HR-Attrition-with-ML.ipynb

hr-attrition-with-ml.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "HR-Attrition-with-ML.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/geoffmomin/7f3820642ce2fb318946c6dbbc39dea2/hr-attrition-with-ml.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "maAu-IofCfmq",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "from sklearn.model_selection import train_test_split, cross_val_score, GridSearchCV\n",
        "from sklearn.pipeline import make_pipeline\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.metrics import confusion_matrix, roc_auc_score, roc_curve"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ZPbZWgDnFI8t",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Next, we create two functions to print the **confusion matrix** and the **ROC curve**. The confusion matrix is a graphical representation of correct guesses, false positives and false negatives, while the ROC (Receiver Operating Characteristic) is a plot of the realtionship between the true positive rate and the false positive rate."
      ]
    },
    {
      "metadata": {
        "id": "zXueySTRE6wM",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def print_matrix(model, X_test, y_test, xlabel='Predicted', ylabel='True class', title=''):\n",
        "    y_pred = model.predict(X_test)\n",
        "    matrix = confusion_matrix(y_pred, y_test, [1,0])\n",
        "    sns.heatmap(matrix, annot=True, fmt='.2f', xticklabels=['Left', 'Stayed'], yticklabels=['Left', 'Stayed'])\n",
        "    plt.ylabel(ylabel)\n",
        "    plt.xlabel(xlabel)\n",
        "    plt.title(title)\n",
        "    plt.show()\n",
        "    \n",
        "def print_roc(model, X_test, y_test, label='', xlabel='False Positive Rate', ylabel='True Positive Rate', title=''):\n",
        "    y_pred = model.predict_proba(X_test)[:, 1]\n",
        "    fpr_lm, tpr_lm, _ = roc_curve(y_test, y_pred)\n",
        "    plt.plot(fpr_lm, tpr_lm, label=label)\n",
        "    plt.plot([0, 1], [0, 1], 'k--')\n",
        "    plt.xlabel(xlabel)\n",
        "    plt.ylabel(ylabel)\n",
        "    plt.title(title)\n",
        "    plt.legend(loc='best')\n",
        "    plt.legend()\n",
        "    plt.grid()\n",
        "    plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "j-ZK0_YzGzXe",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 467
        },
        "outputId": "1351b45c-64f8-4d44-de00-e67cdc9b5d62"
      },
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('https://raw.githubusercontent.com/PwCMoonshot/HR-Attrition-ML/master/HR_comma_sep.csv')\n",
        "df.head()\n",
        "dFrame = df\n",
        "\n",
        "print(df.shape)\n",
        "print()\n",
        "print(df.dtypes)\n",
        "print()\n",
        "print(df.isnull().sum())"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(14999, 10)\n",
            "\n",
            "satisfaction_level       float64\n",
            "last_evaluation          float64\n",
            "number_project             int64\n",
            "average_montly_hours       int64\n",
            "time_spend_company         int64\n",
            "Work_accident              int64\n",
            "left                       int64\n",
            "promotion_last_5years      int64\n",
            "sales                     object\n",
            "salary                    object\n",
            "dtype: object\n",
            "\n",
            "satisfaction_level       0\n",
            "last_evaluation          0\n",
            "number_project           0\n",
            "average_montly_hours     0\n",
            "time_spend_company       0\n",
            "Work_accident            0\n",
            "left                     0\n",
            "promotion_last_5years    0\n",
            "sales                    0\n",
            "salary                   0\n",
            "dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "PyPGan1MHNx9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 222
        },
        "outputId": "2ebcf17e-f33b-45bb-feff-b4591873da75"
      },
      "cell_type": "code",
      "source": [
        "df.rename(columns={\"sales\":\"department\"}, inplace=True) # Renaming the column header Sales to Department\n",
        "df.salary = df.salary.map({\"low\": 0, \"medium\": 1, \"high\": 2}) # Mapping salary levels to numerical values\n",
        "df.head() # Print the first 5 rows to the screen"
      ],
      "execution_count": 4,
      "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>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>number_project</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>left</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "      <th>department</th>\n",
              "      <th>salary</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.38</td>\n",
              "      <td>0.53</td>\n",
              "      <td>2</td>\n",
              "      <td>157</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.80</td>\n",
              "      <td>0.86</td>\n",
              "      <td>5</td>\n",
              "      <td>262</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.11</td>\n",
              "      <td>0.88</td>\n",
              "      <td>7</td>\n",
              "      <td>272</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.72</td>\n",
              "      <td>0.87</td>\n",
              "      <td>5</td>\n",
              "      <td>223</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.37</td>\n",
              "      <td>0.52</td>\n",
              "      <td>2</td>\n",
              "      <td>159</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   satisfaction_level  last_evaluation  number_project  average_montly_hours  \\\n",
              "0                0.38             0.53               2                   157   \n",
              "1                0.80             0.86               5                   262   \n",
              "2                0.11             0.88               7                   272   \n",
              "3                0.72             0.87               5                   223   \n",
              "4                0.37             0.52               2                   159   \n",
              "\n",
              "   time_spend_company  Work_accident  left  promotion_last_5years department  \\\n",
              "0                   3              0     1                      0      sales   \n",
              "1                   6              0     1                      0      sales   \n",
              "2                   4              0     1                      0      sales   \n",
              "3                   5              0     1                      0      sales   \n",
              "4                   3              0     1                      0      sales   \n",
              "\n",
              "   salary  \n",
              "0       0  \n",
              "1       1  \n",
              "2       1  \n",
              "3       0  \n",
              "4       0  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "metadata": {
        "id": "vNcpjzIzHt1m",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 161
        },
        "outputId": "936fd3e4-0631-4b8f-869b-fa0aef38600e"
      },
      "cell_type": "code",
      "source": [
        "df.groupby(\"left\").mean()"
      ],
      "execution_count": 5,
      "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>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>number_project</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "      <th>salary</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>left</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.666810</td>\n",
              "      <td>0.715473</td>\n",
              "      <td>3.786664</td>\n",
              "      <td>199.060203</td>\n",
              "      <td>3.380032</td>\n",
              "      <td>0.175009</td>\n",
              "      <td>0.026251</td>\n",
              "      <td>0.650945</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.440098</td>\n",
              "      <td>0.718113</td>\n",
              "      <td>3.855503</td>\n",
              "      <td>207.419210</td>\n",
              "      <td>3.876505</td>\n",
              "      <td>0.047326</td>\n",
              "      <td>0.005321</td>\n",
              "      <td>0.414730</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "      satisfaction_level  last_evaluation  number_project  \\\n",
              "left                                                        \n",
              "0               0.666810         0.715473        3.786664   \n",
              "1               0.440098         0.718113        3.855503   \n",
              "\n",
              "      average_montly_hours  time_spend_company  Work_accident  \\\n",
              "left                                                            \n",
              "0               199.060203            3.380032       0.175009   \n",
              "1               207.419210            3.876505       0.047326   \n",
              "\n",
              "      promotion_last_5years    salary  \n",
              "left                                   \n",
              "0                  0.026251  0.650945  \n",
              "1                  0.005321  0.414730  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "metadata": {
        "id": "Lr1ZnbTnH8i_",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Exploratory Data Visualization\n",
        "We're going to create a few graphs to display our employee data. This is to help us visualize employee attrition."
      ]
    },
    {
      "metadata": {
        "id": "TxjfJvOnINsu",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 343
        },
        "outputId": "e8586935-47db-4b94-b3a5-cbe4e10d4e31"
      },
      "cell_type": "code",
      "source": [
        "pd.crosstab(df.department, df.left).plot(kind=\"bar\")\n",
        "plt.title(\"Turnover Frequency for Department\")\n",
        "plt.xlabel(\"Department\")\n",
        "plt.ylabel(\"Frequency of Turnover\")\n",
        "plt.show()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAFFCAYAAAD2NXpMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmJlZQ/8PiiaCCI61W4MAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd4f163a0b8>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "lUFbcZT5IQ_O",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 319
        },
        "outputId": "3802f20e-8314-43a2-f9c9-59f29edb1a56"
      },
      "cell_type": "code",
      "source": [
        "sns.countplot(df.left)\n",
        "plt.xticks((0, 1), [\"Didn't leave\", \"Left\"])\n",
        "plt.xlabel(\"Class counts\")\n",
        "plt.ylabel(\"Count\")\n",
        "plt.show()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/seaborn/categorical.py:1428: FutureWarning: remove_na is deprecated and is a private function. Do not use.\n",
            "  stat_data = remove_na(group_data)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEKCAYAAAAFJbKyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAFtRJREFUeJzt3X+cXXV95/HXkEkwCQESnD6CsQJt\n3U8XddlHLaImEWiiCwq4JaS4AYTgbhXRTWAbS7csAitlbdYKBaqg0SBUoaVAwyo/NvwGJRtblx/b\n+tkiitaAzEqIg2Agmdk/zhm8TGYm95vMvXeSeT0fj3nk3u/9nns/J49J3vd7vud8T9fAwACSJDVr\nj04XIEnatRgckqQiBockqYjBIUkqYnBIkooYHJKkIt2tfPOIeDPwt8BnM/PyiPhV4MvAZOBl4OTM\nfDoiTgKWA/3AVZm5KiImA6uBA4CtwNLMfCIiDgE+BwwAj2TmGa3cB0nSq7VsxBER04HLgDsbmj9F\nFQyHAzcBZ9f9zgMWAkcAZ0XELGAJ8FxmzgMuAi6u3+MSYFlmzgX2iYijW7UPkqRttfJQ1WbgvcCG\nhraPAn9TP+4F9gMOA9Zn5qbMfBF4EJgLLKAKF4C1wNyImAIclJnr6/ZbqAJHktQmLTtUlZlbgC0R\n0dj2c4CImAScCVwIzKYKkUHPAPs3tmdmf0QM1G0bh+k7oi1btg50d0/a2d2RpImma6QXWjrHMZw6\nNK4B7srMOyNiyZAuIxU7XPuIOzZo48YXCiuUJPX0zBjxtU6cVfVl4J8y84L6+QaqkcSgOXXbK+31\nRHkX8BTV4a2hfSVJbdLW4KjPnnopMz/Z0LwOODQi9o2IvajmN+4H7gAW132OBe7OzJeB70bEvLr9\neOC29lQvSQLoatXquBHxVuAzwIFUp97+GPgV4BfAz+pu/5CZH42IE4AVVKfYXpaZf1kf0voi8Eaq\nifbTMvNHEXEwcCVV6K3LzLNHq6O3t8/lfyWpUE/PjBGnAloWHOOFwSFJ5UYLDq8clyQVMTgkSUUM\nDklSEYNDklTE4JAkFWn7leO7omUr13S6BI1Dl644rtMlSB3hiEOSVMTgkCQVMTgkSUUMDklSEYND\nklTE4JAkFTE4JElFDA5JUhGDQ5JUxOCQJBUxOCRJRQwOSVIRg0OSVMTgkCQVMTgkSUUMDklSEYND\nklTE4JAkFTE4JElFDA5JUhGDQ5JUpLuVbx4Rbwb+FvhsZl4eEb8KXANMAp4CTsnMzRFxErAc6Aeu\nysxVETEZWA0cAGwFlmbmExFxCPA5YAB4JDPPaOU+SJJerWUjjoiYDlwG3NnQfCFwRWbOBx4HTq/7\nnQcsBI4AzoqIWcAS4LnMnAdcBFxcv8clwLLMnAvsExFHt2ofJEnbauWhqs3Ae4ENDW1HAGvqx7dQ\nhcVhwPrM3JSZLwIPAnOBBcBNdd+1wNyImAIclJnrh7yHJKlNWhYcmbmlDoJG0zNzc/34GWB/YDbQ\n29Bnm/bM7Kc6NDUb2DhMX0lSm7R0jmM7usagfaS+r5g5cxrd3ZOaLkpqVk/PjE6XIHVEu4Pj+YiY\nWo9E5lAdxtpANZIYNAd4qKH94XqivItqQn2/IX0bD4VtY+PGF8aueqlBb29fp0uQWma0L0btPh13\nLbCofrwIuA1YBxwaEftGxF5U8xv3A3cAi+u+xwJ3Z+bLwHcjYl7dfnz9HpKkNmnZiCMi3gp8BjgQ\neDkiTgBOAlZHxIeBJ4GrM/PliDgHuJ1qHuOCzNwUEdcD746IB6gm2k+r33o5cGVE7AGsy8y1rdoH\nSdK2ugYGBjpdQ0v19vbt9A4uW7lm+5004Vy64rhOlyC1TE/PjBHnkL1yXJJUxOCQJBUxOCRJRQwO\nSVIRg0OSVMTgkCQVMTgkSUUMDklSEYNDklTE4JAkFTE4JElFDA5JUhGDQ5JUxOCQJBUxOCRJRQwO\nSVIRg0OSVMTgkCQVMTgkSUUMDklSEYNDklTE4JAkFTE4JElFDA5JUhGDQ5JUxOCQJBUxOCRJRQwO\nSVKR7nZ+WETsBXwFmAnsCVwAPA18DhgAHsnMM+q+K4DFdfsFmfmNiNgH+CqwD/A8sCQzn23nPkjS\nRNfuEcdpQGbmkcAJwKXAJcCyzJwL7BMRR0fEQcAHgHnAMcCfRcQkYDlwT2bOA24E/rDN9UvShNfu\n4Ph/wH7145nAs8BBmbm+brsFWAgcCdyamS9lZi/wJHAwsAC4aUhfSVIbtfVQVWZeFxGnRcTjVMFx\nLHBFQ5dngP2BnwK9w7TPbmgfbBvVzJnT6O6eNAbVS6/W0zOj0yVIHdHuOY6TgR9m5lERcQjV6GFT\nQ5euETYdrn2kvq+yceMLZUVKTert7et0CVLLjPbFqN2HquYCtwNk5sPAVOC1Da/PATbUP7O30z7Y\nJklqo3YHx+PAYQARcQDQB/xjRMyrXz8euA24C3hfREyJiNdRhcQ/AHdQnWkFsKjuK0lqo7YeqgKu\nBL4UEffWn/0RqtNxr4yIPYB1mbkWICK+ANxHdTruGZnZHxF/DlwbEfcDzwEnt7l+SZrwugYGBjpd\nQ0v19vbt9A4uW7lmLErRbubSFcd1ugSpZXp6Zow4j+yV45KkIgaHJKmIwSFJKmJwSJKKGBySpCIG\nhySpiMEhSSpicEiSihgckqQiBockqYjBIUkqYnBIkooYHJKkIgaHJKmIwSFJKmJwSJKKGBySpCIG\nhySpSFPBERG/OUzb28e+HEnSeNc92osRsS+wH/DliFgCDN6DdjLwFeBftLY8SdJ4M2pwAO8AzgL+\nNXBXQ3s/cHuripIkjV+jBkdm3grcGhEfyczPt6kmSdI4tr0Rx6CbI2IZMItfHq4iM89rSVWSpHGr\n2bOqvg4cQnWIamvDjyRpgml2xPF8Zp7e0kokSbuEZkccDw13Sq4kaeJpdsRxFHB2RPQCW6jmOQYy\n8w0tq0ySNC41GxzHjdUHRsRJwCeoAug84BHgGmAS8BRwSmZurvstp5pXuSozV0XEZGA1cADVHMvS\nzHxirGqTJG1fs4eqFozwUyQi9gM+CcwDjgHeD1wIXJGZ84HHgdMjYjpVqCwEjgDOiohZwBLgucyc\nB1wEXFxagyRp5zQ74pjf8HgKcBjwIPClws9bCKzNzD6gD/j9iPg+8JH69VuAPwASWJ+ZmwAi4kFg\nLlVYfaXuu3YHPl+StJOaCo7MXNr4PCKmAV/egc87EJgWEWuAmcD5wPTM3Fy//gywPzAb6G3Ybpv2\nzOyPiIGImJKZL430gTNnTqO7e9IOlCqNrqdnRqdLkDqi2RHHq2TmCxHxGzuwaRfV2le/SzVPcTcN\nFxQOeTx0u5L2V2zc+EJJfVLTenv7Ol2C1DKjfTFqKjgi4n5goKFpDtWkdqmfAN/MzC3A9yKiD9gS\nEVMz88X6fTfUP7OHfN5DDe0P1xPlXaONNiRJY6/ZEce5DY8HgJ8BD+/A590BrI6IT1MdqtqLarHE\nRcC19Z+3AeuAL9ar826hmt9YDuwNLK63OZZqxCJJaqOmzqrKzHupTot9a/0zNTMHRt9q2Pf5MXAD\n1ejhVuDjVGdZnVqPamYBV9ejj3OoAmItcEE9UX49MCkiHgDOBP6otAZJ0s7pGhjY/v//EXEh8B7g\nfqp5hcOBGzNz3J8O29vbVxxwQy1buWYsStFu5tIVY3Z5kzTu9PTMGHEOudlDVUcC78zMfoCI6Abu\nw+soJGnCafYCwD0GQwOgntzuH6W/JGk31eyI4+/qay/W1s/fDXy7NSVJksaz7QZHRBxEdUbT71Fd\nMT4A3JeZK1tcmyRpHBr1UFVELKBaWmRGZl6XmWdRXTF+RkS8tR0FSpLGl+3NcXwSeM/gmlEAmfko\n1TUUn2plYZKk8Wl7wdGVmY8NbczM/wO8pjUlSZLGs+0Fx16jvLbfWBYiSdo1bC84HouIjwxtjIhP\nUC0LIkmaYLZ3VtUK4OaI+CCwnuoufXOp1qp6X4trkySNQ6MGR2Y+Dby9PrvqTVS3a/2rzLyvHcVJ\nksafZm/kdCdwZ4trkSTtAppdckSSJMDgkCQVMjgkSUUMDklSEYNDklTE4JAkFTE4JElFDA5JUhGD\nQ5JUxOCQJBUxOCRJRQwOSVIRg0OSVMTgkCQVMTgkSUWauh/HWIuIqcBjwH+lus/HNVR3F3wKOCUz\nN0fEScByoB+4KjNXRcRkYDVwANVNpZZm5hMd2AVJmrA6NeI4F3i2fnwhcEVmzgceB06PiOnAecBC\n4AjgrIiYBSwBnsvMecBFwMXtLlySJrq2B0dE/CZwMPD1uukIYE39+BaqsDgMWJ+ZmzLzReBBqnud\nLwBuqvuurdskSW3UiUNVnwE+BpxaP5+emZvrx88A+wOzgd6GbbZpz8z+iBiIiCmZ+dJIHzZz5jS6\nuyeN8S5I0NMzo9MlSB3R1uCIiA8C38rM70fEcF26Rti0tP0VGze+0GR1Upne3r5OlyC1zGhfjNo9\n4ngf8GsRcQzwemAz8HxETK0PSc0BNtQ/sxu2mwM81ND+cD1R3jXaaEOSNPbaGhyZeeLg44g4H/gB\n8E5gEXBt/edtwDrgixGxL7CFai5jObA3sBi4HTgWuLt91UuSYHxcx/FJ4NSIuB+YBVxdjz7OoQqI\ntcAFmbkJuB6YFBEPAGcCf9ShmiVpwuoaGBjodA0t1dvbt9M7uGzlmu130oRz6YrjOl2C1DI9PTNG\nnEMeDyMOSdIuxOCQJBXpyJIjksbGiv9xbqdL0Di08phPtfT9HXFIkooYHJKkIgaHJKmIwSFJKmJw\nSJKKGBySpCIGhySpiMEhSSpicEiSihgckqQiBockqYjBIUkqYnBIkooYHJKkIgaHJKmIwSFJKmJw\nSJKKGBySpCIGhySpiMEhSSpicEiSihgckqQiBockqYjBIUkq0t3uD4yIPwXm1599MbAeuAaYBDwF\nnJKZmyPiJGA50A9clZmrImIysBo4ANgKLM3MJ9q9D5I0kbV1xBERRwJvzsx3AEcBlwAXAldk5nzg\nceD0iJgOnAcsBI4AzoqIWcAS4LnMnAdcRBU8kqQ2avehqvuAxfXj54DpVMGwpm67hSosDgPWZ+am\nzHwReBCYCywAbqr7rq3bJElt1NbgyMytmfnz+umHgG8A0zNzc932DLA/MBvobdh0m/bM7AcGImJK\nO2qXJFXaPscBEBHvpwqO9wD/1PBS1wiblLa/YubMaXR3TyorUGpCT8+MTpcgDavVv5udmBz/N8Af\nA0dl5qaIeD4iptaHpOYAG+qf2Q2bzQEeamh/uJ4o78rMl0b7vI0bX2jFbkj09vZ1ugRpWGPxuzla\n+LR7cnwfYCVwTGY+WzevBRbVjxcBtwHrgEMjYt+I2ItqLuN+4A5+OUdyLHB3u2qXJFXaPeI4EXgt\n8FcRMdh2KvDFiPgw8CRwdWa+HBHnALcDA8AF9ejkeuDdEfEAsBk4rc31S9KE19bgyMyrgKuGeend\nw/S9AbhhSNtWYGlrqpMkNcMrxyVJRQwOSVIRg0OSVMTgkCQVMTgkSUUMDklSEYNDklTE4JAkFTE4\nJElFDA5JUhGDQ5JUxOCQJBUxOCRJRQwOSVIRg0OSVMTgkCQVMTgkSUUMDklSEYNDklTE4JAkFTE4\nJElFDA5JUhGDQ5JUxOCQJBUxOCRJRQwOSVIRg0OSVMTgkCQV6e50ATsiIj4LvB0YAJZl5voOlyRJ\nE8YuN+KIiMOBN2bmO4APAX/e4ZIkaULZ5YIDWADcDJCZ/wjMjIi9O1uSJE0cu+KhqtnA3zU8763b\nfjZc556eGV07+4Ff/dOTdvYtpJZYvfTSTpegCWhXHHEMtdPBIElq3q4YHBuoRhiDXgc81aFaJGnC\n2RWD4w7gBICI+C1gQ2b2dbYkSZo4ugYGBjpdQ7GI+G/Au4B+4MzMfLjDJUnShLFLBockqXN2xUNV\nkqQOMjgkSUUMjt1ERBwYEX0RcU9E3BsRd0bEgvq12RFx5TDb/PeIOG2U97yn/vOEYV47PyI+NnZ7\nIG1f/Xv+7Sb7Xh4Rfx8Rew/3O6wdtyteAKiRZWYeARARvw7cEhEfyMxHgA/vxPueA9wwBvVJ7fRe\n4LeAXwBn4+/wmDE4dlOZ+b2IuAg4MyIuBm7IzN+OiJOBPwT+GXgReKwedcwDeoAAVmbmKuCDEbEC\nOCQibszM44f7rPpz5gOTgMsz82sRcQhwBfAy1dlvi4H/AnwnM79Sb/d/qRar/HfAkrrfzZn5mRb8\nlWg3FREHA5dTLXraB5wG/Aeqa7xuAR4H3hIRf5GZH+1UnbsTD1Xt3r4NHDz4JCK6gD+hWu/rOOA3\nGvq+BTge+LfAxwEy84eZuRLYNEpozAcOyMx3Ab8DnBsRU4FfAT6emUcCDwInATcCx9bb/SvgB8A+\nVNflzKM6xXpRRLxhLHZeE8ZlwIczcwHVdV5n1r+3TwNHAxdQjcYNjTHiiGP3NgPY2vB8P6AvM58B\niIgHG177VmZujYh/pvrPvFnvBN4+OB9C9WVkf+AnwKcjYhrVN7+/pAqQVRExBXg/1aGDtwFvBO5u\nqPlA4IcFNWhiexvwhYgA2BPwNgstZnDs3n4b+E7D8y6qw0GDGkecW4b0a9ZLwKrMvLixMSLuBj6d\nmbdFxB8Ae2Vmf91+OPA+qtHHPODrmbkzczCa2F4AjsxML0prEw9V7abqyfGzgc82NP8U2Cci9o2I\nycDcJt9utN+TdcCxEbFHRLwmIi6r218LfC8i9qSapJxSt98IfBD4eWb2Uq10fGRETIuIroi4tD7U\nJTXrYeAogIj4wODZhA368UvymDI4di9Rn477LeBrVMd6Xznkk5n9wPnAvVSHiR5r8n2/ExH/a7gX\nMvObVIeZvgXcxy+XvL+M6r4pf10/PrWeML+L6rjz39Tb/xC4pN72IeDpzHyx2R3WhDT4e35PfYj0\nIuA/R8S9VBPj3xnS/ylgSkT8dXvL3H255IgkqYgjDklSEYNDklTE4JAkFTE4JElFDA5JUhHPbdaE\nFxH7Ayupll0ZvA3x+Zm5tl7Ha2Fmntyp+nZURLyT6vTmJzpdi3Yvjjg0odXrd91MteTKIZk5DzgD\nuLa+iHJXthT4tU4Xod2PIw5NdAuAgcy8YrAhMx+NiH+ZmRvrRRwBiIjfBT5BtUx3N3BKZv4gIpYB\nJ1MtffFC/XhPqvW5uoCpwJWZ+aXGD46INwJfoPoC9wtgaWb+OCLOBY6hWln4MeA/AnOABzLz9fW2\n5wPdmXluRGyiugjuKKp1wn6PagHLxcDbIuIsqtHUq2rMzJ+OxV+gJh5HHJro3sQwi+Jl5sZh+u4L\nnFiv+PsNYPBGVhcCx2Tm4VRXwb8OOBH4bn1/lMOBacO83+eplrB/F/AlYHFEvANYBMzPzPlUS90v\n2c4+7A08mpm/A1wH/PvMvAn438B/ysy7RqhR2iEGhya6rVT3EWnGT4CrG5a2eG3dvgq4LSL+GPh+\nZj4K3AosjIjVVIs5bnMHRuAw4B6AzLwuMy+p2+7NzJfrPvcAhzZR2+Dqwk8Cs4Z5fbgapR1icGii\ne5RqafhXiYi3RMT0hueTgeuB36+/tQ8u5khmnk11H5NngZsj4ujM/C7VvVCuBRZSB8QQA2z7b3Do\nGkBdddvQ9ilDno+6uvFwNQ5Tj9QUg0MTWmbeC/RFxDmDbRHxJmAN8PqGrjOoVln9QUS8hup+IntG\nxMx6vuFHmfk5qrsevi0ilgCHZuZa4KPAGyJi6JziN/nlqq4nRsSfUC30eGQdVFDNwTwE/AyYVa8i\nPInqplfb0w9MHqnGZv5+pOE4OS5V9wb5s4h4jGrp+V9QzWVkPedAZj4bEV+lmg95kur03WuoRhMz\ngPURsZFqQvtDVHdA/HxEbKYaAXw6M7cM+dyPAVdFxJn1dqdn5o8i4jrg/ojYCvw98LX6Xiarqe7q\n+DjbrgA7nP9JdYhs+Qg1SjvE1XElSUU8VCVJKmJwSJKKGBySpCIGhySpiMEhSSpicEiSihgckqQi\n/x+vPDU+KKLdQgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd51c3abcf8>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "sT8AYGXCIarW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 321
        },
        "outputId": "a2bc1dfe-1f30-475a-cc54-a4d335ece541"
      },
      "cell_type": "code",
      "source": [
        "pd.crosstab(df.salary, df.left).div(pd.crosstab(df.salary, df.left).sum(1).astype(float), axis=0).plot(kind='bar', stacked=True)\n",
        "plt.xticks((0, 1, 2), [\"High\", \"Low\", \"Medium\"])\n",
        "plt.title(\"Stacked Bar Chart of Salary Level vs Turnover\")\n",
        "plt.xlabel(\"Salary Level\")\n",
        "plt.ylabel(\"Proportion of Employees\")\n",
        "plt.show()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEvCAYAAABMjRaEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n7mUAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd4eec8d198>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "2DvHyZK7JMJ2",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 253
        },
        "outputId": "48cd1000-e6a1-447c-e8fb-40e673391569"
      },
      "cell_type": "code",
      "source": [
        "plt.pie(df.department.value_counts().values, labels=df.department.value_counts().index)\n",
        "plt.legend()\n",
        "plt.axis(\"equal\")\n",
        "plt.show()"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAADrCAYAAADKbEVrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npNFNkiSpkDS6SZIkFZJGN0mSpEL+AdFiYQGe0ggZAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd4f159f240>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "heZEGWImJP5X",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 337
        },
        "outputId": "3e61e359-abe1-4044-84cc-45c240658ab7"
      },
      "cell_type": "code",
      "source": [
        "pd.crosstab(df.left, df.salary).div(pd.crosstab(df.left, df.salary).sum(1).astype(float), axis=0).plot(kind='bar', stacked=True)\n",
        "plt.title(\"Stacked Chart of Salary vs Class\")\n",
        "plt.xlabel(\"Class\")\n",
        "plt.xticks([0, 1], [\"Didn't leave\", \"Left\"])\n",
        "plt.ylabel(\"Proportion of Salary\")\n",
        "plt.show()"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAE/CAYAAABPWxQfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCcDMrKD+P3HyoS2URG5jAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd4eeb7a978>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "p0HaxamEJjV-",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "7065fbf8-6ed5-4d1d-c99d-350a1f0eabee"
      },
      "cell_type": "code",
      "source": [
        "# Create dummy variables for department feature\n",
        "df = pd.get_dummies(df, columns=[\"department\"], drop_first=True)\n",
        "#df.head()\n",
        "\n",
        "# Convert dataframe into numpy objects and split them into train and test sets: 80/20\n",
        "X = df.loc[:, df.columns != \"left\"].values\n",
        "y = df.loc[:, df.columns == \"left\"].values.flatten()\n",
        "\n",
        "# Original minority class\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y, random_state=1)\n",
        "\n",
        "print (\"Original shape:\", X_train.shape, y_train.shape )"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Original shape: (11999, 17) (11999,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "ebNp0zTgJuZe",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Random Forest ML\n",
        "We're going to use the Random Forest Classification Algorithm to predict employee attrition."
      ]
    },
    {
      "metadata": {
        "id": "HPZH-3uLJrHW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 107
        },
        "outputId": "4776c460-29fd-44f0-8400-331ace92f7f8"
      },
      "cell_type": "code",
      "source": [
        "# Build random forest classifier\n",
        "\n",
        "pip_rf = make_pipeline(StandardScaler(), RandomForestClassifier(n_estimators=500, class_weight=\"balanced\", random_state=123))\n",
        "    \n",
        "hyperparam_grid = {\n",
        "    'randomforestclassifier__n_estimators': [10, 50, 100, 500],\n",
        "    'randomforestclassifier__max_features': [\"sqrt\", \"log2\", 0.4, 0.5],\n",
        "    'randomforestclassifier__min_samples_leaf': [1, 3, 5],\n",
        "    'randomforestclassifier__criterion': [\"gini\", \"entropy\"]}\n",
        "    \n",
        "gs_rf = GridSearchCV(pip_rf,\n",
        "                    hyperparam_grid,\n",
        "                    scoring='f1',\n",
        "                    cv=10,\n",
        "                    n_jobs=-1)\n",
        "gs_rf.fit(X_train, y_train)\n",
        "\n",
        "for hyperparam in gs_rf.best_params_.keys():\n",
        "    print (hyperparam[hyperparam.find('__') + 2:], ':', gs_rf.best_params_[hyperparam]   )\n",
        "    \n",
        "print ('Best 10-folds CV f1-score: {:.2f}%.'.format(gs_rf.best_score_ * 100))"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "criterion : gini\n",
            "max_features : 0.5\n",
            "min_samples_leaf : 1\n",
            "n_estimators : 500\n",
            "Best 10-folds CV f1-score: 98.19%.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "ZYV23599PotK",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 557
        },
        "outputId": "0b422b3f-f627-485e-9bbf-3cdfa170a222"
      },
      "cell_type": "code",
      "source": [
        "print_matrix(gs_rf, X_test, y_test, title='Random Forest')\n",
        "print_roc(gs_rf, X_test, y_test,'Random Forest')"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAEVCAYAAAD91W7rAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nLiISc/8PEO3LCeslttkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd4eec43cf8>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEGCAYAAABsLkJ6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUkHq/wEKG2HMEuQbgwAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd4f1552358>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "ByTaBGhiPzL5",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        },
        "outputId": "b49fea7c-16c2-4ce9-aed7-d2544cd5c0fa"
      },
      "cell_type": "code",
      "source": [
        "features = gs_rf.best_estimator_.steps[1][1].feature_importances_\n",
        "index = dFrame.columns[dFrame.columns != 'left']\n",
        "imp_features = pd.Series(data=features[:9], index=index)\n",
        "imp_features.sort_values(ascending=False, inplace=True)\n",
        "imp_features.plot(kind='bar')\n",
        "plt.show()"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAFUCAYAAADf+HxmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3Xm8HFWd/vFPQgjKGMaoQRZxQfER\nRVEGFX4oi4yOzogo4q4j7iIo7iCICjoMqAyCMIyouKOAyOaAoLIIBCQi4IYPyqJsasSoaEYQcn9/\nnGpSudyb2wm3+3Qqz/v1uq/crurq+qbT+fapU+d8z4yxsTEiIqJbZtYOICIipl+Se0REByW5R0R0\nUJJ7REQHJblHRHRQkntERAfNqh1Az8KFt03LmMy5c9dm0aLF0/FS02oU40pM/UlM/RvFuLoe07x5\nc2ZMtL1zLfdZs9aoHcKERjGuxNSfxNS/UYxrdY2pc8k9IiKS3CMiOinJPSKig5LcIyI6KMk9IqKD\nktwjIjooyT0iooOS3CMiOmhkZqj247UHnzNtr3XsPs+YtteKiBg1ablHRHRQkntERAcluUdEdFCS\ne0REByW5R0R0UJJ7REQHJblHRHRQkntERAcluUdEdFBfM1QlHQZsBYwBe9le0Nr3BuB1wF3AlcAe\ntseWd0xERAzWlC13SdsBm9jempLEj2jtWxt4KfB029sAjwG2Xt4xERExeP10y+wInAJg+ypgrqR1\nmseLbe9o++9Nov9H4DfLOyYiIgavn+S+HrCw9Xhhs+1ukvYBrgFOsH1tP8dERMTgrExVyBnjN9g+\nWNLhwBmSLuznmPHmzl2bWbPWWIlwVs68eXOGdq6a55xKYupPYurfKMa1OsbUT3K/mWVb3RsAtwBI\negCwme3v2f4/SWcC2yzvmMksWrR4ReK+1xYuvG2o55s3b87QzzmVxNSfxNS/UYyr6zFN9iXRT7fM\n2cCuAJK2AG623YtqTeDzku7XPH4K4CmOiYiIAZuy5W57vqTLJM0HlgB7SNoN+JPtkyUdCJwr6U7K\nUMjTmqGQyxwzwL9DRESM01efu+19xm26srXv88Dn+zgmIiKGJDNUIyI6KMk9IqKDktwjIjooyT0i\nooOS3CMiOijJPSKig5LcIyI6KMk9IqKDktwjIjooyT0iooOS3CMiOijJPSKig5LcIyI6KMk9IqKD\nktwjIjooyT0iooOS3CMiOijJPSKig/paZi8mt8c575221zrqGR+dtteKiNVbWu4RER2U5B4R0UFJ\n7hERHZTkHhHRQX3dUJV0GLAVMAbsZXtBa98OwH8CdwEGXg9sC5wI/LR52o9tv3Ua446IiOWYMrlL\n2g7YxPbWkjYFjgW2bj3lGGAH2zdKOhF4NrAYON/2roMIOiIilq+fbpkdgVMAbF8FzJW0Tmv/P9m+\nsfl9IfDA6Q0xIiJWVD/JfT1K0u5Z2GwDwPafASStDzwLOKPZ9VhJp0m6UNIzpyneiIjow8pMYpox\nfoOkdYHTgbfYvlXSL4ADgBOAjYFzJT3K9h2TvejcuWsza9YaKxHOypk3b87QztWvGjHlfehPYurf\nKMa1OsbUT3K/mVZLHdgAuKX3oOmiORPYz/bZALZvAo5vnnKNpN8AGwLXTXaSRYsWr1jk99LChbcN\n9Xz9GHZM8+bNGbn3ITH1ZxRjgtGMq+sxTfYl0U+3zNnArgCStgButt2O6lDgMNvf6m2Q9ApJ725+\nXw94MHDTyoUeERErasqWu+35ki6TNB9YAuwhaTfgT8BZwL8Dm0h6fXPIccBXgeMk7QzMBnZfXpdM\nRERMr7763G3vM27Tla3f15rksJ1WKqKIiLjXMkM1IqKDktwjIjooyT0iooOS3CMiOijJPSKig5Lc\nIyI6KMk9IqKDktwjIjooyT0iooOS3CMiOijJPSKig5LcIyI6KMk9IqKDktwjIjooyT0iooOS3CMi\nOijJPSKig5LcIyI6KMk9IqKDktwjIjooyT0iooOS3CMiOijJPSKig5LcIyI6aFY/T5J0GLAVMAbs\nZXtBa98OwH8CdwEGXm97yfKOiYiIwZqy5S5pO2AT21sDrwOOGPeUY4BdbW8DzAGe3ccxERExQP10\ny+wInAJg+ypgrqR1Wvv/yfaNze8LgQf2cUxERAxQP90y6wGXtR4vbLb9GcD2nwEkrQ88C9if0k0z\n6TETmTt3bWbNWmNFYr9X5s2bM7Rz9atGTHkf+pOY+jeKca2OMfXV5z7OjPEbJK0LnA68xfatkqY8\nZrxFixavRCgrb+HC24Z6vn4MO6Z58+aM3PuQmPozijHBaMbV9Zgm+5LoJ7nfTGl192wA3NJ70HS3\nnAnsZ/vsfo6JiIjB6qfP/WxgVwBJWwA3225/5RwKHGb7WytwTEREDNCULXfb8yVdJmk+sATYQ9Ju\nwJ+As4B/BzaR9PrmkONsHzP+mMGEHxERE+mrz932PuM2Xdn6fa0+j4mIiCHJDNWIiA5Kco+I6KAk\n94iIDkpyj4jooCT3iIgOSnKPiOigJPeIiA5Kco+I6KAk94iIDkpyj4jooCT3iIgOSnKPiOigJPeI\niA5Kco+I6KAk94iIDkpyj4jooCT3iIgOSnKPiOigJPeIiA5Kco+I6KAk94iIDkpyj4jooCT3iIgO\nmtXPkyQdBmwFjAF72V7Q2ncf4FPA42xv2WzbHjgR+GnztB/bfus0xh0REcsxZXKXtB2wie2tJW0K\nHAts3XrKx4ArgMeNO/R827tOW6QREdG3frpldgROAbB9FTBX0jqt/fsCJw8gtoiIWEn9JPf1gIWt\nxwubbQDYvm2S4x4r6TRJF0p65r2IMSIiVlBffe7jzOjjOb8ADgBOADYGzpX0KNt3THbA3LlrM2vW\nGisRzsqZN2/O0M7Vrxox5X3oT2Lq3yjGtTrG1E9yv5lWSx3YALhleQfYvgk4vnl4jaTfABsC1012\nzKJFi/sIZfosXDjZBUc9w45p3rw5I/c+JKb+jGJMMJpxdT2myb4k+umWORvYFUDSFsDNy+mKoXne\nKyS9u/l9PeDBwE0rEnBERKy8KVvutudLukzSfGAJsIek3YA/2T5Z0onARoAknQccA5wGHCdpZ2A2\nsPvyumQiImJ69dXnbnufcZuubO170SSH7bSyQUVExL2TGaoRER2U5B4R0UFJ7hERHbQy49xjFXD1\n63eb+jl9vM6jP/P5extKRFSQlntERAcluUdEdFCSe0REByW5R0R0UJJ7REQHJblHRHRQkntERAdl\nnHsMzdEHnzctr7P7PttPy+tEdFmSe6zWfn35gVM/p8/XeuiTPnDvgomYRumWiYjooCT3iIgOSnKP\niOigJPeIiA5Kco+I6KAk94iIDkpyj4jooCT3iIgOSnKPiOigJPeIiA7qq/yApMOArYAxYC/bC1r7\n7gN8Cnic7S37OSYiIgZrypa7pO2ATWxvDbwOOGLcUz4GXLGCx0RExAD10y2zI3AKgO2rgLmS1mnt\n3xc4eQWPiYiIAeonua8HLGw9XthsA8D2bSt6TEREDNbKlPydMYhj5s5dm1mz1liJl1458+bNGdq5\n+jWdMV09Ta/T9fep33K+/Rj2ezWK/zYwmnGtjjH1k9xvZtlW9wbALdN9zKJFi/sIZfosXDjRBUdd\niak/oxgTDDeuefPmjOT7MIpxdT2myb4k+umWORvYFUDSFsDNk3TF3NtjIiJimkzZcrc9X9JlkuYD\nS4A9JO0G/Mn2yZJOBDYCJOk84Bjbx40/ZnB/hYiIGK+vPnfb+4zbdGVr34v6PCYiIoYkM1QjIjoo\nyT0iooOS3CMiOijJPSKig5LcIyI6KMk9IqKDktwjIjooyT0iooOS3CMiOijJPSKig5LcIyI6KMk9\nIqKDktwjIjooyT0iooOS3CMiOijJPSKig5LcIyI6KMk9IqKDktwjIjooyT0iooOS3CMiOijJPSKi\ng5LcIyI6aFY/T5J0GLAVMAbsZXtBa98/AwcBdwFn2P6wpO2BE4GfNk/7se23TmfgERExuSmTu6Tt\ngE1sby1pU+BYYOvWU44A/gW4CThf0knN9vNt7zrdAUdExNT66ZbZETgFwPZVwFxJ6wBI2hj4g+0b\nbC8BzmieHxERFfWT3NcDFrYeL2y2TbTvd8D6ze+PlXSapAslPfNeRxoREX3rq899nBl97PsFcABw\nArAxcK6kR9m+Y7ID585dm1mz1liJcFbOvHlzhnaufk1nTFdP0+t0/X369bS90vDfq1H8t4HRjGt1\njKmf5H4zS1vqABsAt0yyb0PgZts3Acc3266R9Jtm33WTnWTRosX9xjwtFi68bajn60di6s8oxgTD\njWvevDkj+T6MYlxdj2myL4l+umXOBnYFkLQFJXnfBmD7emAdSQ+XNAt4LnC2pFdIendzzHrAgyk3\nXCMiYgimbLnbni/pMknzgSXAHpJ2A/5k+2Rgd+CrzdOPt321pFuA4yTtDMwGdl9el0xEREyvvvrc\nbe8zbtOVrX3fY9mhkTQt+53udXQREbFSMkM1IqKDktwjIjooyT0iooOS3CMiOijJPSKig5LcIyI6\nKMk9IqKDktwjIjooyT0iooOS3CMiOijJPSKig5LcIyI6KMk9IqKDktwjIjooyT0iooOS3CMiOijJ\nPSKig5LcIyI6KMk9IqKDktwjIjooyT0iooOS3CMiOmhW7QAiYln7LvjFtL3WQU/eZNpeK1YtablH\nRHRQXy13SYcBWwFjwF62F7T2/TNwEHAXcIbtD091TEREDNaUyV3SdsAmtreWtClwLLB16ylHAP8C\n3AScL+kkYN4Ux0TEKuS1B58zba917D7PmLbXisn10y2zI3AKgO2rgLmS1gGQtDHwB9s32F4CnNE8\nf9JjIiJi8PrpllkPuKz1eGGz7c/Nnwtb+34HPBJ40HKOmdC8eXNmTBXI6Yfu3Ee4w3XCS46uHcKE\n5p16Uu0Q7uEDh+5UO4R7mPesj9UO4R4+/a9b1A7hHkbx/96KmDdvTu0Q7mHQMa3MDdXlJeHJ9k2Z\nuCMiYvr003K/mdLq7tkAuGWSfRs22+5YzjERETFg/bTczwZ2BZC0BXCz7dsAbF8PrCPp4ZJmAc9t\nnj/pMRERMXgzxsbGpnySpIOBbYElwB7Ak4A/2T5Z0rbAIc1TT7L98YmOsX3lAOKPiIgJ9JXcIyJi\n1ZIZqhERHZTkHhHRQUnuEREdlOQ+IJL2lDSvdhyTkTRT0v1rxwHQjLQav+0BNWIZT9I6kjaS9NDe\nT8VYtpxg2w41Yhllkp470WdqdbPKvwGSFlCKk403Axiz/ZQhh9SzDnCqpD8CXwW+YfuvlWIBQNI+\nwCLgOOA84FZJl9j+QKV4ZgFrAWdIejZLJ7ut2cT3hBpx9Uj6NPCvlLpJvdjGgKF+piQ9ChBwkKT3\ntXatCRwOPHyY8Ywn6eu2dx237RLbW1UK6XnAwZIuAI6zfUGlOJYhaSNgfduXSnolsCVwtG0P4nyr\nfHKnGU8/amwfRPnPuD6wE3CmpJuA/7F9fqWwdrK9jaQ3AKfY/rCk71SKBeA5wDspyfJnre1LKMm9\nticBD7Fde0jZfSmJYF3gRa3tS4AP1QgIQNILgX2AzSX9jqVfgDOBy2vFZfuNkmYATwWeJ+kDwA+A\nT9u+tlZcwJeBvSRtBbwW2J+lhRen3Sqf3G3/CsrlM7AnsK7ttzeXq9U+YE1MGwAvAZ4P3Ap8E3iN\npBfYfnuFkNaQNBN4OfCmZlu1ohu2TwdOl/RK21+uFcdyXEmpk7RwqicOku0fAz+WdJLtn9SMpc32\nScBJkt7dm98yQtYE1qdc1cwG/gJ8StJZFWO90/YVkj4GfML2RZLWGNTJVvnk3vJ54NvAvzWP16V0\nP/xrjWAkfY/yofoK8ELbv292fUXSxTViAr4B/AY40fbVkvYHvl8plrY7JJ1s+wUAks4GjrH99cpx\nPRK4RtIvgTup39W3i6RzWdoN2Ytn3Urx9HxX0n8B/0irjpTt19YIRtIXKVeD3wQO6U2glHQQsACo\nldxnSdqP0m20v6QnM8DGVZeS+xzbR0t6MYDt4yW9uWI8b7T980n2bT/MQFp+Mi4RHG570kqdQ/QO\n4Nmtx88DzgFqJ/dXVz7/eC8EHl773s0EvkzpXrixdiCNE4DdmjLkd7M91nQl1fJKSjfyLrb/1pRM\nH1iO6lJynynpkTStmuYG3cAuefrwYkl7th7f3cqyfXulmPaUNN/2HwFGJLFD+Xf6v9bjmYxOJdED\ngCdS+rd/AHywYiw/p1xBjJobbH+qdhAtbwEuBP44fkevG7eS/7J99z0T28cP8mRdSu57Ap8CtpR0\nC6W/9I0V43kh8IgRa2WtA9wg6RpK5c7a3Qw9nwR+IukqSqJ/NFBlBM84nwWOptz0nU254voslbr6\nKF96lvRDWkne9osrxdNzWdOPfAHLxnVGpXhG9XP+h6Zr6FJKXMDg3qcuJfcdgVfZHpXSwqPYynpF\n7QAmYvtLkk4GNqWsxftz24srhwWwRnPTsOdrzUijWo6seO7l2aD58wWtbWOUldlqmOhzPgorwc2m\n3ORtr3wysPepS8n9AZSRF/8HnAR83XbNPsBRbGV9iInnBFS58dUj6SGUlvoDbO8q6aWSLq58CQ3l\nRu+LKMMyZwDPAGp1qQFcRBkKuaHtj0vaDBjIGOkVYfs1ktaijOG+vnY8wJ8oCf6BzePZlPsnG1WL\niPI+tR9LWhP470GdrzPJ3faBwIHNRIHnUYY9/aPtp1UKaaJW1noTbBum9g3KNYGn0bo8rOgzlMk4\n+zSPf0cZ/VR79uVrgQOB91O+FC8FXlcxnk9T3pvtKSM+tgf2A15WLySQ9BLKmG2AzSQdASyw/aVK\nIZ0IzAdeChwDbEfptq1K0muBD1OG195O6YL85qDO16nyA81Y962bn/WBKyqGcxFwP+Bhzc8mwEEV\n48H2/7Z+TrH9bsrNwtrWsH0m5aYlts+h4mezaYVCmc37VuD/NT9vp7QKa9nI9t7AYgDbR7K0S6Sm\nPYEtWDof4L2UdR9qmWn7g8Attg+l3CN5zRTHDMObKcNr59teh/KlPH9QJ+tMy13SdykJ/XTgSNuX\nVA7pBOA2SuvqNEor9EMV40HS+BuB6wMb14hlnL9LegZlktWDKX23/zfFMYP0OcpEr5+ybDfWjOZx\nrfdsdlMPqDcibFNK+Yba7rJ9h6Tee1Wz6wrK+7Q5sFjSM4FrgUdVjgngb80QyNmSZto+rZm3cPgg\nTtaZ5A68w/aPJM2yPQo3Mufa3kXSebbf2vyn/B+g1qUqLDt1fQz4M6Nxk/V1LL1cPQu4hIotLdsv\nb359se0F7X3Nl1At+1HG/28i6eeUf8PXV4yn50JJXwIeImlvSrdozbIWe1AmMe5NSZwPZEAJdAUt\naIZHnw2cI+kGYO1BnawzKzE15QY+Aaxl+zGS/gP4nu2zKsUzn3LZ9VnKuNsbKJdjT6oRTyuujYHN\nKaNSLrd9Q8VY1rJ9u6T2B7zXOoYyXXvo9wTahboo9wF6Y+5nAUfYfviwY2qTtC5wu+2aXUTLkPQ0\nStfV7cCltoc+C3vc56itNxSy+gis1md+W0pj5juDmm/SpZb7AZTRDL2bhocDp1JagjXsDzyZ0iI9\nkzIU66hKsQAg6T2UWjcXUS7nPyTp07aPrhTSZN0fUP5Dzpb0I9vPGXJc7UJd7dFNtQt17Q68gWaa\nvyQAbFfpJmoKco13H+CZkp7ZDHIYpt7nqD0Brve4ZncasLT+laR2/auB3VvqUnL/u+1be/1+tn8n\naclUBw2K7e9Kmku5gfIS4OoRmBH6fOCptu+Cu0vunk+ZqDN0ve4P249o4pkLLGm3SCV9rkJckxbq\nkvT+YcfTsgely+O3FWNou7X58ymUVuj5lGS1PfDrYQfT+xxNRNJuQwxlMp9niPWvupTcr5N0IPCg\nZmjW81m2jOxQSdqX0sr6MeUDv6mkoytXz5tBMyKlsYSJx70PlaR/plzV/I3SWl9Cqc1z0fixwUP2\nUEmfp8yhgDJe+kbgI5XiuRRYPCqznm0fBSDpebbvLlsr6RDKVXMVKoua7M2y49zXoyTXmoZa/6pL\nyf2NlEv8CylDIU+jjFip5YXAY3p1ZCTdp4mtZnI/njJV/GLKF85WlLHTtR0IbN+bXdzMVTgOeHrV\nqEoXzIuAL1BG8LyQMgKqlh8Bv5L0W5atUll7xNP6kjZrXeU8iroLiHwS2Bc4BNid8m9Xe/QcDLn+\n1Sqf3McN7/sDy04K+BfqTYH+NffsT7u6RiA9tg+XdCplEYolwMEjMAsU4I522QjbN0j6e82AGn+1\nfV0zbO1W4BhJ36asrFXDm4HHAaNSYqPnHcBnJT2c8rm6EXhPxXgW2z5X0u22L6M0aL7FACcM9Wmo\n9a9W+eTOssP7xqtZ32It4HpJ36d8Oz8JuErSCVCnDEEz9vfVLK27vbOkanW3W66VdBRLp/nvAFxT\nNaLiJkmvAi6X9GXgOko/aS0XA78flW6ZHtvfpax6NCoWS3oepav2IMpnqdraty07Ai+zPZTFX1b5\n5N5Pn2zT1737MOJpOWTI5+vHVyh1t2+qHcg4b6QMG30a5Qv5QuBrVSMqXg3MpbTUX07pw92pYjyP\npHTLXMMILB6iZoEVSQuZYLJXxUVE9qAk8z0ps4qfQLn/VdtQ11Ve5ZN7n1ThnL+ijGwYvzrNsIeH\ntd1g+5iK55/MV5s61zUneE1kvwm27Ua5R1DD67hnLaAH1QgEwM3KWbbn1YphEl8C9gIeSxm5sz+l\nMN1A1irtl4e8rvLqktxrOIMy5r76sLXWfYmfSvoopWU8CnW3e4Za53oF3Nr6fU1gGypc9TRDVtei\nFFh7NstOqjqd0jKtpqmc+XKPzjKJE61VOhK5TkNcV3kk/sId9Svbo7DgBNzzvsSo1N3uGWqd6371\nhvq1fELS6RVCeQ5lwZCnUCbq9JL7XZSx5bW9k9FaJnGitUrvVymWu2nI6yonuQ/OsU0iuJxlW8lD\nv6Tv877EB20fMIx4xmvqgT+B0n22BPiZ7atqxNIm6bHjNq1PWSVqqGyfTlmr4AOVu/UmM2rLJA51\nrdIV8EbbP2+uIjaU9MdWHaztp/tkq0tyr/FB+zAj0i3Tp+1qnVjSkZRSDd+nJIZ9JF1o+x21Ymq0\nW+69Qms1Y9pe0kEjUhivbaSWSWzqJR3WejzQtUqnIulw23s1iX1H4FjgN8C6kt5s+ywPYF3lziR3\nldV8dmHiG5jPqhDSdbZrTlVfUTVbWk9pj/iQNJMB1rnul+3ai4WM91fgF5KuZNl7E1XXUPXoLpM4\nKtr3RD4I7GD7WknrASczoPpXnUnulBtL36JMoFiG7RoTYn7ZjI2+lGW7ZQa2rNa9VLMMwdWSNrB9\nc/N4HqVvuYoJhvb11B7iN9Hs5tqre9FMp39Z+4aqpJo3VEdN+7P0B9vXAtj+zSAn63Upud9q+321\ng2j5ffMzt3Ygq4BHUyYyXU25rN+YkvAXUGEc9/KG9qks/lDLRZThfO2aKe+jlJWo6R2M1g3VUbNZ\nM3lxBqUW/4tsnyjpXcAfB3XSLiX3cyTtAVzAsi3lKsXDbB8gaXvKzNS7gB/Yrt7VsBw1u2WWN8t4\n/aFFMY6kR1Bq8beT6XbUW2h55Fb3aozaDdVRM/7z/Yvmz1sok+PurvM+nSftUnLvtah2bW0bo9R4\nHzpJh1FaoOdTVlvZX9JlNfvhJf2AUpDrq+1aLo1/rxASAMurb9OU/K21+tEXKDXn306ZuLQzA6wF\n0odRXN0LRuyG6qiZbIKS7eNaD89kmj/nnUnutneQdD/KQtR3Ab+wXXMdzn+yvW3r8cGSao9J3ply\nyfwZSTMol81ft/1nV1yRaQo1W4B/t/05SbvZPgk4SdIZlP+INawl6WHAnZIeTVndq8bs62WMu6F6\nJ2DKqkzRv2n/nHcmuUt6BeUS9WeU2XwbS9rb9smVQlpT0n17XzCS/oEBlvfsh+2bKAtzHN3UvD4K\n+GgzHn/fCVrzo6Dmjd4ZkrYDbpX0RkoBqkkXhBiCkVvdC0ay+2pVNO2f884kd0qRoM17Q7CaVvxZ\nlKFGNRwG/Ki5STiTUuO6ZhnU3n/Cl1JmqN5IKW52OqVg10mktTXeqyijUd5G6ZZ5LvDuWsGM6Ope\nMHrdV0G3kvtd7bG1tv8iqdpkD9snSPpfSv/jGOU/Yu2xv18Fvgg82/YfWtvPbeqBjKKa3TJHACdS\nxm3XLos8qqt7weh1X62K0i2zHBdJ+iblBuYMyoiCC2oFM0pjf1uFwz5M+aLZSlraVWv7DNsfGnZc\nPZJeY3uytVKPm2T7MBxOaYW+X9IvKfcoTqvYWh7F1b1g9LqvRpKkh9i+cdy2TZtSG9M+qq8zyd32\n3pKeTlm1fgnwH7YvqhjSKI397Q3FGr8yfG9b7cJhz5J0se2fj99hu9oygLa/B3wPeJekzSjdav9D\nvSJUI7e6V+NVlCGrve6rf6Ni99WokfQg4MGUelO7sfT/4JqUK8NH295jus+7yid3STvbPlXSW5pN\nvbGim0vavOKM0JEZ+9srHCbp9bY/094n6Z01YhpnS8pQur9QptXXngkKgKTZlNVzdgK2paxhulvF\nkNqre80EtgB+poqrezXnvYmlpZCX6b7qLegx/KhGyqaU9+XRQDsfLQG+PKiTrvLJHbh/8+dEswpr\njrQYmbG/zazKZwEvbobQ9awJvBj4rxpx9djepOb5l+Nq4NuUm/Jvtz1+oYxhG8XVvaZy/6mf0m22\nLwAukPQV298BkLQGsI7tRYM67yqf3G1/ofn1Ltsfae+TdGiFkIDlF1PqXW0MMZxLgL9T6oK3a7Ys\noSwAUVVT9O0DlEk6L5L0UuDi5U1uGpKNgQ2Bh9m+YxCzCFfQRZQutg1tf7zpKnKl2kn9qtnAGjVb\nSnokpZ77eZRFai4Z1LoPq3xyl7QLZf3NbZua4D1rUqb+v6tKYJQRO8CCCXbtBQwtudu+DThP0uOB\nx7Ns5cwHTnrg8HyGcvNyn+bx74DPU6bX17QXZcbz/YDNgUMk3WK7Vgv605T3ZnvKTdTtKUsBvqxS\nPLFidrK9jaQ3AKfa/rCk7wzqZONvzqxybH+DcqNrAXAkZVLHUZSuhi0rhrY8tYb3fZNyQ/BtwFub\nnz0rxdK2hu0zKVcS2D6H0fge/BwqAAAPj0lEQVRsPt/2NkBv2Og7KMuj1bKR7b2BxQC2jwQ2qBhP\nrJg1mnLWL2dpsbc5gzrZKt9yB7B9fTMEayfbnwKQtA/wy7qRTarWpepc26M4Uenvkp5B+fA/mDLJ\nqmbpiJ7ejOLev9d9qPt/ZnZTT2YMyjA6yk3WqiTNsD02bts/2P4rMLA+5VXQyZRFOk60fbWk/Sld\npgMxCq2j6fIFlv0g/aTZFktdKOlxtYOYwOsorZkHUWryPxGYcmnAIThO0jmUMq1HU5ZM/GzFePYF\nvkvpu70K+AZl/dLaTm9mhAN338D/PoDtF1aLasTYPsT2uq1hj4cDPxzU+TrRcm/c1/YJvQe2vylp\nVMfa1uqWeQFlzPafWVoWufqQQ9u3SPoEZUbjGGUN1ep1bmz/dzPT8imUIZoH9QqsSXqq7e8PIw5J\n17H06mEG5f/t+pTGzJco5QhqOgr4lqQ9gT0oN6KfVzek0dPUc9qbZWvwrMeAGqFdSu6/kvRxyoiC\nmZTymUMfbSHpocvbb/vXVBp6OKpDDiX9D+Xm9wJK8tpH0kWuv4Yqtq8Hrp9g138yvFLEm1Hel32B\nKygjLWZSbjgPfcHu8Wyf2dRQOhm4wPaOtWMaUZ+k/BseAuxOaWwNrFumS8n91c3PP1OGHl4CfK1C\nHCdRWlmzKeVYr6X03T6Cclm/lctq9kMn6YnAJygtvTUoXVdvm2hm6JA9yfZTew80ImuoTmFoV19N\n3zWStrG9b2vXVyV9e1hxjNdbKau1aRbwKklPBhj2ClqrgMW2z5V0u+3LgMskfYsy0GHadSa5275T\n0iUsXeVkLUp/1uOHHMeTASR9CXhur5ZEU4f7gGHGMoEjgHc0HywkbUWZMVdrMYwe655rqP6kZkB9\nqHFT/PZm7sZ8ysiiJ1O3jHRvYZwNWTpDNSa3WNLzgOskHUSpwbPcK/17ozPJvbm03xR4DGVR6n8C\nPloxpEe3iwTZ/tW42aE13NlL7AC2L5FUbZJJq+U3mzKtvvfF/EhK90Ms64XAKynj22dQFsWoNrW/\nN8lM0hdtb1crjlXIyyl97HtSyiM/gQGugNaZ5A48zvbTVZYg20nSRpTFDWr5vqRLKaMGxihfNj+q\nGA/AHyW9h9JnO4PSYv/Dco8YrF2nfsrIGvpN8WYy2tHDPm8fbpF0EeWeyd0lGmy/t15Io6NVlbVn\nE+AHze8DG8zQpeQ+S9I6AJLm2b5B0ua1grH9tmYc8mObTcfYrt3VsBtl1uX7KZf1C6g45LDV8nsy\nZZZle+YsjCtCNWySnmv7m+O2vcz2V6lbinjUpG778i1vAfiBVWWdMTbWjdIPkl5OWYh6EWVo1t+B\n7/QqIlaI54mUS65lElbNRR9U1k19PKWY0wyafuOmtG01zUiLg4Hftrfb/t9K8TyZMvzxbZSxyD1r\nAu+x/ZAacY2qpnrmyykjnu6itEq/ZntJ1cBWIc2iK7tP52uu8i13SS9sVn/5e6+craTTgDnjVhsa\ntq9QbmDeONUTh+i7lBtwv2ttG6PULK/pKuBz42c5VvQb4C+UewHtaqNLqFvyd1R9ltKoOo+l66fu\nQFk1Kvoz7Qudr/LJHfhPSRsCe0hapuyvJCrWc7+hVwphhMyyvW3tICbwVeByST9i6eSqalc5zUSl\nL6gskzjT9u9Ulq7alLLyUSzrIbZf1Xr8tWZmb1TUheT+BspCCuNbWbVdJuljlKX+2gmr5qpHn5f0\nLsp4+3ZMtVvuH6F0y1SflTrOkZREdQVlxZzjKfcGXlI1qtEzuz2UtSnhvGblmFZ7q3xyt30+cH5T\nO/0Xtm9XWSH+YbZrDqfrVetrD1WrvaTdqyndMlu1to1Ct8zPxq8QNSIebPuUpgjdJ21/uuakoVHT\n1JC5kKbmjaQllJmzS4A31owtOpDcW94E/EDSmZS+5Ysljdl+U41gbL9G0lrA+s0U9lEw0/bTagcx\ngd9L+h7lRlz7iqL2ULq1JW1DM7a8qcg4t3JMo2R3St39aylrA18OXGT7t8s7KCY07UNru5TcN7f9\nVkl7AcfaPqzy1OyXsHSc/WaSjgAW2P5SrZiAb0t6PWWSVzuJTvvK6yvo/OZn1OwPvBc42PbvJb2f\nZUfPrNZs7wIg6TGUrtEXAAdJugU4x/aHa8Y3apruql245wi6AynLYE6rLiX3tZobq68EXiBpFnXX\nb9yTsoDxWc3j91JGE9RM7r2VjV7R2jZG/fIDMILLsdk+Gzi7tekQSrmGmv+GI8f2z5vKlVdTRj49\nlzI0Msl9WadTSlrfYwTdIJZK7FJyP4rSn32c7RslfYRyqVjLXc26m72kVXPtTQBsT7psnaQP2q5V\n+2az1u9rUu4J/AT4Yp1wCkmvAw6k1Jm/nXK/YiBFnlZFkp5NabFvTXlvLqVUZf207YU1YxtRt9p+\n37BO1pnkbvuLtJKB7fdXDAfKwhhfAjaStDelvvUo34yrVhvE9nvaj1VWhq/5xdzzJkqdmzNt79AU\nfXpE5ZhGyaHAP1CuZL4NfL/yAuKj7hxJe3DPEXQD6RZd5ZO7pJNtv0DSQpa9tJ9BxYUobL9f0tOA\nH1Nafe+2fXGNWPpUawERJK09btMGlAJwtd1u+2+SZkuaafs0SeeSfncAbD9O0gOBp1EaL//RjJi5\nmFLXvcoM4xH2zObPdk2lgXWLrvLJ3XZvqOEWvVVyeiQ9doJDhqIpXLYLZebZGLCBpOtHYYWhSdTs\n8/5p6/xjwJ+Bj9cL526XNqsLnU1pdd0A3LdyTCPF9q3AqcCpkjYA/oVyxfNOytyTaDRXf/ejFA67\nizJ0e2BrBa/yyV3Sg4AHA8dK2o2lLdBZlEv7WmV2j6cUl/pKE9PWlIU8RnGB6to+DLyVMopgJuVG\n+H5UWq+0mXw2Rhmv/bBm8xJgG8pqQwFIegSlz31bSuv9L8C5lElptedOjBxJrwA+BPyMst7ExpL2\ntj2Qz9Qqn9wpU8JfS0ni7VIDS4AvV4mo+JvtI1uPfzBB6c9RUq1bBng38HxGZ8GHiap31q7oOYpO\nBc6hjAJ5t+1FUzx/dbcnZcj2YoCmFX8WA2owrPLJ3fYFwAWSvmL7O+19kl5dKSwoyfy9wHcordGn\nAz/vdRXVGlsuaWvK7N2vSVq/1U00sEUD+nC17asrnn8ZtgeyYHHX2H5C7RhWMXf1EjuA7b9IunN5\nB9wbq3xyb/mjpBMZ0srifXhy8+dzxm0/ikpjy5vuhocCj6KsL/smSQ+w/bbx9yuGbKGkiyk34kZp\nhmrEdLpI0jcpE/ZmUFbUumBQJ+tSch/qyuJTaW6e3KcZbfEASt/tFZXL2m7ZxHVuE+OHJA3sw7UC\nLiTVFqPjbO8t6enAlpRu4/+wfdGgztel5D7UlcWnIumTlK6ZMyj9khdTWuxVat001pS0ZhNH72b0\nfSrGA6QbJLpN0s62T5X0lmZTby7A5pI2H1RZ8k4ldw1xZfE+jFStm8Z/Ua5mHtoUWNsUeEfdkCI6\nr1cGZaKS5AO7ku9Sct+DkszbK4vXXAlm1GrdYPsbks4CHkdpPVw9yHG2EbHMleldtj/S3ifp0EGd\nt0vJ/UuUxZ8fS7lRsT/wAcqkihpGrdYNTV/72Lhtd1Gucg4eodLEEZ0haRfKIi/bSmqPMFqTsu7s\nuwZx3i4l9zttX9GMCPmE7Yua1nIV42vdAPv3bqZWLNJ1AWXyxGmUJN8byfNT4HMsrRoZEdOkuWL+\nIWVlryNZOqdkCaWK5kB0KbnPkrQfpcbF/s0K9verHNPdxo2SqVWk6+njKkPOl3S27f1bN3siYprZ\nvr7Vgn8SJbH/gFJFcyC6lNxfSSnIs0sz/HBj4M2VY5pMrdmgazU3eC+ifLi2BB7UTGyqOUM1YnXw\nWWARZV2H2ZRG3g4M6N5gZ5J7MwnnsNbj4yuGM5VaY91fRBkdcwAlmf+y2bYWZXGFiBich9h+Vevx\n1ySdM6iTdSa5x9Rs3yTpQ8ADmk1rAUfbnvYlviLiHmZL2sD2zXD3sntrDupkSe51VOkCkfQBYDdK\niYZfUWbNfqpGLBGrof2A7zY172dSukbfOKiTzRzUC0cp0iXppc3v67d21SrS9RzbGwM/bIo+7UCp\nKx0RA2b7PNubUsojb2X7cYMsP5DkPiDNkMy3A70l5N4k6Qi4+/5ADWOSZlBGFt3X9g8pH7SIGDBJ\nuzdDIn8AXC7pWknXDup8Se6Ds6Xtl1BWFcL2hyhDoGr6OuUL5yvAlU3RsL/WDSlitbEHZUTfE4DH\nt34GIn3ugzOKRbrOtX15E88ZwIOAK+qGFLHauJRS4HAoDaok98E5lNEr0nWopGfZvtP2r4FfV44n\nYnXyI+BXkn5LWbdgBjDW3AebdknuA2L7ZElnM1pFuv4K/ELSlcAdvY22X1wvpIjVxpsp+eCWqZ44\nHZLcB6QpP7wbZdHnGc02bA99BaaWj1c8d8Tq7mLg9+mWWfV9jLIi1G9rB9JyEWVG6oa2Py5pM8CV\nY4pYXTyS0i1zDct2yzxlECdLch+cK4D5tv9WO5CWTwO/o5RE/njz536UYkYRMVivmvop0yfJfXC+\nBVwv6WqWXfS5ZrfMRrZf01pD9UhJL6oYT8Tq5gDgiSytCvnBQZ0oyX1w9qVUqhzKzZM+zZZ0f5YO\nz9yUUl8mIgbvs8DRwDspVSG3b7b96yBOluQ+OJcD59m+c8pnDs++lMW6N5HUWyTgdRXjiVidrGH7\npNbjr0ka2FKgSe6DMwtwM+yw3S1Tc9jh/YGnAnOBO2z/sWIsEaubO5pu0PMoN1OfQRkmPRBJ7oNz\neO0AJrALpeb994GvSzrT9sA+XBGxjNcCBwLvp/S5L2CAV85J7tNM0s62T6VMVpjI+cOMp832ayXN\nBP4fsDPwPknX2M5CHRGD9++2h9YNmuQ+/f6x+fOTlDvjE+2rxvYSSXdQLgdvB9auHFLE6mJdSc+k\ntNjbM8QXD+JkSe7T7y+STgRuBTZj6cIcsyhVId9VKzBJnwW2BX4IfAM4hNJVExGD92/ACygF+8Yo\nOWIJkNoyqwLb32hqNh8JHNXatQS4auKjhuZXwJXAusCbgLcC6wFfqBlUxGriIOAjwHWURt8cYP9B\nnSzJfQBsXw88t3YcE3gOZTjkwcBbKK2IS6pGFLH6eDuwue1b4e4y4N+hrK8w7bJYx+plse1zKcMg\nL7P9fmDP2kFFrCZuAv7QenwrcM2gTpaW++plcVOt8jpJB1E+WA+tHFPE6uLPwBWSzqc0rLemlCj5\nKIDt907nyZLcVy8vp/Sx70lziUi9xbojVjffan56FgzyZDPGxsYG+foREVFB+twjIjooyT0iooOS\n3CMiOijJPSKig5LcIyI66P8DtKmyPcQ3NV8AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd4eec41c88>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}