deepNeuralNetwork.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fXyB6m4hk5jU"
      },
      "source": [
        "# **پروژه اول درس یادگیری عمیق**\n",
        "\n",
        "---\n",
        " fateme padkan \n",
        "970122681002\n",
        "\n",
        "\n",
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "###  ◼ Preprocessing  "
      ],
      "metadata": {
        "id": "iCkUxCbLqVyU"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "j7Lvkeo8kP9a",
        "outputId": "82558b82-86b4-41fc-af69-256fc23bf1b3"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/gdown/cli.py:131: FutureWarning: Option `--id` was deprecated in version 4.3.1 and will be removed in 5.0. You don't need to pass it anymore to use a file ID.\n",
            "  category=FutureWarning,\n",
            "Downloading...\n",
            "From: https://drive.google.com/uc?id=1-Zyp-JP3f9QhPKaErBkPPFNKaPS1v74u\n",
            "To: /content/Dataset.csv\n",
            "100% 38.0k/38.0k [00:00<00:00, 2.43MB/s]\n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "!gdown --id 1-Zyp-JP3f9QhPKaErBkPPFNKaPS1v74u\n",
        "\n",
        "categorical_attr = ['gender', 'NationalITy', 'PlaceofBirth', 'StageID', 'GradeID', 'SectionID', 'Topic', 'Semester', 'Relation', 'ParentAnsweringSurvey', 'ParentschoolSatisfaction', 'StudentAbsenceDays', 'Class']"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PpbcPD9Vk6d5"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 250
        },
        "id": "VlWFj_XCkb3H",
        "outputId": "9d1b8256-72f1-4405-f8d0-f24a7b520e95"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  gender NationalITy PlaceofBirth     StageID GradeID SectionID Topic  \\\n",
              "0      M          KW       KuwaIT  lowerlevel    G-04         A    IT   \n",
              "1      M          KW       KuwaIT  lowerlevel    G-04         A    IT   \n",
              "2      M          KW       KuwaIT  lowerlevel    G-04         A    IT   \n",
              "3      M          KW       KuwaIT  lowerlevel    G-04         A    IT   \n",
              "4      M          KW       KuwaIT  lowerlevel    G-04         A    IT   \n",
              "\n",
              "  Semester Relation  raisedhands  VisITedResources  AnnouncementsView  \\\n",
              "0        F   Father           15                16                  2   \n",
              "1        F   Father           20                20                  3   \n",
              "2        F   Father           10                 7                  0   \n",
              "3        F   Father           30                25                  5   \n",
              "4        F   Father           40                50                 12   \n",
              "\n",
              "   Discussion ParentAnsweringSurvey ParentschoolSatisfaction  \\\n",
              "0          20                   Yes                     Good   \n",
              "1          25                   Yes                     Good   \n",
              "2          30                    No                      Bad   \n",
              "3          35                    No                      Bad   \n",
              "4          50                    No                      Bad   \n",
              "\n",
              "  StudentAbsenceDays Class  \n",
              "0            Under-7     M  \n",
              "1            Under-7     M  \n",
              "2            Above-7     L  \n",
              "3            Above-7     L  \n",
              "4            Above-7     M  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-73d0c52b-f2a1-4aa3-8421-4e61fe04fe11\">\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>gender</th>\n",
              "      <th>NationalITy</th>\n",
              "      <th>PlaceofBirth</th>\n",
              "      <th>StageID</th>\n",
              "      <th>GradeID</th>\n",
              "      <th>SectionID</th>\n",
              "      <th>Topic</th>\n",
              "      <th>Semester</th>\n",
              "      <th>Relation</th>\n",
              "      <th>raisedhands</th>\n",
              "      <th>VisITedResources</th>\n",
              "      <th>AnnouncementsView</th>\n",
              "      <th>Discussion</th>\n",
              "      <th>ParentAnsweringSurvey</th>\n",
              "      <th>ParentschoolSatisfaction</th>\n",
              "      <th>StudentAbsenceDays</th>\n",
              "      <th>Class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>M</td>\n",
              "      <td>KW</td>\n",
              "      <td>KuwaIT</td>\n",
              "      <td>lowerlevel</td>\n",
              "      <td>G-04</td>\n",
              "      <td>A</td>\n",
              "      <td>IT</td>\n",
              "      <td>F</td>\n",
              "      <td>Father</td>\n",
              "      <td>15</td>\n",
              "      <td>16</td>\n",
              "      <td>2</td>\n",
              "      <td>20</td>\n",
              "      <td>Yes</td>\n",
              "      <td>Good</td>\n",
              "      <td>Under-7</td>\n",
              "      <td>M</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>M</td>\n",
              "      <td>KW</td>\n",
              "      <td>KuwaIT</td>\n",
              "      <td>lowerlevel</td>\n",
              "      <td>G-04</td>\n",
              "      <td>A</td>\n",
              "      <td>IT</td>\n",
              "      <td>F</td>\n",
              "      <td>Father</td>\n",
              "      <td>20</td>\n",
              "      <td>20</td>\n",
              "      <td>3</td>\n",
              "      <td>25</td>\n",
              "      <td>Yes</td>\n",
              "      <td>Good</td>\n",
              "      <td>Under-7</td>\n",
              "      <td>M</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>M</td>\n",
              "      <td>KW</td>\n",
              "      <td>KuwaIT</td>\n",
              "      <td>lowerlevel</td>\n",
              "      <td>G-04</td>\n",
              "      <td>A</td>\n",
              "      <td>IT</td>\n",
              "      <td>F</td>\n",
              "      <td>Father</td>\n",
              "      <td>10</td>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>30</td>\n",
              "      <td>No</td>\n",
              "      <td>Bad</td>\n",
              "      <td>Above-7</td>\n",
              "      <td>L</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>M</td>\n",
              "      <td>KW</td>\n",
              "      <td>KuwaIT</td>\n",
              "      <td>lowerlevel</td>\n",
              "      <td>G-04</td>\n",
              "      <td>A</td>\n",
              "      <td>IT</td>\n",
              "      <td>F</td>\n",
              "      <td>Father</td>\n",
              "      <td>30</td>\n",
              "      <td>25</td>\n",
              "      <td>5</td>\n",
              "      <td>35</td>\n",
              "      <td>No</td>\n",
              "      <td>Bad</td>\n",
              "      <td>Above-7</td>\n",
              "      <td>L</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>M</td>\n",
              "      <td>KW</td>\n",
              "      <td>KuwaIT</td>\n",
              "      <td>lowerlevel</td>\n",
              "      <td>G-04</td>\n",
              "      <td>A</td>\n",
              "      <td>IT</td>\n",
              "      <td>F</td>\n",
              "      <td>Father</td>\n",
              "      <td>40</td>\n",
              "      <td>50</td>\n",
              "      <td>12</td>\n",
              "      <td>50</td>\n",
              "      <td>No</td>\n",
              "      <td>Bad</td>\n",
              "      <td>Above-7</td>\n",
              "      <td>M</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-73d0c52b-f2a1-4aa3-8421-4e61fe04fe11')\"\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-73d0c52b-f2a1-4aa3-8421-4e61fe04fe11 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-73d0c52b-f2a1-4aa3-8421-4e61fe04fe11');\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",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 90
        }
      ],
      "source": [
        "df = pd.read_csv('/content/Dataset.csv')\n",
        "df.head()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UmhpmlG7kk4f",
        "outputId": "b354d10c-6922-4c07-9ead-4cf56a38ba6b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(480, 17)\n"
          ]
        }
      ],
      "source": [
        "print(df.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ET9nW-LSkoWC"
      },
      "outputs": [],
      "source": [
        "# Converting Categorical values to scaler values\n",
        "le = LabelEncoder()\n",
        "df[categorical_attr] = df[categorical_attr].apply(le.fit_transform, axis=0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 250
        },
        "id": "icN67SXCkqTo",
        "outputId": "d6ed77dd-0062-4eca-bb33-04ba5a1e1333"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   gender  NationalITy  PlaceofBirth  StageID  GradeID  SectionID  Topic  \\\n",
              "0       1            4             4        2        1          0      7   \n",
              "1       1            4             4        2        1          0      7   \n",
              "2       1            4             4        2        1          0      7   \n",
              "3       1            4             4        2        1          0      7   \n",
              "4       1            4             4        2        1          0      7   \n",
              "\n",
              "   Semester  Relation  raisedhands  VisITedResources  AnnouncementsView  \\\n",
              "0         0         0           15                16                  2   \n",
              "1         0         0           20                20                  3   \n",
              "2         0         0           10                 7                  0   \n",
              "3         0         0           30                25                  5   \n",
              "4         0         0           40                50                 12   \n",
              "\n",
              "   Discussion  ParentAnsweringSurvey  ParentschoolSatisfaction  \\\n",
              "0          20                      1                         1   \n",
              "1          25                      1                         1   \n",
              "2          30                      0                         0   \n",
              "3          35                      0                         0   \n",
              "4          50                      0                         0   \n",
              "\n",
              "   StudentAbsenceDays  Class  \n",
              "0                   1      2  \n",
              "1                   1      2  \n",
              "2                   0      1  \n",
              "3                   0      1  \n",
              "4                   0      2  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-64691194-d8d1-4b63-bf56-b158d428665f\">\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>gender</th>\n",
              "      <th>NationalITy</th>\n",
              "      <th>PlaceofBirth</th>\n",
              "      <th>StageID</th>\n",
              "      <th>GradeID</th>\n",
              "      <th>SectionID</th>\n",
              "      <th>Topic</th>\n",
              "      <th>Semester</th>\n",
              "      <th>Relation</th>\n",
              "      <th>raisedhands</th>\n",
              "      <th>VisITedResources</th>\n",
              "      <th>AnnouncementsView</th>\n",
              "      <th>Discussion</th>\n",
              "      <th>ParentAnsweringSurvey</th>\n",
              "      <th>ParentschoolSatisfaction</th>\n",
              "      <th>StudentAbsenceDays</th>\n",
              "      <th>Class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>15</td>\n",
              "      <td>16</td>\n",
              "      <td>2</td>\n",
              "      <td>20</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>20</td>\n",
              "      <td>20</td>\n",
              "      <td>3</td>\n",
              "      <td>25</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>10</td>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>30</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>30</td>\n",
              "      <td>25</td>\n",
              "      <td>5</td>\n",
              "      <td>35</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>40</td>\n",
              "      <td>50</td>\n",
              "      <td>12</td>\n",
              "      <td>50</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-64691194-d8d1-4b63-bf56-b158d428665f')\"\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-64691194-d8d1-4b63-bf56-b158d428665f 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-64691194-d8d1-4b63-bf56-b158d428665f');\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",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 93
        }
      ],
      "source": [
        "df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "LSAJtfBRktzK"
      },
      "outputs": [],
      "source": [
        "# X: Features, y: Classes\n",
        "X = np.array(df.iloc[:, :-1])\n",
        "y = np.array(df['Class'])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "F3prh-kBuXgG"
      },
      "source": [
        "\n",
        "### ◼ lables must  be 0 or 1\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "yoz0HFq47sz0"
      },
      "outputs": [],
      "source": [
        "temp=np.amax(X, axis=0)\n",
        "X= X/temp\n",
        "\n",
        "for i in range(0,len(y)) : #H,M pass    L=fail\n",
        "  \n",
        "  if y[i]>0 :\n",
        "     y[i]=1\n",
        "  else :\n",
        "     y[i]=0\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "N6QauapCkv0I"
      },
      "outputs": [],
      "source": [
        "X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2, random_state=12)\n",
        "X_train = X_train.T\n",
        "X_val = X_val.T\n",
        "y_train = y_train.T.reshape(1, len(y_train))\n",
        "y_val = y_val.T.reshape(1, len(y_val))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NekPh9RNkx6l",
        "outputId": "5eec6e71-b895-4f22-bbdd-d9ba7bd13f03"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of dataset:  480\n",
            "Number of train set:  16\n",
            "Number of validation set:  16\n"
          ]
        }
      ],
      "source": [
        "print('Number of dataset: ', len(X))\n",
        "print('Number of train set: ', len(X_train))\n",
        "print('Number of validation set: ', len(X_val))"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### ◼ math utils\n"
      ],
      "metadata": {
        "id": "UVk0hf5CrH8J"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "def sigmoid(Z):\n",
        "\n",
        "    A = 1/(1+np.exp(-Z))  \n",
        "    return A\n",
        "\n",
        "\n",
        "\n",
        "def sigmoid_backward(dA, cache):\n",
        "   \n",
        "    \n",
        "    Z = cache\n",
        "    \n",
        "    s = 1/(1+np.exp(-Z))\n",
        "    dZ = dA * s * (1-s)\n",
        "    \n",
        "    \n",
        "    return dZ\n",
        "\n",
        "def relu(Z):\n",
        "    A=np.where(np.asarray(Z) > 0, Z, 0)\n",
        "    \n",
        "    return A\n",
        "def relu_backward(dA, cache):\n",
        "   \n",
        "    Z = cache\n",
        "    dZ = np.array(dA, copy=True) \n",
        "    \n",
        "    dZ[Z <= 0] = 0\n",
        "        \n",
        "    return dZ\n",
        "\n"
      ],
      "metadata": {
        "id": "fdtqwqgkuP_4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "### ◼ *in predict function :* \n",
        "\n",
        "\n",
        "⭕ **converting output of forward function to 0,1** \n",
        "\n",
        "\n",
        "> input:\n",
        "\n",
        "*  shape : shape of train or validation data . \n",
        "\n",
        " ✅    we need it for making a predict matrix which is [1][shape[1]]\n",
        "\n",
        "\n",
        "\n",
        "*   Y : train or validation Y\n",
        "\n",
        "*    output: output of forward prpopagation \n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "> output:\n",
        "\n",
        "* p : predict matrix \n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "MuXXIr8-KMyE"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "b9pCSEwtxjxA"
      },
      "outputs": [],
      "source": [
        "def predict( shape,Y, output):\n",
        "    p = np.zeros((1,shape[1]))\n",
        "    for i in range(0, output.shape[1]):\n",
        "        if output[0,i] > 0.5:\n",
        "            p[0][i] = 1\n",
        "        else:\n",
        "            p[0][i] = 0\n",
        "                   \n",
        "    return p"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "s66mPkcnfhAO"
      },
      "source": [
        "### ◼ forward Function for l layer:\n",
        "\n",
        " \n",
        "\n",
        "> input:\n",
        "\n",
        "*  X: data \n",
        "*  n:number of layers\n",
        "*  parameters : w and b (output of initilizer) \n",
        "\n",
        "\n",
        "\n",
        "other params :\n",
        "\n",
        "*  A_prev: X or A (activation) of a prev layer\n",
        "*  w : weight  \n",
        "*  b : bias\n",
        "* activationType : relu or tanh or sigmoid \n",
        "\n",
        "* A :  the output of the activation function \n",
        "\n",
        "* A_W_B_Z_dict: a dictionary with A,W,b,Z\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "> \n",
        "1.  for each layer z :  z(l)=W(l)A(l-1)+b(l)\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "OnnQWw7SwogY"
      },
      "outputs": [],
      "source": [
        "\n",
        "def forwardFunction(X,n, parameters):\n",
        "    caches = []\n",
        "    A = X\n",
        "    L = n-1    \n",
        "    for l in range(1, L):\n",
        "        previousA = A\n",
        "        W=parameters['W' + str(l)]\n",
        "        b=parameters['b' + str(l)]\n",
        "        A_W_b_dict = (previousA, W, b)\n",
        "        Z = np.dot(W, previousA) + b #linear part\n",
        "        A = relu(Z)    #non linear      \n",
        "        A_W_B_Z_dict = (A_W_b_dict, Z)\n",
        "        caches.append(A_W_B_Z_dict)\n",
        "\n",
        "\n",
        "    W=parameters['W' + str(L)]\n",
        "    b=parameters['b' + str(L)]\n",
        "    Z = np.dot(W, A) + b\n",
        "    A_W_b_dict = (A, W, b)\n",
        "    AL = sigmoid(Z)\n",
        "    A_W_B_Z_dict = (A_W_b_dict, Z)\n",
        "\n",
        "  \n",
        "    caches.append(A_W_B_Z_dict)\n",
        "\n",
        "    return AL, caches\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aSx2-TsDx9Hm"
      },
      "source": [
        "### ◼ compute cost function \n",
        "\n",
        "\n",
        " \n",
        "*compute_cost :*\n",
        "\n",
        "> input:\n",
        "\n",
        "*  AL  : probability vector  and from forward function \n",
        "*  Y\n",
        "\n",
        "\n",
        "> output:\n",
        "\n",
        "* cost \n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "DzryJCISyY78"
      },
      "outputs": [],
      "source": [
        "\n",
        "def compute_cost(AL, Y):\n",
        "    m = Y.shape[1]\n",
        "    cost = (-1/m) * (np.dot(Y, np.log(AL).T) + np.dot((1-Y), np.log(1-AL).T))\n",
        "    return cost"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9fEX-_MC9XXB"
      },
      "source": [
        "### ◼ backward propagation\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "HY_7vSc588p6"
      },
      "outputs": [],
      "source": [
        "\n",
        "def backwardFunction(AL, Y, caches):\n",
        "\n",
        "    L = len(caches) \n",
        "    m = AL.shape[1]\n",
        "    Y = Y.reshape(AL.shape) \n",
        "    gradiants = {}  #gradiant of the loss function\n",
        "\n",
        "\n",
        "    current_cache = caches[L-1]  \n",
        "    AWB, Z = current_cache\n",
        "\n",
        "    dAL = - (np.divide(Y, AL) - np.divide(1 - Y, 1 - AL))  ## derivative of cost with respect to AL\n",
        "    dZ = sigmoid_backward(dAL, Z)  #dz(l)=dA(l)*g'(z(l))\n",
        "\n",
        "    A_prev, W, b = AWB\n",
        "    m = A_prev.shape[1]\n",
        "    dW = (1/m) * np.dot(dZ, A_prev.T)\n",
        "    db = (1/m) * np.sum(dZ, axis=1, keepdims=True)\n",
        "    dA_prev = np.dot(W.T, dZ)        \n",
        "\n",
        "    gradiants[\"dA\" + str(L-1)]=dA_prev\n",
        "    gradiants[\"dW\" + str(L)]=dW\n",
        "    gradiants[\"db\" + str(L)]=db         \n",
        " \n",
        "\n",
        "    for l in reversed(range(L-1)):\n",
        "\n",
        "        current_cache = caches[l]\n",
        "        \n",
        "        AWB, Z = current_cache\n",
        "\n",
        "        dZ = relu_backward( gradiants[\"dA\" + str(l + 1)], Z)\n",
        "\n",
        "        A_prev, W, b = AWB\n",
        "        m = A_prev.shape[1]\n",
        "        dW = (1/m) * np.dot(dZ, A_prev.T)\n",
        "        db = (1/m) * np.sum(dZ, axis=1, keepdims=True)\n",
        "        dA_prev = np.dot(W.T, dZ)\n",
        "\n",
        "        gradiants[\"dA\" + str(l)] = dA_prev\n",
        "        gradiants[\"dW\" + str(l + 1)] = dW\n",
        "        gradiants[\"db\" + str(l + 1)] = db\n",
        "\n",
        "    return gradiants"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6nO-cZWV9u8-"
      },
      "source": [
        "### ◼ update the parameters of the model, using gradient descent\n",
        "\n",
        "\n",
        "*  w(l)=w(l)- lr_rate*dw(l)\n",
        "\n",
        "\n",
        "*  b(l)=b(l)- lr_rate*db(l)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "soKFAekGmj56"
      },
      "outputs": [],
      "source": [
        "\n",
        "def update_parameters(parameters,dims, grads, learning_rate):\n",
        "\n",
        "    L = len(dims)-1\n",
        "  \n",
        "    for l in range(L):\n",
        "        parameters[\"W\" + str(l+1)] = parameters[\"W\" + str(l+1)] -  learning_rate * grads[\"dW\" + str(l+1)]\n",
        "        parameters[\"b\" + str(l+1)] = parameters[\"b\" + str(l+1)] - learning_rate * grads[\"db\" + str(l+1)]\n",
        "    return parameters\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### ◼ function for implementring model\n",
        "\n",
        "\n",
        "### *in initializer function :* \n",
        "\n",
        "⭕ **first inititalize nueral network ** \n",
        "\n",
        "⬜ random initializer for random w and b  and zero initializer for zero w and b\n",
        "\n",
        "* w_b_dict : a dictionary with W and b\n",
        "\n",
        "* layersNum: number of layers \n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "DvduMIGAaBzX"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "0aVQSwkuBfdh"
      },
      "outputs": [],
      "source": [
        "def implementNetwork(X, Y ,X_val, Y_val, layers_dims, lr_rate = 0.1,randomInit=1, itersNum = 1000):#lr was 0.009\n",
        "\n",
        "    np.random.seed(1)\n",
        "    itersValues=[]\n",
        "    dims=layers_dims\n",
        "    trainShape=X.shape\n",
        "    validatonShape=X_val.shape\n",
        "    \n",
        "    n = len(layers_dims) \n",
        "    print(n,\"layer\")\n",
        "\n",
        "    \n",
        "    w_b_dict = {}\n",
        "\n",
        "    if randomInit==1 :\n",
        "\n",
        "      np.random.seed(3)\n",
        "\n",
        "      for l in range(1, len(dims)):  # from 1 to number of layers\n",
        "          w_b_dict['W' + str(l)] = np.random.randn(dims[l],\n",
        "                                                    dims[l-1]) * 0.01\n",
        "          w_b_dict[\"b\" + str(l)] = np.random.randn(dims[l ], 1) * 0.01\n",
        "\n",
        "      layersNum=len(dims)\n",
        "      parameters=w_b_dict;\n",
        "\n",
        "    else :\n",
        "\n",
        "\n",
        "      for l in range(1, len(dims)):  # from 1 to number of layers\n",
        "        w_b_dict['W' + str(l)] =np.zeros((dims[l], dims[l-1]))\n",
        "        w_b_dict[\"b\" + str(l)] = np.zeros((dims[l ], 1)) \n",
        "\n",
        "      layersNum=len(dims)\n",
        "      parameters=w_b_dict;\n",
        "     \n",
        "    trainLoss=[]\n",
        "    validationLoss=[]\n",
        "    trainAccu=[]\n",
        "    validationAccu=[]\n",
        "    for i in range(0, itersNum):\n",
        "\n",
        "       \n",
        "        AL,cachedValues = forwardFunction(X,n, parameters)\n",
        "     \n",
        "        cost = compute_cost(AL, Y)\n",
        "       \n",
        "        grads = backwardFunction(AL, Y,cachedValues)\n",
        "        \n",
        "        parameters = update_parameters(parameters,dims, grads, lr_rate)\n",
        "      \n",
        "\n",
        "\n",
        "        if   i % 100 == 0:\n",
        "\n",
        "           \n",
        "\n",
        "            validationOutput,chaches = forwardFunction(X_val,n, parameters)\n",
        "\n",
        "            predict_train =predict(trainShape,Y,AL)\n",
        "            predict_val = predict(validatonShape,Y_val,validationOutput)\n",
        "\n",
        "             \n",
        "            computeTrainLoss=(compute_cost(AL, Y)[0])\n",
        "            trainLoss.append(computeTrainLoss[0])\n",
        "            \n",
        "            computeValidationLoss=(compute_cost(validationOutput, Y_val)[0])\n",
        "            validationLoss.append(computeValidationLoss[0])\n",
        "\n",
        "    \n",
        "            #percentage of true predictions\n",
        "            trainAccu.append(np.sum((predict_train == Y)/trainShape[1]))\n",
        "            validationAccu.append(np.sum((predict_val == Y_val)/validatonShape[1]))\n",
        "\n",
        "            itersValues.append(i)\n",
        "\n",
        "\n",
        "        \n",
        "  \n",
        "   \n",
        "    print(\"trainLoss\",trainLoss)\n",
        "    print(\"validationLoss\",validationLoss)\n",
        "    print(\"trainAccuracy\",validationLoss)\n",
        "    print(\"validationAccuracy\",validationAccu)\n",
        "\n",
        "    i=trainLoss.index(np.sort(trainLoss)[0])\n",
        "    print(\"\\n \\nfor best trainLoss \\ntrainloss \",trainLoss[i])\n",
        "    print(\"trainAccuracy\",trainAccu[i],\"\\n\")\n",
        "\n",
        "    i=validationLoss.index(np.sort(validationLoss)[0])\n",
        "    print(\"\\n \\nfor best validation loss \\nvalidation loss \",validationLoss[i])\n",
        "    print(\"validation Accuracy\",validationAccu[i],\"\\n\")\n",
        "\n",
        "    i=trainAccu.index(np.sort(trainAccu)[-1])\n",
        "    print(\"\\n \\nfor best trainAccuracy  \\ntrainAccuracy\",trainAccu[i])\n",
        "    print(\"trainLoss\",trainLoss[i],\"\\n\")\n",
        "\n",
        "    i=validationAccu.index(np.sort(validationAccu)[-1])\n",
        "    print(\"\\n \\nfor best validationAccuracy  \\nvalidationAccu\",validationAccu[i])\n",
        "    print(\"trainLoss\",trainLoss[i],\"\\n\")\n",
        "\n",
        "    \n",
        "\n",
        "    \n",
        "    \n",
        "    \n",
        "    return itersValues,trainLoss,validationLoss,trainAccu,validationAccu"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### ◼ drawing graph function"
      ],
      "metadata": {
        "id": "bxCmybd9aXxN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def draw_pot(title,iters,trainLoss,validationLoss,trainAccu,validationAccu):\n",
        "    fig, axs = plt.subplots(2, 2, figsize=(20, 10))\n",
        "\n",
        "    x = np.squeeze(iters)\n",
        "    axs[0, 0].plot(x,np.squeeze(trainAccu))\n",
        "    axs[0, 0].set_title('Accuracy train')\n",
        "    axs[0, 1].plot(x,trainLoss, 'tab:orange')\n",
        "    axs[0, 1].set_title('loss train')\n",
        "    axs[1, 0].plot(x, np.squeeze(validationAccu), 'tab:green')\n",
        "    axs[1, 0].set_title('Accuracy val')\n",
        "    axs[1, 1].plot(x, validationLoss, 'tab:red')\n",
        "    axs[1, 1].set_title('loss val')\n",
        "    plt.suptitle(title, fontsize=20, fontweight='bold')\n",
        "    for ax in axs.flat:\n",
        "        ax.set(xlabel='Itration', ylabel='Percentage', ylim=[0, 1])\n",
        "\n",
        "    # Hide x labels and tick labels for top plots and y ticks for right plots.\n",
        "    # for ax in axs.flat:\n",
        "    #     ax.label_outer()"
      ],
      "metadata": {
        "id": "w4j5DAEfaUP2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "41e1rLFSCply"
      },
      "source": [
        "### ◼ call implement model function"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 2 layer random initialization"
      ],
      "metadata": {
        "id": "bDlJoQ2wKh9t"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QP3JzzUrEgBo",
        "outputId": "c155f1a1-b136-49e9-82ba-fad754dc9f82"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2 layer\n",
            "trainLoss [0.6936859367161485, 0.46167168355550353, 0.41250670725995975, 0.3885543710389004, 0.37351830568280076, 0.3627939182923593, 0.35458115247816646, 0.34801111744449403, 0.342599447463595, 0.3380474939197057]\n",
            "validationLoss [0.6878179498182379, 0.5487724471458746, 0.49782653564337587, 0.4749133744543341, 0.46187343660056684, 0.45352575109831694, 0.4478305242462689, 0.4438018364769299, 0.44089352718779107, 0.4387736024983775]\n",
            "trainAccuracy [0.6878179498182379, 0.5487724471458746, 0.49782653564337587, 0.4749133744543341, 0.46187343660056684, 0.45352575109831694, 0.4478305242462689, 0.4438018364769299, 0.44089352718779107, 0.4387736024983775]\n",
            "validationAccuracy [0.59375, 0.7083333333333334, 0.7291666666666667, 0.7604166666666666, 0.7708333333333333, 0.78125, 0.7916666666666667, 0.8020833333333334, 0.8020833333333334, 0.8125]\n",
            "\n",
            " \n",
            "for best trainLoss \n",
            "trainloss  0.3380474939197057\n",
            "trainAccuracy 0.8359375 \n",
            "\n",
            "\n",
            " \n",
            "for best validation loss \n",
            "validation loss  0.4387736024983775\n",
            "validation Accuracy 0.8125 \n",
            "\n",
            "\n",
            " \n",
            "for best trainAccuracy  \n",
            "trainAccuracy 0.8359375\n",
            "trainLoss 0.342599447463595 \n",
            "\n",
            "\n",
            " \n",
            "for best validationAccuracy  \n",
            "validationAccu 0.8125\n",
            "trainLoss 0.3380474939197057 \n",
            "\n"
          ]
        }
      ],
      "source": [
        "iters,trainLoss,validationLoss,trainAccu,validationAccu = implementNetwork(X_train, y_train,X_val,y_val, [16, 1], itersNum = 1000)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 677
        },
        "id": "A6FIfU95UHlI",
        "outputId": "da8fa532-0995-4dc8-8e0f-4039c5f32d6b"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x720 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "draw_pot(\"2 layer random initialization\",iters,trainLoss,validationLoss,trainAccu,validationAccu)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2-bBWFjwU-2h",
        "outputId": "488df4ae-8558-42c0-e74f-c652673f0d72"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "3 layer\n",
            "trainLoss [0.692044999670896, 0.5762911385815827, 0.5252652906507949, 0.39056478461254895, 0.3389952553145704, 0.31873979691878884, 0.3082017854351299, 0.3022650875964206, 0.29877309331771207, 0.2966744143761594]\n",
            "validationLoss [0.6906726841194886, 0.6999464767788892, 0.6588434837451805, 0.48452455307708675, 0.43819856288362813, 0.4318155084458171, 0.4345624644365052, 0.43923722246189034, 0.44371892942929014, 0.44768835220690156]\n",
            "trainAccuracy [0.6906726841194886, 0.6999464767788892, 0.6588434837451805, 0.48452455307708675, 0.43819856288362813, 0.4318155084458171, 0.4345624644365052, 0.43923722246189034, 0.44371892942929014, 0.44768835220690156]\n",
            "validationAccuracy [0.59375, 0.59375, 0.59375, 0.7291666666666666, 0.7916666666666667, 0.8125, 0.8020833333333333, 0.8125, 0.8020833333333334, 0.8020833333333334]\n",
            "\n",
            " \n",
            "for best trainLoss \n",
            "trainloss  0.2966744143761594\n",
            "trainAccuracy 0.8619791666666666 \n",
            "\n",
            "\n",
            " \n",
            "for best validation loss \n",
            "validation loss  0.4318155084458171\n",
            "validation Accuracy 0.8125 \n",
            "\n",
            "\n",
            " \n",
            "for best trainAccuracy  \n",
            "trainAccuracy 0.8619791666666666\n",
            "trainLoss 0.2966744143761594 \n",
            "\n",
            "\n",
            " \n",
            "for best validationAccuracy  \n",
            "validationAccu 0.8125\n",
            "trainLoss 0.31873979691878884 \n",
            "\n"
          ]
        }
      ],
      "source": [
        "iters,trainLoss,validationLoss,trainAccu,validationAccu = implementNetwork(X_train, y_train,X_val,y_val, [16,25, 1], itersNum = 1000)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 677
        },
        "id": "hgRMUCPVVJWX",
        "outputId": "da1fbf0a-eed9-4f6e-a021-8fafce041bab"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x720 with 4 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABI8AAAKUCAYAAACNNBhgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd5idVbn38e89JZl0CIReQlOkKEKkNxEVC6hHFLGADV4s2LEdj3g81mMXAQWkqCjYRcWCBaVDQA4ggmJIINTQQoCUKff7x3qG2ZnMnpmQmewp3891Pdfsp997zwys+WWt9URmIkmSJEmSJPWlqdEFSJIkSZIkaeQyPJIkSZIkSVJdhkeSJEmSJEmqy/BIkiRJkiRJdRkeSZIkSZIkqS7DI0mSJEmSJNVleCRJGnci4o0RkTXL7EbXpPoiYnav79cbG13T6oqI+TX1nz0SrhkRn6j9XIegngN7fZ8OHK57DZde9X+i0fVIkjRSGB5Jkka8iPjviPhlRNwWEQ9HREdEPBoRN0bEqRGxc6NrlEaSiLi4JgS5uNH1NFp/wZYkSRpYS6MLkCRpED7ex7ZpwE7V8paIeFVm/mLtliUN2qeBGdXrm0bINX8PPDZEtYyke62JE2peX96wKiRJGmEMjyRJo8H9wKXAPOAhYCrwAmBOtb8V+CwwrsOjiJiemY82+hpaVWaePtKumZmXs5YCkrV5rzWRmV9sdA2SJI1EDluTJI14mblhZr4yM0/IzM9m5n8CewD/rDlsq6G4V0TsEhGnRMQVEXFnRDweEcsj4u6IuDAijqg5tqU6pnsozFf6uF7t/EqdEbFZzb7WiHhrRPwhIhZFxIqIeCAifh8Rr+rjWqvM/RMRr4iIyyNiCbB4EO9vleE7EXFMRPwtIp4AbqiOmxkRn4+IiyLi9ohYHBHtEfFgdb8TImJSH9dfaR6eiNg2Is6NiPurz/GmiDi6Tm0bR8S3I+K+iFgaEddHxDFADOJ97V3d5/aIWFZ9326OiK9HxNZ9HH92TZ3zI2LDiDiz+j4sqd73rjWf+3kR8VB13b9ExF4D1dTf51Kzva/v6fMi4o/V0MzHI+LPEbHnYK4Z1dxCwAE1hx7Q+x61x3Yvva59YEScERFzq5/9pdXnekdE/DQiXrCa77/Pe/Xx81hvObDmnBMi4mcRcWv1+9Jefc9uiIgvR83vWHV8An/uVdKfa7//tcfWLJ/o433sFBGnRcQ/I+KJ6nO5rfq5fdZA7zsiJkbERyLilur34d6I+GZETF+dz1OSpLXNnkeSpFElIpqAdYEXAlvU7LpxiG6xL/C2PrZvXC0viojnZuZxmdkREacAn6mOOSoiPpKZy2rOe03N699l5sLqfcwEfgs8p9d91gOeDzw/In4AvD4zu+rU+paq3jXxSWC/PrZvAnywj+0zgb2q5TURsX9mPl7n2s8GrgVq/zDeETg7IsjMc7o3Vn/wXw5sXnPss4DTgF/29wYi4pPAx1g1ZHpGtbw5Il7Xz7DGacAVrBxAHgxcEiUsPJvyfem2P/DHiNg1M2/pr7an4K3A3qz8Xg4E/jRM96vnpZSfr942r5ZXRMRHM/Oza6meWh9i5e8HlN6IO1fLGyNi38y8eShvGhHHAt+g9HSstU21HBUR78rMU/u5zEWs/Pu2IfD/gKcBBw1huZIkDSnDI0nSqBDliWi319n9IPCuIbrVcuBq4G/AA5R5WqYC+1D+iAf4fxFxRmbOpYQb/wVMogQrrwK+W9W8PvC8mmufUfP6O/QER8uA84DbKOHKEZTewUdS5rL5DH3bl/Lez6cM7esdRA3GfsAdwE+BR+kJULqAWyifxb3Aw8AEShhzOKUNsSslaKs31OeZ1XlfoXw+xwDN1b4PA+fUHHsSKwdHl1B6i+wKHFqv+Crc+a+aTfMpn8dk4E2U790U4LyI2Ckz/93HZWZW9X2tOvat1fbJlODqCeCrwDrAG6t9k4B303fQuCb2oXzuPwV2AV68mvfrnlvobUB3j6t5QG2gcc0g6nic8j24gTJU9AnK/EoH0zNc9BMRcU5m3j2I69Xzb1aeZwjKe/0o0FatLwEW1OxfCFxcbXsYSGAz4NWU7+W6wP9SAjCq628DHFdzjW9W94bB9djbm/IZdvfaf4Dye94JHAVsQPmdODkibszMS+tcaj/gZ8DNwOuA2dX250bEHpl51UC1SJLUCIZHkqTR7h/AEZk5JD2PqnlkTo+IHSnhx/pAO/BrylC57qFaLwLmZuaDEfE9SjACpRfBd6vXr6Tn/7X3UfWgiYidgJfU3PbozPxh90pE3ElPr5/3R8TnM7Ozj3IfBXbNzDue6vulBHK7ZebDtRurXhvPiIhNKaHUJpT3fi09E5VD+RzqhUcJPC8z/wYQEcuA91T7to+IaZm5JCI2Al5Wc97F1Xld1XlnUoKgvtT2jloMPCczH6jO+xnwp2pfGyVgfHed6xyTmedW5+1I6VnV7S2ZeV61bwdg92r77gy9O4HdM3NJdb/rKD24BnW/7rmFIuKl9IRHd67uXD6ZeWI1bGtXYAdKINMB/Jye8GgCJRz9bl/XGOR97qTm5yciWqt7dAdHy4BDM/P2mnN2iYhplO/RbErg929K2NX9c3RwRLRmZntmfrEa9lYbHp2fmRevRqnvpyc46gT26+4FFhGnUcKgFkqPsRMoc7T15auZ+d7qvB8B19fs2x0wPJIkjUiGR5Kk0eIhyh9lLcBGlF4F21B6wlwdEW/OzB+s6U0iYhdKj5hnDnBo7bwqX6cnPNonInbMzL+z8pC172Rme/W69zCx8yPi/Dr3mUl5j309Tes7axgcAZzcOzgCiIh1gbOAw+h/zqHN+tl3RXdwVLm11/51Kb1K5vS6x3d7DdU7mz7Co4iYTE+wAvDL7uAIIDP/XM1nM7vatE+dOjuAH9asz6cnPGoHflyz71/0hDjr1rnemvhud3BU+Sc973E47teniHgecDoDzyXW3/d/de/ZROmR193bqgN4dWb+pdcxnwLeB0zs53ITKcHvPUNUXu3w0Ctqhw9m5r8i4lJ6eibW+zkDOKXmdV+/D5IkjUhOmC1JGhUy89HM/GJmfi4z3wNsD/yx2t0GnFH1YHnKokwA/WsGDo6g5g/XzLypphYow9o2psyN0+3bNa9nrmZps+psH4r5b+pd49uUXhwDTVbd3x/w83utL++13t0OWafX9vsGWO+2LivXd28fx9Ruq/e5318T7AGs6LWvo2a99vVwtKPm91qv/czWSrstIjahPLlwMJPQ9/f9X12n0BO4JvDGzOw939U7gY8M8r5DWVvtz85AP2f9hUDza17X+32QJGnEseeRJGlUqiar/iU9cwpNpgwrqzcp8mDsRxme1e0rwOeARZmZEXE/9YOcr9bU8gbgbnr+GLw0M2t7GTzU69zPU+ZQqaeveXqgzEuzpla5RtWj57CaTX8GjgVuz8zOiPghZW6ngbT3Ws8+j4JHeq1vOMB6t+75broDpL7Cw9ptvT/3br3rrNXRz77hMNjPbDi9lDIUrNsJwLcz8+HqZ2Mofu5WEhGfowz57Pau7mGEvdT25rubMjT0b5m5PCLeDpw81LVVHqLMawQD/5yt0pOvW21IWf03ZWiqkyRpmBkeSZJGtIjYH7gpMx/qtb2JMt9OrTX9Q3v9Xuvfy8z7q/sdRP3gCOBCyoTX21J60pxYs++MXsf2ng9leV9z0lQ9qfYagqFpq2sdeia2BvhVZt5W1bQB8Nwhvt+1rBwCvSEizq4ZunZ0Xydl5hMRcT09w7oOjYj1a+Y8ei49Q9YALhviukey2hBq8mqe2/v34MyaoY2v6X3wmoqID1GeoNbt45n5jUHUdm1mXlldo4n+A83eodzqfiaXAa+oXu8VEdvXzHm0HSsPaxtPP2eSpHHC8EiSNNK9GTgyIi6mPAHtEcofkC+mzAXU7VHgL6ucvXp6z0FybkT8ANiYnqds9SkzuyLiJMoTu6Bnwt/FwI96HXtjRPyGnvDr4xGxL+VR9UspvZ/mUCaqvoTydKa16X7K59w9nOxjEbEhJeB5A6uGC2skM++pepF193Y6ELg4Iv7EAE9bA74AfL96PQO4JiLOo4QDb645bjnliW7jxcKa17tFxNcpT9WDMs/V0n7O7f17cGFE/BrYDnjtENZIRBxO6d3XbT6wNCI+0OvQ86vJtW+t6gB4SUScDtxFmYB+DvUt7LX+6Wp+sxWUnkt/7OOcWl8CXk4JOJuBSyLiO5SnEh5FT5s6q2MlSRpTDI8kSaPBBOAF1dKXJcBrMnPAR273JzOvjYgL6Zmwd3vgv6vXF1HCqv4mCD4L+B9ges22H2TmE30c+wbgN5SACOCgamm4akjgZyiPO4cyh0v3U83uonwWzx/i274T2A3YtFrfj56Jxf9Enc8mM38QETtT5sGB0tPow70OWwq8vrv31DjxI3oCzybg+Jp9Z1M+k3ouoDwFbJdqfY9qgfIzXu/Jd0/FTr3WZ1MCwd7mUp5E91nghUAr5X29tdrfAXwPeH1fN8nMBRFxDT2/b7vQ8/5OZuU5y/o6/7KIeAdlcvwWSoD6vl6HdQLvzcy/9nctSZJGIyfmkySNdN+m9Bi5hvLkpBXVch/wV+DjwHaZ+Zshut/hlEeH30UZ6jKf8gfrYZQ/DuuqnpJ1Zq/NvYesdR/7ILA3pXfM7yjvp4PyaPJ5lMeVvws48qm9jTWTmV+gzEHzD8rnsAg4lxIi3D0M97uzuvbZ1b2WA3+nfAbH1D8TMvOjlMnJf0DpXbOCEo7cSgkGnpmZPx3qmkeyzLyQEqz8H6tOzDzQue2U+bvOoHwvVlA+yxPoCWsaIjMvBw6m9MhbRgmO/wQcwAABEPAfwPmU99Q1wLF93ftUSsD5bco8ZMson+3tlJ/b52TmeOrdJkkaRyKzEfMwSpI0NkXEMcBp1er1mfns/o6XJEmSRjqHrUmStIaqia23pwxpq50o+2t9nyFJkiSNHoZHkiStuUMoc8HUuhL4bgNqkSRJkoaUcx5JkjR0kjJX0snASzKz3zmSJEmSpNHAOY8kSZIkSZJUlz2PJEmSJEmSVJfhkSRJkiRJkuoyPJIkSZIkSVJdhkeSJEmSJEmqy/BIkiRJkiRJdRkeSZIkSZIkqS7DI0mSJEmSJNVleCRJkiRJkqS6DI8kSZIkSZJUl+GRJEmSJEmS6jI8kiRJkiRJUl2GR5IkSZIkSarL8EiSJEmSJEl1GR5JkiRJkiSpLsMjSZIkSZIk1WV4JEmSJEmSpLoMjyRJkiRJklSX4ZEkSZIkSZLqMjySJEmSJElSXYZHkiRJkiRJqsvwSJIkSZIkSXUZHkmSJEmSJKkuwyNJkiRJkiTVZXgkSZIkSZKkugyPJEmSJEmSVJfhkSRJkiRJkuoyPJIkSZIkSVJdhkeSJEmSJEmqy/BIkiRJkiRJdRkeSZIkSZIkqS7DI0mSJEmSJNVleCRJkiRJkqS6DI8kSZIkSZJUl+GRJEmSJEmS6jI8kiRJkiRJUl2GR5IkSZIkSarL8EiSBiEiHouIrRtdhyRJGt8iYn5EHNzoOgYrIl4XEb9vdB2S1ozhkTTORcTFEfFwRExsdC3DISIOjIiFa3qdzJyamfOGoiZJkqTRICLOjohPrck1MvPczHzBUNUkqTEMj6RxLCJmA/sBCRy2lu/dsjbv15+RVIskSdJoYRtKGj8Mj6Tx7SjgSuBs4OjaHRGxeUT8NCIWRcSDEfGNmn3HRMQ/ImJJRNwcEbtW2zMitq057sl/reruARQRH4qIe4GzImLdiPhVdY+Hq9eb1Zw/MyLOioi7q/0/r7bfFBGH1hzXGhEPRMSze72HKcBvgE2qYWePRcQmEfGJiPhxRHwvIh4F3hgRu0fEFRHxSETcExHfiIgJNdd68r1V7+vkiPh19RlcFRHbrOk3Q5IkaXVExMSI+GrVVrq7ej2x2rd+1bZ6JCIeiohLIqKp2vehiLirasfcGhHP6+PaxwKvAz5YtaF+WW2fX51/A/B4RLRExIcj4t81bcNX1FznjRFxac16RsRxEfGvqraTIyKG+aOStIYMj6Tx7Sjg3Gp5YURsCBARzcCvgAXAbGBT4Lxq36uAT1TnTqf0WHpwkPfbCJgJbAkcS/lv0FnV+hbAUuAbNcd/F5gM7AhsAHyl2v4d4PU1x70YuCcz/1Z7s8x8HHgRcHc17GxqZt5d7X4Z8GNgner9dwLvBdYH9gKeB7y9n/fyGuC/gXWB24BPD+oTkCRJGjr/CewJ7AI8C9gd+Fi17/3AQmAWsCHwUSAj4unAO4HnZOY04IXA/N4XzszTKG2k/63aUIfW7D4SeAmwTmZ2AP+m9GafQWkffS8iNu6n7pcCzwGeCby6qkHSCGZ4JI1TEbEvJbT5YWZeS/mf/mur3bsDmwAnZObjmbksM7v/xeitlEbENVnclpkLBnnbLuDEzFyemUsz88HM/ElmPpGZSygBzAFVfRtTgp/jMvPhzGzPzL9U1/ke8OKImF6tv4ESNK2OKzLz55nZVdVybWZemZkdmTkf+FZ3LXX8LDOvrhpM51IabZIkSWvT64BPZub9mbmIEty8odrXDmwMbFm1oy7JzKT8g9lEYIeIaM3M+Zn579W879cz887MXAqQmT/KzLurdtX5wL8o7cl6PpeZj2TmHcCfsR0ljXiGR9L4dTTw+8x8oFr/Pj1D1zYHFlTBSG+bU4Kmp2JRZi7rXomIyRHxrYhYUA0f+yuwTtXzaXPgocx8uPdFqt5DlwGvjIh1KCHTuatZy521KxHxtKpr971VLZ+h9EKq596a108AU1fz/pIkSWtqE0pP8W4Lqm0AX6D0jv59RMyLiA8DZOZtwHsoPcnvj4jzImITVk/vdtRREXF9NQztEWAnbEdJY4rhkTQORcQkShfhA6qw5F7KkK1nRcSzKA2CLepMgngnUG9+nycow8y6bdRrf/Zafz/wdGCPzJwO7N9dYnWfmVU41JdzKEPXXkXpRXRXneN637Pe9lOBW4Dtqlo+WtUhSZI0Ut1N6UnebYtqG5m5JDPfn5lbU6YZeF/33EaZ+f3M7O6FnsDn61x/wHZURGwJnE4ZCrdeZq4D3ITtKGlMMTySxqeXU7os70DpJrwL8AzgEspcRlcD9wCfi4gpEdEWEftU554BfCAidoti26rRAHA98NqIaI6IQ+h/2BfANMo8R49ExEzgxO4dmXkPZbLrU6qJtVsjYv+ac38O7Aq8mzIHUj33AetFxIxB1PIo8FhEbA+8bYDjJUmSGu0HwMciYlZErA98nDK8n4h4adVOC2Axpe3XFRFPj4iDqom1l1HaYl11rn8fsPUANUyhhEmLqvu+idLzSNIYYngkjU9HA2dl5h2ZeW/3Qpms+nWUfyk6FNgWuIMy2eIRUMa0U+Ym+j6whBLizKyu++7qvEeq6/x8gDq+CkwCHqA89e23vfa/gTJe/xbgfkoXa6o6lgI/AbYCflrvBpl5C6VhNa/qSl2vW/YHKHM+LaH869n5A9QuSZLUaJ8C5gI3ADcC11XbALYD/gA8BlwBnJKZf6bMd/Q5SvvrXspDST5S5/rfpsyN9EhUT73tLTNvBr5U3eM+YGfK9AKSxpAoc6ZJ0ugTER8HnpaZrx/wYEmSJEnSU9LXfCaSNOJVw9zeQs8TRSRJkiRJw2DYhq1FxJkRcX9E3FRnf0TE1yPitoi4ISJ2Ha5aJI0tEXEMZULt32TmXxtdjySNJLbBJEnSUBvOOY/OBg7pZ/+LKONwtwOOpTzpSJIGlJmnZ+aUzDyu0bVI0gh0NrbBJEnSEBq28KjqDfBQP4e8DPhOFlcC60TExsNVjyRJ0nhgG0ySJA21Rs55tCll2Em3hdW2e3ofGBHHUv5ljClTpuy2/fbbr5UCJUnS2nfttdc+kJmzGl3HGGYbTJIkraK/NtiomDA7M08DTgOYM2dOzp07t8EVSZKk4RIRCxpdgwrbYJIkjR/9tcGGc86jgdwFbF6zvlm1TZIkScPHNpgkSVotjQyPLgCOqp74sSewODNX6S4tSZKkIWUbTJIkrZZhG7YWET8ADgTWj4iFwIlAK0BmfhO4EHgxcBvwBPCm4apFkiRpvLANJkmShtqwhUeZeeQA+xN4x3DdX5IkaTyyDSZJkoZaI4etSZIkSZIkaYQzPJIkSZIkSVJdhkeSJEmSJEmqy/BIkiRJkiRJdRkeSZIkSZIkqS7DI0mSJEmSJNVleCRJkiRJkqS6DI8kSZIkSZJUl+GRJEmSJEmS6jI8kiRJkiRJUl2GR5IkSZIkSarL8EiSJEmSJEl1GR5JkiRJkiSpLsMjSZIkSZIk1WV4JEmSJEmSpLoMjyRJkiRJklSX4ZEkSZIkSZLqMjySJEmSJElSXYZHkiRJkiRJqsvwSJIkSZIkSXUZHkmSJEmSJKkuwyNJkiRJkiTVZXgkSZIkSZKkugyPJEmSJEmSVJfhkSRJkiRJkuoyPJIkSZIkSVJdhkeSJEmSJEmqq6XRBUiSpLWnqytp7+piRUcX7Z1Je2d5vaKze1vPentnrrKte729s4s377MVLc3+O5QkSdJYZ3gkSQ3U0dnFso4ulq7oZFl7WZa2d7KsvYul7Z0sXdHJ8o7ytXb78uq4pSs6Vzl/WUcnQdDUFLQ0Bc0RNDVBc1PQFNW26nVzU80S5ZzmCJqbq6/d5zR3H8+Tx7U01Ry/ynE925tXOg6am5pobmLl+0fv42rOjUZ/l4ZOR1fS3pGs6OxkRUdPcNPe2RPMrOjsor0Kdlbd1n1cPrltRWfvwKcnGFp1W9k+VF6/55aGR5IkSeOA4ZEk9ZKZLO/oWjnIqcKbJ0ObmiBn2YpVQ5/aIKg74Fn2ZADU8/Wp/iE/saWJSROaaWtpLl9bm2lrbWJSazPT2lpIoLMr6cqks6ssKzq66MzS86R7W2dmWa85rvd5Pcfx5HFau1qbg9bmJia0NJWv1esJzU20tsST26ZObCnbao9tiT62Vec2BxNamquvTXWPK+vlPhNr9k1qbW70RyNJkqS1wPBI0pjS2ZUsWdbOo0s7eHRZO48uba++1q53sHjpyvseW95RE/R0kk8hH2lpCia1NjOxtZlJE8of1m3Vss6kVtqmT2RSawl7JnaHPi09x05sbS77q3MmTWh68rgnt7U2M7GliaYGdsfJTLqyVziVSWfnymFUR2fP/vIVOrq6Vgqhunod131+R1fPdZ7K92Kkam6qgpwqjKkNYmpDodaawGZCcxMRY6j7lSRJkkYdwyNJI0pnV/LYshL0LK4T/PQZCFX7Hlve0e/1mwKmtbUyfVIL09tamd7WylbrT2FqW8uTwU5bSxNtvQKb2p49ba09gc7Emm2t42T4TkTQHCUIkSRJkjT2GR5JGlJdXcljK0rAU3r3DCL4qdn+2PKOfnuaRMC0iS1Mn1SCn+mTWthi5uSV1mc8+bqV6W3VsdXrKRNaGtprR5IkSZJGG8MjSSvp6koeX9HRE+h0h0B1g5+V15cMEP4ATGtrWSnc2WzdSUzfeHpPb6Da0KettYRBk8r6VMMfSZIkSVqrDI+kMSYzeXxF50rBTu/5ffoMfqrXS5a1M9B8yFMntjC9rYVpVbCzyTptbD9pWp/BT3cg1N0baGpbi8OdJEmSJGkUMTySRpjM5IkVnSsFO4ufaO9/6FevUGig8GfKhOaVwp2NprfxtA2nVQHPqsFP7frUiS0+mluSJEmSxhHDI2ktae/s4vd/v495ix7rN/h5dFnHgI9CnzyheaVwZ4NpbWw7q4Q+M/oJfqa3tTKtzfBHkiRJkjR4hkfSMHt0WTvnX30nZ152O/csXgbApNbmlcKd9adOYOtZU+pM+LxyEDStrWXcPNVLkiRJktR4hkfSMLn7kaWcddnt/ODqO3lseQd7bb0en37FTuy77SwmtBj+SJIkSZJGB8MjaYjddNdiTr9kHr++4R4SeMnOG3PMfluz82YzGl2aJEmSJEmrzfBIGgKZycX/XMTpf53H5f9+kCkTmnnj3rN5075bsek6kxpdniRJkiRJT9mwhkcRcQjwNaAZOCMzP9dr/xbAOcA61TEfzswLh7MmaSgt7+jkF3+7m9Mvmce/7n+Mjaa38ZEXbc9rdt+CGZNaG12eJGmcsg0mSZKG0rCFRxHRDJwMPB9YCFwTERdk5s01h30M+GFmnhoROwAXArOHqyZpqDzyxArOveoOzr58PouWLOcZG0/nK0c8i5fsvInzGUmSGso2mCRJGmrD2fNod+C2zJwHEBHnAS8DahsuCUyvXs8A7h7GeqQ1tuDBxznz0tv54dyFLG3v5ICnzeLYI7Zm723WIyIaXZ4kSWAbTJIkDbHhDI82Be6sWV8I7NHrmE8Av4+I44EpwMF9XSgijgWOBdhiiy2GvFBpINfd8TCn/3Uev/v7vTQ3BS/bZVPeut9WbL/R9IFPliRp7bINJkmShlSjJ8w+Ejg7M78UEXsB342InTKzq/agzDwNOA1gzpw52YA6NQ51diV/+Md9nP7Xecxd8DDT21o47oBtOHrv2Ww4va3R5UmStCZsg0mSpEEbzvDoLmDzmvXNqm213gIcApCZV0REG7A+cP8w1iX1a+mKTn583UK+fck85j/4BJutO4kTD92BV8/ZnCkTG523SpI0INtgkiRpSA3nX8LXANtFxFaUBstrgNf2OuYO4HnA2RHxDKANWDSMNUl1LVqynO9eMZ/vXrmAh59o51mbzeDk1+7KC3fckJZmJ8GWJI0atsEkSdKQGrbwKDM7IuKdwO8oj4A9MzP/HhGfBOZm5gXA+4HTI+K9lIkb35iZdonWWnXb/Y/x7Uvn8ZPr7qK9s4uDn7Ehx+y3Nc+Zva6TYEuSRh3bYJIkaagN6xiczLyQ8ujX2m0fr3l9M7DPcNYg9SUzuer2hzjjknn84R/3M7GlicN324y37LsV28ya2ujyJElaI7bBJEnSUHICF40rHZ1d/Oamezn9knncsHAxM6dM4D0Hb8cb9tyS9aZObHR5kiRJkiSNOIZHGhceW97B+dfcyZmX3s5djyxlq/Wn8OlX7MQrd92MttbmRpcnSZIkSdKIZXikMe3excs4+/L5nHvVApYs62D32TP5xGE78rztN6CpyfmMJEmSJEkaiOGRxjNUbvsAACAASURBVKR/3PMop18yj1/+3910diUv2mlj3rrfVjx7i3UbXZokSZIkSaOK4ZHGjMzkkn89wOmXzOOSfz3A5AnNvG6PLXnLvlux+czJjS5PkiRJkqRRyfBIo96Kji5++X93c/ol87jl3iVsMG0iJ7zw6bxujy1YZ/KERpcnSZIkSdKoZnikUWvx0na+f9UdnH357dz36HKetuFUvnD4Mzlsl02Y2OIk2JIkSZIkDQXDI406dz70BGddNp/zr7mDx1d0su+26/P5Vz6TA542iwgnwZYkSZIkaSgZHmnUuGHhI5z213lceOM9NEVw6LM24a37bcWOm8xodGmSJEmSJI1ZhkeVW+9dwn2PLmt0GerDI0vbOffKBVx1+0NMm9jCMfttzRv3mc3GMyY1ujRJkiRJksY8w6PKGZfM40fXLmx0GapjkxltfOwlz+CI52zOtLbWRpcjSZIkSdK4YXhUecdzt+U1u2/e6DLUh+amJnbcZDqtzU2NLkWSJEmSpHHH8Kgye/0pzF5/SqPLkCRJkiRJGlHsyiFJkiRJkqS6DI8kSZIkSZJUl+GRJEmSJEmS6jI8kiRJkiRJUl2GR5IkSZIkSarL8EiSJEmSJEl1GR5JkiSpca45AxbObXQVkiSpH4ZHkiRJaoz2ZXDpV+GM58GP3wwPL2h0RZIkqQ+GR5IkSWqM1jZ4+xWw/wfhlgvhG8+Biz4OyxY3ujJJklTD8EiSJEmNM3EaHPSfcPy1sNMr4bKvw9efDVefDp3tja5OkiRheCRJkqSRYMam8IpT4diLYYMd4MIPwCl7lR5JmY2uTpKkcc3wSJIkSSPHJrvA0b+EI88r6+cdCeccCndf39i6JEkaxwyPJEmSNLJEwNNfVOZDevEX4f6b4bQD4WfHweK7Gl2dJEnjjuGRJEmSRqbmVtj9GHjX32Cfd8FNP4WTdoM/fQqWL2l0dZIkjRuGR5IkSRrZ2mbA8z8J77wGtn8x/PUL8PVd4dqzobOj0dVJkjTmGR5JkiRpdFh3Szj8THjrH2Hm1vDLd8O39oN//aHRlUmSNKYZHkmSJGl02WwOvPm38OrvQPtSOPeV8N1XwH1/b3RlkiSNSYZHkiRJGn0iYIeXwTuuhhd+Bu66Dr65L1xwPCy5r9HVSZI0phgeSZIkafRqmQB7vaNMqr3HcXD9D+Drz4a//C+seKLR1UmSNCYYHkmSJGn0mzwTDvksvOMq2PYg+POny5PZrv8+dHU1ujpJkkY1wyNJkiSNHettA0d8D970G5i2Efz8bXDa/jDvL42uTJKkUcvwSJIkSWPPlnuXp7K98tuwdDF85zD4/hGw6J+NrkySpFHH8EiSJEljU1MT7Hw4vPMaOPgTsOByOGVP+PX74fEHGl2dJEmjhuGRJEmSxrbWNtj3vWVS7TlvgrlnlUm1L/0KtC9rdHWSJI14hkeSJEkaH6asDy/5Erz9ijKs7Q+fgG88B278MWQ2ujpJkkYswyNJkiSNL7OeDq89H466ACbNgJ+8Bc54Hiy4otGVSZI0IhkeSZIkaXza+gA49q/w8lPh0XvgrEPg/NfDg/9udGWSJI0owxoeRcQhEXFrRNwWER+uc8yrI+LmiPh7RHx/OOuRJEkaD2yDrYamJtjltXD8tfDc/4Tb/gQn7wG//Qg88VCjq5MkaUQYtvAoIpqBk4EXATsAR0bEDr2O2Q74CLBPZu4IvGe46pEkSRoPbIM9RRMmwwEfhHddB7scCVd9s0yqfcXJ0LGi0dVJktRQw9nzaHfgtsycl5krgPOAl/U65hjg5Mx8GCAz7x/GeiRJksYD22BrYtpGcNhJcNylsOmu8LuPwsm7w82/cFJtSdK4NajwKIrXR8THq/UtImL3AU7bFLizZn1hta3W04CnRcRlEXFlRBxS5/7HRsTciJi7aNGiwZQsSZI06tkGa6ANd4Q3/Axe9xNoaYMfHgVnHgIL5za6MkmS1rrB9jw6BdgLOLJaX0LpDr2mWoDtgAOra58eEev0PigzT8vMOZk5Z9asWUNwW0mSpFHBNlijbXdw6YV06NfgoXnlqWw/fjM8vKDRlUmStNYMNjzaIzPfASwDqLo4TxjgnLuAzWvWN6u21VoIXJCZ7Zl5O/BPSkNGkiRJtsFGhuYW2O2NZT6k/U+AWy6EbzwHLvo4LFvc6OokSRp2gw2P2qvJFxMgImYBXQOccw2wXURsFRETgNcAF/Q65ueUf/EiItandKGeN8iaJEmSxjrbYCPJxGlw0MfKk9l2+g+47GtlUu2rT4fO9kZXJ0nSsBlsePR14GfABhHxaeBS4DP9nZCZHcA7gd8B/wB+mJl/j4hPRsRh1WG/Ax6MiJuBPwMnZOaDT+F9SJIkjUW2wUaiGZvCK74Jx/4FNtgBLvwAnLIX3PobJ9WWJI1JkYP8H1xEbA88Dwjgj5n5j+EsrJ45c+bk3LlOVChJ0lgVEddm5pxG1zFS2AYb4TJLaHTRf8GDt8FGz4Rtngtb7gtb7Alt0xtdoSRJg9JfG6xlkBeYCdwP/KBmW2tm2j9XkiRpmNgGGwUiYPsXw3bPh2vPhpt+AlecUoa0RVMJk2bvC1vuA1vuBZPWbXTFkiSttkGFR8B1lIkXH6b8q9c6wL0RcR9wTGZeO0z1SZIkjWe2wUaL5lbY/ZiytC+FhdfA/Mtg/qVlTqQrvgEEbLgTzN6nCpP2gSnrNbpySZIGNNjw6CLgx5n5O4CIeAHwSuAsyiNk9xie8iRJksY122CjUesk2Gr/sgC0L4O7roUFVZh07Tlw1TfLvlnPWDlMmrZh4+qWJKmOQc15FBE3ZubOvbbdkJnPjIjrM3OXYauwF8fbS5I0tjnnUQ/bYGNUxwq4+28lTFpwGdxxJax4rOxbb7sqTNoXtty7TM4tSdJasMZzHgH3RMSHgPOq9SOA+6pHxw70uFhJkiQ9NbbBxqKWCbDFHmXZ733Q2QH3/l8Z5rbgMrjpZ2X+JIB1Z5cgqbt30rpbNrJySdI4Ndjw6LXAicDPq/XLqm3NwKuHoS5JkiTZBhsfmltg093Kss+7oKsT7rupJ0y69ddw/ffKsTM2LyFSd5g0c+syabckScNoUOFRZj4AHF9n921DV44kSZK62QYbp5qaYeNnlWWvt0NXFyz6RxUmXQq3/QFuqDqjTdukDG/rHuq2/naGSZKkITeo8CgiZgEfBHYE2rq3Z+ZBw1SXJEnSuGcbTAA0NcGGO5Zlj2MhEx74Z5l8u3sS7pt+XI6dskEVJu1beibN2r6cL0nSGhjssLVzgfOBlwLHAUcDi4arKEmSJAG2wdSXCJj19LI85y0lTHpoXk2YdBncXI10nDRz5TBpw50MkyRJq22w4dF6mfntiHh3Zv4F+EtEXDOchUmSJMk2mAYhAtbbpiy7HV3CpEcW9MyZNP9SuOVX5di2GbDF3j1zJm30zDLnkiRJ/Rjs/ynaq6/3RMRLgLuBmcNTkiRJkiq2wbT6IspT2tadDc9+Xdm2eGHPnEnzL4N//qZsnzANttizZ86kTXaB5tZGVS5JGqEGGx59KiJmAO8HTgKmA+8ZtqokSZIEtsE0VGZsBs86oiwAj95TeiV1D3P7w0Vle+sU2Hz3EiZt8zzYdNfG1SxJGjEGGx49nJmLgcXAcwEiYp9hq0qSJElgG0zDZfrGsPPhZQF4bNHKYdKfPlWWZx4BL/wsTFmvsfVKkhpqsLPlnTTIbZIkSRo6tsG0dkydBTu+HF78BXj75fDB2+GAD8NNP4WTdy9fMxtdpSSpQfrteRQRewF7A7Mi4n01u6YDzcNZmCRJ0nhlG0wNN3kmPPcjsMNh8It3wo/fBDf+GF7ypdJrSZI0rgzU82gCMJUSMk2rWR4FDh/e0iRJksYt22AaGTbcEd76B3jBp+Dff4KT94DrvmMvJEkaZ/rteVTzSNizM3PBWqpJkiRpXLMNphGlqRn2Ph6e/mL45bvhguPhxh/BoV+HmVs1ujpJ0low2AmzJ0bEacDs2nMy86DhKEqSJEmAbTCNJOttA0ddANedAxd9HE7dGw76GOxxXAmYJElj1mDDox8B3wTOADqHrxxJkiTVsA2mkaWpCea8CbZ7Afz6ffC7j5bJtF/2DdjgGY2uTpI0TAYbHnVk5qnDWokkSZJ6sw2mkWnGpnDkeXDTT+A3H4Rv7gf7nwD7vhdaJjS6OknSEBtowuxuv4yIt0fExhExs3sZ1sokSZJkG0wjVwTsfDi84xrY8eVw8WfgtAPhrmsbXZkkaYgNtufR0dXXE2q2JbD10JYjSZKkGrbBNPJNWQ9eeQbsdDj86r1wxsGw1zvgwI/ChMmNrk6SNAQGFR5lpo9RkCRJWstsg2lUefohsOVecNGJcPlJcMuvyxPZttqv0ZVJktbQoIatRcTkiPhY9bQPImK7iHjp8JYmSZI0vtkG06jTNgMO/Soc/SvIhHNeCr98Dyxb3OjKJElrYLBzHp0FrAD2rtbvAj41LBVJkiSpm20wjU5b7Qdvuxz2Ph6uOwdO3hP++btGVyVJeooGGx5tk5n/C7QDZOYTQAxbVZIkSQLbYBrNJkyGF3wK3voHmLQufP/V8JO3wuMPNLoySdJqGmx4tCIiJlEmaCQitgGWD1tVkiRJAttgGgs23Q2OvbhMoP33n8PJu8ONPy7D2iRJo8Jgw6MTgd8Cm0fEucAfgQ8OW1WSJEkC22AaK1omwIEfguMugXW3gp+8BX7wGlh8V6MrkyQNwmCftnZRRFwH7EnpKv3uzLS/qSRJ0jCyDaYxZ4NnwFt+D1d9C/70P3DKnvD8T8KuR0PTYP9dW5K0tg32aWuvADoy89eZ+SugIyJePrylSZIkjW+2wTQmNTXDXm8vE2pv8mz41XvgO4fBg/9udGWSpDoGPWwtM598vmZmPkLpRi1JkqThYxtMY9fMreCoX8BhJ8E9N8Cp+8DlJ0FXZ6MrkyT1MtjwqK/jBjXkTZIkSU+ZbTCNbRGw61Hwjqtgm4Pg9x+DMw6G+25udGWSpBqDDY/mRsSXI2KbavkycO1wFiZJkiTbYBonpm8MrzkXDj8LHrkDvrU//Pmz0OHDBSVpJBhseHQ8sAI4HzgPWAa8Y7iKkiRJEmAbTONJBOz0H/DOa8rXv3wOvnUALJzb6MokadwbsNtzRDQDv8rM566FeiRJkoRtMI1jk2fCf5wGOx1eJtM+42DY8+1w0H/ChCmNrk6SxqUBex5lZifQFREz1kI9kiRJwjaYxNNeAG+/Eua8Ga48GU7dG+b9pdFVSdK4NNgJFx8DboyIi4DHuzdm5ruGpSpJkiSBbTCNd23T4aVfhp1eCRccD985rEyw/fz/gUnrNLo6SRo3Bhse/bRaJEmStPbYBpMAZu8Db7sMLv4cXH4S/OsieMmXYfsXN7oySRoXBhUeZeY5ETEJ2CIzbx3mmiRJkoRtMGklrZPg+f8NO74cfnE8nHck7Pgf8KL/hamzGl2dJI1pg3raWkQcClwP/LZa3yUiLhjOwiRJksY722BSHzZ5Nhz7ZzjoY3DLr+Dk3eGGH0JmoyuTpDFrUOER8Algd+ARgMy8Hth6oJMi4pCIuDUibouID/dz3CsjIiNiziDrkSRJGg8+gW0waVXNrbD/CXDcpbDetvDTY+D7r4bFCxtdmSSNSYMNj9ozc3GvbV39nVA9XvZk4EXADsCREbFDH8dNA94NXDXIWiRJksYL22BSf2Y9Hd78Wzjk8zD/Ujh5T7jmDOjq99dEkrSaBhse/T0iXgs0R8R2EXEScPkA5+wO3JaZ8zJzBXAe8LI+jvsf4PPAssEWLUmSNE7YBpMG0tQMex4Hb78CNtsNfv1+OOel8OC/G12ZJI0Zgw2Pjgd2BJYD3wcWA+8Z4JxNgTtr1hdW254UEbsCm2fmr/u7UEQcGxFzI2LuokWLBlmyJEnSqGcbTBqsdWfDG34OLzsZ7rsJTt0bLv0qdHY0ujJJGvX6fdpaRLQBxwHbAjcCe2XmkPzXNyKagC8Dbxzo2Mw8DTgNYM6cOc6EJ0mSxjTbYNJTFAHPfj1se3DpgfSHE+HvP4OXfQM22rnR1UnSqDVQz6NzgDmURsuLgC+uxrXvAjavWd+s2tZtGrATcHFEzAf2BC5wwkZJkiTbYNIambYRHPE9eNU58Ohd8K0D4MwXwZ8+BfMuhhVPNLpCSRpV+u15BOyQmTsDRMS3gatX49rXANtFxFaUBstrgNd276wmf1y/ez0iLgY+kJlzV+MekiRJY5FtMGlNRcCOL4et9ofLTyqh0SVfgr9+AZpaYdPdYPa+Zdl8d5gwpdEVS9KINVB41N79IjM7ImLQF66OfyfwO6AZODMz/x4RnwTmZuYFT6VgSZKkccA2mDRUJs+Eg08EToRlj8KdV8H8S8rT2S79ClzyRWhq6QmTttwHNt8DJk5tdOWSNGJEZv3h6xHRCTzevQpMAp6oXmdmTh/2CnuZM2dOzp3rP4xJkjRWRcS1mTmuh1DZBpPWkuVL4I6rYMGlJUy66zrIzhImbbIrzN6n6pm0p2GSpDGvvzZYvz2PMrN5eEqSJElSPbbBpLVk4jTY7uCyACx/DO68EuZfVsKky08qvZOiGTZ5djXMbT/YYo9yriSNEwMNW5MkSZKk8WHi1PKktm2rMGnF49Uwt6pn0hUnw2VfrcKkXaphbvvCFntC21rvEChJa43hkSRJkiT1ZcIU2OagskAVJl1dgqQFl8EVp8BlX4Nogo13qYa57VeFSTMaW7skDSHDI0mSJEkajAlTYJvnlgVgxROw8OqeYW5XfasMdYsm2OiZNcPc9oRJ6zS2dklaA4ZHkiRJkvRUTJgMWx9YFoD2pbDwmmqY22Vw9WlwxTeAgI2fWYKkLfeBLfeCSes2rGxJWl2GR5IkSZI0FFonwVb7lwWqMGluzzC3q0/vCZM22rnqmbQvbLEXTJ7Z0NIlqT+GR5IkSZI0HFonwVb7lQWgfRncNbca5nYJzD0TrjwFCNhwp54wacu9DZMkjSiGR5IkSZK0NrS29QREfAg6lsNd1/Y8ze3as+CqU8ux3WHSlvuUZcp6DS1d0vhmeCRJkiRJjdAysfQy2nJvOOCDVZh0HSzoDpPOgau+WY7dYMfqaW77wqztYeqG5YluEY19D5LGBcMjSZIkSRoJWiaWybS33Av2PwE6VsDdfytD3OZfCn/7XpmE+8nj20qING2jPr5uBNM2LF8nrwdNTY17X5JGPcMjSZIkSRqJWibAFnuUZf8PlDDp3hvg4fmw5F547F5Ycl/5uuhWuP0vsGzxqtdpaoEpG/SESfW+Tt0AmlvX+tuUNPIZHkmSJEnSaNAyATabU5Z62pfCY/f1hEq9vy5eWCbtfvwBIHudHKWXUp89mXp9bZ00nO9U0ghjeCRJkiRJY0XrJFh3dln609kOjy+qejDd1/fXRbeU110dq54/cUbVY6mf4XLTNoSJ052XSRoDDI8kSZIkabxpboXpm5SlP11dsPShVYfJ1X5deE352rF01fNbJvU/XG7yejBhGkycChOmwoQphk3SCGR4JEmSJEnqW1MTTFm/LOxU/7hMWP5oH+FSTU+m+/8B/74YlvcxL9OTooRI3WHSk1+nPbX1lomGUdIQMDySJEmSJK2ZCGibUZZZT+v/2BVPlEDpsfvgiYdgxWOwfEn1tfv1kvK6e9sjd6x8TOfywdXV1FInXJq6co+nwa43+ye0xid/8iVJkiRJa8+EyTBzq7I8VZ3tK4dJvQOofteXwJJ7qu1VSJWdg7tvS1s/4dKUsr95YpncvKUNmieU3k8tE6vtbWVf88Re27tfT1j1mKbmp/45SUPE8EiSJEmSNLo0t8LkmWVZU5nQsWzlMOnJ0Gmg9cfKxOMP3w4rHoeO5dC5olyvr4nGn4qmlgECpr4Cqe5j6gRSgwqtWsrn3NSy6tLcCtHkkMBxxPBIkiRJkjR+RZSn1LVOAmYN3XW7usrwuo5l0LGiel0tK71e0euYvo7vPqav45fDskf7v+ZQBVm9NbVAU3fA1NwrbGqu2ddShvzVC6Kamle91krhVX/Xaq1zfO31miCaS+DVVH2N5up1DM++McbwSJIkSZKkodbUBE3doVSDdXWufmjV1dGzdNa87mov13tyX+16e/W1s9re0bPe1b7y9XrfY7D3Ihv9aQ7OSsFS9+umnpBp0Pti5XBqo53hpV9e62/H8EiSJEmSpLGsqbnMNcXkRley5rq6Vg2qngyx6oRR2QnZVY7NrrLe1VmGLK60r/Z11hzX1cc1hup6fV27n+s1T2jIx254JEmSJEmSRoemJmiaADQmRBmvmhpdgCRJkiRJkkYuwyNJkiRJkiTVZXgkSZIkSZKkugyPJEmSJEmSVJfhkSRJkiRJkuoyPJIkSZIkSVJdhkeSJEmSJEmqy/BIkiRJkiRJdRkeSZIkSZIkqS7DI0mSJEmSJNVleCRJkiRJkqS6DI8kSZIkSZJUl+GRJEmSJEmS6jI8kiRJkiRJUl2GR5IkSZIkSarL8EiSJEmSJEl1GR5JkiRJkiSpLsMjSZIkSZIk1WV4JEmSJEmSpLqGNTyKiEMi4taIuC0iPtzH/v/P3p3H2VmXdx//XLPPJCEkZBKzEsSIJuxEBFEWlxZqC+qjCIriU1pad1Fqqa0o7qKiVcBKFXEBQSylWPdiAz4qS1hEw6IRgQQCBAgkZCaZ7Xr+OPdMTiY5yYTMmTOZ+bxfr3mdc6/nOnMz8fJ7fr/7vDci7oyIOyLi2ojYs5r1SJIkjQf2YJIkaThVLTyKiHrgAuA4YCFwckQsHLTbbcDizNwf+B5wbrXqkSRJGg/swSRJ0nCr5sijQ4HlmXlvZnYBlwMnlO+Qmf+bmR3F4g3AnCrWI0mSNB7Yg0mSpGFVzfBoNrCibHllsa6S04AfbW1DRJweEUsjYunq1auHsURJkqQxxx5MkiQNq1Fxw+yIOAVYDHxma9sz86LMXJyZi9vb20e2OEmSpDHKHkySJA1FQxXP/SAwt2x5TrFuMxHxcuCfgaMyc2MV65EkSRoP7MEkSdKwqubIo5uBBRGxV0Q0AScB15TvEBEHAV8Bjs/MR6tYiyRJ0nhhDyZJkoZV1cKjzOwB3gH8BLgL+G5mLouIj0TE8cVunwEmAldGxO0RcU2F00mSJGkI7MEkSdJwq+a0NTLzh8APB607u+z5y6v5+pIkSeORPZgkSRpOo+KG2ZIkSZIkSRqdDI8kSZIkSZJUkeGRJEmSJEmSKjI8kiRJkiRJUkWGR5IkSZIkSarI8EiSJEmSJEkVGR5JkiRJkiSpIsMjSZIkSZIkVWR4JEmSJEmSpIoMjyRJkiRJklSR4ZEkSZIkSZIqMjySJEmSJElSRYZHkiRJkiRJqsjwSJIkSZIkSRUZHkmSJEmSJKkiwyNJkiRJkiRVZHgkSZIkSZKkigyPJEmSJEmSVJHhkSRJkiRJkioyPJIkSZIkSVJFhkeSJEmSJEmqyPBIkiRJkiRJFRkeSZIkSZIkqSLDI0mSJEmSJFVkeCRJkiRJkqSKDI8kSZIkSZJUkeGRJEmSJEmSKjI8kiRJkiRJUkWGR5IkSZIkSarI8EiSJEmSJEkVGR5JkiRJkiSpIsMjSZIkSZIkVWR4JEmSJEmSpIoMjyRJkiRJklSR4ZEkSZIkSZIqMjySJEmSJElSRYZHkiRJkiRJqsjwSJIkSZIkSRUZHkmSJEmSJKkiwyNJkiRJkiRVVNXwKCKOjYh7ImJ5RJy1le3NEXFFsf3GiJhfzXokSZLGA3swSZI0nKoWHkVEPXABcBywEDg5IhYO2u00YE1mPgf4PPDpatUjSZI0HtiDSZKk4VbNkUeHAssz897M7AIuB04YtM8JwDeK598DXhYRUcWaJEmSxjp7MEmSNKwaqnju2cCKsuWVwAsr7ZOZPRHxFLAH8Fj5ThFxOnB6sfh0RNxTlYph2uDX1qjhtRmdvC6jl9dmdPK6DM2etS5gF2cPpuHidRm9vDajk9dl9PLaDE3FHqya4dGwycyLgIuq/ToRsTQzF1f7dbTjvDajk9dl9PLajE5eF+1q7MHGN6/L6OW1GZ28LqOX12bnVXPa2oPA3LLlOcW6re4TEQ3AZODxKtYkSZI01tmDSZKkYVXN8OhmYEFE7BURTcBJwDWD9rkGOLV4/lrg55mZVaxJkiRprLMHkyRJw6pq09aK+fPvAH4C1AMXZ+ayiPgIsDQzrwG+BnwrIpYDT1Bqbmqp6sOy9Yx5bUYnr8vo5bUZnbwuqjp7MA0jr8vo5bUZnbwuo5fXZieFHzJJkiRJkiSpkmpOW5MkSZIkSdIuzvBIkiRJkiRJFRkeFSLi2Ii4JyKWR8RZta5nPImIuRHxvxFxZ0Qsi4h3F+unRsTPIuIPxeOUYn1ExBeLa3VHRBxc23cwtkVEfUTcFhH/XSzvFRE3Fr//K4qbsRIRzcXy8mL7/FrWPdZFxO4R8b2IuDsi7oqIw/2bqb2IOKP4d+x3EfGdiGjxb0aqzP6rtuzBRjd7sNHJHmx0sgerPsMjSv8wAxcAxwELgZMjYmFtqxpXeoD3ZeZC4DDg7cXv/yzg2sxcAFxbLEPpOi0ofk4HvjzyJY8r7wbuKlv+NPD5zHwOsAY4rVh/GrCmWP/5Yj9Vz78CP87M5wEHULpG/s3UUETMBt4FLM7MfSndqPgk/JuRtsr+a1SwBxvd7MFGJ3uwUcYebGQYHpUcCizPzHszswu4HDihxjWNG5m5KjNvLZ6vo/QP8GxK1+AbxW7fAF5VPD8B+GaW3ADsHhEzR7jscSEi5gCvBL5aLAfwUuB7xS6Dr0v/9foe8LJifw2ziJgMHEnp25LIzK7MfBL/ZkaDBqA1IhqANmAV/s1Ildh/1Zg92OhlDzY62YONavZgVWZ4VDIbWFG2vLJYpxFWDBk8CLgRmJGZq4pNDwMziuder5HzBeD9QF+xvAfwZGb2FMvlv/uB61Jsf6rYX8NvL2A18PViOPtXI2ICTfwzoQAAIABJREFU/s3UVGY+CHwWeIBSw/IUcAv+zUiV+G/TKGIPNurYg41O9mCjkD3YyDA80qgREROB/wDek5lry7dlZgJZk8LGqYj4S+DRzLyl1rVoCw3AwcCXM/MgYD2bhkcD/s3UQnF/gxMoNZazgAnAsTUtSpKGwB5sdLEHG9XswUYhe7CRYXhU8iAwt2x5TrFOIyQiGik1LZdm5lXF6kf6h3UWj48W671eI+MI4PiIuI/SVIKXUprjvXsxHBQ2/90PXJdi+2Tg8ZEseBxZCazMzBuL5e9RamT8m6mtlwN/yszVmdkNXEXp78i/GWnr/LdpFLAHG5XswUYve7DRyR5sBBgeldwMLCjuxt5E6eZa19S4pnGjmF/6NeCuzDyvbNM1wKnF81OB/ypb/+bi2wsOA54qGyaqYZKZ/5SZczJzPqW/iZ9n5huB/wVeW+w2+Lr0X6/XFvv7qUsVZObDwIqI2KdY9TLgTvybqbUHgMMioq34d63/uvg3I22d/VeN2YONTvZgo5c92KhlDzYCwt9RSUT8BaW5xfXAxZn58RqXNG5ExIuBXwC/ZdO87g9QmnP/XWAecD9wYmY+UfyDcD6loYgdwP/NzKUjXvg4EhFHA2dm5l9GxLMpfQo2FbgNOCUzN0ZEC/AtSvdLeAI4KTPvrVXNY11EHEjpJppNwL3A/6X0gYB/MzUUEecAr6f0DUa3AX9DaV69fzPSVth/1ZY92OhnDzb62IONTvZg1Wd4JEmSJEmSpIqctiZJkiRJkqSKDI8kSZIkSZJUkeGRJEmSJEmSKjI8kiRJkiRJUkWGR5IkSZIkSarI8EhSVUXE08Xj/Ih4wzM4fveIeFvZ8qyI+N5w1ihJkjSW2H9JGm6GR5JGynxgq81LRDRs47jdgYHmJTMfyszXDm9pkiRJY9J87L8kDQPDI0kj5VPASyLi9og4IyLeEhHXRMTPgWsjYmJEXBsRt0bEbyPihLLj9i6O+0zxCdrvACKiJSK+Xux/W0QcU6x/S0RcFRE/jog/RMS5tXnLkiRJNWX/JWlYbCttlqThdBZwZmb+JZQaDOBgYP/MfKL49OvVmbk2IqYBN0TENcVx+2bmgcVx88vO+XYgM3O/iHge8NOIeG6x7UDgIGAjcE9EfCkzV1T9XUqSJI0e9l+ShoXhkaRa+llmPlE8D+ATEXEk0AfMBmZs5/gXA18CyMy7I+J+oL95uTYznwKIiDuBPQGbF0mSNN7Zf0naYYZHkmppfdnzNwLtwCGZ2R0R9wEtO3HujWXPe/HfO0mSJLD/kvQMeM8jSSNlHTBpG9snA48WjcsxlD6p2t5xv6DU9FAMl54H3DM85UqSJO3y7L8kDQvDI0kj5Q6gNyJ+ExFnbGX7pcDiiPgt8GbgboDMfBz4ZUT8LiI+M+iYC4G64pgrgLdk5kYkSZIE9l+ShklkZq1rkCRJkiRJ0ijlyCNJkiRJkiRVZHgkSZIkSZKkigyPJEmSJEmSVJHhkSRJkiRJkioyPJKkHRARl0TEx2pdhyRJUr+IuC8iXl7rOrYnIjIinlPrOiTtOMMjSVuIiCURsSYimmtdiyRJkiSptgyPJG0mIuYDLwESOH6EX7thJF9PkiRJkrR9hkeSBnszcANwCXBq+YaImBsRV0XE6oh4PCLOL9v2txFxV0Ssi4g7I+LgYv1mw5PLp31FxNERsTIi/jEiHga+HhFTIuK/i9dYUzyfU3b81Ij4ekQ8VGy/ulj/u4j4q7L9GiPisYg4aPAbLOr8y7LlhuL1+mu+MiIejoinIuL6iFi0k79TSZKkERERzRHxhaJXeqh43lxsm1b0Vk9GxBMR8YuIqCu2/WNEPFj0cvdExMu2cu4XFj1Sfdm6V0fEHcXzQyPi18X5V0XE+RHRNFLvXVL1GB5JGuzNwKXFz59HxAyAokn4b+B+YD4wG7i82PY64MPFsbtRGrH0+BBf71nAVGBP4HRK/y59vVieB3QC55ft/y2gDVgETAc+X6z/JnBK2X5/AazKzNu28prfAU4uW/5z4LHMvLVY/hGwoDj/rZR+F5IkSbuCfwYOAw4EDgAOBf6l2PY+YCXQDswAPgBkROwDvAN4QWZOotQb3Tf4xJl5I7AeeGnZ6jcAlxXPe4EzgGnA4cDLgLcN31uTVCuGR5IGRMSLKYU2383MW4A/UmoIoNR4zAL+ITPXZ+aGzPx/xba/Ac7NzJuzZHlm3j/El+0DPpSZGzOzMzMfz8z/yMyOzFwHfBw4qqhvJnAc8PeZuSYzuzPzuuI83wb+IiJ2K5bfRClo2prLgOMjoq1YfgOlQAmAzLw4M9dl5kZKodgBETF5iO9HkiSplt4IfCQzH83M1cA5lPoigG5gJrBn0Uf9IjOTUujTDCyMiMbMvC8z/1jh/AMfwkXEJEof2H0HIDNvycwbMrMnM+8DvkLRx0natRkeSSp3KvDTzHysWL6MTVPX5gL3Z2bPVo6bSyloeiZWZ+aG/oWIaIuIr0TE/RGxFrge2L0Y+TQXeCIz1ww+SWY+BPwS+D8RsTulkGmrI4YyczlwF/BXRYB0fPFeiYj6iPhURPyxeP37isOmPcP3J0mSNJJmURop3u/+Yh3AZ4DlwE8j4t6IOAsGeqP3UPrQ7NGIuDwiZrF1lwGvKabCvQa4tf9Dw4h4bjEt7uGij/oE9lDSmGB4JAmAiGgFTgSOKv4H/2FKw44PiIgDgBXAvAo3tV4B7F3h1B2Uppn1e9ag7Tlo+X3APsALM3M34Mj+EovXmVqEQ1vzDUpT114H/DozH6ywH2z61OwE4M6iaYLSKKQTgJcDkylN0et/fUmSpNHuIUojyfvNK9ZRjKx+X2Y+m9KHZ+/tv7dRZl6Wmf2j0BP49NZOnpl3UgqkjmPzKWsAXwbuBhYUfdwHsIeSxgTDI0n9XkVpyPJCSnPkDwSeD/yC0r2MbgJWAZ+KiAkR0RIRRxTHfhU4MyIOiZLnRER/03I78IZiRM+xbH/o8iRK9zl6MiKmAh/q35CZqyjdj+jC4sbajRFxZNmxVwMHA++mdA+kbbkc+DPgrWze9EwCNlK6Z1MbpU/MJEmSdhXfAf4lItojYhpwNqXp/UTEXxZ9WgBPUer9+iJin4h4aTGaaAOlXqxvG69xGaV+60jgyrL1k4C1wNMR8TxKfZakMcDwSFK/U4GvZ+YDmflw/w+lm1W/kdKnRn8FPAd4gNLNFl8PkJlXUro30WXAOkohztTivO8ujnuyOM/V26njC0Ar8Bilb3378aDtb6I0X/9u4FFKQ6wp6ugE/gPYC7hqWy9SBFG/Bl4EXFG26ZuUPk17ELizqEGSJGlX8TFgKXAH8FtKX/7xsWLbAuB/gKcp9UEXZub/Urrf0aco9V8PU/rSkH/axmt8h9IHgj8vu90BwJmURiOtA/6dzXssSbuwKN0fTZLGhog4G3huZp6y3Z0lSZIkSdu1tXuXSNIuqZjmdhqbvlFEkiRJkrSTqjZtLSIujohHI+J3FbZHRHwxIpZHxB0RcXC1apE09kXE31K6ofaPMvP6WtcjSbViDyZJkoZbNe95dAlw7Da2H0dpzu0C4HRKd+aXpGckM/89Mydk5t/XuhZJqrFLsAeTJEnDqGrhUfHJ/xPb2OUE4JtZcgOwe0TMrFY9kiRJ44E9mCRJGm61vOfRbEpTTPqtLNatGrxjRJxO6ZMxJkyYcMjznve8ESlQkiSNvFtuueWxzGyvdR1jmD2YJEnawrZ6sF3ihtmZeRFwEcDixYtz6dKlNa5IkiRVS0TcX+saVGIPJknS+LGtHqya9zzangeBuWXLc4p1kiRJqh57MEmStENqGR5dA7y5+MaPw4CnMnOL4dKSJEkaVvZgkiRph1Rt2lpEfAc4GpgWESuBDwGNAJn5b8APgb8AlgMdwP+tVi2SJEnjhT2YJEkablULjzLz5O1sT+Dt1Xp9SZKk8cgeTJIkDbdaTluTJEmSJEnSKGd4JEmSJEmSpIoMjyRJkiRJklSR4ZEkSZIkSZIqMjySJEmSJElSRYZHkiRJkiRJqsjwSJIkSZIkSRUZHkmSJEmSJKkiwyNJkiRJkiRVZHgkSZIkSZKkigyPJEmSJEmSVJHhkSRJkiRJkioyPJIkSZIkSVJFhkeSJEmSJEmqyPBIkiRJkiRJFRkeSZIkSZIkqSLDI0mSJEmSJFVkeCRJkiRJkqSKDI8kSZIkSZJUkeGRJEmSJEmSKjI8kiRJkiRJUkWGR5IkSZIkSarI8EiSJEmSJEkVGR5JkiRJkiSpIsMjSZIkSZIkVWR4JEmSJEmSpIoMjyRJkiRJklSR4ZEkSZIkSZIqMjySJEmSJElSRYZHkiRJkiRJqsjwSJIkSZIkSRUZHkmSJEmSJKkiwyNJkiRJkiRVZHgkSZIkSZKkigyPJEmSJEmSVJHhkSRJkiRJkioyPJIkSZIkSVJFhkeSJEmSJEmqyPBIkiRJkiRJFTXUugBJGikbejawunM1j3U+xqMdj7K6YzWrO1ezumM1a7vW0trQyoTGCQOP/T9tjW1MaNj0fPByXZjDb09m0tnTyfru9aWfnvV0dHfQ0d2x2fLA9u5ie08H9VG/6TqUX5eGts3W9y/3/zTVN9X6bUuSJEljQlXDo4g4FvhXoB74amZ+atD2ecA3gN2Lfc7KzB9WsyZJY09Xb9dACPRox6MDz1d3lpb7w6K1XWu3OLaxrpH21nZ2a96NDT0bNgUXPR1Dfv3ysKmtoW1TyNGwZeCxrWBqQuMEWhpaRkUY1R/2dPR0bBbqdPRsPeApX17fU/a8OKaju4Mkh/TarQ2tmwVDPX09A+ft6O6gu697SOdpqGvYLFAaHPptLYAqv27l13FC4wQa6xt35lcqjSh7MEmSNJyqFh5FRD1wAfAKYCVwc0Rck5l3lu32L8B3M/PLEbEQ+CEwv1o1Sdq1dPV2bRolNCgQGhg11LmapzY+tcWxDXUNtLe2097Wzvzd5rN4xmKmt02nva2d6a3TmdY2jemt05ncPJmI2OL4vuzbLEwaPDJmcKgyODhZ3bGa+3vuH9jW2dM5pPccxOZhVFmA0drYulmYUSmYam1oZWPvxm0HPMV7Gtg+KPTp6OmgL/uGVHNLfctmtbQ1tDG1ZSpzJ83d6oigtsa2iqFOa0Mr9XX123y97t7uLeofHG5VXNezntWdqzf7XfRkz5DeZ2NdY+UQcND72eL9Fdetqb6JYMv/3nZVsyfNHhVhpzZnDyZJkoZbNUceHQosz8x7ASLicuAEoLxxSWC34vlk4KEq1iNplOju7S6FQp2P8lhH6bF/1FD/+tUdq3ly45NbHNsQDQPBz7xJ8zhkxiGlUKgIitpb25neVgqFdub/1NZF3cDok3bad+btAtDb17tp2lbPejq7t5zCVR5EbTbFq3s9D3c8vFngsaF3wzOupbm+eYsAZHLLZGY1zNr2CKlBwVV/SNJQN7IzoBvrG9m9fnd2Z/edPldm0tXXVXF0Vfnoqa2Fb+u61vHI+kc2C7J6s3cY3uWu4aY33kRrQ2uty9CW7MEkSdKwqmbHPxtYUba8EnjhoH0+DPw0It4JTABevrUTRcTpwOkA8+bNG/ZCJQ2P7r5uHu98vBQEVQiEVnesZs3GNVscWx/1TGudRntrO3MmzuHg6QdvEQi1t7Wze/Puu+RIh/q6eiY2TWRi08RhOV9vX++Wo2yKAGNDzwZa6lu2OlKprbGNxjqnX/WLCJrrm2mub2ZKy5SdPl9mDoz6Kp9q1399unuHNuVuV+F/S6OWPZgkSRpWtb5h9snAJZn5uYg4HPhWROybuflcicy8CLgIYPHixUO7aYakqnno6YdYsmIJ96y5Z7MpZGs2rNnivjb1Uc8eLXvQ3tbOrImzOLD9wIGRQ+1tpVBoWus0prZM3SVDoVqpr6tnUtMkJjVNqnUpKhMRtDS00NLQwh6te9S6HGlb7MEkSdKQVTM8ehCYW7Y8p1hX7jTgWIDM/HVEtADTgEerWJekHdSXffzusd+xZMUSlqxcwh/W/AGAPVr2YMaEGcycMJP92vcbCIT6RwxNb5vOlOYp272HjSRpWNmDSZKkYVXN8OhmYEFE7EWpYTkJeMOgfR4AXgZcEhHPB1qA1VWsSdIQdXR38OtVv+a6Fddx/crreXzD49RHPQdNP4gzF5/JUXOOYv7k+bUuU5K0JXswSZI0rKoWHmVmT0S8A/gJpa+AvTgzl0XER4ClmXkN8D7g3yPiDEo3bnxLZjokWqqRh9c/zHUrrmPJyiXctOomuvq6mNQ4iSNmH8HRc4/mxbNfzOTmybUuU5K0DfZgkiRpuFX1nkeZ+UNKX/1avu7ssud3AkdUswZJlfVlH3c9fhdLVi7huhXXcdcTdwEwd9JcTtznRI6eezQHzzjYm+JK0i7GHkySJA2nWt8wW9II6+zp5MZVN7JkxRKuX3k9qztXUxd1HNB+AGcccgZHzzmavSbvRUTUulRJkiRJ0ihgeCSNA6s7VnPdyuu4bsV13LDqBjb0bmBC4wReNOtFHD33aF4y+yXD8jXlkiRJkqSxx/BIGoMyk7ufuHtgOtqyx5cBMGvCLF6z4DUcNfcoXjDjBTTWOx1NkiRJkrRthkfSGLGxdyM3rbqpNMJo5XU8vP5hgmC/9v1410Hv4qi5R7Fg9wVOR5MkSZIk7RDDI2kX9ljnY/xi5S9YsmIJv171azp7OmltaOXwmYfztgPexkvmvIRprdNqXaYkSZIkaRdmeCTtQjKTPzz5B65bcR1LVi7ht6t/S5LMaJvB8Xsfz1FzjuLQmYfSXN9c61IlSZIkSWOE4ZE0ynX1drH04aUsWVn6drQHn34QgEV7LOKtB76VY+Yewz5T9nE6miRJkiSpKgyPpFFozYY1/OLB0nS0Xz30K9Z3r6elvoXDZh7G3+z3Nxw550imt02vdZmSJEmSpHHA8EgaBTKTe5+6lyUrlnDdyuv4zerf0Jd9tLe2c+z8Yzlm7jEcOvNQWhtaa12qJEmSJGmcMTySaqS7r5tbH7l1IDBasW4FAM+f+nxO3/90jp5zNM/f4/nURV2NK5UkSZIkjWeGR9IIWtu1lutXXs91K67jlw/+knXd62iqa+LQmYdy6sJTOWruUTxrwrNqXaYkSZIkSQMMj6QRkJlc9Yer+NzSz7Guex1TW6by8j1fzlFzj+LwmYfT1thW6xIlSZIkSdoqwyOpylauW8k5vz6HG1bdwOIZi3n3we9m//b9nY4mSZIkSdolGB5JVdKXfXzn7u/wr7f+K0HwwcM+yGuf+1pDI0mSJEnSLsXwSKqCe5+6lw/98kPcvvp2jph9BB867EPMnDiz1mVJkiRJkrTDDI+kYdTT18Mlyy7hy7d/mZaGFj7+4o/zV8/+KyKi1qVJkiRJkvSMGB5Jw+SeJ+7hg7/8IHc9cRev2PMVfOCFH2Ba67RalyVJkiRJ0k4xPJJ2UldvF1+54ytc/NuLmdw8mfOOPo9X7PmKWpclSZIkSdKwMDySdsIdq+/g7F+ezR+f+iPH730873/B+5ncPLnWZUmSpCHITLK7m+zooK+jg77OztJjRyd9HevJzs7SuvVl2zpL+2ZH56ZjOov9OzqJ1lZaFi6kZdHC0uPChdRPnFjrtypJ0k4xPJKegc6eTs6/7Xy+fde3aW9t58KXXchL5ryk1mVJkjQmZWYpyBkIeMrCnf51AwFP//rOzQOhCqEPPT1DL6S+nrq2NupaW6lrayPaSo/1kyfTOHMmda2t9K5dS8dNN7H2+98fOKxp/nxaFi0qQqVFtCxaSP2kSVX4TUmSVB2GR9IOuvnhm/nQrz7EinUrOPG5J3LGIWcwsclPFCVJeiYeOfcz9KxeTV9nRzECaHDo00l2dOzQOaO5efOAp7WNurY2GqfP2Cz06V9f2rfYv7WVurYJA8ubztNGNDYO+Usweh57jA133smGZcvoXLaMjltvZe0PfjCwvXHPebQsXEjrokUDwVL9ZEcvS5JGJ8MjaYie7nqa8245jyt/fyVzJ83l4j+/mBc86wW1LkuSpF1ax0030bt27UBIU9fWRkP7tFKIUx7uTOgPdspCn7bWTeFOaxt1E9qoa2khGmrf4jZMm8bEI49k4pFHDqzreeIJNiwrBUobli1jwx2/Zd2PfjywvXHOnGJk0qKBaW8NU6bUonxJkjZT+/9llXYB16+8no/8+iOs7lzNqQtP5e0HvZ3WhtZalyVJ0i5vr+9dWesSRkzD1KlMfMmLmfiSFw+s61mzphihdOfASKV1P/nJwPbGWbM2hUlFsNQwdWotypckjWOGR9I2PLnhSc69+Vy+f+/32Xvy3px39Hns375/rcuSJEljRMOUKUw84ggmHnHEwLrep54qBUll097W/exnm46ZOXPgptz9094apk2rRfmSpHHC8Eiq4Kf3/ZSP3/hx1m5cy9/t/3ecvv/pNNU31bosSZI0xtVPnsyEww9nwuGHD6zrXbuWDXfdvWnK27JlPH3ttQPbG6ZPHzTlbRGNM6bXonxJ0hhkeCQN8ljnY3z8ho/zPw/8D8+f+nwuesVF7DN1n1qXJUmSxrH63XZjwgsPZcILDx1Y1/v002y86y46ly0bmPb29JIlkFk6pn0arQsXbTbtrWHGjCHf9FuSpH6GR1IhM/n+vd/n0zd9mg09G3jPwe/h1EWn0lDnn4kkSRp96idOpO0FL6DtBZu+wKNv/Xo23N0/QulONty5jKd/8Qvo6ysds8ceAzfjblm0iNZFi2iYOdNASZK0Tf6/YglY9fQqzrnhHH754C85aPpBnPOic9hr8l61LkuSJGmH1E2YQNshh9B2yCED6/o6Othw9z0D91DasGwZj//yV9DbC0D9lCkDYVL/T+PsWQZKkqQBhkca1/qyjyvvuZLzbjmPJDnr0LM4+XknUxd1tS5NkiRpWNS1tdF28EG0HXzQwLq+DRvYeM89xZS3ZWy48y4ev/hi6OkBSvddal28mGed/UEaZ8yoVemSpFHC8Ejj1gNrH+BDv/oQSx9ZymEzD+PDL/owsyfOrnVZkiRJVVfX0kLrAQfQesABA+v6Nm5k4+9/PzDlbe0PfsB9rzuRORdeSOu+i2pYrSSp1gyPNO709vXy7bu+zfm3nU9jXSPnvOgcXv2cVzs0W5IkjWt1zc207rcfrfvtB8CUU05h5Vvfyv2nnMKscz/Nbn/2ZzWuUJJUK87N0biyfM1y3vSjN/HZpZ/lsFmHcfWrruY1C15jcCRJkjRIyz7PZf53r6Dlec/jwXe9m8f+7Stk8U1ukqTxxZFHGhe6e7v56u++ykV3XMSkxkmce+S5HDv/WEMjSZKkbWiYNo1537iEVf/yQVZ/4Qt0/elenvXRj1LX1FTr0iRJI8jwSGPesseXcfYvz+b3a37PcXsdx1mHnsXUlqm1LkuSJGmXUNfczKxzP03zs/di9b9+ka4VK5lz/pdomGo/JUnjhdPWNGZt6NnA52/5PG/8wRtZs2ENXzzmi5x75LkGR5IkSTsoIpj21rcy+wufZ8Odd3Lf605k4x/+UOuyJEkjxPBIY9Ktj9zK677/Oi7+3cWc8JwTuPpVV3PMvGNqXZYkSdIubbdjj2XPb32L7OrivpNO5unrr691SZKkEWB4pDGlo7uDT9z4Cd7y47fQ3dfNRa+4iHNedA67Ne1W69IkSZLGhNb99mX+ld+lcc95rPj7t/LEN7/pjbQlaYwzPNKY8auHfsVrrnkNl999OSc/72SuOv4qDp91eK3LkiRJGnMan/Us5n/720x86TE88olP8vA555Dd3bUuS5JUJd4wW7u8tV1r+ezNn+U/l/8n83ebzzeO+wYHTT+o1mVJkiSNaXVtbcz54hdZ/fkv8Pi//ztd99/PnC98gfrJk2tdmiRpmDnySLu0nz/wc1519au45o/XcNq+p/G9479ncCRJkjRCoq6O6e97LzM/+Uk6lt7Cfa8/ia777qt1WZKkYWZ4pF3SExue4P3XvZ93/++7mdIyhUtfeSnvOeQ9NNc317o0SZKkcWf3V7+KPb9+Mb1PPsmfXn8S62+8qdYlSZKGUVXDo4g4NiLuiYjlEXFWhX1OjIg7I2JZRFxWzXq068tMfnjvD3nV1a/iZw/8jLcf+HYuf+XlLNpjUa1LkyRp1LAHUy20LV7M/Cu/S8O0aTxw2mmsufLKWpckSRomVbvnUUTUAxcArwBWAjdHxDWZeWfZPguAfwKOyMw1ETG9WvVo1/fI+kf42A0fY8nKJew3bT8+8qKP8Jwpz6l1WZIkjSr2YKqlprlzmX/5d3jwjPfy8AfPpuvePzH9zPcR9fW1Lk2StBOGNPIoSk6JiLOL5XkRceh2DjsUWJ6Z92ZmF3A5cMKgff4WuCAz1wBk5qM7Vr7Gg8zkP37/H7z6v17NDatu4MzFZ/Kt475lcCRJGvPswbQrqp80ibn/9mWmnHIKT3z966x8+zvofXp9rcuSJO2EoY48uhDoA14KfARYB/wH8IJtHDMbWFG2vBJ44aB9ngsQEb8E6oEPZ+aPB58oIk4HTgeYN2/eEEveMVfcfQU3P3JzVc6tnbPq6VXc8dgdLJ6xmHNedA7zdqvOfwOSJI1CY74H09gUDQ0861/+maZn78UjH/8E97/xjcy98AIaZ8+udWmSpGdgqOHRCzPz4Ii4DaAY3tw0TK+/ADgamANcHxH7ZeaT5Ttl5kXARQCLFy/OYXjdLTzS8Qi/X/P7apxaO6mhroEPHvZBXvvc11IX3uNdkjSujPkeTGPb1De8gaY99+TB95zBn058PXMvOJ/WAw+sdVmSpB001PCou5g/nwAR0U7pU7BteRCYW7Y8p1hXbiVwY2Z2A3+KiN9TamRGfAjQuw5+F+86+F0j/bKSJEnbMuZ7MI19E484gvlXXM6Kv38r97/5VGZ+4hNM/stX1rosSdIOGOowji8C/wlMj4iPA/8P+MR2jrkZWBARexWfkJ0EXDNon6spfeJFREyjNITjEP0iAAAbBklEQVT63iHWJEmSNNbZg2lMaH72s5l/xeW07r8/D515Jqu/+CWyb3s5qCRptBjSyKPMvDQibgFeBgTwqsy8azvH9ETEO4CfUJpLf3FmLouIjwBLM/OaYtufRcSdQC/wD5n5+E68H0mSpDHDHkxjScOUKcy7+GusOuccHrvwQjb+6V5mffKT1LW01Lo0SdJ2ROb2p69HxNStrF5XDHUeUYsXL86lS5eO9MtKkqQREhG3ZObiWtcxGtiDaSzKTJ64+GIe/eznaNl3X+ZccD6N06fXuixJGve21YMNddrarcBq4PfAH4rn90XErRFxyPCUKUmSpEHswTTmRAR7nHYacy44n41//CP3nfh6Nty1zQF1kqQaG2p49DPgLzJzWmbuARwH/DfwNkpfIStJkqThZw+mMWvSS1/K/Eu/DRHc98ZTWHfttbUuSZJUwVDDo8My8yf9C5n5U+DwzLwBaK5KZZIkSbIH05jW8vzns9d3r6D5Oc9h5TveyeNf/SpDua2GJGlkDTU8WhUR/xgRexY/7wceKb461q9JkCRJqg57MI15De3t7PnNb7Dbccfy6Gc/x6oP/DPZ1VXrsiRJZYYaHr0BmEPpa12vBuYV6+qBE6tTmiRJ0rhnD6Zxoa6lhVmf+xzT3v52nvrP/+T+v/5retasqXVZkqRCw1B2yszHgHdW2Lx8+MqRJElSP3swjScRQfs730HTs/di1T99gPtOfD1z/+3LNO+9d61Lk6Rxb0jhUUS0A+8HFgEt/esz86VVqkuSJGncswfTeDT5la+kac4cVrz9Hdx30snM/vznmfjiI2pdliSNa0OdtnYpcDewF3AOcB9wc5VqkiRJUok9mMal1gMOYK/vXkHjrFms+Lu/44lLL611SZI0rg01PNojM78GdGfmdZn514CfeEmSJFWXPZjGrcZZs9jz0kuZeOSRPPLRj/HwRz5K9vTUuixJGpeGGh51F4+rIuKVEXEQMLVKNUmSJKnEHkzjWv3ECcw5/0tM/eu/Zs1ll7Hi7/6e3rVra12WJI07Qw2PPhYRk4H3AWcCXwXeU7WqJEmSBPZgElFfz4z3/wMzP/ZR1t94I/eddDJdDzxQ67IkaVwZani0JjOfyszfZeYxmXkI8EQ1C5MkSZI9mNRv99e+lnlf+xq9jz/OfSe+no6bvf2XJI2UoYZHXxriOkmSJA0fezCpzIQXHsr8715B/ZQp3P/Xp/HkVf9Z65IkaVxo2NbGiDgceBHQHhHvLdu0G1BfzcIkSZLGK3swqbKmPfdk/uXf4cEzzmDVBz5A171/pP297yXqhvq5uCRpR23vX9gmYCKlkGlS2c9a4LXVLU2SJGncsgeTtqF+8mTmfuUr7H7ySTz+1a+x8p3vom/9+lqXJUlj1jZHHmXmdcB1EXFJZt4/QjVJkiSNa/Zg0vZFYyPPOvtsmp+9N4988pPcd8qbmHvhBTTOnFnr0iRpzNlmeFSmOSIuAuaXH5OZL61GUZIkSQLswaRtigimvukUmubvyYPvOYM/nXgicy+8kNb99qt1aZI0pgw1PLoS+DdKXw/bW71yJEmSVMYeTBqCiS95CfMv/w4r3vo27j/lTcz61CfZ7bjjal2WJI0ZQw2PejLzy1WtRJIkSYPZg0lD1LxgAfO/ewUr3/FOHjzjvWy8916mve1tREStS5OkXd5Qv5Lg+xHxtoiYGRFT+3+qWpkkSZLswaQd0DB1KvMu+TqTTziBx750Pg+d+Q/0bdxY67IkaZc31JFHpxaP/1C2LoFnD285kiRJKmMPJu2guqYmZn7qkzTtvTerzzuPrpUrmHv++TS0t9e6NEnaZQ0pPMrMvapdiCRJkjZnDyY9MxHBtNP/lqb5e/LQP57Fn058Pbu/7rW0HXggLfsfQP3ECbUuUZJ2KUMKjyKiDXgvMC8zT4+IBcA+mfnfVa1OkiRpHLMHk3bObn/2ZzTOns2qD36Qx750PmRCXR3NCxbQeuCBxc8BNM2f772RJGkbhjpt7evALcCLiuUHKX37h42LJElS9diDSTupddEinn3VVfSuW0fnb+6g8/bb6bztNtb+8Ic8ecUVANRPmULrAQdsCpT225e6CY5OkqR+Qw2P9s7M10fEyQCZ2RFG85IkSdVmDyYNk/pJk5j44iOY+OIjAMi+Prr++Ec6br+dzttup/P223l6yZJi53qa93kubQOjkw6kce5cRydJGreGGh51RUQrpRs0EhF7A35tgSRJUnXZg0lVEsX0teYFC5jyutcB0Pvkk3TeURqd1HHbbTx19X+x5rLvAFC/xx4D09zaDjyQln33pa61tZZvQZJGzFDDow8BPwbmRsSlwBHAW6pVlCRJkgB7MGlE1e++OxOPPJKJRx4JQPb2snH58oGRSZ23387T115b2rmhgZZ99ikFSgcdVBqdNHuWo5MkjUmRmUPbMWIP4DAggBsy87FqFlbJ4sWLc+nSpbV4aUmSNAIi4pbMXFzrOkYLezBpdOlZs6YIkn5Tevztb8mODgDq26dtNtWtZdEi6lpaalyxJA3NtnqwoX7b2quBn2fmD4rl3SPiVZl59TDWKUmSpDL2YNLo0zBlCpOOOYZJxxwDQPb0sPH3vy/dO6kIldb97H9KOzc20vL85w9MdWs98EAaZs50dJKkXc6QRh5FxO2ZeeCgdbdl5kFVq6wCP/WSJGlsc+TRJvZg0q6p5/HHB6a5dd52O52/+x25YQMADdOnD0xzaz3wgNLopKamGlcsScMw8gio24ljJUmS9MzYg0m7oIY99mDSy17GpJe9DIDs7mbD3fdsCpRuv511P/kJANHYSMvChZsCpYMOpHHGjFqWL2kIsq+P7OmB7m6yp2fgh7Ln2dNDdvdAT/dmy9lbtl93/77dpXX9y/37dJefr5um2bOZeuqpI/5+h9p8LI2I84ALiuW3A7dUpyRJkiQV7MGkMSAaG2ndb19a99sX3nQKAN2PPkrnb34zcDPuNZddxhOXXAJAw8yZpaluRaDU8rznEY5O0hiSmaXQpT94KX/s7n/sLoUug/fpKl/u32/TMQPHb7FP72YhDN2DQp6e7e9THgzR1zdyv7DGRqKhgWhooPWgA0d1ePRO4IPAFZS+KvZnlJoXSZIkVY89mDRGNU6fTuMrXsFur3gFANnVxYa776bz9tvpuO220r2TfvRjAKK5mZZFiwamujXOnEk0NVPX3EQ0N5d+mppK098aG72n0jiTvb1lwUs32VUWugwOaAYHLAPbi4Bla/vsUFDTsyn06dpyff9zenqq+0tpaCDKApdobISGeqKhbF1DAzQ2DKyra2kprSvbvmmfTftFY9k+9aXlTcdt2mfzdfWbLzeWnX9QrdTXb7k8Cv6mtxseRUQ98N+ZecwI1CNJkiTswaTxJpqaaN1/f1r335+pb34zAN2PPFIamXTbbaXRSd/6Fk9cfPF2ThQDgVJd05bh0sBycxN1TWXPi32iWDcQTDVtbXuxXH7u/u3NzUTd1mbc1lZmQm8v2ddXCj/KH3t7yd7egcfs7YW+PrKnF/p6Nz329kFvD9nbV5pS1Ne32XH09lYIZnrI7q6y0TSbBz1bBjuD13WVRsBU2F7VETCNjZuCjK080ti/3Fj672DChG3vU36OxiJEaWwsC1DK9mkatG9DQzECZ/Njt3w9A9Rq2G54lJm9EdEXEZMz86mRKEqSJGm8sweT1DhjBo3H/jm7HfvnAPR1dbHxrrvoWbOG3NhFdm0kN26kb+PGYrm0bmB548Ziuf956bH3qadKyxs30tdVtu/GjaUwYqcLb9wsqBp4XjzWNW8KqaK+juzpJft6oT+U6S0LZcpCnv5t/YFO9vWWphltJQDqP7b/+YhOMaqkvn5QcFIWfDQ1DgpGGqlra9ty/8bNlwdCl8amLYOZweHL1vbfyjk3jbgxhNEmQ5229jTw24j4GbC+f2VmvqsqVUmSJAnswSSVqWtqovWAA6r6GtnXVxrR0h8ulYVUpbCpa1BwVR5SlYVYg4Opsu19nRvIJ5+ir2sj9PSWphPVN0B9HVFXT9TXl4KWurpSmNHcXNqnrlhfX1/at76BqK+Dunqiob70WHbswDEDx5Yds8WxdZtq6H+NoRzT0FAcWzpmIKTZLPApApn6+qpeO6mahhoeXVX8SJIkaeTYg0kaUVFXVwprmptrXYqkUWRI4VFmfiMiWoF5mXlPlWuSJEkS9mCSJGl0GNKdzCLir4DbgR8XywdGxDXVLEySJGm8sweTJEmjwVBvg/9h4FDgSYDMvB14dpVqkiRJUsmHsQeTJEk1NtTwqHsr3/Kx3dvVR8SxEXFPRCyPiLO2sd//iYiMiMVDrEeSJGk8sAeTJEk1N9TwaFlEvAGoj4gFEfEl4FfbOiAi6oELgOOAhcDJEbFwK/tNAt4N3LhDlUuSJI199mCSJKnmhhoevRNYBGwELgOeAt6znWMOBZZn5r2Z2QVcDpywlf0+Cnwa2DDEWiRJksYLezBJklRz2/y2tYhoAf4eeA7wW+DwzOwZ4rlnAyvKllcCLxx0/oOBuZn5g4j4h23UcTpwOsC8efOG+PKSJEm7JnswSZI0mmxv5NE3gMWUmpbjgM8O1wtHRB1wHvC+7e2bmRdl5uLMXNze3j5cJUiSJI1W9mCSJGnU2ObII2BhZu4HEBFfA27agXM/CMwtW55TrOs3CdgXWBIRAM8CromI4zNz6Q68jiRJ0lhjDyZJkkaN7Y086u5/sgNDpfvdDCyIiL0iogk4Cbim7HxPZea0zJyfmfOBGwCbFkmSJHswSZI0imxv5NEBEbG2eB5Aa7EcQGbmbpUOzMyeiHgH8BOgHrg4M5dFxEeApZl5TaVjJUmSxjl7MEmSNGpsMzzKzPqdOXlm/hD44aB1Z1fY9+ideS1JkqSxwh5MkiSNJtubtiZJkiRJkqRxzPBIkiRJkiRJFRkeSZIkSZIkqSLDI0mSJEmSJFVkeCRJkiRJkqSKDI8kSZIkSZJUkeGRJEmSJEmSKjI8kiRJkiRJUkWGR5IkSZIkSarI8EiSJEmSJEkVGR5JkiRJkiSpIsMjSZIkSZIkVWR4JEmSJEmSpIoMjyRJkiRJklSR4ZEkSZIkSZIqMjySJEmSJElSRYZHkiRJkiRJqsjwSJIkSZIkSRUZHkmSJEmSJKkiwyNJkiRJkiRVZHgkSZIkSZKkigyPJEmSJEmSVJHhkSRJkiRJkioyPJIkSZIkSVJFhkeSJEmSJEmqyPBIkiRJkiRJFRkeSZIkSZIkqSLDI0mSJEmSJFVkeCRJkiRJkqSKDI8kSZIkSZJUkeGRJEmSJEmSKjI8kiRJkiRJUkWGR5IkSZIkSarI8EiSJEmSJEkVGR5JkiRJkiSpIsMjSZIkSZIkVWR4JEmSJEmSpIoMjyRJkiRJklSR4ZEkSZIkSZIqMjySJEmSJElSRVUNjyLi2Ii4JyKWR8RZW9n+3oi4MyLuiIhrI2LPatYjSZI0HtiDSZKk4VS18Cgi6oELgOOAhcDJEbFw0G63AYszc3/ge8C51apHkiRpPLAHkyRJw62aI48OBZZn5r2Z2QVcDpxQvkNm/m9mdhSLNwBzqliPJEnSeGAPJkmShlU1w6PZwIqy5ZXFukpOA360tQ0RcXpELI2IpatXrx7GEiVJksYcezBJkjSsRsUNsyPiFGAx8Jmtbc/MizJzcWYubm9vH9niJEmSxih7MEmSNBQNVTz3g8DcsuU5xbrNRMTLgX8G/n979xpr2VnWAfz/2LFUqOlwMY1MC1NklExQ29pgjYhC+VCk6UjSaAFDS2r6QYjYSEz1g4kkGrDGqohEwiWVECkOTTgxBCRTPhATmhYK9DphUoG2FoudMlxMphQeP6w1ZTvMKqXd++zVOb9fcnLW9Zz37CfvzpP/XmudX+/uwyscDwDAVqAHAwCWapVXHt2YZFdVnVFVJya5OMnG4gFVdVaSf0pyYXffv8KxAABsFXowAGCpVhYedffDSd6Y5ONJ7kjyoe6+rareUlUXjoddleTkJP9aVZ+rqo2JHwcAwGOgBwMAlm2Vt62luz+a5KNHbfuzheWXr/L3AwBsRXowAGCZZvHAbAAAAADmSXgEAAAAwCThEQAAAACThEcAAAAATBIeAQAAADBJeAQAAADAJOERAAAAAJOERwAAAABMEh4BAAAAMEl4BAAAAMAk4REAAAAAk4RHAAAAAEwSHgEAAAAwSXgEAAAAwCThEQAAAACThEcAAAAATBIeAQAAADBJeAQAAADAJOERAAAAAJOERwAAAABMEh4BAAAAMEl4BAAAAMAk4REAAAAAk4RHAAAAAEwSHgEAAAAwSXgEAAAAwCThEQAAAACThEcAAAAATBIeAQAAADBJeAQAAADAJOERAAAAAJOERwAAAABMEh4BAAAAMEl4BAAAAMAk4REAAAAAk4RHAAAAAEwSHgEAAAAwSXgEAAAAwCThEQAAAACThEcAAAAATBIeAQAAADBppeFRVZ1fVfur6kBVXXmM/U+pqmvH/TdU1c5VjgcAYCvQgwEAy7Sy8KiqTkjyjiSvSLI7yauravdRh12W5MHufn6Sq5O8bVXjAQDYCvRgAMCyrfLKoxclOdDdd3X3Q0k+mGTPUcfsSXLNuLw3yXlVVSscEwDA8U4PBgAs1bYV/uwdSe5eWL8nyS9PHdPdD1fVoSTPTPI/iwdV1eVJLh9Xv1VV+1cy4uRZR/9uZkNt5kld5ktt5kldHpvnrnsAT3J6MJZFXeZLbeZJXeZLbR6byR5sleHR0nT3u5K8a9W/p6pu6u5zVv17+NGpzTypy3ypzTypC082erCtTV3mS23mSV3mS22euFXetnZvktMX1k8btx3zmKraluSUJA+scEwAAMc7PRgAsFSrDI9uTLKrqs6oqhOTXJxk46hjNpJcMi5flOT67u4VjgkA4HinBwMAlmplt62N98+/McnHk5yQ5L3dfVtVvSXJTd29keQ9Sd5fVQeSHMzQ3KzTyi/L5nFTm3lSl/lSm3lSF1ZOD8YSqct8qc08qct8qc0TVD5kAgAAAGDKKm9bAwAAAOBJTngEAAAAwCTh0aiqzq+q/VV1oKquXPd4tpKqOr2qPllVt1fVbVX1pnH7M6rqE1X1xfH708ftVVV/P9bqC1V19nr/guNbVZ1QVTdX1b+N62dU1Q3j63/t+DDWVNVTxvUD4/6d6xz38a6qtlfV3qq6s6ruqKpfMWfWr6quGN/Hbq2qf6mqk8wZmKb/Wi892LzpweZJDzZPerDVEx5leGNO8o4kr0iyO8mrq2r3eke1pTyc5I+6e3eSc5O8YXz9r0yyr7t3Jdk3ridDnXaNX5cneefmD3lLeVOSOxbW35bk6u5+fpIHk1w2br8syYPj9qvH41idv0vyse5+QZJfzFAjc2aNqmpHkj9Ick53vzDDg4ovjjkDx6T/mgU92LzpweZJDzYzerDNITwavCjJge6+q7sfSvLBJHvWPKYto7vv6+7PjsvfzPAGvCNDDa4ZD7smyW+Ny3uS/HMPPp1ke1X99CYPe0uoqtOSvDLJu8f1SvKyJHvHQ46uy5F67U1y3ng8S1ZVpyR5SYb/lpTufqi7vx5zZg62JfmJqtqW5KlJ7os5A1P0X2umB5svPdg86cFmTQ+2YsKjwY4kdy+s3zNuY5ONlwyeleSGJKd2933jrq8mOXVcVq/N87dJ/jjJ98b1Zyb5enc/PK4vvvaP1GXcf2g8nuU7I8nXkrxvvJz93VX1tJgza9Xd9yb56yRfydCwHErymZgzMMV704zowWZHDzZPerAZ0oNtDuERs1FVJyf5cJI/7O5vLO7r7k7SaxnYFlVVFyS5v7s/s+6x8AO2JTk7yTu7+6wk3873L49OYs6sw/h8gz0ZGstnJ3lakvPXOiiAx0APNi96sFnTg82QHmxzCI8G9yY5fWH9tHEbm6SqfjxD0/KB7r5u3PzfRy7rHL/fP25Xr83xq0kurKovZbiV4GUZ7vHePl4Omvz/1/6Ruoz7T0nywGYOeAu5J8k93X3DuL43QyNjzqzXy5P8Z3d/rbu/k+S6DPPInIFj8940A3qwWdKDzZcebJ70YJtAeDS4Mcmu8WnsJ2Z4uNbGmse0ZYz3l74nyR3d/TcLuzaSXDIuX5LkIwvbXzf+94JzkxxauEyUJenuP+nu07p7Z4Y5cX13vzbJJ5NcNB52dF2O1Oui8XifuqxAd381yd1V9XPjpvOS3B5zZt2+kuTcqnrq+L52pC7mDByb/mvN9GDzpAebLz3YbOnBNkF5jQZV9ZsZ7i0+Icl7u/sv1jykLaOqXpzkU0luyffv6/7TDPfcfyjJc5J8Oclvd/fB8Q3hHzJcivi/SV7f3Tdt+sC3kKr6jSRv7u4Lqup5GT4Fe0aSm5P8bncfrqqTkrw/w/MSDia5uLvvWteYj3dVdWaGh2iemOSuJK/P8IGAObNGVfXnSX4nw38wujnJ72W4r96cgWPQf62XHmz+9GDzowebJz3Y6gmPAAAAAJjktjUAAAAAJgmPAAAAAJgkPAIAAABgkvAIAAAAgEnCIwAAAAAmCY+Alaqqb43fd1bVax7H+dur6vcX1p9dVXuXOUYAgOOJ/gtYNuERsFl2Jjlm81JV2x7lvO1JHmleuvu/uvui5Q4NAOC4tDP6L2AJhEfAZnlrkl+rqs9V1RVVdWlVbVTV9Un2VdXJVbWvqj5bVbdU1Z6F835mPO+q8RO0W5Okqk6qqveNx99cVS8dt19aVddV1ceq6otV9Vfr+ZMBANZK/wUsxaOlzQDLdGWSN3f3BcnQYCQ5O8kvdPfB8dOvV3X3N6rqWUk+XVUb43kv7O4zx/N2LvzMNyTp7v75qnpBkn+vqp8d952Z5Kwkh5Psr6q3d/fdK/8rAQDmQ/8FLIXwCFinT3T3wXG5kvxlVb0kyfeS7Ehy6g85/8VJ3p4k3X1nVX05yZHmZV93H0qSqro9yXOTaF4AgK1O/wX8yIRHwDp9e2H5tUl+Kskvdfd3qupLSU56Aj/78MLyd+P9DgAg0X8Bj4NnHgGb5ZtJfvJR9p+S5P6xcXlphk+qfth5n8rQ9GS8XPo5SfYvZ7gAAE96+i9gKYRHwGb5QpLvVtXnq+qKY+z/QJJzquqWJK9LcmeSdPcDSf6jqm6tqquOOucfk/zYeM61SS7t7sMBACDRfwFLUt297jEAAAAAMFOuPAIAAABgkvAIAAAAgEnCIwAAAAAmCY8AAAAAmCQ8AgAAAGCS8AgAAACAScIjAAAAACb9H6weJt83jjDwAAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "draw_pot(\"3 layer random initialization\",iters,trainLoss,validationLoss,trainAccu,validationAccu)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eL89QCr9VhKJ",
        "outputId": "7e718a09-7526-467b-d8c3-09c2ddd73634"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "5 layer\n",
            "trainLoss [0.6905746943937583, 0.5828118327328489, 0.5815126848192232, 0.5814899627518391, 0.5814895372977732, 0.5814895214280323, 0.5814895159602556, 0.5814895118708573, 0.5814895079783767, 0.5814895040034584]\n",
            "validationLoss [0.6900941384179446, 0.7057191817154164, 0.717958000101999, 0.7197292797789554, 0.7199724088118116, 0.7200055724782081, 0.720010095334078, 0.7200107122812027, 0.7200107959235507, 0.7200108068307192]\n",
            "trainAccuracy [0.6900941384179446, 0.7057191817154164, 0.717958000101999, 0.7197292797789554, 0.7199724088118116, 0.7200055724782081, 0.720010095334078, 0.7200107122812027, 0.7200107959235507, 0.7200108068307192]\n",
            "validationAccuracy [0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375]\n",
            "\n",
            " \n",
            "for best trainLoss \n",
            "trainloss  0.5814895040034584\n",
            "trainAccuracy 0.7317708333333333 \n",
            "\n",
            "\n",
            " \n",
            "for best validation loss \n",
            "validation loss  0.6900941384179446\n",
            "validation Accuracy 0.59375 \n",
            "\n",
            "\n",
            " \n",
            "for best trainAccuracy  \n",
            "trainAccuracy 0.7317708333333333\n",
            "trainLoss 0.6905746943937583 \n",
            "\n",
            "\n",
            " \n",
            "for best validationAccuracy  \n",
            "validationAccu 0.59375\n",
            "trainLoss 0.6905746943937583 \n",
            "\n"
          ]
        }
      ],
      "source": [
        "iters,trainLoss,validationLoss,trainAccu,validationAccu = implementNetwork(X_train, y_train,X_val,y_val, [16,25,52,26, 1], itersNum = 1000)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 677
        },
        "id": "3zWmrr03Vk10",
        "outputId": "0413499f-7bfa-4f20-86e4-4a33c248aa96"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x720 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "draw_pot(\"5 layer random initialization\",iters,trainLoss,validationLoss,trainAccu,validationAccu)\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "iters,trainLoss,validationLoss,trainAccu,validationAccu = implementNetwork(X_train, y_train,X_val,y_val, [16, 1],randomInit=0, itersNum = 1000)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dK14xNOCV07x",
        "outputId": "70406bad-0d76-4bd1-da18-5c397119e9d0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2 layer\n",
            "trainLoss [0.6931471805599452, 0.46144960049106115, 0.4122574311755374, 0.3883387187118766, 0.3733395072310082, 0.3626461276375533, 0.3544582043719487, 0.3479079879286152, 0.3425122562186079, 0.33797326282324575]\n",
            "validationLoss [0.6880968167370585, 0.5484632574785481, 0.4975394253424857, 0.47470808624007876, 0.4617380488848315, 0.45344186089791416, 0.4477832018330741, 0.44378024502825913, 0.4408900263906812, 0.43878283197781937]\n",
            "trainAccuracy [0.6880968167370585, 0.5484632574785481, 0.4975394253424857, 0.47470808624007876, 0.4617380488848315, 0.45344186089791416, 0.4477832018330741, 0.44378024502825913, 0.4408900263906812, 0.43878283197781937]\n",
            "validationAccuracy [0.59375, 0.7083333333333334, 0.7291666666666667, 0.7604166666666667, 0.7708333333333333, 0.78125, 0.7916666666666667, 0.8020833333333334, 0.8020833333333334, 0.8125]\n",
            "\n",
            " \n",
            "for best trainLoss \n",
            "trainloss  0.33797326282324575\n",
            "trainAccuracy 0.8359375 \n",
            "\n",
            "\n",
            " \n",
            "for best validation loss \n",
            "validation loss  0.43878283197781937\n",
            "validation Accuracy 0.8125 \n",
            "\n",
            "\n",
            " \n",
            "for best trainAccuracy  \n",
            "trainAccuracy 0.8359375\n",
            "trainLoss 0.3425122562186079 \n",
            "\n",
            "\n",
            " \n",
            "for best validationAccuracy  \n",
            "validationAccu 0.8125\n",
            "trainLoss 0.33797326282324575 \n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "draw_pot(\"2 layer zero initialization\",iters,trainLoss,validationLoss,trainAccu,validationAccu)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 677
        },
        "id": "qUnkR8r9WfiX",
        "outputId": "1a8746cc-ef3b-476f-f94d-afe9aead81d0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x720 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "iters,trainLoss,validationLoss,trainAccu,validationAccu = implementNetwork(X_train, y_train,X_val,y_val, [16,25, 1],randomInit=0, itersNum = 1000)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DSKk1IzXZPmb",
        "outputId": "dbaa194d-9bef-47f3-bd71-fd44aa19734c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "3 layer\n",
            "trainLoss [0.6931471805599452, 0.582855331239272, 0.5815139203797091, 0.5814900215866725, 0.5814895720264031, 0.5814895635065576, 0.5814895633449282, 0.5814895633418616, 0.5814895633418034, 0.5814895633418021]\n",
            "validationLoss [0.691041474643566, 0.7055082426456362, 0.7179052791436288, 0.7197189209030622, 0.7199705737747442, 0.7200052722969496, 0.7200100525377378, 0.7200107110103225, 0.720010801712689, 0.7200108142066048]\n",
            "trainAccuracy [0.691041474643566, 0.7055082426456362, 0.7179052791436288, 0.7197189209030622, 0.7199705737747442, 0.7200052722969496, 0.7200100525377378, 0.7200107110103225, 0.720010801712689, 0.7200108142066048]\n",
            "validationAccuracy [0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375]\n",
            "\n",
            " \n",
            "for best trainLoss \n",
            "trainloss  0.5814895633418021\n",
            "trainAccuracy 0.7317708333333333 \n",
            "\n",
            "\n",
            " \n",
            "for best validation loss \n",
            "validation loss  0.691041474643566\n",
            "validation Accuracy 0.59375 \n",
            "\n",
            "\n",
            " \n",
            "for best trainAccuracy  \n",
            "trainAccuracy 0.7317708333333333\n",
            "trainLoss 0.582855331239272 \n",
            "\n",
            "\n",
            " \n",
            "for best validationAccuracy  \n",
            "validationAccu 0.59375\n",
            "trainLoss 0.6931471805599452 \n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "draw_pot(\"3 layer zero initialization\",iters,trainLoss,validationLoss,trainAccu,validationAccu)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 677
        },
        "id": "bhDBPGG6ZTup",
        "outputId": "ff8f0b4c-c4f8-4d68-f468-9a04d4954e9d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x720 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "iters,trainLoss,validationLoss,trainAccu,validationAccu = implementNetwork(X_train, y_train,X_val,y_val, [16,25,52,26, 1],randomInit=0, itersNum = 1000)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AecplmnFZcfk",
        "outputId": "c5ddcbe1-12ac-4b9d-b81b-69b65c93d10b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "5 layer\n",
            "trainLoss [0.6931471805599452, 0.582855331239272, 0.5815139203797091, 0.5814900215866725, 0.5814895720264031, 0.5814895635065576, 0.5814895633449282, 0.5814895633418616, 0.5814895633418034, 0.5814895633418021]\n",
            "validationLoss [0.691041474643566, 0.7055082426456362, 0.7179052791436288, 0.7197189209030622, 0.7199705737747442, 0.7200052722969496, 0.7200100525377378, 0.7200107110103225, 0.720010801712689, 0.7200108142066048]\n",
            "trainAccuracy [0.691041474643566, 0.7055082426456362, 0.7179052791436288, 0.7197189209030622, 0.7199705737747442, 0.7200052722969496, 0.7200100525377378, 0.7200107110103225, 0.720010801712689, 0.7200108142066048]\n",
            "validationAccuracy [0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375, 0.59375]\n",
            "\n",
            " \n",
            "for best trainLoss \n",
            "trainloss  0.5814895633418021\n",
            "trainAccuracy 0.7317708333333333 \n",
            "\n",
            "\n",
            " \n",
            "for best validation loss \n",
            "validation loss  0.691041474643566\n",
            "validation Accuracy 0.59375 \n",
            "\n",
            "\n",
            " \n",
            "for best trainAccuracy  \n",
            "trainAccuracy 0.7317708333333333\n",
            "trainLoss 0.582855331239272 \n",
            "\n",
            "\n",
            " \n",
            "for best validationAccuracy  \n",
            "validationAccu 0.59375\n",
            "trainLoss 0.6931471805599452 \n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "draw_pot(\"5 layer zero initialization\",iters,trainLoss,validationLoss,trainAccu,validationAccu)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 677
        },
        "id": "gi81k26TZjIc",
        "outputId": "89f15097-a7b0-415a-d80b-ea1879f5f81f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x720 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "5Bh4zR2VuBNm"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [
        "41e1rLFSCply"
      ],
      "name": "finalDP",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}