR8fn/r6+vv371+9ehV7giJCyNPT8/HjxxUVFeKTmbx79w4hRKPRxDc3NDRsaWnptWYKhZKXl+fs7Hzs2DFra2ts7hQ5IlTVKYBy++dYhadOnaqpqeFyuRUVFd999x1CaPTo0Qihurq6yMjILVu2fPTRR1QqlclkJiYmsh3NTgEAACAASURBVNns6OhoOT4xkcjISD8/P+mT42lQ/3zBggV79uyRUqDXzDY3NyOEDA0N+61f8XND6bAPWXR1s9ykHNtSyNIeB9rG5aaeJ9FmZmbo/x0wGC6X29HRgZ0+K1K4L0o5xQaDBMa7AQBD5/z582vWrBG/x6SpqYlCoVy7dg2bP6tX2NMaY2JixDcsKChgsVju7u5mZmb37t17//59ZGSkjGF4eHhI/CP3f/7nfxBC/XYy+kWlUnfv3n3//v2RI0e6uLjs3bu3rq4ODXzqsbNnz16/fl1LSwv7Occ+gWPHjhEIhAcPHtBotHHjxpWWlopv0t3dLeOz2mxtbbW1tUtKSuR4gwKBQCAQyPJYeQAAAAAAMARWrFihr69/+fLl/Px8bIJshJCzszODwbh48eKDBw9cXV2xhdiIocTI17t378zNzfuqfPLkyb/99hubzQ4JCUlPTxdNvjwgqj0FUFb/vCds1mzs4y0vL+fz+djYN4ZOpxsbG8vX5cbU19enpKT4+/tLL4ab/nlfmcU+VfFrfQa0uRL95z//QQgtW7ZMwXr6Orb73bDf9ihHG5ed+p9EM5lMfX39qqoq0RJsYvepU6cqWLhfipxig0EC490AgCEiFArT0tK++OIL8YVGRkZeXl58Pj8lJaWvDbFHWj9+/FhieXFxMY/H8/f3t7a2JpPJsj9+qrOzU2Kw+MWLF0iG37a6urqNGzf2Wz+ZTPb393/06NGkSZOwmbUHKikpSfznXHz+buz2NG9v76KiIuxhKQghLpdbVVVlb2/fs6rGxkaJJ15i3XELCwtZIhG/cxAhdP/+faFQOG/ePDnelKa4evWqgYEBdtmRcgvL4vbt205OTnp6egwGIyQkpLOzU1mFhxvIIw5AEnEJ0oo/kFOVI5PJbm5uFy9epFAoWlr/9wRfS0tr1apVP//886hRo3R0dLCFU6ZModFo2CMfMffu3evq6po1a1avNbPZbKzPbGJi8u23386cOVOiCy0LNTkFULx/3lNiYiKTyfzwww8RQth4IjZIh2lpaWlqapKxy92ryMjI9evXiz++D4PX/nlfmR07dqyxsTE227IcmytLfX19TEyMubn5559/rmBVfR3b0snSHgfaxgdE/U+idXR0li9fnp+fLxAIsCXZ2dkEAsHNzU3BwhIUPMUGQwPGuwEAQ+Tu3bt0Ot3JyUli+fbt25HU+xnJZPLGjRtTU1MTEhKam5v5fH5NTU1dXZ2lpSVCKDc3t6Ojo7y8vK9ZyXrl7u5+4cKFd+/evX//PisrKzQ09NNPP5XyUy0UCtvb2y9duiTjfG0IISKR+Pnnn2MPKlG6oKAgKysrX19fFovV2NgYEhLS3t4u6hYcPHjQwMAA6xRSqdRr167l5eU1NzfzeLyioqINGzZQqdSgoCBZdlRbW5uWlvbu3Tsej1dQULB582ZLS0ssZXglFAoHqXC/SkpKFi9evHDhwoaGhoyMjHPnzkn5qAdUeBiCPOIAJBGXIK34AzlVBz4+Pi9evPj444/FF3p5eZWWlrq7u4uWkMnk3bt3Z2RkJCcnNzc3FxcXb9++ncFgbN26tddq2Wz2tm3bysrKurq6ioqKqqqqsDkEjI2N2Wx2ZWVlS0tLvzP/qtUpgIL9cwcHh6qqqu7u7srKyj179uTm5p49exabAZnJZLq6uiYmJubn57e3t1dXV2Of6qZNm2SpWbz3juFwOOfOndu1a1fPwnjtn/eVWRKJtG/fvvz8/ICAgNraWoFA0NLSgg28ih+K2DQU8h0YPQmFwtbWVoFAgF17lJ6e7uTkpK2tnZmZKfv5YF/6Oralb9VXe5SoeUBtfKDU/yT6wIEDHA7n0KFDbW1tBQUF0dHRvr6+EyZMwNZKNDTphaVQ8BQbDJG+7g4AAAAZyfJ84U2bNlGpVB0dnWnTpj169Ei0/JtvvhHNkDVmzJj4+PjIyEjsGREWFhb//ve/sWKdnZ0hISGWlpY6OjomJiYeHh4lJSVCoTAkJMTY2NjQ0NDLy+vEiRMIIRsbGxaLJT2Ya9eueXt729jYkEgkXV3dCRMmHD58uKOjA1ubkZHR87nSIvv375f/k5KX+PXdItXV1WvWrDEyMiKRSA4ODtnZ2aJVBw4c0NfXz8nJwV66ubkxmUwajUYikWxsbHx8fMQfk11QUODk5CTKgpmZmaOj482bN7G1u3fvtrGxwXJnbm7u5+fHZrNlDBsp6fnv6enpw+TXytvbm8lkYh1roVAYHR1NIBCeP3+ueGG5KfHp4ZBHxQvLTVl5HD5Pk1fDJCrrG3UwatMUaphWCfC7OVBqmFMlfk8O6Hjo6uqaNm2a6N1h+Hz+tGnT+Hy++EKBQBAdHT1u3DgikWhkZOTu7v7ixQtsVc9OeGVlpaOjo5GRkba29ujRo8PCwrq7u4VC4aNHj6ysrCgUirOzc319vZTA1OoUQHGLFi0yNDTU0dExMjJasWIFdmG1yJs3bwIDA21tbUkkEo1GEz3QEiO9yy3RexcKhUFBQevXr+81DHXon/dbz507d8aPHy96swsXLpQo0GvGpWT2xIkT9vb2ZDKZTCbPmDEjPj5e2ONQ7HXzwMBAU1NThBCVSl21apX093X58uWpU6fq6enp6upid0sQCARDQ0MHB4cjR440NjaKSsbFxWETQOvp6bm5ucXGxurp6SGExo4de+vWrYiICGyqSVNT0/Pnz6elpWExGBkZpaamCvs4tuPj47FKxo0bV1FRcfr0aWyY2MrK6uXLl722R4kwhFLbuPT6+026ppxE37x508HBgUQiMRiMvXv3iiIU9tbQpBSW3maln2JLN3x61Ko1LHpCAIBBBd/XoFeact4uEAguXLjwz3/+c/B2IQsej0ej0Xx9fUVLnj17hhCKiIhQsLAiNGi8G/IohaaMd0MSpdDc8W5Iq+zgd3NA1DOnqhrvBkBGQzbeDQDoC4yfDA2YzwQAAMDwwufzw8PDJ0yYQKFQRo4cyWQyw8PDV69ejRC6ffu2paUlgUDALglJSEigUql6enpZWVnLli2j0+nm5uapqalYPRKFJQz0eeKvXr1qbW3FbuTEYJdIPH36VMHCeAV5xAFIIi5BWvEHcgoAAAAAzQLj3QAAXCkrK+vrZIlAIPj4+Kg6QKB6kZGRBw8ejI6ObmpqunbtWkdHh6GhIfY0c2dn57t374pK+vv779q1q729XV9fPz09vaKiwtra2s/PD5uqUqKwhIE+T7y+vh4hpK+vL1pCJpMpFAqHw1GwMF5BHnEAkohLkFb8gZyCvkDHG2gWOGL7Ap8MwB8Y7wYA4MrEiRP7OlkSCoVpaWmqDhCoXmZm5qxZs9zc3CgUysyZMz/99NP8/Pyuri4pmzg6OtLpdBMTEx8fn7a2NhaLpfSoOjs7EULa2triC4lEYnt7u4KF8QryiAOQRFyCtOIP5BT0BTreQLPAEdsX+GQA/sB4NwAAgOEFew6J6CWfzycSiRLnwH3R1dVFCGHXqSkXmUxGCHV3d4sv7Orqwp7ko0hhvII84gAkEZcgrfgDOQUAAACAZoHxbgAAAMPL8uXLHz58mJWV1d7e/uDBg8zMzI8//ljG83bZDXQeUuzR6s3NzaIlXC63o6ND9ExwuQvjFeQRByCJuARpxR/IKQAAAAA0i46qAwAAAACG1OHDhx8+fOjr69va2spgMFavXn3s2DGl7wWbh1T28kwmU19fv6qqSrTkr7/+QghNnTpVwcJ4BXnEAUgiLkFa8QdyCgAAAADNAtd3AwBw7urVqwYGBr/99pvSC8vi9u3bTk5Oenp6DAYjJCQEmz5S7sICgSAmJsbR0bHntikpKR988IG+vr6VldXGjRux5zKBXpWUlFRUVDQ0NPB4PBaLlZCQYGRkpOqgkI6OzvLly/Pz8wUCAbYkOzubQCC4ubkpWBivII84AEnEJUgr/kBOwUBpUPdbpKOjY+LEifv37xdfKL2DzePxwsPDbW1tdXV1DQ0Np0yZUllZqax3AXBGgxqFlMJHjhyxs7Oj0+kkEsnW1jY4OLi1tVV8WzgnBeoDxrsBADg3oGuFBlS4XyUlJYsXL164cGFDQ0NGRsa5c+e2b98ud+Hy8nIXF5egoCAulyuxbXp6+rp167y8vGpqarKysvLz85ctWyYxSSUQ2bFjh6WlpUTnTB0cOHCAw+EcOnSora2toKAgOjra19d3woQJ2NqDBw8aGBhcu3ZNlsLDAeQRByCJuARpxR/IKRgoTel+iwsLC3vx4oX4kn472N7e3j///PP58+e5XO7z589tbGzUsJkANaEpjUJ64by8vB07dlRWVr558yY8PDw2NtbLy0u0Fs5JgXqR8gxWAACQhaenp6enp6qjUEfe3t5MJlMgEGAvo6OjCQTC8+fP5Sj8+PHjVatWJScnT58+fdq0aRLburq6jh49WrTtiRMnEEK3b99W/lsaCIRQenq64vWkp6cr99cqLy9vxIgRot9BIpE4adKkS5cuCYXCuLg4bIpPPT09Nze3+Ph4PT09hNC4ceMqKipOnz5Np9MRQlZWVi9fvpQorJTYbt686eDgQCKRGAzG3r17sUeEYQ4cOKCvr5+TkyNLYWVRYuuGPGI0Oo9K/7aHJMpOWd+og1GbBEir3OB3Uw7qllMlfk8OajvFsQF1v0Xu3LmzePFihFBYWJhoofQOdmpqKoFAePr06SC8iUGkrOMKjk8NosRz0hUrVnR3d4sKr169GiHEYrGwl+p5TqqGYPxkaMB4NwBAUer2fS0QCC5cuPDPf/5TtWHweDwajebr6yta8uzZM4RQRESEIoXnzJnTc7zb1tZ21qxZopdZWVkIofPnzyvhbShAbc/b4+PjAwMDRS87Ozt37dpFIpG4XK4S94IP6jzeDXmUndqOd0MSZadB492QVrnB7yYODNvxbk3sfotwuVxHR8fS0lKJ8W7pHWwXFxfxtZoCxruHjCY2ioG2IH9/f4RQWVkZ9lI9z0nVkLqNn+AVzGcCANB4fD4/PDx8woQJFApl5MiRTCYzPDwc+2/z7du3LS0tCQQC9u/lhIQEKpWqp6eXlZW1bNkyOp1ubm6empqK1SNRWEJrayuhD3Pnzu1Z/tWrV62trZaWlqIlNjY2CKGnT58qWLgna2vr169fi15iE6VZW1vLsu1wU19fHxAQsGnTJtESXV1dS0tLHo/H4/FUGBgYEMgjDkAScQnSij+QU9ArHHS/RcLCwr744gsTExOJ5VI62F1dXYWFhdOnT5fpwwLDAw4axUBbUG1tLYVCYTKZ2Es4JwVqBca7AQAaLzIy8uDBg9HR0U1NTdeuXevo6DA0NDQ0NEQIOTs73717V1TS399/165d7e3t+vr66enpFRUV1tbWfn5+2AmbRGEJNBqtr/8cFhYW9iyP/cDr6+uLlpDJZAqFwuFwFCzc0759++rr6+Pi4lpaWkpKSmJjY5csWdJrjwdQKBQikXj27FkOh8Pj8dhs9pkzZw4ePOjj44Pdcw00AuQRByCJuARpxR/IKegVDrrfmDt37lRUVKxdu7bnKikdbDab3dXV9fDhQ1dXVwaDQSaTJ02aFB8fL1TqtMtAs+CgUQyoMJfLzcvL8/Pz09XVxZbAOSlQKzDeDQDQeJmZmbNmzXJzc6NQKDNnzvz000/z8/O7urqkbOLo6Ein001MTHx8fNra2lgsltKjwp5kra2tLb6QSCS2t7crWLinDz/8MCQkJCAggE6nT5kypaWl5cyZM/KHjmvYo6uePXs2fvx4CoViZ2eXlJQUERHx008/qTo0MACQRxyAJOISpBV/IKegVzjofiOE2tvbAwMDExISel0rpYONPZfSxMTk2LFjJSUlHA5n5cqVO3bsSElJUeb7ARoFB41iQIXDw8MZDMbRo0dFS+CcFKgVHVUHAAAAiuro6CCTyaKXfD6fSCRK/E73Bft39GDckIuFJPFA6q6uLgqFomDhnsLCws6cOXP9+vU5c+a8fv06NDR03rx5d+/etbCwkP8N4Nf8+fP/+OMPVUcBFAV5xAFIIi5BWvEHcgp6wkH3GyG0b9++LVu2jBkzpte1UjrYJBIJITR58mRHR0es8Ndff33y5MnTp0+vW7dOmW8JaA4cNArZC2dkZFy4cOHatWviF4PDOSlQK3B9NwBA4y1fvvzhw4dZWVnt7e0PHjzIzMz8+OOPZexbyG6gc6WZmZkhhJqbm0VLuFxuR0cHg8FQsLCEurq6yMjILVu2fPTRR1QqlclkJiYmstns6OhoOd4mAAAAAAAA0uGg+3379u3i4uLNmzf3umvpHWyswjdv3ojK6+rqWllZVVRUKPaOgQbDQaOQsXBaWlpERMSNGzfGjh0rWgjnpEDdwPXdAACNd/jw4YcPH/r6+ra2tjIYjNWrVx87dkzpe8HmSpO9PJPJ1NfXr6qqEi3566+/EEJTp05VsLCE8vJyPp8/evRo0RI6nW5sbFxSUiJ7tAAAAAAAAMgIB93vs2fPXr9+XUvrvy4BPHbs2LFjx+7fv8/lcqV0sGk02rhx40pLS8W37e7uNjAwkD1agDM4aBSyFI6Li8vJycnLy6PRaOLbwjkpUDdwfTcAQOOVlJRUVFQ0NDTweDwWi5WQkGBkZKTqoJCOjs7y5cvz8/MFAgG2JDs7m0AguLm5KVhYgrm5OUKorq5OtKSlpaWpqQluHFOV48ePjxo1ikAgnDp1SlUxCASCmJgY0T22QA4qzyOPxwsPD7e1tdXV1TU0NJwyZUplZaVKItFQqs3gggULel54JXFmCGSk8saI+v5SPXLkiJ2dHZ1OJ5FItra2wcHB2Ky+YEDUOcWgLzjoficlJYk/66+hoQEhFBYWJhQKZ8+e3W8H29vbu6io6NWrV9hLLpdbVVVlb28/qG8QqDMcNArphYVCYUhISHFxcWZmZs8uDZyTAnUD490AAI23Y8cOS0tLNTzDPHDgAIfDOXToUFtbW0FBQXR0tK+v74QJE7C1Bw8exJ4BJUthKZhMpqura2JiYn5+fnt7e3V19datWxFCmzZtGry3BqTYs2ePlIeqD4Hy8nIXF5egoCAul6vCMDSdyvPo7e39888/nz9/nsvlPn/+3MbGRg2/5dSZyjPYk7Ozs6pD0EgqT6WUL9W8vLwdO3ZUVla+efMmPDw8NjbWy8tLJUFqNHVOMegLPrrfUvTbwQ4KCrKysvL19WWxWI2NjSEhIe3t7aGhoYP31oCaw0ejkFK4tLQ0KioqMTGRSCSK/zv/+PHjCM5JgfqB8W4AgMYLDw9/9uyZkZER9ourq6trZ2eXkZGBEDpx4sQHH3yAEAoJCfn0008TEhJiYmIQQlOnTn316lViYuLu3bsRQkuXLi0vL5corHhgkydPzsnJuXbt2ogRIzw8PD7//POTJ0+KFxC/GU164cLCQmdn59GjR9+7d+/JkycMBsPJySk/Px8hRCAQfvnlFx8fn02bNhkZGdnZ2bFYrEuXLs2fP1/xtwA0zpMnT0JDQ7dv3z59+nRVxwLkl5aWlpmZ+csvv8yZM0dHR4fBYGRlZU2ZMkXVcQFZkcnk5uZm8SsHt27dGhwcrOq4wIBJ/1Kl0Whbt241NjbW19dfvXq1u7v7f/7zn+rq6qGPE8gNfjflg4/utxT9drCNjIxu3bplbm4+ffr0MWPG/Pnnn1euXIGjaDjDR6OQUlh624FzUqB2hAAAoBhPT09PT08VBhAfHx8YGCh62dnZuWvXLhKJxOVyVRgVQAilp6crXk96errG/VqVl5cjhE6ePKnCGObMmTNt2jQFK1Fi64Y8DoiLi8usWbOUVZuy8qjyb/uBUoeWiGGxWE5OTorUoKxv1MGobQioQypl+VL19/dHCJWVlcm3C/jdVP8U90uJ35Nq3k6h+62hlHVcqfnxqRLQKICMNK5HraHg+m4AgGarr68PCAgQv09KV1fX0tKSx+PxeDwVBgbUwc2bNx0cHPT09Oh0ur29Pfa08Vu3btnZ2RkYGJDJZHt7+5ycHIRQbGwslUrV0tKaNWuWqakpkUikUqkzZ86cP3++hYUFmUw2NDQUXZv5448/ksnkUaNGbdu2jcFgkMlkR0fHe/fu9RUGn88/ePCgpaUlhUKZOnUqNhLRV3jiBvoEdrwabnns6uoqLCzE0zViwy2DPUVEROzcuVOOj07dQCqlq62tpVAoTCZT8apUBVIMZAHdbwAkQKMAQN3AeDcAQLNRKBQikXj27FkOh8Pj8dhs9pkzZw4ePOjj40On01UdHVCltrY2Nzc3T0/Ppqam8vLy8ePHd3V1IYQ4HI63t3dlZSWbzabRaOvWrUMIBQYG7t27VygUnjx58u+//66vr3dxcSkqKvrqq6+Kioqampo2bNgQHR395MkThFBAQICvry+Xy925c2dlZeWjR4+6u7sXLVrU1z3soaGhUVFRMTExdXV1n3zyydq1ax88eNBXeOKwJ7D3qrCwcJA/P3UxDPPIZrO7uroePnzo6uqKDQxNmjQpPj5eKNst2OpmGGZQQm1t7Y0bNzw8PBT9KFUNUikdl8vNy8vz8/PT1dVVsCpVgRQDGUH3GwAJ0CgAUDcw3g0A0GzY4zWePXs2fvx4CoViZ2eXlJQUERHx008/qTo0oGKVlZXNzc2TJ08mk8mmpqaXLl0aOXIkQsjT0/PQoUNGRkbGxsZubm6NjY0NDQ2irezs7PT09EaMGLFmzRqEkKWl5ciRI/X09NavX48QKisrE5XU0dGZNGkSiUSys7NLSEhoaWlJSkrqGUZHR0dCQoK7u7uHh4ehoeH+/fuJRGJSUlJf4QEJwzCP2JOOTExMjh07VlJSwuFwVq5cuWPHjpSUFAVrVolhmEEJERERX375pZaWxve6IZXShYeHMxiMo0ePDuVOlQtSDGQE3W8AJECjAEDdaHzPGwAA5s+f/8cff7x//767u/vdu3d37tzx9/fX0dFRdVxAxaytrUeNGrV+/frDhw9XVlb2WoZIJCKE+Hx+z1XYBXrd3d3iJfu6IXH27Nl6enriZ/UiL1684HK5oicNUigUMzOzsrIyWcIDaFjmkUQiIYQmT57s6OhobGxsYGDw9ddfGxgYnD59Win1D7FhmEFxbDb78uXLvr6+Sq956A3zVEqXkZFx4cKFnJwcfX39Id61EkGKgeyg+w2ABGgUAKgVGO8GAACATxQKJS8vz9nZ+dixY9bW1j4+Pu3t7QihK1euLFiwwMTEhEQiiaYWVRyJRBK/3k2kra0NIbR//37RFKJVVVVcLrev8MTBPKRoWOaRwWAghN68eSNaoqura2VlVVFRoay3OZSGYQbFRUZG+vn5kclkZb1BFRrmqZQiLS0tIiLixo0bY8eOlbsSdQApBgAAAAA+wHg3AAAA3Jo8efJvv/3GZrNDQkLS09OPHz/OYrHc3d3NzMzu3bv3/v37yMhIpeyIx+O9e/fO3Ny85yoTExOEUExMjPgsogUFBb2GJ7EtzEOKGW55pNFo48aNKy0tFV/Y3d1tYGCglLc59IZbBkXq6+tTUlL8/f2V8u7UwbBNpRRxcXHJycl5eXmjR4+Wrwa1AikGAAAAAA7AeDcAAEhz/PjxUaNGEQiEU6dODf3ejxw5YmdnR6fTSSSSra1tcHAwNrGvLGsRQrdv33ZyctLT02MwGCEhIZ2dnUP+DlSJzWZjI4YmJibffvvtzJkzS0tLi4uLeTyev7+/tbU1mUwmEAhK2deNGzeEQmGv145ZWFiQyeTHjx/LEp5SgsGZ4ZlHb2/voqKiV69eYS+5XG5VVZW9vb3iNQ+94ZlBTGRk5Pr1642NjZVVoWoN51T2SigUhoSEFBcXZ2Zm0mi0Qd3X0IAUA7Wl2g45RiAQxMTEODo6Sizvt0MOgOLUuQlERkZOnDiRQqFQqdSJEyceOHCgublZtJbH4x08eNDa2lpXV3fMmDF79uzpeWsOAIMBxrsBAECaPXv23L17V1V7z8vL27FjR2Vl5Zs3b8LDw2NjY728vGRcW1JSsnjx4oULFzY0NGRkZJw7d2779u2qeBMqw2azt23bVlZW1tXVVVRUVFVVNXfuXEtLS4RQbm5uR0dHeXn5vXv35K5fIBC8ffu2u7v76dOngYGBlpaWvU7RSyaTN27cmJqampCQ0NzczOfza2pq6urqeg1P7mBwbHjmMSgoyMrKytfXl8ViNTY2hoSEtLe3h4aGKl7z0BueGUQIcTicc+fO7dq1Sym1qYNhm8q+lJaWRkVFJSYmEolE8Xkzel50rCkgxUBtqbZDjhAqLy93cXEJCgricrkSq6R3yAFQCnVuArdu3fLz82OxWBwO55tvvomMjPT09BStDQwMjI6ODg8Pb2xsPH/+fGJi4ubNm4c2djBc9XXDFwAAyMjT09PT01PVUQyi8vJyhNDJkyeHftcrVqzo7u4WvVy9ejVCiMViybLW29ubyWQKBALsZXR0NIFAeP78+VDFLkQIpaenK15Penq6fL9WlZWVjo6ORkZG2trao0ePDgsLwz6ukJAQY2NjQ0NDLy+vEydOIIRsbGx2796tp6eHEBo7duytW7ciIiKwuSNMTU3Pnz+flpZmamqKEDIyMkpNTRUKhVu3biUSiWPGjNHR0aHT6StXrqyoqMD2+91332GFqVTqqlWrhEJhZ2dnSEiIpaWljo6OiYmJh4dHSUlJX+EpqKCgwMnJCZsAGiFkZmbm6Oh48+ZN+WpTYuuGPA5UdXX1mjVrjIyMSCSSg4NDdna23FUpK4/y1TNsMxgUFLR+/XqlVCVU3jeqIrUNz1RK+VItLi7u9fQqOjpavn3B76a6pVgOSvzdVG6rxwEVdsgfP368atWq5OTk6dOnT5s2TWKt9A65ulHWcQXH59BT2ybg7u7e3t4ueon9v4fNZguFwoqKCi0trS1btojW7t+/HyFUWlo6NJGrJ9yPn6gJGO8GACgK99/XKuxbSMAmgS0rK+t3LY/Ho9Fovr6+orXPnj1DCEVEeSTVFgAAIABJREFURAxRrGpw3j6otm7damxsrOooBp06jHcPKsijSupRomGSQaF6jHcPquGTSingdxMHYLx78KhDh3zOnDk9B/skSO+uqxyMd2suTWkCgYGBCKGXL18KhcK0tDSE0NmzZ0Vrb9++jXo8oWG4UcMeNS7BfCYAAI1x8+ZNBwcHPT09Op1ub2+PzQt269YtOzs7AwMDMplsb2+fk5ODEIqNjaVSqVpaWrNmzTI1NSUSiVQqdebMmfPnz8cmhTQ0NAwODsaq/fHHH8lk8qhRo7Zt28ZgMMhksqOjo5Tbdfl8/sGDBy0tLSkUytSpU7Gzyr7CE9fa2krog4w35NbW1lIoFCaT2e/aV69etba2YvcgY2xsbBBCT58+lWVHQBZ8Pl/VIQAlgDxqOsggbkAqcQ9SjA/QIZdOencd4AA0AenKy8sNDQ2trKwQQlpaWgghCoUiWjtu3DiE0PPnzxXfEQDSwXg3AEAztLW1ubm5eXp6NjU1lZeXjx8/vqurCyHE4XC8vb0rKyvZbDaNRlu3bh1CKDAwcO/evUKh8OTJk3///Xd9fb2Li0tRUdFXX31VVFTU1NS0YcOG6OjoJ0+eIIQCAgJ8fX25XO7OnTsrKysfPXrU3d29aNGi6urqXiMJDQ2NioqKiYmpq6v75JNP1q5d++DBg77CE0ej0fr632NhYWG/nwCXy83Ly/Pz89PV1e13bX19PUJIX19fVIBMJlMoFA6HI+MHDgAAAAAAgDjokEsnvbsOcACaQF94PF5tbe2JEydyc3Pj4uKwJjBx4kT036PbI0aMQAg1NDTIvSMAZATj3QAAzVBZWdnc3Dx58mQymWxqanrp0qWRI0cihDw9PQ8dOmRkZGRsbOzm5tbY2Cj+82lnZ6enpzdixIg1a9YghCwtLUeOHKmnp7d+/XqEUFlZmaikjo7OpEmTSCSSnZ1dQkJCS0tLUlJSzzA6OjoSEhLc3d09PDwMDQ33799PJBKTkpL6Ck+JwsPDGQzG0aNHZVnb2dmJENLW1hYvQyQS4XHYSrFv376kpKT3798zmcyLFy+qOhwgJ8ijpoMM4gakEvcgxbgBHXLppHfXAQ5AE+iLhYWFubn54cOHo6KivL29sYX29vZLly6Nj4/Py8vr6Oior6/PyMggEAg8Hm9oogLDGYx3AwA0g7W19ahRo9avX3/48OHKyspeyxCJRNTH3bLYf5i7u7vFS/b1Qzt79mw9PT3xnofIixcvuFzulClTsJcUCsXMzKysrEyW8BSRkZFx4cKFnJwc8Uu2pawlk8lI7P1iurq6xO8mA3ILDw/v7OwUCoV///23+PPHgWaBPGo6yCBuQCpxD1KMG8O8Qy6d9O46wAdoAn2prq5+/fp1SkrKTz/9NGPGjNevX2PL09LSvLy8PvvsM2NjYycnp19//VUoFGJXeQMwqGC8GwCgGSgUSl5enrOz87Fjx6ytrX18fLBLla9cubJgwQITExMSiSSa/kxxJBKp19us2traEEL79+8XTXNWVVXF5XL7Ck+c3HOlpaWlRURE3LhxY+zYsTKuNTMzQwiJz9fG5XI7OjoYDIbsHwIAAAAAAAAiw7lDLp307jrADWgCfSESiSYmJosXL05LSyspKQkPD8eWGxgYnDp1qqamhsvlVlRUfPfddwih0aNHy70jAGQE490AAI0xefLk3377jc1mh4SEpKenHz9+nMViubu7m5mZ3bt37/3795GRkUrZEY/He/funbm5ec9VJiYmqMcTpQsKCnoNT2Jb+eZKi4uLS05OzsvL67Vb0NdaJpOpr69fVVUlWvLXX38hhKZOnTqADwIAAAAAAAAxw7NDLp307jrAGWgC0tna2mpra5eUlPS69v79+wghV1dXxXcEgHQw3g0A0AxsNru0tBQhZGJi8u23386cObO0tLS4uJjH4/n7+1tbW5PJZAKBoJR93bhxQygU9vr/bexR2o8fP5YlPAXDEAqFISEhxcXFmZmZNBptQGt1dHSWL1+en58vEAiwJdnZ2QQCwc3NTcGo8OT48eOjRo0iEAinTp1SYRjbtm0TXVWBzeKHELp9+7aTk5Oenh6DwQgJCcEmZJedQCCIiYlxdHQUX5iZmSna0RBPZzmo1DmPPB4vPDzc1tZWV1fX0NBwypQpst9b2tcxgMs8qnMS5W6MCxYs6HnlFPZ1jcskIvXOI+rjixEhdOTIETs7OzqdTiKRbG1tg4ODW1tbZd8dNNWhB7+bKjQMO+TSSe+QA/yBJiChsbFx7dq14kvKy8v5fL6FhUWv5RMTE5lM5ocffjioUQGAYLwbAKAp2Gz2tm3bysrKurq6ioqKqqqq5s6da2lpiRDKzc3t6OgoLy+/d++e3PULBIK3b992d3c/ffo0MDDQ0tLS19e3ZzEymbxx48bU1NSEhITm5mY+n19TU1NXV9dreHIHgyktLY2KikpMTCQSieLDJdh/6aWvRQgdOHCAw+EcOnSora2toKAgOjra19d3woQJCkaFJ3v27Ll7966qo0AIIWNj4+zs7BcvXpw9exYhVFJSsnjx4oULFzY0NGRkZJw7d2779u2y11ZeXu7i4hIUFMTlcsWXf/rppzU1Nfn5+cuXL1fyG1Aptc0jQsjb2/vnn38+f/48l8t9/vy5jY2NjONoUo4BXOZRbZOoYGPsydnZGeE0iUiN84j6/mJECOXl5e3YsaOysvLNmzfh4eGxsbFeXl4y7giaqqrA76aqDMMOuXT9dsgBzkATkEClUq9du5aXl9fc3Mzj8YqKijZs2EClUoOCgrACDg4OVVVV3d3dlZWVe/bsyc3NPXv2LDaPOQCDq68bGQAAQEaenp6enp6DvZfKykpHR0cjIyNtbe3Ro0eHhYV1d3cLhcKQkBBjY2NDQ0MvL68TJ04ghGxsbHbv3q2np4cQGjt27K1btyIiIgwMDBBCpqam58+fT0tLMzU1RQgZGRmlpqYKhcKtW7cSicQxY8bo6OjQ6fSVK1dWVFRg+/3uu++wwlQqddWqVUKhsLOzMyQkxNLSUkdHx8TExMPDo6SkpK/wFFFcXNzr93Z0dHS/azE3b950cHAgkUgMBmPv3r0dHR0KhjQgCKH09HTF60lPTx+8X6vy8nKE0MmTJwepflls3bp1zJgx4ku8vb2ZTKZAIMBeRkdHEwiE58+fy1Lb48ePV61alZycPH369GnTpvVaZufOnSNGjJClNiW27mGYx9TUVAKB8PTpUzlqk+UYGPo8Duq3vXomUZHGuGTJkubmZon6r1+/Lr5E9iQq6xt1MGoTp555lP7FuGLFCvGf7NWrVyOEWCyWLPtSblOF300ZDZPfzcFrp4oYhh1yoVBYUFDg5OQkegqOmZmZo6PjzZs3hbJ1yNWKso4r9Tw+hwA0AYkmIBQK3dzcmEwmjUYjkUg2NjY+Pj7FxcWibRctWmRoaKijo2NkZLRixYr79+8rHo+mG5rxEwDj3QAAReHg+3rr1q3GxsaqjgJv4LxdRhLn7Twej0aj+fr6ipY8e/YMIRQRETGgaufMmaNW5+3DLY9CodDFxWXWrFlyVCXjMQDj3Uo3SI0Rw2KxnJycJBbCePdg6NkYRaR8MYr4+/sjhMrKyvrdkdKbKvxuymiY/G4Ow/FE6JAPARjvVmfQBIYJHIyfaASYzwQAABBCiM/nqzoEIL9JkyYRCAQtLa1Zs2ZhNyMHBwcbGBiQyeR//etfCKFbt27Z2dlhS+zt7XNycnpWEhAQoKura2Zmhr384osvqFQqgUB48+YNtoTP5x88eNDS0pJCoUydOhUbUJCg+BPPX7161drait0XibGxsUEIPX36dECfiSbCUx67uroKCwunT58ux+eg0ccAnpKo3ERERETs3LlTjg1VAk95HKja2loKhcJkMvstCU1VTVKs0YkAEqBDDoY5aAIAKIuOqgMAAAAAFPXs2TNbW1uBQPDnn39qa2sjhKKiorq7u0Vz3nE4HG9v74CAAKFQuGLFinXr1onOxkV+/PHHpqam3Nxc7GV8fDyTydy7d6+oQGhoaFxc3Pnz5xcuXBgdHb127VobG5vZs2eLV4I98VyR91JfX48Q0tfXFy0hk8kUCoXD4ShSrUbAUx7ZbHZXV9fDhw9dXV3Lysrevn3LZDJ37Njh7+9P6O8pRhp9DOApiUpMRG1t7Y0bN+Li4hSJZyjhKY8DwuVy8/Ly/Pz8ZJlaFJqqmqRYoxMBAAAAgMEA13cDAIa7ffv2JSUlvX//nslkXrx4UdXhAHloa2vv3LmTxWJlZGRgS7hc7qVLlz7//HPspaen56FDh4yMjIyNjd3c3BobGxsaGga0i46OjoSEBHd3dw8PD0NDw/379xOJxKSkJCW/E4Q6OzsRQtjogwiRSGxvb1f6vtQNnvKIPZfSxMTk2LFjJSUlHA5n5cqVO3bsSElJ6XdbjT4G8JREJSYiIiLiyy+/1NLSmI43nvI4IOHh4QwG4+jRo7IUhqYqnSY2VaBC0CEHwxw0AQCUS2O63QAAMEjCw8M7OzuFQuHff//t6emp6nCAnDZv3mxgYBAbG4u9TE5OXrlyJZ1O71mSSCSigd8t+OLFCy6XO2XKFOwlhUIxMzMrKytTLOpekMlkhFB3d7f4wq6uLgqFovR9qSHc5JFEIiGEJk+e7OjoaGxsbGBg8PXXXxsYGJw+fbrfbTX9GMBNEpWVCDabffnyZeyaWQ2CmzzKLiMj48KFCzk5OeJXCksBTVU6jWuqQLWgQw6GOWgCACgXjHcDAADAAxqNtmXLlrt37/75558IoZMnTwYEBIjWXrlyZcGCBSYmJiQSKTg4WI7629raEEL79+8XzStaVVWFzXkqTvF5SLGJUJubm0VLuFxuR0eH6JHo+IabPGL5Er/9X1dX18rKqqKiot9tNf0YwE0SlZWIyMhIPz8/bEhOg+AmjzJKS0uLiIi4cePG2LFjZdwEmqp0GtdUAQAAAIAbMN4NAAAAJwICAohEYkxMTH5+voWFBfa4KoQQi8Vyd3c3MzO7d+/e+/fvIyMj5ajcxMQEIRQTEyP+0OeCggKJYtg8pL0qLCyUZUdMJlNfX7+qqkq05K+//kIITZ06VY6wNRE+8kij0caNG1daWiq+sLu728DAoN9tcXAM4COJSklEfX19SkqKv7+/7JuoD3zkURZxcXHJycl5eXmjR4+WfStoqtJpVlMFAAAAAJ7AeDcAAD+OHz8+atQoAoFw6tQpFYaxbds20aVJ69evRwgdOXLEzs6OTqeTSCRbW9vg4GBsbl9xAoEgJibG0dGxZ4UpKSkffPCBvr6+lZXVxo0bsecyyULKfjMzM0VBjhw5UoG3q0bMzc1Xr1598eLFAwcOBAYGipYXFxfzeDx/f39ra2symUzo+2mBOjo6PB6v11UWFhZkMvnx48fKj7tHDMuXL8/PzxcIBNiS7OxsAoHg5uY22LtWE/jII0LI29u7qKjo1atX2Esul1tVVWVvb9/vhjg4BvCRRKUkIjIycv369cbGxoMT4+DCRx6lEwqFISEhxcXFmZmZNBptQNtCU0XqkWIcJALH1LZzvmDBgp63FMj4JXD06FGJDUXz9uCygw0GSm0Pe1nOSaXr64w1MjJy4sSJFAqFSqVOnDjxwIED4vfcSGlu0GTA4IHxbgAAfuzZs+fu3buqjgIhhIyNjbOzs1+8eHH27FmEUF5e3o4dOyorK9+8eRMeHh4bG+vl5SVevry83MXFJSgoqOd9vunp6evWrfPy8qqpqcnKysrPz1+2bJnEJJV9kbLfTz/9tKamJj8/f/ny5cp4x+pi9+7d3d3db9++/eijj0QLLS0tEUK5ubkdHR3l5eX37t3ra3NbW9umpqbMzEwej9fQ0CB+sRiZTN64cWNqampCQkJzczOfz6+pqamrqxuMd3HgwAEOh3Po0KG2traCgoLo6GhfX98JEyZga318fExNTR89ejQYu1YT+MhjUFCQlZWVr68vi8VqbGwMCQlpb28PDQ3F1krPo/RjQCPgI4kKNkYOh3Pu3Lldu3YNRmxDAx95lKK0tDQqKioxMZFIJIqfhx8/fhwrAE1VI1IMv5tqS207571ydnZWcC947WCDAVHbw77fc1LppJyx3rp1y8/Pj8VicTicb775JjIyst8pyLHmBk0GDKK+bh8DAAAZeXp6enp6qjqK/6u8vBwhdPLkSRXGsHXr1jFjxogvWbFiRXd3t+jl6tWrEUIsFgt7+fjx41WrViUnJ0+fPn3atGkStbm6uo4ePVogEGAvT5w4gRC6ffu2LJFI3y9m586dI0aMkPnNDQBCKD09XfF60tPTB/Rr5erqeubMGYmFISEhxsbGhoaGXl5e2GdoY2MTGBhoamqKEKJSqatWrRIKhY2Nja6urmQymclkfvnll3v37kUI2draYh9aZ2dnSEiIpaWljo6OiYmJh4dHSUmJ4m+w5wEjFApv3rzp4OBAIpEYDMbevXs7OjpEq9zd3RFCBw8e7LW2goICJycn0aSlZmZmjo6ON2/eFC8je9KV2LqHZx6rq6vXrFljZGREIpEcHByys7NFq6TnUSj1GMAMfR4HWg8+kih3YxQKhUFBQevXr5eyR9mTqKxvVDlqw0EepXwxFhcX93qKFB0djW07lE0VfjdlNEx+N5Xb6lVIPTvnS5YsaW5ulihz/fp1WWr75ptv/v3vf0svM3gdbAUp67jCzfE5SNTzsJfl3LAv0s9Y3d3d29vbRS+xYXQ2m429lKW5qW2TGQxqNX6CYzDeDQBQlFp9X6tn30ICNpFrWVmZxPI5c+b07D3Y2trOmjVL9DIrKwshdP78eTkC63W/+Bvv1jj9HjAS+Hz+/Pnzz549K/ceNWK8W+MMhzyq1bf9YFDnJKpwvFvjDDSP0mliluH7VoImft8KcdRONaJzzmKxnJycZKwNxruVWA9eacRh39c5qXS9nrFKwCbIevnyZa9re21uattkBgPue9RqAuYzAQCoo0mTJhEIBC0trVmzZmE3TAUHBxsYGJDJ5H/9618IoVu3btnZ2WFL7O3tc3JyelYSEBCgq6trZmaGvfziiy+oVCqBQHjz5g22hM/nHzx40NLSkkKhTJ06FTs/lNDa2tpzujHM3Llz5Xt3tbW1FAqFyWTKUtja2vr169eil9jk3dbW1oO9XzCU2tvbc3JyysvLu7q6pJfk8/mZmZktLS0+Pj4D3YtQKGSz2bdv38Ye5AWUDvKIA5BEfJA9j9JBltUWNNWhh+/OeURExM6dO+XbFuAYvg/7wTs3LC8vNzQ0tLKy6nUtNDcwNGC8GwCgjp49ezZ27FgLC4s///xTT08PIRQVFbVp06aIiAhfX1+EEIfD8fb2rqysZLPZNBpt3bp1PSv58ccfsbu0MPHx8V9//bV4gdDQ0KioqJiYmLq6uk8++WTt2rUPHjyQqIRGo/X1D8PCwkI53hqXy83Ly/Pz89PV1ZWl/L59++rr6+Pi4lpaWkpKSmJjY5csWSJHt2ag+wVDqampaenSpePHj//888+ll7xx48alS5eys7OxdjEgWVlZY8aMmT9//pUrV+SNFEgDecQBSCI+yJ5H6SDLagua6tDDcee8trb2xo0bHh4esm/y1VdfGRkZ6erqMpnMlStX3r9/X479AvWH48N+MM4NeTxebW3tiRMncnNz4+Lieq1ZjuYGgHxgvBsAoI60tbV37tzJYrEyMjKwJVwu99KlS6KzGk9Pz0OHDhkZGRkbG7u5uTU2NjY0NAxoFx0dHQkJCe7u7h4eHoaGhvv37ycSiUlJSUp+Jz2Eh4czGIyjR4/KWP7DDz8MCQkJCAig0+lTpkxpaWk5c+bMEOwXDJlTp06JOqzJycnSCy9cuPD8+fOiK0QGZOXKlaIdia4oAcoCecQBSCI+DCiP0kGW1RM0VZXAcec8IiLiyy+/1NKSdWxkw4YNly9frq6ubm1tTU1NZbFYH374YUlJyaAGCVQCx4f9YJwbWlhYmJubHz58OCoqytvbu9cyA21uAMgNDjIAgJravHmzgYFBbGws9jI5OXnlypV0Or1nSSKRiBDi8/kDqv/FixdcLnfKlCnYSwqFYmZmVlZWpljU/cjIyLhw4UJOTo6+vr6Mm4SFhZ0+ffr69eutra2vXr1ydHScN29edXX1YO8XAAAAAAAAEVx2ztls9uXLl7FrdWVkYWExY8YMGo2mq6s7d+7cpKSk9vb2+Pj4QYsRqBIuD/tBOjesrq5+/fp1SkrKTz/9NGPGDPE5OTFyNDcA5Abj3QAANUWj0bZs2XL37t0///wTIXTy5MmAgADR2itXrixYsMDExIREIgUHB8tRf1tbG0Jo//79ornPqqqqsHnZxClxrrS0tLSIiIgbN26MHTtWxk3q6uoiIyO3bNny0UcfUalUJpOZmJjIZrOjo6MHdb8AAAAAAACIw1/nHCEUGRnp5+dHJpPlCBhjb2+vra398uVLuWsA6gx/h/3gnRsSiUQTE5PFixenpaWVlJSEh4dLFFC8uQEgOxjvBgCor4CAACKRGBMTk5+fb2FhYWNjgy1nsVju7u5mZmb37t17//59ZGSkHJWbmJgghGJiYsSnPysoKJAopqy50uLi4pKTk/Py8kaPHi37VuXl5Xw+X3wTOp1ubGws+y2T8u0XAAAAAAAACXjqnCOE6uvrU1JS/P395YhWRCAQCAQCEomkSCVAneHpsB+ac0NbW1ttbW2JM1alNDcAZKej6gAAAKBP5ubmq1evTk9PZ7PZhw4dEi0vLi7m8Xj+/v7W1tYIIQKB0FcNOjo6PB6v11UWFhZkMvnx48dKD1uCUCgMDQ19+/ZtZmamjs7AvnXNzc0RQnV1daIlLS0tTU1NFhYWg7pfAAAAAAAAJOCjcy4SGRm5fv16Y2PjAW21ZMmSnJwc0cv79+8LhcJ58+YpOzqgLvBx2A/euWFjY+OXX36ZkpIiWoJdsyVxxipfcwNAbnB9NwBAre3evbu7u/vt27cfffSRaKGlpSVCKDc3t6Ojo7y8/N69e31tbmtr29TUlJmZyePxGhoaqqqqRKvIZPLGjRtTU1MTEhKam5v5fH5NTY34yLKylJaWRkVFJSYmEolE8VvPjh8/3u+2TCbT1dU1MTExPz+/vb29urp669atCKFNmzZhBXx8fExNTR89eqTc/QIAAAAAANATDjrnGA6Hc+7cuV27dvVcJaWDjRCqra1NS0t79+4dj8crKCjYvHmzpaXl9u3bBylOoA5wcNj3e24o/bCXgkqlXrt2LS8vr7m5mcfjFRUVbdiwgUqlBgUFicpIaW4ADBK44g8AoNZmzJjh6uq6bt068YX29vYhISHx8fFxcXGLFi1asGDBxYsXnZ2dP/nkk9TUVITQnj17/vjjj0uXLvn7++fl5a1Zs4bBYHz88ccUCgUhNG/evLy8PAsLi9jYWDqdHhkZuXPnTiMjIxcXlyNHjjAYDOW+BaFQKL1AYWHhnj17Xr16hfVsGAyGtbX1t99+6+LiQiAQfvnll6NHj27atKm6uppIJE6bNu3SpUvz58/Htu3q6nr9+nVWVtbMmTMHul/N4uXlpeoQhqPCwkI5psKUAvKoEkrMY2FhISQRH2JiYn755RdVRwEGETRVlVD676YawkHnHBMVFeXm5oYNWUqQ0sFGCC1dunT//v2bN2/u7Ow0MzNbtmzZ119/PWLEiMEIEqgJHBz2/Z4bSj/spZyxkslkJyenzZs3NzQ08Hg8c3PzDz74IDExUfQQTiS1uQEwSAg4GxABAAw97IQKTptFtm3b9vvvv9fU1AzBvgQCwYIFC3x9fT///HP5aggMDExOTn7z5o1yA0MIEQiE9PT01atXK1hPQUHB999/r5SQgBzmzZsnfnWG3CCPqqWUPH7//fc9J5QEQyYoKEhZ98vDSKjaUkqW4ftWtZT1u6msfhRAA++cq3MHW0HKOq7g+FR/Q3/YK0Jtm8xggPGToQHXdwMAgPK1t7fn5ORYW1tbWVnp6uoO0l74fH5WVlZLS4uPj89AtxUKhXV1da9evfrrr78GIzYlmjdvHvQGcADyiANKGcEB6gAaI77B9y0APcneOR8mHWwwHAzNYa8IaDJg8MB4NwAAKF9TU9PSpUsRQuvWrUtOTh6kvdy4cePSpUvZ2dl6enoD3TYrK8vd3R37G27ABAAAAAAAOCZ75xw62AA3huawVwQ0GTB4YD4TAICi4H4c0Cu4zxEAAAAAQD7QjwKDAeYzAUDlYPxkaGipOgAAAAAAAAAAAAAAAAAAQAlgvBsAAAAAAAAAAAAAAAAAHsB4NwAAAAAAAAAAAAAAAAA8gPFuAAAAAAAAAAAAAAAAAHgA490AAAAAAAAAAAAAAAAA8EBH1QEAAPCgsLAQe8qwmmhra6NSqaqOYqi1trbSaDRVRwEAAAAAAJQgJibml19+UXUU0rS0tOjr66s6CqAa6n98Ak30/v17KpWqo4PnscrCwsK5c+eqOgr8g+u7AQCKmjdvnlp9Xzc3N1+/fr2iokLVgQwpNpv9xx9/NDQ0qDqQ/8/T09PCwkLVUQAAAAAAaB5PT09zc3NVRyHN69evc3NzWSyWqgP5L21tbZcvX37//r2qA1FTyuqfq//xOay8f//+8uXLbW1tqg5ECf7888/r16+/e/dO1YEMorlz586bN0/VUeAfQSgUqjoGAABQGjab7ejoaGlpee3aNTKZrOpwho5AIPDx8cnJycnPz582bZqqwwEAAAAAALh1586dpUuXLl26NDU1Va2uxGxpaaHT6VeuXFm+fLmqYwFgiFy5cuXjjz9ubm7Gwf0WHA5nw4YNeXl5+/btO3jwoJYWXKQL5ASHDgAAP1paWlasWKGnp5eZmTmsBrsRQlpaWsnJyQ4ODsuXL6+qqlJ1OAAAAAAAAJ8KCgqWLVu2ePHilJQUtRrsRgjp6+vr6+vX1taqOhAAhk5tbS2dTsfBYDdCyNTUNDs7Ozo6Ojw8fOnSpRx0eNH0AAAgAElEQVQOR9URAU0F490AAJzg8Xienp4cDufq1avGxsaqDkcFdHV1L126ZGJisnz58qamJlWHAwAAAAAA8KawsHDp0qXOzs4pKSlEIlHV4fTC3NwcxrvBsFJbW4un6WUIBMLOnTtv3779119/TZs27dq1a6qOCGgkGO8GAOCBUCj08/MrKCi4cuXK2LFjVR2OytDp9KtXr7a1tS1fvpzL5ao6HAAAAAAAgB9FRUUrVqxwdHT89ddfSSSSqsPp3ZgxY2C8GwwrtbW1Y8aMUXUUSubg4FBUVLRgwYKlS5fu3LmTx+OpOiKgYWC8GwCAB/v27UtJSfnll19mzJih6lhUbPTo0VevXi0vL1+zZg2fz1d1OAAAAAAAAA+ePHmyaNGi2bNnq/NgN4LxbjD84HK8GyFkYGCQlpb2r3/968yZM/Pnz//7779VHRHQJDDeDQDQeKdPn46MjDx9+vSSJUtUHYtasLOzu3LlSm5u7hdffKHqWAAAAAAAgMZ7+vTpP/7xjxkzZqj/Y3JgvBsMNzU1Nbgc78Z89tln9+/f53K5M2bMuHDhgqrDARoDxrsBAJrt999//+KLL7755htfX19Vx6JG5s6dm5qaeubMmW+//VbVsQAAAAAAAA324sWLJUuWTJw48ddff6VQKKoOpx9jxoypqalRdRQADB28Xt8tYmdnd+/evQ0bNnh7e3/22WcwbyeQBYx3AwA02P379318fHx9fcPCwlQdi9pxc3OLj48PCwtLSkpSdSwAAAAAAEAjvXz50tXV1draOjs7m0ajqTqc/o0ZM6apqam9vV3VgQAwFNrb29++fYvv8W6EEIVC+eGHHy5evPj77787ODg8e/ZM1REBdQfj3QAATVVRUfHxxx8vWLDg5MmTqo5FTW3dujU0NHTr1q3Z2dmqjgUAAAAAAGiY8vJyV1fXsWPH/uc//9GIwW6EEDbwx2azVR0IAEMBm70H9+PdGA8Pj6KiIgMDAwcHhx9++EHV4QC1BuPdAACN9ObNm2XLlllaWqanp+vo6Kg6HPV17Nix9evXe3p6FhQUqDoWAAAAAACgMaqqqhYtWmRmZnblyhV9fX1VhyMrbOAPpvAGw8SwGu9GCFlZWd28eTM4ODgoKMjLy+vdu3eqjgioKRjvBgBoHi6X6+bmxufzf//9dyqVqupw1BqBQPjnP//54Ycfurm5vXz5UtXhAAAAAAAADcBisRYsWGBkZPTHH38YGRmpOpwBMDU11dHRgfFuMEzU1tbq6OiYmJioOpCho6Ojc/jw4WvXrt25c2f69OlwXRfoFYx3AwA0DJ/PX7t2bXl5eXZ2tqmpqarD0QBEIvHixYvjxo1btmwZh8NRdTgAAAAAAECtVVdXu7q6GhgY5ObmGhsbqzqcgdHS0jIzM4PxbjBM1NbWMhgMbW1tVQcy1BYuXPj48eNJkya5uLgcPnxYIBCoOiKgXmC8GwCgYXbu3Hnt2rWsrKzx48erOhaNoaend/nyZR0dnRUrVrS2tqo6HAAAAAAAoKZqampcXV2pVGpubu6IESNUHY48xowZA+PdYJiora0dPpOZSBg1atTVq1ePHz8eHh6+ZMmS+vp6VUcE1AiMdwMANMk333xz6tSp8+fPOzo6qjoWDTNy5Mjs7OyamppPP/0/7N17QExp/D/wZ2qmuXRP0V3FkhTJLd3c1mVDpCK0ZFeKtUouIdKPREIhsUJ23Sq0ubWrJVSUzRJdRKL7RSRdppppZn5/nN93fvPtMk3NNGeaPq+/mnOeOec9c86c0/OZM89ZyGAw8I4DAAAAAAAkTnV19ezZs4lEYlJSkrq6Ot5xegnq3WDgKC8v19XVxTsFbggEgre395MnTz5+/Dh27Ni///4b70RAUkC9GwDQb1y5cmXPnj3h4eGOjo54Z+mXjIyM7t69+++//65evZrD4eAdBwAAAAAASJBPnz7NmDGDxWIlJydramriHaf3oN4NBo6BfH0318SJE1+8eDFz5kx7e3tvb28mk4l3IoA/qHcDAPqH5OTk1atX79ixY8OGDXhn6cfGjx9/8+bN69ev79q1C+8sAAAAAABAUtTU1MycOZPBYDx8+FBbWxvvOEKBejcYOKDejVFSUrpy5cqFCxfOnTtnY2Pz4cMHvBMBnEG9GwDQD2RnZy9evNjJySkoKAjvLP3ejBkzoqOjDxw4cPz4cbyzAAAAAAAA/NXV1c2dO7ehoeGff/7p78VuhJCurm5FRQXcvw5IPTabXVVVBfVurpUrV2ZmZra0tIwbNy4mJgbvOABPUO8GAEi6srIye3t7CwuL6OhoAoGAdxxpsHz58qCgoE2bNl2/fh3vLAAAAAAAAE91dXWzZs36/Pnzw4cPDQwM8I4jAjo6Okwms6amBu8gAPStT58+MRgMqHfzGjVq1LNnz9zd3ZctW7Zy5Uo6nY53IoAPqHcDACRabW3trFmzVFRU4uPjyWQy3nGkx86dOzds2PDjjz+mpqbinQUAAAAAAODj27dvc+bMqaqqevjwoaGhId5xRAMr/8GQJkDqYTs51LvboVAox44di4+Pv3PnzoQJE7Kzs/FOBHAA9W4AgORqaWlZuHBhQ0NDYmKiiooK3nGkTVhY2Pz58xcsWPD69Wu8swAAAAAAAHGrr6+fM2dOcXHxP//8Y2RkhHcckYF6NxggsJ1cS0sL7yCSyNHRMSsrS01NbdKkSceOHcM7DhA3qHcDACQUm83+8ccfs7OzExMT9fT08I4jhWRkZC5evDhmzBh7e/uSkhK84wAAAAAAAPGh0+kODg4fP35MTk42NjbGO44o0Wg0FRUVqHcDqVdeXq6qqiovL493EAmlr6//6NEjPz8/X19fJyenr1+/4p0IiA/UuwEAEmrz5s23bt26cePGmDFj8M4itSgUyu3btwcNGmRvbw+nfwAAAACAAYJOp8+fPz8vLy85OdnExATvOKKno6MD9W4g9crLy2EwE/6IRGJgYOA///yTkZFhbm7+5MkTvBMBMYF6NwBAEh0+fPjYsWPnzp2bOXMm3lmknLKycmJiYkNDw6JFi1paWvCOAwAAAAAA+lZzc/OCBQuys7MfPHgwevRovOP0Cah3g4EA6t0CmjFjRlZWlqmp6bRp0wIDA9lsNt6JQJ+DejcAQOLExcX5+fkdOXLEzc0N7ywDgo6OTmJiYnZ29tKlS1ksFt5xAAAAAABAX2EwGM7Ozi9evPj777/NzMzwjtNXoN4NBgKodwtOQ0Pjzp07hw8fPnDgwKxZsyorK/FOBPoW1LsBAJIlJSVl5cqVnp6emzZtwjvLADJ69OiEhISkpKQNGzbgnQUAAAAAAPQJBoPh5OT05MmTpKSk8ePH4x2nD0G9GwwEUO/uEQKB4O3t/eTJk5KSkrFjx/711194JwJ9COrdAAAJkpeXt2jRojlz5pw4cQLvLAOOnZ1dbGxsVFRUSEgI3lkAAAAAAICIMZlMFxeX1NTUpKSkiRMn4h2nb0G9GwwEUO/uhQkTJvz333+zZs2aN2+et7c3g8HAOxHoE1DvBgBIioqKCnt7e1NT09jYWFlZWbzjDEQODg4RERE7duy4cOEC3lkAAAAAAIDIsFisH3/88cGDB7dv3540aRLecfqcjo5OXV1dU1MT3kEA6CtNTU3fvn3T1dXFO0j/o6SkdPny5QsXLpw7d87GxqawsBDvRED0oN4NAJAIDQ0N8+bNo9FoCQkJFAoF7zgDl5eX17Zt29auXXvv3j28swAAAAAAABHAit23b9++e/eura0t3nHEAbvotaKiAu8gAPSVsrIy9D+7OuiFlStXPn/+nMFgWFhYXL16Fe84QMSg3g0AwB+TyXR2dq6urk5MTFRTU8M7zkB34MCBFStWLF68OCMjA+8sAAAAAABAKCwWa9WqVTdv3rxz587UqVPxjiMm2EWvMKQJkGLY7g31bmEYGxtnZGS4u7uvWLFi5cqV8IsQaQL1bgAAzjgcjoeHR3p6+t27dw0MDPCOAxCBQDhz5oytra2Dg0NBQQHecQAAAAAAQC+x2ezVq1fHx8ffunVr+vTpeMcRHw0NDTKZjF0AC4BUKi8vl5OTU1dXxztI/0ahUI4dOxYfH3/37t0JEya8evUK70RANKDeDQDA2c6dO69cuXLt2rVx48bhnQX8PyQSKT4+ftiwYT/88EN1dTXecQAAAAAAQI9xOJz169fHxsZeu3Zt5syZeMcRKwKBoKmpCdd3AylWXl6ura1NIBDwDiINFi1alJWVpaGhYWlpeezYMbzjABGAejcAAE9nzpwJCQk5c+bMnDlz8M4C/hcajXbr1i0ZGZn58+c3NjbiHQcAAAAAAPQAh8P55ZdfoqOjr1+/Pm/ePLzj4EBHRwfq3UCKlZeXw2AmIqSnp5ecnOzn57d582ZHR8fa2lq8EwGhQL0bACAOra2tnz9/bjfxzp07v/zyy759+9zd3fEIBbqhoaHx119/lZaWLl26tK2trd3cqqoqXFIBAAAAAAD+OBzOr7/+evbs2bi4uAULFuAdBx9Q7wZS5urVqz4+PqGhoZcuXUpJSXn79q2mpibeoaQKkUgMDAy8f/9+ZmbmuHHj0tLS8E4Eeg/q3QAAcbh+/bqlpWVhYSF3SmZmpqurq7u7u7+/P47BAH/Dhg27c+dOSkrK6tWrORwOd/qhQ4dmzJjBOwUAAAAAAEgIPz+/06dP//HHHwsXLsQ7C26g3g2kDIFAOHbsmL+//6pVq6ZOnfrPP//cuHFDSUnpu+++mz179k8//ZSRkYF3Rmkwbdq0rKysMWPGTJ8+PTAwkMVi4Z0I9AbUuwEA4hAREVFYWDhx4sTMzEyEUGFh4fz586dNm3bq1Cm8o4FuTJgwITY2NiYmJiAgACHEZrO9vb23b9/+5s2bR48e4Z0OAAAAAAD8Lzt27Dh69Ogff/zh6uqKdxY8Qb0bSBlbW1uEEJPJZLPZ3IkNDQ3v37//559/Ll++bGRkhF86qaKurn7r1q3Dhw8fOHBg1qxZFRUVeCcCPQb1bgBAn8vLy8O+aq6vr7ezs7ty5coPP/ygr68fGxtLJBLxTge6Z29vHx0dvX///rCwsGXLlkVERHA4HCKRGBERgXc0AAAAAADw//n7+4eGhl64cGH58uV4Z8GZjo5OVVUVXJsJpIaOjo62tnans0gkkoeHx+DBg8UcSYoRCARvb++nT5+WlZWZm5vfvXsX70SgZ6DeDQDocydPniSRSAghFovV2tr6448/ysrK3r17V15eHu9oQFBubm47d+4MDw+Pj4/HLihoa2u7efMmXDUDAAAAACAhAgICDhw4cOrUKTc3N7yz4E9HR6etre3Tp094BwFAZGbMmNHpFWNsNnvLli3izyP1xo8f/99//82ZM2fBggXe3t4MBgPvREBQUO8GAPStxsbGCxcuMJlM7CGHw2Gz2fn5+fv374fRn/uRysrKhISEyspK3htXysjInD17FsdUAAAAAAAAs3fv3qCgoMjISA8PD7yzSAQdHR2EEFycAaSJjY1Nx040iURatmyZgYEBHomkn6Ki4sWLFy9cuHD+/HkrK6v379/jnQgIBOrdAIC+deXKldbW1o7TIyIiVq5cya2DA0mWl5c3fvz4d+/etdteTCYzIiICNiIAAAAAAL6OHDkSGBgYERHh5eWFdxZJoaOjQyAQoN4NpImtrW3HIXra2tr8/PxwyTNwrFy58vnz52w228LC4vLly3jHAd2DejcAoG9FRETw3k+Di81mX7582d7evqGhQfypgODS0tIsLS1ramo6rWt/+fLl1q1b4k8FAAAAAAAw4eHhW7duPX78+Pr16/HOIkEoFIqamhrUu4E0GTVqlIqKCu8UEom0YMECU1NTvCINHCNHjkxPT1+9evWPP/64cuXKxsZGvBMBfqDeDQDoQ8+ePcvOzu503BISiSQrK2tqagr3kJFwnz9/VlVV7WrwGRkZmePHj4s5EgAAAAAAwBw/fnzTpk0HDx7csGED3lkkjo6ODtS7gTQhEAjW1tYyMv+/lMdkMrdv345jpAGFTCYfO3YsISEhMTFxwoQJWVlZeCcCXYJ6NwCgD506dUpOTq7dRFlZWYTQ999//+bNm7CwsHZfUANJs2jRovfv30dGRqqqqna8OwqLxUpJScnNzcUlGwAAAADAQHb27FkfH5/g4OBt27bhnUUSQb0bSB87OzusQ40QIhKJU6dOnTJlCr6RBhoHB4esrKwhQ4ZYWloeO3YMbksmmaDeDQDoK3V1dTExMby3MCYQCAQCwdTU9PHjx4mJicOHD8cxHhAciURau3ZtSUmJv7+/nJwciURqN/fMmTN4ZQMAAAAAGJjOnz/v6em5b9++HTt24J1FQuno6JSWlhYXFz958iQuLi48PPzGjRt4hwJAKLa2ttxxJtva2nbv3o1vnoFJV1c3OTl5+/btmzdvdnR0/PLlC96JQHsE+CICANBHwsLCtm7dyh2uhEgkDho0aO/evWvWrOH9BRboX0pLS/fu3Xvu3Dkikcj9T0teXr66ulpeXh7fbAAAAAAA0ufTp09+fn5RUVG8v7S7cOHCzz//vGfPnoCAAByzSZp3797FxMRUVFSUlpZ+/PixtLS03Ri7R48e3bRpE17xABAek8lUVFRsbW2VlZU1MzN7+fIl3okGtMePH7u5uREIhEuXLtnZ2eEdB/x/UHICAPSVyMhI7E6VJBKJQqH4+/t//Phx7dq1UOzu1/T09KKiojIzM7HfzWFbs6WlBe5SDQAAAADQF06cOHHhwoVly5ZxLzWIjY1ds2aNr68vFLvbUVNTO3DgQFRUVGJi4ps3bzreUM7W1haXYACIColEmjBhAkKIzWbDEQB3U6dOzcrKMjc3nzFjRmBgYMebkzU2Nh44cACXbAMcVJ0AAH3i4cOH79+/JxAIMjIyP/30U3FxcWBgIJVKxTsXEI3x48c/fvz41q1bRkZGMjIyLBYL7loJAAAAACByDQ0N4eHhCKE///zTxcWFwWBcu3bNzc3N29s7NDQU73QSR11d3dPTkzu6cTtUKtXc3FzMkQAQuWnTpiGEjIyMFi5ciHcWgAYNGnTr1q3z588fOnTo+++/b3fPgPXr1+/cufPOnTt4xRuwYDwTIFmOHj2anp6OdwogAhkZGWVlZUOGDBk7dqySkhLecUTs2rVrQi4hPT396NGjIgmDLw6H8/Hjx5ycHAaDMX369EGDBuGdCAhF+H0bAAAAACIUGhq6c+fOtrY2hJCsrOzEiROfP3++cePGI0eO4B1NQpWVlRkaGmLvGC8CgTBt2rTk5GRcUg0E0JcXm6qqqrS0tIkTJw4dOhTvLNJD+H5Qbm6uq6trVVVVdHT0/PnzEUKXL1/GRjtRU1PLz89XV1cXRVIgELi+G0iW9PT0jIwMvFMAYbW0tDQ1Ndna2tra2kpZsbusrOz69evCL6e0tFQky8EdgUAwMjKyt7cfNWpUcXEx3nFA74lq3wYAAACAqLS2toaGhnJLtywW6/nz56NGjdq3bx++wSSZrq6um5ubnJxcu+kkEmnq1Km4RBogoC8vNoMGDZKXl9fT08M7iJQQVT9o9OjR//777/Llyx0cHLy9vd+8ebN27VoCgcDhcOrr63/66SfhVwEEB9d3A8ni4uKC4ALD/q+2tlZZWbmrHxL2a3FxcUuXLhX+yCmq5UiUL1++wPXd/ZdU7pMAAABAvxYVFeXl5YXdEYeLSCTa2dnduXMHhgrsytu3b0eNGtXxv5r79+/PnDkTl0gDAfTlxSklJQXujigqIu8HXbx48ZdfflFUVPz06RPvb00uXrzo5uYmqrUA/uD6bgCA6KmpqUllsRvwB8VuAAAAAABRYbFYwcHBHUswbW1tqamps2bN6ngzRoAZOXKko6MjiUTinSgrKzt58mS8IgEgWlDslmQ//vijs7NzdXU1b7GbQCB4eXmVlJTgGGxAgXo3AAAAAAAAAAAgWeLj44uLizu95JDJZD558mT+/PkMBkP8wfqF3bt3txvC28zMTEFBAa88AICB46+//rpw4QKLxeKdyOFwGAyGu7s7/KBWPKDeDQAAAAAAAAAASJagoCAZmU467EQiESE0Z86cQ4cOdRylGmDMzc2nT5+OvVcIITk5uRkzZuAbCQAwEFRXV2P3qOw4i8lkPn78ODIyUvypBiCodwMAAAAAAAAAABIkKSnp9evX7S4PxAbosLW1/ffff//+++9JkybhlK5/4L3Em8lk2tjY4JsHACD12Gz20qVLa2tr2913gbfB5s2bCwoKxBxsAIJ6NwAAAAAAAAAAIEH279/PvTYZIUQkEgkEwvfff5+ZmZmcnDxx4kQcs/UX06ZNmzRpEvdttLa2xjcPAEDq0el0Jycn7MclMjIynf4Eh8ViLVu2rN3XmUDkoN4N+qtXr165uroaGhqSyWR1dfWxY8cGBQXhG2n16tUUCoVAILS0tOCbpCt83rQbN24YGRkROmNgYCDOkGlpadbW1jQaTUtLy8/Pr7W1tdeN+cwNCQkxNjamUqny8vLGxsa7d++ur6/vq5fUW7wbZffu3Z22OXr0KIFAkJGRMTY2TklJEWc8V1fXTncYrjt37ogzj4Rgs9lhYWFWVlbtpjOZzICAACMjIzk5OR0dnS1btjQ3N3e1kGnTpnV8P7kjTvKfGxQU1G6WqalpH71YAAAAAPSFzMzMlJQU7NpkOTk5AoEwa9as58+fJyYmTpgwAe90/Ym/vz/2NhoYGAwePBjvOABZWFhg/6BqaWl5e3t31ezdu3cTJ05UUFCQkZGZO3euOBPyevv27a+//jp69GhFRUUikaisrDxixIh58+alp6fjFakriYmJysrKt2/fxjtI53rUx0ddd6kQQleuXJk4caKiouLQoUNXr15dVVXFnbV3714TExMlJSUymTx8+PBt27aJ/6a+CgoKv/76a3Jycl1dXUJCwqpVq9TU1ND//DoH09bWlpWVdfDgQTFnG2ig3g36pezsbCsrK01NzYcPH3779u3p06dz58599OgRvqmio6O3bNmCbwY++L9pTk5OHz58GDZsmLKyMofD4XA4bW1tdDq9urqaRqOJLWRubu7s2bNnzpxZU1MTHx9//vz5devW9a4x/7mpqakeHh4lJSXV1dX79u0LCQlxdnbu29fWc9yNghA6e/Ysk8ls14DFYh0/fhwhNGPGjPz8fPHfpDspKamuro7JZFZWViKEHBwcGAxGU1PTp0+fPDw8xBxGEhQUFNjZ2fn6+tLp9HazfHx8QkNDg4ODv3z5cvny5aioqDVr1vRo4fx/hAs/0QUAAACkxv79+wkEAolEIhAIixYtys7OTkxMtLCwwDtX/7NgwYKRI0cihKZPn453FoAQQi9evJg9ezaBQHj58uWxY8e401ksFu8A6yNGjMjMzAwODl6+fPnff/+NR1J07tw5MzOz169fHz16tLS0tKmp6eXLl/v27aurq8vOzsYlEh+SfAvEHvXxEd8uVWxs7IoVK1xcXMrKym7evJmSkvLDDz9why1KTk7esGFDUVHR58+fg4ODw8PDXVxc+upVdUdeXn7BggVnzpyprq7Ggunr6yOE5OTkZGRkWCxWYGBgVlYWXvEGAmL3TQCQPIcPH1ZRUQkPD8cejhgxYt++fU5OTuJP0tzcPHPmzKdPn4p/1T3V0zdNVlaWSqVSqdQRI0YIvhYmk3np0qVHjx79/vvvvQi5b98+TU3N//N//g+BQJgyZYqfn9+2bdu2bdtmbGzc08b858rJyf3yyy8UCgUh5OLicu3atWvXrlVWVmppafUidl8bP378f//9l5CQ0O6EfePGDR0dneLiYlxSEQgE7Ft63ikkEolEItFotPHjx/d6yXh9rITce1+9erV3795169Y1NTW1+4/zw4cPp0+fXrNmjaurK0Jo2rRpGzduDAoK2rVr16hRozouikKh1NfXKyoqcqd4eXktWbJEkLkIoYsXL7q5ufXiJQAAAAB9Kj09/ejRo3inkHQNDQ337t3DLoAdNWoUm80ODAwUyZKnTJni6+sr5EKOHj0qgRe38qGsrIwQev36NY6VLxHy9fWdMmUK3imE4urqmpSUdOfOHd6LPx4+fPjw4cO8vDwTExPuxMTExF9++UVU6+1RFyMjI8PT03Pq1Kn37t3jDoljZGRkZGSkoqIigYMvz5s379u3b9yHou1PibOPz6dLhRD67bfftLW1t27dSiAQzM3NfX19N2zY8OzZM2y0IgUFBU9PT1lZWYTQkiVLbty4ERcXV1paqqen14vYvcD/HDdp0iRjY+OKiory8vKvX7+2tbVNnTr1+++/7/S+xKAX2p3j4G0F/dKXL1++fftWW1vLnSInJ4fLj3fOnTv36dOndhM7vRUv7nr9piUkJAiy/NbW1lOnTllYWOTm5vbutzltbW13796dOnUq9w384YcfOBzOzZs3e9q420XFx8djxW6Mjo4OQkj8P3cS0Pr16xFCp06dajf96NGjmzdvxiMRQghdvXqVz7X/np6e8+fP792SO/1Y9Snh916E0NixY2/cuLFixQoymdxuVmZmJpvNnjx5MncK9sPMe/fudbqov//+m7ecXVpampOTw73mhf9cAAAAQGKVlpZev34d7xSS7t27d0OHDp09e/akSZN4z/hCysjIEEmdOj09PSMjQ/jliI2enp6CgoK6ujreQUTg+vXrpaWleKcQ1uLFi+Xk5G7dusU7MSkpSVtbm/f40Nzc/OrVqzlz5ohqvT3qYgQFBbFYrAMHDvAOo4+ZM2fOhg0bRJWqj4iqPyXmPj7i26VCCJWWlmppaXEXhRWyuZd/3blzByt2Y7BPfceLxPtOt+c4JSUlY2PjmTNnLliwYMKECQoKCvn5+WKLJ906OcdxAJAkzs7Ozs7O3TbDrnEYO3ZsWlpax7lhYWE0Go1AIFhYWAwePJhIJNJotHHjxtnY2Ojq6pLJZGVl5a1bt3Lbs9nsIzbqgfgAACAASURBVEeOGBsby8nJqaioLFy48M2bN4LM9fb25t5/YNiwYRwOx9/fX0ZG5saNG3PnzlVSUtLU1Dx37hzW+OTJkzQajUqlJiQkzJ07V1FRUUdH58qVK9wVtbW17d69W09Pj0KhmJmZxcTEYNMfPXo0ceJEKpWqqKhoamr67du3riYK86ZheMczEVxjY+ORI0dMTU0DAwO/fPnCnd7Q0NDVwWjy5Mkdl/P27VuEUEBAAHcKNsDx8uXLe9q4R4vicDjz5s1TUVFpbW3t9sXGxsaK5Mgp+HKGDRv28eNH7Crg/Px87vS0tDQzMzPsTZ45cyZ3elc7UkpKyqhRo7DhzExNTf/++2+OALulgLDxTBYuXNhueqdhoqOj5eXlEUIqKip//vlnZmamvr6+jIzMsmXLOB0+Vr/++iuJRBoyZAi2wPXr12NF9pqaGg6HExISQqVSFRQUqqurfX19tbW18/Pzu3oHOiWqvZfX5MmTx44dyzslLi4OIcT7xlZXVyOE1q5d2+17i73quLg4Aefu27fv4sWLgiy2HVHt2wAAAEBX4FwjiJKSkr5YrID9LLEtR5zi4+PxjiAaCKHY2Fi8U3RJ8H1jwYIFVCqVTqdjDxkMxtatWzdu3GhmZsZtc/v27VWrVmF/8+mSd9od6NhZ7thz56O1tZVCoQwaNKjbF8InmCD9rD/++GP8+PFkMplGow0dOnTv3r2cLnpt2KXQWIkDu/B569atWJvo6OjU1FSs8nvixAlOh/7Uzz//jP1tZGT04sULDofj7u5OpVKVlJRu3rzZ1UvDpY/Pq2OXisPhzJ49W09Pj/vwt99+Qwilp6d3uoSFCxdSqVTJ7ONzCRIPCKLj8Qf+2wCSRcBzJJ1O596qxcTEJCQkhPcQzOFw9uzZgxB69uxZU1PT58+fsUsp7969W1NT09TUtHHjRoRQVlYW1jggIEBOTu7ixYt1dXWvX7+2sLBQV1evqqoSZK6TkxPv+dLf3x8h9ODBg7q6utraWnt7ezKZjJ2QeOd++/bt06dPtra28vLyDAYDm7tlyxYymXz9+vWvX7/u3LlTRkYmMzOzsbFRSUkpJCSkubm5qqpq8eLFNTU1nU4U/k3jdKh3e3t7Z2dn81lmXV1dUFDQmDFjDh8+3NjY2G0G/h4/fowQCg0N5Z1IpVJ5i7kCNhZwUQwGo6ys7MSJE2QyWcASIV71bmx4Ox8fH+50R0fH6OjojvXuTnckDodz7dq1wMDA2traL1++WFpacv+B479bCqirendXYfLy8mg0Gvdf2B07dpw9e5b7rHYfqxUrVnDr3RwOJzQ0lFvv5ub39vY+ceLE4sWL37x509VK2xHt3sur4z9nr1+/Rgjt3r2bOwUbZs7R0bHbpZWVlZmYmLBYLAHn7tu3T1dXV0VFhUQiGRgYLFy48N9//xUkNtQgAAAA9DU41+BoINe7pYbU1LsvX76MELp16xb28O7du/fu3UtNTUUIvX37Fpu4bt26xMRE7G/+XfJ23YHnz5932llu18Xg4927dwghS0vLblsKEqyrflZYWBhC6MCBA1++fKmtrf3tt99WrFjB6aLX1tbWZmBgoK+v39bWxl37pk2bwsLCsL+xC/+xenfHF+vk5CQrK1teXs6dsnz5cu773w6OfXxenda7Hz16RCKRjh8/Xl9fn5OTM2rUqDlz5nT69KamJkVFxY0bNwoSEsd6NxAVqHcDSSf4OZLBYBw7dow75NPgwYMfPXrEnYvVuxsaGrCH2DhT3NLtv//+ixDCrvqk0+kKCgqurq7c52JzsS9X+c/ldFHvbm5uxh7+8ccfCKGcnJxO5548eRIh9P79ew6H09zcTKPRuCui0+lkMnn9+vU5OTkIoTt37vC+9k4nCv+mcTgc7NaIvPjUuxkMxtChQ+3s7Orr63uapFNJSUkIoaNHj/JOVFJSsrKy6mljARc1ZMgQhNCgQYOOHTsmYIUXr3p3XV2dvLy8qqoqdh1EYWGhrq5ua2tru3p3VztSu2UGBwcjhD59+sThu1sKrtN6N/8w2Lfxly5dunLliq+vL+8Te1Hv5uYX8B0Q+d7Lq9N/zubOnaumpvbgwYPm5ubKysq4uDgCgTB//vxul7Zhw4ZTp04JPrekpOTFixcNDQ2tra3p6enjxo2jUqncoxAf8P8ZAACAvgbnGhxBvVsKSE29u6GhgUqlrlmzBnu4efNmJpPJZrO1tbWDgoKwiRYWFlgHrdsuebvuQFedZcHr3c+fP0cIff/99/yb9TQYbz+LwWCoqKhMnz6d+9y2trbw8PB2q+DttWH1ce7POpuamvT19bm/8+Zf775//z5CiPvefvv27bvvvuMtnXPh28fn1WmXisPh7Nq1i1us0NXVLS0t7fTp/v7+I0aMEPBVQL1bCnQ8/sD43aC/IpFIGzdufPPmTUZGxqJFiz59+uTi4vL169dOG2M/5+Het5dEIiGEmEwmQig3N7exsZF74TNCaOLEiXJycs+ePet2riAhuSvqKhU29+3bt3Q63dTUFJtFpVI1NTXz8/ONjIwGDx7s5uYWGBhYVFSEze10ooB5un3T2l3fzX9pr169mjVrlq2tbVhYWFNTk+BJOoUNqM3dTBgGg0GlUnvaWMBFlZaWfvr06cqVK7///vu4cePEPGZ0jygrKy9fvvzr168xMTEIobCwsPXr13N/p8bV1Y7Urhm2Z7JYrI4r4t0thcQ/zNq1a52dnb28vOLi4g4dOiT86gRZKZfI995uxcTEuLi4rFy5Uk1Nzdra+s8//+RwOIMGDeL/rIqKilu3brm7uws+V09Pb9y4cQoKCnJycpaWltHR0c3Nzdi/1wAAAAAAAOBOQUFh3rx5WEm6tbWVSCQSiUQCgeDk5IQNf5yXlzdu3Disz9LTLnmvO8u88ZAA4z73NBhvP+v169d1dXW8o5PLysp27H3z9trWrFmjrKwcHh6Ozbp06dKiRYuUlJQEeUUzZswYMWLE+fPnORwOQigmJsbV1ZV3qGveNeLYx++Wv7//mTNnHjx40NjY+OHDBysrqylTpnQc1D4+Pj4uLu7evXsivAUC6Heg3g36vcmTJ//555/r1q2rqal5+PBhT59eV1eH/ueUxqWiooJdNst/rghhJ5Jdu3YR/kdxcTGdTqdSqcnJyTY2Nvv37zcyMnJ1dW1ubu50Yo9WJ+CbFh4ezq0bdkpZWXnXrl1Pnjxhs9mWlpZ79+7lrZ43NjYSumBpadlxaZqamgih+vp67hQ6nd7S0qKlpdXTxgIuikQiaWhozJ49OyYmJjc3F/v+XGJhd608ffp0XV3dtWvXvLy8OrbpakdCCN29e3fatGkaGhpkMnnbtm1iCMwnDGb//v2NjY2i/Zqh25VyiXbv7ZaysvLp06fLysrodHphYeGRI0cQQtra2vyfFRIS4uHhwXtvVcHnYszMzGRlZbFfZQIAAAAAACAJXF1dq6qqMjMzExMTsdFHEULOzs5ZWVmFhYWJiYlLly7FJva0Sy58Z9nAwIBCoXT7/7MwtQKso6qiotJxVle9NgUFhbVr1z59+hS7ivzUqVPYMK2CIBAIXl5eHz58ePDgAULojz/+4A7q3RGOfXz+KisrQ0JC1q5dO2PGDHl5eUNDw6ioqIqKCuznv1wxMTEHDx589OiRgYFBT1cBpAnUu0G/5OTk1O4bwh9//BH16t672Dmm3Tmprq5OV1e327kipKGhgRDiDr+FwW4vO3r06Nu3b1dUVPj5+cXGxh4+fLirifyJ8E1rR15efvPmzZmZmerq6nZ2dlu3bsVGt1BQUOjqxyad3tXd0NBQUVGRe3tlhND79+8RQmPGjOlp4x4tCiE0fPhwWVnZ3NzcXr0BYmJubm5pafnvv/96enq6uLioqqp2bNPVjlRSUuLo6Kipqfns2bNv376FhISIITCfvRohxGQyvb29jx49mp6eHhQUJJ6VdiSqvbenMjMzEULTp0/n06aqqurKlSvY9xw9ncvFZrPZbHan9zcHAAAAAAAAF/PmzVNUVLx161ZKSoqdnR020cbGRktL6/r168+fP+f+n9yLLnkvOsu8yGTynDlzPn/+/OTJk45za2tr16xZ07tgXNhVL58/f243nX+vbePGjSQSKSwsLCUlRU9Pr+NgpHy4u7tTKJSzZ8++fftWSUlp6NCh/Nvj0sfnr6CggMVi8V4wpKSkpKamxtuLP3HixKVLl5KTk7u9rghIPah3g36ptbU1Ly+Pdwp2299eHDRNTU0VFBSwIbowz549YzAY48eP73auCOnp6VEolKysrHbTKyoqsFeqoaFx4MABCwuLvLy8Tid2u4pev2mVlZWrV6/udvkUCmX9+vUvXrwYNWrU9u3bu23fEZFItLe3T0lJYbPZ2JS//vqLQCA4ODj0tDH/uV++fFm+fDnv0rBzJ3ZXa0mGFTevX7++adOmTht0tSNlZ2czmcz169cbGRlRKBQCgdDnWbsOg/n11189PDw2bdrk6+u7b98+PiVpIpEo+Pgq/FfaFeH33p6KiooyNDScOnUqnzYhISFubm5qamo9msv7u0iEEHavzilTpggZGAAAAOh3EhMTlZWVb9++LfLGgkhLS7O2tqbRaFpaWn5+fq2traJqPKDARpRWFArFwcHh+vXrVCpVRub/FaZkZGQWL178xx9/DB48mEgkYhN72iXvXWe5ncDAQDKZ7Ovr2/Ha8JycHCybMLUCAwMDNTU1bGxrXvx7bbq6ukuWLLl+/fru3bt9fHx69IpUVVWXLl2akJBw+PBhDw8PAZ8l5j4+f9gXCVjZHdPQ0FBbW4v14jkcjp+fX3Z2dkJCQruL7qUVHB75g3o36K8cHR3j4uLq6uq+fft28+bN7du3L1y4sBf1bgqFsnnz5vj4+EuXLtXX12dnZ69bt05LS8vT07PbuQghNTW1ioqKoqKihoYGYYY8plAoq1evvnr1amRkZH19PYvFKisrq6ysrKio8PLyys/PZzAYL1++LC4utrS07HSiIGvp6ZvG4XCam5tv3Lgh4LhgCCESifTTTz9hNwjthd27d1dXV+/Zs6epqSk9PT00NNTd3X3kyJHY3ICAAGVlZe6/Bfwb85krLy+flJSUnJxcX1/PZDJfvny5atUqeXl5X1/f3sUWmyVLlqirqzs6OhoZGXXaoKsdSV9fHyF0//79lpaWgoICAQegF1JXYRBCJ0+e1NHRWbx4MUIoODjYxMRkxYoV3J+5tftYDR8+vLa2NiEhgclk1tTU8F4d0KOVdkvIvZe/SZMmFRcXt7W1FRUVbdmy5f79++fOneOOwN5u30YIVVdXnz9/vqsvNvjMLS8vj4mJqaurYzKZ6enpa9as0dfXX7duXV+8KAAAAECSYSPV9kXjbuXm5s6ePXvmzJk1NTXx8fHnz5/ncy7uUeOBBjaiFHN1dX379u38+fN5J7q4uOTl5Tk6OnKndNslb6erznKPeu7m5uaXL1/OycmxtbVNTEz89u0bk8n8+PFjVFTUzz//jA2r3dNgvMhk8s6dO1NSUjZu3FheXs5msxsaGvLy8rrttW3evLmtre3r168zZszgs/xOX+y6detaW1vv3LmzYMGCbhPyEmcfnw9DQ8Pp06dHRUWlpKQ0NzeXlpZibzU2NkteXt6hQ4eioqJIJBLvKCs9vbq/H4HDYze6+iUCALgQ8J7OSUlJS5cuHTZsGJlMlpOTGzlyZGBgYEtLCzY3PDycRqMhhAwMDFJTUw8ePKisrIwQGjJkyOXLl2NiYoYMGYIQUlVVvXr1KofDYbPZoaGh3333HYlEUlVVdXR0fPv2LXdd/Oe+ePFi6NChVCrVxsbG19cXu+vCd999V1hYeOnSJWzECV1d3ZycnJMnT2KpsLlnzpzBishDhw599+4dh8NpbW318/PT19cnEokaGhpOTk65ublFRUVWVlaqqqqysrLa2tr+/v5YvazjRCHftPj4eD6/h9q1a1dPNqNQHj9+PGnSJDKZrKWltXXrVm5CDoeze/duRUXFe/fuCdKY/1wHBwdDQ0MFBQUymTxs2DBXV9fs7GxB4onz3s3cjaKurr5hwwZs4rZt254+fYr9vWvXLmxANBkZGRMTk9TUVE4XOxKHw/Hz81NTU1NRUXFxcYmIiEAIDRs2bPv27fx3y27V19fb2dlhVxnLyMgMHz58//793LmdhlmwYAGBQFBTU8NeyKZNm7BrOpSVlZ8/f8753x+rqqqqL1++TJ8+nUKhGBoa/vrrr1u3bkUIDR8+vKSkJCQkBPvQ6enpXbx4kc9Ke7R1eic9Pd3a2po7FJ2mpqaVldXjx4+xubNmzVJRUSESiaqqqvPmzcMuu+bquG/7+vq6ubl1tS4+czdv3jxs2DB5eXkikairq+vh4VFRUSFIfrifOAAAgL42cM41S5cuNTQ0ZLPZ2MPQ0FACgfDmzRvhG/eagP0ssS1H8kngRkQIxcbGinaZItTTfYPBYIwdO5b7pmFYLNbYsWNZLBbvRD5d8o7dga46y+26GIIkLCkp2bJli5mZmYKCgqysrIqKyrhx437++ecnT550G6zb7j+Hw4mIiDAzM6NQKBQKZdy4cSdPnuR00WsrKSnhppo+ffrZs2d5c544cQLrFdJoNAcHBz4vdty4cTt27BDktYuW4H18/l2qz58/+/j4DB8+nEwmKygoWFtb//nnn9is7OzsTksZoaGh3cYTZx9fOkjg4bHj8WdAbAnQjwyc/59APwXnQiCtYJ8EAADQ1/r6XMNms+Pi4n777be+W4UgmEymgoKCu7s7d0pOTg5C6ODBg0I2FkZ/qXfDRuRDyurdQPzs7e0/fPiAdwqJ01/6+HB45KPj8QfGMwEAAAAAAAAA0P+wWKzg4OCRI0dSqVR1dXVDQ8Pg4OAlS5YghNLS0vT19QkEAnaBZGRkpLy8PI1Gu3nz5g8//KCkpKSrq3v16lVsOe0at9PY2EjoQqeDCn748KGxsREblwCD/WLv9evXQjaWSrARAehT3PFMXr9+jf1kFt88QHBweBQG1LsBkAb5+fldHaEIBIKrqyveAUH/AzsVAAAAACRcSEhIQEBAaGhobW1tUlJSS0uLioqKiooKQsjGxubp06fcluvXr9+0aVNzc7OiomJsbGxhYaGRkZGHhwdWCWrXuB0FBYWuLijLyMjo2L6qqgohpKioyJ1CoVCoVGp1dbWQjaUSbEQgZgOtm+Pn51dQUPDu3bvVq1fv27cP7zigB+DwKAxiX68AACAGxsbGHJHefwAA2KkAAAAAIOESEhLGjx/v4OCAELKwsFi4cOHZs2cZDAb3jtAdWVlZUSgUhJCrq2tqampJSQmfe9j0TmtrK0JIVlaWdyKJRGpubhaysVSCjQjEbKB1c2g0mrGxsY6OzsmTJ01MTPCOA3oADo/CgOu7AQAAAAAAAAD0P9hNz7gPWSwWiURq16/uClYv4P7SX4SwWkNbWxvvRAaDgd1VT5jGUgk2IgB9KigoiMVilZSULFiwAO8soGfg8CgMqHcDAAAAAAAAAOh/7O3t//vvv5s3bzY3Nz9//jwhIWH+/PkC1gIE19OxTTU1NRFC9fX13Cl0Or2lpUVLS0vIxlIJNiIAAHQKDo/CgPFMAAAAAAAAAAD0P4GBgf/995+7u3tjY6OWltaSJUv2798v8rVgY5sK3t7Q0FBRUbG4uJg75f379wihMWPGCNlYKsFGBACATsHhURhwfTcA4paYmKisrHz79m2RNxZEWlqatbU1jUbT0tLy8/PDRlPqXWM+c5lMZkBAgJGRkZycnI6OzpYtW2D0OsDVXz4CQUFB7b7fNjU1FVUMAAAAAAgvNze3sLCwpqaGyWSWlJRERkaqqqriHQoRiUR7e/uUlBQ2m41N+euvvwgEAjYGqzCNpRJsRNBP9ZdOjSCN2Wx2WFiYlZUVn4W0tLQYGxvv2rVLBOmBYODwKAyodwMgbj366ky0d9LIzc2dPXv2zJkza2pq4uPjz58/v27dut415j/Xx8cnNDQ0ODj4y5cvly9fjoqKWrNmjQhfCOjX+stHAAAAAAASbsOGDfr6+o2NjXgHaW/37t3V1dV79uxpampKT08PDQ11d3cfOXIkNjcgIEBZWTkpKUmQxlIPNiLop/pLp6bbxgUFBXZ2dr6+vnQ6nc9K/f393759K7LXAAQAh0ehcACQJM7Ozs7OzninkFpLly41NDRks9nYw9DQUAKB8ObNm1405jO3sLBQRkZm7dq13EVhXwLn5eX10esSp9jYWJEcOUW1HNAjPfoI7Nu37+LFi2JMhzPYJwEAAPQ1kZ9rkpOTBw0axO3bkkikUaNG3bhxg8PhnDhxAhs2lEajOTg4nDx5kkajIYS+++67wsLCM2fOKCkpIYSGDh367t27do1Fku3x48eTJk0ik8laWlpbt27FbjuG2b17t6Ki4r179wRpLCqi6meJvL8GG1FwCKHY2FiRL1ZUoC8vNiLs12dlZS1evPjSpUvm5uZjx47tao1PnjyZPXs2Qsjf31+kL0UiSGwfHw6Pgut4/IGeLZAsUnmOZLPZcXFxv/32G74xmEymgoKCu7s7d0pOTg5C6ODBgz1tzH9uTEwMQujcuXPcuWlpaQihsLCwvnhdYiax50JJ1h8/AhyodwMAAACiJvJzzcmTJ318fLgPW1tbN23aRCaT6XS6CNciHSS23g0bUXBQ78ZXf+zUCN548uTJXdW76XS6lZVVXl4e1LvFsxwuODwKruPxB8YzAUD0WCxWcHDwyJEjqVSqurq6oaFhcHDwkiVLEEJpaWn6+voEAiEiIgIhFBkZKS8vT6PRbt68+cMPPygpKenq6l69ehVbTrvG7fT0LrofPnxobGzU19fnThk2bBhC6PXr1z1tzH+ujIwMQohKpXLnfvfddwihN2/eCPwWgv5NCj4CAAAAAJBwVVVVGzdu/Pnnn7lT5OTk9PX1mUwmk8nEMRgQHGxEIMmkoFMjkh6Qv7//L7/8oqGhIfhTgPDg8CgkqHcDIHohISEBAQGhoaG1tbVJSUktLS0qKioqKioIIRsbm6dPn3Jbrl+/ftOmTc3NzYqKirGxsYWFhUZGRh4eHtjxq13jdrC76HYqIyOjY/uqqiqEkKKiIncKhUKhUqnV1dU9bcx/rrGxMfrf1W3sNzg1NTXdvXNASkjBRwCzY8cOVVVVOTk5Q0PDRYsWZWZm9ur9AAAAAIDoUalUEol07ty56upqJpNZUVFx9uzZgIAAV1dX7HfcQPLBRgSSTAo6Nb3oAbXz5MmTwsLC5cuXC9geiAocHoUE9W4ARC8hIWH8+PEODg5UKtXCwmLhwoUpKSkMBoPPU6ysrJSUlDQ0NFxdXZuamkpKSkSeCrsLs6ysLO9EEonU3Nzc08b855qZmc2dO/fkyZPJycktLS1VVVXx8fEEAgG+hBw4pOAjgBBatWrVrVu3SktLGxsbr169WlJSMnXq1NzcXJEHAwAAAEAvYLfDysnJGTFiBJVKNTExiY6OPnjw4O+//453NCAo2IhAkklBp6anPaB2mpubfXx8IiMjex8X9BYcHoVExDsAAFKopaWFQqFwH7JYLBKJ1O4c0xU5OTmEUF+UhrFIbW1tvBMZDAbvwCMCNu52UTExMX5+fitXrqytrdXS0po8eTKHw+G90wKQblLwEUAI6enp6enpYX9bWlpGR0ebm5ufPHkS/uEDAAAAJIStre0///yDdwogFNiIQGJJQaempz2gdnbu3Ll27VodHZ3exwVCgMOjMOD6bgBEz97e/r///rt582Zzc/Pz588TEhLmz58v4HlRcD0d5wu7IW99fT13Cp1Ob2lp0dLS6mnjbhelrKx8+vTpsrIyOp1eWFh45MgRhJC2trYoXjfoB6TgI9CRmZmZrKzsu3fvRJUfAAAAAAAAILGkoFMjTA8oLS0tOzt7zZo1vXhRAOAOru8GQPQCAwP/++8/d3f3xsZGLS2tJUuW7N+/X+Rrwcb5Ery9oaGhoqJicXExd8r79+8RQmPGjOlp4x4tCiGEjXo8ffp0wdOCfk0KPgIdsdlsNptNJpN7mhMAAAAAAADQ70hBp0aYHtC5c+cePHggI/O/LpPdv3///v37MzMzJ0yYIHhmAMQPru8GQPRyc3MLCwtramqYTGZJSUlkZKSqqireoRCRSLS3t09JSWGz2diUv/76i0AgODg49LRxjxaFEIqKijI0NJw6daroXxWQSFLwEUAIzZkzh/dhZmYmh8OZMmVKnwcFAAAAgGQ4fPjw4MGDCQTC6dOnxb/2adOmdbzeU0FBQfxJ+i98t2BQUFC7zWdqair+GKDXpKBT09MeEK/o6Gjem2fW1NQghPz9/TkcDhS7pQO+R0gMm80OCwuzsrIS+ZKh3g2A6G3YsEFfX7+xsRHvIO3t3r27urp6z549TU1N6enpoaGh7u7uI0eOxOYGBARgt0QQpDH/uZMmTSouLm5raysqKtqyZcv9+/fPnTuHDWEGBgLp+AiUl5fHxMTU1dUxmcz09PQ1a9bo6+uvW7cOv/gAAAAAEKstW7Y8ffoU7xT/i42NDd4R+hMJ3IKgH5GOTg3/xmAgw/0IWVBQYGdn5+vrS6fTRb5wqHcDIHrBwcE5OTmqqqrY1/hycnImJibx8fEIoYiIiIkTJyKE/Pz8Fi5cGBkZGRYWhhAaM2bMhw8foqKiNm/ejBCaO3duQUFBu8bCBxs9evS9e/eSkpIGDRrk5OT0008/nTp1ircB7w+p+DfmP1dFRcXc3By7jXV+fn5qaioMZjKgSMdHYO7cubt27dLV1aXRaEuWLLG2ts7IyIDbrgIAAABAPCgUSn19Pe/1lZ6entu2bcM7F+iBixcv8m7BnJwcvBOBHpCOTg3/xhkZGTY2Ntra2s+ePXv16pWWlpa1tXVKSorwIQHg79WrV9u3b1+3bp25uXlfLJ/Qo3GCAOhrLi4uCKFr167hHUQokZGRBQUF2AkPmqavYAAAIABJREFUIcRgMLZv3x4ZGfn161cB74MMJFZcXNzSpUuFP3KKajmSCT4C/ZF075MAAAAkQX8817x///677747deqUl5cXvklKS0uXLVuWlpbWu6eLqp/V7/prOG7BoKAgAwMDNzc3US2QQCDExsYuWbJEVAsUrX63b3QLOjUDxEDu40vCOc7S0rKlpSUrK0uYhXQ8/sD13QCIWFVV1caNG3/++WfuFDk5OX19fSaTyWQycQwGgHjARwAAAAAAIvf48eNJkybRaDQlJSUzM7P6+nqEUGpqqomJibKyMoVCMTMzu3fvHkIoPDxcXl5eRkZm/PjxQ4YMIZFI8vLyFhYWtra2enp6FApFRUWFe5X08ePHKRTK4MGDvby8tLS0KBSKlZXVs2fPuorBYrECAgL09fWpVOqYMWNiY2P5xOPV2NjYcTBujKWlpSDvwMGDB729vXvx1kkI2IKgf4FODRAnOEKKHNS7ARAxKpVKIpHOnTtXXV3NZDIrKirOnj0bEBDg6uqqpKSEdzoA+hx8BAAAAAAgWk1NTQ4ODs7OzrW1tQUFBSNGjGAwGAih6urqpUuXFhUVVVRUKCgorFixAiHk4+OzdetWDodz6tSpjx8/VlVV2dnZvXz5cseOHS9fvqytrV21alVoaOirV68QQhs3bnR3d6fT6d7e3kVFRS9evGhra5s1a1ZpaWmnSbZv337o0KGwsLDKysoFCxYsX778+fPnXcXjpaCgwOlCRkZGt+9AeXn5o0ePnJychH0rcTJgt+COHTtUVVXl5OQMDQ0XLVqUmZkpsvcU9DHo1ACxGbBHyD4F9W4ARAy7NUROTs6IESOoVKqJiUl0dPTBgwd///13vKMBIA7wEQAAAACAaBUVFdXX148ePZpCoQwZMuTGjRvq6uoIIWdn5z179qiqqqqpqTk4OHz58qWmpob7LBMTExqNNmjQoGXLliGE9PX11dXVaTQaNr5Efn4+tyWRSBw1ahSZTDYxMYmMjGxoaIiOju4Yo6WlJTIy0tHR0cnJSUVFZdeuXSQSKTo6uqt4InTw4MFff/1VRqa/9t8H5hZctWrVrVu3SktLGxsbr169WlJSMnXq1NzcXOGXDMQAOjVAbAbmEbKvEfEOAIAUsrW1/eeff/BOAQBu4CMAAAAAABEyMjIaPHiwm5ubt7e3u7u7gYFBxzYkEgkhxGKxOs6Sk5NDCLW1tfG27GpEggkTJtBoNN5KAdfbt2/pdLqpqSn2kEqlampq5ufnCxJPGBUVFbdu3QoNDRXtYsVpYG5BPT09PT097G9LS8vo6Ghzc/OTJ09GRkaKZPmgr0GnBojHwDxC9rX++v0wAAAAAAAAAICBgEqlJicn29jY7N+/38jIyNXVtbm5GSF09+7dadOmaWhokMlk7nClwiOTybzX0HE1NTUhhHbt2sUdlrS4uJhOp3cVj5cwY5uGhIR4eHhQKBRRvUDxG+BbEGNmZiYrK/vu3TvhXyAAQJrAEbIvQL0bAAAAAAAAAIBEGz169O3btysqKvz8/GJjYw8fPlxSUuLo6Kipqfns2bNv376FhISIZEVMJrOurk5XV7fjLA0NDYRQWFgY78ik6enpncZr99xej21aVVV15cqV9evXi+TV4WjAbkEuNpvNZrPJZLLQLxEAIG3gCClyUO8GoP85fPjw4MGDCQTC6dOnxb/2vXv3mpiYKCkpkcnk4cOHb9u2rbGxsV0bNpsdFhZmZWXV8elXrlyZOHGioqLi0KFDV69eXVVVJZbUQHpI/v4PAAAAANGqqKjIy8tDCGloaBw4cMDCwiIvLy87O5vJZK5fv97IyIhCoRAIBJGs69GjRxwOp9Pr0fT09CgUSlZWliDxRBIGIRQSEuLm5qampiaqBeJiYG7BOXPm8D7MzMzkcDhTpkwRfslAOuDbr8F01XMPCQkxNjamUqny8vLGxsa7d++ur6/nbQD9ehEamEfIvgb1bgD6ny1btjx9+hSvtScnJ2/YsKGoqOjz58/BwcHh4eEuLi68DQoKCuzs7Hx9fel0ervnxsbGrlixwsXFpays7ObNmykpKT/88AN3nCkABCHh+z8AAAAARK6iosLLyys/P5/BYLx8+bK4uNjS0lJfXx8hdP/+/ZaWloKCgmfPnvV6+Ww2++vXr21tba9fv/bx8dHX13d3d+/YjEKhrF69+urVq5GRkfX19SwWq6ysrLKystN4vQ7Dq7q6+vz585s2bRLJ0nA0MLdgeXl5TExMXV0dk8lMT09fs2aNvr7+unXrhF8ykA749msQ3557amqqh4dHSUlJdXX1vn37QkJCnJ2duXOhXy9aA/MI2ee6uuAcAFw4Ozs7OzvjnaIfKCgoQAidOnVK/KueN29eW1sb9+GSJUsQQiUlJdjDrKysxYsXX7p0ydzcfOzYse2eO336dG1tbTabjT2MiIhACKWlpYknuUjExsaK5MgpquUMTBK7//drsE8CAADoa70+1xQVFVlZWamqqsrKympra/v7+2OnYz8/PzU1NRUVFRcXF+wfy2HDhm3evJlGoyGEDAwMUlNTDx48qKysjBAaMmTI5cuXY2JihgwZghBSVVW9evUqh8Px9PQkkUg6OjpEIlFJSWnRokWFhYXYeo8cOYI1lpeXX7x4MYfDaW1t9fPz09fXJxKJGhoaTk5Oubm5XcUTnq+vr5ubm0gWJap+Vu+WMzC34ObNm4cNGyYvL08kEnV1dT08PCoqKoRZIEIoNjZW+GB9BPryvYBjv4Z/z93R0bG5uZn7ELvEh7sDS0G/nhfuffyBeYRMT0+3trbW0tLCqtOamppWVlaPHz/u3dI6Hn+gZwskC5wjBYTjebEdbDDB/Pz8dtMnT57c8aw5fPjw8ePHcx/evHkTIXT58uU+Tyk6uJ8LAac/7P/9EeyTAAAA+ppknms8PT3V1NTwTtHn8K1396kBsgU5UO+WRpLQr+m0596Oj48PQujdu3fYQyno1/OS7j7+ADlCdjz+wHgmAAjl8ePHkyZNotFoSkpKZmZm2JhWqampJiYmysrKFArFzMzs3r17CKHw8HB5eXkZGZnx48cPGTKERCLJy8tbWFjY2tpiwySpqKhwb7l7/PhxCoUyePBgLy8vLS0tCoViZWXF5wcsLBYrICBAX1+fSqWOGTMGO852FY+X8HfRLS8vp1KphoaGgjQ2MjL69OkT9yE2yJeRkZEgzwUSCPb/Hu3/AAAAAJBMLBYL7whAKLAFgZCgX8NfQUGBiorK0KFDsYfQr+9fBuYREurdAPReU1OTg4ODs7NzbW1tQUHBiBEjGAwGQqi6unrp0qVFRUUVFRUKCgorVqxACPn4+GzdupXD4Zw6derjx49VVVV2dnYvX77csWPHy5cva2trV61aFRoa+urVK4TQxo0b3d3d6XS6t7d3UVHRixcv2traZs2aVVpa2mmS7du3Hzp0KCwsrLKycsGCBcuXL3/+/HlX8XgJeRddOp2enJzs4eEhJycnyDu2c+fOqqqqEydONDQ05ObmhoeHz5kzp3+M/QQ6gP2/p/s/AAAAAAAAQNJAv6YrTCazvLw8IiLi/v37J06c4PZ6oF8PJB/UuwHovaKiovr6+tGjR1MolCFDhty4cUNdXR0h5OzsvGfPHlVVVTU1NQcHhy9fvtTU1HCfZWJiQqPRBg0atGzZMoSQvr6+uro6jUZzc3NDCOXn53NbEonEUaNGkclkExOTyMjIhoaG6OjojjFaWloiIyMdHR2dnJxUVFR27dpFIpGio6O7iidCwcHBWlpaQUFBArafOnWqn5/fxo0blZSUTE1NGxoazp49K9pIQGxg/+/p/g8AAAAASbNz587o6Ohv374ZGhpev34d7zigx2ALAuFBv6Yrenp6urq6gYGBhw4dWrp0KXc69Ov7i4F8hIR6NwC9Z2RkNHjwYDc3t8DAwKKiok7bkEgk1MXvR7BvR7l3McZaMpnMTpczYcIEGo3Ge9bkevv2LZ1ONzU1xR5SqVRNTc38/HxB4gkjPj4+Li7u3r17ioqKAj7F39//zJkzDx48aGxs/PDhg5WV1ZQpU7r6chtIONj/e7r/AwAAAEDSBAcHt7a2cjicjx8/Ojs74x0H9BhsQSC8Ad6v4aO0tPTTp09Xrlz5/fffx40bxx3DBPr1/cVAPkJCvRuA3qNSqcnJyTY2Nvv37zcyMnJ1dW1ubkYI3b17d9q0aRoaGmQymTt0l/DIZDLv98lcTU1NCKFdu3Zxh+gqLi6m0+ldxePV63G+YmJiDh48+OjRIwMDAwHzV1ZWhoSErF27dsaMGfLy8oaGhlFRURUVFaGhoQIuAUgU2P97tP8DAAAAAAAAJNBA7tfwRyKRNDQ0Zs+eHRMTk5ubGxwcjKBfD/oJqHcDIJTRo0ffvn27oqLCz88vNjb28OHDJSUljo6Ompqaz549+/btW0hIiEhWxGQy6+rqdHV1O87S0NBACIWFhfGO0pWent5pvHbP7d04XydOnLh06VJycrK2trbgL6GgoIDFYvE+RUlJSU1NLTc3V/CFAIkC+79IXh0AAAAAAAAARwOzXyO44cOHy8rKYj136NeDfgHq3QD0XkVFRV5eHkJIQ0PjwIEDFhYWeXl52dnZTCZz/fr1RkZGFAqFQCCIZF2PHj3icDidfjeL3QY6KytLkHhCxuBwOH5+ftnZ2QkJCQoKCj16LnZSr6ys5E5paGiora3V09MTMhXABez/Qi4NAAAAAKJy+PDhwYMHEwiE06dP4xjDy8uLe00lNobvtGnTOl5uKfh/EWlpadbW1jQaTUtLy8/Pr7W1FZuekJDAXZrYRvLtaxK7EVHXG6JbQUFB7bY+d7QKqdyI/dQA7Nfw9+XLl+XLl/NOwWrcWM8d+vXiJ8mHR4QQm80OCwuzsrLq6QL79BwH9W4Aeq+iosLLyys/P5/BYLx8+bK4uNjS0lJfXx8hdP/+/ZaWloKCgmfPnvV6+Ww2++vXr21tba9fv/bx8dHX13d3d+/YjEKhrF69+urVq5GRkfX19SwWq6ysrLKystN4vQ6DycvLO3ToUFRUFIlE4v2/reM3zB0ZGhpOnz49KioqJSWlubm5tLTU09MTIfTzzz8LmQrgAvb/Hu3/AAAAAOg7W7Zsefr0Kd4pEEJITU3tr7/+evv27blz57pqY2NjI8iicnNzZ8+ePXPmzJqamvj4+PPnz69btw6btXDhwrKyspSUFHt7e9HklgASuxH5bAhhSOVG7KcGYL+GP3l5+aSkpOTk5Pr6eiaT+fLly1WrVsnLy/v6+iLo1+NBYg+PCKGCggI7OztfX186nd6jRfX5Oa6rnzwAgAtnZ2dnZ2e8UwiqqKjIyspKVVVVVlZWW1vb39+/ra2Nw+H4+fmpqampqKi4uLhEREQghIYNG7Z582YajYYQMjAwSE1NPXjwoLKyMkJoyJAhly9fjomJGTJkCEJIVVX16tWrHA7H09OTRCLp6OgQiUQlJaVFixYVFhZi6z1y5AjWWF5efvHixRwOp7W11c/PT19fn0gkamhoODk55ebmdhVPGNnZ2Z0eSUJDQ7EG6enp1tbWWlpa2HRNTU0rK6vHjx9jcz9//uzj4zN8+HAymaygoGBtbf3nn38KGUnMYmNjRXLkFNVycAT7f8f9v1+Tgn0SAACAhOvTc01BQQFC6NSpU320fEF4enrq6OjwTpkzZ059fX27Ng8ePBBkaUuXLjU0NGSz2djD0NBQAoHw5s0b3jbe3t6DBg0SZGmi6mf1aX9NMjeiIBuiK/v27bt48SL/NoJvRIRQbGysIC1x0b/68rwGYL+G013P3cHBwdDQUEFBgUwmDxs2zNXVNTs7m/tcKejX8+oXfXzJPDxmZWUtXrz40qVL5ubmY8eO7dHS+vocBz1bIFn67zlS5Dw9PdXU1PBOAdrrF+dCKQD7v/jBPgkAAKCvDcBaQDslJSXW1taCLIrJZCooKLi7u3On5OTkIIQOHvy/7N13QFPX3z/wE0JCwh4ioIAyNBT3agXRYv06a1EEEce3ohatVFmiYBWlDiQ4wIW2Vq1WRVQUtVZBS61aBLWOAgpSLUORoQjIJuP3R55fHh5mgJAbkvfrL3LvuTfvcE7G/eTm3LCGzVDvlrpGnShhR7QE9W7AcY386xbH+HL48tjQJ5980q56twze4zCfCYD84vP5VEcAoAzGPwAAgBL66KOPaDSaiorKiBEjRD+OXrNmjY6ODovF+umnnwght2/ftrW1FS0ZNGhQfHx80514e3szmUxjY2PRzW+++UZDQ4NGo719+1a0hM/nb9iwwdzcnM1mDx48WFSkaKSioqLp1NsiHZ5MICwszMfHR5KWL1++rKioEM2oIGJlZUUI+fvvvzt217KkSJ3YrTsC5ASOa0BMkV4eO0MGL62odwMAAAAAAIBcSEtL69u3r5mZ2b1790STBoSHhy9ZsiQsLEw0421hYeGcOXOys7Pz8/M1NTXnz5/fdCd79uxxc3MT39y/f/93333XsEFQUFB4eHhERMSbN2+++OKLefPmPXjwoNFONDU1WzqPLDk5uQMP7fXr1zdv3nRxcZGkcUFBASFES0tLvITFYrHZ7MLCwg7ctYwpUid2viPWrl2rp6fHZDItLCxmzpx5//59CTcEAMWjSC+PnSGD9zjUuwHk0bfffnv06NGysjILC4tz585RHQdApjD+AQAAlBadTvfx8cnNzT1//rxoSVVVVWxs7OLFi0U3XV1dN27cqKenp6+v7+Tk9O7du+Li4nbdRU1NTVRUlLOzs4uLi66u7vr16xkMxtGjR6X8SJoICwtbuXKliopEx+C1tbWEEDqd3nAhg8Gorq7uknBSpUid2MmOWLhw4aVLl/Ly8ioqKqKjo3Nzcz/99NP09HSp5wT5hOMaaESRXh47Qwbvcah3A8ij0NDQ2tpaoVD477//urq6Uh0HQKYw/gEAAJTZV199paOjExkZKbp54sSJmTNnamtrN23JYDBI++cKyMzMrKqqGjhwoOgmm802NjbOyMjoXOo25OfnX7p0SXT6niRYLBYhhMfjNVxYV1fHZrOlnq0rKEwndrIjzMzMhg0bpqmpyWQyR48effTo0erq6v3790s9J8gnHNdAUwrz8tgZMniPQ70bAAAAAAAA5IWmpubSpUuTkpLu3btHCDlw4IC3t7d47ZUrVxwdHQ0NDdXU1NasWdOB/VdWVhJC1q9fL56rNCcnRzSPakPSnduUy+V6enqKjvAlIZqYtby8XLykqqqqpqbGxMSkvXdNCYXpROl2xKBBg+h0+vPnzzuwLQAoBoV5eewMGbzHod4NAAAAAAAAcsTb25vBYERERNy6dcvMzEx0GStCSG5urrOzs7GxcUpKSllZGZfL7cDODQ0NCSERERENpyu9e/duo2ZSnNu0oKDg1KlTXl5ekm9iYWGhpaWVk5MjXvLPP/8QQgYPHtyuu6aQYnSidDtCIBAIBAI1NbUObAsACkMxXh47Qwbvcah3A3ShHTt29OzZk0ajHTx4kMIYX3/9tfibugULFhBCNm3aZGtrq62traamZm1tvWbNmoqKikZbCQSCiIgIe3v7RssdHR2bfgGoqakpeZ5m9xwXFyfeW48ePTr0QEHuyO1TQKSlQS4JDGMAAICuY2pq6ubmdu7cueDgYF9fX/Hy1NTU+vp6Ly8vS0tLFotFo9Fa2oOqqmp9fX2zq8zMzFgs1uPHj6WfuwVcLnfBggX6+vqSb6Kqqjpt2rRbt24JBALRkqtXr9JoNCcnp67JKH2K0Ymd7IjJkyc3vHn//n2hUGhnZyf9oNCVlO2ghhDC5XJtbGzYbLaGhoaNjU1wcHDDU3FbqQnggEgSivHy2BkyeI9DvRugCwUEBCQlJVGdghBC9PX1r169mpmZefjwYUJIYmLiihUrsrOz3759GxoaGhkZOXv27Ibts7Kyxo0b5+/v3/RnL81ycHCQMElLe54xY8arV69u3bo1bdo0CXcF8k9unwKk/YO8IQxjAACArrZq1Soej/f+/fvPPvtMvNDc3JwQcuPGjZqamqysrJSUlJY2t7a2LikpiYuLq6+vLy4ubngSGYvFWrRoUXR0dFRUVHl5OZ/Pf/Xq1Zs3b7rogRQWFh45csTPz6/pKnd3dyMjo4cPHza7YXBwcGFh4caNGysrK+/evbt9+3YPDw8Oh9NFObuCYnRi6x3Reie+fv369OnTpaWl9fX1d+/e/eqrr8zNzZcvX94VOaHrKNtBDSHk9u3bnp6eubm5hYWFmzdv5nK5bc4/LqoJ4IBIQorx8tg6at/jUO8GUApsNnvKlCn9+/cX/XpOU1Nz2bJl+vr6Wlpabm5uzs7O165dy8vLEzV+8uRJUFDQ8uXLhw4d2nRXLBarvLy84a9dli1bJuHEUq3smUaj9e7de+zYsf369evcYwVoRqOnQOuDvHUYxgAAADIwbNiw8ePH+/j4NFw4aNCgwMDA/fv3m5iYrFu3ztHRkRDi4ODg5+cnKrUEBAS4uLgQQry8vMaPHz937lwOh7N582bRJbDs7OxEn3gjIyP9/Py4XK6BgYGJiYmvr+/79++76IGEh4c7OTmJqhiN1NXVFRUVXbx4sdkNBwwYEB8fn5CQYGBg4OLisnjx4gMHDnRRyC6iGJ3Yeke03olTpkxZv369qampurq6m5vbmDFjkpOTDQwMuiInKAPZHNQQQphM5jfffGNoaKipqTl79uyZM2dev35dXDNtpSaAAyIJKcbLY3JysoODQ69evVJSUp48eWJiYjJmzJhbt26J1lL7HqcqxX0BQHfxyy+/NLwp+p2R+EvdIUOGxMbGEkL27t1bU1PTaNtr1641vJmXl5eWlibhD7ta3zOAzHRmKGIYAwAAyEZiYmLThWFhYWFhYeKb33zzjeiPiIiIhs309fUbbR4eHi7+m8lkNtpP19m5c2dLq86dO+fo6NinT5+WGowbN66V8/u6BcXoxFY6ovVO3LFjx44dO7oyGii1rjuoOX/+fMObvXv3JoSIJ0HtTE0AxBTg5XH06NF37txpaS2173E4vxtAIh999BGNRlNRURkxYoSoLrxmzRodHR0Wi/XTTz8RQm7fvm1raytaMmjQoPj4+KY78fb2ZjKZogvREkK++eYbDQ0NGo329u1b0RI+n79hwwZzc3M2mz148OCYmJimO+mKq+i+fv2azWZbWFh0YNuwsLBG30mCQlLspwAAAACAVFRXV8fHx2dlZdXV1bXeks/nx8XFffjwwd3dvb33IhQK8/Pz79y5I7rAF0gXOlGB4aCmY7KysnR1dVsqXKImoDy60csjzu8GkEhaWpq1tbVAILh37x6dTieEhIeH83g8c3NzDw8PQkhhYeGcOXO8vb2FQuHnn38+f/588bud2J49e0pKSm7cuCG6uX//fgsLi9WrV4sbBAUF7d279+TJkxMmTNi+ffu8efOsrKxGjhzZcCeiq+hK8aFVVVUlJiZ6enoymcz2bvv69eubN2/u3btXinlAPinwUwAAAABAWkpKSqZMmUIImT9//okTJ1ppefPmzdjY2KtXr6qrq7f3Xi5evOjs7Cz6G5NjSB06UYHhoKZd6uvri4qKLly4cOPGjR9//LHZigFqAkqlG7084vxuAInQ6XQfH5/c3Fzx73qqqqpiY2MXL14suunq6rpx40Y9PT19fX0nJ6d3794VFxe36y5qamqioqKcnZ1dXFx0dXXXr1/PYDCOHj0q5UfSRGhoqImJyZYtWzqwbVhY2MqVK1VU8Eqi+BT4KQAAAAAgFQcPHhTPZtt6IYAQMmHChJMnT4pPEW2XmTNniu+oaTEOOgOdqNhwUNMuZmZmpqamISEh4eHhc+bMabYNagLKo3u9PGJEAkjqq6++0tHRiYyMFN08ceLEzJkztbW1m7ZkMBiEED6f3679Z2ZmVlVVDRw4UHSTzWYbGxtnZGR0LnUbzp8/f+bMmfj4eC0trfZum5+ff+nSJdHX4KAMFPIpAAAAAAAAygMHNZLLy8srKio6derUsWPHhg0bVlRU1KgBagIgt1DvBpCUpqbm0qVLk5KS7t27Rwg5cOCAt7e3eO2VK1ccHR0NDQ3V1NREFyZur8rKSkLI+vXrxfN25eTkiK8hKSbFeb5Onz4dFhZ28+bNvn37diAwl8v19PRksVgd2Ba6I8V7CgAAAAAAgFLBQY3kGAyGoaHhpEmTTp8+nZ6eHhoa2qgBagIgt1DvBmgHb29vBoMRERFx69YtMzMzKysr0fLc3FxnZ2djY+OUlJSysjIul9uBnRsaGhJCIiIihA3cvXu3UTPRPF/NSk5Olvzu9u7de+LEicTExF69enUgbUFBwalTp7y8vDqwLXRfivQUAAAAAAAAJYSDmvaytram0+np6ekNF6ImAPIM16sEaAdTU1M3N7eYmJj8/PyNGzeKl6emptbX13t5eVlaWhJCaDRaS3tQVVWtr69vdpWZmRmLxXr8+LHUYzciFAqDgoLev38fFxenqtrBFwEul7tgwQJ9fX3pZgM5pxhPAQAAAAAAUFo4qGndu3fvVq5ceerUKfGSrKwsPp9vZmbWsBlqAiDPcH43QPusWrWKx+O9f//+s88+Ey80NzcnhNy4caOmpiYrKyslJaWlza2trUtKSuLi4urr64uLi3NycsSrWCzWokWLoqOjo6KiysvL+Xz+q1ev3rx5I/WH8PTp0/Dw8EOHDjEYjIY/m9qxY4eEeygsLDxy5Iifn1/TVe7u7kZGRg8fPpRqZJAjCvAUaBOGMQAAAACAAsNBTSs0NDQSEhISExPLy8vr6+sfPXq0cOFCDQ0Nf39/cZtWagIA8gD1boD2GTZs2Pjx4318fBouHDRoUGBg4P79+01MTNatW+fo6EgIcXBw8PPzc3BwIIQEBAS4uLgQQry8vMaPHz937lwOh7N582Y2m00IsbOzy8vLI4RERkb6+flxuVwDAwMTExNfX9/3799L/SEIhcLPE64XAAAgAElEQVTWGyQnJzs4OPTq1SslJeXJkycmJiZjxoy5deuWuEF4eLiTk5Po00AjdXV1RUVFFy9e7NieQf4pwFOAtDUUMYwBAAAAABQYDmpa2ZbFYo0ZM+arr77q3bu3lpbW7Nmz+/btm5ycLL4IJ2m1JgAgF1qaMAiAEq6urq6urlSnUDTLli3r3bu3bO6Lz+ePHTv28OHDHd6Dj4+PgYGBFCNJV0xMjFReOaW1H5BEe58CCj+Mm4UxCQAAXU30XgNUkcpxlqurK9WPQ6nFxMR0vhO7CI7lu5rsD2o6oxsdEEn3GB+o0uj1B/N3AyiF6urq+Ph4S0vLPn36MJnMLroXPp9/8eLFDx8+uLu7t3dboVD45s2bly9f/vPPP12RDZSc5E8BDGMAAIAuYm9vj3IAhRrNvdsx/v7+s2fP7vx+oGPs7e2pjgBUks1BTWco8wER3uOo1eg9DvVuAKVQUlIyZcoUQsj8+fNPnDjRRfdy8+bN2NjYq1evqqurt3fbixcvOjs7i/42MDCQdjRQdpI/BTCMAQAAuojoGnFUp4BOsbOzs7OzozoFgJKSzUFNZyjzARHe4+QKTdjWTL4AsiQ6WeDs2bNUBwFo3pkzZ+bMmdP5V05p7QdAWjAmAQAAAKDDcCwP3RSOgxQSrlcJAAAAAAAAAAAAAIoA9W4AAAAAAAAAAAAAUASodwMAAAAAAAAAAACAIkC9GwAAAAAAAAAAAAAUAerdAAAAAAAAAAAAAKAIUO8GAAAAAAAAAAAAAEWgSnUAgMbOnTtHo9GoTgEgCxjqAAAAAACgGHAsDwBygiYUCqnOAPC/7t69m5eXR3WK/yUQCM6fPx8bGzty5Eh/f3+FefNeunTpjBkzPv/8c6qDdFdubm6d3MOrV6+SkpKkEkZ5/PLLL5cvX/7++++pDvJ/XLhw4cyZM35+fh9//DHVWaSg82MbAAAAAJSQvB3LUyg7O3vt2rXffPONg4MD1Vmad+nSpXPnzkVERBgYGFCdRV7gOEjBoN4N0KJXr17Nnz8/JSWFy+V6e3srTLGbEOLo6GhjY3Pw4EGqgwC0w7Jly54/f/77779THaQxf3///fv3X7x4ccqUKVRnAQAAAAAAygiFQkdHRx6Pd+fOHbmtIdTV1Q0cOHDUqFEnT56kOgtAl8D83QDNu3Tp0tChQwsLC5OTk318fOT2japjOBxORkYG1SkA2icjI4PD4VCdohk7d+6cP3++i4vLnTt3qM4CAAAAAACUOXny5J07dyIjI+W5hsBkMsPDw6Ojo2/fvk11FoAugXo3QGO1tbU+Pj4zZ86cNm3agwcPhg4dSnUi6eNwOJmZmVSnAGifzMxM+ax302i0Q4cOTZkyZfr06Y8fP6Y6DgAAAAAAUKCioiIoKMjT03PUqFFUZ2nDzJkzJ06c6OPjIxAIqM4CIH2odwP8H5mZmXZ2dkePHj1+/Pjx48c1NTWpTtQlOBxOQUFBaWkp1UEAJFVaWlpYWCif9W5CCJ1OP3HixLBhwyZPnowvkwAAAAAAlNDWrVurqqo2b95MdRCJREREpKamHjt2jOogANKHejfA/zp+/PjIkSNVVFQePny4YMECquN0IRsbG0LI8+fPqQ4CIClREVk0dOUTm82+ePGiubn5pEmTcK0eAAAAAACl8uLFi4iIiO+++87Q0JDqLBKxtbVdtmzZ2rVry8rKqM4CIGWodwMQQsiHDx/++9//enh4LF68OCkpydramupEXatv374sFgtnoUI3kpmZqaam1qdPH6qDtEZbW/vq1auamppTp0599+4d1XEAAAAAAEBGfH19raysvv76a6qDtMPmzZt5PN62bduoDgIgZah3A5CHDx+OGDHi2rVrly9f3r17N5PJpDpRl6PT6VZWVqh3QzeSmZnZr18/Op1OdZA29OjRIyEhobKyctq0aR8+fKA6DgAAAAAAdLmEhIRffvll3759DAaD6iztoKent2HDhoiIiKysLKqzAEgT6t2g1IRC4e7du+3t7U1NTZ88efL5559TnUh2OBxORkYG1SkAJJWRkSG3k3c30rt37+vXr+fm5s6YMaOmpobqOAAAAAAA0IXq6uq8vb3d3NzGjx9PdZZ28/Ly4nA4a9asoToIgDSh3g3K6+3bt05OTgEBAUFBQdevX+/VqxfViWTKxsYG53dDN5KZmSnPk3c3Ym1tnZCQ8OTJkzlz5vB4PKrjAAAAAABAV4mIiMjNzeVyuVQH6QhVVdWIiIi4uLj4+HiqswBIDerdoKRu3rw5dOjQJ0+e/P777yEhIfI/SYLUcTicrKwsPp9PdRCAtvH5/BcvXnSX87tFBg0adOXKlcTExEWLFgkEAqrjAAAAAACA9BUWFoaGhgYFBfXt25fqLB00YcKEL774wt/fH2fqgMJAvRuUDo/HCwkJ+c9//jNq1KjHjx87ODhQnYgaHA6ntrY2JyeH6iAAbcvOzq6pqele9W5CyOjRoy9cuHD27Flvb2+qswAAAAAAgPStXr1aV1c3ICCA6iCdEhER8eLFi++//57qIADSgXo3KJe8vLzPPvuMy+Xu3LnzwoUL+vr6VCeijKh0iCm8oVsQDdT+/ftTHaTd/vOf/5w+ffr777//7rvvqM4CAAAAAADSdPfu3RMnTkRERKirq1OdpVOsrKy8vb2Dg4PfvXtHdRYAKUC9G5RIXFzc0KFDi4uLk5OTfXx8qI5DMV1dXSMjI0zhDd1CZmamsbGxrq4u1UE6YubMmT/++ON33323a9cuqrMAAAAAAIB0CAQCX1/fzz77bNasWVRnkYLg4GAWi4XTdEAxoN4NSqGmpsbHx2fWrFmff/75gwcPhgwZQnUiucDhcFDvhm4hMzOz201m0tDChQsjIyMDAgIOHz5MdRYAAAAAAJCCw4cPP3z4MCIiguog0qGlpbVp06YDBw6kpqZSnQWgs1SpDgDQ5TIyMubOnfvy5csTJ07MmzeP6jhyxMbGBvVu6BYyMzNtbGyoTtEp3t7eBQUFy5Yt09HRcXV1pToOAAAAAAB03Pv379etW7dy5cpBgwZRnUVqFi9e/MMPP/j5+d24cYPqLACdgvO7QcEdP3585MiRqqqqDx8+RLG7EQ6Hg/m7oVvIyMjo1ud3i4SGhq5YsWL+/Pnx8fFUZwEAAAAAgI4LCQmh0WgbNmygOog0qaioREZGJiYmXrp0ieosAJ2CejcorPLy8gULFnh4eCxZsuTPP/+0srKiOpHc4XA4BQUFpaWlVAcBaE1paWlhYaEC1LsJIbt27XJxcZk1a1ZSUhLVWQAAAAAAoCOePn164MCB0NDQbnqFoVbY29u7ubn5+/vX1tZSnQWg41DvBsX04MGDESNGJCQkXLlyZffu3Uwmk+pE8kg0QcTz58+pDgLQGtGsO919PhMRFRWVY8eOjR8//vPPP3/y5AnVcQAAAAAAoN38/PwGDRq0aNEiqoN0iR07dhQUFOzZs4fqIAAdh3o3KBqhULh79+4xY8aYm5s/efJk6tSpVCeSX3379mWxWJjCG+RcZmammppanz59qA4iHQwG4+zZs0OGDJk2bdq///5LdRwAAAAAAGiHc+fOXb9+ff/+/SoqillSMzU1DQgI2Lx585s3b6jOAtBBivnkBKVVXFw8ffr0gICAtWvXXr9+3cTEhOpEco1Op1tZWaHeDXIuMzOzX79+dDqd6iBSw2azL126ZGJiMnHiRHyIBAAAAADoLqqrq1evXr1w4cLRo0dTnaULBQYG6uvrBwcHUx0EoINQ7wbFkZiYOHTo0LS0tJs3b4aEhCjqd63ShUtWgvxTjItVNqKtrX3t2jUmkzlp0qSSkhKq4wAAAAAAQNu4XO67d+9CQ0OpDtK12Gx2aGjo0aNH79+/T3UWgI5AQRAUAY/HCwkJmThx4ujRox8/fjxmzBiqE3UbHA4H53eDnMvMzFS8ejchpEePHgkJCRUVFVOnTq2oqKA6DgAAAAAAtCYvL2/79u0bNmxQhp+Sz507d8yYMb6+vkKhkOosAO2Gejd0e7m5uY6Ojlwud9euXbGxsXp6elQn6k44HE5WVhafz6c6CEDz+Hz+ixcvFLLeTQgxNTW9fv16bm7ujBkzampqqI4DAAAAAAAt8vPz69Wr18qVK6kOIgs0Gm337t3JyclnzpyhOgtAu6HeDd3bhQsXhg4dWlJSkpKS4uPjQ3Wc7sfGxqa2tjYnJ4fqIADNy87OrqmpsbGxoTpIV7G2to6Pj3/06JG7uzuPx6M6DgAAAAAANCMxMTE2NnbPnj1qampUZ5GRYcOGLVy4MCAgoLKykuosAO2Dejd0VzU1NT4+PrNmzZo+ffqDBw8GDx5MdaJuSXTaLKbwBrklGpz9+/enOkgXGjx48JUrV27cuLF48WKBQEB1HAAAAAAA+D94PJ6vr6+Tk9PUqVOpziJT27Zt+/Dhw44dO6gOAtA+qHdDt/Ts2bNPPvnkp59+io6OPn78uLq6OtWJuitdXV0jIyNM4Q1yKzMz09jYWFdXl+ogXcvOzu7ChQtnzpzB71QAAAAAAOTN/v37MzMzlbDsa2RktHbtWi6Xix+FQ/eCejd0P8ePHx81ahSLxRLNAEB1nG4Pl6wEeaaoF6tsauLEiadOnTpw4MCWLVuozgIAAAAAAP+jqKgoJCRk9erV/fr1ozoLBfz8/ExNTb/99luqgwC0A+rd0J2Ul5fPmzfPw8NjyZIld+7csbS0pDqRIrCxsUG9G+RWZmamAk/e3cisWbN+/PHHDRs2REREUJ0FAAAAAAAIIWTdunUaGhpBQUFUB6EGk8kMDw+Pjo6+ffs21VkAJKVKdQAASd2/f3/u3LkfPnz49ddfp0yZQnUcxcHhcC5dukR1CoDmZWRkzJgxg+oUsuPh4VFQULBq1SodHZ3FixdTHQcAAAAAQKk9fPjwyJEjJ06c0NTUpDoLZWbOnDlp0iQfH58HDx6oqODEWegGMEyhGxAKhbt373ZwcLCwsHjy5AmK3dLF4XAKCgpKS0upDgLQWFlZWWFhoZLMZyIWFBS0Zs2apUuXxsbGUp0FAAAAAEB5CYVCX19fOzs7zKS6a9eu1NTUY8eOUR0EQCKod4O8KyoqmjZtWkBAwNq1a+Pj442NjalOpGhExcTnz59THQSgsYyMDPL/h6hS2bZt21dffTVv3ryEhASqswAAAAAAKKmff/75zz//jIyMpNFoVGehmK2t7bJly9auXVtWVkZ1FoC2od4Ncu23334bOnTos2fPbt26FRISgh/OdAULCwsWi4UpvEEOZWZmqqmp9e3bl+ogskaj0aKiombNmuXq6vrgwQOq4wAAAAAAKJ0PHz6sXbt22bJlI0eOpDqLXNi8eTOPx9u2bRvVQQDahuohyCkejxcSEjJp0qQxY8Y8fvzYzs6O6kQKi06nW1lZod4NcigzM7Nfv350Op3qIBRQUVE5fvz42LFjp06d+vTpU6rjAAAAAAAol82bN1dXV2/atInqIPJCT09v48aNERERWVlZVGcBaAPq3UCl9+/fN7s8Jyfn008/DQ8P37Vr19mzZ3V1dWUcTNlwOBzRxBEAciUjI0MJJzMRYzAY586ds7W1nTRp0r///kt1HAAAAAAAZfHPP//s2bNny5YtPXr0oDqLHFm+fDmHw1mzZg3VQQDagHo3UCYrK2vIkCGvXr1qtDw2NnbYsGFlZWX37t3z8fGhJJuy4XA4OL8b5FBmZqYy17sJIWw2+/Lly0ZGRhMnTnzz5g3VcQAAAAAAlIKPj4+1tfXSpUupDiJfVFVVIyIi4uLi4uPjqc4C0BrUu4EafD5//vz5eXl5c+bM4fP5ooXV1dU+Pj6urq7Tp0+/d+/ewIEDqQ2pPDgcTlZWlrgjAOQBn89/8eKFkte7CSHa2trXrl1jMBiTJ08uKSmhOg4AAAAAgIK7fPnyr7/+um/fPlVVVaqzyJ0JEyZ88cUX/v7+PB6P6iwALUK9G6gRFhb2119/EUJSUlI2b95MCHn69Oknn3xy7NixmJiY48ePq6urU51RidjY2NTW1ubk5FAdBOB/ZWdn19TU2NjYUB2EeoaGhtevXy8vL582bVpFRQXVcQAAAAAAFFZdXV1AQIC7u7ujoyPVWeRURETEixcvvv/+e6qDALQI9W6gwOPHj0NCQgQCASGEz+dv2rTJz89vxIgRurq6qampbm5uVAdUOqJTaDGlCcgV0YDs378/1UHkgqmp6fXr17Ozs2fOnFlbW0t1HAAAAAAAxbRz585Xr16FhYVRHUR+WVlZeXt7BwcHv3v3juosAM1DvRtkrba2dt68eQ2XqKioHDp0aMWKFb///ruZmRlVwZSZrq6ukZERLlkJciUjI8PY2BiXqxXr169ffHz8w4cP3d3d8eNBAAAAAACpe/36dWho6LffftunTx+qs8i14OBgFov13XffUR0EoHmod4OsBQcHP3/+vGGxhs/n19XVpaamqqhgQFIGl6wEeYOLVTY1ZMiQK1euXL9+/auvvhIKhVTHAQAAAABQKGvWrOnZs+eqVauoDiLvtLS0Nm3aFBUVlZqaSnUWgGagvAgy9eeff+7cubPpdRHr6+uvX7++d+9eSlIBIcTGxgb1bpArmZmZmLy7KTs7uwsXLpw+fdrX15fqLAAAAAAAiiMpKSk6OnrXrl0sFovqLN3A4sWLhw8f7ufnR3UQgGag3g2yU1FRMXfuXBqN1uxagUAQEBDw6NEjGacCEQ6H02g+k/r6eqrCgHJqNOQyMjJwfnezJk6cePTo0X379oWGhlKdBQAAAACgWyouLm54UyAQ+Pj4TJgwYcaMGVRF6l5UVFQiIyMTExMvXbrUcHl5eXlVVRVVqQBEVKkOAErE39//zZs3TU/uFmEymXV1dT4+Pn/88UdLNXHoOsbGxgUFBbt373716tXTp0/T0tJYLBbO+AZZGjhwYE1NzcCBA21tbc3MzAoLC42NjakOJafmzp1bWlrq5eWlrq7e6ETv6urq4uJic3NzqrIBAAAAAMi///znP8OHDw8LCzMyMiKE/PDDD48fP8YZeO1ib2/v5ubm7+8/efJkNTU1gUBw9OjRtWvXRkdHT5gwgep0oNRomAAUZCMhIWHKlCmNxhuNRqPT6Twez8zMzNnZ+YsvvnB0dFRVxdcwssDj8Xbt2pWRkZGampqZmfnhwwdCiKqqqqqqam1tLSFk5syZ58+fpzomKJFZs2bFxcUJhUIGgyEQCETfjWlqavbr12/o0KE2Njb+/v54fWgoNDR0/fr1R44c8fDwEC0pKyubNm2atbX1sWPHKI0GAAAAACC/ampqNDU1hUIhm83etGnTggULBgwYsHDhwh07dlAdrZt59eqVjY3Nxo0b7ezsvvnmm7S0NEJIeHg45kAHaqFwALJQWlr65Zdf0mj/8/0KnU4X/TFq1ChnZ+cZM2Zgll7ZU1VVffTo0enTpxsu5PF4okuJMpnMYcOGURQNlNSwYcOuXLlSV1fXcGKTioqKR48ePXr0yN3dHcXuRr799tuysrKlS5f26NFj+vTpxcXFEyZMSEtLu3//PpfLxdnxAAAAAADNevLkiej0msrKyjVr1mzdulUgEGzYsIHqXN2Pqanp0qVLN27cWFNTQ6fTBQIBnU7HafJAOczfDbKwdOnSwsJCFRUVQoi6uvqsWbN+/vnnt2/fJicnBwYGothNldDQ0JYKiHV1dah3g4wNHTq0rq6u2VV0On3Lli0yztMthIWFeXh4uLm5nT9/fuzYsRkZGaJvEw8cOEB1NAAAAAAAOfXw4UPxsTCfzy8rKystLZ0zZ87Lly+pDda9VFVVcbncAwcO8Hg8oVAoOnmOz+enpKRQHQ2UXRfOZ4IpmEH+ubq6nj17luoUVPLy8vrxxx+bvTRlbm6umZmZ7COB0srNze3Tp0/T5QwGw9PTc//+/bKP1C3w+fwvvvji5s2bPB5P/FzW1dV98+YNriwPAAAAANDU0qVLf/rpp0YHwqqqqjQabc2aNWvXrtXQ0KAqW7cgFApPnz7t7+//9u1bUZm7ITqd/uHDBzabTUk2ANLV9W5fX187O7su2j90C3w+/7fffrOxsZHDi6dFRESYmpoqeb27qKiob9++1dXVjZZra2uXlZVREgmUma6ubtOBx2Kx/v33X8zO0ZL09HRHR8fS0tKGHzTpdPoPP/ywePFiCoMBAAAAAMinwYMHp6amtrR2zpw5jWb+hEYSExOnTJkiOq272Qb37t0bNWqUjFMBiHXtXKh2dnZubm5dehcg/+bOnUt1hOYpeaVbpGfPnv7+/lwut9FXskOHDqUqEiizwYMH3759u+ESVVXVgIAAFLtbcu/evYkTJ1ZVVTV6CgsEgvDw8EWLFuG3VgAAAAAADdXV1WVkZDS7SkVFxc3N7aeffpJtou7ns88++/333z///POqqqqmvxen0+mPHz9GvRsohPm7AZTd6tWrG/1Wi8lkjhw5kqo8oMxGjhzJZDIbLmGz2f7+/lTlkXPXrl0bN25cZWVl058QCoXCzMzMxMRESoIBAAAAAMittLS0Zqf0JISsXr361KlTampqMo7UHY0ZM+bBgwe9e/dmMBiNVqmoqDx+/JiSVAAiqHcDKDsdHZ3169fT6XTxEj6fP2TIEAojgdIaMmRIo0k5Nm7cqKenR2EkeZaVlcVms1s6g1tVVXXnzp0yjgQAAAAAIOcePnzY8PiXEEKn0xkMxs8//xwWFobfR0rO2tr6wYMHI0eOFF/8U6S+vv7+/ftUpQIgqHcDACFkxYoVhoaG4vd1Pp+P+UyAEkOHDhUIBKK/aTSagYGBl5cXtZHk2cqVK/Py8rZs2aKhodHoIyYhhMfjXbt2raWfagIAAAAAKKdG9W4Gg6GpqXnjxo0FCxZQmKqbMjAw+P33311cXFRU/k+BMTU1VXxkByB7qHcDAGGxWFu2bBHfVFVVtbGxoTAPKK2PPvpIXLel0Wjbtm3DRb1bp6mpGRgYmJOTs2rVKjU1tUa/JVRVVd27dy9V2QAAAAAA5FBycnJdXZ3obwaDYWFh8ejRo3HjxlGbqvtSU1OLjo5ev359w4U1NTVZWVlURQJAvRsACCHEw8PD2tpa9JUsh8NpNIcygGwwmcz+/fsTQlRUVPr27fvll19Snah7MDAwCAsL++eff77++msGgyGuetfX1x8+fPjdu3fUxgMAAAAAkBN8Pv/p06eiv1VVVR0cHFJSUiwsLKhN1d3RaLTvvvvu8OHDdDpdVFXAFN5ALerr3U+ePHF3d7ewsFBTU+vRo8eQIUManmdKiUWLFrFYLBqNVlNTQ22SlrTyT4uNjbW0tKQ1p2/fvrIMeefOnTFjxqirq5uYmAQGBtbW1na4cStruVyujY0Nm83W0NCwsbEJDg4uLy/vqoek0Oh0+rZt24RCoYqKCi5WCRQaNWqUioqKUCgMDw9vOkcHtMLU1HTPnj3Pnz9fuHChioqK6L/H5/MPHz5MdTQAAAAAALnw7NkzUUmBRqMtWbIkISFBV1eX6lAKYvHixb/++iuLxVJVVUW9G6hFcb07NTXV3t7e2Nj4999/LysrS0pKmjJlys2bN6lNdfTo0YCAAGoztKL1f5qLi8vLly+trKx0dHSEQqFQKOTxeFVVVYWFherq6jILmZ6ePmnSpAkTJhQXF58/f/7IkSPLly/vWOPW196+fdvT0zM3N7ewsHDz5s1cLtfV1bVrH5vicnFxGTlypEAgwOTdQKEhQ4YIBIIhQ4bMmjWL6izdUt++fQ8dOvTw4cOJEycSQng8XmRkZMOrgAIAAAAAKK0HDx4QQlRUVLZt23bw4EGcYSNdkyZNSklJ6dmzJ4/HE/2rAagh7DKEkJiYmNbbfPnll7169Wq4pLa2dvr06V2XqiVVVVV2dnbim+vWrSOEVFdXyz5JmyT5pzWsd4vNmDFD8nupq6s7cuTIl19+2bGQc+bMsbCwEAgEopvbt2+n0WjPnj3rQOPW1zo7OzfsptmzZxNC8vPzJQnp6urq6ura/gfXDIqfxtAJUhkAlKP6v6jsOt+DMTExVD8IpSat9wIAAACAbgFniVGrzVqZJHAE0X3h6EMGKP4i6927d2VlZSUlJfr6+qIlTCbz8uXLsk9y+PDhoqKiRgtpNJrsk7Spw/+0uLg4SfZfW1t75MiRqKioyZMnh4WFdSAhj8e7cuWKq6ur+B84derU1atXX7x4selVEFtv3Oauzp8/33BvvXv3JoRUVFR0IHYn+fr62tnZyf5+pW7r1q2+vr4aGhpUB+lyd+/ejYyMpDqF1CjMCKysrNy9e/e3335LdRCJSHcUddFn1tTU1Hv37i1ZsqQrdq4AIiIiqI4AAAAAIGujR4/28/OjOgUFIiMjZ8yYQeGE3XPmzJHi3uS26l1bW7tnzx5PT09MF9MIjj5kg+J696hRo65cufLZZ5/t379/zJgxjdZGRkauW7euurp62LBhr169KikpYTKZHA5HQ0MjOzu7uLiYxWItXbo0PDxc1F4oFEZERBw6dOjly5fq6uqffvppWFiYuMDaylpfX98DBw7U1dXRaDQrK6t//vmHEKKionLlypVDhw4lJSWpq6tv3bp18eLFhJCoqKjVq1cLhcLo6OiDBw/++eef2tra27dvnzt3ruiO+Hz+d99999NPPxUXF/fr12/dunWil7M//vhj9erVaWlpqqqqffr0EW3Y7MLO/NM6o7Ky8vvvvz969Kirq+sff/whrqdXVFRoaWk1u8knn3ySnJzcaOHLly8rKirMzc3FS6ysrAghf//9d9M9tN64XbsihGRlZenq6vbp06ftRyttdnZ2bm5usr9fqRs5cqSlpSXVKWREkerdCjMCCSGffvppNxqEUhxFXdSDbm5uQqFQPr/BlQdnz56lOgIAAACArJmamirM4UO7TJ48WUdHh8IA0q13y3Mnzps3r6qqqqVqktLC0YdsUDx/95o1a0aOHPnkyRMHB4cBAwaEh52pdG0AACAASURBVIeXlJSI1/r6+ooqywcOHPj3338LCgrGjRv36NGjtWvXPnr0qKSkZOHChdu3b3/y5ImofUhIyNq1a9etW1dUVHTr1q28vLyxY8cWFha2uTYyMvKLL76wsrISCoWiYjchRCAQ6Orqnj59Ojs7e/jw4V5eXlVVVYQQLy8vPz+/6upqLS2tmJiYFy9eWFpaenp61tfXizYMCgoKDw+PiIh48+bNF198MW/evAcPHlRWVjo5Obm6upaUlGRlZfXv37+urq7ZhZ38pzXL19c3LS2tlQZlZWVbt261t7cXCoXJyckbN24UF7sJIZqami39QKBpsZsQUlBQQAhp+KLGYrHYbLa4LyRvLOGu6uvrX79+vW/fvhs3buzdu5fJZLb+D4FWdKM6IygqDEKpQ7EbAAAAAIAQQm2xW6nQ6XQUu4EqFNe72Wx2UlLS7t27bWxsnj59GhgY+NFHH/3xxx+Nmtna2qqrqxsYGIjOoTY3N+/Ro4e6uvqCBQsIIRkZGYSQ6urqXbt2zZo1a8GCBTo6OoMGDTp48ODbt29/+OGHNte2xN7eXkdHR09Pz93dvba29t9//220Vltb29DQ0N3dvbKyMjc3lxBSU1MTFRXl7Ozs4uKiq6u7fv16BoNx9OjR7Ozs8vLyAQMGsFgsIyOj2NjYHj16NLtQWv+0srIy2v+3e/fuVnZYX18/ZMiQhISEO3furFq1qvMTWYgudkyn0xsuZDAY1dXV7W0s4a7MzMxMTU1DQkLCw8Ol+2UpAAAAAAAAAAAAdBcU17sJIQwGw9vb+9mzZ8nJyTNnziwqKpo9e/b79++bbSw6b5fH44m3JYSITqxOT0+vqKgYOXKkuPGoUaOYTGZKSkqbayUJKb6jllKJ1mZmZlZVVQ0cOFC0is1mGxsbZ2RkWFpa9uzZc8GCBSEhIdnZ2aK1zS6UME+b/7SG16v08fFpfW9PnjyZOHHi2LFjIyIiKisrJU/SLBaLRRp0k0hdXR2bzW5vYwl3lZeXV1RUdOrUqWPHjg0bNqzpVOwAAAAAAAAAAACg8Kivd4t98sknFy5cWL58eXFx8e+//97ezUtLSwkhmpqaDRfq6up++PChzbVSJCoWr1+/XnxudU5OTlVVFZvNTkxMdHBw2Lp1q6Wlpbu7e3V1dbML23V3Ev7TIiMjxSX4Zuno6Kxfv/7PP/8UCASjR4/etGlTw+p5RUUFrQWjR49uujdjY2NCSHl5uXhJVVVVTU2NiYlJextLuCsGg2FoaDhp0qTTp0+np6eHhoa28mABAAAAAAAAAABAIVFc73ZxcWl06u5///tfQohopux2EV3ytVH9urS01NTUtM21UmRoaEgIiYiIaDjJ9d27dwkhAwYMuHz5cn5+fmBgYExMzI4dO1pa2Dop/tMa0dDQWLVq1f3793v06DFu3LjVq1e/efOGtH/+bgsLCy0trZycHPES0azogwcPbm/jdu2KEGJtbU2n09PT0zv0DwAAAAAAAAAAAIBujOJ6d21t7dOnTxsuyczMJC1XM1sxcOBATU3NBw8eiJekpKTU1dWNGDGizbVSZGZmxmKxHj9+3Gh5fn6+6JEaGhpu27Zt+PDhT58+bXZhm3fR4X/amzdvFi1a1Ob+WSyWl5fXw4cPP/roo6CgoDbbN6Wqqjpt2rRbt24JBALRkqtXr9JoNCcnp/Y2bn3tu3fv5s2b13BvWVlZfD7fzMysA7EVw6+//qqjo3P58mWpN5bEnTt3xowZo66ubmJiEhgYKJp+XSqNQX5gjCkAdCIAAAAAyAw+fCo8dDHIG+rnM3F2dj5z5kxpaWlZWdnFixeDgoJmzJjRgXo3i8VatWrV+fPnT5w4UV5enpqaunz5chMTk2XLlrW5lhCir6+fn5+fnZ394cOHlubpljDGokWLoqOjo6KiysvL+Xz+q1ev3rx5k5+f//XXX2dkZNTV1T169CgnJ2f06NHNLpTkXtr7TxMKhdXV1bGxsdra2hI+EAaDsXjx4mPHjknYvpHg4ODCwsKNGzdWVlbevXt3+/btHh4eHA5HtHbDhg06OjoJCQmSNG5lrYaGRkJCQmJiYnl5eX19/aNHjxYuXKihoeHv79+x2ApAKBR2UeM2paenT5o0acKECcXFxefPnz9y5Mjy5cul0hjkCsaYAkAnAgAAAIDM4MOnwkMXg9xpaZ6KziOExMTEtN4mISFhzpw5VlZWampqTCaTw+GEhITU1NSI1kZGRqqrqxNC+vbte/v27bCwMB0dHUKIkZHRyZMnT58+bWRkRAjR09OLjo4WCoUCgWD79u39+vVjMBh6enrOzs6ZmZni+2p97cOHD/v06cNmsx0cHPz9/UWXQ+zXr9+LFy9OnDihp6dHCDE1NU1LS9u/f78olWjtDz/8ICoi9+nT5/nz50KhsLa2NjAw0NzcXFVV1dDQ0MXFJT09PTs7297eXk9Pj06n9+rVa926dTwer9mFbf5jW/+nnT9/3srKqqXuXr9+fTu7seP++OOPjz/+WE1NzcTEZPXq1eKEQqEwODhYS0srPj5eksatr3VycrKwsNDU1FRTU7OysnJ3d09NTZUwoaurq6ura+ce5f+QZLQrvDlz5lhYWAgEAtHN7du302i0Z8+edb5xF4mJienS10BZUpIRKIdjTFqjSJFGY+vksBOl+F4AAAAA0C0oz+cfOfzwKa1jN+U5gmidHHZx65Tn2UctiuvdABRSjHq3QCA4c+bM999/T8m9i9XX12tqanp4eIiXpKWlEULCwsI62bjrKNLngy4dgRhjregu9W50YivwiRMAAACUTVd//sGHz1YoRr0bXdxhOPqQDernMwGAduHz+aGhoRwOh81m9+jRw8LCIjQ01M3NjRBy584dc3NzGo22b98+QkhUVJSGhoa6uvrFixenTp2qra1tamoaHR0t2k+jxo1UVFTQWtDsrDsvX76sqKgwNzcXLxH9zuDvv//uZGOQPYwxBYBOBAAAAACZwYdPhYcuhu4F9W75kpGR0dJzm0ajubu7Ux0QqMflcjds2LB9+/aSkpKEhISamhpdXV1dXV1CiIODQ1JSkrill5eXn59fdXW1lpZWTEzMixcvLC0tPT09RTPUN2rciKamZkvfkiUnJzdtX1BQQAjR0tISL2GxWGw2u7CwsJONQfYwxhQAOhEAAAAAZAYfPhUeuhi6F9S75YuNjU1Lz22hUHj69GmqAwL14uLiRowY4eTkxGazhw8fPmPGjFu3btXV1bWyib29vba2tqGhobu7e2VlZW5urtRTiS5qTKfTGy5kMBjV1dWdbAyyhzGmANCJAAAAACAz+PCp8NDF0L2g3g3QzYgu1ym+yefzGQxGo5fsljCZTEKI6GtV6WKxWIQQHo/XcGFdXZ3o0q+daQyyhzGmANCJAAAAACAz+PCp8NDF0L2g3g3QzUybNu2vv/66ePFidXX1gwcP4uLipk+fLuHbjOTaO22WsbExIaS8vFy8pKqqqqamxsTEpJONQfYwxhQAOhEAAAAAZAYfPhUeuhi6F1WqAwBA+4SEhPz1118eHh4VFRUmJiZubm5bt26V+r2Ips2SvL2FhYWWllZOTo54yT///EMIGTx4cCcbg+xhjCkAdCIAAAAAyAw+fCo8dDF0Lzi/u1N+/fVXHR2dy5cvS72xJO7cuTNmzBh1dXUTE5PAwEDRVETN2rRpk62trba2tpqamrW19Zo1ayoqKsRrHR0dm351pqmpKW5w6tSpUaNGaWlp9enTZ9GiRaI5/oEq6enpL168KC4urq+vz83NjYqK0tPTozoUUVVVnTZt2q1btwQCgWjJ1atXaTSak5NTJxuD7GGMKQB0IgAAAADIDD58Kjx0MXQvqHd3Sru+d2pX4zalp6dPmjRpwoQJxcXF58+fP3LkyPLly1tqnJiYuGLFiuzs7Ldv34aGhkZGRs6ePbv1/Ts4OIj+iImJmT9//uzZs1+9enXx4sVbt25NnTq10YRHIEsrVqwwNzdv+I2FnAgODi4sLNy4cWNlZeXdu3e3b9/u4eHB4XBEazds2KCjo5OQkCBJY6AWxpgCQCcCAAAAgMzgw6fCQxdDNyPsMoSQmJiYrtu/kpszZ46FhYVAIBDd3L59O41Ge/bsWbONP//8cx6PJ77p5uZGCMnNzRXdnDx5cnl5ecP2y5Yt++2330R/jx8/vlevXuI72rdvHyHkzp070n04lHB1dXV1dZXKrmQ52hMTEw0MDMRPYQaD8dFHH8XGxgqFwr1794pmpFJXV3dyctq/f7+6ujohpF+/fi9evPjhhx+0tbUJIX369Hn+/HmjxlLJ9scff3z88cdqamomJiarV68WXdFCJDg4WEtLKz4+XpLGshETE9Olr4GyJN0RiDEmOWmNIqmPRnSi5KT4XgAAAADQLUj98w8+fEpOWsduMj6eRRdLC44+ZAP17vYRCARnzpz5/vvvqY1RX1+vqanp4eEhXpKWlkYICQsLk2RzLy8vQkhGRkaza3Nzc8eMGSO+aW1tPWLECPHNixcvEkJOnjzZ0exypJvWu/fv3+/r6yu+WVtb6+fnp6amVlVVJZsACgP17pZgjElObuvd6ETJ4RMnAAAAKBupf/7Bh0/JddN6N7pYWnD0IRuYz6QNfD4/NDSUw+Gw2ewePXpYWFiEhoaKzo++c+eOubk5jUYTnfIcFRWloaGhrq5+8eLFqVOnamtrm5qaRkdHi/bTqHEj7b0E7cuXLysqKszNzcVLrKysCCF///23JA/q9evXbDbbwsKi2bVhYWE+Pj7im5aWlkVFReKbosm7LS0tJbkjkLqCggJvb+8lS5aIlzCZTHNz8/r6+vr6egqDgcLAGFMA6EQAAAAAkBl8+FR46GLodlDvbgOXy92wYcP27dtLSkoSEhJqamp0dXV1dXUJIQ4ODklJSeKWXl5efn5+1dXVWlpaMTExL168sLS09PT0FD35GzVuRHQJ2mYlJyc3bS8qOmtpaYmXsFgsNptdWFjY5iOqqqpKTEz09PRkMplN175+/frmzZsuLi7iJd9++21BQcHevXs/fPiQnp4eGRk5efLkZqvwIANsNpvBYBw+fLiwsLC+vj4/P//HH3/csGGDu7u76CdCAJ2EMaYA0IkAAAAAIDP48Knw0MXQ7aDe3Ya4uLgRI0Y4OTmx2ezhw4fPmDHj1q1bdXV1rWxib2+vra1taGjo7u5eWVmZm5sr9VS1tbWEEDqd3nAhg8Gorq5uc9vQ0FATE5MtW7Y0uzYsLGzlypUqKv87MD799NPAwEBvb29tbe2BAwd++PDhxx9/7Fx86DjRlRbS0tL69+/PZrNtbW2PHj0aFhZ27NgxqqOBgsAYUwDoRAAAAACQGXz4VHjoYuh2VKkOIO9qampYLJb4Jp/PZzAYjQrNLRGdQN0VP+4QReLxeA0X1tXVsdns1jc8f/78mTNnEhISGp4bLpafn3/p0qXt27c3XLhu3boff/zxt99+++STT4qKioKCguzs7JKSkszMzDr9OKAjxo4de/36dapTgCLDGFMA6EQAAAAAkBl8+FR46GLoXnB+dxumTZv2119/Xbx4sbq6+sGDB3FxcdOnT5ew3i259s7fLbqabXl5uXhJVVVVTU2NiYlJK/dy+vTpsLCwmzdv9u3bt9kGXC7X09OzYX3/zZs3XC536dKln332mYaGhoWFxaFDh/Lz8xvVxAEAAAAAAAAAAAAoh/O72xASEvLXX395eHhUVFSYmJi4ublt3bpV6vcimr9b8vYWFhZaWlo5OTniJf/88w8hZPDgwS1tsnfv3vj4+MTERE1NzWYbFBQUnDp1KjMzs+HCrKwsPp/fq1cv8RJtbW19ff309HTJ0wIAAAAAAAAAAADIAM7vbkN6evqLFy+Ki4vr6+tzc3OjoqL09PSoDkVUVVWnTZt269YtgUAgWnL16lUajebk5NS0sVAoDAwMTE1NjYuLa6nYTQjhcrkLFizQ19dvuNDU1JQQ8ubNG/GSDx8+lJSUYDKTbm3Hjh09e/ak0WgHDx6U/b1v2rTJ1tZWW1tbTU3N2tp6zZo1FRUVjdoIBIKIiAh7e/sObAvyj9oRKNLSGAMJoRMBAAAAQDbwyVMxyHM/crlcGxsbNputoaFhY2MTHBzccEIFR0fHppMxtFJeA3mAencbVqxYYW5uLoc1teDg4MLCwo0bN1ZWVt69e3f79u0eHh4cDke0dsOGDaLrCRBCnj59Gh4efujQIQaD0fDJuWPHDvHeCgsLjxw54ufn1+heLCwsxo8ff+jQoVu3blVXV+fl5S1btowQsmTJElk9UJC+gICApKQkqu49MTFxxYoV2dnZb9++DQ0NjYyMnD17dsMGWVlZ48aN8/f3r6qqau+20C1QOwJJq2MMJIROBAAAAADZwCdPxSDP/Xj79m1PT8/c3NzCwsLNmzdzuVxXV9fW9+bg4NBlSUEKUO9uQ2hoaFpamp6enqhGzGQybW1tz58/TwjZt2/fqFGjCCGBgYEzZsyIioqKiIgghAwePPjly5eHDh1atWoVIWTKlClZWVmNGnc+2IABA+Lj4xMSEgwMDFxcXBYvXnzgwIGGDcQTpEgyU0p4eLiTk5O5uXmj5TQa7ezZs+7u7kuWLNHT07O1tc3NzY2NjR07dmznHwIoJ01NzWXLlunr62tpabm5uTk7O1+7di0vL0+09smTJ0FBQcuXLx86dGh7twWQROtjDLoFdCIAAAAAyAY+eSqG1vuRyWR+8803hoaGmpqas2fPnjlz5vXr18VTHbBYrPLycmEDy5YtW7NmjWwfAbQP5u9uw7Nnz/773/+KCtmEkLq6uqCgoHnz5r1//37FihUrVqxo2NjLy0v8t6Wlpaenp/hmv379GjXuvHHjxqWkpDS7atOmTZs2bRL9PXDgwDZL3jt37mxplYGBQUREhPg/ANBJv/zyS8ObPXr0IISIv18dMmRIbGwsIWTv3r01NTXt2hZAEq2PMegW0IkAAAAAIBv45KkYWu9H0VmtYr179yaEiGd6uHbtWsO1eXl5aWlpFM7KApLA+d2tKSgo8Pb2bjh3B5PJNDc3r6+vr6+vpzAYKIw//vjj448/VldX19bWHjRokGiKqNu3b9va2uro6LBYrEGDBsXHxxNCIiMjNTQ0VFRURowYYWRkxGAwNDQ0hg8fPnbsWDMzMxaLpaurK/6Ccc+ePSwWq2fPnl9//bWJiQmLxbK3t2/p2xFCCJ/P37Bhg7m5OZvNHjx4cExMTCvxGqqoqGg6j5XI6NGjJfkPvH79ms1mW1hYdOC/15ltQQQjUAGgEwEAAABANvDJUzGgH1uXlZWlq6vbp0+fZteGhYX5+Ph0/l6gawm7DCEkJiam6/YvA6WlpSwWy9fXt6CgoK6u7vXr14cOHdLS0po3bx7V0UAKXF1dXV1dpbKrjo32iooKbW1tLpdbXV1dUFAwa9as4uJioVB49uzZkJCQkpKSd+/ejR492sDAQNR+48aNhJCUlJTKysq3b99OmTKFEHLlypXi4uLKykpvb29CyOPHj0WNly1bpqGh8fTp05qamvT09FGjRmlpaeXm5orWZmVlEUIOHDgguhkQEKCmpnbu3Ln3799/++23Kioq9+/fbymeFFVWVmppaXl7ezdd9cknnwwZMqRj20pI9G7a4c3lCkZgB7Q5xiQhrVHU4f2gEzvfiVJ8LwAAAADoFjr2+QefPDv/yVMovVoZjiA6ppV+rKure/Xq1d69e9XU1H7++edm27x69crW1pbP53c4AI4+ZAPnd7dGdMnHtLS0/v37s9lsW1vbo0ePhoWFHTt2jOpooAiys7PLy8sHDBjAYrGMjIxiY2NFE3S4urpu3LhRT09PX1/fycnp3bt3xcXF4q1sbW3V1dUNDAzmzp1LCDE3N+/Ro4e6uvqCBQsIIRkZGeKWqqqqH330kZqamq2tbVRU1IcPH44ePdo0Rk1NTVRUlLOzs4uLi66u7vr16xkMxtGjR1uKJ0WhoaEmJiZbtmyR8bYgghGoANCJAAAAACAb+OSpGNCPLTEzMzM1NQ0JCQkPD58zZ06zbcLCwlauXKmigmqqvEMPtWHs2LHXr18vKyvj8XilpaV//vmnl5eXqirmPQcpsLS07Nmz54IFC0JCQrKzs5ttw2AwCCF8Pr/pKiaTSQjh8XgNW7Y0087IkSPV1dUbvgmJZWZmVlVVDRw4UHSTzWYbGxtnZGRIEq8zzp8/f+bMmfj4eC0tLVluC2JKPgIVAzoRAAAAAGQDnzwVA/qxJXl5eUVFRadOnTp27NiwYcOKiooaNcjPz7906ZKHh4csU0HHoN4NQBk2m52YmOjg4LB161ZLS0t3d/fq6mpCyJUrVxwdHQ0NDdXU1KR4zV81NbWGX8+KVVZWEkLWr18vnvEqJyenqqqqpXgNdXjarNOnT4eFhd28ebNv377tfSCd2RYaUuYRqDDQiQAAAAAgG/jkqRjQjy1hMBiGhoaTJk06ffp0enp6aGhoowZcLtfT05PFYnX4LkBmUO8GoNKAAQMuX76cn58fGBgYExOzY8eO3NxcZ2dnY2PjlJSUsrIyLpcrlTuqr68vLS01NTVtusrQ0JAQEhER0XCqo7t37zYbr9G2mpqaLU2WlJyc3FKYvXv3njhxIjExsVevXu19IJ3ZFppSzhGoYNCJAAAAACAb+OSpGNCPrbO2tqbT6enp6Q0XFhQUnDp1ysvLq/P7BxlAvVt+7dixo2fPnjQa7eDBg5QEqK+vDw0Ntba2ZjKZurq6AwcOxG+CpCs/P//p06eEEENDw23btg0fPvzp06epqan19fVeXl6WlpYsFotGo0nlvm7evCkUCpv9qlN0VeXHjx9LEq+TMYRCYWBgYGpqalxcnKampsy2hWYp4QhUPOhEAAAAAJANfPJUDOjHRt69ezdv3ryGS7Kysvh8vpmZWcOFXC53wYIF+vr6XRoGpAX1bvkVEBCQlJREYYA5c+YcP3785MmTVVVVz549s7KyqqiooDCP4snPz//6668zMjLq6uoePXqUk5MzevRoc3NzQsiNGzdqamqysrJSUlI6vH+BQPD+/Xsej/f333/7+vqam5s3O88Ui8VatGhRdHR0VFRUeXk5n89/9erVmzdvmo3X4TAiT58+DQ8PP3ToEIPBaPibo6Zf2Ep3W2iWEo5AxYNOBAAAAADZwCdPxYB+bERDQyMhISExMbG8vLy+vv7Ro0cLFy7U0NDw9/cXtyksLDxy5Iifn1+XJgFpauknAJ1HCImJiem6/SuDrKwsQsiBAwdkf9fR0dE0Gu3vv/+W/V3LjKurq6urq1R21bHRnp2dbW9vr6enR6fTe/XqtW7dOh6PJxQKAwMD9fX1dXV1Z8+evW/fPkKIlZXVqlWr1NXVCSF9+/a9fft2WFiYjo4OIcTIyOjkyZOnT582MjIihOjp6UVHRwuFwmXLljEYjN69e6uqqmpra8+cOfPFixei+925c6eosYaGxqxZ/6+9e4+LstoXP74GGJgBFFBRJmEKNUwU87J3WxFv9aptViSba+prR2aF1sFrYoVaWzNR98atgb12F+xkKaIerTxuzGOGbm7eL3gkygAJxQsmcZNhmN8fc36zJ0AEZoZnmPm8/3LWs561vuN3OT7z5WE9f9LpdHfu3ElISFCr1U5OTt7e3uHh4QUFBXcLzxTnzp1r9YNo3bp1+g45OTnjxo1TqVT6dh8fn+Dg4O+++64953ZUenq6RT8DuxIrsP3aWGOdYK5V1OlxSKLpSTTj/wUAAADdQueuf7jyNP3KU2e+WhnfIDqk7TyGhob6+/u7u7u7uLgMHDgwJibm3LlzxqcvXLhw5syZpoeh49tHV5HpdLr2FcY7TCaTpaenR0VFWWh8e/DDDz88+OCDmzdvjouL6+KpJ06cWFNTc/z48S6etytFRkYKITIyMkwfygpXe1xcXEZGxs2bN6UOxHrt2LEjOjracp+BXYkVKBVzrSLrXI12kkQz/l8AAADQLVjh9Y+dXHkK83134xtEN2WF//psEvuZtMt33333yCOPuLq69uzZMygoqKqqSghx5MiRwMBADw8PhUIRFBSUmZkphNiwYYObm5uDg8Po0aP79esnl8vd3NxGjRo1fvx4/eZEnp6ehgfdbty4UaFQ9O3bNy4uTqVSKRSK4ODgNn5tRKvVLl++XK1WK5XK4cOH63+ad7fwjHX02bUNDQ25ubkjRowwy98epKLVaqUOAXaNFWgDSCIAAAC6BleetoE8whpQ7763mpqa0NDQiIiIysrKoqKigICAhoYGIURFRUV0dHRxcXF5ebm7u/uMGTOEEPPnz3/99dd1Ot3mzZt/+umnq1evTpgw4dSpU2+88capU6cqKyuff/75devWnTlzRggRHx8fGxtbW1s7b9684uLikydPNjY2Pv7445cvX241kqVLl65duzY5OfnKlSvPPPPM9OnTjx8/frfwjHX02bXl5eUNDQ0nTpyYPHmyvhA/ZMiQlJQUa/vJIQAAAAAAAAAYUO++t+Li4qqqqqFDhyoUin79+u3atatPnz5CiIiIiBUrVnh5efXq1Ss0NPTmzZvXr183nBUYGOjq6tq7d+/nnntOCKFWq/v06ePq6jpz5kwhxMWLFw09nZychgwZ4uLiEhgYmJqa+uuvv6alpbUMo76+PjU1NSwsLDw83NPTMzExUS6Xp6Wl3S08U+ifS+nt7f3uu+8WFBRUVFRMmzbttdde++KLL0wcGV3jzTffTEtLu337tr+//86dO6UOB3aHFWgDSCIAAAC6BleetoE8wnpQ7763AQMG9O3bd+bMmW+//XZxcXGrfeRyubjLb204OzsLIRobG417ajSaVsf53e9+5+rqalwNNygsLKytrR02bJj+pVKp9PHxuXjxYnvC6ygXFxchxNChQ4ODg3v16uXh4fHOO+94eHj84x//MMv4sLTVq1ffuXNHp9P99NNPERER1ESPmgAAIABJREFUUocDu8MKtAEkEQAAAF2DK0/bQB5hPah335tSqTx06FBISMi77747YMCAmJiYuro6IcS+ffsmTZrk7e3t4uJi2JLbdC4uLsb3iRvU1NQIIRITEw1bb5eUlNTW1t4tPGMd3b9b/8jaGzduGFqcnZ3vv//+H3/80VxvEwAAAAAAAADMi3p3uwwdOvSrr74qLy9PSEhIT09fv359aWlpWFiYj49PXl7e7du3k5KSzDKRRqP55ZdffH19Wx7y9vYWQiQnJxvvvp2Tk9NqeM3O7ej+3e7u7g8++OCFCxeMGxsbGz08PMzyNgEAAAAAAADA7Kh331t5ebm+8uvt7f3ee++NGjXqwoUL586d02g0c+fOHTBggEKhkMlkZpnr8OHDOp2u1Xuu/fz8FArF6dOn2xOe6ZFER0efOnXq0qVL+pe1tbUlJSVBQUGmj4z2W79+fd++fWUy2QcffCBhGHFxcYZfCNBvQP+Xv/wlMDCwZ8+eLi4ugwYNWrJkiX7Pd2NNTU3JycnBwcHN2ttzbhuOHj06btw4V1dXlUqVkJBw584dffuePXsMQZq+hT30rHYFiruvhHZqdX3a5Cqy5iSKu39Q3BMfBQAAAFaIi08bYM1JtMTXQCFEUlLSQw89pFQq3dzcHnrooWXLllVVVRmOTpo0qeVmCe7u7sJGF4BtoN59b+Xl5XFxcRcvXmxoaDh16lRJScmYMWPUarUQ4uDBg/X19UVFRXl5eZ0ev6mp6datW42NjWfPnp0/f75arY6NjW3ZTaFQvPDCC9u2bUtNTa2qqtJqtWVlZVeuXGk1vE4HY7Bw4cL7778/Nja2tLT05s2bCQkJdXV1S5cuNX1ktN/ixYuzs7OljkIIIXr16rV///7CwsKPP/5YCHHo0KHXXnutuLj4xo0bq1ev3rBhQ2RkpHH/oqKiCRMmLFy4sLa2ttlQ9zy3DQUFBU888cRjjz12/fr13bt3f/LJJ3PmzNEfevbZZ8vKyrKysqZOnWry28X/sdoV2MZKaI+7rU+bXEVWm0TR5gdF2/goAAAAsE5cfNoAq02ihb4GCiGOHDny0ksvlZaWVlRUrFy5Mikp6Z77j4eEhAgbXQC2gXr3vXl7e2u12uDgYFdX16effjouLu61114LCgpKSEhISUlRqVRvvfXWpEmThBAhISGLFy9et26dECIoKOjo0aNJSUlxcXFCiClTpnzxxRfp6elTpkwRQsTHx2/fvl0/fn19fVBQkFKpHD9+fEBAwLfffqt/XOTf/vY3/b+fxYsXh4eHCyE2bNiwYMGCpKSk3r17q1Sq+fPn37p1q9XwTH/XXl5eR44c8fX1HTFiRP/+/fPz8/ft2zdixAjTR0Z3pFQqp0yZEhAQoF+c7u7ur7zySq9evXr06BEVFRUWFvbPf/7z8uXL+s5nzpxZunTpnDlzWl0wbZ/btpUrV/r4+Lzzzjtubm5jx45NSEjYsmWL/vmuMpmsf//+48ePf/DBB833vmEtmq3ANlbCPbWxPllFFtUsiW1/ULSNjwIAAAC0jYtPG9A1XwOFEM7Ozq+++qq3t7e7u3tkZOS0adO++eabK1eu6I8qFIqqqirjnYFfeeUV/WP8WABWy0nqALqB+++//1//+lfL9jVr1qxZs8bw8tVXX9X/wXj77JCQkISEBOOzoqOjm43To0ePsrKyluMvXLhw4cKFxi3Ozs7NJtVrNTzT+fr6fvHFF5YYGd3d119/bfxS/2s7hp+RPvzww7t27RJCbNq0qb6+vkPntqGxsXHfvn0RERGG7YOefPLJ119/fe/evQ899FAn3wm6IRNXQtvrE12m04ngowAAAAAdxcVnd2fRr4G7d+82ftm/f38hhGHn1X/+85/GRy9fvnz+/HlpN3vBPXF/t/S0Wq3UIcDihgwZIpPJHBwcRo8era/tLlmyxMPDQ6FQbNmyRQhx5MiRwMBAfUtQUFBmZmbLQeLj452dnX18fPQvX331VTc3N5lMduPGDX2LVqtdvny5Wq1WKpXDhw9PT09vOUh1dXXLnaf0Or0Tzs8//6xUKv39/S167qVLl6qrq/VbCekNHDhQCHH27NlOzGtvbGkF2u1KsKUkmsJuFwAAAEBX4uJTr1tffNpSErsyEUVFRZ6envfff3+rR9esWTNv3jyzTwrz4v5uoCucP39+0KBBTU1N+fn5jo6OQoi1a9c2NjYatmuvqKiIjo6Oj4/X6XRPPfXUjBkzDP95GGzcuLGysvLgwYP6lykpKf7+/q+//rqhw9KlSzdt2vT5558/9thj69atmz59+sCBA3/3u98ZD+Lu7q7T6cz41mpraw8dOvTSSy85Oztb9NyrV68KIXr06GFoUSgUSqWyoqKio/PaIVtagXa7Emwpiaaw2wUAAADQlbj41OvWF5+2lMQuSIRGo7l27dp//dd/HTx48KOPPmq1TPHzzz8fPnx406ZN5poUFsL93VJ6880309LSbt++7e/vv3PnTqnDgQU5OjrOmzevtLTU8GsytbW1u3btmjVrlv5lRETEihUrvLy8evXqFRoaevPmzevXr3doivr6+tTU1LCwsPDwcE9Pz8TERLlcnpaWZuZ30sLq1atVKtWqVassfa7+ycv6/6QN5HJ5XV1dJ6a2N7a0Au12JdhSEk1htwsAAACgK3HxqdetLz5tKYldkAg/Pz9fX9+333577dq1Lfci1luzZs1//Md/ODhQTbV2ZEhKq1evvnPnjk6n++mnn+757Fd0d7Nnz/bw8NiwYYP+5datW6dNm9azZ8+WPeVyuej4RjeFhYW1tbXDhg3Tv1QqlT4+Pu18dEOn7d69e8eOHZmZmcY/ZbXQuQqFQgjR2Nho3NjQ0KBUKjs6tX2ymRVozyvBZpJoCnteAAAAAF2Ji0/R/S8+bSaJXZCIy5cvX7t27Ysvvvj0009Hjhx57dq1Zh3Ky8u//PJL/a3xsHLUu4Eu4u7u/vLLL2dnZ+fn5wshNm/eHB8fbzi6b9++SZMmeXt7u7i46J/z21E1NTVCiMTERMM2WCUlJS2fA2nGvc+2b9++Zs2aw4cPP/DAAx2NthPn6vcLq6qqMrTU1tbW19erVKqOzm6fbGYF2vNKsJkkmsKeFwAAAEBX4uJTdP+LT5tJYhckQi6Xe3t7P/HEE9u3by8oKFi9enWzDklJSS+99JK+8g4rR70b6Drx8fFyuTw5OTkrK8vPz0//dAUhRGlpaVhYmI+PT15e3u3bt5OSkjoxuLe3txAiOTlZZyQnJ6dZN/22Wa3Kzc1t/3SbNm3aunXroUOH7rvvvo6G2rlz/f39e/ToUVJSYmj54YcfhBDDhw/vaAB2yzZWoJ2vBNtIoinsfAEAAAB0JS4+beDi0zaS2JWJGDRokKOjY0FBgXHj1atXv/jii7lz55p9OlgC9W7zWL9+fd++fWUy2QcffCBhGHFxcYYfkc2cOVMIsWrVqmY/OjP8mkk7NTU1JScnBwcHGzfu2bPHMGCfPn3M+R5smq+vb1RU1M6dO5ctWzZ//nxD+7lz5zQazdy5cwcMGKBQKGQy2d1GcHJy0mg0rR7y8/NTKBSnT582f9y/pdPpEhISzp07t2fPHnd39y4718nJaerUqVlZWU1NTfqW/fv3y2Sy0NDQDo1jz2xjBdr5SrCNJJrCzhcAAABAV+Li0wYuPm0jiZZLxM2bN6dPn27cUlRUpNVq/fz8jBuTkpJmzpzZq1cvE6dD16DebR6LFy/Ozs6WOgohhOjVq9f+/fsLCws//vhj00crKiqaMGHCwoULm/02yrPPPltWVpaVlTV16lTTZ7ErixYtamxsvHXr1qOPPmpoVKvVQoiDBw/W19cXFRXl5eXd7fRBgwZVVlbu2bNHo9Fcv37d+GebCoXihRde2LZtW2pqalVVlVarLSsru3LlitnfwoULF9auXfvhhx/K5XLjH6WsX7/e9HNjYmL69et38uTJVk9ftmxZRUXFihUrampqcnJy1q1bFxsbO3jwYHO+PVtnAytQ3GsltL2KbIBtJLFtfBQAAABYCS4+beDi0zaSaKGvgW5ubgcOHDh06FBVVZVGozl16tTzzz/v5ua2cOFCQ5+KiopPPvlkwYIFZnszsDDq3bZGqVROmTIlICDAxcVF3/LZZ58Z/6rI+fPn2znUmTNnli5dOmfOnBEjRjQ7JJPJ+vfvP378+AcffNCc0duBkSNHTp48ed68ecaNQUFBCQkJKSkpKpXqrbfemjRpkhAiJCRkwYIFISEhQojFixeHh4cLIebOnTt58uTnnntu8ODBK1eu1D+ZYezYsZcvXxZCbNiwYcGCBUlJSb1791apVPPnz79165bZ34JOp2u7Q25ubkhIyH333ZeXl3fmzBmVSjVu3LisrKz2nNvQ0HDt2rW9e/e2enTo0KGZmZkHDhzo3bt3eHj4rFmzNm/e3Ll3YbdsYAWKe62EtldRG+uzu7CNJLadCD4KAAAArAQXnzZw8WkbSbTQ10CFQjFu3LjZs2f379+/R48ekZGRDzzwQG5urvHuCGvXrg0NDdX/hADdw9020DGdECI9Pd1y41uboqIiIcTmzZsljOGVV17p37+/ccvKlSub1bs74Q9/+MPDDz/c6qF58+b17t3bxPGlEhERERERYZahbHu1t1xXlqPVasePH//xxx93eoT2r8n09HSLfgZ2JVagse64imxpNbbKmpNoxv8LAAAAugWbv/6x5otPc3134xtEM6Yn0RR8+7A23N/d3JAhQ2QymYODw+jRo/WbeCxZssTDw0OhUGzZskUIceTIkcDAQH1LUFBQZmZmy0Hi4+OdnZ31T48VQrz66qtubm4ymezGjRv6Fq1Wu3z5crVarVQqhw8frv+cakbaZxADxurq6jIzM4uKihoaGiw3i1ar3bNnz6+//hoTE9PRc3U6XXl5+dGjR/XPrICNaf8KZBVZLZIIAACALsPFpw3omiSaggVgtah3N3f+/PkHHnjAz88vPz/f1dVVCLF27doXX3xxzZo1sbGxQoiKioro6Oji4uLy8nJ3d/cZM2a0HGTjxo1RUVGGlykpKe+8845xh6VLl65duzY5OfnKlSvPPPPM9OnTjx8/3mwQcz2D+I033vDy8nJ2dvb39582bdqxY8fafy6gV1lZqd8nZ9asWZab5fDhw7t27dq/f7/+n16H7N27V7/Hzr59+ywRG6TV/hXIKrJaJBEAAABdhotPG9A1STQFC8BqUe9uztHRcd68eaWlpbt379a31NbW7tq1y/CvKyIiYsWKFV5eXr169QoNDb158+b169c7NEV9fX1qampYWFh4eLinp2diYqJcLk9LSzPzOxFCCPH8889/+eWXly9frq6u3rZtW2lp6cSJEwsKCiwxF2zVBx98YPhZy9atWy030WOPPfb5558bfjGiQ6ZNm2YI0vCLFLANHVqBrCLrRBIBAADQZbj4tAFdlkRTsACsFvXuVsyePdvDw2PDhg36l1u3bp02bVrPnj1b9pTL5UIIrVbbofELCwtra2sNO98rlUofH5+LFy+aFnXr/Pz8Ro4c6e7u7uzsPGbMmLS0tLq6upSUFEvMBQAAAAAAAAASot7dCnd395dffjk7Ozs/P18IsXnz5vj4eMPRffv2TZo0ydvb28XFZcmSJZ0Yv6amRgiRmJho2I+7pKREv1e4MUvs3x0UFOTo6Pj999937nQAAAAAAAAAsFrUu1sXHx8vl8uTk5OzsrL8/PwGDhyoby8tLQ0LC/Px8cnLy7t9+3ZSUlInBvf29hZCJCcnG2/JnZOT06ybufbvNtbU1NTU1OTi4tK50wEAAAAAAADAajlJHYCV8vX1jYqKSk9PLy8vX7FihaH93LlzGo1m7ty5AwYMEELIZLK7jeDk5KTRaFo95Ofnp1AoTp8+bfawW/rjH/+YmZlpeHns2DGdTjd27NgumBoAAAAAAAAAuhL3d9/VokWLGhsbb9269eijjxoa1Wq1EOLgwYP19fVFRUV5eXl3O33QoEGVlZV79uzRaDTXr18vKSkxHFIoFC+88MK2bdtSU1Orqqq0Wm1ZWdmVK1cs8S5+/vnn7du3//LLLxqNJicnZ/bs2Wq1es6cOfqjMTEx/fr1O3nypCWmBgAAAAAAAICuxP3ddzVy5MjJkyfPmDHDuDEoKCghISElJWXTpk2PP/74pEmTdu7cGRIS8swzz2zbtk0IsXjx4m+++WbXrl1z5849dOjQc889p1Kpnn76aaVSKYQYO3bsoUOH/Pz8NmzY0LNnz6SkpHnz5nl5eU2YMOEvf/mLSqUy+7uYMmVKYmLi7Nmz79y54+Pj8+STT77zzju9e/fWH21oaLh27drevXtHjRrV8tzc3NzFixdfunRJX4tXqVQDBgx47733JkyYYPY4bUDLHWlg5WwsZTb2droL8/6179ixw4yjoZ3Kysp8fX2ljgIAAKBLlZWVcfFpA0hit8O3j64h0+l0lhpaJktPT4+KirLQ+GgpLi7u66+/Lisra2f/pqamSZMmxcbGzpo1q3Mzzp8/f+vWrTdu3Ojc6dKKjIwUQmRkZJg+VBs728DKWe4zsCuxAqVl+irasWNHdHS0WYJBJ0RERJjl/wIAAIBuITIycufOnVJHYb/MUivjG0T3xbePLsD93bamrq4uMzNzwIAB999/v7Ozcxs9tVrt3r17f/3115iYmI7OotPprly5cunSpR9++MGEYG2HbdRM0X2xAru7qKgofjwMAACArkGtzQbwDQJoA/VuW1NZWTllyhQhxIwZM7Zu3dpGz8OHD+/atWv//v2urq4dnWXv3r1hYWH6Pxt2RwEAAAAAAAAACbGfCeyXGfczAQAAAAAAACA5B6kDAAAAAAAAAADADKh3AwAAAAAAAABsAfVuAAAAAAAAAIAtoN4NAAAAAAAAALAF1LsBAAAAAAAAALbAyaKj5+TkWHR8wBRlZWW+vr5SRwEAAAAAAADAPGQ6nc5SQ8tkFhoZMJeIiIiMjAypowAAAAAAAABgBhasd9u5TZs2LV68+Ny5cwEBAVLH0rqqqqqHHnooIiJi48aNUscCAAAAAAAAAKai3m0RlZWVAQEBL774YlJSktSxtOWTTz55+eWXT5w48fDDD0sdCwAAAAAAAACYhHq3RcydO3f37t2FhYUeHh5Sx9KWpqam4OBguVyelZXF/jMAAAAAAAAAujUHqQOwQQUFBR9++OGaNWusvNgthHBwcEhJScnOzk5PT5c6FgAAAAAAAAAwCfd3m98TTzxx8+bNY8eOOTh0jx8nvPjii5mZmRcvXnR3d5c6FgAAAAAAAADopO5RkO1Gdu7cefDgwb///e/dpdgthHjvvfdqampWr14tdSAAAAAAAAAA0Hnc321O9fX1Q4YMGT9+/H/+539KHUvHbNy48fXXXz979uzgwYOljgUAAAAAAAAAOoN6tzmtXLly7dq1hYWF9913n9SxdIxWqx01apSPj09mZqbUsQAAAAAAAABAZ3SbPTes388//5yUlPTmm292u2K3EMLR0TElJeWbb775+uuvpY4FAAAAAAAAADqD+7vNZvr06Xl5eQUFBQqFQupYOik6OvrEiRPnz5/vvm8BAAAAAAAAgN3i/m7zyM7O3r59+1//+tduXSn+61//WlFR8be//U3qQAAAAAAAAACgw7i/2wyamprGjBnTo0eP//mf/5E6FlOtXr363XffvXDhwv333y91LAAAAAAAAADQAdS7zeCjjz6Ki4s7depUUFCQ1LGYqqGhISgoaMSIEenp6VLHAgAAAAAAAAAdQL3bVL/++mtAQEBkZOTGjRuljsU8vv7662eeeebQoUOTJ0+WOhYAAAAAAAAAaC/q3aZatGjRli1bvv/++969e0sdi9k8/fTTP/300+nTp+VyudSxAAAAAAAAAEC78LxKk/zwww8pKSmrVq2ypWK3EOLvf//7pUuXUlNTpQ4EAAAAAAAAANqL+7tN8tRTT5WUlJw+fdrJyUnqWMzszTffTElJuXjxokqlkjoWAAAAAAAAALg36t2dt2/fvqeffvrAgQOPP/641LGYX21tbWBg4GOPPfbxxx9LHQsAAAAAAAAA3Bv17k5qaGgYPnx4UFBQRkaG1LFYyvbt22fMmJGdnf2HP/xB6lgAAAAAAAAA4B6od3fS+vXrly1bduHCBX9/f6ljsaBHH320qqoqPz/fwYGt3gEAAAAAAABYNYqYnXHt2rVVq1a9/vrrtl3sFkJs2rTp7NmzaWlpUgcCAAAAAAAAAPfA/d2d8eKLL2ZmZhYWFrq5uUkdi8XFx8enp6cXFhZ6enpKHQsAAAAAAAAA3BX3d3fYqVOntmzZsnbtWnsodgshVq1a5eDgsGLFCqkDAQAAAAAAAIC2cH93x+h0uokTJzY2Nv7rX/+SyWRSh9NFPvroo7i4uJMnTw4fPlzqWAAAAAAAAACgddS7O+bzzz//85//nJub+/vf/17qWLpOU1NTcHCwXC7Pysqynyo/AAAAAAAAgO6FencH1NbWDhky5Iknnvjwww+ljqWrnThx4pFHHvniiy+io6OljgUAAAAAAAAAWkG9uwMSExM3btxYWFioUqmkjkUCs2bN+uabb/73f//X3d1d6lgAAAAAAAAAoDmeV9lepaWlycnJK1assM9itxBizZo11dXV7733ntSBAAAAAAAAAEAruL+7vcLDw8+ePXv+/HkXFxepY5HMhg0bEhISzp07FxAQIHUsAAAAAAAAAPAb1Lvb5dtvv3300Uf37ds3depUqWORUmNj4+jRo9Vq9VdffSV1LAAAAAAAAADwG9S7702r1Y4aNapfv34HDhyQOhbpHTlyZOLEiV999dVTTz0ldSwAAAAAAAAA8G/Uu+8tJSVlwYIF586dGzx4sNSxWIWoqKiTJ0+eP39eoVBIHQsAAAAAAAAA/B+eV9lccnLyZ599ZvgxwK1bt95+++34+HiK3QZ/+9vfrl69mpycbGipqKiYNWvW7du3JYwKAAAAAAAAgJ2j3t3cl19++ec///mRRx7Jz88XQixfvtzBwWHZsmVSx2VFfH19ly5dumrVqpKSEo1Gk5ycPHDgwLS0tFOnTkkdGgAAAAAAAAD7xX4mzXl6et6+fdvJyUmr1U6bNu2rr77avHnz7NmzpY7LutTX1w8bNmzQoEE//fTTjz/+qNVq5XL5e++9t2jRIqlDAwAAAAAAAGCnnKQOwLqUlpbqN+VobGwUQnz99dcODg6XL1+ur69nr2pjN2/eDAgI2L9/v6Ojo1arFUI0NTUdP35c6rgAAAAAAAAA2C/2M/mNZjtyaDSahoaGd999d/DgwRkZGVJFZVUaGhr+/ve/Dxo06ODBg0IIfbFb/4ecnBxJQwMAAAAAAABg16h3/8bp06ednZ2bNWq12rKysqioqClTptTU1EgSmJU4fPjwoEGDFi1aVF9fr9Fomh0tLS2tqqqSJDAAAAAAAAAAoN79GydOnNDvZNKSXC6PjY11c3Pr4pCsytChQ/v37y+TyVo9qtPpeGQlAAAAAAAAAKlQ7/6N/Pz8pqamZo1OTk49evQ4ePBgTEyMJFFZD29v7++++2769OmtlrzlcvmJEye6PioAAAAAAAAAENS7jd28ebOioqJZo1wu79+/f35+/oQJEySJyto4Ozt/+umnycnJMpmsWdWbR1YCAAAAAAAAkBD17n9ruReHk5PTqFGjjh8/HhAQIElIVmvevHkZGRnOzs5OTk6GRh5ZCQAAAAAAAEBC1Lv/7dSpU3K53PDSwcFh+vTpWVlZffr0kTAqqxUeHp6bm9unTx/jv7SSkpLbt29LGBUAAAAAAAAAu0W9+99OnjxpvHn3smXLPv30U2dnZwlDsnIjRow4ceLEkCFDDHd563S6M2fOSBsVAAAAAAAAAPtEvfvf8vPztVqtg4ODXC7funXr22+/LXVE3cB9992XnZ395JNPOjg4CCGcnZ15ZCUAAAAAAAAASVDv/j+1tbXFxcUymaxnz56HDx+eMWOG1BF1G25ubnv27Fm0aJEQoqGhgUdWAgAAAAAAAJCETKfTGV6UlZVlZ2dLGI2EioqKEhMT+/Xr99Zbb/Xr10+SGPz8/MaOHWviIDk5OZcvXzZLPB317bfffvjhh3369Nm4caMkAViD4OBgX19fqaMAAAAAAAAA7NFv6t07duyIjo6WMBo7FxERkZGRYeIgkZGRO3fuNEs86IT09PSoqCipowAAAAAAAADskVPLJuMKuP349ttvx40bJ+HTKSMjI801lFnq5p1WVFQkl8sfeOABqQKQkEwmkzoEAAAAAAAAwH61Uu+2T5MnT5Y6BBvx4IMPSh0CAAAAAAAAAHvE8yoBAAAAAAAAALaAejcAAAAAAAAAwBZQ7wYAAAAAAAAA2ALq3QAAAAAAAAAAW0C9GwAAAAAAAABgC6Ssd//3f/+3h4fHV199ZfbO7XH06NFx48a5urqqVKqEhIQ7d+6Yq7NdIYkAAAAAAAAArISU9W6dTmehzvdUUFDwxBNPPPbYY9evX9+9e/cnn3wyZ84cs3S2NyQRAAAAAAAAgJWQGZcgd+zYER0dbd6ipHWKiYnJz8//8ccfZTKZEGL9+vVLliy5cOHCQw89ZGLnTouMjBRCZGRkWMk41s8KkyiTydLT06Oiosw4JgAAAAAAAIB26tL7u3U6XUZGxj/+8Y+unLSlxsbGffv2TZw4UV/6FEI8+eSTOp1u7969Jna2ByQRAAAAAAAAgHWybL1bq9WuXr168ODBSqWyT58+/v7+q1ev1t/9evToUbVaLZPJ3n//fSFEamqqm5ubq6vr3r17n3zyyZ49e/r6+m7btk0/TrPOzVRXV8vuYsyYMS37X7p0qbq6Wq1WG1oGDhwohDh79qyJnW0SSQQAAAAAAADQLVi23p2UlLR8+fJ169ZVVlYeOHCgvr7e09PT09NTCBESEpKdnW3oOXfu3AULFtTV1fXo0SNYKgh3AAASbklEQVQ9Pf3HH38cMGDASy+9pNFoWnZuxt3dXXcXubm5LftfvXpVCNGjRw9Di0KhUCqVFRUVJna2SSQRAAAAAAAAQLdg2Xr3nj17Ro8eHRoaqlQqR40a9eyzz2ZlZTU0NLRxSnBwcM+ePb29vWNiYmpqakpLS80e1Z07d4QQjo6Oxo1yubyurs7EzjaJJAIAAAAAAADoFixb766vrzd++qVWq5XL5c3Kjnfj7OwshNDfGmxeCoVCCNHY2Gjc2NDQoFQqTexsk0giAAAAAAAAgG7BsvXuqVOnnjhxYu/evXV1dcePH9+zZ8/TTz/dzlJp+3V062cfHx8hRFVVlaGltra2vr5epVKZ2NkmkUQAAAAAAAAA3YKTRUd/++23T5w4ERsbW11drVKpoqKi3n33XbPPot/6uf39/f39e/ToUVJSYmj54YcfhBDDhw83sbNNIokAAAAAAAAAugXL1rsLCgp+/PHH69evOzlZdqIOcXJymjp1alZWVlNTk4ODgxBi//79MpksNDTUxM42iSQCAAAAAAAA6BYsu5/Ja6+9plarq6urLTpLJyxbtqyiomLFihU1NTU5OTnr1q2LjY0dPHiw/ujy5cs9PDwOHDjQns42jyQCAAAAAAAA6BYsW+9evXr1+fPnvby89BsxOzs7BwYG7t69Wwjx/vvv//73vxdCJCQkPPvss6mpqcnJyUKI4cOHX7p06cMPP1y0aJEQYsqUKUVFRc06mx7Y0KFDMzMzDxw40Lt37/Dw8FmzZm3evNm4g/HeGvfsbNtIIgAAAAAAAIBuQWZcE9yxY0d0dHSHtlFuW2pqalFRkb4GKoRoaGhYunRpamrqrVu3lEqluWaxDZGRkUKIjIwMKxnHgCS2n0wmS09Pj4qKkjoQAAAAAAAAwB5ZcEfmq1evxsfHnz592tDi7OysVqs1Go1Go6FU2i2QRAAAAAAAAADdhQX3M1EqlXK5/OOPP66oqNBoNOXl5R999NHy5ctjYmJ69uxpuXlhRiQRAAAAAAAAQHdhwXq3/mmB58+fDwgIUCqVgYGBaWlpa9as+fTTTy03KcyLJAIAAAAAAADoLiy4n4kQYvz48d98841Fp4ClkUQAAAAAAAAA3YIF7+8GAAAAAAAAAKDLUO8GAAAAAAAAANiC7l3vXr9+fd++fWUy2QcffND1s69atUr2W8OGDev6MLo1aTOo19TUlJycHBwcLFUAAAAAAAAAAMyie9e7Fy9enJ2dLXUU6DzJM1hUVDRhwoSFCxfW1tZKGAYAAAAAAAAA03XverfkPvvsM52R8+fPSx0ROuDMmTNLly6dM2fOiBEjpI4FAAAAAAAAgKmod8N+Pfzww7t27ZoxY4aLi4vUsQAAAAAAAAAwldnq3d99990jjzzi6uras2fPoKCgqqoqIcSRI0cCAwM9PDwUCkVQUFBmZqYQYsOGDW5ubg4ODqNHj+7Xr59cLndzcxs1atT48eP9/PwUCoWnp+eSJUv0w27cuFGhUPTt2zcuLk6lUikUiuDg4Ly8vLuFodVqly9frlarlUrl8OHD09PT2wjPWHV1tewuxowZY66/JWtGBgEAAAAAAAB0a+apd9fU1ISGhkZERFRWVhYVFQUEBDQ0NAghKioqoqOji4uLy8vL3d3dZ8yYIYSYP3/+66+/rtPpNm/e/NNPP129enXChAmnTp164403Tp06VVlZ+fzzz69bt+7MmTNCiPj4+NjY2Nra2nnz5hUXF588ebKxsfHxxx+/fPlyq5EsXbp07dq1ycnJV65ceeaZZ6ZPn378+PG7hWfM3d1ddxe5ubl3e+NvvPGGl5eXs7Ozv7//tGnTjh07Zpa/z65ntxkEAAAAAAAAYDPMU+8uLi6uqqoaOnSoQqHo16/frl27+vTpI4SIiIhYsWKFl5dXr169QkNDb968ef36dcNZgYGBrq6uvXv3fu6554QQarW6T58+rq6uM2fOFEJcvHjR0NPJyWnIkCEuLi6BgYGpqam//vprWlpayzDq6+tTU1PDwsLCw8M9PT0TExPlcnlaWtrdwjPR888//+WXX16+fLm6unrbtm2lpaUTJ04sKCgwfeSuZ58ZBAAAAAAAAGBLzFPvHjBgQN++fWfOnPn2228XFxe32kculwshtFpty0POzs5CiMbGRuOeGo2m1XF+97vfubq6GtdSDQoLC2tra4cNG6Z/qVQqfXx8Ll682J7wOsHPz2/kyJHu7u7Ozs5jxoxJS0urq6tLSUkx1/hdyT4zCAAAAAAAAMCWmKferVQqDx06FBIS8u677w4YMCAmJqaurk4IsW/fvkmTJnl7e7u4uBg2dDadi4uL8V3GBjU1NUKIxMREw8bNJSUltbW1dwvPmOm7PwcFBTk6On7//femv8GuRwYBAAAAAAAAdHdme17l0KFDv/rqq/Ly8oSEhPT09PXr15eWloaFhfn4+OTl5d2+fTspKcksE2k0ml9++cXX17flIW9vbyFEcnKy8d7NOTk5rYbX7FzTd39uampqampycXEx+S1KgwwCAAAAAAAA6NbMU+8uLy+/cOGCEMLb2/u9994bNWrUhQsXzp07p9Fo5s6dO2DAAIVCIZPJzDLX4cOHdTpdq3fs+vn5KRSK06dPtyc80yP54x//aPzy2LFjOp1u7Nixpo/c9ewzgwAAAAAAAABsidnq3XFxcRcvXmxoaDh16lRJScmYMWPUarUQ4uDBg/X19UVFRXl5eZ0ev6mp6datW42NjWfPnp0/f75arY6NjW3ZTaFQvPDCC9u2bUtNTa2qqtJqtWVlZVeuXGk1vE4HY/Dzzz9v3779l19+0Wg0OTk5s2fPVqvVc+bMMX3krmefGQQAAAAAAABgU4y3fUhPT2/W0k7FxcXBwcFeXl6Ojo733XffW2+91djYqNPpEhISevXq5enpGRkZ+f777wshBg4cuGjRIldXVyHEAw88cOTIkTVr1nh4eAgh+vXr9/nnn2/fvr1fv35CCC8vr23btul0uldeeUUul/fv39/Jyalnz57Tpk378ccf9fP+9a9/1Xd2c3P705/+pNPp7ty5k5CQoFarnZycvL29w8PDCwoK7haeiRYtWjRw4EA3NzcnJydfX9+XXnqpvLy806NFRERERESYHlXnxrHPDObk5IwbN06lUun/Lfj4+AQHB3/33XedHlAIkZ6ebnpgAAAAAAAAADpBptPpDLXvHTt2REdHG7dYg7i4uIyMjJs3b0odiGVFRkYKITIyMqxkHDOykwwKIWQyWXp6elRUlNSBAAAAAAAAAPbIbM+rtCitVit1CDAJGQQAAAAAAABgad2j3g0AAAAAAAAAQNusvd795ptvpqWl3b5929/ff+fOnVKHgw4jgwAAAAAAAAC6hpPUAdzD6tWrV69eLXUU6DwyCAAAAAAAAKBrWPv93QAAAAAAAAAAtAf1bgAAAAAAAACALeiievf69ev79u0rk8k++OCDrpmxVXFxcbL/b+bMmUKIVatWyX5r2LBh7R/w6NGj48aNc3V1ValUCQkJd+7c0bfv2bPHMGCfPn0s8ma6nNUmUdw9Ee3U1NSUnJwcHBxs3GiTSQQAAAAAAABsWBfVuxcvXpydnd01c7WtV69e+/fvLyws/Pjjj00cqqCg4IknnnjssceuX7++e/fuTz75ZM6cOfpDzz77bFlZWVZW1tSpU00O2VpYbRLbSER7FBUVTZgwYeHChbW1tcbtNplEAAAAAAAAwIbZ3X4mSqVyypQpAQEBLi4u+pbPPvtMZ+T8+fPtHGrlypU+Pj7vvPOOm5vb2LFjExIStmzZcvHiRSGETCbr37//+PHjH3zwQUu9EzvWLIltJOKezpw5s3Tp0jlz5owYMaLZIZIIAAAAAAAAdC92V+82l8bGxn379k2cOFEmk+lbnnzySZ1Ot3fvXmkDszcmJuLhhx/etWvXjBkzDD//AAAAAAAAANBNdbjePWTIEJlM5uDgMHr0aP3+D0uWLPHw8FAoFFu2bBFCHDlyJDAwUN8SFBSUmZnZcpD4+HhnZ2cfHx/9y1dffdXNzU0mk924cUPfotVqly9frlarlUrl8OHD09PTWw5SXV0tu4sxY8Z09H111KVLl6qrq9VqtaFl4MCBQoizZ89aemrT2VISu3UiAAAAAAAAAJhRh+vd58+ff+CBB/z8/PLz811dXYUQa9euffHFF9esWRMbGyuEqKioiI6OLi4uLi8vd3d3nzFjRstBNm7cGBUVZXiZkpLyzjvvGHdYunTp2rVrk5OTr1y58swzz0yfPv348ePNBnF3d9fdRW5ubvvf0RtvvOHl5eXs7Ozv7z9t2rRjx46156yrV68KIXr06GFoUSgUSqWyoqKi/VNLxZaS2K0TAQAAAAAAAMCMOlzvdnR0nDdvXmlp6e7du/UttbW1u3btmjVrlv5lRETEihUrvLy8evXqFRoaevPmzevXr3doivr6+tTU1LCwsPDwcE9Pz8TERLlcnpaW1tFQ2+P555//8ssvL1++XF1dvW3bttLS0okTJxYUFNzzxDt37gghHB0djRvlcnldXZ0l4jQvW0pit04EAAAAAAAAADPqzP7ds2fP9vDw2LBhg/7l1q1bp02b1rNnz5Y95XK5EEKr1XZo/MLCwtra2mHDhulfKpVKHx+fdj5+sKP8/PxGjhzp7u7u7Ow8ZsyYtLS0urq6lJSUe56oUCiEEI2NjcaNDQ0NSqXSEnGanc0ksbsnAgAAAAAAAIC5dKbe7e7u/vLLL2dnZ+fn5wshNm/eHB8fbzi6b9++SZMmeXt7u7i4LFmypBPj19TUCCESExMNWzmXlJTot5k2Zon9u4OCghwdHb///vt79tTvW11VVWVoqa2tra+vV6lUnZu6i9lMErt7IgAAAAAAAACYS2fq3UKI+Ph4uVyenJyclZXl5+enf0KgEKK0tDQsLMzHxycvL+/27dtJSUmdGNzb21sIkZycbLybc05OTrNu5tq/21hTU1NTU5OLi8s9e/r7+/fo0aOkpMTQ8sMPPwghhg8f3rmpu55tJNEGEgEAAAAAAADALJw6d5qvr29UVFR6enp5efmKFSsM7efOndNoNHPnzh0wYIAQQiaT3XViJyeNRtPqIT8/P4VCcfr06c7F1iF//OMfMzMzDS+PHTum0+nGjh17zxOdnJymTp2alZXV1NTk4OAghNi/f79MJgsNDbVguGZlG0m0gUQAAAAAAAAAMItO3t8thFi0aFFjY+OtW7ceffRRQ6NarRZCHDx4sL6+vqioKC8v726nDxo0qLKycs+ePRqN5vr168b35yoUihdeeGHbtm2pqalVVVVarbasrOzKlSudDrUNP//88/bt23/55ReNRpOTkzN79my1Wj1nzhz90ZiYmH79+p08ebLVc5ctW1ZRUbFixYqampqcnJx169bFxsYOHjzYEnFaiG0kse1EtJ1EAAAAAAAAADaj8/XukSNHTp48ed68ecaNQUFBCQkJKSkpKpXqrbfemjRpkhAiJCRkwYIFISEhQojFixeHh4cLIebOnTt58uTnnntu8ODBK1eu1D9dcOzYsZcvXxZCbNiwYcGCBUlJSb1791apVPPnz79165YJb/OupkyZkpiY6Ovr6+rqGhUVNW7cuNzc3N69e+uPNjQ0XLt2be/eva2eO3To0MzMzAMHDvTu3Ts8PHzWrFmbN2+2RJCWYxtJbDsRbScxNzc3JCTkvvvuy8vLO3PmjEqlGjduXFZWliXiBAAAAAAAAGBRMp1OZ3ixY8eO6Oho4xYbExcX9/XXX5eVlbWzf1NT06RJk2JjY2fNmtW5GefPn79169YbN27cs2dkZKQQIiMjo3MTmX0cq2XNSZTJZOnp6VFRUZ2bCAAAAAAAAIApOn9/dzdVV1eXmZlZVFTU0NDQdk+tVrtnz55ff/01Jiamo7PodLry8vKjR4/qn50I8yKJAAAAAAAAAFrq5PMqu6/KysopU6YIIWbMmLF169Y2eh4+fHjXrl379+93dXXt6Cx79+4NCwvT/9mwOwrMhSQCAAAAAAAAaMm+9jOxZuxnYgPYzwQAAAAAAACQkN3tZwIAAAAAAAAAsEnUuwEAAAAAAAAAtoB6NwAAAAAAAADAFlDvBgAAAAAAAADYAurdAAAAAAAAAABb4NSyKTIysuvjQG5u7pgxY8w1FEkEAAAAAAAAYG9+U+/28/OLiIiQKhQ7N2bMmLFjx5o+jlkGQedERET4+flJHQUAAAAAAABgp2Q6nU7qGAAAAAAAAAAAMBX7dwMAAAAAAAAAbAH1bgAAAAAAAACALaDeDQAAAAAAAACwBf8PbckPWMlqzkcAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 1515
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jkTz9t9gmgJq"
      },
      "source": [
        ""
      ],
      "execution_count": 1515,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w5u-ZLbrmjix"
      },
      "source": [
        "The predicticton result for head data, this should really be fresh data not reusing training data\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RasjOyHOmj72",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "74cd36ff-6c2b-459d-e778-828a66c8d158"
      },
      "source": [
        "#When we use our training data above we should use our validation data here to test the predictions\n",
        "print(\"Making predictions for the following 5 patients:\")\n",
        "print(X_test)\n",
        "print(\"The predictions are\")\n",
        "print(clf.predict(X_test))"
      ],
      "execution_count": 1516,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Making predictions for the following 5 patients:\n",
            "     Mean_Radius  Mean_Texture  ...  Worst_Symmetry  Worst_Fractal_Dimension\n",
            "517       19.890         20.26  ...          0.2549                  0.09136\n",
            "314        8.597         18.60  ...          0.3142                  0.08116\n",
            "116        8.950         15.76  ...          0.1652                  0.07722\n",
            "474       10.880         15.62  ...          0.2581                  0.10800\n",
            "288       11.260         19.96  ...          0.2955                  0.07009\n",
            "..           ...           ...  ...             ...                      ...\n",
            "377       13.460         28.21  ...          0.2694                  0.07061\n",
            "60        10.170         14.88  ...          0.3557                  0.08020\n",
            "300       19.530         18.90  ...          0.2968                  0.09929\n",
            "543       13.210         28.06  ...          0.2473                  0.06443\n",
            "479       16.250         19.51  ...          0.3318                  0.09136\n",
            "\n",
            "[171 rows x 30 columns]\n",
            "The predictions are\n",
            "[1 0 0 0 0 0 0 1 0 1 0 0 1 0 0 0 1 0 0 0 0 0 1 0 1 0 0 0 0 1 0 1 0 1 0 0 0\n",
            " 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 1 1 1 1 1 0 1 0 0 0 0 1 1 0 1 1 0 1 0 0\n",
            " 0 1 0 0 0 0 0 1 1 0 0 1 0 0 1 0 0 1 1 1 0 1 0 1 0 0 1 0 1 0 0 1 0 0 0 0 0\n",
            " 0 0 0 0 0 1 1 0 1 0 0 1 0 0 1 1 0 1 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0\n",
            " 0 0 0 0 1 0 1 0 1 0 1 0 1 0 0 1 0 0 0 0 1 0 1]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Dwy10a0omoB_"
      },
      "source": [
        "Desicion tree test results\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cy2rvis0mo16",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9d97405d-68de-4228-c35f-0125affd2658"
      },
      "source": [
        "#Predict the response for test dataset\n",
        "y_pred = clf.predict(X_test)\n",
        "print(y_pred)\n",
        "# Test decision tree accuracy\n",
        "# Model Accuracy, how often is the classifier correct?\n",
        "print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred))"
      ],
      "execution_count": 1517,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[1 0 0 0 0 0 0 1 0 1 0 0 1 0 0 0 1 0 0 0 0 0 1 0 1 0 0 0 0 1 0 1 0 1 0 0 0\n",
            " 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 1 1 1 1 1 0 1 0 0 0 0 1 1 0 1 1 0 1 0 0\n",
            " 0 1 0 0 0 0 0 1 1 0 0 1 0 0 1 0 0 1 1 1 0 1 0 1 0 0 1 0 1 0 0 1 0 0 0 0 0\n",
            " 0 0 0 0 0 1 1 0 1 0 0 1 0 0 1 1 0 1 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0\n",
            " 0 0 0 0 1 0 1 0 1 0 1 0 1 0 0 1 0 0 0 0 1 0 1]\n",
            "Accuracy: 0.9239766081871345\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Dw0fHkLQmrdZ"
      },
      "source": [
        "Random forest"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JRo8cf3Fmv2_",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "12c10fb9-09d0-414b-a2fe-7e3ea5f91e4e"
      },
      "source": [
        "\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "\n",
        "rfclf = RandomForestClassifier(n_estimators=10)\n",
        "rfclf = rfclf.fit(X,y)\n",
        "#Predict result from the data given\n",
        "y_pred=rfclf.predict(X_test)\n",
        "print(y_pred)\n",
        "\n",
        "#Test Random forrrest accuracy\n",
        "# Model Accuracy, how often is the classifier correct?\n",
        "print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred))"
      ],
      "execution_count": 1518,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[1 0 0 0 0 0 0 1 1 0 0 0 1 0 0 0 1 0 0 0 0 0 1 0 1 0 0 0 0 1 0 1 1 1 1 1 0\n",
            " 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 1 0 1 1 1 0 1 0 0 0 0 0 1 0 0 1 0 1 0 0\n",
            " 0 1 0 0 1 0 0 1 1 0 0 1 0 0 1 0 0 1 1 1 0 1 0 1 0 0 1 0 1 0 1 1 0 0 0 0 0\n",
            " 0 0 0 0 1 1 0 0 1 0 0 1 0 0 1 1 0 1 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 0 1 0 0\n",
            " 0 0 0 0 1 0 1 0 1 0 1 0 1 0 0 1 0 0 0 0 1 0 1]\n",
            "Accuracy: 1.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BbN4Z5RUmzBK"
      },
      "source": [
        "So we can see that the Random forrest is more accurate\n",
        "\n",
        "We also explored the alternative to the decision tree using entropy criterion rather than the default gini\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "saPj7NNemzWp"
      },
      "source": [
        "\n",
        "entclf = tree.DecisionTreeClassifier(criterion=\"entropy\")\n",
        "\n",
        "# Train Decision Tree Classifer\n",
        "entclf = entclf.fit(X_train,y_train)"
      ],
      "execution_count": 1519,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uTkRTkehm2TZ"
      },
      "source": [
        "Display the entropy tree\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BUnbJSavm2gG",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 930
        },
        "outputId": "9a21c5c2-0aac-4e68-e822-b3db8508a61a"
      },
      "source": [
        "from IPython.display import Image\n",
        "from six import StringIO\n",
        "import pydot\n",
        "dot_data = StringIO()\n",
        "tree.export_graphviz(entclf, out_file=dot_data, feature_names=features)\n",
        "graph, = pydot.graph_from_dot_data(dot_data.getvalue())\n",
        "Image(graph.create_png())"
      ],
      "execution_count": 1520,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 1520
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3ObjuSEFm5bj"
      },
      "source": [
        "entropy tree results"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gyh4DGHqnBRc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "393391ec-a7f0-4933-b32c-842745827799"
      },
      "source": [
        "\n",
        "#Predict the response for test dataset\n",
        "y_pred = entclf.predict(X_test)\n",
        "print(y_pred)\n",
        "# Test decision tree accuracy\n",
        "# Model Accuracy, how often is the classifier correct?\n",
        "print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred))"
      ],
      "execution_count": 1521,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[1 0 0 0 0 0 0 1 0 1 0 0 1 0 0 0 1 0 0 0 0 0 1 0 1 0 0 0 1 1 0 1 1 1 0 0 0\n",
            " 1 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 1 1 1 1 1 0 1 0 0 0 0 1 1 0 0 1 0 1 0 0\n",
            " 0 1 0 0 0 0 0 1 1 0 0 1 0 0 1 0 0 1 1 1 0 1 0 1 0 0 1 0 1 0 0 1 0 0 0 0 0\n",
            " 0 0 0 0 0 1 0 0 1 0 0 1 0 0 1 1 0 1 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0\n",
            " 0 0 0 0 1 0 1 0 1 0 1 0 1 0 0 1 0 0 0 0 1 0 1]\n",
            "Accuracy: 0.9298245614035088\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DC3f-Ba9nCix"
      },
      "source": [
        "Random Entropy forrest"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CER-A40EnEny",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "43ddad45-623a-4c29-aa66-31f19074d455"
      },
      "source": [
        "\n",
        "rfentclf = RandomForestClassifier(criterion=\"entropy\",n_estimators=10)\n",
        "rfentclf = rfentclf.fit(X_train,y_train)\n",
        "#Predict result from the data given\n",
        "y_pred=rfentclf.predict(X_test)\n",
        "print(y_pred)\n",
        "\n",
        "#Test Random forrrest accuracy\n",
        "# Model Accuracy, how often is the classifier correct?\n",
        "print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred))"
      ],
      "execution_count": 1522,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[1 0 0 0 0 0 0 1 0 1 0 0 1 0 0 0 1 0 0 0 0 0 1 0 1 0 0 0 1 1 0 1 1 1 0 0 0\n",
            " 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 1 1 1 0 1 0 0 0 0 1 1 0 0 1 0 1 0 0\n",
            " 0 1 0 0 1 0 0 1 1 0 0 1 0 0 1 0 0 1 1 1 0 1 0 1 0 0 1 0 0 0 0 1 0 0 0 0 0\n",
            " 0 0 0 0 0 1 0 0 1 0 0 1 0 0 1 1 0 1 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0\n",
            " 0 0 0 0 1 0 1 0 1 0 1 0 1 0 0 1 0 0 0 0 1 0 1]\n",
            "Accuracy: 0.9298245614035088\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ujk0p7munIMh"
      },
      "source": [
        "testing via crossvalidation\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rm-TZBi4nI8j",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e59157c6-0566-4728-bcf7-d8958fe70980"
      },
      "source": [
        "from sklearn.model_selection import cross_val_score\n",
        "np.random.seed(1)\n",
        "scores = cross_val_score(clf, X, y)\n",
        "print('gini tree accuracy = {} +/- {}'\\\n",
        "        .format(scores.mean(axis=0), scores.std(axis=0)))\n",
        "scores = cross_val_score(rfclf, X, y)\n",
        "print('gini random forest accuracy = {} +/- {}'\\\n",
        "        .format(scores.mean(axis=0), scores.std(axis=0)))\n",
        "scores = cross_val_score(entclf, X, y)\n",
        "print('entropy accuracy = {} +/- {}'\\\n",
        "        .format(scores.mean(axis=0), scores.std(axis=0)))\n",
        "scores = cross_val_score(rfentclf, X, y)\n",
        "print('entropy random forest accuracy = {} +/- {}'\\\n",
        "        .format(scores.mean(axis=0), scores.std(axis=0)))\n"
      ],
      "execution_count": 1523,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "gini tree accuracy = 0.9121099208197485 +/- 0.011235106844788624\n",
            "gini random forest accuracy = 0.9525694767893185 +/- 0.016223037334090338\n",
            "entropy accuracy = 0.9331936034777207 +/- 0.01540870493426004\n",
            "entropy random forest accuracy = 0.9525384257102936 +/- 0.018922161707975938\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sPmn-_O-nL-6"
      },
      "source": [
        "Creating the Model For DecisionTreeRegression() (Problematic due to M/B classification rather than a series of numeric variables, may not be a viable model for this data)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "enY-8AB8sIMs"
      },
      "source": [
        ""
      ],
      "execution_count": 1523,
      "outputs": []
    }
  ]
}