Assignment3_LSTM.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyOUc6Dh/b8MEUfYZZyck3Aa",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jasminaaa20/MHealthActivityRecognition/blob/main/Assignment3_LSTM.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Mount Drive"
      ],
      "metadata": {
        "id": "9waDhhUZSoP3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import files, drive\n",
        "drive.mount('/content/drive/')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bqQZ5_l6P4H5",
        "outputId": "c9ee777b-bb78-4f34-8737-6fa0309de0ee"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive/\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Create the combined dataframe using the dataset"
      ],
      "metadata": {
        "id": "RKyKpTv0MZ90"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd"
      ],
      "metadata": {
        "id": "HDyTF9woN0tk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "comdf = pd.DataFrame()"
      ],
      "metadata": {
        "id": "z_BMrcEoQNTB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Loop through and add all 10 subjects' sensor data to dataframe\n",
        "for i in range(1, 11):\n",
        "    df = pd.read_csv(f'/content/drive/MyDrive/datasets/MHEALTHDATASET/mHealth_subject{i}.log', header=None, sep='\\t')\n",
        "    # Note: Excluding the ECG data collected with the chest sensor\n",
        "    df = df.rename(columns= {\n",
        "        0: 'acx',\n",
        "        1: 'acy',\n",
        "        2: 'acz',\n",
        "        3: 'ecsig1',\n",
        "        4: 'ecsig2',\n",
        "        5: 'alx',\n",
        "        6: 'aly',\n",
        "        7: 'alz',\n",
        "        8: 'glx',\n",
        "        9: 'gly',\n",
        "        10: 'glz',\n",
        "        11: 'mlx',\n",
        "        12: 'mly',\n",
        "        13: 'mlz',\n",
        "        14: 'arx',\n",
        "        15: 'ary',\n",
        "        16: 'arz',\n",
        "        17: 'grx',\n",
        "        18: 'gry',\n",
        "        19: 'grz',\n",
        "        20: 'mrx',\n",
        "        21: 'mry',\n",
        "        22: 'mrz',\n",
        "        23: 'activity'\n",
        "    })\n",
        "    # introduce new column to differentiate between the subjects for creating the train-test split\n",
        "    df['subject'] = f'subject{i}'\n",
        "    comdf = pd.concat([comdf, df])"
      ],
      "metadata": {
        "id": "rIzzQoLZQSt4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "comdf.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HW3CDlxdZwXp",
        "outputId": "40db3ee6-8c65-4bf0-c9b0-0ab156524dce"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1215745, 25)"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Additional adjustments to the dataset"
      ],
      "metadata": {
        "id": "cd_E9ccYKBu9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Drop the rows where the subject is inactive\n",
        "comdf = comdf[comdf['activity']!=0]\n",
        "comdf.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G3Hu9BEYbZeh",
        "outputId": "e405aae2-5d7f-4eba-ced1-14b8677bc1b1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(343195, 25)"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "comdf.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-MRY0k08Z3Jm",
        "outputId": "f70d8284-99ad-4bda-d225-90631b715236"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                 acx            acy            acz         ecsig1  \\\n",
              "count  343195.000000  343195.000000  343195.000000  343195.000000   \n",
              "mean       -7.485271      -0.140923      -0.938418       0.003649   \n",
              "std         5.701949       2.799380       4.611164       0.839074   \n",
              "min       -22.438000     -20.188000     -18.401000      -8.619600   \n",
              "25%        -9.707100      -1.294300      -3.686550      -0.213500   \n",
              "50%        -8.800300      -0.292570      -0.881070      -0.071167   \n",
              "75%        -5.117400       0.984775       1.092650       0.163270   \n",
              "max        19.094000      20.927000      26.196000       8.506500   \n",
              "\n",
              "              ecsig2            alx            aly            alz  \\\n",
              "count  343195.000000  343195.000000  343195.000000  343195.000000   \n",
              "mean       -0.008048       1.805184      -9.061487      -0.701985   \n",
              "std         0.857724       4.214175       5.200389       6.457622   \n",
              "min        -8.619600     -22.146000     -19.619000     -19.373000   \n",
              "25%        -0.171640       0.159875     -10.063000      -3.494350   \n",
              "50%        -0.046049       1.372500      -9.597700       0.340070   \n",
              "75%         0.138150       2.924450      -7.549800       1.819300   \n",
              "max         8.519100      20.024000      21.161000      25.015000   \n",
              "\n",
              "                 glx            gly  ...            arx            ary  \\\n",
              "count  343195.000000  343195.000000  ...  343195.000000  343195.000000   \n",
              "mean        0.101167      -0.549942  ...      -3.448201      -5.781446   \n",
              "std         0.458913       0.434721  ...       5.891595       6.578371   \n",
              "min        -1.779200      -2.660400  ...     -22.345000     -18.972000   \n",
              "25%        -0.345080      -0.810510  ...      -5.013900      -9.558100   \n",
              "50%         0.204080      -0.688560  ...      -2.480000      -7.738900   \n",
              "75%         0.487940      -0.487800  ...      -0.539050      -2.359550   \n",
              "max         1.710600       1.752300  ...      19.801000      21.965000   \n",
              "\n",
              "                 arz            grx            gry            grz  \\\n",
              "count  343195.000000  343195.000000  343195.000000  343195.000000   \n",
              "mean        2.381165      -0.201574      -0.410515       0.380693   \n",
              "std         4.177971       0.549524       0.546359       0.515658   \n",
              "min       -18.238000      -1.170600      -2.256700      -1.114200   \n",
              "25%         0.139070      -0.682350      -0.831620      -0.040948   \n",
              "50%         1.819100      -0.300000      -0.597540       0.448280   \n",
              "75%         5.324750       0.278430      -0.045175       0.840520   \n",
              "max        25.741000       1.415700       1.121100       1.528000   \n",
              "\n",
              "                 mrx            mry            mrz       activity  \n",
              "count  343195.000000  343195.000000  343195.000000  343195.000000  \n",
              "mean       -0.532629       1.422140       0.030670       6.169021  \n",
              "std        34.216257      30.285262      82.663004       3.298902  \n",
              "min      -319.030000    -358.130000    -702.570000       1.000000  \n",
              "25%        -6.228350      -7.272000     -12.170000       3.000000  \n",
              "50%         0.360740       0.353930      -0.709300       6.000000  \n",
              "75%         5.224000       8.350250      10.302500       9.000000  \n",
              "max       239.690000     335.250000     657.180000      12.000000  \n",
              "\n",
              "[8 rows x 24 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-b2bc972e-1080-47b4-87bd-5ddc78c21e4b\" 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>acx</th>\n",
              "      <th>acy</th>\n",
              "      <th>acz</th>\n",
              "      <th>ecsig1</th>\n",
              "      <th>ecsig2</th>\n",
              "      <th>alx</th>\n",
              "      <th>aly</th>\n",
              "      <th>alz</th>\n",
              "      <th>glx</th>\n",
              "      <th>gly</th>\n",
              "      <th>...</th>\n",
              "      <th>arx</th>\n",
              "      <th>ary</th>\n",
              "      <th>arz</th>\n",
              "      <th>grx</th>\n",
              "      <th>gry</th>\n",
              "      <th>grz</th>\n",
              "      <th>mrx</th>\n",
              "      <th>mry</th>\n",
              "      <th>mrz</th>\n",
              "      <th>activity</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "      <td>343195.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>-7.485271</td>\n",
              "      <td>-0.140923</td>\n",
              "      <td>-0.938418</td>\n",
              "      <td>0.003649</td>\n",
              "      <td>-0.008048</td>\n",
              "      <td>1.805184</td>\n",
              "      <td>-9.061487</td>\n",
              "      <td>-0.701985</td>\n",
              "      <td>0.101167</td>\n",
              "      <td>-0.549942</td>\n",
              "      <td>...</td>\n",
              "      <td>-3.448201</td>\n",
              "      <td>-5.781446</td>\n",
              "      <td>2.381165</td>\n",
              "      <td>-0.201574</td>\n",
              "      <td>-0.410515</td>\n",
              "      <td>0.380693</td>\n",
              "      <td>-0.532629</td>\n",
              "      <td>1.422140</td>\n",
              "      <td>0.030670</td>\n",
              "      <td>6.169021</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>5.701949</td>\n",
              "      <td>2.799380</td>\n",
              "      <td>4.611164</td>\n",
              "      <td>0.839074</td>\n",
              "      <td>0.857724</td>\n",
              "      <td>4.214175</td>\n",
              "      <td>5.200389</td>\n",
              "      <td>6.457622</td>\n",
              "      <td>0.458913</td>\n",
              "      <td>0.434721</td>\n",
              "      <td>...</td>\n",
              "      <td>5.891595</td>\n",
              "      <td>6.578371</td>\n",
              "      <td>4.177971</td>\n",
              "      <td>0.549524</td>\n",
              "      <td>0.546359</td>\n",
              "      <td>0.515658</td>\n",
              "      <td>34.216257</td>\n",
              "      <td>30.285262</td>\n",
              "      <td>82.663004</td>\n",
              "      <td>3.298902</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-22.438000</td>\n",
              "      <td>-20.188000</td>\n",
              "      <td>-18.401000</td>\n",
              "      <td>-8.619600</td>\n",
              "      <td>-8.619600</td>\n",
              "      <td>-22.146000</td>\n",
              "      <td>-19.619000</td>\n",
              "      <td>-19.373000</td>\n",
              "      <td>-1.779200</td>\n",
              "      <td>-2.660400</td>\n",
              "      <td>...</td>\n",
              "      <td>-22.345000</td>\n",
              "      <td>-18.972000</td>\n",
              "      <td>-18.238000</td>\n",
              "      <td>-1.170600</td>\n",
              "      <td>-2.256700</td>\n",
              "      <td>-1.114200</td>\n",
              "      <td>-319.030000</td>\n",
              "      <td>-358.130000</td>\n",
              "      <td>-702.570000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>-9.707100</td>\n",
              "      <td>-1.294300</td>\n",
              "      <td>-3.686550</td>\n",
              "      <td>-0.213500</td>\n",
              "      <td>-0.171640</td>\n",
              "      <td>0.159875</td>\n",
              "      <td>-10.063000</td>\n",
              "      <td>-3.494350</td>\n",
              "      <td>-0.345080</td>\n",
              "      <td>-0.810510</td>\n",
              "      <td>...</td>\n",
              "      <td>-5.013900</td>\n",
              "      <td>-9.558100</td>\n",
              "      <td>0.139070</td>\n",
              "      <td>-0.682350</td>\n",
              "      <td>-0.831620</td>\n",
              "      <td>-0.040948</td>\n",
              "      <td>-6.228350</td>\n",
              "      <td>-7.272000</td>\n",
              "      <td>-12.170000</td>\n",
              "      <td>3.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>-8.800300</td>\n",
              "      <td>-0.292570</td>\n",
              "      <td>-0.881070</td>\n",
              "      <td>-0.071167</td>\n",
              "      <td>-0.046049</td>\n",
              "      <td>1.372500</td>\n",
              "      <td>-9.597700</td>\n",
              "      <td>0.340070</td>\n",
              "      <td>0.204080</td>\n",
              "      <td>-0.688560</td>\n",
              "      <td>...</td>\n",
              "      <td>-2.480000</td>\n",
              "      <td>-7.738900</td>\n",
              "      <td>1.819100</td>\n",
              "      <td>-0.300000</td>\n",
              "      <td>-0.597540</td>\n",
              "      <td>0.448280</td>\n",
              "      <td>0.360740</td>\n",
              "      <td>0.353930</td>\n",
              "      <td>-0.709300</td>\n",
              "      <td>6.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>-5.117400</td>\n",
              "      <td>0.984775</td>\n",
              "      <td>1.092650</td>\n",
              "      <td>0.163270</td>\n",
              "      <td>0.138150</td>\n",
              "      <td>2.924450</td>\n",
              "      <td>-7.549800</td>\n",
              "      <td>1.819300</td>\n",
              "      <td>0.487940</td>\n",
              "      <td>-0.487800</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.539050</td>\n",
              "      <td>-2.359550</td>\n",
              "      <td>5.324750</td>\n",
              "      <td>0.278430</td>\n",
              "      <td>-0.045175</td>\n",
              "      <td>0.840520</td>\n",
              "      <td>5.224000</td>\n",
              "      <td>8.350250</td>\n",
              "      <td>10.302500</td>\n",
              "      <td>9.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>19.094000</td>\n",
              "      <td>20.927000</td>\n",
              "      <td>26.196000</td>\n",
              "      <td>8.506500</td>\n",
              "      <td>8.519100</td>\n",
              "      <td>20.024000</td>\n",
              "      <td>21.161000</td>\n",
              "      <td>25.015000</td>\n",
              "      <td>1.710600</td>\n",
              "      <td>1.752300</td>\n",
              "      <td>...</td>\n",
              "      <td>19.801000</td>\n",
              "      <td>21.965000</td>\n",
              "      <td>25.741000</td>\n",
              "      <td>1.415700</td>\n",
              "      <td>1.121100</td>\n",
              "      <td>1.528000</td>\n",
              "      <td>239.690000</td>\n",
              "      <td>335.250000</td>\n",
              "      <td>657.180000</td>\n",
              "      <td>12.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>8 rows × 24 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b2bc972e-1080-47b4-87bd-5ddc78c21e4b')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-b2bc972e-1080-47b4-87bd-5ddc78c21e4b 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-b2bc972e-1080-47b4-87bd-5ddc78c21e4b');\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",
              "\n",
              "\n",
              "<div id=\"df-01cceee1-2dc1-459d-85ec-45c58da28119\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-01cceee1-2dc1-459d-85ec-45c58da28119')\"\n",
              "            title=\"Suggest charts.\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-01cceee1-2dc1-459d-85ec-45c58da28119 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "comdf['activity'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "4lqY5yh-j3Gf",
        "outputId": "ca9d11d6-0f43-475f-b2ee-95fada8832df"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1     30720\n",
              "2     30720\n",
              "3     30720\n",
              "4     30720\n",
              "9     30720\n",
              "10    30720\n",
              "11    30720\n",
              "5     30720\n",
              "7     29441\n",
              "8     29337\n",
              "6     28315\n",
              "12    10342\n",
              "Name: activity, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Convert the activity from range [1,12] to range [0,11] for indexing\n",
        "comdf['activity'] = comdf['activity'] - 1"
      ],
      "metadata": {
        "id": "hYtr7GFXl3IV"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "comdf['activity'].unique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SDmOLHHLl_7p",
        "outputId": "783229f7-4f26-4468-85bd-4c01d46a86f9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 0,  1,  2,  3,  5,  6,  7,  8,  9, 10, 11,  4])"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "comdf['subject'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "GDXaFoqzmAx0",
        "outputId": "170159f1-ab33-4c7f-c714-b6a1d3d63cb3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "subject2     35532\n",
              "subject3     35380\n",
              "subject4     35328\n",
              "subject1     35174\n",
              "subject9     34354\n",
              "subject7     34253\n",
              "subject5     33947\n",
              "subject10    33690\n",
              "subject8     33332\n",
              "subject6     32205\n",
              "Name: subject, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Create the train-test split"
      ],
      "metadata": {
        "id": "sZNoZT8KKnOE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split"
      ],
      "metadata": {
        "id": "D7EEqDUWNvcy"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "trainx = pd.DataFrame()\n",
        "trainy = pd.DataFrame()\n",
        "testx = pd.DataFrame()\n",
        "testy = pd.DataFrame()"
      ],
      "metadata": {
        "id": "46MvsXQzmLjM"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# For each activity within each test subject, take 70% of data for training and 30% of data for testing\n",
        "for i in range(1,11):\n",
        "  for activity in range(0,12):\n",
        "    sub = comdf[(comdf['activity'] == activity) & (comdf['subject'] == 'subject'+str(i))]\n",
        "    subx = sub.drop(['activity', 'subject'], axis=1)\n",
        "    suby = sub['activity']\n",
        "    subtrainx, subtestx, subtrainy, subtesty = train_test_split(subx,suby,test_size=0.3,random_state=3)\n",
        "    trainx = pd.concat([trainx, subtrainx])\n",
        "    trainy = pd.concat([trainy, subtrainy])\n",
        "    testx = pd.concat([testx, subtestx])\n",
        "    testy = pd.concat([testy, subtesty])"
      ],
      "metadata": {
        "id": "0d4JvShooj_Z"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(trainx.shape)\n",
        "print(trainy.shape)\n",
        "print(testx.shape)\n",
        "print(testy.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8y_hAvqs6R-H",
        "outputId": "3d2f1097-2305-4047-b63b-9afb9eb7ee66"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(240182, 23)\n",
            "(240182, 1)\n",
            "(103013, 23)\n",
            "(103013, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Standardize the train and test data"
      ],
      "metadata": {
        "id": "7HwkX65VLFXn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import StandardScaler"
      ],
      "metadata": {
        "id": "3Z1QgSfyPZUc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "scaler = StandardScaler()\n",
        "trainx_scaled = scaler.fit_transform(trainx)\n",
        "testx_scaled = scaler.transform(testx)"
      ],
      "metadata": {
        "id": "OfGzYyJU7AcY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "trainx.head(1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PjB-4m23-Z0U",
        "outputId": "2adcc16e-a5b9-492a-a2f0-862fc5198c51"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        acx      acy      acz    ecsig1    ecsig2     alx     aly      alz  \\\n",
              "8136 -9.871  0.61958  0.89787 -0.004186 -0.037677  2.4974 -9.5005  0.44578   \n",
              "\n",
              "          glx      gly  ...      mlz     arx     ary     arz       grx  \\\n",
              "8136 -0.21707 -0.87805  ... -0.58665 -3.0288 -8.9639  1.6143 -0.070588   \n",
              "\n",
              "          gry     grz      mrx     mry      mrz  \n",
              "8136 -0.93018 -0.3556  0.56029  2.1685 -0.37149  \n",
              "\n",
              "[1 rows x 23 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-870f0940-711f-4c1d-8331-22c4fa8a942e\" 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>acx</th>\n",
              "      <th>acy</th>\n",
              "      <th>acz</th>\n",
              "      <th>ecsig1</th>\n",
              "      <th>ecsig2</th>\n",
              "      <th>alx</th>\n",
              "      <th>aly</th>\n",
              "      <th>alz</th>\n",
              "      <th>glx</th>\n",
              "      <th>gly</th>\n",
              "      <th>...</th>\n",
              "      <th>mlz</th>\n",
              "      <th>arx</th>\n",
              "      <th>ary</th>\n",
              "      <th>arz</th>\n",
              "      <th>grx</th>\n",
              "      <th>gry</th>\n",
              "      <th>grz</th>\n",
              "      <th>mrx</th>\n",
              "      <th>mry</th>\n",
              "      <th>mrz</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>8136</th>\n",
              "      <td>-9.871</td>\n",
              "      <td>0.61958</td>\n",
              "      <td>0.89787</td>\n",
              "      <td>-0.004186</td>\n",
              "      <td>-0.037677</td>\n",
              "      <td>2.4974</td>\n",
              "      <td>-9.5005</td>\n",
              "      <td>0.44578</td>\n",
              "      <td>-0.21707</td>\n",
              "      <td>-0.87805</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.58665</td>\n",
              "      <td>-3.0288</td>\n",
              "      <td>-8.9639</td>\n",
              "      <td>1.6143</td>\n",
              "      <td>-0.070588</td>\n",
              "      <td>-0.93018</td>\n",
              "      <td>-0.3556</td>\n",
              "      <td>0.56029</td>\n",
              "      <td>2.1685</td>\n",
              "      <td>-0.37149</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1 rows × 23 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-870f0940-711f-4c1d-8331-22c4fa8a942e')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-870f0940-711f-4c1d-8331-22c4fa8a942e 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-870f0940-711f-4c1d-8331-22c4fa8a942e');\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",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "trainx_scaled.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FFreEQcu7FCa",
        "outputId": "174d3eb1-e0d8-4e8e-d757-8b9f2c7902f5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(240182, 23)"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "trainx_scaled[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3vz3Ct6S-XrB",
        "outputId": "f7c043a9-61a1-4b17-a482-317e624abeb9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([-4.19634023e-01,  2.71040698e-01,  3.99268186e-01, -9.49042352e-03,\n",
              "       -3.42951976e-02,  1.65037174e-01, -8.35496150e-02,  1.77706611e-01,\n",
              "       -6.92872148e-01, -7.55933725e-01, -6.80157716e-01, -6.96913110e-04,\n",
              "        4.96967212e-02, -1.06417976e-02,  7.19747709e-02, -4.85388461e-01,\n",
              "       -1.83587436e-01,  2.38116900e-01, -9.52483883e-01, -1.42626522e+00,\n",
              "        3.08686024e-02,  2.45171784e-02, -5.24630760e-03])"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Models"
      ],
      "metadata": {
        "id": "oHXD3BK9Lefw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim"
      ],
      "metadata": {
        "id": "ESxs3FE56S7X"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Model 1"
      ],
      "metadata": {
        "id": "saUuan_CX_99"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Define the LSTM model\n",
        "class LSTMModel(nn.Module):\n",
        "    def __init__(self, input_size, hidden_size, num_layers, num_classes):\n",
        "        super(LSTMModel, self).__init__()\n",
        "        self.hidden_size = hidden_size\n",
        "        self.num_layers = num_layers\n",
        "        self.lstm = nn.LSTM(input_size, hidden_size, num_layers, batch_first=True)\n",
        "        self.fc = nn.Linear(hidden_size, num_classes)\n",
        "\n",
        "    def forward(self, x):\n",
        "        h0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(device)\n",
        "        c0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(device)\n",
        "        out, _ = self.lstm(x, (h0, c0))\n",
        "        out = self.fc(out[:, -1, :])  # Take the last time step's output\n",
        "        return out"
      ],
      "metadata": {
        "id": "n0BhNPFs6rMC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Model with dropout"
      ],
      "metadata": {
        "id": "r6IQiTLtXpo1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# class LSTMModelWithDropout(nn.Module):\n",
        "#     def __init__(self, input_size, hidden_size, num_layers, num_classes):\n",
        "#         super(LSTMModelWithDropout, self).__init__()\n",
        "#         self.hidden_size = hidden_size\n",
        "#         self.num_layers = num_layers\n",
        "#         self.lstm = nn.LSTM(input_size, hidden_size, num_layers, batch_first=True)\n",
        "#         self.dropout = nn.Dropout(0.25)  # Dropout layer with the specified dropout rate\n",
        "#         self.fc = nn.Linear(hidden_size, num_classes)\n",
        "\n",
        "#     def forward(self, x):\n",
        "#         h0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(device)\n",
        "#         c0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(device)\n",
        "\n",
        "#         # Apply dropout after each LSTM layer\n",
        "#         out, _ = self.lstm(x, (h0, c0))\n",
        "#         out = self.dropout(out)  # Apply dropout after each LSTM layer\n",
        "#         out = self.fc(out[:, -1, :])  # Take the last time step's output\n",
        "#         return out"
      ],
      "metadata": {
        "id": "DnCC9Rr-XoP0"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Hyperparameters 1"
      ],
      "metadata": {
        "id": "ee-GS68TMtGc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "input_size = 23  # Number of features\n",
        "hidden_size = 128  # Number of LSTM units\n",
        "num_layers = 2  # Number of LSTM layers\n",
        "num_classes = 12  # Number of activity classes\n",
        "batch_size = 64\n",
        "learning_rate = 0.001\n",
        "num_epochs = 10\n",
        "sequence_length = 1"
      ],
      "metadata": {
        "id": "K-mTfdyzH6mi"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Hyperparameters 2"
      ],
      "metadata": {
        "id": "60QGX2XPV9Bq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# # Hyperparameters\n",
        "# input_size = 23  # Number of features\n",
        "# hidden_size = 64  # Number of LSTM units\n",
        "# num_layers = 3  # Number of LSTM layers\n",
        "# num_classes = 12  # Number of activity classes\n",
        "# batch_size = 64\n",
        "# learning_rate = 0.001\n",
        "# num_epochs = 10\n",
        "# sequence_length = 1"
      ],
      "metadata": {
        "id": "T3yxdVxNV3vu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Training"
      ],
      "metadata": {
        "id": "0jt34m0SNJpg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from torch.utils.data import DataLoader, TensorDataset\n",
        "from sklearn.metrics import accuracy_score"
      ],
      "metadata": {
        "id": "DDXKHRUYQ3YY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data = TensorDataset(torch.Tensor(trainx_scaled), torch.Tensor(trainy.values))\n",
        "train_loader = DataLoader(train_data, batch_size=batch_size, shuffle=True)\n",
        "n_total_steps = len(train_loader)"
      ],
      "metadata": {
        "id": "4e6RdmmGFNA3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dataset_shape = next(iter(train_loader))[0].shape\n",
        "dataset_shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PUvb6H54K2dD",
        "outputId": "61ee88e3-c7d2-43d9-bd0c-98a6244794fa"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([64, 23])"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Create the LSTM model\n",
        "model = LSTMModelWithDropout(input_size, hidden_size, num_layers, num_classes)\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "model.to(device)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rVTyXbZnH0N2",
        "outputId": "44d9f82e-3b0b-4315-9bba-c0b8a2491e42"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LSTMModelWithDropout(\n",
              "  (lstm): LSTM(23, 128, num_layers=2, batch_first=True)\n",
              "  (dropout): Dropout(p=0.25, inplace=False)\n",
              "  (fc): Linear(in_features=128, out_features=12, bias=True)\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Loss and optimizer\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = optim.Adam(model.parameters(), lr=learning_rate)"
      ],
      "metadata": {
        "id": "dRNmQ3O8H9FR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Training loop\n",
        "for epoch in range(num_epochs):\n",
        "  for i, (inputs, labels) in enumerate(train_loader):\n",
        "    inputs = inputs.reshape(inputs.size()[0], sequence_length, input_size).to(device)\n",
        "    labels = labels.view(-1).long().to(device)  # Ensure labels are of type torch.long\n",
        "\n",
        "    optimizer.zero_grad()\n",
        "    outputs = model(inputs)\n",
        "    loss = criterion(outputs, labels)\n",
        "    loss.backward()\n",
        "    optimizer.step()\n",
        "\n",
        "    if(i+1)%1000 == 0:\n",
        "      print(f'Epoch[{epoch+1}/{num_epochs}], Step[{i+1}/{n_total_steps}], Loss: {loss.item():.4f}')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H_THFNS2IAlJ",
        "outputId": "25ac91eb-584a-4006-eec3-2a60d3a69acc"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch[1/10], Step[1000/3753], Loss: 0.3810\n",
            "Epoch[1/10], Step[2000/3753], Loss: 0.4314\n",
            "Epoch[1/10], Step[3000/3753], Loss: 0.1272\n",
            "Epoch[2/10], Step[1000/3753], Loss: 0.0909\n",
            "Epoch[2/10], Step[2000/3753], Loss: 0.0952\n",
            "Epoch[2/10], Step[3000/3753], Loss: 0.0465\n",
            "Epoch[3/10], Step[1000/3753], Loss: 0.0942\n",
            "Epoch[3/10], Step[2000/3753], Loss: 0.0759\n",
            "Epoch[3/10], Step[3000/3753], Loss: 0.0489\n",
            "Epoch[4/10], Step[1000/3753], Loss: 0.0532\n",
            "Epoch[4/10], Step[2000/3753], Loss: 0.1597\n",
            "Epoch[4/10], Step[3000/3753], Loss: 0.0222\n",
            "Epoch[5/10], Step[1000/3753], Loss: 0.0320\n",
            "Epoch[5/10], Step[2000/3753], Loss: 0.0571\n",
            "Epoch[5/10], Step[3000/3753], Loss: 0.0223\n",
            "Epoch[6/10], Step[1000/3753], Loss: 0.0084\n",
            "Epoch[6/10], Step[2000/3753], Loss: 0.0163\n",
            "Epoch[6/10], Step[3000/3753], Loss: 0.0224\n",
            "Epoch[7/10], Step[1000/3753], Loss: 0.0286\n",
            "Epoch[7/10], Step[2000/3753], Loss: 0.0100\n",
            "Epoch[7/10], Step[3000/3753], Loss: 0.0088\n",
            "Epoch[8/10], Step[1000/3753], Loss: 0.0010\n",
            "Epoch[8/10], Step[2000/3753], Loss: 0.0003\n",
            "Epoch[8/10], Step[3000/3753], Loss: 0.0279\n",
            "Epoch[9/10], Step[1000/3753], Loss: 0.0634\n",
            "Epoch[9/10], Step[2000/3753], Loss: 0.0920\n",
            "Epoch[9/10], Step[3000/3753], Loss: 0.0604\n",
            "Epoch[10/10], Step[1000/3753], Loss: 0.0343\n",
            "Epoch[10/10], Step[2000/3753], Loss: 0.0009\n",
            "Epoch[10/10], Step[3000/3753], Loss: 0.0030\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Evaluation"
      ],
      "metadata": {
        "id": "ZhoaJ4X2NNa4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import (classification_report, accuracy_score,\n",
        "                             recall_score, precision_score, f1_score,\n",
        "                             confusion_matrix)"
      ],
      "metadata": {
        "id": "odgOVN1tSehm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Evaluation 1"
      ],
      "metadata": {
        "id": "L3_F_PgyVKOa"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# model.eval()\n",
        "# with torch.no_grad():\n",
        "#     test_inputs = torch.Tensor(testx_scaled).unsqueeze(1).to(device)\n",
        "#     test_labels = torch.Tensor(testy.values).to(device).long()\n",
        "#     test_outputs = model(test_inputs)\n",
        "#     _, predicted = torch.max(test_outputs, 1)\n",
        "\n",
        "# accuracy = accuracy_score(test_labels.cpu().numpy(), predicted.cpu().numpy())\n",
        "# print(f'Test Accuracy: {accuracy * 100:.2f}%')  # Test Accuracy: 99.56%"
      ],
      "metadata": {
        "id": "QisGcY-IIDcL"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Evaluation 2"
      ],
      "metadata": {
        "id": "dh_P1QRAVc_g"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# model.eval()\n",
        "# with torch.no_grad():\n",
        "#     test_inputs = torch.Tensor(testx_scaled).unsqueeze(1).to(device)\n",
        "#     test_labels = torch.Tensor(testy.values).to(device).long()\n",
        "#     test_outputs = model(test_inputs)\n",
        "#     _, predicted = torch.max(test_outputs, 1)\n",
        "\n",
        "# accuracy = accuracy_score(test_labels.cpu().numpy(), predicted.cpu().numpy())\n",
        "# print(f'Test Accuracy: {accuracy * 100:.2f}%')  # Test Accuracy: 99.33%"
      ],
      "metadata": {
        "id": "15vIQS3oVYfo"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Evaluation 3"
      ],
      "metadata": {
        "id": "IbFqGK19Vnto"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.eval()\n",
        "with torch.no_grad():\n",
        "    test_inputs = torch.Tensor(testx_scaled).unsqueeze(1).to(device)\n",
        "    test_labels = torch.Tensor(testy.values).to(device).long()\n",
        "    test_outputs = model(test_inputs)\n",
        "    _, predicted = torch.max(test_outputs, 1)\n",
        "\n",
        "accuracy = accuracy_score(test_labels.cpu().numpy(), predicted.cpu().numpy())\n",
        "print(f'Test Accuracy: {accuracy * 100:.2f}%')  # Test Accuracy: 99.24%"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8qjshgHoVmhC",
        "outputId": "2ebe5239-0f71-44dc-83d6-ab9dc43ea6f0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Test Accuracy: 99.24%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Training and Evaluation"
      ],
      "metadata": {
        "id": "6yAsRkKGBKRs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "# Initialize a list to store accuracy values\n",
        "epoch_accuracies = []\n",
        "\n",
        "for epoch in range(num_epochs):\n",
        "    for inputs, labels in train_loader:\n",
        "        inputs = inputs.reshape(inputs.size()[0], sequence_length, input_size).to(device)\n",
        "        labels = labels.view(-1).long().to(device)\n",
        "\n",
        "        optimizer.zero_grad()\n",
        "        outputs = model(inputs)\n",
        "        loss = criterion(outputs, labels)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "\n",
        "    # Evaluate the model after each epoch\n",
        "    model.eval()\n",
        "    with torch.no_grad():\n",
        "        test_inputs = torch.Tensor(testx_scaled).unsqueeze(1).to(device)\n",
        "        test_labels = torch.Tensor(testy.values).to(device).long()\n",
        "        test_outputs = model(test_inputs)\n",
        "        _, predicted = torch.max(test_outputs, 1)\n",
        "\n",
        "        accuracy = accuracy_score(test_labels.cpu().numpy(), predicted.cpu().numpy())\n",
        "        epoch_accuracies.append(accuracy)\n",
        "\n",
        "    model.train()  # Set the model back to training mode\n",
        "\n",
        "    print(f'Epoch [{epoch+1}/{num_epochs}] - Test Accuracy: {accuracy * 100:.2f}%')\n",
        "\n",
        "# Print the accuracy values after each epoch\n",
        "print(\"Epoch Accuracies:\", epoch_accuracies)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Sm9K3e-2BJil",
        "outputId": "e411f953-c6ef-4669-9de6-8d13501bec88"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch [1/10] - Test Accuracy: 99.58%\n",
            "Epoch [2/10] - Test Accuracy: 99.66%\n",
            "Epoch [3/10] - Test Accuracy: 99.58%\n",
            "Epoch [4/10] - Test Accuracy: 99.64%\n",
            "Epoch [5/10] - Test Accuracy: 99.61%\n",
            "Epoch [6/10] - Test Accuracy: 99.62%\n",
            "Epoch [7/10] - Test Accuracy: 99.65%\n",
            "Epoch [8/10] - Test Accuracy: 99.63%\n",
            "Epoch [9/10] - Test Accuracy: 99.63%\n",
            "Epoch [10/10] - Test Accuracy: 99.65%\n",
            "Epoch Accuracies: [0.9957675244872006, 0.9965926630619436, 0.9958451845883529, 0.996437342859639, 0.9961364099676739, 0.9962043625561823, 0.9964761729102152, 0.9962723151446905, 0.9962917301699785, 0.9965441254987235]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Visualization"
      ],
      "metadata": {
        "id": "9sJoDBMSNji9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Your code for training and storing accuracies here\n",
        "\n",
        "# Plot the accuracies\n",
        "plt.figure(figsize=(8, 6))\n",
        "plt.plot(range(1, num_epochs + 1), epoch_accuracies, marker='o', linestyle='-', color='b')\n",
        "plt.title('Accuracy vs. Epochs', fontsize=18)\n",
        "# plt.xlabel('Epochs')\n",
        "# plt.ylabel('Accuracy')\n",
        "plt.grid(True)\n",
        "plt.xticks(np.arange(1, num_epochs + 1))\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6bssKCA-EPc_",
        "outputId": "dd5612e4-e42a-41b3-a3b3-55f3db7de4ff"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x600 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAroAAAIVCAYAAAA+t1wiAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACIc0lEQVR4nO3deVhUZf8G8HvYEcENRUAEWQQ0xbXSVNJckjKVStteTdPSV3PhTXNfsjItTfNnLplWlLZppi0YkWKmuZumuOAu7qaCbA4z5/fH0xkYWQdm5sycuT/X5TXH4czM9wDKzTPP8300kiRJICIiIiJSGSelCyAiIiIisgQGXSIiIiJSJQZdIiIiIlIlBl0iIiIiUiUGXSIiIiJSJQZdIiIiIlIlBl0iIiIiUiUGXSIiIiJSJQZdIiIiIlIlBl0iIiIr02g00Gg02LJli9KlEKkagy6RA7t58yY8PT0NP3RPnDihdEnkgGbMmGH4HqzIHyKiinJRugAiUs4XX3yBvLw8w99XrlyJ2bNnK1gROTo/Pz+lSyAiFeGILpED+/jjjwEAr776KgDg008/hU6nU7IkcnCXL18u9w8RUUUx6BI5qH379uHAgQOoWbMm5s6di0aNGuHSpUv46aeflC6NiIjILBh0iRyUPJrbv39/eHh4YMCAAQDE9IWK+OWXX/DMM88gODgYnp6eqF27Npo3b45XX30VO3bsKPExd+/exYoVK/Doo4/Cz88P7u7u8Pf3R7t27fDGG2/g9OnTRuc//PDD0Gg0mDFjRql1yPM7H3744WIfK/p4rVaLefPmoU2bNqhZs6bRQiC9Xo+UlBSMGjUKDz74IBo0aAA3NzfUqVMHsbGxWLp0KbRardk+J8888ww0Gg3i4uLKfL709HQ4OTlVeNGSVquFr68vNBoNPvjggzLPXblyJTQaDXx8fJCTk2O4v6CgAMuXL8fDDz8MX19fuLq6ok6dOoiMjET//v0N3ze2JCQkBBqNBp988gmysrIwceJEREZGwtPTE76+vujTpw927txZ5nPodDqsXLkSXbp0ga+vL9zd3REYGIinn366Qp/78+fPY/z48WjRogVq1KgBT09PhIWFoXfv3vjss8+MpgjdKysrC1OmTEFUVBQ8PT1Rp04dPP7442XWfPPmTUybNg2tWrWCj48P3NzcUL9+fTRv3hzDhg1DSkpKuTUTOQSJiBxObm6uVLNmTQmA9Mcff0iSJEknT56UNBqN5OLiIl2+fLnUx2ZnZ0tPP/20BMDwx9vbW6pRo4bh7zExMcUed+rUKem+++4znKPRaKRatWpJ1apVM9w3evRoo8fExsZKAKTp06eXWs/06dMlAFJsbGyxj8mPf/3116X27dtLACQXFxepVq1akkajkTZv3ixJkiSdPn3a6HqqV69udD0ApI4dO0o5OTlm+Zxs3rxZAiA5OTlJZ8+eLfXaXn/9dQmA1Lhx41LPudeIESMkAFKbNm3KPO/hhx+WAEgvvvii4b6CggKpW7duRtdRo0YNyd3d3eg+c5O/hpV97uDgYAmANH/+fCkyMlICILm5uUk+Pj6G53VycpI+/vjjEh9/69Ytw+cDgOTs7CzVrFlT0mg0hvtee+21Ul//s88+kzw8PAznurm5SXXq1JFcXFwM9+3fv9/oMfL9q1evlsLDwyUAkoeHh9G/Bzc3N2nTpk3FXu/8+fNSw4YNja6tVq1akrOzs+G+kv49EDkiBl0iB/T5559LAKTw8HCj+zt27CgBkObOnVvqY/v162f44fr6669L58+fN3zs2rVr0hdffCENGzbM6DG3b9+WIiIiJABSrVq1pOXLl0u3bt0yfPzkyZPSvHnzpPnz5xs9zlxBt3r16lL16tWlVatWGcLq9evXpRs3bkiSJILD888/L23YsMFwnyRJUlZWlrRq1SopICBAAiCNHTvWbJ+T6OhoCYA0bdq0Ep/z7t27kp+fnwRAeu+990q9/nvt3LnTEHbS0tJKPOfs2bOGEPfbb78Z7k9MTDQErhUrVkhZWVmSJEmSXq+Xrly5Iq1bt0566qmnKlxLRZkr6NaoUUOqVauW9PXXX0tarVaSJEk6cuSI4fvAxcVF2rt3b7HHP/nkk4Zg+cEHH0jZ2dmSJEnSpUuXpMGDBxtqW7JkSbHH/vDDD4bP5UMPPST9/vvvkk6nkyRJkvLz86Xff/9dGjp0qHT48GGjx8nPWatWLalJkybSb7/9Jul0Okmv10u7du0yBPbg4GDD88leeuklCYAUEhIi/frrr1JBQYEkSeIXlTNnzkhLliyRXn/99Up9LonUhkGXyAF17txZAiC98cYbRvd/9NFHEgApKiqqxMf9+uuvhh/QH374YYVfb8qUKRIAyd3dXdq3b1+FH2euoAtA2rBhQ4Vf9167d++WAEheXl5Sbm6u0ccq+zlZsGCBBEBq0KCBIagU9e233xo+Z9euXTOpXjkkTZw4scSPv/322xIAqWHDhpJerzfcP3z4cAmA9PLLL5v0elVVNOj6+fmV+WfUqFHFHi8HXQDSr7/+WuzjOTk5hl+04uLijD72559/Gh67bNmyEuuTg7Cvr6/R11+r1UqNGjWSAEgdOnSQ8vPzK3zN8mvWrVtXunLlSrGPHzx40HDOtm3bjD4m/5K0evXqCr8ekaPiHF0iB3Pq1Cls2bIFGo0G//nPf4w+1q9fP3h6euLo0aPYvn17scfK83fvu+8+DB8+vMKvKT9uyJAhaNmyZRWqr5ymTZuiV69elX58mzZtUK9ePWRnZ+PAgQNGH6vs52TgwIGoVq0aLly4UOICwI8++ggAEB8fD19fX5Pqlb+uX3zxBSRJKvbxxMREAMDzzz9v1Je2Zs2aAKBoZ4MrV66U+ef27dulPvahhx7CI488Uux+T09PjBs3DgCQlJRk9BxfffUVAKBBgwYYMmRIic87a9YsAMD169eRnJxsuH/z5s2GeeXvv/8+3NzcTLxa4OWXX0a9evWK3d+sWTM0atQIAHDw4EGjj8lfp0uXLpn8ekSOhkGXyMGsWrUKkiShY8eOCAkJMfqYj48P+vTpAwAlLjqSw+/jjz9e4dc7e/YsLl68CABVCptV8dBDD5V7zt27d7F06VJ0794dAQEBcHd3N9qk4OrVqwCACxcuGD2uMp8TQISV/v37AygMtbKzZ88aAtXLL79s0vMCIuhqNBqcO3cOqampRh/bu3cv0tLSAMCwAFEWFxcHjUaDDRs2oGfPnlizZo3ha2ctkninsdQ/n3zySamP7dKlS7kf0+v12Ldvn+H+PXv2AAA6d+4MJ6eSfyRGR0cjMDDQ6Hyg8Gtfv359tGnTpmIXeI8HHnig1I8FBAQAAP755x+j++XvtQkTJuDll19GUlISMjMzK/X6RGrHoEvkQPR6vSEo3BtyZAMHDgQAfP3117hz547Rx+SRvuDg4Aq/ZtHRQVMeZ04ljZgVdfXqVbRp0wbDhw9HcnIyLl26BCcnJ/j6+sLPzw9+fn6GEJSdnW302Mp8TmTDhg0DAPz000/IyMgw3L9ixQro9XpERkaW2E2iPA0bNkRsbCyAwtFbmfz3tm3bIioqyuhjHTp0wJw5c+Dm5oakpCQ899xzCAwMRFBQEAYNGoTNmzebXIs1yWG0vI/Jv7QUPS7rsYAY8b33sVX52su8vb1L/ZiLi9jT6d6OH+PGjUO/fv2g1Wrx0UcfoWfPnqhZsyaaNWuGcePG4dixY5Wuh0htGHSJHMimTZsMI5JDhgwpcXvVRx99FABw584dfP3110aPr8z2q7awZauzs3OZHx87diwOHTqEOnXqYOXKlbh06RJyc3Nx7do1wyYF8ujavVMBqnJ9999/P1q1agWdTmcYQdfpdFi1ahUAYOjQoZV+bvkXmW+//Ra5ubkAROuwNWvWAECxaSuycePG4fTp03j//ffRp08f1KtXDxcuXMAnn3yCLl264Omnn65QqzVHoNT3tqurK7766iscOHAA06ZNQ5cuXVCtWjX8/fffeO+999C0aVPMmzdPkdqIbA2DLpEDMbUH6r3n169fH4B4a72i5MeY+jigcESrrB6kZc3ZrAitVot169YBAP7v//4PgwYNMqoZEOHz+vXrJT6+Mp+TouRR3ZUrV0Kv1xtGd93d3Q2j65Xx1FNPwdPTE5mZmfj+++8BiD6/V69ehaurK5599tlSHxsQEIAxY8bgu+++w5UrV3Dw4EHD/NVvv/0WS5YsqXRdllR0VLysjxUd4ZeP752Sci/540UfW9WvfVXFxMRg5syZSElJwa1bt/Drr7+iU6dO0Ol0GDduHP766y9F6iKyJQy6RA7i2rVr2LBhAwARVrKyskr9s2vXLgBiDmLRt0Hbt28PANi4cWOFX7dhw4aGt4VNeRwA1KpVC4Boxl+a8jYCKM+1a9cMQbq0hXLbtm0rNWxX5nNS1HPPPQcfHx+cPXsWmzZtqtIitKK8vb0N863l6Qrybc+ePU167mbNmuGjjz4yzHUuuiDLlpQ1tUL+mJOTk9HXWZ5bu3nzZuj1+hIfe/ToUUNQbtu2reF++Wt/+fJlo7m7SnBxccEjjzyCH3/8Ee7u7pAkCb/++quiNRHZAgZdIgeRmJgIrVaLGjVqoFevXqhevXqpf4rO3yw6qvvSSy8BAA4fPmzSqJ78uBUrVmD//v0VflxMTAwAMeXi3rmxAPDbb7+VugtbRfn4+Bjegi5pBKygoACTJ08u9fGV/ZzIvLy8DNMI3nzzTUMHhsosQruXPH3hl19+wYkTJwwju6XNz87Pzy/z+Tw9PQGg1EVbStu2bVuJu5jl5eUZ3srv0aOHoWsBIHapA8SI74oVK0p83mnTpgEAfH190bVrV8P9nTt3RmhoKAAx/eXu3bvmuIxylfV1cnd3N0zVsdWvE5FVKdPVjIisrUmTJhIAacCAARU6f+rUqYa+pnLzfUmSpGeeecawOcKECROKbY7w0UcfSYMHDzZ6rszMzGIbRty+fdvw8fT0dGnmzJnSu+++a/S4Y8eOSU5OThIAqVevXobXysnJkT755BPJx8dHql27drl9dMvqwytJktShQwcJgBQYGCilpKQYGvQfOnRI6tatm+Tu7i55eXlJAKRVq1YVe3xlPidFHTp0yGjnscjIyDLrraiCggKpfv36hp3S5M9/Xl5eiec/+uij0qBBg6SffvpJunnzpuH+GzduSLNmzTJsjHBvv9lVq1YZapd3mzOFOTeMqF27tvTNN98YvmfT0tKkLl26GHY82717d7HHF90wYtGiRUYbRgwZMqTMDSN++uknw+elQ4cOxTaM2Lx5s/T888+XumFEWZ+v0r5//fz8pAkTJkg7duww+lqeOHHCsEOfk5NTsdckckQMukQOYMeOHYYfrBs3bqzQY4o2rF+/fr3h/uzsbCk+Pt4omPn4+JS7BfDJkycNYVv+QVy7du0ytwCWJEmaNm2a0WvVqFHDsLVqnz59DJtRVCXo7tmzxxBk8e8mDd7e3obdtD777DNDmCop6Fb2c1KUHLZh4k5o5UlISDCq65VXXin13KIbbMjXUHQbXQDSU089VWynLnMG3fI2jPDz8zNsWy0raQtgd3d3o8+/RqORli9fXuLr37p1y+jai24TLd9X1hbAn376qdE2ye7u7hXeArgyQbfo10Pe/rfoFsQajUZ6//33y/qUEzkMvq9B5ADk6Qc1atRA9+7dK/SYZs2aITo62ujxAFCtWjWsXbsWP/zwA/r27YuAgADk5eXBxcUFzZs3x6hRo7B8+fJizxcaGor9+/fjww8/xMMPP4xatWohKysLNWvWRLt27TBr1iyMHTu22ONmzpyJxMREPPjgg/Dy8oJOp0OLFi2wdOlSrFu3rtyOChXRunVr7Nq1C/369YOvry/0ej28vb3Rr18/bN++vdQOBVX9nBT19NNPA0CVF6Hd695pCqVNWwCARYsWYc6cOYiLi0NERAQkSUJubi4CAgLwxBNPYO3atfjmm28s+pZ4eRtGXLlypdQpArVq1cKuXbswYcIENGzYEPn5+ahduzZ69eqFP/74o9QuFjVq1EBKSgo+/vhjPPzww/D29sadO3dQv359PPnkk9i8eTPefffdUmseMGAAjh49ijFjxqBJkyZwcXFBbm4ugoOD0adPHyQmJhr+LZnDL7/8gokTJ6Jjx44ICgoydNUIDw/HoEGDsHv3bowZM8Zsr0dkzzSSVMK2OUREZFW9evXCDz/8gGeffRarV69Wuhy7EhISgrNnz2LVqlV48cUXlS6HiGwIR3SJiBR26tQpwyI0U7YRJiKisjHoEhEpKDMzE8OHD4der8cDDzyAjh07Kl0SEZFquChdABGRI3rttdfwzTff4PLly7h79y5cXFywYMECpcsiIlIVjugSESng+vXrOHfuHNzc3NCuXTskJSXhwQcfVLosIiJV4WI0IiIiIlIljugSERERkSpxjm4Rer0eFy9ehLe3t2FLUCIiIiKyHZIkISsrCwEBAeX29WbQLeLixYsICgpSugwiIiIiKsf58+fRoEGDMs9h0C3C29sbgPjE+fj4WOU1tVotfvnlF3Tv3h2urq5WeU1b4KjXDTjutTvqdQO8dke8dke9boDX7ojXbu3rzszMRFBQkCG3lYVBtwh5uoKPj49Vg261atXg4+PjcP8oHPG6Ace9dke9boDX7ojX7qjXDfDaHfHalbruikwz5WI0IiIiIlIlBl0iIiIiUiUGXSIiIiJSJQZdIiIiIlIlBl0iIiIiUiUGXSIiIiJSJQZdIiIiIlIlBl0iIiIiUiUGXSIiIiJSJQZdIiIiIlKlSgXdxYsXIyQkBB4eHnjggQewa9euUs/VarV44403EBYWBg8PD8TExCApKcnonKysLIwZMwbBwcHw9PRE+/btsXv37mLPlZaWhieeeAI1atSAl5cX2rZti3Pnzhmds2PHDnTp0gVeXl7w8fFBp06dkJubW5nLJCIiIiI7ZnLQ/eqrr5CQkIDp06dj3759iImJQY8ePXD16tUSz58yZQqWLVuGRYsW4ciRIxg2bBj69u2L/fv3G84ZMmQIkpOTkZiYiEOHDqF79+7o2rUrMjIyDOecPHkSHTp0QFRUFLZs2YKDBw9i6tSp8PDwMJyzY8cOPProo+jevTt27dqF3bt3Y+TIkXBy4sA1ERERkaMxOQHOnz8fQ4cOxaBBg9CkSRMsXboU1apVw8qVK0s8PzExEZMmTUJcXBxCQ0MxfPhwxMXFYd68eQCA3NxcrF27FnPnzkWnTp0QHh6OGTNmIDw8HEuWLDE8z+TJkxEXF4e5c+eiZcuWCAsLwxNPPIF69eoZzhk7dixGjRqFCRMmoGnTpoiMjES/fv3g7u5u6mUSERERkZ0zKejevXsXe/fuRdeuXQufwMkJXbt2xY4dO0p8TH5+vtGoKwB4enpi27ZtAICCggLodLoyz9Hr9fjxxx/RuHFj9OjRA/Xq1cMDDzyA9evXG86/evUqdu7ciXr16qF9+/bw8/NDbGys4TnIduh0QGqqBlu3BiI1VQOdTumKiIiISI1cTDn5+vXr0Ol08PPzM7rfz88PR48eLfExPXr0wPz589GpUyeEhYUhJSUF69atg+7fdOPt7Y127dph1qxZiI6Ohp+fH9asWYMdO3YgPDwcgAixd+7cwTvvvIM333wTc+bMQVJSEuLj47F582bExsbi1KlTAIAZM2bgvffeQ4sWLfDZZ5/hkUcewd9//42IiIhiteXn5yM/P9/w98zMTABiXrFWqzXlU1Np8utY6/WU9t13GiQkOCMjwwVAG8yfDwQGSpg/X4e+fSWly7MKR/uayxz1ugFee9FbR+Go1w3w2oveOgprX7cpr6ORJKnC6eLixYsIDAzE9u3b0a5dO8P948ePR2pqKnbu3FnsMdeuXcPQoUOxceNGaDQahIWFoWvXrli5cqVhkdjJkycxePBgbN26Fc7OzmjVqhUaN26MvXv3Ii0tzfC6zz77LFavXm147ieeeAJeXl5Ys2YNtm/fjoceeggTJ07E22+/bTinefPmeOyxxzB79uxitc2YMQMzZ84sdv/q1atRrVq1in5aqIJ27PDHnDlt//2bpshHxLfg66/vRrt2l6xeFxEREdmPnJwcPPfcc7h9+zZ8fHzKPNekEV1fX184OzvjypUrRvdfuXIF9evXL/ExdevWxfr165GXl4cbN24gICAAEyZMQGhoqOGcsLAwpKamIjs7G5mZmfD390f//v0N5/j6+sLFxQVNmjQxeu7o6GjD1AR/f38AKPGcezszyCZOnIiEhATD3zMzMxEUFITu3buX+4kzF61Wi+TkZHTr1g2urq5WeU0l6HTAiBHyt5vmno9qoNFI+OKLtpgxowDOztauzroc5Wt+L0e9boDX7ojX7qjXDfDaHfHarX3d8jvwFWFS0HVzc0Pr1q2RkpKCPn36ABDzZ1NSUjBy5MgyH+vh4YHAwEBotVqsXbsW/fr1K3aOl5cXvLy8cPPmTWzatAlz5841vG7btm1x7Ngxo/OPHz+O4OBgAEBISAgCAgJKPKdnz54l1uTu7l7iQjVXV1erf4Mq8ZrW9McfQJEmGsVIkgYXLgB//umKhx+2WlmKUvvXvDSOet0Ar90Rr91RrxvgtTvitVvruk15DZOCLgAkJCRg4MCBaNOmDe6//34sWLAA2dnZGDRoEABgwIABCAwMNEwV2LlzJzIyMtCiRQtkZGRgxowZ0Ov1GD9+vOE5N23aBEmSEBkZifT0dIwbNw5RUVGG5wSAcePGoX///ujUqRM6d+6MpKQkbNy4EVu2bAEAaDQajBs3DtOnT0dMTAxatGiBTz/9FEePHsW3335r6mWSmV2q4IyEip5HREREVB6Tg27//v1x7do1TJs2DZcvX0aLFi2QlJRkWKB27tw5o761eXl5mDJlCk6dOoXq1asjLi4OiYmJqFmzpuGc27dvY+LEibhw4QJq166NJ598Em+99ZZRYu/bty+WLl2K2bNnY9SoUYiMjMTatWvRoUMHwzljxoxBXl4exo4di3/++QcxMTFITk5GWFhYZT43ZEb/ziwx23lERERE5TE56ALAyJEjS52qII+wymJjY3HkyJEyn69fv34lTmW41+DBgzF48OAyz5kwYQImTJhQ7nORdXXsCDRoIKYvlLT8UaMRH+/Y0fq1ERERkTpxyzCyCmdnYOHCkj+m+Xdt2oIFUP1CNCIiIrIeBl2ymvh44NtvAW9v4/sbNBD3x8crUxcRERGpE4MuWVV8PPDII8b37dnDkEtERETmx6BLVnfihPHfjx9Xpg4iIiJSNwZdsiqdDkhPF8eBgVkAgHLWKhIRERFVCoMuWdX580B+PuDmJqFVq6sAGHSJiIjIMhh0yarkaQphYUBwsNjCj0GXiIiILIFBl6xK3qE5IkJCUBCnLhAREZHlMOiSVckjuhEREho0EEE3IwO4dUu5moiIiEidGHTJquSg27ixBC+vAgQGim3S0tIULIqIiIhUiUGXrKow6Irb6GgRdDl9gYiIiMyNQZesJi8POHtWHEdEiIDLoEtERESWwqBLVnPyJCBJQI0aQN264r4mTRh0iYiIyDIYdMlq5I4LjRsDGo04jo4Wtwy6REREZG4MumQ1987PBYCoKDGie+4ckJWlQFFERESkWgy6ZDUlBd3atYH69cXx0aPWr4mIiIjUi0GXrEYOupGRxvc3aSJuDx+2bj1ERESkbgy6ZDUljegChUGX83SJiIjsi04HpKZqsHVrIFJTNdDplK7IGIMuWcXNm8C1a+I4IsL4Ywy6RERE9mfdOiAkBOjWzQXz57dBt24uCAkR99sKBl2yCnk0NyAAqF7d+GNNm4pbBl0iIiL7sG4d8NRTwIULxvdnZIj7bSXsMuiSVZQ2bQEoHNE9cwbIzrZaSURERFQJOh0werTojX8v+b4xY2AT0xgYdMkqygq6vr5iAwlJKuy1S0RERLbp99+Lj+QWJUnA+fPiPKUx6JJVlNZxQcZ5ukRERPbh0iXznmdJDLpkFWWN6AJsMUZERGQv/P3Ne54lMeiSxen1FQ+6HNElIiKybR07Ag0alP5xjQYIChLnKY1Blyzu4kUgJwdwdgYaNSr5HAZdIiIi++DsDCxcWPLHNBpxu2CBOE9pDLpkcfJobmgo4Opa8jlyi7FTp4DcXOvURURERJXTsGHJ9zdoAHz7LRAfb916SsOgSxZX3rQFAKhXD6hd23iaAxEREdmmN98Ut889ByQnFyAhYQ+Skwtw+rTthFyAQZesoLyOC4B4q4PTF4iIiGzfwYPA99+Ln91TpwKxsRI6dcpAbKxkE9MVimLQJYuryIguwKBLRERkD+TR3KefBqKilK2lPAy6ZHHyJhAVDbpsMUZERGSb0tLEHFwAmDJF2VoqgkGXLOruXeD0aXHMEV0iIiL79vbbYuezPn2AZs2UrqZ8DLpkUadPi72uvbyAgICyz5WDbno6kJ9v+dqIiIio4tLTgdWrxbE9jOYCDLpkYUXn58q99UoTEAD4+IhgfOKE5WsjIiKiips9W3RH6tkTaN1a6WoqhkGXLKqiC9EAEYTlfrqcvkBERGQ7zpwBPvtMHE+dqmgpJmHQJYsyJegCnKdLRERki+bMAQoKgEceAdq1U7qaimPQJYuqaMcFGYMuERGRbcnIAFauFMf2NJoLMOiShVV2RJctxoiIiGzDu++KLkodOgCdOildjWkYdMlisrKAS5fEsalB9/hxQKu1TF1ERERUMVeuAMuWieOpU8tfWG5rGHTJYuTOCfXqATVrVuwxQUFA9epiHlB6usVKIyIiogqYNw/IywPuvx/o1k3pakzHoEsWY+q0BUD8phgdLY45T5eIiEg5N24AH34oju1xNBdg0CULMnUhmowtxoiIiJS3YAGQnQ20aAE89pjS1VQOgy5ZTGVGdAF2XiAiIlLarVvABx+I4ylT7HM0F2DQJQti0CUiIrJPixYBmZniXda+fZWupvIYdMkiJKnqQffYMbEojYiIiKwnK0tMWwCAyZMBJztOi3ZcOtmyq1fFb4IaDRAebtpjg4MBT08gPx84dcoy9REREVHJliwB/vkHiIgA+vVTupqqYdAli5BHc0NCAHd30x7r5MTOC0RERErIyREtxQBg0iTA2VnZeqqKQZcsorIdF2Scp0tERGR9y5eLd2VDQoDnn1e6mqpj0CWLqOz8XBlbjBEREVlXXh4wd644njgRcHVVth5zYNAli6hq0OWILhERkXWtXAlcugQ0aAAMHKh0NebBoEsWYa6gm5YG6HTmqYmIiIhKdvcuMGeOOH79ddPX19gqBl0yO50OSE8Xx5GRlXuORo3EP7K8PODsWfPVRkRERMUlJgLnzgF+fsBLLyldjfkw6JLZnT0LaLUiqAYFVe45nJ2BqChxfPiw+WojIrI0nQ5ITdVg69ZApKZq+K4U2byCAuDtt8XxuHGixadaMOiS2ckdFyIiqtZkmvN0icjerFsnVqt36+aC+fPboFs3F4SEiPuJbNWaNaJvva8vMGyY0tWYF4MumV1V5+fKGHSJyJ6sWwc89RRw4YLx/RkZ4n6GXbJFOh3w1lviOCEB8PJSth5zY9AlszNX0GWLMSKyFzodMHq02P78XvJ9Y8ZwcS3ZnrVrxTuxNWsCI0YoXY35MeiS2Zl7RDctDdDrq/ZcRESW9PvvxUdyi5Ik4Px5cR6RrdDrgTffFMejRwM+PsrWYwkMumR2ctCtbMcFWViYaFadnS1+QBAR2apLl8x7HpE1bNgAHDoEeHuLoKtGDLpkVrm5oj0JUPURXReXwrDM6QtEZMv8/c17HpGlSRIwa5Y4HjkSqFVL2XoshUGXzOrECXFbqxZQp07Vn0+evsAWY0Rkyzp2FLtJaTSlnxMUJM4jsgVJScC+fUC1asDYsUpXYzkMumRWRefnlvUffkWx8wIR2QNnZ2DhwpIXo8nefVecR6S0oqO5w4cDdesqW48lMeiSWZlrIZqMQZeI7EXbtiX3DpfvO3nSuvUQlea334AdO8TGTv/7n9LVWBaDLpmVuRaiyYoG3bJGSoiIlPbBB2IV+8MPA8nJBUhI2IPk5AJ88on4+JtvAmfOKFgg0b/k0dyhQ9U/b5xBl8zK3CO6ERFiUVpWlmi6TkRki27fBpYtE8fjxgGxsRI6dcpAbKyEF14AYmPFYl01z4Uk+/D770BqquhqNH680tVYHoMumZW5g66bmwi7AKcvEJHtWrFC/EIeHQ08+qjxxzQaYPFi8Uv7+vXATz8pUiIRgMK+uYMGiQWSasegS2Zz44b4AwDh4eZ7Xs7TJSJbptUCCxaI4//9r+R5uk2bip3RAODVV4G8PGtVR1Ro1y7gl1/EosgJE5SuxjoYdMls5NHcBg3Mu1c2gy4R2bKvvxa7ovn5Ac8/X/p506YBAQHAqVPAnDnWq49IJs/NfeEFoFEjZWuxFgZdMhtzT1uQsZcuEdkqSQLee08cv/oq4OFR+rne3sD774vj2bNF4CWylv37gR9+EO84TJqkdDXWw6BLZmPujgsydl4gIlv122/AgQOi6f6wYeWf//TTQNeuQH4+MGoU/08j65Hn5vbvb/4BKVvGoEtmY6kR3caNxW+gt24Bly+b97mJiKpCHs0dPLhiu0FqNMCiRWLF+48/Ahs3WrY+IkC8I7punTiePFnZWqyNQZfM5tgxcWvuoOvhUbi4jfN0ichW/P232EbVyalwoVlFREUVNukfNQrIybFIeUQGb70lbuPjxcJIR8KgS2ah1wMnTohjS7wlwgVp9k2nA1JTNdi6NRCpqRrodEpXRFR18+eL2/h4ICzMtMdOmSJaO509K+brElnK8ePAV1+J4ylTlK1FCQy6ZBYXLoh2OS4uQEiI+Z+fQdd+rVsnvie6dXPB/Plt0K2bC0JCCt9GI7JHly4Bn38ujiuzhaqXV2FLsrlzCwcKiMzt7bfFYNTjjwMtWypdjfUx6JJZyPNzw8JE2DU3Bl37tG4d8NRT4hehojIyxP0Mu2SvFi0S/XMfegh48MHKPUffvmJzibt3gZEjuTCNzO/06cJfyKZOVbYWpTDokllYquOCrGiLMf4wsA86HTB6dMlfL/m+MWPAaQxkd+7cAZYsEcevvVb559FogA8+EDtA/vILf/Ej83vnHfF/bPfuwP33K12NMhh0ySws1XFBFhkpfijcuAFcu2aZ1yDz+v334iO5RUkScP68OI/InqxcKbrAREQAvXpV7bkiIoDXXxfHY8YA2dlVrY5IOH8eWLVKHDvi3FwZgy6ZhaU6LsiqVSvcxYXTF+zDpUvmPY/IFhQUFG76kJAgtlKtqgkTxDz2CxcKd64iqqq5c8X0mthYoGNHpatRDoMumYWlR3SBwpYoDLr2wd/fvOcR2YJ164AzZwBfX2DAAPM8Z7VqYgoDAMybB6Slmed5yXFdugR89JE4dtS5uTIGXaqy/HzxHz9g2aDLBWn2pWNHoEEDMeWkJBqNaK/kyCMNZF+Kbvc7YoQIqObSq5dYFV9QwIVpVHXz5omfze3aAV26KF2Nshh0qcpOnRKtS6pXB+rXt9zrMOjaF2dnYOHCss9ZsMA8b/0SWcO2bcDu3WITm//+1/zP/8EH4rl/+w34+mvzPz85hmvXChdLTp1a+mCDo2DQpSor2nHBkv+gGHTtT3w88O23gLd38Y/NmiU+TmQv5NHcAQOAevXM//yNGgETJ4rjhAQgK8v8r0Hq9/77Yre91q1F+zpHx6BLVWaN+bmA2DYTAK5cEd0XyD7ExwNt24rjrl3P4NFH9QD4CwvZl2PHgA0bxHFCguVeZ/x40Y/84kVg5kzLvQ6p0z//AP/3f+J4yhSO5gIMumQGlu64IKteHQgOFscMSfZDrwf27hXHcXFnMGOGaJz77bdsFUf2Q97u94knLNcvHBBTFxYtEscLFgB//2251yL1+eAD8U5As2bie5UYdMkMrDWiC3D6gj1KTwdu3wY8PCQ0bJiJVq2ANm3EblByj0ciW3b1KvDpp+K4KhtEVFTPnmLXNJ1OLHrjwjSqiMzMwnURU6YATkx4ABh0yQysGXTZYsz+7N4tbmNiJLi4iJ/Yw4eL+5YtEyO+RLZs8WKxgv3++4EOHazzmu+/D3h6Alu3Al98YZ3XJPu2eLHYyCQqCnjySaWrsR0MulQlt2+LObMAR3SpZLt2idu2bQuHpZ55BqhRQ3TsSE5WqDCiCsjJEQECEKO51przGBxc2P/0tdfE/7VEpcnOLpxeM2kSu9kUxaBLVXLihLitXx/w8bH86zHo2h95RLd168KgW60aMHCgOJbb4BDZos8+E4tfGzUS0wmsKSFBDCBcuQJMm2bd1yb7snQpcP26WMj47LNKV2NbGHSpSqw5bQEAoqPF7cWL4i0asm1aLbB/vzhu08Z4ouGwYeJ240axJzuRrdHpCkfJxo4FXFys+/ru7oUr6P/v/4C//rLu65N9yM0F3n1XHE+caP3vU1vHoEtVYq2OCzIfH7HbFsBtMu3B4cNAXp74ukVEGH8sOhp4+GExR3fFCkXKIyrTxo3iXauaNYFBg5SpoVs34Omnxb+T//6Xc9qpuI8/FqP+DRsC//mP0tXYHgZdqhJrj+gChdMXDh+23mtS5cjTFtq0KXkFsDyq+9FHYvSXyJbIG0QMHy7aGypl/nzAywvYvl1MpSCS5ecDc+aI4wkTADc3ZeuxRQy6VCVKBl3O07V98kK0++8v+eN9+wJ+fsClS4XN+IlswY4dwB9/AK6uwKuvKltLgwbA9OniePx44OZNZesh2/Hpp8CFC4C/v3LvOti6SgXdxYsXIyQkBB4eHnjggQewS/5pVgKtVos33ngDYWFh8PDwQExMDJKSkozOycrKwpgxYxAcHAxPT0+0b98eu+WhoCLS0tLwxBNPoEaNGvDy8kLbtm1x7ty5YudJkoSePXtCo9Fg/fr1lblEqgBJMt7+11oYdO2H/M9Y3hntXm5uwEsvieOlS61TE1FFzJsnbl94QYQIpY0ZI/7vu3ZN9Egl0mqB2bPF8fjxYrMRKs7koPvVV18hISEB06dPx759+xATE4MePXrg6tWrJZ4/ZcoULFu2DIsWLcKRI0cwbNgw9O3bF/vlFSoAhgwZguTkZCQmJuLQoUPo3r07unbtioyMDMM5J0+eRIcOHRAVFYUtW7bg4MGDmDp1KjxK+MouWLAAGu57Z3GXLwN37oi3pENDrfe67KVrH3JyCnd1Ki3oAsDQoaJl06+/FnbxIFLSyZPAunXi+H//U7YWmatrYZuzJUsKdxskx7V6NXDmDFCvHvDyy0pXY7tMDrrz58/H0KFDMWjQIDRp0gRLly5FtWrVsHLlyhLPT0xMxKRJkxAXF4fQ0FAMHz4ccXFxmPfvr8u5ublYu3Yt5s6di06dOiE8PBwzZsxAeHg4lhTpOzR58mTExcVh7ty5aNmyJcLCwvDEE0+gXr16Rq934MABzJs3r9R6yHzkhWiNGll3XpDceeH8ebETDNmmAwfEqnU/v8IFhCUJCQHi4sQxR3XJFrz/vnjHqmfPwl+sbcHDDwPPPSdq48I0x6bTAW+/LY7/9z/RspFKZlITirt372Lv3r2YOHGi4T4nJyd07doVO3bsKPEx+fn5xUZdPT09sW3bNgBAQUEBdDpdmefo9Xr8+OOPGD9+PHr06IH9+/ejUaNGmDhxIvr06WN4TE5ODp577jksXrwY9evXL/d68vPzkZ+fb/h75r+pSavVQmullTHy61jr9cwpLU0DwAUREXpotTqTHluV665eHfD3d8GlSxr8/XeB0UYE9sCev+am+PNPJwDOaN1aj4ICXZnXPXSoBj/+6IJPPpEwfXoBPD2tXKyFOcrXvCT2du03bgCrVrkA0GDMmAJotZX7/8VS1z17NrBxowt27dJg+fICvPSS7f3/Z29fc3Oy1rV/+aUGx4+7oHZtCUOGFCi+mNfaX3NTXsekoHv9+nXodDr4+fkZ3e/n54ejR4+W+JgePXpg/vz56NSpE8LCwpCSkoJ169ZBpxPByNvbG+3atcOsWbMQHR0NPz8/rFmzBjt27EB4eDgA4OrVq7hz5w7eeecdvPnmm5gzZw6SkpIQHx+PzZs3IzY2FgAwduxYtG/fHr17967Q9cyePRszZ84sdv8vv/yCalb+9SjZDreH2rSpKYBwuLicxk8//V2p56jsddet2x6XLtXFl18exLVr9tmE1R6/5qbYsKEVgCDUqHEMP/103HB/Sdet0wF163bDtWvVMG3aIXTubJ9f0/Ko/WteFnu59m++aYycnGiEht5CTk4qfvqpas9niet++ulQrFzZDOPH6+Hl9St8fGwzUNrL19wSLHntej0wZUpnAD7o0eMofv/9eLmPsRZrfc1zcnIqfK7F2wovXLgQQ4cORVRUFDQaDcLCwjBo0CCjqQWJiYkYPHgwAgMD4ezsjFatWuHZZ5/F3n8nIen/fX+md+/eGDt2LACgRYsW2L59O5YuXYrY2Fhs2LABv/32m9Hc3/JMnDgRCQkJhr9nZmYiKCgI3bt3h481tvmC+K0kOTkZ3bp1g6urq1Ve01w++kjsMdi9ewji4hqa9NiqXndyshMOHgRcXWMQF9fM5McryZ6/5qZ47TXx38tzz0WgR4/wcq/78GEnTJsG/PlnC7z7rn19TcvjKF/zktjTteflAa+8Ir5vp02rjscei6v0c1nyurt3B3btkvD3325ITX0US5aY9o6apdnT19zcrHHt69ZpcO6cC3x8JCxcGI6aNcMt8jqmsPbXPNOEeYsmBV1fX184OzvjypUrRvdfuXKl1KkCdevWxfr165GXl4cbN24gICAAEyZMQGiR1UthYWFITU1FdnY2MjMz4e/vj/79+xvO8fX1hYuLC5rIy+3/FR0dbZje8Ntvv+HkyZOoWbOm0TlPPvkkOnbsiC1bthSrzd3dHe7u7sXud3V1tfo/TiVes6rS08VtkybOcHWt3Mbalb3u++4Tt0ePVv61lWaPX/OKunWr8PvjwQddUPQyS7vul18G3ngD2LnTCYcPO6FFC6uUalVq/pqXxx6u/bPPROP9Bg2A554z/r6tLEtct6sr8OGHQKdOwMqVThg61AkPPGDWlzALe/iaW4qlrl2SgHfeEcejRmlQt65tfX6t9TU35TVMWozm5uaG1q1bIyUlxXCfXq9HSkoK2rVrV+ZjPTw8EBgYiIKCAqxdu7bE6QVeXl7w9/fHzZs3sWnTJsM5bm5uaNu2LY7Jq5/+dfz4cQQHBwMAJkyYgIMHD+LAgQOGPwDw/vvvY9WqVaZcJlVAQYFYmQxYt4eujC3GbNuePeK2USPA17dij/HzA+LjxTEXpZG16fWFLcXGjIFZQq4ldewIDBhQuDBNZ1uDumQhP/4oFvp6eYnvUyqfyVMXEhISMHDgQLRp0wb3338/FixYgOzsbAz6t1PxgAEDEBgYiNn/NnfbuXMnMjIy0KJFC2RkZGDGjBnQ6/UYP3684Tk3bdoESZIQGRmJ9PR0jBs3DlFRUYbnBIBx48ahf//+6NSpEzp37oykpCRs3LjRMFJbv379EkeVGzZsiEaNGpl6mVSO06dF2PX0BAIDrf/68kroM2eA7Gzxj55sR3n9c0szfDjw9dfA558Dc+eKrYOJrOHnn8W24j4+ouWdPZg7F/j+e2DfPmD5cvHvh9RLkoBZs8Txf/8L1KmjbD32wuT2Yv3798d7772HadOmoUWLFjhw4ACSkpIMC9TOnTuHS5cuGc7Py8vDlClT0KRJE/Tt2xeBgYHYtm2b0RSD27dvY8SIEYiKisKAAQPQoUMHbNq0yWhoum/fvli6dCnmzp2LZs2aYcWKFVi7di06dOhQhcunypI3ioiIKHlrV0urU0f0DgSAUtZBkoLkoFvajmiliY0FoqLELy9ffGH+uohKI2/3+/LL9vMLlp8f8Oab4njSJKCUdvakEsnJYrdJT0/b6e9sDyq1GG3kyJEYOXJkiR+7dy5sbGwsjpTz/nK/fv3Qr1+/cl938ODBGDx4cIXrlCTba7uiFkps/XuvJk3Ef+xHjgCtWytXBxUnb5Zo6oiuRgMMGybekluyRBxz7xeytL17gS1bABcXYNQopasxzfDhwMqVwP79wIQJ4pjUp+ho7ssvi19yqGIUGIsjNbCVoAtwnq6tuXQJyMgQI/2tWpn++IEDxYjFoUPA9u3mr4/oXvLc3GeeAYKClK3FVM7OhTumrVoF/PGHsvWQZWzdCmzbJjZnGjdO6WrsC4MuVYocdCMjlatBDrqHDytXAxUnT1uIjhabe5iqZk3g2WfFMRelkaWdPSvmhQP2+3Zwu3bASy+J4xEjxPoJUhd5NPell5RZF2PPGHSpUjiiS6Wp7EK0ooYNE7dffw1cv171mohKs3Ch6FjwyCOw65Z2s2cDtWoBf/0lWo+ReuzYAaSkiKk1r7+udDX2h0GXTJadDVy4II5tIeieOgXk5ipXBxmr7EK0otq2FfOu794Vb8cSWcKtW8BHH4nj115TtJQqq1tXhF0AmDoVuHxZ2XrIfOTR3AEDgH87qpIJGHTJZCdOiNs6dYDatZWro149UYMkAfe0WCaFSJJ5RnSBwlZJy5aJHqdE5rZ8OXDnjtiApkcPpaupuiFDgDZtgMxMzuNUiz17ROs7Jydg4kSlq7FPDLpkMluYtgCI1ficvmBbTp0C/vlHLJho3rxqz/XMM0CNGmJjkl9/NU99RLK7d8W0BUDMzVVDdw9nZzFtQaMRvai3blW6Iqqqt94St889B4Qrv9OvXWLQJZPZStAFGHRtjTyaGxMjwm5VeHmJt+oA0WqMyJy+/BK4eBHw9y9c/KgGbduK9lOA2FRAq1W2Hqq8gweB9evFLy6TJildjf1i0CWT2ULHBRmDrm0x17QFmbwobePGwnnhRFUlSYUtxUaNAtzdla3H3N5+W0zrOnwYWLRI6WqosuTR3KeeEl1sqHIYdMlkHNGl0pg76DZpAnTqJFbFr1hhnuck+vVXMVrm5QW88orS1Zhf7drAnDniePp00dea7EtaGvDNN+J4yhRla7F3DLpkkqILv2wp6KanA/n5ytbi6AoKxA5TQNU6LtxLXpT20Ud8G5bMQ97ud8gQ0ZJLjQYNAh58UCy2s/eOEo5o9mzx87Z376qvd3B0DLpkkuvXRUsewDYmxvv7iwVLOl3hSDMpIy0NyMkRm0SYc1pLfLxonXTxIvDDD+Z7XnJMBw8Cv/wiVrGPHq10NZbj5CQWpjk5ifnIKSlKV0QVdfIksHq1OJ46Vdla1IBBl0wih8mGDcU2rUpj5wXbIU9baN1arP42Fze3wl2fuCiNqkqem/vUU0CjRsrWYmktW4oFaQAwcqToNEG2b/ZsMXjTs6f4/5SqhkGXTGJL83NlTZuKWwZdZZl7fm5Rr7wifqlJTi7s40xkqgsXCkfKHOXt/FmzRM/xo0eB999Xuhoqz9mzwKefimPOzTUPBl0yiS11XJBxRNc2WDLohoSI0Q1ANPknqoxFi8Rc8k6dLPN9aotq1gTefVccv/EGcP68ouVQOebMEd+jXboA7dsrXY06MOiSSWxxRJdBV3l5ecBff4ljcy5EK0puNbZqlXg9IlNkZYld9gDHGc2V/ec/QIcOYg792LFKV0OluXgR+Phjccy5uebDoEsmsaWOCzI56B4/zlX5SvnrLzEK4etrub3Y4+LE3PAbNwrb7hBV1McfA7dvi3ejHntM6WqsS6MBFi8Wc+fXrgU2bVK6IirJu++KedQdOgCxsUpXox4MulRhOp1o4wXYVtBt0ECs9C8o4PxNpRSdtmCprVSdnQt3fFq61DKvQepUUFA4P/V//xOdCBxN8+bAq6+K41dfZTtGW3P1auE7DlOnqmNLalvhgP/cqbLOnxf/Obq5WW7UrjLYeUF5lpyfW9RLLwEuLsD27aJNFFFFfPstcO6cWJT1n/8oXY1yZs4E6tcXAwJyL2GyDfPmAbm54v/Qbt2UrkZdGHSpwuT5ueHh5m0fZQ4MusqyVtCtXx/o21ccs9UYVYQkFS7GGjEC8PBQth4l+fgUtld76y3gzBlFy6F/3bghppYAHM21BAZdqjBbXIgmY4sx5WRmitZFgHVWsss7pX3+uVhgRFSW1FRg3z4RcOWeso7s2WeBhx8Wo4djxihdDQHAwoVAdjbQogXw+ONKV6M+DLpUYba4EE3GEV3l7N0rRs0aNgT8/Cz/eg8/LBYU3bkDfPGF5V+P7Jv8Fv2gQWKxpKOTF6a5uADffw/8+KPSFTm227eBDz4Qx1OmcDTXEhh0qcJseURXDrrHjomFJ2Q91pq2INNoCluNLVkiQjZRSY4cEUFOo2FbraKaNCn8fLz6qhjdJWUsWiTCbpMmhdOyyLwYdKnCbDnoNmwIVKsmWrOcOqV0NY7F2kEXAAYOFG9FHzwI/Pmn9V6X7Mv8+eK2Tx8gIkLRUmzOtGlAYCBw+rTYpICsLyursBvI5MmO2Q3EGvhppQrJyxNbEwK2GXSdnIDoaHF8+LCytTgaJYJurVrAM8+IYy5Ko5JcvgwkJopjR9sgoiKqVy8MWe+8A5w8qWw9jmjJEuCff8QvYf37K12NejHoUoWcPCneIq5RQ7TosUWcp2t9V6+KX4A0GqB1a+u+trwo7euvxaploqIWLxbv8LRrx61US/PUU0DXrqJt5KhRnAZkTTk5hR0wJk2yvU5GasKgSxVSdNqCrU6WZ9C1Pnk0NzJS/BJkTW3bAq1aiR/Sn3xi3dcm25adDXz4oTjmaG7pNBrg//4PcHUFfvoJ2LBB6Yocx0cfiYGCkBDg+eeVrkbdGHSpQmy544KMLcasT4lpC7Kii9KWLgX0euvXQLbpk0/EW8JhYUDv3kpXY9siIwt/GRg9Wow0kmXl5QFz54rjCRPELxpkOQy6VCG2vBBNJo/oHj0qtismy1My6ALAc8+JJvjp6UBKijI1kG3R6QoXoY0dy7eEK2LyZLGg9+xZ4O23la5G/VatAi5eFNvXv/ii0tWoH4MuVYg9BN2QELESPy+PO/5YgyQpH3S9vIABA8Tx0qXK1EC2Zf160Xmldm2GiIry8gIWLBDH775b+P89mZ9WKxb/AcD48YC7u7L1OAIGXaoQewi6zs5AVJQ45vQFyzt3Drh2TTSeb9FCuTrk6Qvffw9kZChXBymv6Ha///2vCHBUMX36AD17igV8r77KhWmWkpgo/u/08wOGDFG6GsfAoEvlunlTBBrAtoMuwAVp1rRrl7ht3lyMpCulaVOgY0fxlvWKFcrVQcrbvh3YuRNwcwNGjlS6Gvui0YgdutzdgV9+AdauVboi9SkoKJwaMm4c4OmpbD2OgkGXynXihLgNCBC9F22ZHHTZS9fylJ62UJTcauyjj7gzniOT2zUNGGCd7ajVJjwceP11cTx2rNhmm8znyy9Fq846dYBXXlG6GsfBoEvlsoeOCzKO6FqPLQXd+Higbl0xdeGHH5SuhpRw4oSYnwsACQmKlmLXJkwAGjUCLlwAZs1Suhr10OmAt94SxwkJtj9opCYMulQue5ifK5NbjKWlsd2UJen1wN694tgWgq67OzB4sDjmTmmO6f33xbzSxx8v3CWRTOfpKaYwAKJ7RVqasvWoxbp1oiNQzZqcVmNtDLpULnsKuqGhYn5eTo6Y8E+WceyY2Kfd07NwFF1pL78s5hn+8gu3M3U0166Jlk0A8L//KVuLGjz+ONCrl5gGNGIEF6ZVlV4PvPmmOB49WrREJOth0KVy2VPQdXERDdABTl+wJHkhWuvW4nNuC0JDgR49xPGyZcrWQta1ZIloK9i6NRAbq3Q16rBwoVhkunkz8NVXSldj3zZuBA4eBLy9xVbLZF0MulQmSSoMunKAtHWcp2t5tjQ/tyh5UdrKlSL4kPrl5optbAGxw5etblFubxo1AiZNEscJCUBmprL12CtJKpzrPGKE6O9M1sWgS2W6eFFMA3B2Fv/x2QMGXcuz1aD72GNAUBBw4wbbIzmKxEQxdaFhQ+Cpp5SuRl3GjROdGC5dAmbOVLoa+5SUJNYzVKvGRZJKYdClMskdF0JD7Wc/brYYs6y7d4EDB8SxrQVdZ2dg6FBxzEVp6qfXF7YUGzvWdqbRqIWHB7BokTheuBA4dEjZeuxN0dHcYcNEZxiyPgZdKpM9zc+VFR3R5SIK8zt0SITdWrWAsDClqyluyBAReP74gz+Y1e7HH8X/UTVqAC+9pHQ16vToo6J9n07HhWmm2rwZ2LFDdIV57TWlq3FcDLpUJnsMuuHhIujcuSN6QZJ5yQvR2ra1zfmQ/v5iO1MAWLpU0VLIwt57T9wOGyYW+pBlvP++eOv999+Bzz9Xuhr7IY/mDh0q/l8iZTDoUpnsMei6uRXWy3m65mer83OLGjZM3CYmcncntdq1C9i6VUypevVVpatRt4YNgalTxfFrrwG3bilajl344w8NtmwR35/jxytdjWNj0KUy2VvHBRkXpFmOPQTdLl3ELztZWcAXXyhdDVmCPDf32WeBwEBla3EECQni58DVq8C0aUpXY/tmzxbx6sUXxQJZUg6DLpVKqwVOnRLH9jSiCzDoWkp2duHn1JaDrkZTOKq7ZAnnFarN6dPAt9+KY24QYR1uboVt3BYvLlyQSoV0OiA1VYOvvorAL784wclJbKlMymLQpVKdOiX+4VarBgQEKF2NaRh0LWPfPrHSPSDA9r8nBg4Uq8b/+gvYuVPpasicFiwQ34fduwPNmytdjePo2hXo10987v/7X26zXtS6dUBICNCtmwvWrBE/gDw8+AuBLWDQpVIVnZ9ri4uOylK0xRhH88xHnrZw//3K1lERtWsD/fuLY7YaU49//gE+/lgccyW79c2fD3h5iW4Cn36qdDW2Yd060cP53sXPubni/nXrlKmLBAZdKpU9LkSTNW4MODkBt2+LZudkHkU7LtgDeae0r74SAYns37JlYgpN8+ZihJGsKzAQmDFDHI8fz39XOh0wenTJAyryfWPGiPNIGQy6VCp7XYgGiL6F4eHimNMXzMceFqIVdf/9QIsWQH4+8MknSldDVZWfX7iBAbf7Vc7o0eJds+vXgSlTlK5GWampZbexlCTg/HnRmo2UwaBLpbLnEV0AaNpU3DLomseNG4WLE9u0UbaWitJoCkd1ly7lnEJ7t2aNeIcmMLBwWgpZn6urWJAGiH9Xe/YoW4815OWJDWi++QZ44w3gueeAli3FhhoVwXcWlcMNE6lU8va/9hp0mzQBvvuOQddc5B9m4eFiVzR78dxzYvTvxAmxU9EjjyhdEVWGJBVuEDF6tOgCQMp5+GHg+edF+77//hf4808xXcze/fMPkJYGHD0q/sjHp09X7RdlbhihHAZdKlFWVuFvoBERytZSWey8YF72tBCtqOrVgf/8B/jwQ7EojUHXPm3aJBaXensDL7+sdDUEAO++C2zcKP5vWLHCfr4uej1w7pxxkJWPr10r/XE1agDR0UBUVOFt48ZAt25ARkbJ83Q1GqBBA6BjR8tdD5WNQZdKdOKEuK1b175G74q6t/MC5/NVjb0tRCtq+HARdNevBy5etP3WaFScPJo7ZIgIHKQ8f3/xNv6YMcDEiUB8PODrq3RVhfLyxM8yOczKt8eOiY4IpQkKKgyyRUOtn1/JP0cWLhTdFTQa47Arn7tgAeDsbNZLIxMw6FKJ7H1+LiAW0Wk04q2oq1fFf1JUOZJkfwvRirrvPqBDB2DbNtGaSt7OlOzD/v1ASooIC6NHK10NFTViBLByJXDwoAi7H31k/Rpu3Ch5dPb06dLbS7q6ip9vRYNsVJT4uVG9ummvHx8vNjAZPdp4YVqDBiLkxsdX+tLIDBh0qUT23HFB5ukJhIYCJ0+K6QsMupWXkQFcviyCRsuWSldTOcOGiaC7fLn4gezC//3shrzdb79+QHCwsrWQMRcXsTCtY0cxfeHFF4HcXA22bg2El5cGnTubZzRTnm5w7+js0aMVm25w75SDRo3M+39AfDzQuzeweXMBfv75AHr2bIHOnV04kmsD+F89lUgNI7qAmL4gB93OnZWuxn7Jo7lNm4qd8uzRU0+Jt1gvXAB+/FH8UCLbd/488OWX4pjb/dqmDh3EToSffir+n9VqXQC0wfz5YlRz4cKKj2rm5orpBkXDbFqa+JlU1nSDhg2Lj85GRwP16llv2pqzMxAbKyE7OwOxsTEMuTaCQZdKZO8dF2RNm4rFElyQVjX2PG1B5u4ODB4MzJ0rWiIx6NqHDz4QzfY7dwZat1a6GipNbKwIulqt8f0ZGeKXzG+/NQ67N26U3t2gtOkGbm5icXRJC8JMnW5AjoNBl4qRJHWN6AIMulUlL0Szt44L93r5ZRF0N20SPYFDQ5WuiMpy+7bYCQ3gdr+2TKcDpk0r+WNyaH3pJeCnn8QgytGjYrOJ0tSsWTzMRkcDISGcckSm47cMFXP1KpCZKd7uCQtTupqqYdCtOr2+sIeuPY/oAuL7uUcPEXSXLQPmzFG6IirLihWi1WGTJhVvzE/W9/vvZe8OBgC3bomFoEXdO91AvrXmdANSPwZdKkYezQ0OBjw8lK2lqqKixO3Vq2IEwZZa39iL9HQxsubhIboX2Lvhw0XQXblStEZyd1e6IiqJVitWrANibq4aNiNQq4ru+tW3r5jGIHc38PKybF1EALcAphKooeOCzMtLvN0FiDlgZDp5fm6LFqIlj7177DGxQOb6dWDtWqWrodJ8/bUYJfTzEztwke2q6K5fo0aJnQpbtWLIJeth0KVi1DI/V1Z04wgynRoWohXl4gIMHSqOlyxRthYqWdHtfl99laPutq5jR/HLY2nTDTQasQkDdwcjJTDoUjFq6bgg4zzdqlHLQrSihgwRrYC2bQP+/lvpauhev/0GHDggWtkNG6Z0NVQeZ2fRQgwoHna5OxgpjUGXilHbiG7TpuKWQdd0Wq3YlQpQz4guILYAltuLLV2qbC1UnLxBxODBQJ06ytZCFSPvDhYYaHx/gwbFW4sRWRODLhnR6cTiI0A9QZcjupV3+LDYL97HR/SvVJPhw8XtZ58Bd+4oWwsV+vtv4OefxeKzMWOUroZMER8PnDkDJCcXICFhD5KTC3D6NEMuKYtBl4ycPStG8dzdxZwqNYiOFreXLgE3bypbi72R5+e2aaO+Ve9duojwnpUFrFmjdDUkmz9f3MbH2397Q0ck7w7WqVMGYmMlTlcgxansRxdVlTxtISJCPfOpvL0LQzs7L5hGbQvRinJyAl55RRwvWVL6bkxkPZcuAZ9/Lo65QQQRmQODLhlR2/xcGacvVI4cdNW0EK2oF18U717s31+46I6Us2iReEepQwfggQeUroaI1IBBl4yoreOCjC3GTJeTAxw6JI7VOKILiIVO/fuLYy5KU9adO4Xt3v73P2VrISL1YNAlIxzRJdmBA2Jxop+fWDmtVnL7qi+/BP75R9laHNnKlWKb2IgIoFcvpashIrVg0CUjDLokKzo/V837zj/4IBATI7pLfPqp0tU4poIC4P33xXFCgnrWBxCR8hh0ySA3Fzh3ThyrNeheuABkZipbi71Q80K0ojSawlZjS5dyUZoSvvtOtKXy9QUGDFC6GiJSEwZdMpD759aqJX7gqEnNmmKTAICdFypK7QvRinruOaB6dfGOxubNSlfjWCQJePddcTxihNgNjYjIXBh0yaDoQjQ1vlXN6QsVd+tW4TSWNm0ULcUqvL2B//xHHMsLosg6tm0Tv1R5eAD//a/S1RCR2jDokoFa5+fKGHQrbs8ecduokfpG90sjL0pbv170cyXreO89cTtwIFCvnrK1EJH6MOiSgaMEXbYYK5+jzM8tqnlzoH17sTDq44+VrsYxHDsGbNgg3kEaO1bpaohIjRh0ycBRgi5HdMvniEEXKFyUtny5aK1GliVv99urFxAZqWwtRKRODLpkIAddtf7AkYPu2bOiOT2VzlGD7lNPiU0kzp8HfvpJ6WrU7erVwnZu3O6XiCyFQZcAADduiD8AEB6ubC2WUqeO2PwAAI4eVbYWW3bpkmjD5uQEtG6tdDXW5eEBDBokjrkozbIWLwby80VXjw4dlK6GiNSKQZcAFI7mNmgAeHkpW4slcfpC+eTR3Oho0XLL0bzyirhNSgJOn1a2FrXKyQE+/FAcv/aaOru8EJFtYNAlAOqfnytj0C2fo05bkIWHA927i/6uy5crXY06ffYZcP266OrRt6/S1RCRmjHoEgAGXSrk6EEXKGw19vHH4u11Mh+drnAR2tixgIuLsvUQkbox6BIABl0SJIlBFxBdAAIDgWvXgHXrlK5GXTZuBE6cEDswyvOhiYgshUGXAKi/44JMDrqnTol5gmTs1Cngn38ANzfRV9ZRubgAQ4eK46VLla1FbeQNIoYPd8w54ERkXQy6BL1ejLAA6h/RrVtXdF+QpMItj6mQPJobEwO4uytbi9KGDAGcnYGtW7nJiLns2AH88Yf4RWrkSKWrISJHwKBLuHAByM0Vo1ghIUpXY1kaDdC0qTjm9IXiOG2hUGAg8MQT4pijuuYxb564ff55wN9f2VqIyDEw6JJh2kJYmGMsDOE83dIx6BqTd0r77DMgO1vZWuzdyZOF853/9z9layEix8GgSw6zEE3GoFuyggJg715xzKArPPKI+AUwMxNYs0bpauzbokVOkCSgZ8/Cd1WIiCyNQZcYdAkAkJYmFuhVrw5ERSldjW1wcipsNbZkiZjbTabLzHTFJ5+IHzfc7peIrIlBlxym44JMDrrp6eyRWpQ8baF1a7EIi4QXXxQL8/btA/bsUboa+6LTAampGixbFoOcHA1atAA6d1a6KiJyJAy65HAjuvXrAzVrim4T7LxQiPNzS+brCzz9tDheskTZWuzJunVicWu3bi74449AAMC5c8B33ylbFxE5FgZdB5efD5w+LY4dJehqNJy+UBIG3dLJi9K+/BK4eVPZWuzBunXAU0+Jji5F3bwp7ucmHERkLZUKuosXL0ZISAg8PDzwwAMPYNeuXaWeq9Vq8cYbbyAsLAweHh6IiYlBUlKS0TlZWVkYM2YMgoOD4enpifbt22O3/FO3iLS0NDzxxBOoUaMGvLy80LZtW5w7dw4A8M8//+DVV19FZGQkPD090bBhQ4waNQq3b9+uzCU6jFOnxMhm9epipNNRsMWYsfx84OBBccygW1y7dmIDjdxc0YGBSqfTAaNHlzyfWb5vzBhxHhGRpZkcdL/66iskJCRg+vTp2LdvH2JiYtCjRw9cvXq1xPOnTJmCZcuWYdGiRThy5AiGDRuGvn37Yv/+/YZzhgwZguTkZCQmJuLQoUPo3r07unbtioyMDMM5J0+eRIcOHRAVFYUtW7bg4MGDmDp1Kjw8PAAAFy9exMWLF/Hee+/h77//xieffIKkpCS89NJLpl6iQyk6bUGjUbYWa+KIrrG//gK0WvE2vdp7KVeGRlO4KG3pUi5KK8vvvxcfyS1KkoDz58V5RESWZnLQnT9/PoYOHYpBgwahSZMmWLp0KapVq4aVK1eWeH5iYiImTZqEuLg4hIaGYvjw4YiLi8O8fzuH5+bmYu3atZg7dy46deqE8PBwzJgxA+Hh4VhSZELc5MmTERcXh7lz56Jly5YICwvDE088gXr16gEA7rvvPqxduxa9evVCWFgYunTpgrfeegsbN25EQUFBZT43DsHR5ufKGHSNyW/KtG3rWL/wmOKFF8Q7H0ePAlu2KF2N7Tp1qmLnXbpk2TqIiADApO0B7t69i71792LixImG+5ycnNC1a1fs2LGjxMfk5+cbRl1lnp6e2LZtGwCgoKAAOp2uzHP0ej1+/PFHjB8/Hj169MD+/fvRqFEjTJw4EX369Cm13tu3b8PHxwcupeyCkJ+fj/wiy+4zMzMBiOkWWq221Oc1J/l1rPV69zp61BmAE8LDddBq9VZ7XaWvOyICAFxx4oSE7OwCuLlZ77WVvvaS7Nwpvg9atbLc94EtXrcpPDyA555zwvLlzvjwQz06dKj4e+/2fu0VodUCK1Y4YepUJwDl/7ZUt24BtFr1Do07wte8NLx2x7t2a1+3Ka9jUtC9fv06dDod/Pz8jO738/PD0aNHS3xMjx49MH/+fHTq1AlhYWFISUnBunXroPt3gpa3tzfatWuHWbNmITo6Gn5+flizZg127NiB8PBwAMDVq1dx584dvPPOO3jzzTcxZ84cJCUlIT4+Hps3b0ZsbGyJtc6aNQsvv/xyqdcze/ZszJw5s9j9v/zyC6pVq1bhz4s5JCcnW/X1ZDt3PgTAF9nZ+/HTTxnlnm9uSl23JAGennHIzXXFqlW/Iygoy+o1KHXtJdmypQsAbwC78dNPVyz6WrZ03aaKjvYB0BnffQd88UUKatUyrT+dPV97aSQJ2LmzPj77rCkuXqwOAHBy0kOv16DkwCvB1zcXmZnJ+Oknq5aqCDV+zSuK1+54rHXdOTk5FT5XI0kVn2128eJFBAYGYvv27WjXrp3h/vHjxyM1NRU7d+4s9phr165h6NCh2LhxIzQaDcLCwtC1a1esXLkSubm5AMT828GDB2Pr1q1wdnZGq1at0LhxY+zduxdpaWmG13322WexevVqw3M/8cQT8PLywpp7tizKzMxEt27dULt2bWzYsAGurq4lXk9JI7pBQUG4fv06fHx8KvppqRKtVovk5GR069at1DotKSjIBVeuaLBjRwFat7be6IrS1w0AHTo4Y9cuJ6xeXYCnnnKsay8qKwvw9XWBJGlw/rwW9/weaza2dt2V1amTM/780wkzZ+owcWLFRr/Vcu332rVLg9dfd8Iff4hZcHXrSpg2TY86dSQ8/7xoxixJhWFXoxH/zr78Uoe+fdU7mguo92teEbx2x7t2a193ZmYmfH19De/cl8WkEV1fX184OzvjyhXjEZ8rV66gfilL9uvWrYv169cjLy8PN27cQEBAACZMmIDQ0FDDOWFhYUhNTUV2djYyMzPh7++P/v37G87x9fWFi4sLmsgTK/8VHR1tmN4gy8rKwqOPPgpvb2989913ZX7C3d3d4e7uXux+V1dXq3+DKvGat28D8pcyOtoFSvybVOK6ZU2birmpx4873rUXdfCgGJULCgIaNLB8PbZy3ZX13/8Cf/4JfPyxMyZPdjZpcw17v3bZqVPApEnAV1+Jv3t6AgkJwPjxGvj4iE+Iu7vovlB0YVqDBhosWADEx5v0o8euqeVrXhm8dse7dmtdtymvYdJiNDc3N7Ru3RopKSmG+/R6PVJSUoxGeEvi4eGBwMBAFBQUYO3atejdu3exc7y8vODv74+bN29i06ZNhnPc3NzQtm1bHLunu//x48cRHBxs+HtmZia6d+8ONzc3bNiwodi8XzJ24oS49fMDatRQthYlcEGaIC9Eu/9+ZeuwF08/DdSuLTY/+Plnpauxrn/+EYE2KkqEXI1G7Bx3/Djw5ptA0YGV+HjgzBkgObkACQl7kJxcgNOnxf1ERNZi8q/VCQkJGDhwINq0aYP7778fCxYsQHZ2NgYNGgQAGDBgAAIDAzF79mwAwM6dO5GRkYEWLVogIyMDM2bMgF6vx/jx4w3PuWnTJkiShMjISKSnp2PcuHGIiooyPCcAjBs3Dv3790enTp3QuXNnJCUlYePGjdjy7/JnOeTm5OTg888/R2ZmpmFxWd26deHMPU2LcdSOCzL20hW4UYRpPDyAQYOAefPETmmPP650RZaXnw/83/+JMHvrlrivWzfg3XeBmJjSH+fsDMTGSsjOzkBsbAy3liYiqzM56Pbv3x/Xrl3DtGnTcPnyZbRo0QJJSUmGBWrnzp2Dk1PhQHFeXh6mTJmCU6dOoXr16oiLi0NiYiJq1qxpOOf27duYOHEiLly4gNq1a+PJJ5/EW2+9ZTQ03bdvXyxduhSzZ8/GqFGjEBkZibVr16JDhw4AgH379hnmCMuL2GSnT59GCJuDFiMH3chIZetQijyie+wYUFAAlNKcQ/UYdE33yisi6P78sxi1VOt/L5IkRm4nTSrcQbFZMxFwe/RQtjYiooqo1I/2kSNHYuTIkSV+bMs9DSZjY2NxpJwhs379+qFfv37lvu7gwYMxePDgEj/28MMPw4R1dQQR8ADHHdENCgK8vIDsbODkSccM/NeuAWfPiuPWrZWtxZ5ERABduwK//gosXw68/bbSFZnf1q3Aa68V/iIUEADMmgUMHAiOzBKR3ajUFsCkDo4+dcHJCYiOFseOOn1BDjGRkY45T7sqhg8Xtx9/DNy9q2wt5nTsGNCnDxAbK74/qlcH3nhD/H8xeDBDLhHZFwZdByVJDLoAF6RxIVrl9eolRjmvXgW++07paqru6lVgxAgxd/3770WgHTYMSE8Hpk4V734QEdkbBl0HdfkycOeOGNUs0unN4chB9/BhZetQCufnVp6rKzBkiDguslu53cnJEVMvwsOBDz8EdDoR4g8dEtdlqb7KRETWwKDroOTR3EaNRL9LR+XII7qSxKBbVUOHipHP1FT7+x7S6YBPPhHv6EyeLDYOad0a2LwZ2LChcFoPEZE9Y9B1UJy2IMgtxo4eFT/4Hcm5c2IxmosL0KKF0tXYpwYNxOgnACxbpmwtpkhOFqF20CAgIwNo2BD44gsxleXhh5WujojIfBh0HZSjd1yQBQeLXZ3y8wvbJzkKeTS3WTPRG5YqZ9gwcfvpp6KDhy07dAjo2RPo3h346y+xAHHuXPH/wXPPialMRERqwv/WHBRHdAVnZ7HLE2B/bz1XFReimUe3bkBYmNhS+8svla6mZBcvAi+9JEbuk5LE/OLRo0VbvXHj+IsOEakXg66DYtAt5KjzdDk/1zycnMQGEgCwdKmytdwrKwuYNk30/V25EtDrgaeeEt/rCxYAdeooXSERkWUx6DqgggIxkgMw6AKOGXT1emDvXnHMoFt1L74IuLkBe/aIP0orKBBzhiMixCYPOTlA+/bA9u3AN9+IDgtERI6AQdcBnTkjfhB6eorFNI7OEVuMHTsmRvs8PQuvnyqvbl3g6afFsZKtxiQJ+OEHoHlzMXf4yhURar/9Fti2DWjXTrnaiIiUwKDrgORpCxERXHwCFAa9tDQx0ukI5GkLrVqJrgtUdfJOaWvWALduWf/19+4FunQRXSDS0sS0hIULxS9wTz4JaDTWr4mISGmMOQ6IHReMhYaKXsK5ucDZs0pXYx1y0OVCNPNp3x647z7xffTZZ9Z73bNngRdeANq0AbZsEd/Lr78udjQbNUpMqSAiclQMug6IC9GMubgAkZHi2FHm6codFzg/13w0msJR3aVLxTQCS7p1SwTayEjRAxcQgffYMeCdd4CaNS37+kRE9oBB1wEx6BbnSAvS7t4FDhwQxwy65vXCC4CXl5g6sHWrZV7j7l0xJSE8XPTAzc8HOncWi+ASE0VvaCIiEhh0HRCDbnGOFHQPHRJhqVYt0f+VzMfHB3j+eXFs7kVpkiQWlTVpAowZA9y4Ibbp/eEHICVF7HRGRETGGHQdTHY2cOGCOJbfrifHCrry/Nw2bbhAyRLk6Qvr1omuB+awfTvw0EOis8PJk4Cfn2gfdvAg8Nhj/DoSEZWGQdfBpKeL2zp1gNq1la3FlhQNupaeW6k0LkSzrBYtgAcfBLRasUlDVaSniw0eHnoI2LEDqFZNbABx4gTw8svsmEFEVB4GXQfDjgslCw8XoeHOHeD8eaWrsSwuRLO8YcPE7bJlgE5n+uOvXxdb9DZpAqxdK9oAvvSSCLgzZwLe3uatl4hIrRh0HQzn55bM1bXwc6Lm6QvZ2YXXx6BrOf36iTnQZ88CmzZV/HF5eWKBWXg48MEHYlT40UfF4sEVK4CAAIuVTESkSgy6DoZBt3RNm4pbNQfdffvEphgBAQxNluTpCQwaJI4rsihNrxctwiIjRcuw27eBmBggORn4+WegWTPL1ktEpFYMug6GQbd0jrAgTZ6fy9Fcy3vlFXH7449lb0SyebP4erzwAnDunNiW+5NPxE5nXbtapVQiItVi0HUgklQ4R5cdF4pj0CVzatwYeOQR8e9u2jQnbN0aiNRUjWHO7pEjYrveLl3ESLu3N/D22+KX0YEDAWdnZesnIlIDrtl1INevi92UADEHkIzd23lBjS2b5IVo7LhgHS1bih63a9Y4A2iD+fMBf3+xVXBKipiy4OwsFq9NmwbUq6d0xURE6sKg60DkaQsNG4o5hGQsIkKEjtu3gUuX1DeH9cYN4NQpcdymjbK1OIJ164B584rff+mS+AMAffqI7Xr5DgsRkWVw6oID4fzcsrm7F450Hz6sbC2WsGePuA0PFx0ByHJ0OtEerKyezPXqiZ3OGHKJiCyHQdeBMOiWT83zdDk/13p+/71wB8LSXL0qziMiIsth0HUgDLrlU3OLMQZd65GnJpjrPCIiqhwGXQciB12+VVo6tY7oShIXolmTv795zyMiosph0HUQOp3YPhTgiG5Z5KB7+HDZ8yvtTUYGcPmyWGzXsqXS1ahfx46iH25pnTs0GiAoSJxHRESWw6DrIM6fB/LzxVa3wcFKV2O7GjcGnJyAmzfFHEq1kKctNG0KVKumbC2OwNkZWLhQHN8bduW/L1jAXrlERJbGoOsg5GkL4eH84VoWT08gNFQcq2n6AufnWl98vOiqEBhofH+DBuL++Hhl6iIiciQMug6CC9Eqruj0BbVg0FVGfDxw5gyQnFyAhIQ9SE4uwOnTDLlERNbCoOsguBCt4tS2IE2vLwy6XIhmfc7OQGyshE6dMhAbK/EdFSIiK2LQdRAc0a04tbUYS08Xu715eIitZ4mIiBwFg66DOHZM3DLolk9tI7ryaG6LFmIxIhERkaNg0HUAeXnA2bPimEG3fFFRYmX8tWvij73j/FwiInJUDLoO4ORJ0RPWxweoV0/pamxftWpASIg4TktTtBSzYNAlIiJHxaDrAIrOzy2tgT0ZU8v0Ba0W2L9fHHMhGhERORoGXQfAjgumU0uLscOHgdxcMZofEaF0NURERNbFoOsA2HHBdGoZ0ZWnLbRpI3Z8IyIiciT80ecA2HHBdGppMcb5uURE5MgYdB0AR3RNFxUlbi9fBv75R9laqoJBl4iIHBmDrsrdvFnYIotzNCvO2xto2FAc22vnhdxc4NAhccygS0REjohBV+VOnBC3/v4ivFHF2fs83f37AZ0O8PMDgoKUroaIiMj6GHRVjh0XKs/eg27RaQtsK0dERI6IQVflOD+38uy9xRjn5xIRkaNj0FU5dlyoPDWN6BIRETkiBl2V44hu5UVHi9uMDOD2bWVrMdWtW4VfewZdIiJyVAy6KiZJDLpVUbMmEBgoju2t88KePeK2USPA11fZWoiIiJTCoKtiFy8COTmAs7MIPGQ6e52+wGkLREREDLqqJo/mhoYCbm7K1mKvGHSJiIjsF4OuinEhWtUx6BIREdkvBl0V4/zcqrPHoHvpEnDhguid26qV0tUQEREph0FXxRh0q07uvHD2LHDnjrK1VJQ8mtukCXfDIyIix8agq2IMulVXp47YQhewn84LnLZAREQkMOiqlFYLnDoljhl0q6ZpU3FrL9MXGHSJiIgEBl2VOn0a0OmAatUKe8FS5djTPF1JYtAlIiKSMeiqVNGOCxqNsrXYO3sKuqdPA//8A7i6As2bK10NERGRshh0VYrzc83HnoLurl3itkULwN1d0VKIiIgUx6CrUgy65iMH3dOnxU5ztozTFoiIiAox6KoUg6751K0L+PqK+a9HjypdTdkYdImIiAox6KqUHHQjI5WtQy3sYfqCTgfs2yeOGXSJiIgYdFXpzh3g4kVxHBGhbC1qYQ8txtLSgOxswMsLiIpSuhoiIiLlMeiqkDyaW7cuUKuWsrWohT2M6MrTFtq0AZydla2FiIjIFjDoqhDn55qfPQRdueMCpy0QEREJDLoqxKBrfnLQPXkSyMtTtpbScCEaERGRMQZdFWLQNT8/PzENRK8v/Pzakvx84OBBccygS0REJDDoqhA7LpifRlM4qnv4sLK1lOSvvwCtFqhTBwgJUboaIiIi28CgqzKSxBFdS7HlebpFpy1wy2ciIiKBQVdlrl4Fbt8WYScsTOlq1MWWW4zJC9Huv1/ZOoiIiGwJg67KyKO5wcGAh4eytaiNvYzoEhERkcCgqzKctmA5ctA9cQK4e1fZWorKyircmphBl4iIqBCDrsow6FpOQADg4yO22j1xQulqCu3dK+ZmBwWJ7hBEREQkMOiqDDsuWE7Rzgu2NH2B0xaIiIhKxqCrMhzRtSxbDLpciEZERFQyBl0V0emA9HRxzKBrGbbYS5cjukRERCVj0FWRs2fFIil3dzFfk8zP1kZ0r10TX3cAaN1a2VqIiIhsDYOuisjTFsLDAWdnZWtRK7mX7vHjYicypcmjuZGRQI0aytZCRERkaxh0VYTzcy0vKAioXl2E3JMnla6G0xaIiIjKwqCrIuy4YHkaDRAdLY5tYfoCF6IRERGVjkFXRY4dE7cc0bUsW5mnK0kc0SUiIioLg66KcOqCddhK0D13TixGc3EBWrRQthYiIiJbxKCrErm5IvgADLqWZistxuTR3GbNAA8PZWshIiKyRQy6KiH3z61ZE/D1VbQU1ZOD7rFjQEGBcnVw2gIREVHZGHRVoui0BY1G2VrULiQE8PQE8vOB06eVq0MOulyIRkREVDIGXZVgxwXrcXJSvvOCXg/s2SOOOaJLRERUMgZdlWDHBetSekHasWNAVpYYWZZrISIiImMMuirBjgvWpXTQlacttGolui4QERFRcQy6KsGga122EnQ5bYGIiKh0DLoqcOOG+AMAERHK1uIo5KCblibmy1obgy4REVH5GHRV4MQJcdugAeDlpWwtjiI0FHB3F/2Lz5yx7mvfvQvs3y+O2XGBiIiodJUKuosXL0ZISAg8PDzwwAMPYNeuXaWeq9Vq8cYbbyAsLAweHh6IiYlBUlKS0TlZWVkYM2YMgoOD4enpifbt22O3PGRVRFpaGp544gnUqFEDXl5eaNu2Lc7JuyQAyMvLw4gRI1CnTh1Ur14dTz75JK5cuVKZS7QrnLZgfc7OQFSUOLb29IVDh0TYrVULCAuz7msTERHZE5OD7ldffYWEhARMnz4d+/btQ0xMDHr06IGrV6+WeP6UKVOwbNkyLFq0CEeOHMGwYcPQt29f7JeHpAAMGTIEycnJSExMxKFDh9C9e3d07doVGRkZhnNOnjyJDh06ICoqClu2bMHBgwcxdepUeBTZEmrs2LHYuHEjvvnmG6SmpuLixYuIj4839RLtDjsuKEOpebry74Bt2rBnMhERUVlMDrrz58/H0KFDMWjQIDRp0gRLly5FtWrVsHLlyhLPT0xMxKRJkxAXF4fQ0FAMHz4ccXFxmDdvHgAgNzcXa9euxdy5c9GpUyeEh4djxowZCA8Px5IlSwzPM3nyZMTFxWHu3Llo2bIlwsLC8MQTT6BevXoAgNu3b+Pjjz/G/Pnz0aVLF7Ru3RqrVq3C9u3b8eeff1bmc2M3OKKrDKWDLufnEhERlc2kxkR3797F3r17MXHiRMN9Tk5O6Nq1K3bs2FHiY/Lz841GXQHA09MT27ZtAwAUFBRAp9OVeY5er8ePP/6I8ePHo0ePHti/fz8aNWqEiRMnok+fPgCAvXv3QqvVomvXrobniIqKQsOGDbFjxw48+OCDJdaWn59v+HtmZiYAMd1Cq9VW9NNSJfLrVOX1jh1zAaBBaGgBtFrJTJVZljmuW2mNG2sAuODwYT20Wl2FH1fVa9+1S3y9W7a0n683oI6veWXx2h3v2h31ugFee9FbR2Ht6zbldUwKutevX4dOp4Ofn5/R/X5+fjh69GiJj+nRowfmz5+PTp06ISwsDCkpKVi3bh10OhEMvL290a5dO8yaNQvR0dHw8/PDmjVrsGPHDoSHhwMArl69ijt37uCdd97Bm2++iTlz5iApKQnx8fHYvHkzYmNjcfnyZbi5uaFmzZrFart8+XKJtc2ePRszZ84sdv8vv/yCatWqmfKpqbLk5ORKPU6vB44dewyACy5e3IKffso2b2EWVtnrtgXXrlUH8Aj+/luPH3/8yeRpBJW59rw8Zxw58hgAICsrBT/9lGfycyjNnr/mVcVrdzyOet0Ar90RWeu6c3JyKnyuxVvNL1y4EEOHDkVUVBQ0Gg3CwsIwaNAgo6kOiYmJGDx4MAIDA+Hs7IxWrVrh2Wefxd69ewGIEV0A6N27N8aOHQsAaNGiBbZv346lS5ciNja2UrVNnDgRCQkJhr9nZmYiKCgI3bt3h4+PT2Uv2SRarRbJycno1q0bXF1dTX78+fPA3bsucHGR8OKLsXazeUBVr9sWaLXA2LES8vJc0KxZHBo2rOjjKn/t27ZpoNdrEBAg4YUXulSiauWo4WteWbx2x7t2R71ugNfuiNdu7euW34GvCJNika+vL5ydnYt1Mrhy5Qrq169f4mPq1q2L9evXIy8vDzdu3EBAQAAmTJiA0NBQwzlhYWFITU1FdnY2MjMz4e/vj/79+xvO8fX1hYuLC5rcs9dpdHS0YXpD/fr1cffuXdy6dctoVLes2tzd3eHu7l7sfldXV6t/g1b2NU+fFrdhYRp4etrfPyolPtfm4uoq5kUfPgwcP+5qcgeEyly7vIazbVuNHX/e7PdrXlW8dse7dke9boDX7ojXbq3rNuU1TFqM5ubmhtatWyMlJcVwn16vR0pKCtq1a1fmYz08PBAYGIiCggKsXbsWvXv3LnaOl5cX/P39cfPmTWzatMlwjpubG9q2bYtjcnuBfx0/fhzBwcEAgNatW8PV1dWotmPHjuHcuXPl1mbP2HFBWU2biltrLUjjQjQiIqKKM/mN7oSEBAwcOBBt2rTB/fffjwULFiA7OxuDBg0CAAwYMACBgYGYPXs2AGDnzp3IyMhAixYtkJGRgRkzZkCv12P8+PGG59y0aRMkSUJkZCTS09Mxbtw4REVFGZ4TAMaNG4f+/fujU6dO6Ny5M5KSkrBx40Zs2bIFAFCjRg289NJLSEhIQO3ateHj44NXX30V7dq1K3Ehmlqw44KyrN15gUGXiIio4kwOuv3798e1a9cwbdo0XL58GS1atEBSUpJhgdq5c+fg5FQ4UJyXl4cpU6bg1KlTqF69OuLi4pCYmGg0veD27duYOHEiLly4gNq1a+PJJ5/EW2+9ZTQ03bdvXyxduhSzZ8/GqFGjEBkZibVr16JDhw6Gc95//304OTnhySefRH5+Pnr06IEPP/ywMp8Xu8GgqyxrBt0bN4CTJ8VxmzaWfz0iIiJ7V6mlSyNHjsTIkSNL/Jg8wiqLjY3FkXJSQL9+/dCvX79yX3fw4MEYPHhwqR/38PDA4sWLsXjx4nKfSy0YdJVVNOhKkmU3cNizR9yGhwO1a1vudYiIiNSiUlsAk224e7dwMVpkpLK1OKqICLEdcGYmcPGiZV+L0xaIiIhMw6Brx06eFH10q1cHSmksQRbm5ibCLmD56QsMukRERKZh0LVjRactWPItcyqbPH3h8GHLvYYkAbt2iWMGXSIiooph0LVjnJ9rG6yxIC0jA7h8WUyTaNnScq9DRESkJgy6doxB1zZYo5euPG2haVPAy8tyr0NERKQmDLp2jEHXNtzbecESOD+XiIjIdAy6dkwOuuy4oKzGjQEnJ+DmTeCe3bHNhkGXiIjIdAy6diozU8zZBApX/ZMyPDyAsDBxbInpC5JU2EOXQZeIiKjiGHTtlDya6+cH1KihbC1k2QVp6enArVuAuzvQrJn5n5+IiEitGHTtFOfn2hZLthiT24q1bAkU2RWbiIiIysGga6cYdG2LJUd0OT+XiIiochh07RSDrm2xZIsxBl0iIqLKYdC1U+y4YFsiI8XudNevA9eume95CwqA/fvFMYMuERGRaRh07ZAkAceOiWOO6NqGatWARo3EsTlHdQ8fBnJzAR8ffq2JiIhMxaBrhy5fBu7cEb1bQ0OVroZklpinKy9Ea9NGfL2JiIio4vij0w7J0xZCQkTLKbINlgi6nJ9LRERUeQy6dogL0WwTgy4REZFtYdC1Q1yIZpvM3Us3Nxc4dEgcM+gSERGZjkHXDnFE1zZFR4vbK1eAGzeq/nwHDgA6HVCvHhAUVPXnIyIicjQMunaIHRdsU/XqQHCwOE5Lq/rzyQvR7r9ftC4jIiIi0zDo2pmCAuDkSXHMoGt7zDlPl/NziYiIqoZB186cOSPCrocH0KCB0tXQvRh0iYiIbAeDrp2R5+dGRLCvqi0yV9C9davwa82gS0REVDmMSnaGHRdsm7mC7t694jYkBPD1rdpzEREROSoGXTvDjgu2Te68kJEhRmUrq+hCNCIiIqocBl07w44Ltq1GDSAwUBxXpfMC5+cSERFVHYOuneGIru1r2lTcVmX6AoMuERFR1THo2pHsbODCBXHMoGu7qjpP9/Jl8XXWaIBWrcxXFxERkaNh0LUj6enitnZtoE4dZWuh0lU16MqjudHRgLe3eWoiIiJyRAy6doQdF+yDuYIuF6IRERFVDYOuHeFCNPsgd144dw7IyjL98XLHBc7PJSIiqhoGXTvChWj2oXZtoH59cWxq5wVJ4kI0IiIic2HQtSMMuvajstMXTp8G/vkHcHUFmjc3f11ERESOhEHXTkgSpy7Yk8q2GJNHc2NiAHd389ZERETkaBh07cSNG4U7bYWHK1oKVUBlR3Q5bYGIiMh8GHTthDxtoWFDoFo1ZWuh8lU26HLrXyIiIvNh0LUTnLZgX+Sge+aM2OijInQ6YN8+ccwRXSIioqpj0LUTXIhmX3x9gbp1jedWlyctTYRiLy8gKsqy9RERETkCBl07waBrf+RR3cOHK3a+PD+3dWvA2dkyNRERETkSBl07waBrf0ydp8uFaERERObFoGsH9HrgxAlxzKBrP0xtMcaFaERERObFoGsHzp8H8vPFJgIhIUpXQxVlyohufj5w8KA45oguERGReTDo2gF5MVN4OOdu2hM56J46BeTmln3uX38BWi1Qpw5/mSEiIjIXBl07wPm59qlePaB2bTH1RP4alqbo/FyNxvK1EREROQIGXTvAoGufNJqKT1/gQjQiIiLzY9C1Awy69quiQZcL0YiIiMyPQdcOyEE3MlLZOsh0Femlm5UFHD0qjjmiS0REZD4MujYuP19sIwtwRNceVaTF2N69Yge1oCDAz886dRERETkCBl0bl54uQpCPj1jcRPZFHtFNTxe/tJSE83OJiIgsg0HXxhWdn8vV+PbH3x+oUQPQ6Qo3/bgXgy4REZFlMOjaOC5Es28V6bzAhWhERESWwaBr4xh07V9ZQffaNeDsWXHcurX1aiIiInIEDLo2jh0X7F9ZQVeethAZKaY4EBERkfkw6No4eftfjujar7JajHF+LhERkeUw6NqwmzfFW9sAEBGhbC1UeXLQPX4c0GqNP8agS0REZDkMujZMXqXv7w94eytbC1VeUBBQvTpQUCDajMkkqTDociEaERGR+THo2jAuRFOH0jovnDsHXL0KuLgALVooUhoREZGqMejaMAZd9Sgp6O7ZIxojN2sGeHgoUBQREZHKMejaMHZcUI+ygi7n5xIREVkGg64NY8cF9Sgp6O7dy6BLRERkSQy6NkqSOHVBTeSge/SoWJSm1zPoEhERWRqDro26eBHIyQGcnYFGjZSuhqoqOBioVg24exc4dQrIyKiOrCwNPD2Bpk2Vro6IiEidGHRtlDya26gR4OambC1UdU5OQHS0OE5L0yA9vRYAoFUr0XWBiIiIzI9B10Zx2oL6yNMX0tI0OHGiJgBOWyAiIrIkBl0bxY4L6lM06Kan1wTAoEtERGRJDLo2ih0X1EcOun/9pcHp0zUAMOgSERFZEmcH2ihOXVCfwhZjGgDOqFlTQni4RtGaiIiI1IwjujZIqxUr8wEGXTW5d2FhaKgEvV65eoiIiNSOQdcGnT4N6HSiHVVAgNLVkLl8/73ojyzbt88JISHAunWKlURERKRqDLo2SJ62EBEh2lKR/Vu3DnjqKTFaX1RGhrifYZeIiMj8GKNsEDsuqItOB4webTyaK5PvGzNGnEdERETmw6Brg9hxQV1+/x24cKH0j0sScP68OI+IiIjMh0HXBrHjgrpcumTe84iIiKhiGHRtEIOuuvj7m/c8IiIiqhgGXRtz5w5w8aI4ZtBVh44dgQYNAE0pLXM1GiAoSJxHRERE5sOga2NOnBC3desCtWopWwuZh7MzsHChOL437Mp/X7BAnEdERETmw6BrYzhtQZ3i44FvvwUCA43vb9BA3B8fr0xdREREasYtgG0MOy6oV3w80Ls3sHlzAX7++QB69myBzp1dOJJLRERkIQy6NoYjuurm7AzExkrIzs5AbGwMQy4REZEFceqCjWHQJSIiIjIPBl0bIkkMukRERETmwqBrQ65dA27fFivxw8OVroaIiIjIvjHo2hB5IVpwMODhoWwtRERERPaOQdeGcNoCERERkfkw6NoQBl0iIiIi82HQtSEMukRERETmw6BrQxh0iYiIiMyHQddG6HRAero4joxUthYiIiIiNWDQtRFnzwJ37wLu7kBQkNLVEBEREdk/Bl0bceKEBoDon8ttYYmIiIiqrlJBd/HixQgJCYGHhwceeOAB7Nq1q9RztVot3njjDYSFhcHDwwMxMTFISkoyOicrKwtjxoxBcHAwPD090b59e+zevdvonBdffBEajcboz6OPPmp0zvHjx9G7d2/4+vrCx8cHHTp0wObNmytziVYnB13OzyUiIiIyD5OD7ldffYWEhARMnz4d+/btQ0xMDHr06IGrV6+WeP6UKVOwbNkyLFq0CEeOHMGwYcPQt29f7N+/33DOkCFDkJycjMTERBw6dAjdu3dH165dkZGRYfRcjz76KC5dumT4s2bNGqOPP/744ygoKMBvv/2GvXv3IiYmBo8//jguX75s6mVa3YkT4pZBl4iIiMg8TA668+fPx9ChQzFo0CA0adIES5cuRbVq1bBy5coSz09MTMSkSZMQFxeH0NBQDB8+HHFxcZg3bx4AIDc3F2vXrsXcuXPRqVMnhIeHY8aMGQgPD8eSJUuMnsvd3R3169c3/KlVq5bhY9evX8eJEycwYcIENG/eHBEREXjnnXeQk5ODv//+29TLtDqO6BIRERGZl0lB9+7du9i7dy+6du1a+AROTujatSt27NhR4mPy8/Phcc9+tp6enti2bRsAoKCgADqdrsxzZFu2bEG9evUQGRmJ4cOH48aNG4aP1alTB5GRkfjss8+QnZ2NgoICLFu2DPXq1UPr1q1NuUxFyEGXHReIiIiIzMPFlJOvX78OnU4HPz8/o/v9/Pxw9OjREh/To0cPzJ8/H506dUJYWBhSUlKwbt066HQ6AIC3tzfatWuHWbNmITo6Gn5+flizZg127NiB8PBww/M8+uijiI+PR6NGjXDy5ElMmjQJPXv2xI4dO+Ds7AyNRoNff/0Vffr0gbe3N5ycnFCvXj0kJSUZjfwWlZ+fj/z8fMPfMzMzAYh5xVqt1pRPTaVptVrk5zvh3Dnx90aNtLDSSytK/vxa6/NsSxz12h31ugFee9FbR+Go1w3w2oveOgprX7cpr6ORJEmq6MkXL15EYGAgtm/fjnbt2hnuHz9+PFJTU7Fz585ij7l27RqGDh2KjRs3QqPRICwsDF27dsXKlSuRm5sLADh58iQGDx6MrVu3wtnZGa1atULjxo2xd+9epKWllVjLqVOnEBYWhl9//RWPPPIIJElCnz59oNVqMXnyZHh6emLFihXYsGEDdu/eDX9//2LPMWPGDMycObPY/atXr0a1atUq+mmpsjNnvDFmTBd4ed3F55//DI3Gai9NREREZFdycnLw3HPP4fbt2/Dx8SnzXJNGdH19feHs7IwrV64Y3X/lyhXUr1+/xMfUrVsX69evR15eHm7cuIGAgABMmDABoaGhhnPCwsKQmpqK7OxsZGZmwt/fH/379zc6516hoaHw9fVFeno6HnnkEfz222/44YcfcPPmTcNFf/jhh0hOTsann36KCRMmFHuOiRMnIiEhwfD3zMxMBAUFoXv37uV+4sxFq9Vi5kwxh7hJExc89licVV5XaVqtFsnJyejWrRtcXV2VLseqHPXaHfW6AV67I167o143wGt3xGu39nXL78BXhElB183NDa1bt0ZKSgr69OkDANDr9UhJScHIkSPLfKyHhwcCAwOh1Wqxdu1a9OvXr9g5Xl5e8PLyws2bN7Fp0ybMnTu31Oe7cOECbty4YRipzcnJASDmDBfl5OQEvV5f4nO4u7vD3d292P2urq5W/Qa9eLE6ACAy0gmuro7V2tjan2tb4qjX7qjXDfDaHfHaHfW6AV67I167ta7blNcwOVUlJCTgo48+wqeffoq0tDQMHz4c2dnZGDRoEABgwIABmDhxouH8nTt3Yt26dTh16hR+//13PProo9Dr9Rg/frzhnE2bNiEpKQmnT59GcnIyOnfujKioKMNz3rlzB+PGjcOff/6JM2fOICUlBb1790Z4eDh69OgBAGjXrh1q1aqFgQMH4q+//sLx48cxbtw4nD59Go899pipl2lVctBlxwUiIiIi8zFpRBcA+vfvj2vXrmHatGm4fPkyWrRogaSkJMMCtXPnzhmNqubl5WHKlCk4deoUqlevjri4OCQmJqJmzZqGc27fvo2JEyfiwoULqF27Np588km89dZbhsTu7OyMgwcP4tNPP8WtW7cQEBCA7t27Y9asWYYRWV9fXyQlJWHy5Mno0qULtFotmjZtiu+//x4xMTFV+RxZ3MWLXgDYcYGIiIjInEwOugAwcuTIUqcqbNmyxejvsbGxOHLkSJnP169fvxKnMsg8PT2xadOmcutq06ZNhc6zNRkZHNElIiIiMjfHmhBqg27cALKyxKh0kW5qRERERFRFDLoKS08XvcQCAyVUr65wMUREREQqwqCrsOPHxW1ERIXbGRMRERFRBTDoKkze+rdxYwZdIiIiInNi0FWYHHQjIhQuhIiIiEhlGHQVpNMB+/aJoJuXJ/5ORERERObBoKuQdeuAkBDg9GkRdKdOdUZIiLifiIiIiKqOQVcB69YBTz0FXLhgfH9GhrifYZeIiIio6hh0rUynA0aPBqQS1p7J940Zw2kMRERERFXFoGtlv/9efCS3KEkCzp8X5xERERFR5THoWtmlS+Y9j4iIiIhKxqBrZf7+5j2PiIiIiErGoGtlHTsCDRoAGk3JH9dogKAgcR4RERERVR6DrpU5OwMLF4rje8Ou/PcFC8R5RERERFR5DLoKiI8Hvv0WCAw0vr9BA3F/fLwydRERERGpiYvSBTiq+Higd29g8+YC/PzzAfTs2QKdO7twJJeIiIjITBh0FeTsDMTGSsjOzkBsbAxDLhEREZEZceoCEREREakSgy4RERERqRKDLhERERGpEoMuEREREakSgy4RERERqRKDLhERERGpEoMuEREREakSgy4RERERqRKDLhERERGpEoMuEREREakSgy4RERERqRKDLhERERGpEoMuEREREamSi9IF2BJJkgAAmZmZVntNrVaLnJwcZGZmwtXV1WqvqzRHvW7Aca/dUa8b4LU74rU76nUDvHZHvHZrX7ec0+TcVhYG3SKysrIAAEFBQQpXQkRERERlycrKQo0aNco8RyNVJA47CL1ej4sXL8Lb2xsajcYqr5mZmYmgoCCcP38ePj4+VnlNW+Co1w047rU76nUDvHZHvHZHvW6A1+6I127t65YkCVlZWQgICICTU9mzcDmiW4STkxMaNGigyGv7+Pg41D8KmaNeN+C41+6o1w3w2h3x2h31ugFeuyNeuzWvu7yRXBkXoxERERGRKjHoEhEREZEqMegqzN3dHdOnT4e7u7vSpViVo1434LjX7qjXDfDaHfHaHfW6AV67I167LV83F6MRERERkSpxRJeIiIiIVIlBl4iIiIhUiUGXiIiIiFSJQZeIiIiIVIlBVyFbt25Fr169EBAQAI1Gg/Xr1ytdklXMnj0bbdu2hbe3N+rVq4c+ffrg2LFjSpdlcUuWLEHz5s0NzbTbtWuHn3/+WemyFPHOO+9Ao9FgzJgxSpdicTNmzIBGozH6ExUVpXRZVpGRkYEXXngBderUgaenJ5o1a4Y9e/YoXZbFhYSEFPuaazQajBgxQunSLEqn02Hq1Klo1KgRPD09ERYWhlmzZsFR1rtnZWVhzJgxCA4OhqenJ9q3b4/du3crXZbZlZddJEnCtGnT4O/vD09PT3Tt2hUnTpxQpth/MegqJDs7GzExMVi8eLHSpVhVamoqRowYgT///BPJycnQarXo3r07srOzlS7Noho0aIB33nkHe/fuxZ49e9ClSxf07t0bhw8fVro0q9q9ezeWLVuG5s2bK12K1TRt2hSXLl0y/Nm2bZvSJVnczZs38dBDD8HV1RU///wzjhw5gnnz5qFWrVpKl2Zxu3fvNvp6JycnAwCefvpphSuzrDlz5mDJkiX4v//7P6SlpWHOnDmYO3cuFi1apHRpVjFkyBAkJycjMTERhw4dQvfu3dG1a1dkZGQoXZpZlZdd5s6diw8++ABLly7Fzp074eXlhR49eiAvL8/KlRYhkeIASN99953SZSji6tWrEgApNTVV6VKsrlatWtKKFSuULsNqsrKypIiICCk5OVmKjY2VRo8erXRJFjd9+nQpJiZG6TKs7vXXX5c6dOigdBk2YfTo0VJYWJik1+uVLsWiHnvsMWnw4MFG98XHx0vPP/+8QhVZT05OjuTs7Cz98MMPRve3atVKmjx5skJVWd692UWv10v169eX3n33XcN9t27dktzd3aU1a9YoUKHAEV1S1O3btwEAtWvXVrgS69HpdPjyyy+RnZ2Ndu3aKV2O1YwYMQKPPfYYunbtqnQpVnXixAkEBAQgNDQUzz//PM6dO6d0SRa3YcMGtGnTBk8//TTq1auHli1b4qOPPlK6LKu7e/cuPv/8cwwePBgajUbpciyqffv2SElJwfHjxwEAf/31F7Zt24aePXsqXJnlFRQUQKfTwcPDw+h+T09Ph3gHR3b69GlcvnzZ6P/4GjVq4IEHHsCOHTsUq8tFsVcmh6fX6zFmzBg89NBDuO+++5Qux+IOHTqEdu3aIS8vD9WrV8d3332HJk2aKF2WVXz55ZfYt2+fKuesleWBBx7AJ598gsjISFy6dAkzZ85Ex44d8ffff8Pb21vp8izm1KlTWLJkCRISEjBp0iTs3r0bo0aNgpubGwYOHKh0eVazfv163Lp1Cy+++KLSpVjchAkTkJmZiaioKDg7O0On0+Gtt97C888/r3RpFuft7Y127dph1qxZiI6Ohp+fH9asWYMdO3YgPDxc6fKs5vLlywAAPz8/o/v9/PwMH1MCgy4pZsSIEfj7778d5jfeyMhIHDhwALdv38a3336LgQMHIjU1VfVh9/z58xg9ejSSk5OLjXioXdHRrObNm+OBBx5AcHAwvv76a7z00ksKVmZZer0ebdq0wdtvvw0AaNmyJf7++28sXbrUoYLuxx9/jJ49eyIgIEDpUizu66+/xhdffIHVq1ejadOmOHDgAMaMGYOAgACH+JonJiZi8ODBCAwMhLOzM1q1aoVnn30We/fuVbo0h8epC6SIkSNH4ocffsDmzZvRoEEDpcuxCjc3N4SHh6N169aYPXs2YmJisHDhQqXLsri9e/fi6tWraNWqFVxcXODi4oLU1FR88MEHcHFxgU6nU7pEq6lZsyYaN26M9PR0pUuxKH9//2K/wEVHRzvEtA3Z2bNn8euvv2LIkCFKl2IV48aNw4QJE/DMM8+gWbNm+M9//oOxY8di9uzZSpdmFWFhYUhNTcWdO3dw/vx57Nq1C1qtFqGhoUqXZjX169cHAFy5csXo/itXrhg+pgQGXbIqSZIwcuRIfPfdd/jtt9/QqFEjpUtSjF6vR35+vtJlWNwjjzyCQ4cO4cCBA4Y/bdq0wfPPP48DBw7A2dlZ6RKt5s6dOzh58iT8/f2VLsWiHnrooWJtA48fP47g4GCFKrK+VatWoV69enjssceULsUqcnJy4ORkHCmcnZ2h1+sVqkgZXl5e8Pf3x82bN7Fp0yb07t1b6ZKsplGjRqhfvz5SUlIM92VmZmLnzp2Krkfh1AWF3Llzx2hU5/Tp0zhw4ABq166Nhg0bKliZZY0YMQKrV6/G999/D29vb8O8nRo1asDT01Ph6ixn4sSJ6NmzJxo2bIisrCysXr0aW7ZswaZNm5QuzeK8vb2LzcH28vJCnTp1VD83+7XXXkOvXr0QHByMixcvYvr06XB2dsazzz6rdGkWNXbsWLRv3x5vv/02+vXrh127dmH58uVYvny50qVZhV6vx6pVqzBw4EC4uDjGj9levXrhrbfeQsOGDdG0aVPs378f8+fPx+DBg5UuzSo2bdoESZIQGRmJ9PR0jBs3DlFRURg0aJDSpZlVedllzJgxePPNNxEREYFGjRph6tSpCAgIQJ8+fZQrWrF+Dw5u8+bNEoBifwYOHKh0aRZV0jUDkFatWqV0aRY1ePBgKTg4WHJzc5Pq1q0rPfLII9Ivv/yidFmKcZT2Yv3795f8/f0lNzc3KTAwUOrfv7+Unp6udFlWsXHjRum+++6T3N3dpaioKGn58uVKl2Q1mzZtkgBIx44dU7oUq8nMzJRGjx4tNWzYUPLw8JBCQ0OlyZMnS/n5+UqXZhVfffWVFBoaKrm5uUn169eXRowYId26dUvpssyuvOyi1+ulqVOnSn5+fpK7u7v0yCOPKP7vQCNJDrJtCRERERE5FM7RJSIiIiJVYtAlIiIiIlVi0CUiIiIiVWLQJSIiIiJVYtAlIiIiIlVi0CUiIiIiVWLQJSIiIiJVYtAlIiIiIlVi0CUiIiIiVWLQJSIiIiJVYtAlIiIiIlVi0CUiIiIiVfp/6tEZhUCaQukAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "tRdJTM2UFs0p"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}