AhAzizPy · July 5, 2022 06:21
diff --git a/Bank_Churn_Prediction_via_Neural_networks_Project_Ahmad_Abdelaziz.ipynb b/Bank_Churn_Prediction_via_Neural_networks_Project_Ahmad_Abdelaziz.ipynb
 {
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TBOMe0eo3GBG"
   },
   "source": [
    "# Bank Churn Prediction using Artificial Neural networks Project\n",
    "*by Ahmad Abdelaziz*\n",
    "\n",
    "Background and Context\n",
    "\n",
    "Businesses like banks that provide service have to worry about the problem of 'Churn' i.e. customers leaving and joining another service provider. It is important to understand which aspects of the service influence a customer's decision in this regard. Management can concentrate efforts on the improvement of service, keeping in mind these priorities.\n",
    "\n",
    "Objective\n",
    "\n",
    "Given a Bank customer, build a neural network-based classifier that can determine whether they will leave or not in the next 6 months.\n",
    "\n",
    "Data Dictionary:\n",
    "- `CustomerId`: Unique ID which is assigned to each customer\n",
    "- `Surname`: Last name of the customer \n",
    "- `CreditScore`: It defines the credit history of the customer.  \n",
    "- `Geography`: A customer’s location    \n",
    "- `Gender`: It defines the Gender of the customer   \n",
    "- `Age`: Age of the customer     \n",
    "- `Tenure`: Number of years for which the customer has been with the bank\n",
    "- `NumOfProducts`: It refers to the number of products that a customer has purchased through the bank.\n",
    "- `Balance`: Account balance\n",
    "- `HasCrCard`: It is a categorical variable that decides whether the customer has a credit card or not.\n",
    "- `EstimatedSalary`: Estimated salary \n",
    "- `isActiveMember`: It is a categorical variable that decides whether the customer is an active member of the bank or not ( Active member in the sense, using bank products regularly, making transactions, etc )\n",
    "- `Exited`: It is a categorical variable that decides whether the customer left the bank within six months or not. It can take two values \n",
    "                    0=No ( Customer did not leave the bank )\n",
    "\n",
    "                    1=Yes ( Customer left the bank )\n",
    "\n",
    "### Contents:\n",
    "\n",
    "### A. Data Overview\n",
    "* Column Data check\n",
    "* Check the column Data types and its applicability for analysis / transformation\n",
    "* Missing value check\n",
    "* Duplicate observations check\n",
    "\n",
    "### B. EDA\n",
    "* Univariate analysis\n",
    "* Bivariate analysis\n",
    "* EDA Observations\n",
    "\n",
    "### C. Data Pre-processing\n",
    "* Rescaling the Data\n",
    "* Data Encoding\n",
    "* Split the target variable and predictors - Split the data into train and test\n",
    "\n",
    "### D. Model building\n",
    "Model 1 --> 7\n",
    "* Identifying the correct metric for model evaluation\n",
    "* Build Neural Network\n",
    "  * Locating optimal threshold using ROC-AUC curves\n",
    "* Model improvement\n",
    "* Model Performance Evaluation\n",
    "\n",
    "### E. Conclusion and key takeaways"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gYKvggX0PYl0"
   },
   "source": [
    "## Data and libraries importation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "ynK6TZb66yDc"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "import matplotlib.pylab as plt\n",
    "%matplotlib inline\n",
    "\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection  import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics import roc_curve\n",
    "from matplotlib import pyplot\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.layers import Dense, Input, Dropout,BatchNormalization\n",
    "from tensorflow.keras.wrappers.scikit_learn import KerasClassifier\n",
    "import random\n",
    "from tensorflow.keras import backend\n",
    "random.seed(1)\n",
    "np.random.seed(1) \n",
    "tf.random.set_seed(1)\n",
    "\n",
    "# To supress warnings\n",
    "import warnings\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6FyysKM38cOd",
    "outputId": "172e2706-1dc8-4a56-afb4-8550c182f32d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Drive already mounted at /content/drive/; to attempt to forcibly remount, call drive.mount(\"/content/drive/\", force_remount=True).\n"
     ]
    }
   ],
   "source": [
    "from google.colab import drive\n",
    "drive.mount('/content/drive/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "jNchbgY27Wz3"
   },
   "outputs": [],
   "source": [
    "# Loading the dataset\n",
    "import pandas as pd\n",
    "data = pd.read_csv(\"/content/drive/My Drive/AIML/Churn.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "G834slNL8Dss",
    "outputId": "875d6bf8-081d-4dfa-8bfc-2d69940f23c2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 14)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking the overall size of the dataset\n",
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "h4G4_veq9-Gd",
    "outputId": "1f0e3e32-069c-40a8-ab2a-28e835fe93b8"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-ef6a63d7-f7c7-478d-b529-69ba6ec3061e\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>RowNumber</th>\n",
       "      <th>CustomerId</th>\n",
       "      <th>Surname</th>\n",
       "      <th>CreditScore</th>\n",
       "      <th>Geography</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Tenure</th>\n",
       "      <th>Balance</th>\n",
       "      <th>NumOfProducts</th>\n",
       "      <th>HasCrCard</th>\n",
       "      <th>IsActiveMember</th>\n",
       "      <th>EstimatedSalary</th>\n",
       "      <th>Exited</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9953</th>\n",
       "      <td>9954</td>\n",
       "      <td>15655952</td>\n",
       "      <td>Burke</td>\n",
       "      <td>550</td>\n",
       "      <td>France</td>\n",
       "      <td>Male</td>\n",
       "      <td>47</td>\n",
       "      <td>2</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>97057.28</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3850</th>\n",
       "      <td>3851</td>\n",
       "      <td>15775293</td>\n",
       "      <td>Stephenson</td>\n",
       "      <td>680</td>\n",
       "      <td>France</td>\n",
       "      <td>Male</td>\n",
       "      <td>34</td>\n",
       "      <td>3</td>\n",
       "      <td>143292.95</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>66526.01</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4962</th>\n",
       "      <td>4963</td>\n",
       "      <td>15665088</td>\n",
       "      <td>Gordon</td>\n",
       "      <td>531</td>\n",
       "      <td>France</td>\n",
       "      <td>Female</td>\n",
       "      <td>42</td>\n",
       "      <td>2</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>90537.47</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3886</th>\n",
       "      <td>3887</td>\n",
       "      <td>15720941</td>\n",
       "      <td>Tien</td>\n",
       "      <td>710</td>\n",
       "      <td>Germany</td>\n",
       "      <td>Male</td>\n",
       "      <td>34</td>\n",
       "      <td>8</td>\n",
       "      <td>147833.30</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1561.58</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5437</th>\n",
       "      <td>5438</td>\n",
       "      <td>15733476</td>\n",
       "      <td>Gonzalez</td>\n",
       "      <td>543</td>\n",
       "      <td>Germany</td>\n",
       "      <td>Male</td>\n",
       "      <td>30</td>\n",
       "      <td>6</td>\n",
       "      <td>73481.05</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>176692.65</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8517</th>\n",
       "      <td>8518</td>\n",
       "      <td>15671800</td>\n",
       "      <td>Robinson</td>\n",
       "      <td>688</td>\n",
       "      <td>France</td>\n",
       "      <td>Male</td>\n",
       "      <td>20</td>\n",
       "      <td>8</td>\n",
       "      <td>137624.40</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>197582.79</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2041</th>\n",
       "      <td>2042</td>\n",
       "      <td>15709846</td>\n",
       "      <td>Yeh</td>\n",
       "      <td>840</td>\n",
       "      <td>France</td>\n",
       "      <td>Female</td>\n",
       "      <td>39</td>\n",
       "      <td>1</td>\n",
       "      <td>94968.97</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>84487.62</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1989</th>\n",
       "      <td>1990</td>\n",
       "      <td>15622454</td>\n",
       "      <td>Zaitsev</td>\n",
       "      <td>695</td>\n",
       "      <td>Spain</td>\n",
       "      <td>Male</td>\n",
       "      <td>28</td>\n",
       "      <td>0</td>\n",
       "      <td>96020.86</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>57992.49</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1933</th>\n",
       "      <td>1934</td>\n",
       "      <td>15815560</td>\n",
       "      <td>Bogle</td>\n",
       "      <td>666</td>\n",
       "      <td>Germany</td>\n",
       "      <td>Male</td>\n",
       "      <td>74</td>\n",
       "      <td>7</td>\n",
       "      <td>105102.50</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>46172.47</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9984</th>\n",
       "      <td>9985</td>\n",
       "      <td>15696175</td>\n",
       "      <td>Echezonachukwu</td>\n",
       "      <td>602</td>\n",
       "      <td>Germany</td>\n",
       "      <td>Male</td>\n",
       "      <td>35</td>\n",
       "      <td>7</td>\n",
       "      <td>90602.42</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>51695.41</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ef6a63d7-f7c7-478d-b529-69ba6ec3061e')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-ef6a63d7-f7c7-478d-b529-69ba6ec3061e button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-ef6a63d7-f7c7-478d-b529-69ba6ec3061e');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "      RowNumber  CustomerId         Surname  CreditScore Geography  Gender  \\\n",
       "9953       9954    15655952           Burke          550    France    Male   \n",
       "3850       3851    15775293      Stephenson          680    France    Male   \n",
       "4962       4963    15665088          Gordon          531    France  Female   \n",
       "3886       3887    15720941            Tien          710   Germany    Male   \n",
       "5437       5438    15733476        Gonzalez          543   Germany    Male   \n",
       "8517       8518    15671800        Robinson          688    France    Male   \n",
       "2041       2042    15709846             Yeh          840    France  Female   \n",
       "1989       1990    15622454         Zaitsev          695     Spain    Male   \n",
       "1933       1934    15815560           Bogle          666   Germany    Male   \n",
       "9984       9985    15696175  Echezonachukwu          602   Germany    Male   \n",
       "\n",
       "      Age  Tenure    Balance  NumOfProducts  HasCrCard  IsActiveMember  \\\n",
       "9953   47       2       0.00              2          1               1   \n",
       "3850   34       3  143292.95              1          1               0   \n",
       "4962   42       2       0.00              2          0               1   \n",
       "3886   34       8  147833.30              2          0               1   \n",
       "5437   30       6   73481.05              1          1               1   \n",
       "8517   20       8  137624.40              2          1               1   \n",
       "2041   39       1   94968.97              1          1               0   \n",
       "1989   28       0   96020.86              1          1               1   \n",
       "1933   74       7  105102.50              1          1               1   \n",
       "9984   35       7   90602.42              2          1               1   \n",
       "\n",
       "      EstimatedSalary  Exited  \n",
       "9953         97057.28       0  \n",
       "3850         66526.01       0  \n",
       "4962         90537.47       0  \n",
       "3886          1561.58       0  \n",
       "5437        176692.65       0  \n",
       "8517        197582.79       0  \n",
       "2041         84487.62       0  \n",
       "1989         57992.49       0  \n",
       "1933         46172.47       0  \n",
       "9984         51695.41       0  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.sample(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wBD_9sb__RHJ"
   },
   "source": [
    "## A. **Data Overview**\n",
    "#### 1. First, Check if all columns contain important data to be analysed and for the model that will be built later"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "WXWv4R0dPRrQ",
    "outputId": "fc903633-30ce-43eb-fd69-bc741e541adb"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-deb4d589-a066-4143-8980-a1606f98e03d\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>count</th>\n",
       "      <th>unique</th>\n",
       "      <th>top</th>\n",
       "      <th>freq</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>RowNumber</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5000.5</td>\n",
       "      <td>2886.89568</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2500.75</td>\n",
       "      <td>5000.5</td>\n",
       "      <td>7500.25</td>\n",
       "      <td>10000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CustomerId</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>15690940.5694</td>\n",
       "      <td>71936.186123</td>\n",
       "      <td>15565701.0</td>\n",
       "      <td>15628528.25</td>\n",
       "      <td>15690738.0</td>\n",
       "      <td>15753233.75</td>\n",
       "      <td>15815690.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Surname</th>\n",
       "      <td>10000</td>\n",
       "      <td>2932</td>\n",
       "      <td>Smith</td>\n",
       "      <td>32</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CreditScore</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>650.5288</td>\n",
       "      <td>96.653299</td>\n",
       "      <td>350.0</td>\n",
       "      <td>584.0</td>\n",
       "      <td>652.0</td>\n",
       "      <td>718.0</td>\n",
       "      <td>850.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Geography</th>\n",
       "      <td>10000</td>\n",
       "      <td>3</td>\n",
       "      <td>France</td>\n",
       "      <td>5014</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gender</th>\n",
       "      <td>10000</td>\n",
       "      <td>2</td>\n",
       "      <td>Male</td>\n",
       "      <td>5457</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Age</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>38.9218</td>\n",
       "      <td>10.487806</td>\n",
       "      <td>18.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>92.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tenure</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.0128</td>\n",
       "      <td>2.892174</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Balance</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>76485.889288</td>\n",
       "      <td>62397.405202</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>97198.54</td>\n",
       "      <td>127644.24</td>\n",
       "      <td>250898.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NumOfProducts</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.5302</td>\n",
       "      <td>0.581654</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HasCrCard</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.7055</td>\n",
       "      <td>0.45584</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IsActiveMember</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.5151</td>\n",
       "      <td>0.499797</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EstimatedSalary</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100090.239881</td>\n",
       "      <td>57510.492818</td>\n",
       "      <td>11.58</td>\n",
       "      <td>51002.11</td>\n",
       "      <td>100193.915</td>\n",
       "      <td>149388.2475</td>\n",
       "      <td>199992.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exited</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.2037</td>\n",
       "      <td>0.402769</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-deb4d589-a066-4143-8980-a1606f98e03d')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-deb4d589-a066-4143-8980-a1606f98e03d button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-deb4d589-a066-4143-8980-a1606f98e03d');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "                   count unique     top  freq           mean           std  \\\n",
       "RowNumber        10000.0    NaN     NaN   NaN         5000.5    2886.89568   \n",
       "CustomerId       10000.0    NaN     NaN   NaN  15690940.5694  71936.186123   \n",
       "Surname            10000   2932   Smith    32            NaN           NaN   \n",
       "CreditScore      10000.0    NaN     NaN   NaN       650.5288     96.653299   \n",
       "Geography          10000      3  France  5014            NaN           NaN   \n",
       "Gender             10000      2    Male  5457            NaN           NaN   \n",
       "Age              10000.0    NaN     NaN   NaN        38.9218     10.487806   \n",
       "Tenure           10000.0    NaN     NaN   NaN         5.0128      2.892174   \n",
       "Balance          10000.0    NaN     NaN   NaN   76485.889288  62397.405202   \n",
       "NumOfProducts    10000.0    NaN     NaN   NaN         1.5302      0.581654   \n",
       "HasCrCard        10000.0    NaN     NaN   NaN         0.7055       0.45584   \n",
       "IsActiveMember   10000.0    NaN     NaN   NaN         0.5151      0.499797   \n",
       "EstimatedSalary  10000.0    NaN     NaN   NaN  100090.239881  57510.492818   \n",
       "Exited           10000.0    NaN     NaN   NaN         0.2037      0.402769   \n",
       "\n",
       "                        min          25%         50%          75%         max  \n",
       "RowNumber               1.0      2500.75      5000.5      7500.25     10000.0  \n",
       "CustomerId       15565701.0  15628528.25  15690738.0  15753233.75  15815690.0  \n",
       "Surname                 NaN          NaN         NaN          NaN         NaN  \n",
       "CreditScore           350.0        584.0       652.0        718.0       850.0  \n",
       "Geography               NaN          NaN         NaN          NaN         NaN  \n",
       "Gender                  NaN          NaN         NaN          NaN         NaN  \n",
       "Age                    18.0         32.0        37.0         44.0        92.0  \n",
       "Tenure                  0.0          3.0         5.0          7.0        10.0  \n",
       "Balance                 0.0          0.0    97198.54    127644.24   250898.09  \n",
       "NumOfProducts           1.0          1.0         1.0          2.0         4.0  \n",
       "HasCrCard               0.0          0.0         1.0          1.0         1.0  \n",
       "IsActiveMember          0.0          0.0         1.0          1.0         1.0  \n",
       "EstimatedSalary       11.58     51002.11  100193.915  149388.2475   199992.48  \n",
       "Exited                  0.0          0.0         0.0          0.0         1.0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Display basic descriptive statistics around the data\n",
    "data.describe(include=\"all\").T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ePVvUMFUUgWi"
   },
   "source": [
    "- `CreditScore` is having a minimum of 350 and a max of 850 with 75% of customers having score ranging from 350 upto 718.\n",
    "- 3 unique locations are available in `Geography` with France as top location.\n",
    "- For `Gender`; Males are dominating the list\n",
    "- Client `Age` is ranging between 18 and 92, 75% of the customers are between 18 and 44.\n",
    "- `Tenure` is ranging between 0 and 10, 75% of customers are ranging between 0 and 7\n",
    "- Clients `Balance` is ranging between 0 and 250K with std of 62K.\n",
    "- Customers are having a maximum of 4 products per client.\n",
    "- `EstimatedSalary` is ranging between 11 and 200K with std 57K.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "fP1RN7-RAbNg",
    "outputId": "62fb929e-9ca3-419b-dd7a-9c3522274ff8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# let's check for duplicate values in the data\n",
    "data.duplicated(keep='last').sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "uA6dymf6Utqp"
   },
   "source": [
    "No duplicates to be dropped"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "Z534ui9s_R_N"
   },
   "outputs": [],
   "source": [
    "#Dropping the CustomerId as it is a bank specific number unique to the client and not important for analysis\n",
    "#Dropping Rownumber as it is a duplication for the pandas index\n",
    "#Dropping Surname as it is customer unique feature and is not important for analysis\n",
    "data.drop('RowNumber',axis=1,inplace=True)\n",
    "data.drop('CustomerId',axis=1,inplace=True)\n",
    "data.drop('Surname',axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "TlO7wfxIAiOu",
    "outputId": "588aba43-bbb0-476a-b9aa-22b0023bc061"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 10000 entries, 0 to 9999\n",
      "Data columns (total 11 columns):\n",
      " #   Column           Non-Null Count  Dtype  \n",
      "---  ------           --------------  -----  \n",
      " 0   CreditScore      10000 non-null  int64  \n",
      " 1   Geography        10000 non-null  object \n",
      " 2   Gender           10000 non-null  object \n",
      " 3   Age              10000 non-null  int64  \n",
      " 4   Tenure           10000 non-null  int64  \n",
      " 5   Balance          10000 non-null  float64\n",
      " 6   NumOfProducts    10000 non-null  int64  \n",
      " 7   HasCrCard        10000 non-null  int64  \n",
      " 8   IsActiveMember   10000 non-null  int64  \n",
      " 9   EstimatedSalary  10000 non-null  float64\n",
      " 10  Exited           10000 non-null  int64  \n",
      "dtypes: float64(2), int64(7), object(2)\n",
      "memory usage: 859.5+ KB\n"
     ]
    }
   ],
   "source": [
    "#using info method to check on available datapoints and their type\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Ups7VQx4Ume8"
   },
   "source": [
    "There is no missing data, all data seem to be numeric form"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1VdAKWduC9Rt"
   },
   "source": [
    "## B. **EDA**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9aoZlAZ5C_jh"
   },
   "source": [
    "### Univariate Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "KNIrdJ2pAj2a",
    "outputId": "efc08727-61b3-41a9-8f49-cdd1542a8c2e"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x1008 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# creating histograms for numerical variables\n",
    "data.hist(figsize=(14, 14))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "OhpxDwNvZTHa"
   },
   "source": [
    "- `CreditScore` and `Age` are partialy normaly distributed\n",
    "- Majority of customers have a `Tenure` between 1 and 9 years.\n",
    "- Around 3500 customers have a 0 balance, rest of custumers have a normally distrubuted `Balance` between 50K and 200K.\n",
    "- More than half of the customers are using 1 product, vast majority are using 2 products.\n",
    "- Around 7000 clients have credit cards\n",
    "- Around 50% of customers are active.\n",
    "- Around 20% of customers left the bank within 6 months according to given dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "QlsGFqFcHMvy"
   },
   "outputs": [],
   "source": [
    "# Function to plot a boxplot and a histogram along the same scale.\n",
    "\n",
    "\n",
    "def histogram_boxplot(data, feature, figsize=(12, 7), kde=False, bins=None):\n",
    "    \"\"\"\n",
    "    Boxplot and histogram combined\n",
    "\n",
    "    data: dataframe\n",
    "    feature: dataframe column\n",
    "    figsize: size of figure (default (12,7))\n",
    "    kde: whether to the show density curve (default False)\n",
    "    bins: number of bins for histogram (default None)\n",
    "    \"\"\"\n",
    "    f2, (ax_box2, ax_hist2) = plt.subplots(\n",
    "        nrows=2,  # Number of rows of the subplot grid= 2\n",
    "        sharex=True,  # x-axis will be shared among all subplots\n",
    "        gridspec_kw={\"height_ratios\": (0.25, 0.75)},\n",
    "        figsize=figsize,\n",
    "    )  # creating the 2 subplots\n",
    "    sns.boxplot(\n",
    "        data=data, x=feature, ax=ax_box2, showmeans=True, color=\"violet\"\n",
    "    )  # boxplot will be created and a star will indicate the mean value of the column\n",
    "    sns.histplot(\n",
    "        data=data, x=feature, kde=kde, ax=ax_hist2, bins=bins, palette=\"winter\"\n",
    "    ) if bins else sns.histplot(\n",
    "        data=data, x=feature, kde=kde, ax=ax_hist2\n",
    "    )  # For histogram\n",
    "    ax_hist2.axvline(\n",
    "        data[feature].mean(), color=\"green\", linestyle=\"--\"\n",
    "    )  # Add mean to the histogram\n",
    "    ax_hist2.axvline(\n",
    "        data[feature].median(), color=\"black\", linestyle=\"-\"\n",
    "    )  # Add median to the histogram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "6xYPeo6wHaIw",
    "outputId": "96129cbe-c77d-482f-cbac-f1dd7907160e"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "histogram_boxplot(data, \"CreditScore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "SgYDAD7Ya_Mh"
   },
   "source": [
    "Credit score is normally distributed with mean very close to median value at around 650.\n",
    "Some outliers lie on both ends."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "1oK3RHyVHfZn",
    "outputId": "dbf77f59-6520-4311-90b8-8418dbccbbf2"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "histogram_boxplot(data, \"Age\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mef5zVeFbwHb"
   },
   "source": [
    "Age is almost uniformly distributed with mean close to median around 35+"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "4LEWVtRHHfeT",
    "outputId": "65aac8f9-a060-4fa9-c5cd-b61309041dc9"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "histogram_boxplot(data, \"Tenure\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cWC4gPUrcA6x"
   },
   "source": [
    "Mean Teunure of the customers is for 5 years, In terms of count most of customer's tenure is between 1 and 9 years."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "2O56isCBHfhQ",
    "outputId": "644b18b4-16c0-4342-cf86-5e451b0df119"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "histogram_boxplot(data, \"EstimatedSalary\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cTYFXRxfctQ0"
   },
   "source": [
    "Mean salary of clients is around 100K, bank is having a fair share from each salary category and have no specific range overcoming the other."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "OHMDHSslDQ7F"
   },
   "outputs": [],
   "source": [
    "# function to create labeled barplots\n",
    "\n",
    "\n",
    "def labeled_barplot(data, feature, perc=False, n=None):\n",
    "    \"\"\"\n",
    "    Barplot with percentage at the top\n",
    "\n",
    "    data: dataframe\n",
    "    feature: dataframe column\n",
    "    perc: whether to display percentages instead of count (default is False)\n",
    "    n: displays the top n category levels (default is None, i.e., display all levels)\n",
    "    \"\"\"\n",
    "\n",
    "    total = len(data[feature])  # length of the column\n",
    "    count = data[feature].nunique()\n",
    "    if n is None:\n",
    "        plt.figure(figsize=(count + 1, 5))\n",
    "    else:\n",
    "        plt.figure(figsize=(n + 1, 5))\n",
    "\n",
    "    plt.xticks(rotation=90, fontsize=15)\n",
    "    ax = sns.countplot(\n",
    "        data=data,\n",
    "        x=feature,\n",
    "        palette=\"Paired\",\n",
    "        order=data[feature].value_counts().index[:n].sort_values(),\n",
    "    )\n",
    "\n",
    "    for p in ax.patches:\n",
    "        if perc == True:\n",
    "            label = \"{:.1f}%\".format(\n",
    "                100 * p.get_height() / total\n",
    "            )  # percentage of each class of the category\n",
    "        else:\n",
    "            label = p.get_height()  # count of each level of the category\n",
    "\n",
    "        x = p.get_x() + p.get_width() / 2  # width of the plot\n",
    "        y = p.get_height()  # height of the plot\n",
    "\n",
    "        ax.annotate(\n",
    "            label,\n",
    "            (x, y),\n",
    "            ha=\"center\",\n",
    "            va=\"center\",\n",
    "            size=12,\n",
    "            xytext=(0, 5),\n",
    "            textcoords=\"offset points\",\n",
    "        )  # annotate the percentage\n",
    "\n",
    "    plt.show()  # show the plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "O7pxy2JYEMN3",
    "outputId": "d24a6ecf-bdfe-462a-8b69-b9e9f7b7cbdb"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "labeled_barplot(data, \"Geography\", perc=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "d1mW8uRmc_GE"
   },
   "source": [
    "50% of customers are from france, other 25% are from Germany and Spain."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "rWg3AgPMEWEJ",
    "outputId": "6b186ecb-af7a-4906-c033-4f66c676da96"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "labeled_barplot(data, \"Gender\", perc=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bPjt0JsIhUUx"
   },
   "source": [
    "Around 55% of customers are Males"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "rDzZ8h-jEWT0",
    "outputId": "5316c4d5-ae07-4098-a672-db0fd25c6dc2"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "labeled_barplot(data, \"NumOfProducts\", perc=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "L5Yml54ahZhA"
   },
   "source": [
    "50% of customer are tied to 1 product, 45% are tied to 2 products"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "UL3BbKHUEWdC",
    "outputId": "6f0f65b5-f4c1-4f9a-e1d2-7275b9422e1b"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOMAAAE8CAYAAAAykQYCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAbPUlEQVR4nO3df3RV5Z3v8fdXEPkRhCAROoCQiwwsk4qDUfxxq4NYwB8jVsdf9UfqZRrlUoud8c4o61oUtLVLRkfqLS1LacHliCh6wbk6GEHoGqotoYaf2iZI+KVANIAKBqF+7x/nSTzBE4h4dvIkfF5rnXX2fvZznv1sVj7sfZ6zz3nM3RGRlndcS3dARFIURpFIKIwikVAYRSKhMIpEQmEUiUT7lu5AEnr27OkDBgxo6W6IZLRy5coP3D3v0PI2GcYBAwZQVlbW0t0QycjMNmUq12WqSCQURmlxOTk5DR7t2rXjjjvuqN++ePFihgwZQufOnRkxYgSbNmU8sQCpq6JOnTrVtzVq1KgG7eTn59O7d2/mzp1bX757926GDRvGxx9/nMwBNpW7t7nHmWee6dI6ffzxx96lSxdftmyZu7tXV1f7iSee6PPmzfNPP/3U77rrLh8+fHijr+/fv7+XlpZm3FZYWOhr1qzx8vJyz83N9YMHD7q7++233+7PPvts9g+mEUCZZ/i7bZPvGaX1mj9/PieffDLf+ta3AHjhhRcoKCjgmmuuAeC+++6jZ8+evPPOOwwZMuQrtb13714KCwsB6NChAx9++CFVVVVs3LiRGTNmZPdAjoIuUyUqs2fP5pZbbsHMAFi3bh1Dhw6t396lSxcGDhzIunXrGm3jxhtvJC8vj1GjRrFq1ar68pNPPplVq1axatUqjjvuOHJzc5k4cSLTp09P7oC+AoVRorFp0yaWLVtGcXFxfdknn3xCt27dGtTr1q1bo+/vnn76aaqqqti0aRMjRoxg9OjR7N69G4Bf/vKXTJw4kZKSEp566ilmzJjBxRdfTG1tLaNHj2bEiBEsW7YsuQM8kkzXrq39ofeMrdPUqVP9ggsuaFD2wx/+0MePH9+grLCw0J9//vkmtTl48GBfuHDhl8rfe+89Hzp0qO/bt8/POussX758uW/cuNH79evnn3/++dEfRBPQyHtGnRklGnPmzGlwVgQoKChocKm5d+9eNmzYQEFBQZPaNDM8w3d2f/SjH/HAAw/QqVMn1qxZQ1FREQMGDODAgQNUV1d/vQM5SgqjROF3v/sd27Ztqx+oqfOd73yHtWvXMn/+fGpra5kyZQqnn356xsGbzZs3s3z5cj777DNqa2t5+OGH+eCDDzj//PMb1CstLaW2tpbLL78cgPz8fJYsWcK6devYv38/J510UnIHejiZTpfZeACDgfK0x0fAnUAPoBSoCM+5ob4B04FKYDUwLK2t4lC/Aig+0r51mdr6lJSU+E033ZRxW2lpqQ8ePNg7duzoF154oW/cuLF+22233ea33Xabu7uvXbvWv/nNb3rnzp29R48eftFFF/mKFSsatFVbW+tDhw71qqqq+rLXXnvN+/fv77179/Znnnkm+wd3CBq5TDXPcArPNjNrB2wDhgMTgBp3f8jM7g5h/BczuxS4A7g01HvM3YebWQ+gDCgCHFgJnOnuuxrbX1FRket2uOZz5v+a09JdiMbKh285Yh0zW+nuRYeWN9dl6khgg7tvAsYCs0P5bODKsDwWmBP+83gT6G5m3wBGA6XuXhMCWAqMaaZ+izSb5grj9cAzYbmXu78flrcDvcJyH2BL2mu2hrLGyhswsxIzKzOzspZ6Ay7ydSQeRjPrAFwBPHfotnD9nJXrZHef6e5F7l6Ul/elb6eIRK85zoyXAH909x1hfUe4/CQ87wzl24B+aa/rG8oaKxdpU5ojjDfwxSUqwEJSo6OE5wVp5bdYyjnAnnA5uwgYZWa5ZpYLjAplIm1KojeKm1kX4NvAbWnFDwHzzGwcsAm4NpS/TGoktRLYB9wK4O41ZjYVWBHqTXH3miT7LdISEg2ju+8FTjqk7ENSo6uH1nVSH3tkamcWMCuJPorEQnfgiERCYRSJhMIoEgmFUSQSCqNIJBRGkUgojCKRUBhFIqEwikRCYRSJhMIoEgmFUSQSCqNIJBRGkUgojCKRUBhFIqEwikRCYRSJhMIoEgmFUSQSCqNIJBRGkUgojCKRUBhFIqEwikRCYRSJhMIoEgmFUSQSiYbRzLqb2fNm9o6ZvW1m55pZDzMrNbOK8Jwb6pqZTTezSjNbbWbD0topDvUrzKy48T2KtF5JnxkfA/7T3YcAQ4G3gbuBxe4+CFgc1iE1qeqg8CgBZgCYWQ9gMjAcOBuYXBdgkbYksTCaWTfgAuBJAHf/zN13A2OB2aHabODKsDwWmOMpbwLdw8zGo4FSd69x911AKTAmqX6LtJQkz4z5QDXwazN7y8yeCJOn9gozEgNsB3qF5T7AlrTXbw1ljZU3YGYlZlZmZmXV1dVZPhSR5CUZxvbAMGCGu/8NsJcvLkmB+glSPRs7c/eZ7l7k7kV5eXnZaFKkWSUZxq3AVnf/fVh/nlQ4d4TLT8LzzrB9G9Av7fV9Q1lj5SJtSmJhdPftwBYzGxyKRgLrgYVA3YhoMbAgLC8EbgmjqucAe8Ll7CJglJnlhoGbUaFMpE1pn3D7dwBPm1kH4F3gVlL/Acwzs3HAJuDaUPdl4FKgEtgX6uLuNWY2FVgR6k1x95qE+y3S7BINo7uXA0UZNo3MUNeBCY20MwuYld3eicRFd+CIREJhFImEwigSCYVRJBIKo0gkFEaRSCiMIpFQGEUioTCKREJhFImEwigSCYVRJBIKo0gkFEaRSCiMIpFQGEUioTCKREJhFImEwigSCYVRJBIKo0gkFEaRSCiMIpFQGEUioTCKREJhFImEwigSCYVRJBKJhtHMqsxsjZmVm1lZKOthZqVmVhGec0O5mdl0M6s0s9VmNiytneJQv8LMihvbn0hr1hxnxhHufoa7181GdTew2N0HAYv5YjbjS4BB4VECzIBUeIHJwHDgbGByXYBF2pKWuEwdC8wOy7OBK9PK53jKm0D3MLPxaKDU3WvcfRdQCoxp7k6LJC3pMDrwqpmtNLOSUNYrzEgMsB3oFZb7AFvSXrs1lDVW3oCZlZhZmZmVVVdXZ/MYRJpF0jMX/3d332ZmJwOlZvZO+kZ3dzPzbOzI3WcCMwGKioqy0qZIc0r0zOju28LzTuBFUu/5doTLT8LzzlB9G9Av7eV9Q1lj5SJtSmJhNLMuZta1bhkYBawFFgJ1I6LFwIKwvBC4JYyqngPsCZezi4BRZpYbBm5GhTKRNiXJy9RewItmVreff3f3/zSzFcA8MxsHbAKuDfVfBi4FKoF9wK0A7l5jZlOBFaHeFHevSbDfIi0isTC6+7vA0AzlHwIjM5Q7MKGRtmYBs7LdR5GY6A4ckUgojCKRUBhFIqEwikRCYRSJhMIoEgmFUSQSCqNIJBRGkUgojCKRUBhFIqEwikRCYRSJhMIoEgmFUSQSCqNIJBRGkUgojCKRUBhFIqEwikRCYRSJhMIoEgmFUSQSCqNIJBRGkUgojCKRUBhFIpF4GM2snZm9ZWb/Edbzzez3ZlZpZs+aWYdQfkJYrwzbB6S1cU8o/5OZjU66zyItoTnOjBOBt9PWfwY86u6nAruAcaF8HLArlD8a6mFmpwHXAwWkpg//hZm1a4Z+izSrRMNoZn2By4AnwroBFwHPhyqzgSvD8tiwTtg+MtQfC8x19/3uvpHUlHFnJ9lvkZbQpDCa2eKmlGXwb8A/A5+H9ZOA3e5+MKxvBfqE5T7AFoCwfU+oX1+e4TUibcZhw2hmHc2sB9AzzBzcIzwGcIRAmNnlwE53X5m13h5+fyVmVmZmZdXV1c2xS5GsOtJkqbcBdwJ/BawELJR/BDx+hNeeD1xhZpcCHYETgceA7mbWPpz9+gLbQv1tQD9gq5m1B7oBH6aV10l/TT13nwnMBCgqKvIj9E0kOoc9M7r7Y+6eD9zl7v/N3fPDY6i7HzaM7n6Pu/d19wGkBmCWuPuNwOvA34dqxcCCsLwwrBO2LwmzGS8Erg+jrfnAIOAPX/1QReLWpGnE3f3nZnYeMCD9Ne4+5yj2+S/AXDN7AHgLeDKUPwk8ZWaVQA2pAOPu68xsHrAeOAhMcPe/HMV+RaLWpDCa2VPAQKAcqAuCA00Ko7svBZaG5XfJMBrq7rXANY28/kHgwabsS6S1alIYgSLgtHDZKCIJaOrnjGuB3kl2RORY19QzY09gvZn9AdhfV+juVyTSK5FjUFPDeF+SnRCRpo+mLku6IyLHuqaOpn5MavQUoANwPLDX3U9MqmMix5qmnhm71i2n3bx9TlKdEjkWfeVvbXjK/wX0vUKRLGrqZepVaavHkfrcsTaRHokco5o6mvp3acsHgSpSl6oikiVNfc94a9IdETnWNfXLxX3N7EUz2xke88O3+EUkS5o6gPNrUl9l+qvweCmUiUiWNDWMee7+a3c/GB6/AfIS7JfIMaepYfzQzG4KP7vYzsxuIvUtfBHJkqaG8X8A1wLbgfdJfRP/ewn1SeSY1NSPNqYAxe6+CyD8SNU0UiEVkSxo6pnx9LogArh7DfA3yXRJ5NjU1DAeZ2a5dSvhzNjUs6qINEFTA/WvwBtm9lxYvwb9Jo1IVjX1Dpw5ZlZG6qf5Aa5y9/XJdUvk2NPkS80QPgVQJCGan1EkEgrjUdi/fz/jxo2jf//+dO3alTPOOINXXnmlfvsTTzzBqaeeSk5ODmPGjOG99947YpsVFRV07NiRm266qb5s1apVFBQU0LNnTx555JH68gMHDjB8+HC2bNmSqSlppRTGo3Dw4EH69evHsmXL2LNnDw888ADXXnstVVVVLF26lEmTJrFgwQJqamrIz8/nhhtuOGKbEyZM4KyzzmpQds899zBt2jRWrVrFgw8+yPbt2wF45JFHuPrqq+nXr1+mpqSV0scTR6FLly7cd9999euXX345+fn5rFy5kjfeeINrrrmGgoICAO6991769OnDhg0bGDhwYMb25s6dS/fu3TnvvPOorKysL9+4cSMXXXQRJ5xwAoMGDWLz5s3s37+f+fPns3z58kSPUZqfzoxZsGPHDv785z/XBzD9h9frlteuXZvxtR999BE//vGPG1yG1iksLOTVV19l69atVFVVMXDgQCZOnMjDDz/M8ccfn8CRSEtSGL+mAwcOcOONN1JcXMyQIUMYM2YM8+bNY/Xq1Xz66adMmTIFM2Pfvn0ZX3/vvfcybtw4+vb98tdDp02bxowZM7jiiit49NFHWb58OV27diU/P5+xY8dy4YUX8txzz2VoVVqjxC5Tzawj8FvghLCf5919cpjWbS6pWYlXAje7+2dmdgKpiXTOJPWNkOvcvSq0dQ8wjtSkOz9090VJ9fur+Pzzz7n55pvp0KEDjz+emiHv4osv5v777+fqq6/mo48+4s4776Rr164Zw1ZeXs5rr73GW2+9lbH9/v378/LLLwOwb98+zj33XF599VXuuOMOrrvuOi677DIKCwsZOXIkPXr0SO5ApVkkeWbcD1zk7kOBM4AxZnYO8DPgUXc/FdhFKmSE512h/NFQDzM7jdT0cAXAGOAXZtYuwX43ibszbtw4duzYwfz58xtcNk6YMIGKigp27NjB1VdfzcGDByksLPxSG0uXLqWqqopTTjmF3r17M23aNObPn8+wYcO+VHfKlCl8//vfp1evXqxZs4aioiK6detG3759G7zPlNYrsTCGn3T8JKweHx5O6i6e50P5bODKsDw2rBO2j0z7jda57r7f3TcClWSYUq65jR8/nrfffpuXXnqJTp061ZfX1taydu1a3J3NmzdTUlLCxIkTyc3N/VIbJSUlbNiwgfLycsrLy7n99tu57LLLWLSo4Yl//fr1LF26lPHjxwOQn5/PkiVL2LFjBxUVFZxyyinJHqw0i0TfM4YvIpcDO4FSYAOwO0whDrAV6BOW+wBbAML2PaQuZevLM7wmfV8lZlZmZmXV1dVJHE69TZs28atf/Yry8nJ69+5NTk4OOTk5PP3009TW1vLd736XnJwczj77bM4991ymTp1a/9qf/OQnXHLJJQB07tyZ3r171z9ycnLo2LEjeXkNf0RhwoQJPPbYY7Rrl7og+OlPf8r06dMpKChg0qRJ9O6tCcLagkQ/2ggzDJ9hZt2BF4EhCe5rJjAToKioKNF5JPv3799gxPRQq1evbnTbpEmTGt2W/nFJutdff73B+tChQ1m/XncmtjXN8jmju+82s9eBc4HuZtY+nP36AttCtW1AP2CrmbUHupEayKkrr5P+mqP24kq9z6rznTNPbekuCAlepppZXjgjYmadgG8DbwOvk/rZDoBiYEFYXhjWCduXhJmSFwLXm9kJYSR2EPCHpPot0lKSPDN+A5gdRj6PA+a5+3+Y2Xpgrpk9ALwFPBnqPwk8ZWaVQA2pEVTcfZ2ZzSP1jZGDwIRw+SvSpiQWRndfTYaf5nD3d8kwGurutaS+tJyprQfRl5mljdMdOCKRUBhFIqEwikRCYRSJhMIoEgmFUSQSCqNIJBRGkUgojCKRUBhFIqEwikRCYRSJhMIoEgmFUSQSCqNIJBRGkUgojCKRUBhFIqEwikRCYRSJhMIoEgmFUSQSCqNIJBRGkUgojCKRUBhFIqEwikRCYRSJRJJTwvUzs9fNbL2ZrTOziaG8h5mVmllFeM4N5WZm082s0sxWm9mwtLaKQ/0KMytubJ8irVmSZ8aDwD+5+2nAOcAEMzsNuBtY7O6DgMVhHeASUnMvDgJKgBmQCi8wGRhOavaqyXUBFmlLEguju7/v7n8Myx+Tmii1DzAWmB2qzQauDMtjgTme8iapGY6/AYwGSt29xt13AaXAmKT6LdJSmuU9o5kNIDVX4++BXu7+fti0HegVlvsAW9JetjWUNVZ+6D5KzKzMzMqqq6uz2n+R5pB4GM0sB5gP3OnuH6VvC9OEezb24+4z3b3I3Yvy8vKy0aRIs0o0jGZ2PKkgPu3uL4TiHeHyk/C8M5RvA/qlvbxvKGusXKRNSXI01YAngbfd/ZG0TQuBuhHRYmBBWvktYVT1HGBPuJxdBIwys9wwcDMqlIm0Ke0TbPt84GZgjZmVh7JJwEPAPDMbB2wCrg3bXgYuBSqBfcCtAO5eY2ZTgRWh3hR3r0mw3yItIrEwuvt/AdbI5pEZ6jswoZG2ZgGzstc7kfjoDhyRSCiMIpFQGEUioTCKREJhFImEwigSCYVRJBIKo0gkFEaRSCiMIpFQGEUioTCKREJhFImEwigSCYVRJBIKo0gkFEaRSCiMIpFQGEUioTCKREJhFImEwigSCYVRJBIKo0gkFEaRSCiMIpFQGEUikeQsVLPMbKeZrU0r62FmpWZWEZ5zQ7mZ2XQzqzSz1WY2LO01xaF+hZkVZ9qXSFuQ5JnxN3x5uu+7gcXuPghYHNYBLgEGhUcJMANS4QUmA8OBs4HJdQEWaWsSC6O7/xY4dOq2scDssDwbuDKtfI6nvAl0DxOpjgZK3b3G3XcBpXw54CJtQnO/Z+wVJkAF2A70Cst9gC1p9baGssbKRdqcFhvACfMxerbaM7MSMyszs7Lq6upsNSvSbJo7jDvC5SfheWco3wb0S6vXN5Q1Vv4l7j7T3YvcvSgvLy/rHRdJWnOHcSFQNyJaDCxIK78ljKqeA+wJl7OLgFFmlhsGbkaFMpE2J7FpxM3sGeBvgZ5mtpXUqOhDwDwzGwdsAq4N1V8GLgUqgX3ArQDuXmNmU4EVod4Udz90UEikTUgsjO5+QyObRmao68CERtqZBczKYtdEoqQ7cEQioTCKREJhFImEwigSCYVRJBIKo0gkFEaRSCiMIpFQGEUioTCKREJhFImEwigSCYVRJBIKo0gkFEaRSCiMIpFQGEUioTCKREJhFImEwigSCYVRJBIKo0gkFEaRSCiMIpFQGEUioTCKREJhFImEwigSiVYTRjMbY2Z/MrNKM7u7pfsjkm2tIoxm1g74P8AlwGnADWZ2Wsv2SiS7WkUYgbOBSnd/190/A+YCY1u4TyJZldj8jFnWB9iStr4VGJ5ewcxKgJKw+omZ/amZ+vZ19AQ+aOlOtCEt/u9p04qPXAn6ZypsLWE8InefCcxs6X58FWZW5u5FLd2PtqK1/3u2lsvUbUC/tPW+oUykzWgtYVwBDDKzfDPrAFwPLGzhPolkVau4THX3g2b2A2AR0A6Y5e7rWrhb2dCqLqtbgVb972nu3tJ9EBFaz2WqSJunMIpEQmEUiYTCKBIJhVEkEq3io422wMxygAuBIUBuKN4FvAMsc/dPWqpvbZGZdQROdvfNLd2XplIYE2ZmBtwP/CPQGdhHKoSQCmVnYJ+Z/Stwn+uzpmy5DJhH6nPpVkFhTN59pIJ4PzDX3dNveMfM+pK6o2gy4KG+HIP0oX/CzGwbcH+4kf1w9UqAye7ep3l61jqZ2ZImVs0DTnN3nRmlXndgQxPqbQh15fAuAP4ErD9CvY7N0Jes0pkxYWa2GDgIXOXuexup0wV4ETjO3S9uzv61Nma2CnjH3a87Qr2/B57VmVHS/QB4DdhsZotIjZ7uDtu6kRpdHQ3sB0a2SA9blzeBMU2o54Al3Jes0pmxGZhZd2A8qT+iTB9tvAL80t13Z25B6pjZQKDA3Q/7FToz60Tqo41NzdOzr09hFImE7sARiYTCKBIJhbEVM7NPDln/npk9fpRt/bWZvWxmFWb2RzObZ2a9vm7dJu77N2H085im0VSpu4/z/wH/6O4vhbK/JfXB+Y60eu1J/c0cse5h9tXe3Q9m+RDaBIWxjTKzvwP+N9AB+BC40d13mNmFwGOhmpP6EP0a4I26cAG4+9LQzveAq4AcUvd5zj5M3QHAU0CXsOkH7v67ENappEaPh5jZYODnwLdJ/R7uZ9k89tZKYWzdOplZedp6D7741bz/As5xdzezfwD+Gfgn4C5ggrsvD98kqQUKgZWH2c8w4HR3rzGzRw5TdyfwbXevNbNBwDNAUVobhe6+0cyuAgaTmqqhF6m7aWZ9pSNvgxTG1u1Tdz+jbiWcxer++PsCz5rZN0idHTeG8uXAI2b2NPCCu29NfbHksErdvaYJ/TkeeNzMzgD+Avx12rY/uHtdHy4AnnH3vwDvfYX7Tds0DeC0XT8HHnf3bwK3Ee7VdPeHgH8AOgHLzWwIsA448zBtpd/Gd7i6PyL1vnEoqf8UOjTShmSgMLZd3fjiV9frJ4Aws4Huvsbdf0bqx6GHAP8OnGdml6XVu8DMCjO0e7i63YD33f1z4GYa/y7hb4HrzKxdOHOPOOqjbEMUxrbrPuA5M1tJw8lg7jSztWa2GjgAvOLunwKXA3eEjyvWA/8TqD600SPU/QVQHG7mHkLjZ8MXgQpS7xXnAG987aNtA3Q7nEgkdGYUiYTCKBIJhVEkEgqjSCQURpFIKIwikVAYRSLx/wGPEELPFEEWAQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 216x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "labeled_barplot(data, \"HasCrCard\", perc=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "C-LAsG0shgb1"
   },
   "source": [
    "70% of customers have credit cards"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "7rCroY44ElQs",
    "outputId": "2578db02-cd1f-41ad-f192-6e3a29d4505b"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 216x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "labeled_barplot(data, \"IsActiveMember\", perc=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Qt0RqjSVhlr8"
   },
   "source": [
    "51.5% of customers are active"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "Zt_7XKdpF9wO",
    "outputId": "9d7847d4-a99e-4986-8ee0-a3be4b58a303"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    7963\n",
      "1    2037\n",
      "Name: Exited, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(data.Exited.value_counts())\n",
    "labels = 'Customer left the bank', 'Customer did not leave the bank'\n",
    "#sizes = [ds.is_promoted[ds['is_promoted']==1].count(), ds.is_promoted[ds['is_promoted']==0].count()]\n",
    "sizes = [data.Exited[data['Exited']==1].count(),data.Exited[data['Exited']==0].count()]\n",
    "explode = (0, 0.1)\n",
    "fig1, ax1 = plt.subplots(figsize=(10, 8))\n",
    "ax1.pie(sizes, explode=explode, labels=labels, autopct='%1.1f%%',\n",
    "        shadow=True, startangle=90)\n",
    "ax1.axis('equal')\n",
    "plt.title(\"Customers\", size = 20)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rWdxbtk-h2X_"
   },
   "source": [
    "20% of customers left the bank within 6 months."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0iilSSEYIA7B"
   },
   "source": [
    "### Bivariate Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "ZCrGwKFUII4p"
   },
   "outputs": [],
   "source": [
    "### Function to plot distributions\n",
    "\n",
    "\n",
    "def distribution_plot_wrt_target(data, predictor, target):\n",
    "\n",
    "    fig, axs = plt.subplots(2, 2, figsize=(12, 10))\n",
    "\n",
    "    target_uniq = data[target].unique()\n",
    "\n",
    "    axs[0, 0].set_title(\"Distribution of target for target=\" + str(target_uniq[0]))\n",
    "    sns.histplot(\n",
    "        data=data[data[target] == target_uniq[0]],\n",
    "        x=predictor,\n",
    "        kde=True,\n",
    "        ax=axs[0, 0],\n",
    "        color=\"teal\",\n",
    "    )\n",
    "\n",
    "    axs[0, 1].set_title(\"Distribution of target for target=\" + str(target_uniq[1]))\n",
    "    sns.histplot(\n",
    "        data=data[data[target] == target_uniq[1]],\n",
    "        x=predictor,\n",
    "        kde=True,\n",
    "        ax=axs[0, 1],\n",
    "        color=\"orange\",\n",
    "    )\n",
    "\n",
    "    axs[1, 0].set_title(\"Boxplot w.r.t target\")\n",
    "    sns.boxplot(data=data, x=target, y=predictor, ax=axs[1, 0], palette=\"gist_rainbow\")\n",
    "\n",
    "    axs[1, 1].set_title(\"Boxplot (without outliers) w.r.t target\")\n",
    "    sns.boxplot(\n",
    "        data=data,\n",
    "        x=target,\n",
    "        y=predictor,\n",
    "        ax=axs[1, 1],\n",
    "        showfliers=False,\n",
    "        palette=\"gist_rainbow\",\n",
    "    )\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "v7cB0G8bBoxc",
    "outputId": "29264975-c63b-4056-92ed-b7a07261757b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 10000 entries, 0 to 9999\n",
      "Data columns (total 11 columns):\n",
      " #   Column           Non-Null Count  Dtype  \n",
      "---  ------           --------------  -----  \n",
      " 0   CreditScore      10000 non-null  int64  \n",
      " 1   Geography        10000 non-null  object \n",
      " 2   Gender           10000 non-null  object \n",
      " 3   Age              10000 non-null  int64  \n",
      " 4   Tenure           10000 non-null  int64  \n",
      " 5   Balance          10000 non-null  float64\n",
      " 6   NumOfProducts    10000 non-null  int64  \n",
      " 7   HasCrCard        10000 non-null  int64  \n",
      " 8   IsActiveMember   10000 non-null  int64  \n",
      " 9   EstimatedSalary  10000 non-null  float64\n",
      " 10  Exited           10000 non-null  int64  \n",
      "dtypes: float64(2), int64(7), object(2)\n",
      "memory usage: 859.5+ KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "G5q1mjh_IPd7",
    "outputId": "4c28b7a4-2d9e-4776-800f-43b1edcbb3bf"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "distribution_plot_wrt_target(data, 'CreditScore', 'Exited')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ir4HNGhTiDiD"
   },
   "source": [
    "From this chart, hard to identify any relation between credit score and leaving the bank.\n",
    "Both customer who chose to stay and one who left the bank share the same distribution of credit score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "U44mNuouBTTe",
    "outputId": "d283ef31-d9e6-4df8-d251-102d8671c15f"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "distribution_plot_wrt_target(data, 'Age', 'Exited')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "A_ZzYibSiar_"
   },
   "source": [
    "Customers who left the bank tend to be higher is age than who stayed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "m59_ROXDBTfS",
    "outputId": "af9b54ba-282b-48ae-f94e-d7c081799356"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "distribution_plot_wrt_target(data, 'Tenure', 'Exited')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "aYszItABiqcT"
   },
   "source": [
    "Customers who have have left the bank and thos who stayed have a shared range of tenure. It is hard to draw an observation from this trend."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "AdXUBWM8BTiG",
    "outputId": "36d51431-c602-4399-b714-e392c749035e"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "distribution_plot_wrt_target(data, 'Balance', 'Exited')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "h_EChn0SjLhi"
   },
   "source": [
    "Customers who stayed and thos who left share the same balance distribution, those with lower balance have stayed as customers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "Qcp9uj-lBhGW",
    "outputId": "6ef6ecee-ca1d-4d69-e0ea-d167b5c69150"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "distribution_plot_wrt_target(data, 'EstimatedSalary', 'Exited')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_C6qiRYejgR8"
   },
   "source": [
    "Same also for Estimated salary, both who left and stayed have the same range of estimated salaries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "iZYIn4x6BTlI",
    "outputId": "625a630b-178b-4bbd-88a2-8e9d88778aa6"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 7))\n",
    "sns.heatmap(data.corr(), annot=True, vmin=-1, vmax=1, fmt=\".2f\", cmap=\"Spectral\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RrtDKx3Cjqax"
   },
   "source": [
    "Similar to the previous observations in previous bivariate analysis, there are no varaibles/features in high coorelation."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WASTeKpBkQZC"
   },
   "source": [
    "### EDA Observations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "FU3YaWr5kY_f"
   },
   "source": [
    "- `CreditScore` is having a minimum of 350 and a max of 850 with 75% of customers having score ranging from 350 upto 718. Credit score is normally distributed with mean very close to median value at around 650.\n",
    "Some outliers lie on both ends.\n",
    "- 3 unique locations are available in `Geography` with France as top location. 50% of customers are from France, other 25% are from Germany and Spain.\n",
    "- For `Gender`; Males are dominating the list. Around 55% of customers are Males.\n",
    "- Client `Age` is ranging between 18 and 92, 75% of the customers are between 18 and 44. Age is almost uniformly distributed with mean close to median around 35+\n",
    "- `Tenure` is ranging between 0 and 10, 75% of customers are ranging between 0 and 7. Mean Teunure of the customers is for 5 years, In terms of count most of customer's tenure is between 1 and 9 years.\n",
    "- Clients `Balance` is ranging between 0 and 250K with std of 62K. Around 3500 customers have a 0 balance, rest of custumers have a normally distrubuted `Balance` between 50K and 200K.\n",
    "- Customers are having a maximum of 4 products per client. More than half of the customers are using 1 product, vast majority are using 2 products. 50% of customer are tied to 1 product, 45% are tied to 2 products\n",
    "- `EstimatedSalary` is ranging between 11 and 200K with std 57K. Mean salary of clients is around 100K, bank is having a fair share from each salary category and have no specific range overcoming the other.\n",
    "- 70% of customers have credit cards\n",
    "- Around 51.5% of customers are active\n",
    "- Around 20% of customers left the bank within 6 months according to given dataset.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "YZ8CTDnVU0vI"
   },
   "source": [
    "# C. **Data Pre-processing**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "4AZ5rLGAgEfJ"
   },
   "source": [
    "### Data rescaling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pbjy7DJFB7tr"
   },
   "source": [
    "Rescale the Data, The numerical columns present in this dataset are having different units (scores and currencies), so scaling would help them all be in the same range"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QuxdnZ0BDR2y"
   },
   "source": [
    "Since there is no mising data and imputation did not take place so rescaling and one hot encoding can be applied on the whole dataset at once "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "id": "l1iUSVUVCis1"
   },
   "outputs": [],
   "source": [
    "## Scaling the data\n",
    "sc=StandardScaler()\n",
    "temp = sc.fit(data[[\"CreditScore\",\"Tenure\",\"Balance\",\"NumOfProducts\",\"Age\",\"EstimatedSalary\"]])\n",
    "data[[\"CreditScore\",\"Tenure\",\"Balance\",\"NumOfProducts\",\"Age\",\"EstimatedSalary\"]] = temp.transform(data[[\"CreditScore\",\"Tenure\",\"Balance\",\"NumOfProducts\",\"Age\",\"EstimatedSalary\"]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9vfM_4j_Fx3w"
   },
   "source": [
    "### Data Onehot encoding"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cCMx3gpogSue"
   },
   "source": [
    "Applying One hot encoding for Gender and Geography columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "id": "PG3no7icgLLC"
   },
   "outputs": [],
   "source": [
    "data = pd.get_dummies(data,['Geography','Gender'],drop_first= True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nYkmtAKEKV5d"
   },
   "source": [
    "### Splitting the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "id": "luEm0LBKKli1"
   },
   "outputs": [],
   "source": [
    "## Separating Independent and Dependent Columns\n",
    "X = data.drop(['Exited'],axis=1)\n",
    "Y = data[['Exited']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "id": "TdvoqiIFKa3G"
   },
   "outputs": [],
   "source": [
    "# Splitting the dataset into the Training and Testing set.\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,Y, test_size = 0.2, random_state = 1,stratify = Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pk_0uwUXMOvW"
   },
   "source": [
    "# D. **Model building**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gNaQDWcqMv_0"
   },
   "source": [
    "#### **A model can make wrong predictions in the following ways:**\n",
    "* Predicting a customer is churning in 6 months, when he/she is not leaving the bank.\n",
    "* Predicting a customer is not churning in 6 months, when he/she leaves the bank within the mentioned period\n",
    "\n",
    "#### **Which case is more important?**\n",
    "\n",
    "Predicting a customer is not leaving while he/she actually leaves in 6 month, that is a customer base loss.\n",
    "\n",
    "#### **How to reduce this loss i.e need to reduce False Negatives**\n",
    "\n",
    "The bank would focus on the Recall Score evaluation metric to be maximized.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2COuWweGO_EY"
   },
   "source": [
    "### Model 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "zW9gcIC7UeRV",
    "outputId": "c736deba-cb54-46c5-c0d6-5c2bbe45cc96"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8000, 11)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking the final shape of X_train\n",
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "id": "UtX0DdrMV8_i"
   },
   "outputs": [],
   "source": [
    "#clearing the memory\n",
    "backend.clear_session()\n",
    "#Fixing the seed for random number generators so that we can ensure we receive the same output everytime\n",
    "np.random.seed(1)\n",
    "random.seed(1)\n",
    "tf.random.set_seed(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "id": "aq6L8q_tPCeJ"
   },
   "outputs": [],
   "source": [
    "# Initializing the ANN\n",
    "model = Sequential()\n",
    "# The amount of nodes (dimensions) in hidden layer should be the average of input and output layers, in this case 64.\n",
    "# This adds the input layer (by specifying input dimension) AND the first hidden layer (units)\n",
    "model.add(Dense(activation = 'relu', input_dim = 11, units=64))\n",
    "#Add 2nd hidden layer\n",
    "model.add(Dense(32, activation='relu'))\n",
    "# Adding the output layer \n",
    "# we have an output of 1 node, which is the the desired dimensions of our output (stay with the bank or not)\n",
    "# We use the sigmoid because we want probability outcomes\n",
    "model.add(Dense(1, activation = 'sigmoid')) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "id": "EeNvLaNcVH26"
   },
   "outputs": [],
   "source": [
    "# Create optimizer with default learning rate\n",
    "# Compile the model\n",
    "model.compile(optimizer='SGD', loss='binary_crossentropy', metrics=[tf.keras.metrics.Recall()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "1gfz72NMWMKA",
    "outputId": "21e0f905-823b-4eea-9e2f-8586cc31167d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 64)                768       \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 32)                2080      \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 1)                 33        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 2,881\n",
      "Trainable params: 2,881\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "vOao7HtIWOZE",
    "outputId": "8fe6234d-25e3-4096-9d43-94a7ca595712"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "200/200 [==============================] - 2s 7ms/step - loss: 0.5387 - recall: 0.0950 - val_loss: 0.4827 - val_recall: 0.0000e+00\n",
      "Epoch 2/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.4679 - recall: 0.0000e+00 - val_loss: 0.4653 - val_recall: 0.0000e+00\n",
      "Epoch 3/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.4533 - recall: 0.0015 - val_loss: 0.4522 - val_recall: 0.0000e+00\n",
      "Epoch 4/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.4419 - recall: 0.0201 - val_loss: 0.4419 - val_recall: 0.0119\n",
      "Epoch 5/50\n",
      "200/200 [==============================] - 1s 4ms/step - loss: 0.4325 - recall: 0.0479 - val_loss: 0.4323 - val_recall: 0.0388\n",
      "Epoch 6/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.4238 - recall: 0.0896 - val_loss: 0.4237 - val_recall: 0.0836\n",
      "Epoch 7/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.4158 - recall: 0.1459 - val_loss: 0.4162 - val_recall: 0.1134\n",
      "Epoch 8/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.4085 - recall: 0.1815 - val_loss: 0.4080 - val_recall: 0.1672\n",
      "Epoch 9/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.4012 - recall: 0.2193 - val_loss: 0.4006 - val_recall: 0.2149\n",
      "Epoch 10/50\n",
      "200/200 [==============================] - 1s 4ms/step - loss: 0.3943 - recall: 0.2571 - val_loss: 0.3941 - val_recall: 0.2716\n",
      "Epoch 11/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.3878 - recall: 0.2873 - val_loss: 0.3880 - val_recall: 0.2896\n",
      "Epoch 12/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.3819 - recall: 0.3189 - val_loss: 0.3832 - val_recall: 0.2836\n",
      "Epoch 13/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.3767 - recall: 0.3320 - val_loss: 0.3775 - val_recall: 0.3403\n",
      "Epoch 14/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.3720 - recall: 0.3483 - val_loss: 0.3734 - val_recall: 0.3851\n",
      "Epoch 15/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.3683 - recall: 0.3691 - val_loss: 0.3696 - val_recall: 0.3791\n",
      "Epoch 16/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3648 - recall: 0.3753 - val_loss: 0.3664 - val_recall: 0.3970\n",
      "Epoch 17/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3619 - recall: 0.3900 - val_loss: 0.3641 - val_recall: 0.4060\n",
      "Epoch 18/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3594 - recall: 0.4046 - val_loss: 0.3619 - val_recall: 0.4030\n",
      "Epoch 19/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3573 - recall: 0.4108 - val_loss: 0.3604 - val_recall: 0.3881\n",
      "Epoch 20/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3557 - recall: 0.4093 - val_loss: 0.3582 - val_recall: 0.4388\n",
      "Epoch 21/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3539 - recall: 0.4263 - val_loss: 0.3577 - val_recall: 0.4269\n",
      "Epoch 22/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3526 - recall: 0.4278 - val_loss: 0.3571 - val_recall: 0.4119\n",
      "Epoch 23/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3514 - recall: 0.4286 - val_loss: 0.3553 - val_recall: 0.4537\n",
      "Epoch 24/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3501 - recall: 0.4378 - val_loss: 0.3546 - val_recall: 0.4388\n",
      "Epoch 25/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3491 - recall: 0.4440 - val_loss: 0.3544 - val_recall: 0.4478\n",
      "Epoch 26/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3482 - recall: 0.4347 - val_loss: 0.3533 - val_recall: 0.4597\n",
      "Epoch 27/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3474 - recall: 0.4409 - val_loss: 0.3527 - val_recall: 0.4746\n",
      "Epoch 28/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3464 - recall: 0.4402 - val_loss: 0.3517 - val_recall: 0.4776\n",
      "Epoch 29/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3458 - recall: 0.4525 - val_loss: 0.3513 - val_recall: 0.4418\n",
      "Epoch 30/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3449 - recall: 0.4533 - val_loss: 0.3511 - val_recall: 0.4478\n",
      "Epoch 31/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3442 - recall: 0.4463 - val_loss: 0.3507 - val_recall: 0.4836\n",
      "Epoch 32/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3434 - recall: 0.4602 - val_loss: 0.3501 - val_recall: 0.4418\n",
      "Epoch 33/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3429 - recall: 0.4541 - val_loss: 0.3499 - val_recall: 0.4507\n",
      "Epoch 34/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3422 - recall: 0.4564 - val_loss: 0.3490 - val_recall: 0.4687\n",
      "Epoch 35/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3414 - recall: 0.4579 - val_loss: 0.3493 - val_recall: 0.4806\n",
      "Epoch 36/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3411 - recall: 0.4602 - val_loss: 0.3481 - val_recall: 0.4716\n",
      "Epoch 37/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3404 - recall: 0.4564 - val_loss: 0.3481 - val_recall: 0.4806\n",
      "Epoch 38/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3396 - recall: 0.4571 - val_loss: 0.3479 - val_recall: 0.4866\n",
      "Epoch 39/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3393 - recall: 0.4633 - val_loss: 0.3471 - val_recall: 0.4507\n",
      "Epoch 40/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3386 - recall: 0.4587 - val_loss: 0.3467 - val_recall: 0.4627\n",
      "Epoch 41/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3380 - recall: 0.4641 - val_loss: 0.3467 - val_recall: 0.5015\n",
      "Epoch 42/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3375 - recall: 0.4718 - val_loss: 0.3460 - val_recall: 0.4866\n",
      "Epoch 43/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3369 - recall: 0.4710 - val_loss: 0.3465 - val_recall: 0.4507\n",
      "Epoch 44/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3364 - recall: 0.4633 - val_loss: 0.3458 - val_recall: 0.4955\n",
      "Epoch 45/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3359 - recall: 0.4780 - val_loss: 0.3457 - val_recall: 0.5015\n",
      "Epoch 46/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3355 - recall: 0.4741 - val_loss: 0.3455 - val_recall: 0.4507\n",
      "Epoch 47/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3350 - recall: 0.4672 - val_loss: 0.3450 - val_recall: 0.5015\n",
      "Epoch 48/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3347 - recall: 0.4780 - val_loss: 0.3452 - val_recall: 0.4866\n",
      "Epoch 49/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3339 - recall: 0.4703 - val_loss: 0.3451 - val_recall: 0.4507\n",
      "Epoch 50/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3337 - recall: 0.4710 - val_loss: 0.3444 - val_recall: 0.4746\n"
     ]
    }
   ],
   "source": [
    "history=model.fit(X_train, y_train, validation_split=0.2, epochs=50, batch_size=32, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "od5o8OZXWgUl",
    "outputId": "ae25f247-91db-40c4-b3b7-e3ca32cae6ff"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "63/63 [==============================] - 0s 2ms/step - loss: 0.3595 - recall: 0.4472\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Capturing learning history per epoch\n",
    "hist  = pd.DataFrame(history.history)\n",
    "hist['epoch'] = history.epoch\n",
    "\n",
    "# Plotting accuracy at different epochs\n",
    "plt.figure(figsize=(8,6))\n",
    "plt.plot(hist['loss'])\n",
    "plt.plot(hist['val_loss'])\n",
    "plt.legend((\"train\" , \"valid\") , loc =0)\n",
    "\n",
    "#Printing results\n",
    "results = model.evaluate(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mghsJOVPdYgp"
   },
   "source": [
    "We have a very smooth loss curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "HsmjYo-jXVhO",
    "outputId": "5a522f27-7f4b-45bb-c86a-7f41e47175d2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False],\n",
       "       [False],\n",
       "       [False],\n",
       "       ...,\n",
       "       [False],\n",
       "       [False],\n",
       "       [False]])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred=model.predict(X_test)\n",
    "y_pred = (y_pred > 0.5)\n",
    "y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "id": "KcmqE278XkPC"
   },
   "outputs": [],
   "source": [
    "def make_confusion_matrix(cf,\n",
    "                          group_names=None,\n",
    "                          categories='auto',\n",
    "                          count=True,\n",
    "                          percent=True,\n",
    "                          cbar=True,\n",
    "                          xyticks=True,\n",
    "                          xyplotlabels=True,\n",
    "                          sum_stats=True,\n",
    "                          figsize=None,\n",
    "                          cmap='Blues',\n",
    "                          title=None):\n",
    "    '''\n",
    "    This function will make a pretty plot of an sklearn Confusion Matrix cm using a Seaborn heatmap visualization.\n",
    "    Arguments\n",
    "    '''\n",
    "\n",
    "\n",
    "    # CODE TO GENERATE TEXT INSIDE EACH SQUARE\n",
    "    blanks = ['' for i in range(cf.size)]\n",
    "\n",
    "    if group_names and len(group_names)==cf.size:\n",
    "        group_labels = [\"{}\\n\".format(value) for value in group_names]\n",
    "    else:\n",
    "        group_labels = blanks\n",
    "\n",
    "    if count:\n",
    "        group_counts = [\"{0:0.0f}\\n\".format(value) for value in cf.flatten()]\n",
    "    else:\n",
    "        group_counts = blanks\n",
    "\n",
    "    if percent:\n",
    "        group_percentages = [\"{0:.2%}\".format(value) for value in cf.flatten()/np.sum(cf)]\n",
    "    else:\n",
    "        group_percentages = blanks\n",
    "\n",
    "    box_labels = [f\"{v1}{v2}{v3}\".strip() for v1, v2, v3 in zip(group_labels,group_counts,group_percentages)]\n",
    "    box_labels = np.asarray(box_labels).reshape(cf.shape[0],cf.shape[1])\n",
    "\n",
    "\n",
    "    # CODE TO GENERATE SUMMARY STATISTICS & TEXT FOR SUMMARY STATS\n",
    "    if sum_stats:\n",
    "        #Accuracy is sum of diagonal divided by total observations\n",
    "        accuracy  = np.trace(cf) / float(np.sum(cf))\n",
    "\n",
    "\n",
    "\n",
    "    # SET FIGURE PARAMETERS ACCORDING TO OTHER ARGUMENTS\n",
    "    if figsize==None:\n",
    "        #Get default figure size if not set\n",
    "        figsize = plt.rcParams.get('figure.figsize')\n",
    "\n",
    "    if xyticks==False:\n",
    "        #Do not show categories if xyticks is False\n",
    "        categories=False\n",
    "\n",
    "\n",
    "    # MAKE THE HEATMAP VISUALIZATION\n",
    "    plt.figure(figsize=figsize)\n",
    "    sns.heatmap(cf,annot=box_labels,fmt=\"\",cmap=cmap,cbar=cbar,xticklabels=categories,yticklabels=categories)\n",
    "\n",
    "    \n",
    "    if title:\n",
    "        plt.title(title)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "Wr40Zz5kXmtJ",
    "outputId": "e0b37e1e-fd23-4268-a74b-91e82b6edc7a"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Calculating the confusion matrix \n",
    "from sklearn.metrics import confusion_matrix\n",
    "cm=confusion_matrix(y_test, y_pred)\n",
    "labels = ['True Positive','False Negative','False Positive','True Negative']\n",
    "categories = [ 'Not Churning','Churning']\n",
    "make_confusion_matrix(cm, \n",
    "                      group_names=labels,\n",
    "                      categories=categories, \n",
    "                      cmap='Blues')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "T9pDMOkGXpSC",
    "outputId": "1aa97fed-dc47-40aa-837e-5886135e0177"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.87      0.96      0.91      1593\n",
      "           1       0.73      0.45      0.55       407\n",
      "\n",
      "    accuracy                           0.85      2000\n",
      "   macro avg       0.80      0.70      0.73      2000\n",
      "weighted avg       0.84      0.85      0.84      2000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Accuracy as per the classification report \n",
    "from sklearn import metrics\n",
    "cr=metrics.classification_report(y_test,y_pred)\n",
    "print(cr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "C7zm7p7YdC3y"
   },
   "source": [
    "We have got a good result for the recall , let us see if we can improve it by finding the optimal threshold on ROC-AUC curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "AMgQdyJJe_rD",
    "outputId": "65055a03-dabc-4d8a-8efb-d9559beee5e2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold=0.180811, G-Mean=0.766\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# predict probabilities\n",
    "yhat = model.predict(X_test)\n",
    "# keep probabilities for the positive outcome only\n",
    "yhat = yhat[:, 0]\n",
    "# calculate roc curves\n",
    "fpr, tpr, thresholds = roc_curve(y_test, yhat)\n",
    "# calculate the g-mean for each threshold\n",
    "gmeans = np.sqrt(tpr * (1-fpr))\n",
    "# locate the index of the largest g-mean\n",
    "ix = np.argmax(gmeans)\n",
    "print('Best Threshold=%f, G-Mean=%.3f' % (thresholds[ix], gmeans[ix]))\n",
    "# plot the roc curve for the model\n",
    "pyplot.plot([0,1], [0,1], linestyle='--', label='No Skill')\n",
    "pyplot.plot(fpr, tpr, marker='.')\n",
    "pyplot.scatter(fpr[ix], tpr[ix], marker='o', color='black', label='Best')\n",
    "# axis labels\n",
    "pyplot.xlabel('False Positive Rate')\n",
    "pyplot.ylabel('True Positive Rate')\n",
    "pyplot.legend()\n",
    "# show the plot\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RWNizflK5eG0"
   },
   "source": [
    "The Geometric Mean or G-Mean is a metric for imbalanced classification that, if optimized, will seek a balance between the sensitivity and the specificity.\n",
    "\n",
    "G-Mean = sqrt(Sensitivity * Specificity)\n",
    "\n",
    "That is why the performance of recall on the '0' class reduced as the ROC-AUC optimal point specified was a balance between True positve and False positive\n",
    "\n",
    "Recall = Sensitivity = True Positive Rate\n",
    "\n",
    "Specificity = 1 – False Positive Rate\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "RsZN44RLkh7o",
    "outputId": "0859bfa1-62c3-4cbf-9a21-c2818c510ac5"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False],\n",
       "       [False],\n",
       "       [False],\n",
       "       ...,\n",
       "       [ True],\n",
       "       [False],\n",
       "       [ True]])"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Predicting the results using best as a threshold\n",
    "y_pred_e1=model.predict(X_test)\n",
    "y_pred_e1 = (y_pred_e1 > thresholds[ix])\n",
    "y_pred_e1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "hwNYEdJfkmXE",
    "outputId": "5cd7ddae-1f43-4a8d-eb11-0716afb896ed"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Calculating the confusion matrix \n",
    "from sklearn.metrics import confusion_matrix\n",
    "cm1=confusion_matrix(y_test, y_pred_e1)\n",
    "labels = ['True Positive','False Negative','False Positive','True Negative']\n",
    "categories = [ 'Not Churning','Churning']\n",
    "make_confusion_matrix(cm1, \n",
    "                      group_names=labels,\n",
    "                      categories=categories, \n",
    "                      cmap='Blues')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_T5X7qnXk5rV",
    "outputId": "e8d62180-df39-492f-f14f-abc5aa2efccd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.93      0.75      0.83      1593\n",
      "           1       0.44      0.78      0.56       407\n",
      "\n",
      "    accuracy                           0.75      2000\n",
      "   macro avg       0.69      0.76      0.70      2000\n",
      "weighted avg       0.83      0.75      0.78      2000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Accuracy as per the classification report \n",
    "from sklearn import metrics\n",
    "cr=metrics.classification_report(y_test,y_pred_e1)\n",
    "print(cr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vEUVHvpgnoxL"
   },
   "source": [
    "The model delivered good result and smooth loss, let's see if increasing the number of layers can improve the performance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "l9-7WCEWdvDE"
   },
   "source": [
    "### Model 2\n",
    "(Inreasing number of hidden layers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "id": "17f9c66Idwhn"
   },
   "outputs": [],
   "source": [
    "#clearing the memory\n",
    "backend.clear_session()\n",
    "#Fixing the seed for random number generators so that we can ensure we receive the same output everytime\n",
    "np.random.seed(1)\n",
    "random.seed(1)\n",
    "tf.random.set_seed(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "id": "gyM_HhBwd3_-"
   },
   "outputs": [],
   "source": [
    "# Initializing the ANN\n",
    "model2 = Sequential()\n",
    "model2.add(Dense(128,activation='relu',input_dim = 11))\n",
    "model2.add(Dense(64,activation='relu'))\n",
    "model2.add(Dense(64,activation='relu'))\n",
    "model2.add(Dense(32,activation='relu'))\n",
    "model2.add(Dense(16,activation='relu'))\n",
    "model2.add(Dense(1, activation = 'sigmoid')) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "id": "g34R_-Bjx7Jh"
   },
   "outputs": [],
   "source": [
    "# Create optimizer with default learning rate\n",
    "# Compile the model\n",
    "model2.compile(optimizer='SGD', loss='binary_crossentropy', metrics=[tf.keras.metrics.Recall()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "FCuJFyPAym8W",
    "outputId": "788eb7d7-cf61-47eb-9f12-bc4dab71518b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 128)               1536      \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 64)                8256      \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 64)                4160      \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 32)                2080      \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 16)                528       \n",
      "                                                                 \n",
      " dense_5 (Dense)             (None, 1)                 17        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 16,577\n",
      "Trainable params: 16,577\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model2.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "S4jAxvGYyuXY",
    "outputId": "c9030378-873b-4b62-9afc-ef9ceb997e3c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.5513 - recall: 0.0633 - val_loss: 0.4980 - val_recall: 0.0000e+00\n",
      "Epoch 2/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4769 - recall: 0.0000e+00 - val_loss: 0.4713 - val_recall: 0.0000e+00\n",
      "Epoch 3/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4493 - recall: 0.0000e+00 - val_loss: 0.4440 - val_recall: 0.0000e+00\n",
      "Epoch 4/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4261 - recall: 0.0131 - val_loss: 0.4277 - val_recall: 0.0179\n",
      "Epoch 5/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4126 - recall: 0.1097 - val_loss: 0.4157 - val_recall: 0.1731\n",
      "Epoch 6/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4016 - recall: 0.2178 - val_loss: 0.4064 - val_recall: 0.2597\n",
      "Epoch 7/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3916 - recall: 0.2896 - val_loss: 0.3976 - val_recall: 0.2448\n",
      "Epoch 8/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3824 - recall: 0.3313 - val_loss: 0.3866 - val_recall: 0.3701\n",
      "Epoch 9/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3725 - recall: 0.3776 - val_loss: 0.3773 - val_recall: 0.3851\n",
      "Epoch 10/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3649 - recall: 0.4046 - val_loss: 0.3711 - val_recall: 0.3940\n",
      "Epoch 11/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3579 - recall: 0.4201 - val_loss: 0.3661 - val_recall: 0.4955\n",
      "Epoch 12/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3526 - recall: 0.4440 - val_loss: 0.3618 - val_recall: 0.3821\n",
      "Epoch 13/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3485 - recall: 0.4471 - val_loss: 0.3585 - val_recall: 0.4149\n",
      "Epoch 14/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3445 - recall: 0.4641 - val_loss: 0.3569 - val_recall: 0.4209\n",
      "Epoch 15/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3429 - recall: 0.4517 - val_loss: 0.3536 - val_recall: 0.4179\n",
      "Epoch 16/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3400 - recall: 0.4610 - val_loss: 0.3524 - val_recall: 0.4597\n",
      "Epoch 17/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3382 - recall: 0.4602 - val_loss: 0.3564 - val_recall: 0.3791\n",
      "Epoch 18/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3366 - recall: 0.4680 - val_loss: 0.3500 - val_recall: 0.4716\n",
      "Epoch 19/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3346 - recall: 0.4749 - val_loss: 0.3510 - val_recall: 0.4209\n",
      "Epoch 20/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3336 - recall: 0.4795 - val_loss: 0.3486 - val_recall: 0.5015\n",
      "Epoch 21/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3315 - recall: 0.4880 - val_loss: 0.3548 - val_recall: 0.3881\n",
      "Epoch 22/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3312 - recall: 0.4803 - val_loss: 0.3496 - val_recall: 0.4388\n",
      "Epoch 23/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3298 - recall: 0.4950 - val_loss: 0.3518 - val_recall: 0.5373\n",
      "Epoch 24/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3277 - recall: 0.4958 - val_loss: 0.3508 - val_recall: 0.4179\n",
      "Epoch 25/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3265 - recall: 0.4865 - val_loss: 0.3524 - val_recall: 0.5194\n",
      "Epoch 26/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3263 - recall: 0.5004 - val_loss: 0.3476 - val_recall: 0.5015\n",
      "Epoch 27/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3252 - recall: 0.4958 - val_loss: 0.3515 - val_recall: 0.5493\n",
      "Epoch 28/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3241 - recall: 0.5027 - val_loss: 0.3489 - val_recall: 0.5373\n",
      "Epoch 29/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3226 - recall: 0.5019 - val_loss: 0.3463 - val_recall: 0.4657\n",
      "Epoch 30/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3219 - recall: 0.5089 - val_loss: 0.3471 - val_recall: 0.4657\n",
      "Epoch 31/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3215 - recall: 0.5004 - val_loss: 0.3503 - val_recall: 0.5522\n",
      "Epoch 32/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3199 - recall: 0.5104 - val_loss: 0.3489 - val_recall: 0.4418\n",
      "Epoch 33/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3184 - recall: 0.5019 - val_loss: 0.3496 - val_recall: 0.5313\n",
      "Epoch 34/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3178 - recall: 0.5158 - val_loss: 0.3470 - val_recall: 0.4567\n",
      "Epoch 35/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3158 - recall: 0.5135 - val_loss: 0.3543 - val_recall: 0.5015\n",
      "Epoch 36/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3165 - recall: 0.5120 - val_loss: 0.3507 - val_recall: 0.4388\n",
      "Epoch 37/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3161 - recall: 0.5166 - val_loss: 0.3490 - val_recall: 0.5284\n",
      "Epoch 38/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3128 - recall: 0.5073 - val_loss: 0.3504 - val_recall: 0.5224\n",
      "Epoch 39/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3136 - recall: 0.5104 - val_loss: 0.3473 - val_recall: 0.5194\n",
      "Epoch 40/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3124 - recall: 0.5228 - val_loss: 0.3469 - val_recall: 0.5075\n",
      "Epoch 41/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3107 - recall: 0.5135 - val_loss: 0.3546 - val_recall: 0.5910\n",
      "Epoch 42/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3104 - recall: 0.5290 - val_loss: 0.3481 - val_recall: 0.4746\n",
      "Epoch 43/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3083 - recall: 0.5305 - val_loss: 0.3702 - val_recall: 0.3701\n",
      "Epoch 44/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3080 - recall: 0.5220 - val_loss: 0.3505 - val_recall: 0.5433\n",
      "Epoch 45/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3064 - recall: 0.5375 - val_loss: 0.3494 - val_recall: 0.5015\n",
      "Epoch 46/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3073 - recall: 0.5205 - val_loss: 0.3503 - val_recall: 0.5075\n",
      "Epoch 47/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3038 - recall: 0.5359 - val_loss: 0.3549 - val_recall: 0.5851\n",
      "Epoch 48/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3053 - recall: 0.5359 - val_loss: 0.3532 - val_recall: 0.5612\n",
      "Epoch 49/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3035 - recall: 0.5436 - val_loss: 0.3576 - val_recall: 0.4269\n",
      "Epoch 50/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3026 - recall: 0.5382 - val_loss: 0.3511 - val_recall: 0.4537\n"
     ]
    }
   ],
   "source": [
    "history2=model2.fit(X_train, y_train, validation_split=0.2, epochs=50, batch_size=32, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "-ISreDFqy0ru",
    "outputId": "3cf6c12f-5c2c-400d-c969-c051ba75e2b4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "63/63 [==============================] - 0s 2ms/step - loss: 0.3552 - recall: 0.4177\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Capturing learning history per epoch\n",
    "hist2  = pd.DataFrame(history.history)\n",
    "hist2['epoch'] = history2.epoch\n",
    "\n",
    "# Plotting accuracy at different epochs\n",
    "plt.figure(figsize=(8,6))\n",
    "plt.plot(hist2['loss'])\n",
    "plt.plot(hist2['val_loss'])\n",
    "plt.legend((\"train\" , \"valid\") , loc =0)\n",
    "\n",
    "#Printing results\n",
    "results = model2.evaluate(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "aQFUuJaMPv-4"
   },
   "source": [
    "The loss curve is smoothe and no overfitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "NGrnxmSdzIYU",
    "outputId": "7e266bb4-5903-424a-b9e8-07925b749d88"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False],\n",
       "       [False],\n",
       "       [False],\n",
       "       ...,\n",
       "       [False],\n",
       "       [False],\n",
       "       [False]])"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred2=model2.predict(X_test)\n",
    "y_pred2 = (y_pred2 > 0.5)\n",
    "y_pred2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "D3_n7RiWzAAE",
    "outputId": "945a4954-b660-4b8d-99b9-aec12978b6da"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Calculating the confusion matrix \n",
    "cm2=confusion_matrix(y_test, y_pred2)\n",
    "labels = ['True Positive','False Negative','False Positive','True Negative']\n",
    "categories = [ 'Not Churning','Churning']\n",
    "make_confusion_matrix(cm2, \n",
    "                      group_names=labels,\n",
    "                      categories=categories, \n",
    "                      cmap='Blues')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "0FqCLB69zTDo",
    "outputId": "791029d0-e1ba-4d3c-c7a6-1c38a705cfae"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.87      0.97      0.92      1593\n",
      "           1       0.79      0.42      0.55       407\n",
      "\n",
      "    accuracy                           0.86      2000\n",
      "   macro avg       0.83      0.69      0.73      2000\n",
      "weighted avg       0.85      0.86      0.84      2000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Accuracy as per the classification report \n",
    "cr2=metrics.classification_report(y_test,y_pred2)\n",
    "print(cr2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZX2em_NRP7AP"
   },
   "source": [
    "The performance dropped slightly, let's check if ROC curve can offer any improvement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "h7RRywvD0rtD",
    "outputId": "336293e4-9705-47cb-d151-090ee1ccbea0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold=0.133170, G-Mean=0.772\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# predict probabilities\n",
    "yhat2 = model2.predict(X_test)\n",
    "# keep probabilities for the positive outcome only\n",
    "yhat2 = yhat2[:, 0]\n",
    "# calculate roc curves\n",
    "fpr2, tpr2, thresholds2 = roc_curve(y_test, yhat2)\n",
    "# calculate the g-mean for each threshold\n",
    "gmeans2 = np.sqrt(tpr2 * (1-fpr2))\n",
    "# locate the index of the largest g-mean\n",
    "ix2 = np.argmax(gmeans2)\n",
    "print('Best Threshold=%f, G-Mean=%.3f' % (thresholds2[ix2], gmeans2[ix2]))\n",
    "# plot the roc curve for the model\n",
    "pyplot.plot([0,1], [0,1], linestyle='--', label='No Skill')\n",
    "pyplot.plot(fpr2, tpr2, marker='.')\n",
    "pyplot.scatter(fpr2[ix2], tpr2[ix2], marker='o', color='black', label='Best')\n",
    "# axis labels\n",
    "pyplot.xlabel('False Positive Rate')\n",
    "pyplot.ylabel('True Positive Rate')\n",
    "pyplot.legend()\n",
    "# show the plot\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "HnygJSRX000y",
    "outputId": "1ee445e9-4284-48d5-e1f7-7cc4a9ef32f2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False],\n",
       "       [False],\n",
       "       [False],\n",
       "       ...,\n",
       "       [ True],\n",
       "       [False],\n",
       "       [False]])"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Predicting the results using best as a threshold\n",
    "y_pred_e2=model2.predict(X_test)\n",
    "y_pred_e2 = (y_pred_e2 > thresholds2[ix2])\n",
    "y_pred_e2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "vuUwxGud1Uu9",
    "outputId": "1ff689c7-aed7-423b-9ef9-abd0c3278779"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Calculating the confusion matrix \n",
    "cm22=confusion_matrix(y_test, y_pred_e2)\n",
    "labels = ['True Positive','False Negative','False Positive','True Negative']\n",
    "categories = [ 'Not Churning','Churning']\n",
    "make_confusion_matrix(cm22, \n",
    "                      group_names=labels,\n",
    "                      categories=categories, \n",
    "                      cmap='Blues')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8HgwF0PT1Ztb",
    "outputId": "43d7cddd-5ce5-4ba0-dd06-b42b8a984139"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.93      0.76      0.84      1593\n",
      "           1       0.46      0.78      0.58       407\n",
      "\n",
      "    accuracy                           0.77      2000\n",
      "   macro avg       0.69      0.77      0.71      2000\n",
      "weighted avg       0.83      0.77      0.79      2000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Accuracy as per the classification report \n",
    "cr22=metrics.classification_report(y_test,y_pred_e2)\n",
    "print(cr22)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Hs_Smk526MbF"
   },
   "source": [
    "In creasing the number of layers improved the performance, yet after applying the ROC curve and spotting the best threshold the number of False Negatives remained the same."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "K_55Yk9MHkKz"
   },
   "source": [
    "### Model 3\n",
    "(Dropout Technique)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cC07AowaLSeV"
   },
   "source": [
    "This model will try dropout technique with SGD optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "id": "K4DTxkayHtHU"
   },
   "outputs": [],
   "source": [
    "backend.clear_session()\n",
    "np.random.seed(42)\n",
    "random.seed(42)\n",
    "tf.random.set_seed(42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "id": "jBxLG2nxHxcV"
   },
   "outputs": [],
   "source": [
    "model3 = Sequential()\n",
    "model3.add(Dense(256,activation='relu',input_dim = X_train.shape[1]))\n",
    "model3.add(Dropout(0.2))\n",
    "model3.add(Dense(128,activation='relu',kernel_initializer='he_uniform'))\n",
    "model3.add(Dropout(0.2))\n",
    "model3.add(Dense(64,activation='relu'))\n",
    "model3.add(Dropout(0.2))\n",
    "model3.add(Dense(32,activation='relu',kernel_initializer='he_uniform'))\n",
    "model3.add(Dense(1, activation = 'sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "h2bnz2fWH0H8",
    "outputId": "d644c353-9ca2-4965-865f-83a7d01960cd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 256)               3072      \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 256)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 128)               32896     \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 64)                8256      \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 64)                0         \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 32)                2080      \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 1)                 33        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 46,337\n",
      "Trainable params: 46,337\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model3.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "id": "6FM2hT70H3-J"
   },
   "outputs": [],
   "source": [
    "optimizer = tf.keras.optimizers.SGD(0.001)\n",
    "model3.compile(loss='binary_crossentropy',optimizer=optimizer,metrics=[tf.keras.metrics.Recall()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "QCjoHyFcH7Ua",
    "outputId": "81e62af5-563a-41a8-fdc4-22bd5e0b58f4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "100/100 [==============================] - 1s 5ms/step - loss: 0.7247 - recall: 0.6031 - val_loss: 0.6939 - val_recall: 0.3015\n",
      "Epoch 2/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.6783 - recall: 0.2803 - val_loss: 0.6558 - val_recall: 0.0627\n",
      "Epoch 3/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.6464 - recall: 0.0880 - val_loss: 0.6253 - val_recall: 0.0030\n",
      "Epoch 4/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.6187 - recall: 0.0355 - val_loss: 0.6009 - val_recall: 0.0000e+00\n",
      "Epoch 5/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5951 - recall: 0.0085 - val_loss: 0.5811 - val_recall: 0.0000e+00\n",
      "Epoch 6/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5791 - recall: 0.0023 - val_loss: 0.5653 - val_recall: 0.0000e+00\n",
      "Epoch 7/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5624 - recall: 0.0023 - val_loss: 0.5524 - val_recall: 0.0000e+00\n",
      "Epoch 8/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5507 - recall: 0.0000e+00 - val_loss: 0.5421 - val_recall: 0.0000e+00\n",
      "Epoch 9/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5399 - recall: 0.0000e+00 - val_loss: 0.5337 - val_recall: 0.0000e+00\n",
      "Epoch 10/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5325 - recall: 0.0000e+00 - val_loss: 0.5270 - val_recall: 0.0000e+00\n",
      "Epoch 11/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5266 - recall: 0.0000e+00 - val_loss: 0.5215 - val_recall: 0.0000e+00\n",
      "Epoch 12/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5207 - recall: 0.0000e+00 - val_loss: 0.5171 - val_recall: 0.0000e+00\n",
      "Epoch 13/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5193 - recall: 0.0000e+00 - val_loss: 0.5134 - val_recall: 0.0000e+00\n",
      "Epoch 14/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5156 - recall: 0.0000e+00 - val_loss: 0.5103 - val_recall: 0.0000e+00\n",
      "Epoch 15/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5089 - recall: 0.0000e+00 - val_loss: 0.5075 - val_recall: 0.0000e+00\n",
      "Epoch 16/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5076 - recall: 0.0000e+00 - val_loss: 0.5051 - val_recall: 0.0000e+00\n",
      "Epoch 17/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5078 - recall: 0.0000e+00 - val_loss: 0.5030 - val_recall: 0.0000e+00\n",
      "Epoch 18/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5040 - recall: 0.0000e+00 - val_loss: 0.5010 - val_recall: 0.0000e+00\n",
      "Epoch 19/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5033 - recall: 0.0000e+00 - val_loss: 0.4991 - val_recall: 0.0000e+00\n",
      "Epoch 20/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4992 - recall: 0.0000e+00 - val_loss: 0.4973 - val_recall: 0.0000e+00\n",
      "Epoch 21/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.5004 - recall: 0.0000e+00 - val_loss: 0.4956 - val_recall: 0.0000e+00\n",
      "Epoch 22/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4982 - recall: 0.0000e+00 - val_loss: 0.4939 - val_recall: 0.0000e+00\n",
      "Epoch 23/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4959 - recall: 0.0000e+00 - val_loss: 0.4923 - val_recall: 0.0000e+00\n",
      "Epoch 24/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4943 - recall: 0.0000e+00 - val_loss: 0.4908 - val_recall: 0.0000e+00\n",
      "Epoch 25/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4937 - recall: 0.0000e+00 - val_loss: 0.4892 - val_recall: 0.0000e+00\n",
      "Epoch 26/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4933 - recall: 0.0000e+00 - val_loss: 0.4877 - val_recall: 0.0000e+00\n",
      "Epoch 27/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4900 - recall: 0.0000e+00 - val_loss: 0.4862 - val_recall: 0.0000e+00\n",
      "Epoch 28/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4882 - recall: 0.0000e+00 - val_loss: 0.4847 - val_recall: 0.0000e+00\n",
      "Epoch 29/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4844 - recall: 0.0000e+00 - val_loss: 0.4832 - val_recall: 0.0000e+00\n",
      "Epoch 30/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4856 - recall: 0.0000e+00 - val_loss: 0.4818 - val_recall: 0.0000e+00\n",
      "Epoch 31/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4859 - recall: 0.0000e+00 - val_loss: 0.4804 - val_recall: 0.0000e+00\n",
      "Epoch 32/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4832 - recall: 0.0000e+00 - val_loss: 0.4790 - val_recall: 0.0000e+00\n",
      "Epoch 33/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4833 - recall: 0.0000e+00 - val_loss: 0.4776 - val_recall: 0.0000e+00\n",
      "Epoch 34/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4798 - recall: 0.0000e+00 - val_loss: 0.4762 - val_recall: 0.0000e+00\n",
      "Epoch 35/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4788 - recall: 0.0000e+00 - val_loss: 0.4749 - val_recall: 0.0000e+00\n",
      "Epoch 36/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4794 - recall: 0.0000e+00 - val_loss: 0.4735 - val_recall: 0.0000e+00\n",
      "Epoch 37/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4766 - recall: 0.0000e+00 - val_loss: 0.4721 - val_recall: 0.0000e+00\n",
      "Epoch 38/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4751 - recall: 0.0000e+00 - val_loss: 0.4708 - val_recall: 0.0000e+00\n",
      "Epoch 39/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4756 - recall: 0.0000e+00 - val_loss: 0.4695 - val_recall: 0.0000e+00\n",
      "Epoch 40/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4743 - recall: 0.0000e+00 - val_loss: 0.4683 - val_recall: 0.0000e+00\n",
      "Epoch 41/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4752 - recall: 7.7220e-04 - val_loss: 0.4670 - val_recall: 0.0000e+00\n",
      "Epoch 42/100\n",
      "100/100 [==============================] - 0s 4ms/step - loss: 0.4706 - recall: 7.7220e-04 - val_loss: 0.4658 - val_recall: 0.0000e+00\n",
      "Epoch 43/100\n",
      "100/100 [==============================] - 1s 8ms/step - loss: 0.4704 - recall: 0.0000e+00 - val_loss: 0.4645 - val_recall: 0.0000e+00\n",
      "Epoch 44/100\n",
      "100/100 [==============================] - 0s 4ms/step - loss: 0.4677 - recall: 7.7220e-04 - val_loss: 0.4633 - val_recall: 0.0000e+00\n",
      "Epoch 45/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4684 - recall: 7.7220e-04 - val_loss: 0.4621 - val_recall: 0.0000e+00\n",
      "Epoch 46/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4661 - recall: 7.7220e-04 - val_loss: 0.4609 - val_recall: 0.0000e+00\n",
      "Epoch 47/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4671 - recall: 0.0015 - val_loss: 0.4597 - val_recall: 0.0000e+00\n",
      "Epoch 48/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4674 - recall: 7.7220e-04 - val_loss: 0.4586 - val_recall: 0.0000e+00\n",
      "Epoch 49/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4607 - recall: 0.0023 - val_loss: 0.4574 - val_recall: 0.0000e+00\n",
      "Epoch 50/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4638 - recall: 0.0031 - val_loss: 0.4563 - val_recall: 0.0000e+00\n",
      "Epoch 51/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4656 - recall: 0.0015 - val_loss: 0.4552 - val_recall: 0.0000e+00\n",
      "Epoch 52/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4619 - recall: 0.0031 - val_loss: 0.4541 - val_recall: 0.0000e+00\n",
      "Epoch 53/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4606 - recall: 0.0023 - val_loss: 0.4531 - val_recall: 0.0000e+00\n",
      "Epoch 54/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4606 - recall: 0.0031 - val_loss: 0.4521 - val_recall: 0.0000e+00\n",
      "Epoch 55/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4588 - recall: 0.0039 - val_loss: 0.4511 - val_recall: 0.0000e+00\n",
      "Epoch 56/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4589 - recall: 0.0031 - val_loss: 0.4502 - val_recall: 0.0000e+00\n",
      "Epoch 57/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4587 - recall: 0.0069 - val_loss: 0.4493 - val_recall: 0.0000e+00\n",
      "Epoch 58/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4581 - recall: 0.0054 - val_loss: 0.4483 - val_recall: 0.0000e+00\n",
      "Epoch 59/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4554 - recall: 0.0054 - val_loss: 0.4474 - val_recall: 0.0000e+00\n",
      "Epoch 60/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4558 - recall: 0.0062 - val_loss: 0.4464 - val_recall: 0.0000e+00\n",
      "Epoch 61/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4538 - recall: 0.0069 - val_loss: 0.4455 - val_recall: 0.0000e+00\n",
      "Epoch 62/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4545 - recall: 0.0124 - val_loss: 0.4447 - val_recall: 0.0030\n",
      "Epoch 63/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4547 - recall: 0.0100 - val_loss: 0.4439 - val_recall: 0.0030\n",
      "Epoch 64/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4498 - recall: 0.0185 - val_loss: 0.4431 - val_recall: 0.0030\n",
      "Epoch 65/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4508 - recall: 0.0201 - val_loss: 0.4423 - val_recall: 0.0030\n",
      "Epoch 66/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4501 - recall: 0.0116 - val_loss: 0.4414 - val_recall: 0.0030\n",
      "Epoch 67/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4495 - recall: 0.0116 - val_loss: 0.4406 - val_recall: 0.0030\n",
      "Epoch 68/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4465 - recall: 0.0124 - val_loss: 0.4399 - val_recall: 0.0030\n",
      "Epoch 69/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4465 - recall: 0.0162 - val_loss: 0.4391 - val_recall: 0.0030\n",
      "Epoch 70/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4475 - recall: 0.0216 - val_loss: 0.4384 - val_recall: 0.0030\n",
      "Epoch 71/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4483 - recall: 0.0255 - val_loss: 0.4377 - val_recall: 0.0030\n",
      "Epoch 72/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4452 - recall: 0.0286 - val_loss: 0.4369 - val_recall: 0.0030\n",
      "Epoch 73/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4463 - recall: 0.0255 - val_loss: 0.4362 - val_recall: 0.0060\n",
      "Epoch 74/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4461 - recall: 0.0208 - val_loss: 0.4356 - val_recall: 0.0060\n",
      "Epoch 75/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4448 - recall: 0.0332 - val_loss: 0.4349 - val_recall: 0.0090\n",
      "Epoch 76/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4447 - recall: 0.0232 - val_loss: 0.4343 - val_recall: 0.0090\n",
      "Epoch 77/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4414 - recall: 0.0278 - val_loss: 0.4336 - val_recall: 0.0119\n",
      "Epoch 78/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4397 - recall: 0.0402 - val_loss: 0.4330 - val_recall: 0.0179\n",
      "Epoch 79/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4407 - recall: 0.0301 - val_loss: 0.4324 - val_recall: 0.0239\n",
      "Epoch 80/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4398 - recall: 0.0402 - val_loss: 0.4317 - val_recall: 0.0269\n",
      "Epoch 81/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4400 - recall: 0.0378 - val_loss: 0.4311 - val_recall: 0.0299\n",
      "Epoch 82/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4419 - recall: 0.0363 - val_loss: 0.4305 - val_recall: 0.0299\n",
      "Epoch 83/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4397 - recall: 0.0571 - val_loss: 0.4299 - val_recall: 0.0358\n",
      "Epoch 84/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4397 - recall: 0.0471 - val_loss: 0.4293 - val_recall: 0.0388\n",
      "Epoch 85/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4384 - recall: 0.0386 - val_loss: 0.4288 - val_recall: 0.0418\n",
      "Epoch 86/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4391 - recall: 0.0533 - val_loss: 0.4282 - val_recall: 0.0478\n",
      "Epoch 87/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4400 - recall: 0.0641 - val_loss: 0.4277 - val_recall: 0.0507\n",
      "Epoch 88/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4373 - recall: 0.0587 - val_loss: 0.4271 - val_recall: 0.0597\n",
      "Epoch 89/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4341 - recall: 0.0641 - val_loss: 0.4266 - val_recall: 0.0627\n",
      "Epoch 90/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4349 - recall: 0.0618 - val_loss: 0.4261 - val_recall: 0.0687\n",
      "Epoch 91/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4351 - recall: 0.0548 - val_loss: 0.4255 - val_recall: 0.0776\n",
      "Epoch 92/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4374 - recall: 0.0649 - val_loss: 0.4250 - val_recall: 0.0806\n",
      "Epoch 93/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4348 - recall: 0.0664 - val_loss: 0.4245 - val_recall: 0.0806\n",
      "Epoch 94/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4356 - recall: 0.0649 - val_loss: 0.4240 - val_recall: 0.0866\n",
      "Epoch 95/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4339 - recall: 0.0695 - val_loss: 0.4235 - val_recall: 0.0925\n",
      "Epoch 96/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4336 - recall: 0.0726 - val_loss: 0.4231 - val_recall: 0.1015\n",
      "Epoch 97/100\n",
      "100/100 [==============================] - 0s 3ms/step - loss: 0.4339 - recall: 0.0842 - val_loss: 0.4226 - val_recall: 0.1075\n",
      "Epoch 98/100\n",
      "100/100 [==============================] - 1s 8ms/step - loss: 0.4321 - recall: 0.0826 - val_loss: 0.4222 - val_recall: 0.1194\n",
      "Epoch 99/100\n",
      "100/100 [==============================] - 0s 4ms/step - loss: 0.4317 - recall: 0.0888 - val_loss: 0.4217 - val_recall: 0.1224\n",
      "Epoch 100/100\n",
      "100/100 [==============================] - 1s 5ms/step - loss: 0.4336 - recall: 0.0826 - val_loss: 0.4212 - val_recall: 0.1254\n"
     ]
    }
   ],
   "source": [
    "history_3 = model3.fit(X_train,y_train,batch_size=64,epochs=100,verbose=1,validation_split = 0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "V7GKjd2pH92L",
    "outputId": "c323382b-637d-4386-cbda-2d31d81e78e5"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXyU9bn//9c1k8m+L0BICAlL2APBCCIuKC6orYq7tT3ac6qtv3rUrkf7PT22tp56zvG0drGn1Va7KrXUhdaFukDdAAGRfU+AJED2hWwkmbl+f9w3MGICiTAMyVzPx2MeM3Mvc39uRuedz3J/blFVjDHGmKN5wl0AY4wxpycLCGOMMT2ygDDGGNMjCwhjjDE9soAwxhjTIwsIY4wxPbKAMOYkEJHfiMj3+7jtLhG56EQ/x5hQs4AwxhjTIwsIY4wxPbKAMBHDbdr5hoisE5FWEfm1iAwVkVdE5ICIvC4iaUHbXykiG0WkUUSWisiEoHXFIvKBu9+fgNijjvUpEfnQ3fc9ESn6hGW+XUR2iEi9iCwSkeHuchGRH4lItYg0i8h6EZnsrrtcRDa5ZasUka9/on8wE/EsIEykuRa4GCgEPg28AnwLyML5/+FuABEpBJ4B7nXXvQz8VUSiRSQaeAH4PZAO/Nn9XNx9i4EngS8CGcAvgUUiEtOfgorIhcAPgBuAbGA3sMBdfQlwnnseKe42de66XwNfVNUkYDLwZn+Oa8whFhAm0vxUVatUtRJ4G1ihqmtUtQN4Hih2t7sReElVX1PVLuARIA44GzgL8AGPqmqXqi4EVgYd4w7gl6q6QlX9qvpb4KC7X3/cAjypqh+o6kHgfmCWiOQDXUASMB4QVd2sqvvc/bqAiSKSrKoNqvpBP49rDGABYSJPVdDr9h7eJ7qvh+P8xQ6AqgaAciDHXVepH53pcnfQ65HA19zmpUYRaQRGuPv1x9FlaMGpJeSo6pvAz4DHgGoReVxEkt1NrwUuB3aLyD9EZFY/j2sMYAFhTG/24vzQA06bP86PfCWwD8hxlx2SF/S6HHhIVVODHvGq+swJliEBp8mqEkBVf6KqZwATcZqavuEuX6mqVwFDcJrCnu3ncY0BLCCM6c2zwBUiMldEfMDXcJqJ3gOWAd3A3SLiE5FrgBlB+z4BfElEZrqdyQkicoWIJPWzDM8AnxeRaW7/xX/iNIntEpEz3c/3Aa1ABxBw+0huEZEUt2msGQicwL+DiWAWEMb0QFW3Ap8FfgrU4nRof1pVO1W1E7gGuA2ox+mveC5o31XA7ThNQA3ADnfb/pbhdeDbwF9wai2jgZvc1ck4QdSA0wxVB/yPu+5zwC4RaQa+hNOXYUy/id0wyBhjTE+sBmGMMaZHFhDGGGN6ZAFhjDGmRxYQxhhjehQV7gKcLJmZmZqfnx/uYhhjzICyevXqWlXN6mndoAmI/Px8Vq1aFe5iGGPMgCIiu3tbZ01MxhhjemQBYYwxpkcWEMYYY3o0aPogetLV1UVFRQUdHR3hLsqgERsbS25uLj6fL9xFMcaE2KAOiIqKCpKSksjPz+ejE2+aT0JVqauro6KigoKCgnAXxxgTYoO6iamjo4OMjAwLh5NERMjIyLAamTERYlAHBGDhcJLZv6cxkWPQB8Tx+AMBqpo7aOvsDndRjDHmtBLxAQFQ1dxB60F/SD67sbGRn//85/3e7/LLL6exsTEEJTLGmL6J+IDwiOARocsfmptu9RYQ3d3HrrG8/PLLpKamhqRMxhjTF4N6FFNfiAg+r9DtD82Nk+677z527tzJtGnT8Pl8xMbGkpaWxpYtW9i2bRtXX3015eXldHR0cM8993DHHXcAR6YOaWlp4bLLLuOcc87hvffeIycnhxdffJG4uLiQlNcYYw6JmID47l83smlvc4/rOrqc5qVYn7dfnzlxeDIPfHrSMbd5+OGH2bBhAx9++CFLly7liiuuYMOGDYeHiT755JOkp6fT3t7OmWeeybXXXktGRsZHPmP79u0888wzPPHEE9xwww385S9/4bOf/Wy/ymqMMf0VMQFxLCJCIHBqbr06Y8aMj1xD8JOf/ITnn38egPLycrZv3/6xgCgoKGDatGkAnHHGGezateuUlNUYE9kiJiCO9Zf+3sZ26ls7mTQ8OeTDOBMSEg6/Xrp0Ka+//jrLli0jPj6eOXPm9HiNQUxMzOHXXq+X9vb2kJbRGGPAOqkB8Hk9BFQJ6MmvRSQlJXHgwIEe1zU1NZGWlkZ8fDxbtmxh+fLlJ/34xhjzSUVMDeJYfF6n1tDlV7wnOTIzMjKYPXs2kydPJi4ujqFDhx5eN2/ePH7xi18wYcIExo0bx1lnnXVyD26MMSdANAR/NYdDSUmJHn3DoM2bNzNhwoTj7ttysJvSmhZGZSaQGGuT0B1PX/9djTGnPxFZraolPa2zJibA5zlSgzDGGOOwgACi3HalrkBoLpYzxpiByAIC8HoEryd0F8sZY8xAZAHhivJ4QjbdhjHGDEQhDQgRmSciW0Vkh4jc18P6H4nIh+5jm4g0Bq27VUS2u49bQ1lOcEYyWR+EMcYcEbJhriLiBR4DLgYqgJUiskhVNx3aRlW/ErT9vwLF7ut04AGgBFBgtbtvQ6jK6/N6aD1oU34bY8whoaxBzAB2qGqpqnYCC4CrjrH9zcAz7utLgddUtd4NhdeAeSEsK1FeoSughHvYb2JiIgB79+7luuuu63GbOXPmcPSQ3qM9+uijtLW1HX5v04cbY/orlAGRA5QHva9wl32MiIwECoA3+7OviNwhIqtEZFVNTc0JFdbn9aCq+E/RnEzHM3z4cBYuXPiJ9z86IGz6cGNMf50undQ3AQtVtV937VHVx1W1RFVLsrKyTqgAoboW4r777uOxxx47/P473/kO3//+95k7dy7Tp09nypQpvPjiix/bb9euXUyePBmA9vZ2brrpJiZMmMD8+fM/MhfTnXfeSUlJCZMmTeKBBx4AnAkA9+7dywUXXMAFF1wAONOH19bWAvDDH/6QyZMnM3nyZB599NHDx5swYQK33347kyZN4pJLLrE5n4yJcKGcaqMSGBH0Ptdd1pObgC8fte+co/ZdekKleeU+2L++19WJqozq9OPzecDTx9wcNgUue/iYm9x4443ce++9fPnLzuk9++yzLF68mLvvvpvk5GRqa2s566yzuPLKK3udKPD//u//iI+PZ/Pmzaxbt47p06cfXvfQQw+Rnp6O3+9n7ty5rFu3jrvvvpsf/vCHLFmyhMzMzI981urVq3nqqadYsWIFqsrMmTM5//zzSUtLs2nFjTEfEcoaxEpgrIgUiEg0TggsOnojERkPpAHLghYvBi4RkTQRSQMucZeFgIK/E9HAoXcnVXFxMdXV1ezdu5e1a9eSlpbGsGHD+Na3vkVRUREXXXQRlZWVVFVV9foZb7311uEf6qKiIoqKig6ve/bZZ5k+fTrFxcVs3LiRTZs29fYxALzzzjvMnz+fhIQEEhMTueaaa3j77bcBm1bcGPNRIatBqGq3iNyF88PuBZ5U1Y0i8iCwSlUPhcVNwAIN6h1W1XoR+R5OyAA8qKr1J1Sg3v7SD3TD/vVI0nBKm2IYmhzL0OTYEzrU0a6//noWLlzI/v37ufHGG/njH/9ITU0Nq1evxufzkZ+f3+M038dTVlbGI488wsqVK0lLS+O22277RJ9ziE0rbowJFtI+CFV9WVULVXW0qj7kLvuPoHBAVb+jqh+7RkJVn1TVMe7jqZAV0hMF4kUCnUR5hO4QXCx34403smDBAhYuXMj1119PU1MTQ4YMwefzsWTJEnbv3n3M/c877zyefvppADZs2MC6desAaG5uJiEhgZSUFKqqqnjllVcO79PbNOPnnnsuL7zwAm1tbbS2tvL8889z7rnnnsSzNcYMFjbdN4A3Gro7ifJ6QnKx3KRJkzhw4AA5OTlkZ2dzyy238OlPf5opU6ZQUlLC+PHjj7n/nXfeyec//3kmTJjAhAkTOOOMMwCYOnUqxcXFjB8/nhEjRjB79uzD+9xxxx3MmzeP4cOHs2TJksPLp0+fzm233caMGTMA+MIXvkBxcbE1JxljPsam+waoKwX/Qco8eXQHAowdkhSiUg4ONt23MYOHTfd9PFHR4O/EZxP2GWPMYRYQ4DQxaYBor9LtD4T9ampjjDkdDPqA6NOPvTcagGi6UaD7NLma+nRk4WlM5BjUAREbG0tdXd3xf9SCAgKwab97oarU1dURG3tyhwEbY05Pg3oUU25uLhUVFRx3nqZAAJqr6Y7ppKrDh78+mlif99QUcoCJjY0lNzc33MUwxpwCgzogfD4fBQUFx99QFX5wCa2TbubTyy7gofmTuaVoZOgLaIwxp7FB3cTUZyKQmkdcayVej7Cv8ZNfjWyMMYOFBcQhKSPwNJczIi2OsrrWcJfGGGPCzgLikNQ8aNxDQWYCZTUWEMYYYwFxSGoedDQxLk3ZVddqwzmNMRHPAuKQVOfWFZPim2jr9FN94GCYC2SMMeFlAXFIah4Ao33OrOKl1sxkjIlwFhCHpDrDWofjXDNRVmsBYYyJbBYQh8RnQFQcKQf3ERPloay2JdwlMsaYsLKAOMS9FkKa9pCfkUBZbVu4S2SMMWFlAREseKir1SCMMRHOAiJY6ghoLKcgK4E99W0huf2oMcYMFBYQwVLzoL2esSnQ5VcqG9vDXSJjjAkbC4hg7lDXwtgGwEYyGWMimwVEsBQnIEZ46gALCGNMZLOACObWIJI79pIUG2UBYYyJaBYQwRKHgC8eaShjVGaCBYQxJqKFNCBEZJ6IbBWRHSJyXy/b3CAim0Rko4g8HbTcLyIfuo9FoSxnUGEgYwzUbiffAsIYE+FCdkc5EfECjwEXAxXAShFZpKqbgrYZC9wPzFbVBhEZEvQR7ao6LVTl61XmWKhYRcHkBBat3UtHl99uP2qMiUihrEHMAHaoaqmqdgILgKuO2uZ24DFVbQBQ1eoQlqdvMguhcQ+j07yowp56u6LaGBOZQhkQOUB50PsKd1mwQqBQRN4VkeUiMi9oXayIrHKXX93TAUTkDnebVTU1NSen1JljAWWcrxawWV2NMZErZE1M/Tj+WGAOkAu8JSJTVLURGKmqlSIyCnhTRNar6s7gnVX1ceBxgJKSkpNzh5+MsQCMCJQDceyy248aYyJUKGsQlcCIoPe57rJgFcAiVe1S1TJgG05goKqV7nMpsBQoDmFZj8gYA0BcUylZSTHsqLY5mYwxkSmUAbESGCsiBSISDdwEHD0a6QWc2gMikonT5FQqImkiEhO0fDawiVMhOt65YK52O4VDE9ledeCUHNYYY043IQsIVe0G7gIWA5uBZ1V1o4g8KCJXupstBupEZBOwBPiGqtYBE4BVIrLWXf5w8OinkMscC7XbKByaxLaqFgIBuz+1MSbyhLQPQlVfBl4+atl/BL1W4KvuI3ib94ApoSzbMWWOhQ+WM25aIu1dfioa2snLiA9bcYwxJhzsSuqeZI6FrlYmJTsd1NusmckYE4EsIHqSWQjAaNkHwFYLCGNMBLKA6IkbEPHNpeSkxlkNwhgTkSwgepI4FKKT3I7qRLZV2VBXY0zksYDoiYg7kmk7hUOT2FndYrcfNcZEHAuI3mQWHg6ITn+AXXU2J5MxJrJYQPQmcww0VzAhw/knsn4IY0yksYDojdtRPcqzHxELCGNM5LGA6I0bELFNpYxMj7eAMMZEHAuI3qSPAvFAzVYKhyaxdb8FhDEmslhA9CYqBtJHQ9VGCocmsauujYPd/nCXyhhjThkLiGPJngr711E4LAl/QO3mQcaYiGIBcSzZRdBUzoSUbsA6qo0xkcUC4liypwKQ37mDKI9YP4QxJqJYQBzLsCIAfNXrGJWVwOZ9zWEukDHGnDoWEMcSnw4pI2D/OibnpLC+shnnFhbGGDP4WUAcT/ZU2LeWKTkp1LYcpKr5YLhLZIwxp4QFxPEMK4K6nUwd4tx8b31lU5gLZIwxp4YFxPFkTwWUiZ49eMQCwhgTOSwgjifb6aiOrd3A6KxENlhAGGMihAXE8SRlQ0IW7FvHlJwUq0EYYyKGBcTxiDj9EPvWMjknhZoDB6lq7gh3qYwxJuQsIPoiuwhqNjM1OxaA9RVWizDGDH4hDQgRmSciW0Vkh4jc18s2N4jIJhHZKCJPBy2/VUS2u49bQ1nO48qeCoFuJkZVWke1MSZiRIXqg0XECzwGXAxUACtFZJGqbgraZixwPzBbVRtEZIi7PB14ACgBFFjt7tsQqvIek3tFdVztRkZn5VtHtTEmIoSyBjED2KGqparaCSwArjpqm9uBxw798Ktqtbv8UuA1Va13170GzAthWY8trQBikmHfh0zJSWGdBYQxJgKEMiBygPKg9xXusmCFQKGIvCsiy0VkXj/2RUTuEJFVIrKqpqbmJBb9KB4P5EyH8veto9oYEzHC3UkdBYwF5gA3A0+ISGpfd1bVx1W1RFVLsrKyQlREV97ZULWRaVkCWEe1MWbwC2VAVAIjgt7nusuCVQCLVLVLVcuAbTiB0Zd9T62RswBlYvcmxDqqjTERIJQBsRIYKyIFIhIN3AQsOmqbF3BqD4hIJk6TUymwGLhERNJEJA24xF0WPjkl4PERu3cFY7IS+bC8MazFMcaYUAtZQKhqN3AXzg/7ZuBZVd0oIg+KyJXuZouBOhHZBCwBvqGqdapaD3wPJ2RWAg+6y8InOh6GT4M9y5g9JpMVZXV0dNk9qo0xg1dI+yBU9WVVLVTV0ar6kLvsP1R1kftaVfWrqjpRVaeo6oKgfZ9U1THu46lQlrPP8mZB5QfMHZNMR1eAZTvrwl0iY4wJmXB3Ug8sI8+GQBczosuIj/byxpaqcJfIGGNCxgKiP0bMBCCmcgWzx2SyZEuN3WHOGDNoWUD0R3w6DJkIe95j7vghVDa2s7XqQLhLZYwxIWEB0V95s6D8fS4oTAfgjc3Vx9nBGGMGpj4FhIgkiIjHfV0oIleKiC+0RTtNjTwbOlsY2radyTnJLNliAWGMGZz6WoN4C4gVkRzg78DngN+EqlCntbxZzvPuZVw4figf7GmgobUzvGUyxpgQ6GtAiKq2AdcAP1fV64FJoSvWaSwlB1LzYPe7zB0/hIDCP7aFcB4oY4wJkz4HhIjMAm4BXnKXeUNTpAFg9IVQupQpw+LITIzhDWtmMsYMQn0NiHtx7tvwvHs19CicK58jU+E86GzBs+c95ozL4q1tNfgDNtzVGDO49CkgVPUfqnqlqv6X21ldq6p3h7hsp6+C8yEqFrYt5vzCLJrau1hbYXMzGWMGl76OYnpaRJJFJAHYAGwSkW+Etminseh4JyS2vcK5YzLwCPxjq/VDGGMGl742MU1U1WbgauAVoABnJFPkKrwUGnaR2raLaSNSWWod1caYQaavAeFzr3u4Gvf+DTj3io5chZc6z9te5fzCIayraKTehrsaYwaRvgbEL4FdQALwloiMBJpDVagBISUXhk6Bra9y/rgsVOHt7VaLMMYMHn3tpP6Jquao6uXuFN27gQtCXLbT37h5UL6cKekB0uJ9dj2EMWZQ6WsndYqI/FBEVrmP/8WpTUS2wnmgAbw73+C8wize2lZLwIa7GmMGib42MT0JHABucB/NwOlxE59wGj4d4jNh2yucX5hFbctBNu2L7JY3Y8zg0deAGK2qD6hqqfv4LjAqlAUbEDweGH85bH2Vc/PjAZt2wxgzePQ1INpF5JxDb0RkNtAemiINMEU3QVcrWeWvMTknmaVbbdoNY8zg0NeA+BLwmIjsEpFdwM+AL4asVANJ3ixn8r51C7h04jBW7mpgV21ruEtljDEnrK+jmNaq6lSgCChS1WLgwpCWbKDweKDoRihdyk0TfHg9wjMr94S7VMYYc8L6dUc5VW12r6gG+GoIyjMwFd0EGiCrbBEXTRjCwlUVHOz2h7tUxhhzQk7klqNy0kox0GWOgdwzYe0CPjNzJHWtnSzeWBXuUhljzAk5kYCwAf/Bim6E6o2cm7iPEelxPL1id7hLZIwxJ+SYASEiB0SkuYfHAWD48T5cROaJyFYR2SEi9/Ww/jYRqRGRD93HF4LW+YOWL/pEZ3cqTb4WPD486//EzTPyWF5az47qlnCXyhhjPrFjBoSqJqlqcg+PJFWNOta+IuIFHgMuAyYCN4vIxB42/ZOqTnMfvwpa3h60/Mr+ntgpF5/uTOC3dgHXF2US5RGeed86q40xA9eJNDEdzwxgh3thXSewALgqhMcLv5lfgrZaskqf49LJw1i4uoL2TuusNsYMTKEMiBygPOh9hbvsaNeKyDoRWSgiI4KWx7rzPi0Xkat7OoCI3HFofqiamtPgCub8c2B4Mbz3M26dOYKm9i6eX1MZ7lIZY8wnEsqA6Iu/AvmqWgS8Bvw2aN1IVS0BPgM8KiKjj95ZVR9X1RJVLcnKyjo1JT4WETj7bqjfyZkHlzE5J5kn3y1D1frzjTEDTygDohIIrhHkussOU9U6VT3ovv0VcEbQukr3uRRYChSHsKwnz4QrIXUk8t5P+Oez89lR3cLb22vDXSpjjOm3UAbESmCsiBSISDRwE/CR0Ugikh309kpgs7s8TURi3NeZwGxgUwjLevJ4o2DWXVCxkk+l7SErKYYn3y0Ld6mMMabfQhYQqtoN3AUsxvnhf1ZVN4rIgyJyaFTS3SKyUUTWAncDt7nLJwCr3OVLgIdVdWAEBEDxLRCXRvTyn/LZmSNZurWGnTU25NUYM7DIYGkfLykp0VWrVoW7GEcsfRiW/oDGz7zEjN8e4MYzR/C9qyeHu1TGGPMRIrLa7e/9mHB3Ug9es+6CxGGkvvUdrpqazZ9Xl1PZaDOkG2MGDguIUIlJhLnfhoqV3DdiI4Lw7Rc22IgmY8yAYQERSlM/A8OKyFj2n3xzbh5vbqnmb+v2hbtUxhjTJxYQoeTxwKX/Cc0V3CovU5Sbwnf/upHGts5wl8wYY47LAiLUCs6F8Z/C884PeeSiZBrauvjPlzeHu1TGGHNcFhCnwryHwRNF4bvf4I5zRvLsqgpesqYmY8xpzgLiVEgdAVc8AuXL+Vr8y0zPS+UbC9eydf+BcJfMGGN6ZQFxqky5HibNJ+qth3niYh8JMVHc8ftVNLV1hbtkxhjTIwuIU0UErvghJAwhY/GXefzGQvY2tnP3gjX4Azb01Rhz+rGAOJXi02H+L6BuJ8XLv8qDnxrPP7bV8L9/3xrukhljzMdYQJxqo86HK/4XdrzGzXU/5eYzR/DzpTut09oYc9qxgAiHks/D7Htg1ZN8b8ibnDEyja//eS1b9jeHu2TGGHOYBUS4zP0OTLyaqDce4DeT15EcF8Xtv1vFix9WUtdy8Li7G2NMqEWFuwARy+OB+b+E7oMkvfFvPH/GN7ly7UzuWfAhIlCUm8q3r5hASX56uEtqjIlQVoMIJ18s3Ph7mHwdw1f/N+/PfIcX/7+z+cpFhdS3HuSmx5fz+2W7bII/Y0xYWA0i3Lw+uOZxiEnE8+6PmNpQytSrHuPWWfnc+6c1fPvFjayraOL78ycTE+UNd2mNMRHEahCnA48XPvUoXPRd2PxXeOJCUlpL+fWtZ3L3hWP48+oKHvzrwLmhnjFmcLCAOF2IwDn3wj+9CG318MSFeNb8jq9eXMjt5xbwxxV7eHNLVbhLaYyJIBYQp5uC8+CLb8HwYvjr3fD7q/n6zDjGD0vimwvXUWsjnIwxp4gFxOkoJQf+aZEzNUfFKmJ+OZvfjV9Oe3sH9/1lvXVaG2NOCRksPzYlJSW6atWqcBfj5Gssh5e/DttepSkuj6803UDbyLlcMH4o54zNJCXOx4GObloPdjM+O5nEGBt3YIzpOxFZraolPa6zgBggtr+OLr4fqd3GWu8Uvtc2n1U6/iOb5KbF8fjnSpg4PDlMhTTGDDQWEIOFvwtWPQlvPQKt1VQNmc2G0V/i4PAzCajy/b9tpqm9i0eun8oVRdnhLq0xZgAIW0CIyDzgx4AX+JWqPnzU+tuA/wEq3UU/U9VfuetuBf7dXf59Vf3tsY4VEQFxSGcbrHwC3nkU2uthxEw4+26qcy7kzj9+yOrdDcwdP4Tx2UmMHZLErNEZDE2ODXepjTGnobAEhIh4gW3AxUAFsBK4WVU3BW1zG1CiqncdtW86sAooARRYDZyhqg29HS+iAuKQzlZY8wdY9jNo3ANpBXSX/AuP1s7g1Z0d7KptpTugpMT5+MVnz2DW6Ixwl9gYc5o5VkCEchTTDGCHqpaqaiewALiqj/teCrymqvVuKLwGzAtROQeu6ASY+UX41zVw3VOQOJSo1/6dr2+Yz+tjn2PzncNYdNdshiTF8Llfr+DZleUAtHf6eW9HLW9vryFgNysyxvQilENecoDyoPcVwMwetrtWRM7DqW18RVXLe9k3J1QFHfC8UTD5Guexby2seBzW/gnf6t9QlD2VRbM+y70bRvPNv6zjyXfL2FHdQrcbDKOyErj93FHML84h1mdTeRhjjgj3dRB/BfJVtQinlnDMfoajicgdIrJKRFbV1NSEpIADTvZUuPox+NoWuPwRCPiJW/wNfrH/Zl7OeYpzPeu449yRPHXbmfz4pmnER3u5/7n1nPvfS3jirVJaD3aH+wyMMaeJUPZBzAK+o6qXuu/vB1DVH/SyvReoV9UUEbkZmKOqX3TX/RJYqqrP9Ha8iOyD6AtV2PchrPkjrP8zdDRC4jAouh6KbkKHTGRZaT2PLd3BuzvqSIv38fnZBcwvzmFEeny4S2+MCbFwdVJH4TQbzcUZpbQS+IyqbgzaJltV97mv5wP/pqpnuZ3Uq4Hp7qYf4HRS1/d2PAuIPujqgO2LYe2fnOdAN2SNh8nXwZRrWX0gjZ+9uZ0lW53aWFFuCpdOGkbh0CRGZsSTlx5vzVDGDDLhHOZ6OfAozjDXJ1X1IRF5EFilqotE5AfAlUA3UA/cqapb3H3/GfiW+1EPqepTxzqWBUQ/tdbBphdg/ULY856zLHsaTJrP3tzL+OvuKF5av491FU0f2S0+2ktqnI+s5FjunTuWC8YPCUPhjTEni10oZ46tsdwJiw3Pwd4PnGXDi2Hi1TSPuoJSfxa761qpaGinobWThrYuPixvYGdNK587ayTfunwCcdFWs2JYIsgAABedSURBVDBmILKAMH1XX+aExcYXnL4LOBwWTLoa0vIB6Ojy88jirfzqnTJGZyVwz0WFXDppqN3UyJgBxgLCfDL1ZbDpRScw9q5xlg0rgolXwoQrIWsc7+6o5f7n1rOnvo20eB/zi3O5c85ospJiwlt2Y0yfWECYE9ewCzYtcu54V/G+syxjLIy/gsC4T/FuRx4LVlby9037SU+I5ue3nMEZI9PCWmRjzPFZQJiTq3kfbPmb8yh7G9QPScNh/OXsHnIht70ZTUVzFw98ehJTclJYUVbH+2X17G/uoKWjm5aD3UwbkcZ9l41nzJDEcJ+NMRHNAsKETls9bFvshMWON6C7nUBMCsu8JfyucQpvBYpoJ5ZRmQmMzIgnKdZHdJSHxRv209bl57Mz87i6OIeAKl1+pSAzwSYWNOYUsoAwp0ZnG+x8E7a8hG57BWlvwO+JpjvvXGImXgbjLoOUXADqWg7yo9e38fSKPQRPBxUf7eXBqyZz7fQcRCRMJ2JM5LCAMKeev9u5vmLrq7DtFagvdZYPmwKFl8G4eZBdTFl9O6U1LUR5PXgEHluyg+Wl9Vw9bTjfnz/F7pBnTIhZQJjwUoXa7U5QbH0FyleABiBxKIy9BAovhVFzICYJf0B5bMkOHn19GwnRUUwdkcq0Ealkp8ayv6mDysZ2ojzCly8Yw8iMhHCfmTEDngWEOb201sGO12Dbq06/xcFm8Pgg76zDgbG6NYu/rKlkbXkjW/YfwB9QPALDkmNpbO+iO6B86bxR3DlnjF2kZ8wJsIAwpy9/F+xZ7gTG9teh2p2qK3UkFM6DsRfTPnwWDV1ehiTFEOX1UNXcwUMvbWbR2r0MS47l01OzmTd5GMUj0jjYHaCysZ3Gtk4mDk8mPtqaqIw5FgsIM3A0lsP2vzsjo8regu52iIqF/HNh7MUw5iLIGA3Asp11/OIfO3lvZy1dfiXW56GjK3D4o6KjPMwsSOesURnsa2pnfUUTu+rauHPOaL50/uhwnaExpxULCDMwdbXD7ndh+2vOo36nszytwOm3GHsJ5J9Dc7eHJVuqWbOnkaykGIanxpIY42NFaR1Ltlazs6aVpJgoJuekALCstI6vXFTIPReNDePJGXN6sIAwg0N9qdNnsf3vbu2iA3wJkH8OjL4ARl8ImYVw1PDYhtZOUuJ8eDyCP6B8c+E6/vJBBf964Ri+enGhDac1Ec0Cwgw+nW1OSOx4DXYuOVK7SM6FMRfC6LlQcB7Ep39s10BA+X8vrOeZ98vxeYVYn5c4n5eLJw7la5eMIz0h+hSfjDHhYwFhBr+G3c5FejvfgNK34GATIM4tWEfNgVHnQ94s8MUBTkj8eXU5u+vaaO/yU9vSycvr95EQ7eVrl4zjjJFp7KhuobSmhbyMBK6aNhyfN9x36DXm5LOAMJHF3w2Vq6HsH1C6FMrfh0AXeKNhxEynOWrUBc4NkjxHfvS3VR3gu3/dyLs76j72kblpcXz5gjFMz0tjfWUT6yoa2V3XRkNbJ/WtnRQOTeLnt0y3O+6ZAccCwkS2zlbYvQxKl0DpP6BqvbM8Lt1phho1x3mkF6CqvLOjlpaObsYMSSQvI553d9Ty49e3szbo7noJ0V5GZSWSkRhNfLSXl9fv5/ZzC/h/V0z82OHrWg7y/JpKtlUd4K4LxpKXYff6NqcPCwhjgrVUO0Gx802nhnFgr7M8daTTFDVqDuSfB4lZh3dRVd7dUUdVcwdFuSmMykrE6znSuf3vL6znD8v38PQXZnL2mEwANu1t5qdvbuf1zVV0+ZVor4dYn4cf3TiNuROGnrrzNeYYLCCM6c2haUBKlzpNUmVvu/0XQOY4Z4TUoUdi7/ffbu/0c8VP36a908+iu87h98t28fOlO0mMjeLa6bnceOYI4nxevvSH1Wzc28ydc0bzz7ML7MZKJuwsIIzpK3+3c6vVXW/Drnecq7w7W5x1xwmMdRWNXPPz94jyCh1dAa4pzuHbn5pIWtCoqI4uP99ZtJEFK8sBmJKTwuwxmSTFRhEIKB6PcN7YLCbnJNvwW3NKWEAY80n5u2HfWjcw3j5uYPz6nTL+uGI3375iIheM773GsXFvE0u2VLN0aw0f7Gn4yJTnABOyk7mhJJec1Dj8AcWvSpTHQ4zPQ2yUlyHJMeSmxdk9wM0Js4Aw5mTxdwUFxtE1jEIYeTaMnO0MqU0d0aeP7PIHUAWPQOtBP4vW7eVPK/ewobL5mPuJQHZyLBdOGMK9FxWSmWjNVab/LCCMCZXDgfGOMy3InuXO7LTgXLSXd5b7mAVDJoCn73/xl9W20nqwmyiv4BWh0x+goytAR5efquYOdte1saO6hcUb9xPr8/LlC8Zw+ZRhNLR1UddykKHJsYenFzGmNxYQxpwqAT/sX+/c82LPMmd4bct+Z11MMuSWwIizYMQMyJkOsSf+A76zpoUfvLyF1zdXfWzd1BGpfP7sfM4alUFpbQs7qlvY29jBwW4/nd0BhibHcuec0XYRYAQLW0CIyDzgx4AX+JWqPtzLdtcCC4EzVXWViOQDm4Gt7ibLVfVLxzqWBYQ5LalC427Y4wZG+ftQvQlQQJxmqZwznLDILYGhk8Hr+0SHWrmrnrLaVjISoklLiGZ9RRO/fW8XpbWtH9nO5xVio7xER3moa+3kiinZ/PimaURZSESksASEiHiBbcDFQAWwErhZVTcdtV0S8BIQDdwVFBB/U9XJfT2eBYQZMDqaoGIVVH4Alauc1221zrqoWOe2rMOLYfh0GD7NCZF+NE0FCwScC/9Ka1oYMySJsUMTGZIUc3iE1K/eLuX7L21mfnEOj1w/Fa9HUFXqWztJi4/G47GRVIPdsQIilHdTmQHsUNVStxALgKuATUdt9z3gv4BvhLAsxpw+YlNgzFznAW4tY48TFpUfOI81f4T3H3fW++Kd0Mie5gRGthsa3uP/7+vxCOcVZnFeYVaP679w7igOdgf4n8VbaWzrRIE1exppau8i1uchPyOBgswE8jMTKMhIYFRWAtPz0iw4IkQoAyIHKA96XwHMDN5ARKYDI1T1JRE5OiAKRGQN0Az8u6q+ffQBROQO4A6AvLy8k1l2Y04dEUgb6TwmX+ssC/ihdhvs/dDpBN/3Iaz5A7z/S2e9Lx6GFbk1jWInODLGfKKaxpcvGEOXP8DPl+6kICOByyYPY8yQRPY3dVBW28rW/Qd4bVMV3e5Y3Ol5qfznNVMYPyy5189sbOtk075mtle1UJSbQnFeWr/LZcIvlE1M1wHzVPUL7vvPATNV9S73vQd4E7hNVXeJyFLg624TUwyQqKp1InIG8AIwSVV7HfdnTUxm0Av4oW4H7F3jBMfeNU54dLc7630JkF3kzGB76JFZ2Oc+jUMX6vWk2x9gb2MH7+2s5b9e3cKBjm7+5dwCSkamo6p0B5Sd1S2sr2xi495mKhvbD+8rArfOyueb88YdvgVsy8FuGlo7iYny4PN6SIqNsj6QMAlXH8Qs4Duqeqn7/n4AVf2B+z4F2Am4g8gZBtQDV6rqqqM+ayluePR2PAsIE5H83U5NY19QYOxfD11tznpvjDO8NrsIhk5xmqqGToLY3v/6P5761k5+8PJm/ry64mPrCjITmJyTwqThyUzITmZUZgK/fqeM37y3ixHpccwencmH5Y1srTpA8E9PQrSXWaMzOX9cFsUjUomO8hDlETISY0iJ+2Sd9qZvwhUQUTid1HOBSpxO6s+o6sZetl/KkRpEFlCvqn4RGQW8DUxR1frejmcBYYwr4Hfml9q/zg2MdU5otDcc2SY51wmOIeMh69DzeIhO6PNhdte10tzejQh4RMhNjyM5tucf8/fL6rn/uXXUtnQybUQqxXmpDE+No8sfoLM7wM6aFpZuraGiof0j+8VEefjW5RP4p1kjbeqREAlLJ7WqdovIXcBinGGuT6rqRhF5EFilqouOsft5wIMi0gUEgC8dKxyMMUE8XucHf8h4KLrBWaYKzXuhaoPzqN4C1ZudCQr9nUf2Tc1zgiJrvFPTGDrJaaaK+vhV2iMz+h4mMwrSeeNrc1DVXn/oVZXS2la2V7XQHQjgDyjPr6nkgUUbeWNLNf9zXRFDk2P79U8BTtPZpn3N5GcmkBgTym7XwcculDMmkvm7oaHMCYuaLc6jegvUbT8SHOJ1OsAP1TayCp3QSB8N0aG9t4Wq8oflu3no5c1EeTx8emo284tzKRmZRmVjO2vKG9lZ3cLoIYlMzU0hLz0eEWeobm1LJ899UMHT7+9hd10bybFR3Hp2PrednU9GP6YlOVaoDQZ2JbUxpn/8XVC3E6o3QtVGt8axCRp24VzkByBBNY5xziNznBMgJ+EK8WA7a1p47M0dvLpxP22dfuJ8Xtq7/B/b7lANoa2z+/AEiDPy07mqeDhvb6vl1Y37ifV5yEmNQxUCqhRkJhweCpwYE0VZbevh0Vub9zWzZf8B0hOi+eal45g3edigCwsLCGPMydHZBvU7nY7x2u1Qs9V5BNc4ABKGOLWOjNGQOfZIcKSO/MQX/YHzw794435W725g3LBkikekMjorkZ01LaytaGTb/gN4PR4SYrwkxkRx4fghjB2adHj/HdUH+N2y3dS1duJxf+g3VDZRdtTV5gBxPi/js5MYPyyZD3Y3sLXqADMK0vmPT03s8xxXzR1drC1vZGZBBtFRp+coLQsIY0xo+budKUVqt7mBscOpgdRth9aaI9t5oyGtwAmO9FGQlu+8Ty9waiOfcJqRE7Wnro23d9TgDzg1ivyMBIanxh2+a2C3P8CCleX86LVtNLZ38cXzRnHPRWN7nW69qb2Lp94t48l3ymju6GZkRjz3XzaBSycNRUTo6PJTXt9GfmZC2OfBsoAwxoRPewPUbIParU6to77UCY+GXUeu4QCnryM1zwmLQ6GRPupIkPjiwnUGhzW1d/HQS5t4dlUFY4ckcteFY6huPsj26gNUNrbT3umnvStAeX0bLQe7uWTiUC6ZNIxf/mMn26tbmJyTTEdXgNKaFgIKmYnRXDk1h/nFOYwdmkis70jgBAJKa2c3iTFRx23W6uwOfOIaigWEMeb0owoH9jud5PVl7nOp+yiDjsaPbp803A0MNzjSRjpNVqkjISHTuSLvFFmytZr7/7Ke/c0dgPNDn5ceT0JMFDFRXjITo/ncrJFMGu40RR2qgTzz/h6yU+KYODyZnNRYlmyp4c0t1XT6AwDE+jwkx/o42B3gQEcXAYW89HiuPyOX60pyyU45EpKd3QEWb9zP0yv2kBATxa9u7fE3/rgsIIwxA097w5GwOFzrcF8HN1uBM5V6eoEzsiq9wAmNtHynRpKcA1HRPR7iRLQc7GZ71QHyMxI+clvZ/mps6+SNzdXsb+6gqb2LprYuYnweUuJ8xEV7eXtbLctK6/CIM7Q4PSGa1DgfaysaqW3pJDctjs+eNZIvnjfqE3WgW0AYYwaXgwecCQ4bdjtNVYdrHjuhqQIC3UEbCyRlO3f4SxkR9DzSCZCU3JAP1z1Re+raeG5NBTtrWmlo7aSutZO89DhunpHHeWOzTmjyRAsIY0zk8HdDc6UTHE3lTpA0lh953Vx5VIAA8RlOUKQEh0iuc8V5So4zKstzeo5COlHhmu7bGGNOPW/UkdlxexLwO30fhwKjqdypdTSWO6Ovdi6BrqOGvXp8kJx9JDCSc9xAyXVeJ+dAfPop7Qc5FSwgjDGRxeN1fuRTcpz7hR9NFdrqobkCmiqdGkdThTNVSXOlc1fA5r0Q6ProflFxkDzcDZBcJ1CSsp1lSdlOiCRkDaiaiAWEMcYEE4GEDOeRPbXnbQIBaK0+EiCHQ8R9Ll0KLVWgR13t7YlyRmMlDw8KDjdIEodC0jDnOSbptKiNWEAYY0x/eTzOj3nSMOCMnrcJ+J3RVs174cC+IzWQ5n3O874PYdurR6ZmDxYVB4lD3NAYConDjhwvcai7bpgzvPcErkw/HgsIY4wJBY83KER6oQoHm53QaNkPB6qc55ZqpwZyYL9zkWHZW869zI8mXicwRs6C65486adgAWGMMeEi4kxsGJvizJZ7LF3tTmi0VDvBcShADux3ahkhYAFhjDEDgS/Onbsq/5QdcuB0pxtjjDmlLCCMMcb0yALCGGNMjywgjDHG9MgCwhhjTI8sIIwxxvTIAsIYY0yPLCCMMcb0aNDcD0JEaoDdJ/ARmUDtSSrOQBGJ5wyRed6ReM4Qmefd33MeqapZPa0YNAFxokRkVW83zRisIvGcITLPOxLPGSLzvE/mOVsTkzHGmB5ZQBhjjOmRBcQRj4e7AGEQiecMkXnekXjOEJnnfdLO2fogjDHG9MhqEMYYY3pkAWGMMaZHER8QIjJPRLaKyA4RuS/c5QkVERkhIktEZJOIbBSRe9zl6SLymohsd5/Twl3Wk01EvCKyRkT+5r4vEJEV7nf+JxGJDncZTzYRSRWRhSKyRUQ2i8iswf5di8hX3P+2N4jIMyISOxi/axF5UkSqRWRD0LIev1tx/MQ9/3UiMr0/x4rogBARL/AYcBkwEbhZRCaGt1Qh0w18TVUnAmcBX3bP9T7gDVUdC7zhvh9s7gE2B73/L+BHqjoGaAD+JSylCq0fA6+q6nhgKs75D9rvWkRygLuBElWdDHiBmxic3/VvgHlHLevtu70MGOs+7gD+rz8HiuiAAGYAO1S1VFU7gQXAVWEuU0io6j5V/cB9fQDnByMH53x/6272W+Dq8JQwNEQkF7gC+JX7XoALgYXuJoPxnFOA84BfA6hqp6o2Msi/a5xbKMeJSBQQD+xjEH7XqvoWUH/U4t6+26uA36ljOZAqItl9PVakB0QOUB70vsJdNqiJSD5QDKwAhqrqPnfVfiA0dz8Pn0eBbwIB930G0Kiq3e77wfidFwA1wFNu09qvRCSBQfxdq2ol8AiwBycYmoDVDP7v+pDevtsT+o2L9ICIOCKSCPwFuFdVm4PXqTPmedCMexaRTwHVqro63GU5xaKA6cD/qWox0MpRzUmD8LtOw/lruQAYDiTw8WaYiHAyv9tID4hKYETQ+1x32aAkIj6ccPijqj7nLq46VOV0n6vDVb4QmA1cKSK7cJoPL8Rpm091myFgcH7nFUCFqq5w3y/ECYzB/F1fBJSpao2qdgHP4Xz/g/27PqS37/aEfuMiPSBWAmPdkQ7ROJ1ai8JcppBw295/DWxW1R8GrVoE3Oq+vhV48VSXLVRU9X5VzVXVfJzv9k1VvQVYAlznbjaozhlAVfcD5SIyzl00F9jEIP6ucZqWzhKRePe/9UPnPKi/6yC9fbeLgH9yRzOdBTQFNUUdV8RfSS0il+O0U3uBJ1X1oTAXKSRE5BzgbWA9R9rjv4XTD/EskIczXfoNqnp0B9iAJyJzgK+r6qdEZBROjSIdWAN8VlUPhrN8J5uITMPpmI8GSoHP4/xBOGi/axH5LnAjzoi9NcAXcNrbB9V3LSLPAHNwpvWuAh4AXqCH79YNy5/hNLe1AZ9X1VV9PlakB4QxxpieRXoTkzHGmF5YQBhjjOmRBYQxxpgeWUAYY4zpkQWEMcaYHllAGNMPIuIXkQ+DHidtwjsRyQ+eodOYcIs6/ibGmCDtqjot3IUw5lSwGoQxJ4GI7BKR/xaR9SLyvoiMcZfni8ib7lz8b4hInrt8qIg8LyJr3cfZ7kd5ReQJ974GfxeRuLCdlIl4FhDG9E/cUU1MNwata1LVKThXrj7qLvsp8FtVLQL+CPzEXf4T4B+qOhVnnqSN7vKxwGOqOgloBK4N8fkY0yu7ktqYfhCRFlVN7GH5LuBCVS11J0Xcr6oZIlILZKtql7t8n6pmikgNkBs87YM7Dftr7k1fEJF/A3yq+v3Qn5kxH2c1CGNOHu3ldX8EzxPkx/oJTRhZQBhz8twY9LzMff0ezkyyALfgTJgIzm0h74TD98xOOVWFNKav7K8TY/onTkQ+DHr/qqoeGuqaJiLrcGoBN7vL/hXnzm7fwLnL2+fd5fcAj4vIv+DUFO7EuROaMacN64Mw5iRw+yBKVLU23GUx5mSxJiZjjDE9shqEMcaYHlkNwhhjTI8sIIwxxvTIAsIYY0yPLCCMMcb0yALCGGNMj/5/LckR/fCqYjwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotting Train Loss vs Validation Loss\n",
    "plt.plot(history_3.history['loss'])\n",
    "plt.plot(history_3.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KBsmjEVwQ1Jz"
   },
   "source": [
    "The curve is smooth and no overfitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "vEmnRQXgIDD2",
    "outputId": "93b7d1b2-d53b-4ba8-f95e-1d50b398fa70"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold=0.240535, G-Mean=0.738\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# predict probabilities\n",
    "yhat3 = model3.predict(X_test)\n",
    "# keep probabilities for the positive outcome only\n",
    "yhat3 = yhat3[:, 0]\n",
    "# calculate roc curves\n",
    "fpr3, tpr3, thresholds3 = roc_curve(y_test, yhat3)\n",
    "# calculate the g-mean for each threshold\n",
    "gmeans3 = np.sqrt(tpr3 * (1-fpr3))\n",
    "# locate the index of the largest g-mean\n",
    "ix3 = np.argmax(gmeans3)\n",
    "print('Best Threshold=%f, G-Mean=%.3f' % (thresholds3[ix3], gmeans3[ix3]))\n",
    "# plot the roc curve for the model\n",
    "pyplot.plot([0,1], [0,1], linestyle='--', label='No Skill')\n",
    "pyplot.plot(fpr3, tpr3, marker='.')\n",
    "pyplot.scatter(fpr3[ix3], tpr3[ix3], marker='o', color='black', label='Best')\n",
    "# axis labels\n",
    "pyplot.xlabel('False Positive Rate')\n",
    "pyplot.ylabel('True Positive Rate')\n",
    "pyplot.legend()\n",
    "# show the plot\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "lVNhYY9MIG3m",
    "outputId": "a7c60315-ed6f-4de4-8860-e5c4d917e587"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False],\n",
       "       [False],\n",
       "       [False],\n",
       "       ...,\n",
       "       [ True],\n",
       "       [False],\n",
       "       [False]])"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_e3=model3.predict(X_test)\n",
    "y_pred_e3 = (y_pred_e3 > thresholds3[ix3])\n",
    "y_pred_e3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "_2pu2TKhIJRW",
    "outputId": "c42f459c-4350-47ab-a479-95bdbb98b788"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Calculating the confusion matrix \n",
    "cm3=confusion_matrix(y_test, y_pred_e3)\n",
    "labels = ['True Positive','False Negative','False Positive','True Negative']\n",
    "categories = ['Not Churning','Churning']\n",
    "make_confusion_matrix(cm2, \n",
    "                      group_names=labels,\n",
    "                      categories=categories, \n",
    "                      cmap='Blues')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ft_stmDkIORX",
    "outputId": "59beb95d-e912-4e3c-bdc6-5d3ef6689a47"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.91      0.78      0.84      1593\n",
      "           1       0.44      0.70      0.54       407\n",
      "\n",
      "    accuracy                           0.76      2000\n",
      "   macro avg       0.68      0.74      0.69      2000\n",
      "weighted avg       0.82      0.76      0.78      2000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Accuracy as per the classification report \n",
    "cr3=metrics.classification_report(y_test,y_pred_e3)\n",
    "print(cr3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "AU4AF-y2RsXT"
   },
   "source": [
    "The validation loss was smooth for this model, False negative percentage slightly improved, overall model performance slightly dropped."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UrQcL7LHjfBk"
   },
   "source": [
    "### Model 4\n",
    "(Random Search)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "uH4fJN1Hk8wB"
   },
   "source": [
    "In this model will use random search to optimize batch size and learning rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "id": "F4SsIzudjn7H"
   },
   "outputs": [],
   "source": [
    "backend.clear_session()\n",
    "np.random.seed(1)\n",
    "random.seed(1)\n",
    "tf.random.set_seed(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "id": "JfZgt6_CjvU_"
   },
   "outputs": [],
   "source": [
    "def cr_model4(lr,batch_size):\n",
    "  model4 = Sequential()\n",
    "  model4.add(Dense(256,activation='relu',input_dim = X_train.shape[1]))\n",
    "  model4.add(Dense(128,activation='relu'))\n",
    "  model4.add(Dense(64,activation='relu'))\n",
    "  model4.add(Dense(32,activation='relu')) \n",
    "  model4.add(Dense(1, activation='sigmoid'))\n",
    "\n",
    "  optimizer4 = tf.keras.optimizers.SGD(learning_rate=lr)\n",
    "  model4.compile(optimizer = optimizer4,loss = 'binary_crossentropy', metrics = [tf.keras.metrics.Recall()])\n",
    "  return model4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "id": "iC7cmIqQwlmb"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.model_selection import RandomizedSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "id": "OchfthHxlLhk"
   },
   "outputs": [],
   "source": [
    "keras_estimator = KerasClassifier(build_fn=cr_model4, verbose=1)\n",
    "# define the grid search parameters\n",
    "param_random = {\n",
    "    'batch_size':[32, 64, 128],\n",
    "    \"lr\":[0.01,0.1,0.001],}\n",
    "\n",
    "kfold_splits = 3\n",
    "random4= RandomizedSearchCV(estimator=keras_estimator,  \n",
    "                    verbose=1,\n",
    "                    cv=kfold_splits,  \n",
    "                    param_distributions=param_random,n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "RcvdPAmqlQtA",
    "outputId": "1222501f-2465-4361-f10a-a1e15535d67e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 9 candidates, totalling 27 fits\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.5323 - recall: 0.0000e+00 - val_loss: 0.4884 - val_recall: 0.0000e+00\n",
      "Best: nan using {'lr': 0.01, 'batch_size': 32}\n"
     ]
    }
   ],
   "source": [
    "random_result = random4.fit(X_train, y_train,validation_split=0.2,verbose=1) \n",
    "\n",
    "# Summarize results\n",
    "print(\"Best: %f using %s\" % (random_result.best_score_, random_result.best_params_))\n",
    "means = random_result.cv_results_['mean_test_score']\n",
    "stds = random_result.cv_results_['std_test_score']\n",
    "params = random_result.cv_results_['params']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "YEN2lryTlyjt",
    "outputId": "ce43cf1f-c132-4bce-f51e-3055ff764a7b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense_5 (Dense)             (None, 256)               3072      \n",
      "                                                                 \n",
      " dense_6 (Dense)             (None, 128)               32896     \n",
      "                                                                 \n",
      " dense_7 (Dense)             (None, 64)                8256      \n",
      "                                                                 \n",
      " dense_8 (Dense)             (None, 32)                2080      \n",
      "                                                                 \n",
      " dense_9 (Dense)             (None, 1)                 33        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 46,337\n",
      "Trainable params: 46,337\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "estimator_v4=cr_model4(batch_size=random_result.best_params_['batch_size'],lr=random_result.best_params_['lr'])\n",
    "estimator_v4.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8-QzX51nl1NU",
    "outputId": "7a81a995-7a09-4043-8f88-f40a38b021c7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "200/200 [==============================] - 1s 3ms/step - loss: 0.5191 - accuracy: 0.7977 - val_loss: 0.4848 - val_accuracy: 0.7906\n",
      "Epoch 2/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4612 - accuracy: 0.7977 - val_loss: 0.4582 - val_accuracy: 0.7906\n",
      "Epoch 3/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4397 - accuracy: 0.7977 - val_loss: 0.4410 - val_accuracy: 0.7906\n",
      "Epoch 4/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4258 - accuracy: 0.7977 - val_loss: 0.4302 - val_accuracy: 0.7906\n",
      "Epoch 5/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4157 - accuracy: 0.7987 - val_loss: 0.4201 - val_accuracy: 0.7931\n",
      "Epoch 6/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.4059 - accuracy: 0.8080 - val_loss: 0.4104 - val_accuracy: 0.8169\n",
      "Epoch 7/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3964 - accuracy: 0.8250 - val_loss: 0.4020 - val_accuracy: 0.8275\n",
      "Epoch 8/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3872 - accuracy: 0.8364 - val_loss: 0.3911 - val_accuracy: 0.8388\n",
      "Epoch 9/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3773 - accuracy: 0.8450 - val_loss: 0.3817 - val_accuracy: 0.8469\n",
      "Epoch 10/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3687 - accuracy: 0.8487 - val_loss: 0.3744 - val_accuracy: 0.8506\n",
      "Epoch 11/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3612 - accuracy: 0.8498 - val_loss: 0.3677 - val_accuracy: 0.8537\n",
      "Epoch 12/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3552 - accuracy: 0.8552 - val_loss: 0.3628 - val_accuracy: 0.8512\n",
      "Epoch 13/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3502 - accuracy: 0.8542 - val_loss: 0.3586 - val_accuracy: 0.8619\n",
      "Epoch 14/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3457 - accuracy: 0.8550 - val_loss: 0.3555 - val_accuracy: 0.8594\n",
      "Epoch 15/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3434 - accuracy: 0.8577 - val_loss: 0.3524 - val_accuracy: 0.8581\n",
      "Epoch 16/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3401 - accuracy: 0.8570 - val_loss: 0.3508 - val_accuracy: 0.8594\n",
      "Epoch 17/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3376 - accuracy: 0.8586 - val_loss: 0.3517 - val_accuracy: 0.8587\n",
      "Epoch 18/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3355 - accuracy: 0.8606 - val_loss: 0.3487 - val_accuracy: 0.8581\n",
      "Epoch 19/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3332 - accuracy: 0.8598 - val_loss: 0.3487 - val_accuracy: 0.8569\n",
      "Epoch 20/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3318 - accuracy: 0.8636 - val_loss: 0.3471 - val_accuracy: 0.8600\n",
      "Epoch 21/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3294 - accuracy: 0.8631 - val_loss: 0.3504 - val_accuracy: 0.8581\n",
      "Epoch 22/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3285 - accuracy: 0.8634 - val_loss: 0.3476 - val_accuracy: 0.8612\n",
      "Epoch 23/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3267 - accuracy: 0.8622 - val_loss: 0.3485 - val_accuracy: 0.8594\n",
      "Epoch 24/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3245 - accuracy: 0.8634 - val_loss: 0.3465 - val_accuracy: 0.8575\n",
      "Epoch 25/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3234 - accuracy: 0.8661 - val_loss: 0.3486 - val_accuracy: 0.8562\n",
      "Epoch 26/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3226 - accuracy: 0.8661 - val_loss: 0.3465 - val_accuracy: 0.8581\n",
      "Epoch 27/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3216 - accuracy: 0.8684 - val_loss: 0.3477 - val_accuracy: 0.8556\n",
      "Epoch 28/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3201 - accuracy: 0.8681 - val_loss: 0.3464 - val_accuracy: 0.8575\n",
      "Epoch 29/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3190 - accuracy: 0.8664 - val_loss: 0.3436 - val_accuracy: 0.8569\n",
      "Epoch 30/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3179 - accuracy: 0.8691 - val_loss: 0.3451 - val_accuracy: 0.8594\n",
      "Epoch 31/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3171 - accuracy: 0.8683 - val_loss: 0.3472 - val_accuracy: 0.8550\n",
      "Epoch 32/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3155 - accuracy: 0.8675 - val_loss: 0.3463 - val_accuracy: 0.8600\n",
      "Epoch 33/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3143 - accuracy: 0.8656 - val_loss: 0.3461 - val_accuracy: 0.8550\n",
      "Epoch 34/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3137 - accuracy: 0.8695 - val_loss: 0.3435 - val_accuracy: 0.8581\n",
      "Epoch 35/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3113 - accuracy: 0.8678 - val_loss: 0.3517 - val_accuracy: 0.8512\n",
      "Epoch 36/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3123 - accuracy: 0.8691 - val_loss: 0.3460 - val_accuracy: 0.8594\n",
      "Epoch 37/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3107 - accuracy: 0.8683 - val_loss: 0.3461 - val_accuracy: 0.8556\n",
      "Epoch 38/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3087 - accuracy: 0.8686 - val_loss: 0.3493 - val_accuracy: 0.8537\n",
      "Epoch 39/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3089 - accuracy: 0.8705 - val_loss: 0.3450 - val_accuracy: 0.8569\n",
      "Epoch 40/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3080 - accuracy: 0.8694 - val_loss: 0.3438 - val_accuracy: 0.8587\n",
      "Epoch 41/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3064 - accuracy: 0.8719 - val_loss: 0.3487 - val_accuracy: 0.8556\n",
      "Epoch 42/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3059 - accuracy: 0.8695 - val_loss: 0.3432 - val_accuracy: 0.8525\n",
      "Epoch 43/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3039 - accuracy: 0.8730 - val_loss: 0.3559 - val_accuracy: 0.8581\n",
      "Epoch 44/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3036 - accuracy: 0.8709 - val_loss: 0.3488 - val_accuracy: 0.8569\n",
      "Epoch 45/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3018 - accuracy: 0.8725 - val_loss: 0.3477 - val_accuracy: 0.8594\n",
      "Epoch 46/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3027 - accuracy: 0.8706 - val_loss: 0.3476 - val_accuracy: 0.8569\n",
      "Epoch 47/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3002 - accuracy: 0.8733 - val_loss: 0.3503 - val_accuracy: 0.8550\n",
      "Epoch 48/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.3007 - accuracy: 0.8750 - val_loss: 0.3495 - val_accuracy: 0.8537\n",
      "Epoch 49/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.2988 - accuracy: 0.8731 - val_loss: 0.3491 - val_accuracy: 0.8544\n",
      "Epoch 50/50\n",
      "200/200 [==============================] - 0s 2ms/step - loss: 0.2982 - accuracy: 0.8739 - val_loss: 0.3461 - val_accuracy: 0.8587\n"
     ]
    }
   ],
   "source": [
    "optimizer = tf.keras.optimizers.SGD(random_result.best_params_['lr'])\n",
    "estimator_v4.compile(loss='binary_crossentropy',optimizer=optimizer,metrics=['accuracy'])\n",
    "history_4=estimator_v4.fit(X_train, y_train, epochs=50, batch_size=random_result.best_params_['batch_size'], verbose=1,validation_split=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "_Fr0ufvCl3v0",
    "outputId": "780fd4d9-4154-4fef-ac97-706101034914"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotting Train Loss vs Validation Loss\n",
    "plt.plot(history_4.history['loss'])\n",
    "plt.plot(history_4.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RxpQZ5bFRpO8"
   },
   "source": [
    "Noise and overfitting is visible on the trend"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "o4hGttctl6Hc",
    "outputId": "25d8f1b3-10f5-4225-81ca-1d50a2b2bc45"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold=0.172570, G-Mean=0.775\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# predict probabilities\n",
    "yhat4 = estimator_v4.predict(X_test)\n",
    "# keep probabilities for the positive outcome only\n",
    "yhat4 = yhat4[:, 0]\n",
    "# calculate roc curves\n",
    "fpr4, tpr4, thresholds4 = roc_curve(y_test, yhat4)\n",
    "# calculate the g-mean for each threshold\n",
    "gmeans4 = np.sqrt(tpr4 * (1-fpr4))\n",
    "# locate the index of the largest g-mean\n",
    "ix4 = np.argmax(gmeans4)\n",
    "print('Best Threshold=%f, G-Mean=%.3f' % (thresholds4[ix4], gmeans4[ix4]))\n",
    "# plot the roc curve for the model\n",
    "pyplot.plot([0,1], [0,1], linestyle='--', label='No Skill')\n",
    "pyplot.plot(fpr4, tpr4, marker='.')\n",
    "pyplot.scatter(fpr4[ix4], tpr4[ix4], marker='o', color='black', label='Best')\n",
    "# axis labels\n",
    "pyplot.xlabel('False Positive Rate')\n",
    "pyplot.ylabel('True Positive Rate')\n",
    "pyplot.legend()\n",
    "# show the plot\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Wm1HlG53tpZh",
    "outputId": "07f48d8d-164d-4c36-cb65-c81c23b81413"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False],\n",
       "       [False],\n",
       "       [False],\n",
       "       ...,\n",
       "       [ True],\n",
       "       [False],\n",
       "       [ True]])"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_e4=estimator_v4.predict(X_test)\n",
    "y_pred_e4 = (y_pred_e4 > thresholds4[ix])\n",
    "y_pred_e4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 0
    },
    "id": "_J36o_gYt25y",
    "outputId": "71fd6a0d-c7e7-4daa-8f0c-d958bfe502d4"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWYAAAD4CAYAAADfPUyRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd3wURRvA8d+TBEgCBEKRrvQuNfQqHQRROtJEBFFBEBCliKAoKqLYaBakKR1BQJAqHem89N47wQRIvWTeP+4SEki5hCRcLs/Xz35yOzu7M5vgk8nszKwYY1BKKeU4XJ50BZRSSkWngVkppRyMBmallHIwGpiVUsrBaGBWSikH45bcBXhU7KfDPtQjZkwf/qSroBxQhwp55XGvkZCYE7jv+8cuLzloi1kppRxMsreYlVIqRUnqb29qYFZKORcX1yddg8emgVkp5VzEIbuNE0QDs1LKuWhXhlJKORhtMSullIPRFrNSSjkYbTErpZSD0VEZSinlYLQrQymlHIx2ZSillIPRFrNSSjkYDcxKKeVgXPXhn1JKORbtY1ZKKQejXRlKKeVgtMWslFIORlvMSinlYLTFrJRSDkanZCullIPRrgyllHIw2pWhlFIOJi20mEWkUgzJfsB5Y4wl6auklFKPIS0EZmASUAk4CAhQFjgMZBGRN4wxfydj/ZRSKmGc4OGfPb9argAVjTE+xpjKQEXgDNAY+CI5K6eUUgkmYv/moOwJzMWNMYcjdowxR4CSxpgzyVctpZRKJHGxf4vvUiK/iMgNETkUJS2biKwRkZO2r962dBGRb0XklIgcjNoNLCI9bPlPikiP+Mq1JzAfFpHJIlLPtk0CjohIBiDUjvOVUirlJG2L+Veg2UNp7wPrjDHFgHW2fYDmQDHb1geYbK2OZAM+BKoBVYEPI4J5bOwJzK8Ap4CBtu2MLS0UeM6O85VSKsWIiN1bfIwxmwDfh5JbAzNsn2cAL0ZJn2msdgBZRSQP0BRYY4zxNcbcAdbwaLCPJt6Hf8aYQGCCbXvYvfjOV0qplGRPwH1MuYwxV22frwG5bJ/zARej5LtkS4stPVb2DJerBYwGnoma3xhTOL5zlVIqpYmL/YFZRPpg7XaIMM0YM83e840xRkRMAqpnF3uGy/0MvAPsAcKSugJKKZWUEtJitgVhuwOxzXURyWOMuWrrqrhhS78MFIiSL78t7TJQ/6H0jXEVYE8fs58x5i9jzA1jzO2Izd47UEqplJSUfcyxWAZEjKzoASyNkt7dNjqjOtbYeRVYDTQREW/bQ78mtrRY2dNi3iAi44HFQHBEojFmb4JuRSmlUkBS9jGLyO9YW7s5ROQS1tEVnwHzRaQXcB7oYMu+EmiBdbBEANATwBjjKyIfA7ts+T4yxjz8QDEaewJzNdtXnyhpBmhgx7lKKZWykvDZnzGmcyyHGsaQ1wBvxXKdX4Bf7C3XnlEZOiROKZVqpMCojGQXa2AWka7GmNkiMiim48aYr5KvWkoplTguLs69iFFG29fMKVERpZRKCk7dYjbGTLV9HZNy1Ul+2bJkZOXU/gDkyu5FeHg4N+9Y58nU6TqeUMvjjwhc/eMAcufwIigklPsBwbw+eg4nz9+I/8Qolnz3Bq8M+xWAjs19mLZgMwB5cmZhwtB2vPzuz49dz7RqVKeG5Hq6UOT+y0PG4v1U7hjzfty9OR/M/Ouxyls86TNOHdzDoO/m4JYuPff9/Zgy/HUGfz/3sa77sCO7tpAjT36eyl8QgHXzf6FgyfIUKVc5SctxeKk/Lts1wSQn0BsoSPQJJq8mX7WSj6/ffap3+gyAEa+34H5AMBNnrYs87urqQlhY+GOX03PEDPYeucCrbWrx6Tsv0X7g1ASd/1L/yQA8nScbfTrUiQzMV2/6aVB+TOnSp+etL35K0TJdXFzYu+EvqjZpnWxlHNu1heKVakQG5oYdUuX/oo/NqVvMUSwFNgNrcdIJJtPGdCUoxEKFEvnZfuAM/veCogXs3QuG0+btKVy46kunFlV4q3M90qVzY9f/zjFg3DzCw2Of+LNl7yn6dakPwKcDX6RJrdIYA5//tIqFf+8ldw4vZn3+KpkzuuPm6sKAT+exdd9pjq0YQ60uXzB2QGsK58/Bjrnvs37HMabM28Tib/vi0/5T/pkxmL5j5nD0zDXA2lIf9vUSjp25xlfvtadM0Ty4ubnyydSVLN/4v2T/PqZWwUGB/DZ+BIH37hEWZqFRx1cpVaV2tDx379xm3sSPCA68T3hYGK1ee4eCpcpx6sAu1i/4FUtoKNly5eWlN98jg7vHI2XUaNGWbSsXUrlhy0eObVk2l0PbN2KxhFKqSm0adugJwIZFMzmweS0ZvbKQJftT5C1cnNqtOrJ73XJ2r12OxRJK9tz5aNtvONfOneLY7m2cPXKAfxbPovOgMWxcPIvilWqQwd2DPetX0mnQaADOHt7PluXz6PbeOLvrn5qklcDsaYx5L9lr8oTleyor9V+ZQHi4YcTrLWLMU6JQLto1qcRzPb/CYgln4rAOdGpRhd+W/xvrdZ+vW5bDJ6/wYsMKlCuRn6odx5Ejaya2zH6XLXtP0bG5D2u2HeWLn1fj4iJ4uqePdv7Ib5ZSukieyFb+03myRR5b+Pde2japxNgpK8mdw4vcObzYe+QCY/q1YuOuE/QdM4csmTzYPPtd1u84TkBQSBJ8p1K/0JAQfhj6GgDeT+Wh4zuj6Tz4Y9w9M3Lf349pI9+kpE+taP+DH9yyjqLlq1C/TVfCw8MIDQ7mvr8fGxfP5pWRX5Le3YNNS39n2/L5PNfu0VUds+TIxTMlynJg09+UqFwzMv3UgV3cvnaJ1z+djDGGOeNHcO7IAdzSZ+DIzk289cVPhIdZmPR+H/IWLg5A6ap18LEF+LVzf2bv+pVUb96Gkj41KV6pBmWr14tWduFnK7N02gRCggJJ7+7B/7Zv4NmaDRJU/9QkIVOyHZU9gXm5iLQwxqxM9to8QYvX7ouz5QvwXNUSVCr9NFtmDwXAI0M6bvrGvI7T9E96EBgcyoUrtxn0+QLe7tqA+at2Ex5uuOF7l817TlG5zDPsPnyeqR92JZ2bK39uOMDBE5ftrvOiv/fy5+R+jJ2ykraNK7Fk7X4AGtYoxfP1nmVgd+tQS/f0bhTI483xs9ftvrYze7grI8xiYe3cnzh39CAigr/vLe753SFz1ge/BPMVKcGSKV8QHmahVJXa5ClYlHNHtnPz8jl+HNU/8joFipeOtdy6L3ZhzviRFK9UPTLt1MHdnDq4m0nv9QYgJCiQ29cuERwYSEmfWqRLnx5IT8nKNSLPuX7xLOvm/ULg/XuEBAVSrHyVOO/X1dWVYhWqcGzPdspUr8eJvTto2uV1zh05kKD6pxZppcU8ABguIsFYl/oUrGOpvZK1ZiksIDByUiOWsDBcovzWdU+fDrD+wGf/uZNR3y2L93oRfczx2br3NI1fm0iz2mWY9lE3vp29Ps4WeFRXbvrh63efssXy0q5pJfp/Yn2YJEDnIT8l+IFjWnVgy1ru+//HG+Om4urmxoR+nbCERP/romDp8vQa/Q0n9u1g8aTPqPl8ezwyZabIsz50GPCBXeVkz5Of3AWLcGj7xsg0Ywx1W79MlcYvRMu7bcXCWK+zeNLnvDzkY/IULMrejas4d2R/vGU/W7MBO1b9gWemzOQtXIIMHp4YTILqn1o4Q2COc8CfiLgAzYwxLsYYD2OMlzEms7MF5Yedv+JLhVLWtUgqlMxPwXzZAdjw73FealSBnN6ZAPD28uTpPHGudx1p677TtGtSGRcXIYd3JmpXLsruQ+d4Oo8312/7M33JNn5dso2KJQtEO+9eQDCZM7rHet2Fq/cyqEcjvDJ5cOjkFQDWbj/Km50e/DlbvkR++28+DQoOuEdGL29c3dw4c2gf/9189C+L/25eI1NWb3watqRyg+e5evYkBYqV5sLxQ9y+Zv0rJyQokFtXLj5yblT1XurK1uXzIveLla/C3o1/ERwUCIC/703u+d3h6RJlOb5nO6EhIQQHBXJ8747Ic0KCAsjsnZ0wi4WDW9ZGpqd39yQkMCDGcguWLs/VcyfYvW4Fz9ayzhlLTP1TgxRYKyPZxdliNsaEi8j3WN/zl2b8sW4/XVpWZc/CEez637nIluexM9cY88Ny/pzcDxcRQi1hvPPZfC5cvRPvNZeuP0C1coX4d94wjIERE//g+u27dGlVjXe6NyTUEsb9gGB6fTAr2nm+fvfZvv8MuxcM5++tR5gyb1O040vW7uPLd9sy7sdVkWnjflzF+CFt2TV/OC4uwrnLt2k7YEoSfGecU7najZnzxXC+G/Iq+QqXIEfepx/Jc/bIfrYsm4eLmxsZ3D1o+9YwMnplpc2b77Hgm4+xWKwv82nY8VVy5C3wyPkRchUoRJ6Cxbl67gQARctX4ebl80wbaZ3Jm8Hdg3b9hpO/aElK+tTkh6G9yJTFm1wFCuHuaZ1a0LDDq0wd8SYZvbKSv2gpQoKswfjZms+xdNoEtq9aTOd3Rkcr18XFlRKVarBv4yravGV94UZi6p8aOHLAtZdYp3fHkUHkS2A7sNjElzkGHhX7JflapSr1mzF9+JOugsMLDgokg7sHIcFB/Dx6AK17D458AOisOlTI+9hRNW/fxXbHnCtT2jhkFLenj/l1YBBgEZEgnLSPWSlHs3Tal9y8dB5LaAgV6jZ1+qCcVJx9SjYAxhidkq3UE9Dhbed6KJdSnKErw56Zf3VjSre9pFAppRxL6o/Ldr3B5N0o2wfAn1jfAaiAKR924fy6cexe8KDPtE2jiuxZOIL7e76lUulHHyQVyO3Nza0TGNjtwZKub3Wuz+4Fw9mzcAT9Xq6fElVXKSg8PIwf3uvNrM+HRUtfMf1bPu7e/JH8h3f+wwcdn+Py6eMpVUWn4QyjMuINzMaYVlG2xkBZIP5hCGnErD930PqtH6KlHT59hU6Df2TL3tMxnvP54Db8vfVw5H7pInno2aYmdbqNp2rHcTSvW5bCBXIka71Vytq+chE580X/JX359HEC7z86QSk4MIDtKxeTv2iplKqeU0kTgTkGlwD9F2Ozde9pfP2ijx09fvZ6rJM7WtUvx7nLtzly+lpkWslCudl16ByBQaGEhYWzec8pXmxQIVnrrVKO3+2bnNi3A58Gz0emhYeHsXr2FJp2ef2R/Ovm/UKd1p1wS5/+kWMqfmkiMIvIdyLyrW37HuuCRvq+v0TI6JGewT0b88nU6LPbD5++Qq2KRcmWJSMe7uloVrsM+XPbN3FFOb6VM76nSZfXsc7Xstq5agklfWqS2Tt7tLxXzpzA7/YNSlSq8fBllJ3ERezeHJU9w+V2R/lsAX43xmyN6wQR6QP0AXDLXx+3HGUSX0MnMrLv83w3ez33A6NP9z1+9joTfl3Dn5PeIiAohAPHLyXJ0qPqyTu+ZzuZvLKSr3AJzh62Tp32973FoR3/8OqHE6PlDQ8P569Zk2jzxvtPoqpOw5FbwvayZ7jcjIRe1BgzDZgGOsEkqipln+GlRhX4ZOCLZMnsQXi4ISgklCnzNjHjj+3M+GM7AGP6teLy9f+ecG1VUjh//BDH9mzjxP6dWEJCCA4M4LshPXFzS8fEAV0ACA0J5uu3u/DGZ1O5cfEsv3w0EIB7//kyZ/wIurz7CfmKlHiSt5GqpInALCK1sI7CeMaWP2KCSeHkrZrzadTrQQspYpH+iCnWOb0zcfPOPQrk9qZ1g/LU6z7hSVVTJaEmL/emycvWleOiroMc1cfdm/POt3MAGPbT0sj0n8cMpFnXNzQoJ5ATxGW7ujJ+Bt4B9uCkC+U/jhnjXqFO5WLkyJqJU6s+5uMpK7njd5+v3mtPDu9MLP62LwePX+aFh0ZuPOz3L18jW9aMhFrCGPjZfPzuBabQHSjlXJyhxWzPWhk7jTHVEluAdmWomOhaGSomSbFWRon3Vtsdc45/3tQho3isLWYRqWT7uEFExgOLgchFi40xOjJDKeVwnKDBHGdXxsOdnD5RPhugQdJXRymlHo+LAw+Ds1esgdkY81xKVsTRHVsxhrv3gwkLD8cSFk7tLl8w67OeFCuYC4CsmT34725g5Lv5HubiImydM5QrN/wi10aeNqYrdSoXxe9eEAB9Rs3i4InLvNiwAh+88Tx3/O7TYdCP+Prdp1D+HHzUrxXd3p+eMjes4uV36waLfhjHPb87IFClYUtqtGjHqtlTOL5nG65u6awvOH3jPTwyZrLrXICAe/7Mn/gRd25ewztnbjoO/BCPTJk5vPMf1s2fjmcmL14e8jGembPge+0ya+b+RMeBHz6Jb4FDcuoWs4gMAvyMMT8/lN4LyGyMmRjzmc6rWZ9vuP3f/cj9qEHys0EvxfnArt/Lz3H87PVH3kYyfOIfke/qi/BGp3rU7voFrRtUoGNzHybP/YfRb7Vk9KTlSXQnKim4uLrSrNsb5C1cnODAACYPe50i5Xwo+mxlGnfujaurK6vnTGXTH3MemeEX27lP5S/I5j9+o3DZStR98WU2/fEbm5b+RtMur7Nj1RL6fjqFI/9u5uCWdVRv3oa1836hUcdeT+g74Jic4eFfXDP/ugAzY0ifBbyaPNVJvdo2rsT8VXtiPJbvqaw0q12G6Uu22XWt8PBwMqRzw9M9PaGWMGpVLML1W/6cvnAzKausHlNm7+yRayRn8PAkZ76n8fe9RdHyVXB1dQWsr2/yv/3ozy22cwGO7t5GxXpNAahYrylHd1nnc4m4EBYaSmhwEC5ubpw7epBMWbORPY++OiwqEfs3RxVXYHYzxoQ+nGiMCcEpFtZLGGMMf07qx9Y5Q3m1Ta1ox2pVKsJ137uxBs7x77ZlxDd/xPgW7tFvteLfecP4YnAb0qez/gEz/pc1rJjSnxZ1yzJ/1W7e790s2qujlOO5c+MaV8+eemThob0b/qJYxbgHNT187n0/38ip2pmyZuO+ny8AdV98mV/HDub4nu2Uq9WAjYtnUb9tt2S4m9TNxcXF7s1RxfXwz0VEchljor2ZUkRyJXOdHFLDnl9z5aYfOb0zsXxKP46fu8ZW2+pxHZr5sGDV7hjPa16nLDd877Lv6EXqVC4W7dio75Zx7ZY/6dO58cMHnRncsxHjpq1i/c5jrO9yDICXW1Zl9ZbDFHvmKQZ2b8gd/wCGjF9IYNAjvzPVExIcFMjcr0bRvMdbke/lA9i4eDYurq6Ur90owedGkChNu6LlfChazvoMft8/qyleoRq3r1xi6fJ5eGTMTItX+pE+Q+wv7k0rHLklbK+4fmWMB1aISD0RyWzb6gPLgS9TpHYO5MpNPwBu3rnHsvUHqVKmIACuri60blCehatjHj1Yo0JhWtZ7lmMrxjDzs57Ur1KcX8Z2B+DaLX8AQkItzFy6Ax/bNSN4uKejW6tqTJm/iZF9n+e1D2axbf8ZOjWvkjw3qRIszGJh7oRRlKvdiDLVHrxTYu/GVZzYu512/UfE2ucZ27kZs2Tj7p3bANy9c5uMXtEXtAoJDmLfP6up1vRF1i+YTts33+eZkmWjvTE7LXPq1eWMMTOxLoz/EXAOOAuMAUYlZv2M1MzTPT2ZPDNEfm5UoySHT18BoEG1Epw4d53LN2Je22LUd8so2uwDSj7/Id3fn87GXSd4daS16z53jgevTXzhuXIcsV0zwjvdGzHp93+wWMLxcE+HwRAeHo6nuy4H6QiMMSyZ8gU58z1DrZYdItNP7v+XLcvm0mXoJ7G2YGM7F6CkT032/bMasLaMS/nUjHZ8y7J5VG/eBlc3N0JDQkAEERdCg4OS+A5Tp6TsYxaRd0TksIgcEpHfRcRdRAqJyE4ROSUi80QkvS1vBtv+Kdvxgom9hzinZBtj/gL+SuzFncVT2TMz7yvregdurq7M+2s3a7YdBaB908qPPPTLkzMLk0a9zEv9J8d53emf9CCHd2ZE4ODxS/T/ZG60a/iUfYZPp1m//ZN//4cts4fidzeADoN+TMrbU4l04fghDmxeQ66nC/PD0NcAaNz5NVZM/w6LJZRfxw4BrA8AX+g9CH/fW/wx9Uu6D/ss1nOLV6xO3dadmTdxDHs2rCRrjlx0fOfBUDh/31tcPn2UBu17AFC92UtMGd4XD89MvPzu2BT+DjimpGoJi0g+4G2gtDEmUETmA52AFsDXxpi5IjIF6AVMtn29Y4wpKiKdgM+BjokqO74p2Y9Lp2SrmOiUbBWTpJiS7TN2g90xZ/fI52ItzxaYdwDlAX/gD+A7YA6Q2xhjEZEawGhjTFMRWW37vF1E3IBrQE6TiCDruI8llVIqEVxcxO5NRPqIyO4oW5+I6xhjLmN9nnYBuAr4YV3M7T9jjMWW7RKQz/Y5H3DRdq7Flj/6mxDsZM+yn4WMMWfjS1NKKUeQkK6MqGvHx3Adb6A1UAj4D1gANEuCKsbLnhbzohjSFiZ1RZRSKikk4cO/RsBZY8xN25yOxUAtIKutqwIgP3DZ9vkyUMBaB3EDsgC3E3MPcU3JLgmUAbKISJsoh7wAHSyplHJISTgM7gJQXUQ8gUCgIdZX7W0A2gFzgR5AxNsNltn2t9uOr09M/zLE3ZVRAmgJZAVaRUm/C/ROTGFKKZXckiouG2N2ishCrC+ftgD7sHZ7rADmishYW1rEekI/A7NE5BTgi3UER6LEtbrcUmCpiNQwxmxPbAFKKZWSknLZT2PMh8DDS/edAarGkDcIaJ8U5drTx3xRRJaIyA3btkhEdNUUpZRDcuqZf1FMx9p3kte2/WlLU0oph5NWAvNTxpjpxhiLbfsVyJnM9VJKqURx9mU/I9wSka4i4mrbupLIISBKKZXc0kqL+VWgA9bphVexDgPpmZyVUkqpxHKGFnO8M/+MMeeBF1KgLkop9dic+mWsIjIqjvOMMebjZKiPUko9FhdHbgrbKa4W8/0Y0jJiXdouO6CBWSnlcJwgLsc5wWRCxGcRyQwMwNq3PBeYENt5Sin1JDnyQz17xdnHLCLZgEFY35g9A6hkjLmTEhVTSqnEcIIu5jj7mMcDbbDODX/WGHMvxWqllFKJ5AwP/+IaLjcY60y/kcAVEfG3bXdFxD9lqqeUUgkjCfjPUcXVx6xvN1FKpTpO0GCOfxyzUkqlJk7/8E8ppVIbJ4jLGpiVUs7F2SeYKKVUquMMozI0MCulnIoTNJg1MCulnIt2ZSillINJ/WFZA7NSysnocDmllHIwTvDsTwOzUsq56KgMpZRyMNqVoZRSDsYJGswamJVSzkVbzEop5WBSf1jWwKyUcjKuTtCXoYFZKeVUtCtDKaUcjBPEZQ3MSinn4gxrZejro5RSTkXE/i3+a0lWEVkoIsdE5KiI1BCRbCKyRkRO2r562/KKiHwrIqdE5KCIVErsPSR7i/nOru+TuwiVCh047/ekq6CcVBL3MX8DrDLGtBOR9IAnMBxYZ4z5TETeB94H3gOaA8VsWzVgsu1rgmmLWSnlVFxF7N7iIiJZgLrAzwDGmBBjzH9Aa2CGLdsM4EXb59bATGO1A8gqInkScw8amJVSTsVF7N/iUQi4CUwXkX0i8pOIZARyGWOu2vJcA3LZPucDLkY5/5ItLeH3kJiTlFLKUSUkMItIHxHZHWXrE+VSbkAlYLIxpiJwH2u3RSRjjAFMUt+DjspQSjmVhPQxG2OmAdNiOXwJuGSM2WnbX4g1MF8XkTzGmKu2roobtuOXgQJRzs9vS0swbTErpZxKUnVlGGOuARdFpIQtqSFwBFgG9LCl9QCW2j4vA7rbRmdUB/yidHkkiLaYlVJOJYmHMfcH5thGZJwBemJt0M4XkV7AeaCDLe9KoAVwCgiw5U0UDcxKKafiloSR2RizH/CJ4VDDGPIa4K2kKFcDs1LKqTjBxD8NzEop5+IMU7I1MCulnIoTxGUNzEop5+IEyzFrYFZKORddKF8ppRyME8RlDcxKKeciTvDWPw3MSimnoi1mpZRyMBqYlVLKwejLWJVSysG4OsHSbBqYlVJORWf+KaWUg9E+ZqWUcjBO0GDWwKyUci4uOo5ZKaUci7aYlVLKwbg5QSezBmallFPRFrNSSjkYHS6nlFIOxgnisgZmpZRzcYKJfxqYlVLORbsylFLKwWhgVkopB5P6w7IGZqWUk3GCBrMGZqWUc9H1mJVSysGkmVEZIvInYB5K9gN2A1ONMUFJXTGllEoMZ3j4Z+8vlzPAPeBH2+YP3AWK2/aVUsohiIjdm6OytyujpjGmSpT9P0VklzGmiogcTo6KKaVUYjhDV4a995BJRJ6O2LF9zmTbDUnyWimlVCI5Q4vZ3sA8GNgiIhtEZCOwGRgiIhmBGclVuaRW8dlSdGjTOnK7fPlSrHmr+1R87PI+GP4+zZs0oEOb1nRs9xIH9u9L8DVGjxrB6VOnAPhp2pRox7p36fTYdUyr7vr/x8h+XRjZrwv9uzRjQLfnI/ctoaFJUsan7/Vl1NvdI/fPnDjCp+/1TZJrR7V5zXLu3L4Zuf/zxLFcvnAmyctJLSQBm6OyqyvDGLNSRIoBJW1Jx6M88JuYLDVLBhkyuDN/8dIULXPQ4KE0btqMbVu38PGYUSxc8meCzh/90SeRn3+aNpXX+jz4H3vmnLlJVs+0JrNXVsZ+PweAxbOn4e7hSYu2XSOPh4VZcHV9/EFL/n53OLBrG+Wr1Hzsa8Vm89rl5HumMN7ZcwLQa+DIZCsrNXBN4pawiLhiHehw2RjTUkQKAXOB7MAeoJsxJkREMgAzgcrAbaCjMeZcYspMyL+8ykBB2znlRQRjzMzEFOooAu7fZ0D/N/H398disdDv7QE816BRtDw3b95g6OB3uH/vHpawMEaOGk2lyj5s27qFyT98R0hICAUKFOCjsePwzJgx1rIq+1Th4oULAMz8dTp/LFkEQJu27eja/RUCAgIYOngg169dIyw8nD5936RZ8xb0eqUbg4YMZc3fqwkODqJDm9YUKVqUcV9MoLpPRXbs3sfQIe/QslVr6tarD1hb6nXr1adBo8Z88/WX7P73X0JCQ+jYuQvtO2grOzbTvhpDunTpOX/6BMVKl8PDM2O0gD3sjU4MGv0VOXPlZev6v1izbB4WSyhFSpSlx5tDcXF1feSaLdp2Zdm86Y8E5vCwMOb9+gPHDu7BEhpKw5btaIH2g4MAABkzSURBVNCiDeHh4cycPJ6jB3aTLUcuXN3cqNukFVVrN+SP335i387NhIQEU6xUOXr2H8aures5e/IoU8aPIn2GDIya8DNfjhpIp15vc+7kUa5fu0znXm8D1pb12ZNH6f7mu3bXPzVKhh6KAcBRwMu2/znwtTFmrohMAXoBk21f7xhjiopIJ1u+jokp0K6uDBGZBXwJ1Aaq2DafxBT4JEUEtg5tWjPw7bdInyEDX3/7A/MWLuGn6TOY8MXnGBN9VODKFcupWas28xcvZcHipZQoWZI7d3z5cepkpv40nXkLl1C6TFlmzpgeZ9n/bFxP0WLFOXL4EEv/WMzs3+cz+7d5LFq4gKNHj7Bty2Zy5nyKBUuWsXjpcmrVrhPt/IGDhkS2+Md9MSHasabNWvD3qr8ACA0JYefO7dSpV58lixaSKVNmfpu/iN/mLWLxwvlcunQxCb6Tzsv31g1GTfiJLn3eiTXP5Qtn2blpDSO//Imx389BXFzYtnFVjHmLlnwWNzc3jhzYHS39n7+X4emZkTHfzGD0N7+ycdUf3Lx2md3bNnDr+lXGTZnH60NGc+rY/yLPadSqPWO+mcG4yXMJCQ5m/79bqFq7IYWKlaLvux8x9vs5pM/gHpnfp3YD9mzbGLm/c9MaqtVrnKD6p0aSgP/ivZZIfuB54CfbvgANgIW2LDOAF22fW/Oga3ch0FAS2ZFtb4vZByhtHo5aqczDXRmhoaF8O/Er9u7ZhYu4cOPGdW7fukWOnDkj85Qt+ywfjhyOxWLhuQaNKFmqFLt3beDM6VO80rVz5HXKVagQY5lfTfiCH6dOxjtbNkZ//An/7thOg4aN8PT0BKBho8bs3bObWrXrMGH853w9YTz16j9Hpcr2/96rXacuX4z7hJCQELZu2UTlyj64u7uzfdtWTpw4ztq/VwNw995dLpw/T/78BRL8vUsrqtZpGG/L8ciBXZw7dYzRA3sAEBIcjFcW71jzt+70Ksvm/kKHnv0i0w7t3cmFsyfZtXU9AAH373HtykVOHD5A1doNcXFxIWu2HJQqVznynKMH9rBi4SxCgoO4f8+ffM8UpmK1Oo+UF8ErizdP5c7HqWP/I1feAly5dI7ipcuzdvmCBNU/tUlIKBSRPkCfKEnTjDHTouxPBIYCmW372YH/jDEW2/4lIJ/tcz7gIoAxxiIifrb8txJ4C3YH5kNAbuBqQgtwZCuX/8mdO778Pn8x6dKlo3njBgSHBEfLU9mnCr/MnM3mf/5h1Ij36dajJ5m9vKheoxaff/lVvGVE9DFH+HfH9hjzFSxYiLkLFrN58z98/+1EqlarTt83+8WY92EZMmTAp2pVtm3ZzOq//qJZ8xYAGGN4f/jIR1rfKnYZ3D0iP7u6umLCwyP3Q0OsA5CMMdRu+Dwder5l1zVLV6jCwllTOH38UGSaMYZubwyhXOUa0fIe2LUtxmuEhAQzY9IXjPlmBtlz5mLx7GmEPvRvNSbV6jVm56a15C1QkMo16mPrgkxQ/VObhLwl2xaEp8V0TERaAjeMMXtEpH7S1M4+9o7KyAEcEZHVIrIsYkvOiqWEe/fuki1bdtKlS8e/O3dw5crlR/JcuXKZ7Nlz0LZ9B15q256jRw5TrnwF9u/by4Xz5wEICAjg3LmzdpVZqbIPG9avJTAwkICAANavW0ulyj7cuHEddw8PWrZqTY+evTh29Mgj57qlcyM0lhEDTZu14I8/FrN37+7IQFyzVm0WzPs98pxz584SEBBgVz0V5MiVl3OnjwNw7tQxbl6/AkCZClXYtXU9/v/5AnDvrh+3rsfdZmnd6VVWLJwVuf9s5eqsX7EIi8Xa8Lp66TzBQYEUL12OXVvXEx4ejt+d2xw7uBd48Eshs1cWggIDIlvaAO4engQFxvxz9alRn307NrHjn7+pXq9JouufmojYv8WjFvCCiJzD+rCvAfANkFVEIhq1+YGIwHEZKGCtg7gBWbA+BEwwe1vMoxNzcUfXomUr3n7rDdq+2IrSZcpSqHDhR/Ls/vdffp3+M25ubnh6ejJ23Odky5aNjz4Zx/vvDiIk1Po/TL/+AylYsFC8ZZYqXYYXWrehS6f2gPXhX6lSpdm6ZTNfT/gCF3HBzc2NEaNGP3Ju23YdaN/mBUqVKv1IP3ONmrUY8f5Q6jdoSLr06a3XbteeK1cu06l9G4wxeHt7M/G7SQn9NqVZVWo9x9Z1KxnWtyOFS5Qldz7rUP58Txembbe+fDGyPybc4OrmRvc33yVHrjyxXqt8lVrRugvqNW3NretXGdW/GwaDVxZvBnwwHp9aDTi8fxfD+nYkW45cPFO0BJ6emciYKTP1m7Zm+BudyeKdncLFSkdeq06jlvz6/WeRD/+iypjZizwFCnLl4lmKlCiT6PqnJkk1JdsYMwwYBmBrMQ8xxnQRkQVAO6zBugcQ0T+6zLa/3XZ8fWK7fyW5u42DLI+ssaEUB877PekqOKygwADcPTy56/8fYwb2ZOSXP5I1W44nXa0UUa1IlseOquuO3bI75jQsmcOu8qIE5pYiUhhrUM4G7AO6GmOCRcQdmAVUBHyBTsaYRA0oj7PFLCJbjDG1ReQu0RcxEsAYY7xiOVUplUhfjR5EwP27WEIttO7cK80E5aRiz2iLhDLGbAQ22j6fAarGkCcIaJ8U5cUZmI0xtW1fM8eVTymVdIZ/PiX+TCpWDjzT2m52TzCxzX7JFfUcY8yF5KiUMzh39gxDBz8YC3vp0kXe7Pc2d+/eZdHC+WTzzgZA/4GDqFO33pOqpkpmISHBfDr0dUJDQwgPC6NK7Ya06dqHnyZ+zNmTR8FA7nxP03vQKNw9PPlr8Rz+Wb0MV1dXMmfJymsDP3Cavt+Ukhwt5pRmVx+ziPQHPgSuAxHjh4wxplx852ofM4SFhdH4ubrMnjufpUsW4+npSY+evZ50tZ6otNLHbIwhOCgQdw9PLBYLY4f0pmvfQeR7uhAentZ1wOZM+xqvrNlo1aEHRw7spkiJsmRwd2fdioUcPbiXfsM+fcJ3kXKSoo950wlfu2NO3eLZHDKK29tiHgCUMMYkauhHWrdzx3YKFChA3rz54s+snIqI4O5hnUwUZrEQFmZBkMigbIwhNCQ48s/v0uUfTCwqUvJZtq13nhl5KcUZFsq3NzBfxPrGEpUIq/5aQbMWLSP35/42hz+X/UHpMmUZ8u77eGXJ8gRrp5JbeFgYowZ05/qVSzRq2Y4iJcsC8ONXH3Fg9zbyPV2Izq8NfOS8TauXUc6nxiPpKm6pPywn7A0mG0VkmIgMithiyywifURkt4js/vnHGCfVpBmhISH8s2E9TWyz/zp07MzyVWuYv2gpOXM+xZfjP3vCNVTJzcXVlbHfz2HizOWcOXGES+dOA9B70Ci+nbWCPAUKsnPTmmjnbF3/F2dPHqVFu25PosqpmouI3ZujsjcwXwDWAOmxzhmP2GJkjJlmjPExxvj06t0ntmxpwpYtmyhZugzZc1iHPGXPkQNXV1dcXFxo0649h/73v3iuoJxFxkyZKVWuMgf3PJiW7+LqSvW6jaPN5Du071+WzZvOOx9+Sbp06Z9EVVO1NLEes200RnFjTJcUqI/T+WvlCpq3eD5y/+bNG+TM+RQA69eupWixYk+qaioF+PvdwdXVjYyZMhMSHMShfTt5vl13rl+5SK68BTDGsG/nZvIWKAjAudPH+fW7cQz5+Bu8smZ7spVPrRw54top3sBsjAkTkWdEJL0xRl8jlQABAQHs2LaNDz78KDLt6wnjOX7sGCKQN28+Phj9URxXUKndf763mDZhDCY8nHATTrU6jShfpRafvNuHwID7GAxPFyrGK/3eA2Duz98SFBTI9+OGAZA9Z27e+XBCXEWohzhyF4W97B0uNxMohXUu+P2IdGNMvMur6XA5FZO0MlxOJUxSDJfbdcbP7phTpfDjl5cc7B2Vcdq2uRBH37JSSj1xDhlqE8bed/6NSe6KKKVUUnCGmX92BWYR2QCPdkkYYxokeY2UUuoxOEEXs91dGUOifHYH2gKWWPI6teDgYHp270JoSAiWsDAaN2nKm/3ejpZn/GefsuvfnQAEBgVxx/c2W3ZY3/lW8dlSFCtWHIDcefLw7Q/WBWuGDR3MyZMnqFvvOd4eaB0iPm3KJIoWK06DhtFfEKscw49ff8z+f7fgldWbcZMfvLH872XzWLd8IS4uLpSvUotOvd5+5Nz79+7yyzefcOn8aRDhtYEjKVaqHPfu+vHDuBHcunGVHE/lod+wT8mY2YtdW9azePZUMmb2YsAH48nslZXrVy+x4NdJaWrKtj2cIC7b3ZWx56GkrSLybzLUx+GlT5+en36ZgWfGjISGhvJKt5epXacu5co/eOffu+8Pj/z825xZ0d5G8vB7BwFOHD9GBnd3Fi75k9df68ndu3cJCgrkfwcP0qfvm8l/UypR6jR6nsat2jN1wujItCMHdrN3xybG/jCHdOnSR74l5GGzp07g2crV6T/iMyyhoQQHBwGwfP4MSleoQqsOPfhz/gyWL5hBx1f7s+bP+YyeOIPd2zawfeNqmrzQkUUzp9Cue9+UuNVUJZHvP3Uo9r4lO1uULYeINMX62pQ0R0TwzJgRAIvFYn01UBz/EFatXEHzKNOxY+Lmlo7goCDCw8OxWCy4urgw6btvebNf/yStu0paJZ+tRMbM0ZckX79iES3b94icGBLTWOSA+/c4fmgf9Zq2BsAtXToyZrI+U9+7YxN1GlnHvddp9Dx7tv8DWP/dWUJDCAkOws3VjeOH9pHFO1vkW1XUA0n4aqknxt6ujD1Y+5gFaxfGWSDNLo8WFhZG5/ZtuHDhAh07v0y5cuVjzHflymUuX7pE1WrVI9NCQoLp3KENrq5uvPpaHxo0bEThIkXw9s5Gp3Yv8fwLrblw4QLhJpxSpcuk1C2pJHLtygVOHN7PwhmTSZc+PZ1fG0Dh4qWj5bl57QpeWbz58euPuHDmJIWKlqRr38FkcPfA/z/fyIXxs3hnj2xxt+rwCp+P6EfWbDno++5HfP/pMN58f2yK319q4MDx1m72dmXE/zK7NMTV1ZX5i5fi7+/PO2+/xcmTJyL7jaNatXIFjZo0xdXVNTLtrzUbyJUrF5cuXqT3qz0oVqw4BZ5+mqHDRkTm6f9mXz4YPYYfp07mxPFjVK9Ri7btO6TIvanHExYWxr27fnz49S+cOXGE78cNY8Ivf0T78zoszMK5U8fp1ncIRUqWZfaUCfw5f8Yj3RISpVlXtlI1ylaqBsCWdSsoV6Um1y5f4K9Fc/DMlJmurw8mg7t7yt2oI3OCyGzvWhmISE0ReVlEukdsyVmx1MDLy4sqVauxbcvmGI+v+mtltOnYALly5QIgf4EC+FSp+sjbsDesX0vpMmUICAjg4sULjP/qG9b8vZrAwMDkuQmVpLLleAqfms8hIhQpUQYXceGu/3+P5MmW46nIVeaq1G7AedvbuL2yZuM/31uAddZg1Be4AgQHBbF5zXIatWzP4tnT6DP4Q4qXKc+2jbo8aARJwH+Oyt4+5lnAl0BtoIpt84nzJCfl6+uLv78/AEFBQezYvo2ChR59u/bZM6e56+9P+QoVI9P8/fwIsb2G/s4dX/bv20vhIkUjj4eGhjJ75gxeefU1goOCI1tZ4eFhhIaGJudtqSRSuXo9jh60Piu/euk8Fksomb2yRsuTNVsOsuV8iquXzgNweP8u8j5t/aO0YvW6bF67AoDNa1dQqXrdaOeuXDSLJi90xM3NjdDgYEBwERdCgoKS+c5Sj7TUx+wDlE7sq7idya2bNxg5/H3Cw8MIDzc0adqMevWf44fvvqFMmbLUb9AQsLaWmzZvEe1P2DNnTvPxmA9xESHcGHq+1psiRR8E5nm/z+GF1i/h4eFB8RIlCAoMou2Lrahdpy5eXvreW0cz6fORHD24h3v+/zGgW0vadO1N3SYv8NPEjxn2Rifc3NLRZ9CHiAh3bt/k528+YchHEwHo1vddJn/xAWEWCzlz56X3O6MAaNm+Oz+MG86mv5eR/anc0YbC3bl9kzMnjvBSl94ANH6hA6MH9sAzY2YGjBqf8t8AB+XIAdde9q6VsQB42xhzNaEF6FoZKia6VoaKSVKslXH48n27Y06ZfBkdMozH2WIWkT+xjsbIDByxjV0OjjhujHkheaunlFIJ4wwt5vi6MpZhfTP2w0+36gAJbj0rpVRyc4K4HG9gbg0MM8ZEe82GiPgCnwI/J1fFlFIqUZwgMscXmHM9HJQBjDH/E5GCyVIjpZR6DM6wUH58gTlrHMc8krIiSimVFFJ/WI5/HPNuEen9cKKIvIZ1mrZSSjkWJ3gba3wt5oHAEhHpwoNA7IP1bdkvJWfFlFIqMRx5Rp+94gzMxpjrQE0ReQ4oa0teYYxZH8dpSin1xDhBF7PdixhtADYkc12UUuqxOUFctn8RI6WUSg1ExO4tnusUEJENInJERA6LyABbejYRWSMiJ21fvW3pIiLfisgpETkoIpUSew8amJVSTiUJFzGyAIONMaWB6sBbIlIaeB9YZ4wpBqyz7QM0B4rZtj7A5MTegwZmpZRTSapBGcaYq8aYvbbPd4GjQD6sE+9m2LLNAF60fW4NzDRWO4CsIpInMfeggVkp5VwSEJlFpI+I7I6y9YnxktYJdRWBnVgn3kUsSXEN67IVYA3aF6OcdsmWlmD2LvuplFKpQkKGyxljpgHT4ryeSCZgETDQGOMftW/aGGNEJMlX0NQWs1LKqSTlQvkikg5rUJ5jjFlsS74e0UVh+3rDln4ZKBDl9Py2tATTwKyUciouYv8WF7E2jX8GjhpjvopyaBnQw/a5B7A0Snp32+iM6oBfYtawB+3KUEo5nSQbyVwL6Ab8T0T229KGA58B80WkF3AeiHhT8kqgBXAKCAB6JrZgDcxKKaeSVDP/jDFbiD3KN4whvwHeSoqyNTArpZyKM8z808CslHIqaWatDKWUSi3im2qdGmhgVko5ldQfljUwK6WcjBM0mDUwK6Wci9MvlK+UUqlO6o/LGpiVUs7FCeKyBmallHNxcYJOZg3MSimn4gRxWRcxUkopR6MtZqWUU3GGFrMGZqWUU9Hhckop5WC0xayUUg5GA7NSSjkY7cpQSikHoy1mpZRyME4QlzUwK6WcjBNEZg3MSimn4gxTssX6/kCVEkSkjzFm2pOuh3Is+u9CPUynZKesPk+6Asoh6b8LFY0GZqWUcjAamJVSysFoYE5Z2o+oYqL/LlQ0+vBPKaUcjLaYlVLKwWhgVkopB5OmArOIGBGZEGV/iIiMjuecF0WkdBzHu4vIIRH5n4jsE5EhtvSNIuKTZJWPXmZfEemeHNdWcROR3CIyV0ROi8geEVkpIn1EZHkylrktua6tHFOaCsxAMNBGRHIk4JwXgRgDs4g0BwYCTYwxzwLVAb/HrqX12q6xHTPGTDHGzEyKcpT9RESAJcBGY0wRY0xlYBiQ6zGvG+cMXGNMzce5vkp90lpgtmB9Av7OwwdEpKCIrBeRgyKyTkSeFpGawAvAeBHZLyJFHjptGDDEGHMFwBgTbIz5Mcrx9iLyr4icEJE6tnJeEZHvo5S7XETq2z7fE5EJInIAqGHb/0REDojIDhHJZcs3+qGW+ecxlOMpIvNF5IiILBGRncnVgk9DngNCjTFTIhKMMQeAzUAmEVkoIsdEZI4tiCMi5yIaAiLiIyIbbZ9Hi8gsEdkKzLLt/2L7eZ4RkbcjyhCRe7av9W3HYyqnhS1tj4h8m5wteJX80lpgBvgB6CIiWR5K/w6YYYwpB8wBvjXGbAOWAe8aYyoYY04/dE5ZYE8cZbkZY6pibVV/aEfdMgI7jTHljTFbbPs7jDHlgU1A7wSU8yZwxxhTGvgAqGxH+Spucf28K2L9/pcGCgO17LheaaCRMaazbb8k0BSoCnwoIunsKUdE3IGpQHNbKz6nfbejHFWaC8zGGH9gJvD2Q4dqAL/ZPs8CaidBcYttX/cABe3IHwYsirIfAkS0fOK6Rkzl1AbmAhhjDgEH7ShfJd6/xphLxphwYD/2/byXGWMCo+yvsP3VdQu4QcxdJDGVUxI4Y4w5a8vze2JvQjmGNBeYbSYCvbC2SB/HYeJui
No results found