activity_tracker/activity_tracker_motionSense_gsoc20.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "activity_tracker_motionSense_gsoc20",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8yrf769yy5l9",
        "colab_type": "text"
      },
      "source": [
        "# Activity Tracker Buffer Model Generation for Arduino library : Arduino_TensorFlowLite\n",
        "\n",
        "**Author**: Prashant Dandriyal\n",
        "\n",
        "**Subject**: In support of proposal for GSoC 2020 to Arduino.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3NAmBsLpGgdp",
        "colab_type": "text"
      },
      "source": [
        "**Model**: Motion Sense Dataset from the link: https://github.com/mmalekzadeh/motion-sense/tree/master/data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Pge9gQs_zDKZ",
        "colab_type": "code",
        "outputId": "34fa5564-e0a1-4983-f1a6-043800e34905",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "#Get DeviceMotion data\n",
        "!wget \"https://github.com/mmalekzadeh/motion-sense/raw/master/data/A_DeviceMotion_data.zip\"\n",
        "!unzip \"/content/A_DeviceMotion_data.zip\"\n",
        "!wget \"https://raw.githubusercontent.com/mmalekzadeh/motion-sense/master/data/data_subjects_info.csv\""
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-04-04 10:54:39--  https://github.com/mmalekzadeh/motion-sense/raw/master/data/A_DeviceMotion_data.zip\n",
            "Resolving github.com (github.com)... 140.82.113.3\n",
            "Connecting to github.com (github.com)|140.82.113.3|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://raw.githubusercontent.com/mmalekzadeh/motion-sense/master/data/A_DeviceMotion_data.zip [following]\n",
            "--2020-04-04 10:54:40--  https://raw.githubusercontent.com/mmalekzadeh/motion-sense/master/data/A_DeviceMotion_data.zip\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 151.101.0.133, 151.101.64.133, 151.101.128.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.0.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 73696396 (70M) [application/zip]\n",
            "Saving to: ‘A_DeviceMotion_data.zip’\n",
            "\n",
            "A_DeviceMotion_data 100%[===================>]  70.28M  70.5MB/s    in 1.0s    \n",
            "\n",
            "2020-04-04 10:54:45 (70.5 MB/s) - ‘A_DeviceMotion_data.zip’ saved [73696396/73696396]\n",
            "\n",
            "Archive:  /content/A_DeviceMotion_data.zip\n",
            "   creating: A_DeviceMotion_data/\n",
            "   creating: A_DeviceMotion_data/dws_11/\n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_11/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/ups_12/\n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_12/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/wlk_7/\n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_13.csv  \n",
            "   creating: __MACOSX/\n",
            "   creating: __MACOSX/A_DeviceMotion_data/\n",
            "   creating: __MACOSX/A_DeviceMotion_data/wlk_7/\n",
            "  inflating: __MACOSX/A_DeviceMotion_data/wlk_7/._sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_7/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/std_14/\n",
            "  inflating: A_DeviceMotion_data/std_14/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/std_14/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/wlk_15/\n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_15/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/wlk_8/\n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/wlk_8/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/dws_2/\n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_2/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/sit_13/\n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_13/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/jog_9/\n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_4.csv  \n",
            "   creating: __MACOSX/A_DeviceMotion_data/jog_9/\n",
            "  inflating: __MACOSX/A_DeviceMotion_data/jog_9/._sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_5.csv  \n",
            "  inflating: __MACOSX/A_DeviceMotion_data/jog_9/._sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_9/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/ups_3/\n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_3/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/ups_4/\n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/ups_4/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/jog_16/\n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/jog_16/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/dws_1/\n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_1.csv  \n",
            "   creating: __MACOSX/A_DeviceMotion_data/dws_1/\n",
            "  inflating: __MACOSX/A_DeviceMotion_data/dws_1/._sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/dws_1/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/sit_5/\n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/sit_5/sub_23.csv  \n",
            "   creating: A_DeviceMotion_data/std_6/\n",
            "  inflating: A_DeviceMotion_data/std_6/sub_12.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_1.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_13.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_11.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_2.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_3.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_10.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_14.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_7.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_6.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_15.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_17.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_4.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_5.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_16.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_8.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_9.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_18.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_24.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_19.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_21.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_20.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_22.csv  \n",
            "  inflating: A_DeviceMotion_data/std_6/sub_23.csv  \n",
            "--2020-04-04 10:54:55--  https://raw.githubusercontent.com/mmalekzadeh/motion-sense/master/data/data_subjects_info.csv\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 151.101.0.133, 151.101.64.133, 151.101.128.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.0.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 385 [text/plain]\n",
            "Saving to: ‘data_subjects_info.csv’\n",
            "\n",
            "data_subjects_info. 100%[===================>]     385  --.-KB/s    in 0s      \n",
            "\n",
            "2020-04-04 10:54:55 (18.5 MB/s) - ‘data_subjects_info.csv’ saved [385/385]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_rKz5SDOykcf",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "def get_ds_infos():\n",
        "    \"\"\"\n",
        "    Read the file includes data subject information.\n",
        "    \n",
        "    Data Columns:\n",
        "    0: code [1-24]\n",
        "    1: weight [kg]\n",
        "    2: height [cm]\n",
        "    3: age [years]\n",
        "    4: gender [0:Female, 1:Male]\n",
        "    \n",
        "    Returns:\n",
        "        A pandas DataFrame that contains inforamtion about data subjects' attributes \n",
        "    \"\"\" \n",
        "\n",
        "    dss = pd.read_csv(\"data_subjects_info.csv\")\n",
        "    print(\"[INFO] -- Data subjects' information is imported.\")\n",
        "    \n",
        "    return dss\n",
        "\n",
        "def set_data_types(data_types=[\"userAcceleration\"]):\n",
        "    \"\"\"\n",
        "    Select the sensors and the mode to shape the final dataset.\n",
        "    \n",
        "    Args:\n",
        "        data_types: A list of sensor data type from this list: [attitude, gravity, rotationRate, userAcceleration] \n",
        "\n",
        "    Returns:\n",
        "        It returns a list of columns to use for creating time-series from files.\n",
        "    \"\"\"\n",
        "    dt_list = []\n",
        "    for t in data_types:\n",
        "        if t != \"attitude\":\n",
        "            dt_list.append([t+\".x\",t+\".y\",t+\".z\"])\n",
        "        else:\n",
        "            dt_list.append([t+\".roll\", t+\".pitch\", t+\".yaw\"])\n",
        "\n",
        "    return dt_list\n",
        "\n",
        "\n",
        "def creat_time_series(dt_list, act_labels, trial_codes, mode=\"mag\", labeled=True):\n",
        "    \"\"\" data_subjects_info\n",
        "    Args:\n",
        "        dt_list: A list of columns that shows the type of data we want.\n",
        "        act_labels: list of activites\n",
        "        trial_codes: list of trials\n",
        "        mode: It can be \"raw\" which means you want raw data\n",
        "        for every dimention of each data type,\n",
        "        [attitude(roll, pitch, yaw); gravity(x, y, z); rotationRate(x, y, z); userAcceleration(x,y,z)].\n",
        "        or it can be \"mag\" which means you only want the magnitude for each data type: (x^2+y^2+z^2)^(1/2)\n",
        "        labeled: True, if we want a labeld dataset. False, if we only want sensor values.\n",
        "\n",
        "    Returns:\n",
        "        It returns a time-series of sensor data.\n",
        "    \n",
        "    \"\"\"\n",
        "    num_data_cols = len(dt_list) if mode == \"mag\" else len(dt_list*3)\n",
        "\n",
        "    if labeled:\n",
        "        dataset = np.zeros((0,num_data_cols+7)) # \"7\" --> [act, code, weight, height, age, gender, trial] \n",
        "    else:\n",
        "        dataset = np.zeros((0,num_data_cols))\n",
        "        \n",
        "    ds_list = get_ds_infos()\n",
        "    \n",
        "    print(\"[INFO] -- Creating Time-Series\")\n",
        "    for sub_id in ds_list[\"code\"]:\n",
        "        for act_id, act in enumerate(act_labels):\n",
        "            for trial in trial_codes[act_id]:\n",
        "                fname = 'A_DeviceMotion_data/'+act+'_'+str(trial)+'/sub_'+str(int(sub_id))+'.csv'\n",
        "                raw_data = pd.read_csv(fname)\n",
        "                raw_data = raw_data.drop(['Unnamed: 0'], axis=1)\n",
        "                vals = np.zeros((len(raw_data), num_data_cols))\n",
        "                for x_id, axes in enumerate(dt_list):\n",
        "                    if mode == \"mag\":\n",
        "                        vals[:,x_id] = (raw_data[axes]**2).sum(axis=1)**0.5        \n",
        "                    else:\n",
        "                        vals[:,x_id*3:(x_id+1)*3] = raw_data[axes].values\n",
        "                    vals = vals[:,:num_data_cols]\n",
        "                if labeled:\n",
        "                    lbls = np.array([[act_id,\n",
        "                            sub_id-1,\n",
        "                            ds_list[\"weight\"][sub_id-1],\n",
        "                            ds_list[\"height\"][sub_id-1],\n",
        "                            ds_list[\"age\"][sub_id-1],\n",
        "                            ds_list[\"gender\"][sub_id-1],\n",
        "                            trial          \n",
        "                           ]]*len(raw_data))\n",
        "                    vals = np.concatenate((vals, lbls), axis=1)\n",
        "                dataset = np.append(dataset,vals, axis=0)\n",
        "    cols = []\n",
        "    for axes in dt_list:\n",
        "        if mode == \"raw\":\n",
        "            cols += axes\n",
        "        else:\n",
        "            cols += [str(axes[0][:-2])]\n",
        "            \n",
        "    if labeled:\n",
        "        cols += [\"act\", \"id\", \"weight\", \"height\", \"age\", \"gender\", \"trial\"]\n",
        "    np_dataset = dataset\n",
        "    dataset = pd.DataFrame(data=dataset, columns=cols)\n",
        "    return dataset, np_dataset\n",
        "\n",
        "\n",
        "ACT_LABELS = [\"dws\",\"ups\", \"wlk\", \"jog\", \"std\", \"sit\"]\n",
        "TRIAL_CODES = {\n",
        "    ACT_LABELS[0]:[1,2,11],\n",
        "    ACT_LABELS[1]:[3,4,12],\n",
        "    ACT_LABELS[2]:[7,8,15],\n",
        "    ACT_LABELS[3]:[9,16],\n",
        "    ACT_LABELS[4]:[6,14],\n",
        "    ACT_LABELS[5]:[5,13]\n",
        "}\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZiS1opvQPZjw",
        "colab_type": "code",
        "outputId": "b3552584-8a0d-4ee9-bd62-041d5ceceedd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        }
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "def get_ds_infos():\n",
        "    ## 0:Code, 1:Weight, 2:Height, 3:Age, 4:Gender\n",
        "    dss = np.genfromtxt(\"data_subjects_info.csv\",delimiter=',')\n",
        "    dss = dss[1:]\n",
        "    print(\"----> Data subjects information is imported.\")\n",
        "    return dss\n",
        "##____________\n",
        "\n",
        "def creat_time_series(num_features, num_act_labels, num_gen_labels, label_codes, trial_codes):\n",
        "    dataset_columns = num_features+num_act_labels+num_gen_labels\n",
        "    ds_list = get_ds_infos()\n",
        "    train_data = np.zeros((0,dataset_columns))\n",
        "    test_data = np.zeros((0,dataset_columns))\n",
        "    for i, sub_id in enumerate(ds_list[:,0]):\n",
        "        for j, act in enumerate(label_codes):\n",
        "            for trial in trial_codes[act]:\n",
        "                fname = 'A_DeviceMotion_data/'+act+'_'+str(trial)+'/sub_'+str(int(sub_id))+'.csv'\n",
        "                raw_data = pd.read_csv(fname)\n",
        "                raw_data = raw_data.drop(['Unnamed: 0'], axis=1)\n",
        "                unlabel_data = raw_data.values\n",
        "                label_data = np.zeros((len(unlabel_data), dataset_columns))\n",
        "                label_data[:,:-(num_act_labels + num_gen_labels)] = unlabel_data\n",
        "                label_data[:,label_codes[act]] = 1\n",
        "                label_data[:,-(num_gen_labels)] = int(ds_list[i,4])\n",
        "                ## We consider long trials as training dataset and short trials as test dataset\n",
        "                if trial > 10:\n",
        "                    test_data = np.append(test_data, label_data, axis = 0)\n",
        "                else:    \n",
        "                    train_data = np.append(train_data, label_data, axis = 0)\n",
        "    return train_data , test_data\n",
        "#________________________________\n",
        "\n",
        "\n",
        "print(\"--> Start...\")\n",
        "## Here we set parameter to build labeld time-series from dataset of \"(A)DeviceMotion_data\"\n",
        "num_features = 12 # attitude(roll, pitch, yaw); gravity(x, y, z); rotationRate(x, y, z); userAcceleration(x,y,z)\n",
        "num_act_labels = 4 # dws, ups, wlk, jog\n",
        "num_gen_labels = 1 # 0/1(female/male)\n",
        "label_codes = {\"dws\":num_features, \"ups\":num_features+1, \"wlk\":num_features+2, \"jog\":num_features+3}\n",
        "trial_codes = {\"dws\":[1,2,11], \"ups\":[3,4,12], \"wlk\":[7,8,15], \"jog\":[9,16]}    \n",
        "## Calling 'creat_time_series()' to build time-series\n",
        "print(\"--> Building Training and Test Datasets...\")\n",
        "train_ts, test_ts = creat_time_series(num_features, num_act_labels, num_gen_labels, label_codes, trial_codes)\n",
        "print(\"--> Shape of Training Time-Seires:\", train_ts.shape)\n",
        "print(\"--> Shape of Test Time-Series:\", test_ts.shape)"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--> Start...\n",
            "--> Building Training and Test Datasets...\n",
            "----> Data subjects information is imported.\n",
            "--> Shape of Training Time-Seires: (621973, 17)\n",
            "--> Shape of Test Time-Series: (145687, 17)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WY8pjoQq_TqM",
        "colab_type": "text"
      },
      "source": [
        "The dataset contains several cases like : \"downStairs\",\"upStairs\", \"walk\", \"jog\", \"stand\" and \"sit\" but we only extract the last 4 for our analysis.\n",
        "\n",
        "Also, out of the 4 parameters: attitude(roll, pitch, yaw); gravity(x, y, z); rotationRate(x, y, z); userAcceleration(x,y,z), we only need gravity, userAcceleration and rotationRate.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3y61eFK2GZWq",
        "colab_type": "text"
      },
      "source": [
        "## Removing the redundant columns: "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jdkOIwXyAwC3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "dataset = dataset.drop([\"id\", \"weight\", \"height\", \"age\", \"gender\", \"trial\"], axis = 1)\n",
        "dataset.head()\n",
        "\n",
        "#Doing same for the numpy version\n",
        "#i.e., remove last 6 columns from the data\n",
        "np_dataset = np.delete(np_dataset, np.s_[-6:], axis = 1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TQ5FOs_ZHxDV",
        "colab_type": "text"
      },
      "source": [
        "## Visualizing data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d8lfkJSyWtNX",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 71
        },
        "outputId": "b43eaba0-db00-4848-c31f-ee0a94c01958"
      },
      "source": [
        "# Save previous df to csv\n",
        "dataset.to_csv (r'data.csv', index = False, header=True)\n",
        "\n",
        "# making data frame for different activities, from csv file \n",
        "data = pd.read_csv(\"data.csv\", index_col =\"act\") \n",
        "\n",
        "# Filter rows by loc method \n",
        "# Selected activites: ['wlk', 'jog', 'std', 'sit'] -> 0, 1, 2, 3\n",
        "df_wlk = data.loc[0] \n",
        "df_jog = data.loc[1] \n",
        "df_std = data.loc[2]\n",
        "df_sit = data.loc[3]   "
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/numpy/lib/arraysetops.py:569: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
            "  mask |= (ar1 == a)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aGFy0akoHAFd",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "1e38ce81-8efb-4a5c-dd1f-09c173493a80"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "def vis(df, name):\n",
        "  index = range(1, len(df[\"gravity.x\"]) + 1)\n",
        "\n",
        "  plt.rcParams[\"figure.figsize\"] = (10,5)\n",
        "\n",
        "  plt.plot(index, df['rotationRate.x'], 'g.', label='x', linestyle='solid', marker=',')\n",
        "  plt.plot(index, df['rotationRate.y'], 'b.', label='y', linestyle='solid', marker=',')\n",
        "  plt.plot(index, df['rotationRate.z'], 'r.', label='z', linestyle='solid', marker=',')\n",
        "  plt.title(name+\"Gyroscope\")\n",
        "  plt.xlabel(\"Samples\")\n",
        "  plt.ylabel(\"Gyroscope (deg/sec)\")\n",
        "  plt.legend()\n",
        "  plt.show()\n",
        "\n",
        "  plt.plot(index, df['userAcceleration.x'], 'g.', label='x', linestyle='solid', marker=',')\n",
        "  plt.plot(index, df['userAcceleration.y'], 'b.', label='y', linestyle='solid', marker=',')\n",
        "  plt.plot(index, df['userAcceleration.z'], 'r.', label='z', linestyle='solid', marker=',')\n",
        "  plt.title(name+\"Acceleration\")\n",
        "  plt.xlabel(\"Samples\")\n",
        "  plt.ylabel(\"Acceleration (G)\")\n",
        "  plt.legend()\n",
        "  plt.show()\n",
        "\n",
        "\n",
        "  plt.plot(index, df['gravity.x'], 'g.', label='x', linestyle='solid', marker=',')\n",
        "  plt.plot(index, df['gravity.y'], 'b.', label='y', linestyle='solid', marker=',')\n",
        "  plt.plot(index, df['gravity.z'], 'r.', label='z', linestyle='solid', marker=',')\n",
        "  plt.title(name+\"Gravity\")\n",
        "  plt.xlabel(\"Samples\")\n",
        "  plt.ylabel(\"gravity (G)\")\n",
        "  plt.legend()\n",
        "  plt.show()\n",
        "\n",
        "#Plot for walk\n",
        "vis(df_wlk, \"Walk\")\n",
        "#Plot for sit and see the difference\n",
        "vis(df_sit, \"Sit\")"
      ],
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmsAAAFNCAYAAABfUShSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmIAAAFNCAYAAACjRAOYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCn9rUxn5cAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnEAAAFNCAYAAABv3TlzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmoAAAFNCAYAAACwk0NsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0\ndHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdd5gUVdbA4d+ZGWCIEpUoYELBgAgG\nBBXMimFdw6qI4XMNa0JRMS+ioq5ZMSdEjJjWFUVEBDMCSpIMktOQ48DM9Pn+qE7Vsaane7qB8z5P\nP92Vbt2urq46deveuqKqGGOMMcaY3JOX7QwYY4wxxpjYLFAzxhhjjMlRFqgZY4wxxuQoC9SMMcYY\nY3KUBWrGGGOMMTnKAjVjjDHGmBxlgZoxJutE5GIRGZHtfBhjTK6xQM0YU2lEpIuI/Cwi60VkjYj8\nJCKdVPUdVT0pbD4VkX0ilq0tIk+KyHwR2SwiC0XkIxE5ovK/iTHGVI6CbGfAGLNrEJE6wBfAtcCH\nQFWgK7DNw7LVgFHAOqAHMB0oBE71v8aWMy8FqlpanmWMMSYbrETNGFNZ9gNQ1fdUtUxVt6rqCFWd\nLCKXiciPACLyvX/+SSKySUQuAC4BmgNnq+pU//KbVfUjVe3nX+55EXkifIUi8rmI3Oz/PF9E+orI\nZGCziBSIyJki8qeIrBOR0SJyQNiyfUVkiYhsFJGZInK8f3y+iNwlInP90yaISAv/tM4iMs5fYjhO\nRDqHpTdaRB4Wkd9EZIOI/FdE6odNP9Jf2rhORCaJyHFp3v7GmB2QBWrGmMoyCygTkbdE5FQRqRdr\nJlU9xv/xEFWtpaofACcAX6vq5gTpvwVcKCJ5ACLS0L/cu2HzXAicDtQF9gLeA3oDjYAvgf+JSFUR\naQNcD3RS1drAycB8fxq3+NM5DagDXAFs8Qddw4BngQbAk8AwEWkQtv5e/vmbAKX+eRGRZv5lHwTq\nA7cCH4tIowTf1xizC7BAzRhTKVR1A9AFUOBVoMhf4rWHh8UbAssDAyLS3l/ytEFEZvrT/w1YDxzv\nn+0fwGhVXRGWzrOqukhVtwIXAMNU9RtVLQEeB6oDnYEyoBrQVkSqqOp8VZ3rT+NK4B5VnamOSaq6\nGicAnK2qb6tqqaq+B8wAzghb/9v+EsHNwL3A+SKSD/QEvlTVL1XVp6rfAONxgkFjzC7MAjVjTKVR\n1emqepmqNgcOBJoCT3tYdDVOKVQgnYmqWhc4ByegCngLJ+jB//52RDqLwj43BRaEpenzT2+mqnNw\nStr6AStF5H0RaeqftQUwl2iu9PwWAM3irH8BUAUnCG0JnOcPPteJyDqcoLYJxphdmgVqxpisUNUZ\nwCCcgC2Zb4GTRKRmkvmGAGeJyCHAAcBnkasN+7wUJ0ACQEQEJwhb4s/fu6raxT+PAo/6Z10E7B1j\n3a70/PYMpOfXImJaCbDKn+bbqlo37FVTVR9J8n2NMTs5C9SMMZVCRPYXkT4i0tw/3AKnrtevMWZf\ngVOHLGAwsAz4VEQO9FfoLwQ6hi+kqouBcTglaR/7b3HG8yFwuogcLyJVgD44LVB/FpE2ItLd39q0\nGNgK+PzLvQY8ICL7iuNgfz20L4H9ROQif0OFC4C2OC1dA3qKSFsRqQH0Bz5S1TKcAPMMETk58N1E\n5LjAtjLG7LosUDPGVJaNwBHAWBHZjBOgTcUJkCL1A97y3wY8X1WLgW7ANJxK9xuAmUAn4PyIZd8C\nDiL6tqeLqs7EuT36HE6p1hnAGaq6Hed26iP+8cuB3YE7/Ys+iRPkjfDn43Wgur+eWg//91kN3A70\nUNVVYat9G6cUcTnO40Vu9OdlEXAWcBdQhFPCdht2jDZmlyeqmnwuY4zZQYjIMTglVC01hw5wIjIa\nGKKqr2U7L8aYHYddrRljdhr+W5g3Aa/lUpBmjDGpskDNGLNT8D+sdh1OS0kvLUmNMSbn2a1PY4wx\nxpgcZSVqxhhjjDE5ygI1Y4wxxpgcVZDtDJRHw4YNtVWrVtnOhjHGGGNMUhMmTFilqhXqs3eHCtRa\ntWrF+PHjs50NY4wxxpikRCSyW7lys1ufxhhjjDE5ygI1Y4wxxpgcZYGaMcYYY0yO2qHqqBljjDFm\n11JSUsLixYspLi7OdlbiKiwspHnz5lSpUiXtaVugZowxxpictXjxYmrXrk2rVq0QkWxnJ4qqsnr1\nahYvXkzr1q3Tnr7d+jTGGGNMziouLqZBgwY5GaQBiAgNGjTIWImfBWrGGGOMyWm5GqQFZDJ/FqgZ\nY4wxxuQoC9SMMcYYY3KUBWrGGLOLW78ecvzOkjFZM27cOA4++GCKi4vZvHkz7dq1Y+rUqZW2fmv1\naYwxu7gePZz37duhatXs5sWYXNOpUyfOPPNM7rnnHrZu3UrPnj058MADK239FqgZY8wubtw4593n\ny24+jEmm9/DeTFw+0dO8YxaM4diWxyadr33j9jx9ytMJ57nvvvvo1KkThYWFPPvss57Wny5269MY\nY4wxOx0vQZpXq1evZtOmTWzcuLHSH7xrJWrGGGOM2SEkK/nKlKuvvpoHHniAv/76i759+zJw4MBK\nW7cFasYYY4wxcQwePJgqVapw0UUXUVZWRufOnRk1ahTdu3evlPVboGaMMcYYE0evXr3o1asXAPn5\n+YwdO7ZS12911IwxxhhjclRWAzURqSsiH4nIDBGZLiJHZTM/xhhjjDG5JNu3Pp8BhqvquSJSFaiR\n5fwYY4wxxuSMrAVqIrIbcAxwGYCqbge2Zys/xhhjjDG5Jpu3PlsDRcCbIvKHiLwmIjWzmB9jjDHG\nmJySzUCtAOgAvKiqhwKbgTsiZxKRq0RkvIiMLyoqquw8GmOMMcZkTTYDtcXAYlUNtHP9CCdwc1HV\nV1S1o6p2bNSoUaVm0BhjjDEmm7IWqKnqcmCRiLTxjzoemJat/BhjjDHG5JpsP0ftBuAdEZkMtAcG\nZDk/xhhjjDFB9913H08/Heq66u677+aZZ56ptPVnNVBT1Yn+25oHq+rZqro2m/kxxhhjjAl3xRVX\nMHjwYAB8Ph/vv/8+PXv2rLT1Z/s5asYYY4wxnvTuDRMnept3zBg49tjk87VvD08n6Ou9VatWNGjQ\ngD/++IMVK1Zw6KGH0qBBA2+ZSAML1Iwxxhiz0/ESpHl15ZVXMmjQIJYvX84VV1yRvoQ9sEDNGGOM\nMTuERCVfmfS3v/2N++67j5KSEt59991KXbcFasYYY4wxCVStWpVu3bpRt25d8vPzK3XdFqgZY4wx\nxiTg8/n49ddfGTp0aKWvO9uP5zDGGGOMyVnTpk1jn3324fjjj2ffffet9PVbiZoxxhhjTBxt27Zl\n3rx5WVu/lagZY4wxxuQoC9SMMcYYY3KUBWrGGGOMMTnKAjVjjDHGmBxlgZoxxhhjTI6yQM0YY4wx\nJkdZoGaMMcYYE8dLL71E+/btad++Pa1bt6Zbt26Vun4L1Iwxxhhj4rjmmmuYOHEi48aNo3nz5txy\nyy2Vun574K0xxhhjdgy9e8PEid7mHTMGjj02+Xzt23vq7f2mm26ie/funHHGGd7WnyYWqBljjDFm\n5+MlSPNo0KBBLFiwgIEDB6YtTa8sUDPGGGPMjsFDyVe6TZgwgccff5wffviBvLzKrzFmddSMMcYY\nY+IYOHAga9asoVu3brRv354rr7yyUtdvJWrGGGOMMXG8+eabWV2/lagZY4wxxuQoC9SMMcYYY3KU\nBWrGGGOMMTnKAjVjjDHG5DRVzXYWEspk/ixQM8YYY0zOKiwsZPXq1TkbrKkqq1evprCwMCPpW6tP\nY4wxxuSs5s2bs3jxYoqKirKdlbgKCwtp3rx5RtK2QM0YY4wxOatKlSq0bt0629nIGrv1aYwxxhiT\noyxQM8YYY4zJURaoGWOMMcbkKAvUjDHGGGNylAVqxhhjjDE5ygI1Y4wxxpgcZYGaMcYYY0yOskDN\nGGOMMSZHWaBmjDHGGJOjLFAzxhhjjMlRFqgZY4wxxuQoC9SMMcYYY3KUBWrGGGOMMTnKAjVjjDHG\nmBxlgZoxxhhjTI7KeqAmIvki8oeIfJHtvBhjjDHG5JKsB2rATcD0bGfCGGOMMSbXZDVQE5HmwOnA\na9nMhzHGGGNMLsp2idrTwO2AL8v5MMYYY4zJOVkL1ESkB7BSVSckme8qERkvIuOLiooqKXfGGGOM\nMdmXzRK1o4EzRWQ+8D7QXUSGRM6kqq+oakdV7dioUaPKzqMxxhhjTNZkLVBT1TtVtbmqtgL+AYxS\n1Z7Zyo8xxhhjTK7Jdh01Y4wxxhgTR0G2MwCgqqOB0VnOhjHGGGNMTrESNWOMMcaYHGWBmjHGGGNM\njrJAzRhjjDEmR1mgZowxxhiToyxQM8YYY4zJURaoGWOMMcbkKAvUjDHGGGNyVMLnqIlIc5xeA7oC\nTYGtwFRgGPCVqlpn6sYYY4wxGRI3UBORN4FmwBfAo8BKoBDYDzgFuFtE7lDV7ysjo8YYY4wxu5pE\nJWpPqOrUGOOnAp+ISFVgz8xkyxhjjDHGxK2jFgjSRKSmiATnE5E8EamhqttVdU5lZNIYY4wxZlfk\npTHBt0CNsOEawMjMZMcYY4wxxgR4CdQKVXVTYMD/uUaC+Y0xxhhjTBp4CdQ2i0iHwICIHIbT+tMY\nY4wxxmRQwsdz+PUGhorIUkCAxsAFGc2VMcYYY4xJHqip6jgR2R9o4x81U1VLMpstY4wxxhiT9Nan\niNQA+gI3+VuCthKRHhnPmTHGGGNSUlQEItnOhUkHL3XU3gS2A0f5h5cAD2YsR8YYY4ypkI4ds50D\nky5eArW9VfU/QAmAqm7BqatmjDHGmBy0cGG2c2DSxUugtl1EqgMKICJ7A9symitjjDHGGOOp1ee/\ngeFACxF5BzgauCyTmTLGGFN5FB/ertuNMZXNS6vPb0Tkd+BInFueN6nqqoznzBhjTKXYXrYdKKS4\ntJhCCrOdHWNMGC+tPo8GilV1GFAXuEtEWmY8Z8YYY4wxuzgvZd0vAltE5BDgFmAuMDijuTLGGGOM\nMZ4CtVJVVeAs4HlVfR6ondlsGWOMMcYYL40JNorInUBP4BgRyQOqZDZbxhhjjDHGS4naBTiP4/g/\nVV0ONAcey2iujDHGGGNM/BI1Efka57EcX6nqk4HxqroQq6NmjDHGGJNxiUrULgXWAv1E5HcReVFE\nzhKRmpWUN2OMMcaYXVrcEjX/bc5BwCB/vbQjgFOB20VkKzDC37WUMcYYY4zJAC+NCVBVH/CL/3Wf\niDQETs5kxowxxhhjdnVJAzUReQ5/P59h1gPjM5IjY4wxxhgDeGv1WQ1oD8z2vw7Gafn5fyLydAbz\nZowxZiehqlw37Drkfsl2VozZoXi59XkwcLSqlgGIyIvAD0AXYEoG82aMMWYnsaVkCy+Mf4EaVWpk\nOyvG7FC8lKjVA2qFDdcE6vsDt20ZyZUxxhhjjPFUovYfYKKIjAYEOAYY4H9Mx8gM5s0YY4wxZpeW\nNFBT1ddF5EvgcP+ou1R1qf/zbRnLmTHGmJ2O03W0McarpLc+RUSA44FDVPW/QIGIHJ5kMWOMMSbI\nOZUYY8rLSx21F4CjgAv9wxuB5zOWI2OMMeW2ZAk0aAC5Gg9t2QI8sZit92zJdlaM2aF4CdSOUNXr\ngGIAVV0LVM1orowxxpTLJ5/AmjVw/fXZzklss2YKbGyGNJmU7awYs0PxEqiViEg+/ofeikgjwJfR\nXBljjNkp6bJDsp0FY3YoXgK1Z4FPgd1F5CHgR2BARnNljDEmJQMHZjsHsQk5ek/WmBznpdXnOyIy\nAadBgQBnq+r0jOfMGGPMTkcaTwKsVM0Yr+KWqIlI/cALWAm8B7wLrPCPqxARaSEi34nINBH5U0Ru\nqmiaxpgdy6ZNcMkluVsB3qSP/cbGpCZRidoEnHppAuwJrPV/rgssBFpXcN2lQB9V/V1EagMTROQb\nVZ1WwXSNMTuI11+HIUPgxhuznZOdR642JsB/61OXW2maMeURt0RNVVur6l44vQ+coaoNVbUB0AMY\nUdEVq+oyVf3d/3kjMB1oVtF0jTFmV5arddSMManx0pjgSFX9MjCgql8BndOZCRFpBRwKjE1nusYY\ns6vJ1RI165HAmNR4CdSWisg9ItLK/7obWJp0KY9EpBbwMdBbVTfEmH6ViIwXkfFFRUXpWq0xxhhj\nTM7zEqhdCDTCeUTHJ/7PFyZcwiMRqYITpL2jqp/EmkdVX1HVjqrasVGjRulYrTHGmBjqPVI3Y2nn\n5Tl11JxWn8YYr7w8nmMNkPYWmf4+RF8Hpqvqk+lO3xhjdiUVurPoX7bznkfHnLx8OTRpUrF12J1P\nY1KT6PEcr4rIQXGm1RSRK0Tk4gqs+2jgEqC7iEz0v06rQHrGGLPLE4E2bVJ7HEZJWUnM8U2awDHH\nVCxfZVoGWKtPY8orUYna88C9/mBtKlAEFAL7AnWAN4B3Ul2xqv4I9qhqY4xJt1mzyjd/oW5lK9Xp\nRJe483z/fcXyNGHpBKBrxRIxZhcUN1BT1YnA+f7K/h2BJsBWnFuVMyspf8YYD1QhL89uL5nUDPU5\nN0fyfJnbgdS6iDYmJV7qqG0CRmc+K8aYVL35pvP+7bdw/PHZzYvJjuLSYqCQ554r/7KnBB6NmdFA\nza4ijEmFl1afxphcsG5d3EnjxzvvM62se5f1zbxvADj63AnlXlYqIYjSQIHaHhMzvi5jdiYWqBmz\nIxg8GOrVg0n2aAMTm8/nREI/Lfy53MsGAjWfWImaMbnGc6AmIjUymRGz85s/3zpmTtnXXzvvU6dm\nNx8mZ+XP+RZFaPrbOSmnkclQyuqoGZOapIGaiHQWkWnADP/wISLyQsZzZnY6rVtnOwc7L2tEYC6c\nPgqAUxhe7mXz/CFaJrt5si6kjEmNlxK1p4CTgdUAqjoJqOATdYwxKUlysrMSSyOtRmY7CzEFb32u\naJ/djBizg/F061NVF0WMKstAXowx8VgEZiJs2r6JtVvXBofVv4+8Nv/9lNPMZB214I1Va0xgTLl4\nCdQWiUhnQEWkiojcCkzPcL6MMbHY7SPj1/qZ1tT/T/3gcJc1S7KYm+R8aiVqxqTCS6B2DXAd0AxY\nCrT3DxtjKkuSEjWL33Y9q7ascg232rK+wmlmtI6alagZkxIvD7xdBVSkT09jdilbt0JhYXbuVtod\n0l1Xrsfq1pjAmNR4afW5l4j8T0SKRGSliPxXRPaqjMyZncfEXeQies0aqFHD6c4pI+xkZ+LQNETp\nmXzWmdqtzx3XihXOFajJCi+nk3eBD3H6+mwKDAXey2SmzM7n0EOznYPKsWKF877//tnNh0nNnDVz\nkPt3zGJJJbfzbQ+8rVwtmZ++faJxY+cK1GSFl0Cthqq+raql/tcQoDDTGTPGeGcFbemx73P7ZjsL\nKfNp0posSdlz1HYe83EeXGnbfcfnJVD7SkTuEJFWItJSRG4HvhSR+iJSP+nSxpiK83hba2evoxYo\n8dpRS70ySTW/4onE2YGqUUxFa8FZY4LsWFu8NvlMJqd5CdTOB64GvgNGA9cC/wAmAOMzljNjTLRd\n/Or4o2kfAdD36L6e5v9j2R+7TFCXjttcMW9Prl5NMdVR8iq0LRWlLmthxSEVyKEpLytR2/F5afVp\nHf8Yk207e1FZOXk9+XR4pUOGc5I7MlZHbelSAKZwIDCVjds2Urta7XInU6doOWsJ3ISx4CEjVJ2W\nTL5Qv6rZqBu4ZusaOrzcgQXrF6D/tt+6opIGaiJSBacULdBt1GjgZVUtyWC+jDHlsKtcNEuKwciW\nLVCz5s69nXzeOppJKFEAfBBTnfVoap2rN1gU2cGNSbshQ5z390O9U2SjRO3bed+yYP0Czm17bqWv\ne2fk5Z/9InAY8IL/dZh/nDEmQsaPidbXJ1D+UoLr/I/onjAhA5nJlqI20C+0HdJRovbnTTFqs/j3\nOadELXV5ZaUVWt54MGuW8z57dnBUNlvbBqoqmIrx0kyok6qGVyoYJSKTMpUhY3YG6Q6YVFItS9q5\nSIobdv585319xR/enzuen+EarKzHc9hjNnJYjP9HNuuoWYlaengpUSsTkb0DA/6H3Vqn7MZUotmr\nnSvkJRtzuz/HylLek0/g/LUz3/rMVKAWSDdw6zNVddasSj7TTmbz9s2c+s6pld+gRcNLWit/pw+s\n00rU0sNLoHYb8J2IjBaRMcAooE9ms2V2ahdemFqRU0mJs9zOfLaNY8nGxQAs27As5vRdZZOkWq64\npng1ACs3FaUzOzklY4GaLz07V7tff0lLOjuS4XOGM3zOcM454JzKWWGMK5Jslqid1/a8rK17Z5I0\nUFPVb4F9gRuBG4A2qvpdpjNmdmJhFV3L5cEHnffPP09fXnYwya6OrY5abFN/d56qPnnJ7CRz7rgy\nduvT5248kOqJX3y73o2YwH76yfRPKmV9c9fMi5uHbBg6bWjW1r0z8dLX53VAdVWdrKqTgRoi8q/M\nZ82YCAsXOu+rV2c3HxWx994pRlPxl1m/Hl57LfUsZVJJSXpL+1Kto1a2rToAC2bVSVteFi/OrcA4\nKyVqo0Z5fxizL7XWosa7t6f4W33mQomaL1MdHu96vGzJf6rqusCAqq4F/pm5LBkTR+CAk6Gzoyo0\naJDhk++86CteLwJXxS+MeyFq2gMPVChHGVNcDFWrZqaD+oydfJa1d7WkTKRFi8xkIVUZK1GL+EO4\nguXjj/ecTF6OBmrbSrdx+ze3e65HdufIO3P2IcoaI1vZKFFbuaQ69C/z/F8yiXk5hOZL2D9TRPKB\nqpnLkjFxBE7OmTjzA6WlsGYN5FegJ55MxQ8L1julic1qNYmaVpKjTzTctMl5b9AgjYmqwKzTePKU\nJ9KYaJiX/8hMupUgYyVqJWGP1einbNuW2k6el6O3Pl//43Ue+/kxbj3qVk/zP/LTIxnOURr075+1\nVc+dCzecfEbW1r8z8nLGGw58ICLHi8jxwHv+ccZUrsAVeQZL1ADK0nA+SfvjOfzvSzdGNyYIDw5z\n6VYcPh9DuJhVq9OXqSmj94V3h9H1So/1FOd3dT9rLEkkfQPPVtpjLnYUec894xoe+n5q1+lSVvkl\namMXj01a+rW9bDsAJb4cveIph+Defe+9oXGVfOvzX2EVoxofuuNe+OQSL4FaX5yWntf6X98Ct2cy\nU1lTUgK9eqV+tnv77aycKSevmJyzRfFpleFbnzl6ZwYI1VB7I0Z8ko5833hj+jerrF3DxbzLaklf\nkdqG1bWc9xX1k8zp92tvANox1VMA9iw3pZy3bMtYidoy98VBqvtbwfbtaciNd9OLpnPk60dW6joj\nBYKkvx/w9+zlIYuNCZb/cWjW1r0z8dLq06eqL6nqucBVwC+qmptl2BX1669OsNWlS2rL9+qV3vx4\ndMjF70XXBdi4EerVy7EilgqqhDpqueryifGnpSPfzz1X8TQipeuxDuECP/2k/3n9jzoLTOUg/2AO\n/8gVlKlAbcumyMgstW24x/LYj5bJlKItufMolo+nf1y5K8xQYwIvh94RI8IG9h2WtnVXREnZjl1a\n6qXV52gRqSMi9YEJwKsi8lTms5ZFP/6Y7RyUz7cPR4+bPBnWrYPOnSs/P5mS4TpqgeSrV89I8hnj\n80EhW6nJpmxnJaZ0xmt5/kDr4NM9/kdj1a7eSaWjr89YJJevYBLwGqCoKkcsgqdPfQamVuyhvvFU\nVolaYHd/8MHMPPC2Q4cYI//4I34EN/v0tK07VQvWLaDqg1V36LtOXv7Zu6nqBuAcYLCqHgF4b+qz\ng+o9vHfKP+xXs7/aoXcKl/r1c6dULsN11ALJb92akeQzRhXmsjebqF3hTZPWc7I/sUYkf5zKs2Of\nSToPELwHPHmYxxK1onYRi++YQUdW5XKdgAQCnccf2/LYpPOeN83/4euv05qHLRsLYOCffHyB84T+\nTD/SJbB338NDoXHp/VO7Bxcvjhm9FVDCU/TOifqes1bPynYWKsxLoFYgIk2A84EvMpyfnOA7+mie\nGfsMBXleukKNdtq7p6U5RxXw888VW37t2vTkIx0yfOuzrExpxEryKEs5WMt04cOqGtHjfD5oSnpu\nK23YkJZkHB42xuw1/gfQjvXXDRs3LuHvW+6ffs2+7iztRIHa4/RxnQi1SnHoc/h4hXfeSbDttm1L\nvKIdNFALGLNgjPeZ0/wHnjp2d1jVFvb/FICDDl8DwJatsWsPTZ3q/E4pH+I2NwLgIe4KjlI0bX3c\n/r40onHAunVR8/ydjyihKr3xePGVYXnihDndW3fPck5S5yVQ6w98DcxV1XH+vj53zsd7+/+keT/9\nBECprzTR3KaSzZju/D6/TcjMLZ6lC6exkj0oo4AaMQKiXNBwS/S48HPLttIkJ904mrA04uSuPDf2\nuYqVDHs46a3ctNI94vDDE84fOIG1OyXFC5A0xvgnMzyrJQZ9eNI9Ik5Whg6Fnj3h/vvjJFRYGPwY\n8/vsqLc+MxWUr20Z9/lgJw852fWfWRmoJzfjbwCsW1kTgLI41byHfuGfv9PAlLKmvzkXPOElx6O+\nrk7duvD99ykl6bY8ovQsxr7xT14Nfv5vvRPTsNKKWbG4BvRTRl36bbazkjIvjQmGqurBqnqtf3ie\nqmavCYvxJgcPrucytEIntqlTnCv7997LzMkxb71TnPQXrVJOI9N3ieOVqAVc92VqnYbMYH8glP+v\n54zgxiNvSLXeuMPDPljengaCs3vMV+TjNhpsWFqu9SUynFPTllY6+OL8t1b5+0Jfvjx5GtP9+0G4\n1UU7Zoma5zpqKHnbnWC1tCzx/rhmDfDMfCD2f33E3BGu4cGTBjkf2nzmvJdVS5h+tfm/oAg67oYk\nuY4t1vH1px+qAE5buQpr/LtrsDRJJf2z1n6ThpVWzM0Xtc92FirMS2OC5iLyqYis9L8+FpHmlZE5\ns3MZyvnpSShT0ZD/fNSa+ZlJPw2Slaj9vW1q11B12AiENu2XH/sfqXF/JQX8zcZ6+lnF35hAPTYS\niHzcRpMFP+dOncs0S8cvdQAzosYtX+ou/cm9S8DYAnXUAk4+Of5P32zKcQD89nXiqh6TZ4bdQ9zL\nQxCy5AjnfebZyecF9lkwztN8yYSXqH3+vdMbStGmVWlJOdyqTdEta3OhXlq4PZc4j+Y5lN+Tz5yj\nvNxDehP4HGjqf/3PP26ns+AnClYAACAASURBVL44TTfyTUYEDj55efDp9E/T3mBD/Cf/ebSucFrT\niv6scBpehZeofTztowqlFTiRrVhYs8IHXC+P5wiuIXBCS+KHjw4BYNqIo1LK093fPpR8ph2UJglA\nX3wxtXTziOyUPbV0sm3JiPjP0ltc2xlfZY/Ez+dbtnYdpzHMSWfeiZSUlSD3S9SxKDj8453lymNF\nG7vE+n5rpjrVCWbNrnjJqEbc+oy1NY9gbIXWEav72EnLJ6V8vD8bpzTzdw6rUL6yyUug1khV31TV\nUv9rENAow/nKimWbnArZP9RpVbE+yvopbK+8ZzzUYDMNcF8tlZbtfPXrAgexEi3hnA/PSXv6gWoj\ne/FXymks9d9aq16l8n7/8BPneW3P87RM//6xSxcC4w6f55QWVChY81AJff76BQB8xSme1rV0TiMO\nZlKlXLVvK93GPaPu8XyCKCuDFSsynKkE4m2TigRWn32247SU/fFH9z69aov7mBh8ll4s4uyrnd7v\nk3AdVbZuZBg9gsObtjuPxKlbWNc130l7n+T/pFzJq1G/jfiHN2+O/B9WLJhq4G9hfSehbq5q41Tp\nWDCvOkuWpLdtiMTY5+oRamDwWf2ToqYnE9l97Pil42n/cuq3Lw9hUsrL5govgdpqEekpIvn+V0/w\n0N5+B5S/zXlydtcN8xPOd+lZ60Ccq6jHf3489oE8TgnBBx/AmfJ5cPlkJ4Fly5K3AprEIayKiJ3n\nr0/8HXLKrFmebkcFruxLMvS85R9mTA9+TjUQCFQSrsyGKOEH3sU/equE/O9/xx4f+Bn2XD2zgrny\nZn2xc1A/BW+PRTiFr5iE/6BdoeZxyb00/iUe+sF7Cdztt0Pjxv56TBW0aP2icpcgxAunygJ9bHZ6\nvtz5GDLEHagpQsHm9LcEX7iw4j9l167u4f9N/Q6eWFzui+6jj46fl61F3v7XgZaGJzCSV7mKl7nK\nNb12Tafe2OkRjxnbXs7GQAN/G4jcL2zc5lRduJXoPnDzcX7/omVVad68Yn0ZR0r2jL2z14xION2L\nBesWVGj5KuzYD7sFb4HaFTiP5lgOLAPOBS7PZKaypdHbnwY/F1CC9IsdEJzz+aUAHPBrd2775jYA\nDr0m9M8WfFR5K/aJ5x//gM85yzVO7hc49dSYR4dYLXWKi51ZAyfofZjrmq6qfD95bvSCWbJoUeLp\n2r2b8yHJo0ACJwzJi95OJaNGxj26Lli3wHXSCzyfccqUiHz+kvgP7VOfs9ETPT7C/5eqzCdhhx8r\n9/2xYvUAA1+tvJX8ASYsneAOLjwU5SzasLhc6ziA6clnSpPi0uLkM4X573+d98hATVWD+8Nfs37z\nFJFMOGJPtF+5Vk+TkuhHJQD8tuS38iUUJi8vdKIP+P32b9LSH264li0rnsapfOm6wDpI/gEbm0XN\n9/Oin137aWSjg3hPNJoyBe641n1B/P2C72FiL9bdsZYvpo0An5Pu8DnDkfuFWv6HUF/NKzHTHBPx\n5JD1G8t3gXfD4IHQT3lnwqdJ5126LP2NQmRbqFuwWD2EvcuFFV5HZF1DL8rKnM55dhYJAzURyQcG\nqOqZqtpIVXdX1bNVdWEl5a9S5ZWG/iQlVMVHPtxyS9SBtQVO5FF9bdPguIFhPWUM5Hq2Uw2++ML7\nZeLwUD/3JUnO8YFm9h98EHv6G7+9x5uPtos90YP5851sR/7xSktTK8Ro0ybx9M3rnAqpW7fHqCkf\nJhCoaZ7AgA2uK+UqxzvNwGPlrdOrnVzDT77mBAevf+BuBtdwdPwOhBetX0R+/3xuP9HpkFq/GRlz\nvlDHzlppDz0OP8+8xWUVSiveb/vEE8l/946vdoybr3g2TzzZY84cgaDhcRLfokomWWm2SNjjHZZ0\nTPjdWz3dKiqtZRuXBU8wvYf3puqDVdH+/WndJlTSvr54PWe9f1Zo2cWhoPXschRo/vSTk98avth9\naQZK1No2agvAt9+Gfstv5sapEP/XXyBCXh7sxbyoyUOGeM+fF8czstyl2OHHouJi+BJ38dT0sc3Y\ni7lR6R69p7u3lnVTCrlllXNh3aedex9evdpZx5w5cPDB7qBVEc4edCl89hY0mMEZ7U6C/j7nuFRS\nDXzu06vcL5zHh06rzjjftUpROftDHf60810n7O5hZmd/7tQpyWzlsHVz6DsOjFGYX6N+xW+++dTH\nbuV8rmXv3lCnTvJHBO4oEgZq/j49W4pI1UrKT1b5fDEuE5+K31uWah55ZcDouzk6rGDgX/hr7Z5x\nRuIVKuTNPYaCfqE/5/ffQ9Wq0Vda4cbOngPARRfBphi9Bg14eCEvbEq90vS11zrvIyNikftfSn6v\n/403nANb+HMQkz08NhDcJCuFCtz6XLxpEWyvHXe+yZPdw43nFblKJ1bO/xVFKFribq++cX38ewIz\nVs3gpFFH8Z9Rzj4y/pPY1yozV/nPsEUHusYvnb2Z/Z6JfvRBMks3Lk0a8E2fmr6aCP5HCEadSG69\nFQ5iMogwfXrsoO0f71zobOfARC+R2sJjYo5+4IHY6wicKMuo2P2bAv9ffeS8kTG3b4cO/pKWpR3g\n1XHce2/8tBasd9+aWbZxGXs92pQq/Zw8Pj/OueUoEfeb3/jjDT6f+TkAHz31T2jRIqX7f4GuiVfR\nIOb0PH8J9LQvnZLrE05wxn/9xyRO2ifGc66OPRb22guA3batDLYIDpfu3jtGUrHnbZ17bvS43adP\nYy77ACT8DzV+bU7w8xN/jndN26vNZgD29T83Ofw2Win58Kj/QLd5D3eiDxVD/4jzST9lMKH+oMN/\n6kD+6qxMcvshUp5TuFBa6qHUyV8Pb1w6GpY++iiI8PxTobq4sYKidNz6LPhxAesexXMJc2lpKGgM\ne0TgDs3Lrc95wE8icq+I3BJ4ZTpj2bBkffzbMOvHOZetdWUdHXBKXn7/bTBlD4COTh4UzT2iG4hw\nIKH7bXo/lL39vasOyLH+3k4CJ6qCbZv5mpNcJ86Zq0NN6PteE91S9fyZP3AQqfdZV+DvkKE0ohR+\n9l+hf2Ks88nq1fB//+d8nvuXe+EiGkYvEOek9N/XbgMRFm9Y7Cr5CJaorUpcWhh5wJj8kvMeKOG4\nYI7TCuidQX9zzZfokQ/VZszm6+9/Ca2jOHYQkldaykiOd/1eS/5YTNP9ajGrt7uYZMyPxcFNsHy5\nczKI3CQvjX8Jtid++u4e0ye4hitS12fK1OiLFZ+/9eZknBaXbdsCJ0UfAt6b/Z47H2G3LOLl6XcO\njVm6cN99secPBOt9+U/sGTwqeQDq9VvNiXufEDtfv8OaFTXhFWfbRgb/Lgs7Qz9lwzbnvzh/7QK2\nPgRl/Z3/0O0/xT7JyLbt9B/lTJv4v9dipy0SKNyirCx0IRTLUkIl/Ntwrq1ffRXeffRo13zj6Igi\n/PJzKQOI0SoxrL7F5b9HP89LUEbGr2mQEWvWxC/NnzEDhsXo+7vF5lAAlqie2qpNoRYgfdp15APO\nd7bP5OVsWF0zOK06W/iVI4PDBeG3hIvrefoehSQu4imJVViQSJ4zf0k57pjWaDofuaNe6iX+paVw\nxx1sqApFRaELprvuSrBMBWz+X/kqfVap4h7Oq56bfSCXh5dAbS5O11F5QO2wV4WJyCkiMlNE5ojI\nHelIs0L5STBtVC/nsvUUhieYK7b5a/9i799GAzCFg6Omb8f9EMTdWMe3/ocoV/vtQ07CfXui/cZZ\nKEJNNtHpHfdzogB8M9ylFGvrVivXUTWwo0fegt24zdnh67A+Zgfgzz4b+jx5+WQ2b99MtyedSrQN\ny9H+5Kx/Pg5A891aoP1CJ7nT+RJwSnt68VbME7wilGnsukWBW0ClJbEfOikRx/LgJisupsljEXVA\nysqCQeS20m20vacnIvDyeXtzPKOcefopIvD3DrEvAC7ruiz4Hd56p5iVc5wT/bSiacixDzgn5+er\noQPct4Qjf8qTy75yDSviOqsl6mEgsu/BmRMGAe4S0ZIYFZKarr0gZnrhIm+PFYU9cikQfBzKxLjL\nH8GvUV821m24VK0JlEAtj/5PAsz9oSUPcweKBOufxfSGUwy5bptTx3LStND+V1BFuH9k7JPvuocG\nca8/Jjo0wcNoD/Zn7/XXnQuhxx+Hn6aF6sAGbu+Fd8peDaeU/qqwOuyTa7UGETriBJ9N/prqah0Y\nS50Nse/Bzvg4/qMuvJD7hbVb1/LqXX+5fuPIi8OARx91Dwda1xZQwiEHxA5+qhd7eMIvoBpa6RN/\njud8hgLQ+ZDGHMOY4Pc8il9oFNG6PnDLVhGW0RhFaMFCFKE5i7iNxxKuewB3urajr5ybtAabUITS\n0viB6FrqU4uNwQNcy6VbUH9JoNwvwW6rwCkED9R9XrbM3/3YYPf+u9pfl7jOdtCVyRuWiIRKDNu2\ndf+lJ08ODQdu4UdqtmJO9MgwRUWxlzuaH53zQXmD31ykqglfwEHJ5knlBeTjBIF7AVWBSUDbRMsc\ndthhmknD96yv6uybWXudyNeqoP/lDFXQFy68OziNfjgZDZt/KH8PTUd1TX6D+Ol7dMk5m1RB339f\nXWnXP+NBV3qg+tNPoaT79fMFJz/8xkQ9pMfPmkepXszbifPk/1zv9ujvF76uwOdT+DL4ecMGVZ/P\nvcy3T7ylJ+x7Y1T6geHna1ztSvfaa53BB2teH7Xeld+OjpmfG3la/85QVdBf5k7WFizQmmzU3jwZ\nle8ufB+1/Kv33+0aPpofgp9f3qe5KmgnxsbdbqDamrmqoKuIv9/OWT1H6Re276izvaJ+Bv+HYQ1b\nqqrq/btdGxy3btMWvYw3XOtufMDI0O+iqmVlEYmq6j95OSrPK1Y4859wxIZy/S9KirfE/45zQoMT\nxpWpgk6ZVBZ3X4ralv84M7Tzh41/6sTe7v0vkn9aTwarglZtuEAbsSIq/U3UiBrn8/n0tv06J/1P\nKOjuLFcFHTAgNPrxf7RM/r3CBmuzPmr6a9e9WK7fIOHLi/C83dZQB08cHDPPA4fMVVAtLVX98ENn\nUmGhz7WqP/9UzaNU/+AQVdC6rHHlxedTffuwfwXHVWFb1Lquumm1guptPBp3+wU+H8Cf6dtW/pdQ\nFrFPqL5Z85SUtmmfG19Our83ZbF2Z2To+/Xzv8JWd+45ZVqdzQqqFzFEFfQj+bsrnb9mFQU/L6Vx\nwn03fD9UVW1AkVZjq34/dLmCqogzbcmS0CLH8p1r2aI6jV3D554bmveii0Kf161T7dFDtRD3seIr\nTg5+/v135+PWraobNnjbxBUFjFetYLyUdAb4AfgN+BewW0VXGJbuUcDXYcN3AncmWibTgdqw/O5p\n/zNW9PUWl7iGn+ZG1/BM9vWenkdfNnP+mO/1GZYwvfDBU85frKfxRXDENQ8/5SlPt3Xc0zV8Jp/F\nnK8O6xLmI960DZtKsv4b5trro7cXJJ3n33UuC35evjb6JB+5/ZcuXuMer/F/k32YlfVtEP56qHVH\nVdDVm9e6xj/TPSJwjxSRzs2FD8RMfy27xRz/PufHzlOCbZfKqwPjM74NA8H/2LHRm2r2bP+4qP0m\n+ooBfNqP+4L7VeB1BL+45om1bOAVuIAJP0HHejWgKOH0yn7VZKO+wWVh3zN6l9u8OfY+eO+5j1R4\nv9m+XbWIBBf6/lc7psSfniAPeZRGjavLGm3KYv30w216Pc96yufVhC4whDK9k4c8f8fAxz1YprXY\n4FzkZ1g6AjVx0klMRPbDeSTHef6gbZCqVqiWoIicC5yiqlf6hy8BjlDV6+Mt07FjRx0/fny8yRW2\nVJrSlGUZS98YY4wxOcRDDFQRIjJBVTsmnzM+L3XUUNVZwD1AX+BY4BkRmSEi6X88fAQRuUpExovI\n+KKi6H7F0smCNGOMMcbkEi+dsh8sIk8B04HuwBmqeoD/c/xnVyS3BGgRNtzcP85FVV9R1Y6q2rFR\no52y56pKtYzGAHzE39lOqHnMt3SnlHx+I/SQnbnsVen5MybXrZW6yWfaxXzJqXxPV/6iFV9zEl9y\nKrfW6c+LNa9mtv8RGV6s3Dl7J0zJTPajL49wGsPoyyP04988xq08RW+A4Hsueb7VpdnOQrksWbyD\ndLWY7N4oMAa4BKgeY9olqd5zBQpwHv3RmlBjgnaJlsl0HbVs11FQ0Ne4QhX0Oa7To/hJezHINf22\nHr+4hkdwQvDz+gH9Yqb5E0c5FZ49OmHQScplXXX47OHBNA5ksvbnnuDw21ysoDpm/hidv3a+Llix\nVl+ud15w+gcDf9areVG/pVvc73roP9F7mlwcHH6z/5e6qkY5t1mM3+61fR7UvZij+ZRETfMtXKSf\nVz9dFfRzeuipDNPfajgVkh+vdZ3n9S5n9+Dn0zu/Vu7fudqRD2o7pmgDivREvtYT+To4uXuL2L9j\n5Otl/qnt+T1q/K38J/j5ndc362GM03xKtMznVLDfvl31b3ys/2Kggk/bMjU4/6z8PXX5qgU6vOXB\n7m0csY79mg1SUD39gmWqqrplS2ieubSOWqaA7fr++z59860SrcI2rUqxk3SS7/g/Tk86T4e91mpd\n1iiEGm4clzdaB/f2VjfrGl6I2o9aM1ffOrx8ddSGVztaFXR+yw4J1wc+rXIP+k7j/aKmza5dLe6x\nqGjkr8HP8xpW8fTdevOkq95T+OvOC/5X7v028FpPbb2UN+Nvmxie4QZXGqcfGr3vhg9u3qx68qnb\ntMeZ27SA7cEJTetu0hNP3aqvcYXeQ3+9gwH6Baepgl7Hc7obaxVUf+bIhN/h/24eqlzRWSdwaMzp\n+zJTT2CEXs+zehpfBBsPpfrqyhhdT+3486jq0BadXcOR1hevdw3Px6nj26f3f+LuN8H/SURdxYsY\noicwQruedJh2vLQw6fKB16LVobp9nxQe4/rtYqVxHc/pwUxUWn+je7BMD2ai+urV05Lps/U0vtDn\nuVZ9Pp9u3qxaVhZavg7r9Eh+dqUllAUH69Rx3huyUv/Bu9qAIm3JX675I4+PpaWqjzxSrt22wqiM\nxgSZfAGnAbNwWn/enWz+jLf6rHVQhf6I8V4Drurved7IUfcd+2poYNgwVVX9+SWnwn19VrkCNVXV\n7R0Oj5umVye9fZLSD1egtmGD6r3c70rzkUfcy33R5lJV0Hvor4M+WRhc7wH8qYPoFZUvn8+nr5zT\nJji8bWmRTtnH+ff99b8h2ulKtG7fJNssIGxc/zvWuibvwTLtw2O6Zb7T5PC12pcH512yRPXXeiep\ngr7S6lpXOg33+Fp1zRrVbdv09Tv6BsdP4FB984nV2o/79NbLJ+v20u36wZUf6OndNmtHfgvO15Ux\nCqoFskl/4ih3a6Z+KPdUVfo01u/GL9HTnrlFOf8c5d4Cnb16thbejRbejdL1AVfrsGKq6sP0dW2C\nZTXcrQqrVPHpgJM66fffvaWqqiNHRrdwCsw+4JES/1HAGXHVHk7Q8tt+Xd3bOCz9S3lTQfWvtX8F\n09u+XV3fO3KZuQu2uNbf490eyn15+ilnJf0/DP1zqKqq+mbGaYQQ5u67VauwTfv1i94v4r0u5/Wo\nDQOqH558VWi+7duj/ygR6Ry01wI9m090ykd/JP1Os1fP1uvaHhM1bVDnmqqqeiaf6TBO1Ye4U6f0\n6aUl+aK+sJrPa34PXbA9TN+4LYT5Z0e9/oCurnHPcIO+wDX6yKvTPG2fWK+beUIfeihsG3gQ3rry\nNL7QIUNUW/KX1mdVcPwPP6g2aKD62GPuZT/4aJvWZ5Uezzeq6m4lCKrV2ax7MUfBaf3366+qv9ZI\nHDAvWLdAVTVuIBs56lw+dI0IbzwVmF9B/82/tQvf6w0841oGVOfRKjjcink6Hef41/nSfFVV7XnU\nTeXapoHlB9w+OOY+GXj1ZHAwf4tpqk6DDNXzznOnN+W3LTrgjvWqJSWqf/yhun27vt/4Jlday9ev\nDn7u0vAl9/ePkYfAx/XF6/Wee1SHDw+tL+ZXjVjuPD5QBf2Qc3XVKtWZM1WLi51Zb7nFmWfrVtXO\nnVWfeEK1BaHGUrWIaF3u95//OK39K0NGAzXgLOC6sOGx/hKwecC5FV1xKq9MB2p9uvaJu6OvK3Cu\nYKfsG31wVfylCHGWnfVZ6NEG8a6omrFIa7IxatJr/wp73EMYn895fXznOJ1La23KYmdCSXQrxxdf\nVP3mG+/bYeKyidrh5Q66cdtG17qHdLkg+k8ZZvRBTgnE9Tyrgz9bqL2uWen+j0R+OVXd0v9e1/Cm\nB5wWXzp/vv77u3/r2MVjVRcuDDUdw2np2oxFzqMQAsLS7dffCT66d3dPCvy5P27ttKS9ofatqqp6\n1pHL9Q4G6FUnDXEfYPqFvuS4ESNc3/3VV2Nvu5+GjXEfaO6qrlxzUNQ2eGX8K67HZjz1y1NKP/Sm\nr27Sjds2Kv3Q58Y+pxx3n3s5rtSpU1WXLg2tc2HtmjEPRokEZi0pK3FakvlH/K3XaFVVHXbEee70\nIg68Dz23MG6iffpE/yaRVm1epc+NfU5vrHNzzP/DBx+ojh6t+umn7uXCH0cTK+2iItXTTnPeo/IQ\n7/X221Hfgb519c2RT+rwxg11j1vibFP/vMcxSpuxSN980xm1fl1Yi8RPPolaXyDLvQ84OjjunQOd\n91uu3VtVVTuftCz5zxm+n8X6f/n34evbt49aP6g+9MqfrtLUWK8p1faOOb4+q9zb14PbecSVhw8/\nVG3dukwvu7zEUzqRswSz0/UBrV07evq4Pdol/t39XufymNNXRDxp5cKq78Xcjt34NliKF/k6nm9c\n889hL9dv05yFejmvB48DPbv0if4iCUxjf1XQB3oPidgoodflvB4crMO64OM3nn7a0yq0rHi7K72i\nDeu0LVO1Fht0/73d309Vo1o5vzT2NT381cM9/abhIwPTfnrolrjbpKzMX5ofsXhDVuoZ/FdBdR11\nyrVN0y3TgdpPQIuw4YlAA2BP4NuKrjiVV6YDtRtvCz03KfL16vn7qIJOvK+3bqBW1PS/aBl3Wd/q\n0BXI3jWWxpwn3rFk6JQP9NbD9tI6gWeMRfj88+h98I3j3CVDFRKWxgP3fKUz2E9bsCBmsmt6hgK1\nt/+7IDqJWAfKqf4TxfjxzrDPp7pqVcK8TMvbJ/qr4dwm248Z+uBDpa7pgc+BQpFfDr/BlYcuXZyP\nfR8KBWqn8YX7u21ZozeeUEVP6f6tguprr8XO4o//G+3+Tfuh/Ktt9DZQjRmohQeHqqoPPVTmWu5l\n/hm1zsW77Ra9XZPghNsUyoK3QwPLPvyFc8B/pcftGnMjgm7cGC9RZ/rWrRq1TDzX1e3rni/J/OEn\ne0/fN84f60PO1ffwX3iER4P+6Z9O/1RfHv9yzN8kct6Y2SgoCI5cQ92o/7qqavMTz9MiGuh+zNC6\nfdE7u6Nj5n0XTEKqrU/89SLXH+N7vjXxLX26uxN0ry7YTXXUqODkt97bEHe5wOuX6gdHjevOyFC+\nyrHPXUAo0DnkENW1a6O/SyJjx7pL8bueuEYHPLtE56+drwceGJ3EuGYHJPxuATPy3BfZBzNRj2F0\nVLbGPnFr9DF7n2GhURef7Lw3Hasc+YSC6lH8FJx/2KRfdCTOI6C+pZuzb93cXLngLC0tK1VV1d4n\nDijXNp1KW1XQgQ+96t6Ornz6okb/9puW79EUYQuv37pBOepxLby2qw4dFPp+77zj3+4c5vn/GWuW\n7ozUY/lOP/vMKSH/9eHry7VNwld9++2q79fqVa7l0y0dgVqixgRVVTW847EfVXW1Oh2y14y30I6s\ngALX8BkXhj7/en5nru4BGy49i5ltoiu8tsLd3991HQ8Ifpb69Tm+F7T7F/wyv0nslZ96A9xdyN13\nO4NHHQXffQcqwuNnzGND4l6EXFoesNT7zElcyiB68RYAV910Avs3+JwaR23nxhuj560XVse62W6N\no6ZfwPu0J6Lj83btnL/QYYc5wyLQIHafhQHqqxI1TlDO4Atm0YYTT3C6NYl8Mn5+oLcTce/2t9/u\nvDdvtSE4Lq9AXfPUq16PJ7/eSuM9nf4S8+L8c3xlznJ/0YqLL4Yltyxh0Jmxe7B+qPtDNKnl7A/H\ntHR6kxjR0/3Um5tvzuOOO+Bj4jewrlrVQz9/kbo8Bv3ykYiny0te4HvHfkT6SXxNrVqJky5P10J1\nCneLGjeEi+POX0L0b5+K8xnKNbwE/fvH7JP37P3PdkIYL25uAXfUcY+bNy/YcerhhV9xNS9FLXZe\n79Y0OuYmZt3Zkbev+B8/9OxCl1ahXkXG/r6VZ9+Z5f1LxdDrkF40KHS6bxu2+8nQrVtwWu1qSTqY\neecdJldvEzVaER58sELZYuJEqFvONhmHHw59+4aGvx9RjztvaErLui15KXrz4ssL7YiTwnqEOfbE\njvQ4OtQx+6Jm7oxM5hC+59io9MK7+gPo0G0hXPD34PCSF9+APk3gii5wSh+4vKtrme5t29OLwTzK\n7ZzQ2em67OD96sMB/yU/L3BwKt9/WfKdp+7XrRu7N5ZAziN16pR6F2D5eXlw8q0U7/EDNaqH8nvR\nRamlF2kUxzOG4zjrLHjwQdiyu7fuuQJGhB1C+/SBfUtmxJ95B5EoUHNtHXU/32ynbJrTcnd3/Dku\n0HXeiy/yzBkvcNQDb9Kl9bHsUa961LJLI05e0+rWgAMOgOrOvLfc/QVfPDiPeA1XRw48m5V3Lgr0\nhcw118Bxx4X6pyyPgrz0dZkxmEt529+R8O4NC9BVbZjx8z4880yMmfv2ZX37Y/ml9cUc3rFq1OQP\nuYBJtK9wnnbbPXGP0Icf7sR+F0ec7wMHJl/Ebn/GGc78teqEWgDlx9iG+Xn5dOjgJLJPnIZsBQXO\n76UIQ4ZA09pN2b+V/0TQ/W7XvHd1vYulfZygukOTDvju83Hi3u7OqatXh4cfhhGcBMBVvBq1znkH\n7B78fJvHPjD/71CnU1aJOFoHBiUv4ij+2GN0YALf+PORLqru30JQLiF2YBuYnhadBrK+zwFw771h\nEbzTsfkUDgTg4oMvUsch4gAAIABJREFU5ty257KsT+zH9rRgIW2YAbstptv+EY9JatECOjvBwJyr\nL+SVs38OTeu9JwAPnHAfV9wynzX3LqTHfj344fIfyAu7iOjUdg9uuGi/lL/iHjhdKG2t5hzXLln6\nYfkSuOgi7tnjZibQwTVakWCfvrni6EB3pi1+Co4bv2fb4Of2t4fKHL4/ejzDTgz9Hr4mrTytwxex\n7/32zZ70O+EOOKkPXL8fTWs3hdrLocDpe+/DPjcykfZMZ3864+RrKc24g0fRo5z+9sZcNoaJV4d1\no1bO4En9f1iN6Hsq0NdrJoTvo7H+jWn7j/ot73QAXS6HvDj9/0Y68UTYuNG5Ttp9d9iSl+TKcgeQ\nKFAbKyL/jBwpIlfjPPR2p3PQPu4rqxW1cXpCvvpqalatyWXtL0NEaF67WdSyTSO6vlxWowr8+Sds\n3gzA6fudTut6reOu+/i9jqdRzUZcfjl88w1ccokzPvJEGqlhrL7Ofen9o3jWvDm7/TGa8fPqUzOD\nZa6Sn9qlYLA/O/+BZsSJ7n748sP+DvmRHX/6XX89TJkCXbvGXsfuuzvBXkG9UKnmEW1a8+GE4awf\nlrg722S/NcArRP0l2VIr1Hep1/4XX+7xMpvv2hw1vlbVOD/crbeycZ8Osaf51WYDtdmQcJ5I5e0v\nMnASeJw+5Vou0r2PrmTjgNlR4xuxioOZAkCtqrUYet5QGteKLh0GWEwLZtGGrXdvZcQlCZ7/XW8+\ntB8cHNz2uNN3Yc2qNXn9rNepV718JQZerWQPABZ1uYwbeJZquEtdAgWG5/r7toylNK8qYyJKl1rs\ntYXGsTeJJ++TvJ/YlPStC5d2Dw5+0PFETuErBB/UWMudDOAWnohabN7DXYKfL2Gwa9pnnzmFowB5\nEYeE/Hw4s82Z0PlJaOjel5bespTz2p3HFmrSlun8Qmeq5Yf1Mey/EKxbWJdDGh8SHK3lDHJ8/v+P\nlrn/R4G+XtPtOL6jWkE1alSpwUunxyjGBFenyaNrHJEwvd2iC9SjnNnmTOqfeAZ/3Tzfcz5r1Qpe\nJ/FBg+sA+IpTPC+faxIFajcDl4vIdyLyhP81GrgMcvABLmlQEGtz5OVFlRFL2C2R1SvmA/Dc4e7F\nZrZa4eoYO5Yhh13BtIhASwROOCG02Nn7n831na5n5a0rY6Zx1FHR47bWKcd90mR67wm3J74VmUyN\nNGYHYI+GNfnpJ+f2iRfHRtzFWNjc+QevaOIu3csnVLJSe7fYpXYicOCBCVYW2Dfy3SWh53U4hTqF\nSW41JUo2QUCTStian5dPjSrRP8zZB5wVd5lJk2DNmvhpNt6nNpuonfItFS/SdbXev1t/alVNz5V2\nYUEhBXkFyWf0q5qf3tKOnj1h4ED3uEBpGsC+jdoxkBs4v2c1YvmYc90jfvgBrnNObkJ0CfTBh5cv\nGK801ddz/iFnBwebHDSLrzkFLjuWN896k0e4k6e4JWqxkppVGNqwHQDF7T5zTTvrLGjtv74+9JTL\nopZtt3u7mFlpUtup0vDkW6FO7cMvxM478Fx+vPzHqOVaHf5nnC8XmwaSLHX/6e5kQNxlDj20XKsA\n4BS+Yl9mMYbjyJM8Nt+1mas7Xh2WgTB5oWPf9i11oqeHWbIE/vor8bprVq3J5xd+Tsu6LcufcUAL\nnP3+VIantHwuiHt0UdWVQGcR6Q4E9sZhqjqqUnKWBflhRbpbEx13n3vOua0JaK2a1L8d8uvV44bf\n1obmqZ9k7wP2PHApHU6BqmXELYeoml+V5057zkPuQ6b06MT2T4bSI7rAoPzqLko+TxJffAHduyef\nz6v8/ILg1VKkTZuix33xBSwNq7Y3bf9zaMoSrtm7qWu+fAn96G3ar0otc/5Abc9VJaktH0cgQHFu\nfb7iXmXY58iTannl+yvfxSrdq1EjcdA9erRzu6FqOeIQTTGquzVGyYgXP9CF++jPdyktnZveftv/\nIaxySqA0DULXDoFNvdZ/mBrlP5L/7W/Ap2EJdunivIDaVetEXSTUqpta1YratUthY0qLeqL/Vle9\nwqIqf0A/J++Xtf+ey/3jT9zrRBasD9Up9lq9pKDdQbB1a7A6CzjH50U3L6JRjdh1Wuo3ch8Hqtcs\nZevmAl4643nq14v+r1Y99SheGP4+A7rCYi+Z8v+ovogStYe5K+bsZSnWivk6TmlUrH9vtVobYL3z\n+SS+SZhuzZpk9O4LhKpxfMmpnJbZVWVM0qO6qo5S1ef8r502SAPID7s6qJ7ogcX77x/8mCd5rK0B\nq7atTbCA2zcXHeksW1DGtiqwsbDcWU0sL48n4gQy5TXsomH8+n+/ViiNbt1CJ4sKCTQyiCwiCxPr\nT1+rFuwXVtVHFZbRNKrkp0nN0MktX1KooO+kDsD8Rump9B7pFbkyydorVpwVvD3sH37LXz/Ri2bN\n4Pzzw/JSLXYJjks594tAwPpkFW+F+sM52TV8K48zmm5x5t45+fy7cqABTN26zivwn/z009jLAexT\nf++o4N8XqxTFg+o1nIz8gw9SWt6L8AuMbaXbXNNefBEGDYIRl4xg5vWhkq7L2l8WLF2VZJX5C6MP\n1s3rNKeav9Rm7JVjKbotrKvDwEZuMgGAMd8VcNVVUK9u7FNv3doNua4HLPFwSxDi3/qMJy8vfkOo\nVBx36J5R46rVWx38/HWa67SmIljvNs115ypTGn+ynUDYDfMhB3lbpH71+rx4+oss7L3Q82oCBxNV\nZdw/x/Hx+R+XK5vJnLrvqYxuDTNf/w9s25Z8gQRO2/c0jmieuJ5BpQkUy8W43/vDD07dsfKIDNTC\nb0nllbP1VaRWRektUQu4Sl+LHhn2RfbvUrEWToGDeKAtwaURdXbKY80f85LOU7dxakUst5Q87Wm+\nLbiLACsayKaiVd1WAFzFy3zDCZW+/sgStUjnJOyxOY+luEueK3pvO2N11CIc3eJo1/A118Cll0bP\nt1vhbjSr4w84RDniCGj5/+3deZgU9Z3H8fd3mAPkvm/Bk3OB4KCoGA8U1DUhsmgEEw94ZI13zJoV\nT2LibpLVaNRNkERNssYrHiExB5KomydGEKKoKEHRYDzwIJ7RBRV/+0dVMz19VvdUd1V1f17PM8/U\nVFdX/eo3Vd3f+p3l1bKx99C96bdTW3uWHZ05N3u92qdMgeuvz5+FqV7Ynx8XLI/aqj7brz+f/wya\n5NJc2D5Q3WmnwtdCp67Z7WCrbcRIL41JrvpUoJZm5JTDmH4C7HI2nPy54tunnNp6KsN7Dmf2f0yk\n9RTod17h7VO9ZhyO1iGtzB4T7tz2Y/uPxV3qGDX/vNLqoeKuwDfOtGlF2o7l2E2mhrTboeiTdR4f\n7TyMh4bDF84eXnzjEqSeBq9nYa4Xcy+Xc5wdT58dD2hcv+ItzvsMD14SDfAbjgDggM4Pwm23wRNP\nFNw+M+COIlD748l/5J7P38MPWFi0KqgSUiVq+YKDu+/2hpPJp1CenTsDLo5pAWW+TiA5+Z8J/zLm\naFauhE2bwknDzP0HwcEX8c1f5+/JnC71EH/7UwFLHVM92TOqNM84I3vTUDSV9uD/4fbo59JMVcCs\n7pfczgTBW8DWgYHdBrL8po9o+np51VZ3L1qLfa34F0G3Zq9Rea7G3GVZMBVenQQ5xmqqKakIq4Nl\n9/nivYa0L6Syqz6bmpi2AEb1DbkHhc8NXZW9Lv2LtMw45LucxZlcuyNPtnbueMORYk/bAKQNz+Hu\nvIvMdu2Zuszaii1zsBWCFMzEIVAb2mMoQ3tk9xSvtAn+0GH5bpvU+tmz4Z2789e1ZVZ9urS2vFf5\nTSy+HiRB/gG9qs/bgrwjFOfsU7yafMSnR8AGOPToPONclmlAt/64+zs46FwBTT3eha3Qf/AH7dYP\nGVqZ63zWqIzORkVKV4/cmv15FZUpW1SiVjOCDnLpmspvg9R6qNfuZ/L0L5S9j3aGr4Ip14ezrzgr\nVjQQUL5ALb0UqaUl+ifBdF17eh/Ep778eNZr6afhrLxb+hy+Syc+2ZEna8ceyJf4Hl3GZ4/bFlSQ\n3r6pwOmewZ8pVgcHQK9BbwOw77EPFdnS04nwxhTMdNNNcEHuNtuReI+2Xqzvvw+rV3vL+W6b9Ptg\nA97AtrkG5s0Mbl2ZbdSicvWq4tXkPa/9BtxxB32PO6zotpVUakl2U3ev6UDPnpVpapFp2YZl7VcM\nGeKNKrtuXc7tf8lRVUhVEf6FH4f2cuVSoJYh6Dg2tnkzvFhej8iGuXNh7Voajjm2+MbSplhjmxIV\nKlHb8/Do21ak6zM0WBXhinEd+zDaEVyZsYQvMahP9sj0gZnx+o3XsvW5/KPrp3p9ll7VHOw+fS2t\n96P3LmPq1BIPlcdJJ8Hll4ezrzD8rOWYHcs77dTW6qHYbXPXXW3B2Lu0H07BuY73JM5UrTZqJWlp\ngWOOCe2zpVxNnUorAEjFzFmpznEey5ZlrSrNuBzVsWZwxRXeDDNtqdqx9Bnu7eBBO+7jrl5p8UwK\njHUYcwrUMqSXqM2fND//hn37wrBh5R9o4sTi20h7IQVq+QpNB3RpG+H/jWkBG7zFQVp+fNgcoKdl\nAamOmqn2epveKd4hoJABJ59B5133KLrd5zb/CjP4/e/h9dxDBnpK/NefxTUs4Ic8ijd41NLr4eGH\nS9tHUlzQNXcD8nxVn3v7Yz/+8peFq4Szqj7LTF8U1c5J89lRn+W8/c5r33O0gB2BWsaHWq6cHjIk\nx8qA7lx3D8w5ruT3XTz21pLfM3x4/plfyvHW6H2Zyy3MOjReD9+lUKCWIb1E7YZZN5S/n0ur1xV4\n/qT5dG8ufzDVxAipjVoqvh6a0WyoW9q8k2P6jSFOUh+8S0YWnlJo0swSu74WO+DjObrJVdAhh5B3\nmrVyvE83bmQBf8LrAfhRzxxTedSI/7Pc7QpTVZ/XZ7SOGDHCu6WOOgoe9aeJeoHs7o5ZVZ9pXxvL\nv7Cch+YHq4ZOqeTwHOmCzPQRN40NjXz7sG+36zlaSOp/k2ciFQAmTvKacXRkiKSGBkp+SNqLNTw8\nuPTg7m9/g2fDGAM0zW3M5cPGyrQbrgYFahmaOzXTrbkbS49aWnzjHC6YdkHVv+RvmHUD7y6K6Wjh\nYQqpjdppp3lTw6Sm6UpxQ4dxCV9jNza2m9alHBv+vqH4RiVIfRCfuim7GjFVfXgOV9GyUzhtVQZ0\n9aOlCeUPzxFE6ovmx733L7Jl5vtKc/e07zCWp9g2qMxxF2LoKs7ZMS9pSiurGU/7YP0Ir6Msa9bk\n39eVfIVJPMbDZA/A+GHGvJHpgduM3Waw3/Bggzamei5Xu+rz7H3OrurxqumTPFWf6QNJhxGvpv7H\nD5xYfKjobvVQaFBl6vWZocEaeG9R+cNnXz79ci6fHqOGK7UkpKrPhgZvapgsZnydgDP/FjGqbwfa\nduVQKDBJz44whtUA2K3PbgAcP+GLRbbsmB0lAm/tGmj7cs9ve0MT6xlbfMMEOZergLRrwxx/pjVr\nu112KV6a4mjgcSZlr3ewlYxBXsu8xFr3ex9+rirQMKWqPjPnIW1oyA7UOlKiNrDbwMC1RN6wKK+0\nO7Z0jErUpOrWr3gJnn669DeGVPWZT2rQ8dNOK38fvTr3AsIvUXMBblXDhVbdE+Z4akEO1HtwwDY5\nyR1cPLGyh+coc2aCrl6J+NwqDs0BwXvyJ1FbG7X824QRqJXC/A/QTYyMVaD22+SOzqFATapv+6Ch\nO+ZKLUlIVZ/5dO7szYN4zTXl72NQt0GsP3092y7q2IwQme4dMosfsoC+R2YPNFbJEopKf7inhnr4\nzOZSP0VLS9iZZ3q/x9ZWoVrovnrB/2Wty+5M0Clrm1JUq+qz4g8ZMZCvM0E71c6GBQtY/lvHm/St\n8oFzO/RQb1aKMgdpiAUFaiJpevWCTh37HmJ0v9HtpqMKw8eNXTmFH/Jmgfkzw5zLrlpPwqkg895B\nwYYVKffLd84cL+gMs6NC3IQxvtnQYdnDpGR3JihPjxAGUS7HNY904Mkrlw0b4J13wt1nmXbv63Uu\nmjhgQt5tJkz0xhLs06cqSQJCH0mpQ5qavHleOzJIQ9QUqEnVlX3zxunur7KvzPfacP15cY4BaFPZ\ncdCliStF+LjRC2iPejW5YxwFddll5c8hWS0HT88e6DmscdQmjvPaFlSr12fKWXufFe4O99wTevQo\nvl0VdG32gt+W1IPh6NHe77SnzSuu+oiHH4Y9io+SE7o6/KiuCAVqUnXDy50Gc/FiGDw456Tste64\nY5v46COYPKFL3m2s+R9l73/1aliao6Nzpas+V+6xP5dxMVMP/l2J70xeu6OLLw5vDslK6dSQUXrm\nsie2/7ixvLH6BpUw9aYElNkAbcaM9uuBvt27hTbIc1A13CwwEgrUpOrKfhidOhVeecWrn6xDjfn6\naIfw2NraCqeckr3LW27p8K4Lcg2duJTLeL85WLWYHtDzs0ItygPvIzuH72MG5/g9TAFeGxDiaKRV\nEHS2mUTK11Mg4qKsOq78qAgFaiIJl17dGfYgn3Pnhrq7HPRJHicNOa8f47u0TWyelNKSJA54W7Lz\nz/d+77VXtOnII8k9LeNEgZpINS1fDitWhLrLRw71etD+fHR4+6xaZ4LUl34dfKcmQUWDm0MO8eYo\nW7WqcsfI4dpHrq3q8apq+nTvJuobjx6WKQccABMmwGOPRZ2S2qBATarmyivhwQejTkXEZszw+ouH\naPOu/bHF8HwFenXdWvpUfWUqrZgmIYU6iTd9etvy1XRwhP+BA2Hr1rZJRqvkzL3PrOrxBLp3h8cf\nh0nZYyhLGRSoSdWcey4ceGDUqahtYfX6TBWsHFf6VH2lHqnEzUvbfuDA0nZf7zKrPi+4AF54wVv+\nMldjFWrx9b8j4E8JHj5BpJI0hZRIwlWyjVrF2yMFGVq9AzZtgu3bK7LrmpR5/TQ0wM47Z2xUgX/V\nQSdXbNc1PTOB1AeVqIlIlqq3wy45UAu2fefO0DWacVYTqVNDB0d7jpHUA8x1q6+LOCUiHaNATSTh\n0ktBkjbgbalVn0k7u6Rpacw9o8bNN6f/lYz/wpDuQ6JOgkgoFKiJJNzY/pWbwPL2Kg0iH/SrP4xp\nkmpVJUtBjz8+/a9kVCXOHjObX837FdsvqaO673nzvN8zZ0abDgmVAjWRhFvwqQXsP3z/UPeZ+tI/\n9thQd5ul/OZDyQgWalFSgmUz48g9jqTB6uhrbp99vJtqdIhj9Ujk6ugKFqlNZsbUYVN3LIdht928\n33fcEcru8tsxjlqwwKvSbedefRW2bKnsMeKsHsaIFUkaBWoikqW1FTZsgE8+qfSRUpFBPErIBg6M\n3dihge03ey0AI1vXR5wSEQmTAjWRGlCJIQj23LMKJSx+sp98/YlAm/fu3BuAVS9Xd3T7JOjR930A\nNn2kvBGpJQrURGpI0np9jurntaWZPeboQNtPGDgRgPP2+2rF0pRUqVB9bP9xZe8jUGCekDZqIrVC\ngZqIRGZQt8EADOkRbCiFVBu8uphwu0SpRvNPb1lX0eN8ogFkRapKgZpIDUhN7JO0AEbf+eE5aORB\n3sLakyNNh4iES4GaSA1ItVFLWtVnynUaPL7DPvvPzTQ3w5o15e8jWNVn+fvPZ/GBi1k0bVH4Oxap\nAZrrU0QiM28e3HJLFYYBqQODB8O2bVGnojyXHnRp1EkQiS0FaiI1wMVkeItS9esHK1dGnQopxah/\nej/qJIjUFVV9itSQpLVRk3gpdPl0avkAgN3HKFATqSYFaiI1oBLjqImk0xUmEg0FaiI1JKmdCSQe\nghTIJrWaXSSpIgnUzOy/zOwvZvaEmd1jZr2iSIeIJIsKDqOjRwCRaERVorYCGO+cmwA8A6hftkgH\n1Fsph5riVV99XWHJN4DXGMaLUSdDQhBJr0/n3H1pf64E5kSRDpFao84E0hFBLh9dYsnwBgOiToKE\nJA5t1OYDv4k6ESIiIiJxU7ESNTP7HTAox0sXOueW+dtcCHwM/LTAfhYCCwF23nnnCqRUJPn6dukL\nQM+WnhGnRJKod294660iG2kydpFIVCxQc84dWuh1MzsJOAqY7gqMLeCcWwosBWhtbVUzCZEcFh2w\niMHdB/PFiV+MOimSYIGqNU0fwyLVFEkbNTM7HPgqcKBz7oMo0iBSS5o7NbNwr4VRJ0NEREIWVRu1\n64DuwAozW2tmSyJKh4hI3Qsy7MnQmbcD0NysEjWRaoqq1+fuURxXRETyK1T1OeLoG3hxr5NpbPxD\n9RIkIrHo9SkiIiIiOShQExGpc5rxQSS+FKiJiAigwWxF4kiBmoiIFDWu/zgAenfpHXFKROqLAjUR\nSYy+3ri+fOtb0aYjDpqawttXkKrPqw+/mvtPuJ/xA8aHd2ARKSqSXp8iIuVYsABaWmDevKhTEr2X\nXoJ33w13n4WqPjs3dubgXQ4O94AiUpQCNRFJjIYGOOGEqFMRDwMGeD8iUttU9SkiUufU61MkvhSo\niYgIoF6fInGkQE1EREQkphSoiYjUOVV9isSXAjUREQFU9SkSRwrURERERGJKgZqISJ1T1adIfClQ\nExERQFWfInGkQE1EREQkphSoiYjUOVV91p5LL4WePaNOhYRBgZqIiACq+qwlixfD229HnQoJgwI1\nERERkZhSoCYiUudU9SkSXwrUREQEUNWnSBwpUBMRERGJKQVqIiJ1TlWfIvGlQE1ERABVfYrEkQI1\nERERkZhSoCYiUueOP9773dgYbTpEJJsCNRGROrdkCbz5JjQ1RZ0SEcmkQE1EpM41NkLv3lGnQkRy\nUaAmIiIiElMK1ERERERiSoGaiIiISEwpUBMRERGJKQVqIiIiIjGlQE1EREQkphSoiYiIiMSUAjUR\nERGRmFKgJiIiIhJTCtREREREYsqcc1GnITAzewN4ocKH6QdsqfAxaoXyKjjlVXDKq+CUV8Epr4JT\nXgUTJJ9GOOf6d+QgiQrUqsHM1jjnWqNORxIor4JTXgWnvApOeRWc8io45VUw1conVX2KiIiIxJQC\nNREREZGYUqCWbWnUCUgQ5VVwyqvglFfBKa+CU14Fp7wKpir5pDZqIiIiIjGlEjURERGRmFKglsbM\nDjezDWa20czOjzo91WJmm8zsSTNba2Zr/HV9zGyFmT3r/+7trzczu8bPoyfMbHLafk70t3/WzE5M\nW7+Xv/+N/nut+mdZHjO70cxeN7N1aesqnjf5jhFnefJqsZm97F9ba83syLTXFvnnvcHMZqatz3kf\nmtkuZrbKX3+7mTX761v8vzf6r4+szhmXz8yGm9kDZva0mT1lZmf763VtZSiQV7q2MphZZzN7xMwe\n9/Pqa/76ks8vrDyMowL59CMz+2vaNTXJXx/t/eec049X/dsJeA7YFWgGHgfGRp2uKp37JqBfxrpv\nA+f7y+cD3/KXjwR+AxgwFVjlr+8DPO//7u0v9/Zfe8Tf1vz3HhH1OZeQN58GJgPrqpk3+Y4R5588\nebUY+Lcc247177EWYBf/3utU6D4E7gCO85eXAF/yl08DlvjLxwG3R50XAfJqMDDZX+4OPOPnia6t\n4Hmlayv73A3o5i83Aav8a6Ck8wszD+P4UyCffgTMybF9pPefStTa7A1sdM4975z7ELgNmBVxmqI0\nC/ixv/xj4HNp63/iPCuBXmY2GJgJrHDOvemcewtYARzuv9bDObfSeVfmT9L2FXvOuT8Ab2asrkbe\n5DtGbOXJq3xmAbc557Y55/4KbMS7B3Peh/7T6CHAnf77M/M9lVd3AtNTT69x5Zzb7Jx71F9+D1gP\nDEXXVpYCeZVP3V5b/vXxD//PJv/HUfr5hZmHsVMgn/KJ9P5ToNZmKPBi2t8vUfjDoJY44D4z+7OZ\nLfTXDXTObfaXXwUG+sv58qnQ+pdyrE+yauRNvmMk0Rl+dcGNacX8peZVX+Bt59zHGevb7ct//R1/\n+0Twq5s+hfdUr2urgIy8Al1bWcysk5mtBV7HCxyeo/TzCzMPYykzn5xzqWvqcv+ausrMWvx1kd5/\nCtQEYJpzbjJwBHC6mX06/UX/iUDdg3OoRt4kPP+/D+wGTAI2A1dGm5x4MbNuwF3AOc65d9Nf07XV\nXo680rWVg3Nuu3NuEjAMrwRsdMRJiqXMfDKz8cAivPyagled+e8VTkOg+0+BWpuXgeFpfw/z19U8\n59zL/u/XgXvwbu7X/OJb/N+v+5vny6dC64flWJ9k1cibfMdIFOfca/4H4ifAD/CuLSg9r/6OV93Q\nmLG+3b7813v628eamTXhBR4/dc7d7a/WtZVDrrzStVWYc+5t4AFgX0o/vzDzMNbS8ulwv5rdOee2\nATdR/jUV6v2nQK3NamAPv+dKM17Dyl9EnKaKM7OuZtY9tQzMANbhnXuqB8uJwDJ/+RfACX4vmKnA\nO34x7nJghpn19qsgZgDL/dfeNbOpfjuGE9L2lVTVyJt8x0iU1AeS72i8awu88zvOvF5nuwB74DW+\nzXkf+k+eDwBz/Pdn5nsqr+YA9/vbx5b//74BWO+c+07aS7q2MuTLK11b2cysv5n18pe7AIfhtekr\n9fzCzMPYyZNPf0kLoAyv7Vj6NRXd/edi0AMjLj94PTuewavTvzDq9FTpnHfF67nzOPBU6rzx2hz8\nHngW+B3Qx19vwH/7efQk0Jq2r/l4jU43AienrW/1L/jngOvwB1pOwg9wK161ykd47QwWVCNv8h0j\nzj958up//Lx4wv+AGpy2/YX+eW8grSdwvvvQv1Yf8fPwZ0CLv76z//dG//Vdo86LAHk1Da/K4wlg\nrf9zpK6tkvJK11Z2Xk0AHvPzZB1wSbnnF1YexvGnQD7d719T64CbaesZGun9p5kJRERERGJKVZ8i\nIiIiMaVATURERCSmFKiJiIiIxJQCNREREZGYUqAmIiIiElMK1EQkcczsQjN7yp/qZa2Z7VPBYz1o\nZq2V2r+ISCGNxTcREYkPM9sXOAqY7JzbZmb9gOaIkyUiUhEqURORpBkMbHHeNC8457Y4514xs0vM\nbLWZrTOzpf61HtmHAAACIElEQVSI4KkSsavMbI2ZrTezKWZ2t5k9a2bf8LcZaWZ/MbOf+tvcaWY7\nZR7YzGaY2cNm9qiZ/cy8+Scxs2+a2dN+Cd8VVcwLEalxCtREJGnuA4ab2TNm9j0zO9Bff51zbopz\nbjzQBa/ULeVD51wrsARvypbTgfHASWbW199mFPA959wY4F3gtPSD+iV3FwGHOucmA2uAc/33Hw2M\nc85NAL5RgXMWkTqlQE1EEsU59w9gL2Ah8AZwu5mdBBxsZqvM7EngEGBc2ttS8/Y+CTzlvMmXtwHP\n0zap8ovOuYf85Zvxpi5KNxUYCzxkZmvx5ukbAbwDbAVuMLPZwAehnayI1D21URORxHHObQceBB70\nA7N/xZu/r9U596KZLcabxzBlm//7k7Tl1N+pz8HM+fQy/zZghXNubmZ6zGxvYDrepNRn4AWKIiId\nphI1EUkUMxtlZnukrZqEN3E0wBa/3dicMna9s99RAWAe8MeM11cC+5vZ7n46uprZnv7xejrnfg18\nGZhYxrFFRHJSiZqIJE034Foz6wV8DGzEqwZ9G1gHvAqsLmO/G4DTzexG4Gng++kvOufe8KtYbzWz\nFn/1RcB7wDIz64xX6nZuGccWEcnJnMss3RcRqS9mNhK41++IICISG6r6FBEREYkplaiJiIiIxJRK\n1ERERERiSoGaiIiISEwpUBMRERGJKQVqIiIiIjGlQE1EREQkphSoiYiIiMTU/wO/XDX8y2kLVgAA\nAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnQAAAFNCAYAAACJ9PI3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhVkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnoAAAFNCAYAAACXPcKEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBEEQshQR9ARBEARBELKU/wcUGvxg8AhCkgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xdn3znBvhLkr",
        "colab_type": "text"
      },
      "source": [
        "## Standardize the values"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ve6PFl8IU9gy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# For gyroscope data\n",
        "dataset['rotationRate.x'] = (dataset['rotationRate.x']+7.5)/15\n",
        "dataset['rotationRate.y'] = (dataset['rotationRate.y']+7.5)/15\n",
        "dataset['rotationRate.z'] = (dataset['rotationRate.z']+7.5)/15\n",
        "\n",
        "np_dataset[:,3:6] = (np_dataset[:,3:6]+7.5)/15\n",
        "\n",
        "# For accelerometer\n",
        "dataset['userAcceleration.x'] = (dataset['userAcceleration.x']+4)/8\n",
        "dataset['userAcceleration.y'] = (dataset['userAcceleration.y']+4)/8\n",
        "dataset['userAcceleration.z'] = (dataset['userAcceleration.z']+4)/8\n",
        "\n",
        "np_dataset[:,6:9] = (np_dataset[:,6:9]+4)/8"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KXz7rMMT_vb4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "86220a48-4dc3-44f0-98da-a080fcb63c37"
      },
      "source": [
        "np_dataset.shape"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1123724, 10)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HK649qtXP2Xq",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 71
        },
        "outputId": "4be57285-9556-4e55-cf95-a48c83920079"
      },
      "source": [
        "\n",
        "def time_series_to_section(dataset, num_act_labels, sliding_window_size, step_size_of_sliding_window):\n",
        "    data = dataset[: , 0:-(num_act_labels)]         #(1123724, 9)\n",
        "    act_labels = dataset[: , -num_act_labels: ]     #(1123724, 1)    \n",
        "\n",
        "    ## We want the Rows of matrices show each Feature and the Columns show time points.\n",
        "    data = data.T    #(9, 1123724))\n",
        "\n",
        "    size_features = data.shape[0]     #9\n",
        "    size_data = data.shape[1]         #1123724\n",
        "    number_of_secs = round(((size_data - sliding_window_size)/step_size_of_sliding_window)) #112367 steps \n",
        "    # ... before all the data can be traversed \n",
        "            \n",
        "    ##  Create a 3D matrix for Storing Snapshots  \n",
        "    ## set of training inputs (x)\n",
        "    secs_data = np.zeros((number_of_secs , size_features , sliding_window_size )) #(112367, 9, 50)\n",
        "    ## set of training outputs (f(x))\n",
        "    act_secs_labels = np.zeros((number_of_secs, 1)) #(112367, 1)\n",
        "    \n",
        "    k=0    \n",
        "    for i in range(0 ,(size_data)-sliding_window_size  , step_size_of_sliding_window): # range(0, 1123674, 10)\n",
        "        j = i // step_size_of_sliding_window #Using floor division () , j = 0, 1 , ...112365, 112366, 112367. (last elem = 112367 because = round(1123674/10) )\n",
        "        if(j>=number_of_secs):      # \n",
        "            break\n",
        "        ### Note: \n",
        "        if(not (act_labels[i] == act_labels[i+sliding_window_size-1]).all()): \n",
        "            continue    \n",
        "        \n",
        "        secs_data[k] = data[0:size_features, i:i+sliding_window_size]   # data[9, 0:50]\n",
        "        act_secs_labels[k] = act_labels[i].astype(int)\n",
        "        k = k+1\n",
        "    secs_data = secs_data[0:k]\n",
        "    act_secs_labels = act_secs_labels[0:k]\n",
        "    \n",
        "    return secs_data, act_secs_labels\n",
        "\n",
        "## This Variable Defines the Size of Sliding Window\n",
        "## ( e.g. 100 means in each snapshot we just consider 100 consecutive observations of each sensor) \n",
        "sliding_window_size = 50 # 50 Equals to 1 second for MotionSense Dataset (it is on 50Hz samplig rate)\n",
        "## Here We Choose Step Size for Building Diffrent Snapshots from Time-Series Data\n",
        "## ( smaller step size will increase the amount of the instances and higher computational cost may be incurred )\n",
        "step_size_of_sliding_window = 10 \n",
        "num_act_labels = 1 # All 4 classes [walk, sit, jog, stand (but in order)] in single column \n",
        "num_features = 9\n",
        "\n",
        "print(\"--> Sectioning Training and Test datasets: shape of each section will be: (\",num_features,\"x\",sliding_window_size,\")\")\n",
        "train_data, act_train_labels = time_series_to_section(np_dataset.copy(),\n",
        "                                                      num_act_labels,\n",
        "                                                      sliding_window_size,\n",
        "                                                      step_size_of_sliding_window)\n",
        "                                                                                                                                                                                             \n",
        "print(\"--> Shape of Training Sections:\", train_data.shape)"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--> Sectioning Training and Test datasets: shape of each section will be: ( 9 x 50 )\n",
            "--> Shape of Training Sections: (111897, 9, 50)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jd3gBbpQiBfG",
        "colab_type": "text"
      },
      "source": [
        "## Split data into train and validation\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bEwSXLmz_Oa9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "619f57d7-1300-4da8-fb3b-8a4fdb87f2e1"
      },
      "source": [
        "# Note: All the data should be \"float32\" ! For converting into TF Lite format\n",
        "## Let's check our data\n",
        "train_data.dtype\n",
        "\n",
        "train_data = train_data.astype('float32')\n",
        "train_data.dtype"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "dtype('float32')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rO4i4rnZiMo2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "# make the x for train and test (also called validation data) \n",
        "xtrain,xtest,ytrain,ytest = train_test_split(train_data,act_train_labels,\n",
        "                                             train_size=0.7,     # 70 % : Train\n",
        "                                             random_state=123)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dZT3aE2JLze7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "c3d460f9-ad6b-4def-86f3-2da9d45d09c8"
      },
      "source": [
        "print(\"Train input shape: \",xtrain.shape)\n",
        "print(\"Train output shape: \",ytrain.shape)"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train input shape:  (78327, 9, 50)\n",
            "Train output shape:  (78327, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "70TBJIzRqYdB",
        "colab_type": "text"
      },
      "source": [
        "## Start Training\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3B-ZywO8mjiV",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 377
        },
        "outputId": "2edd8682-9465-4ba5-e285-ea51f9902239"
      },
      "source": [
        "import tensorflow as tf\n",
        "\n",
        "model = tf.keras.Sequential([\n",
        "    tf.keras.layers.Flatten(input_shape=(9, 50)),\n",
        "    tf.keras.layers.Dense(128, activation='relu'),\n",
        "    tf.keras.layers.Dense(4)\n",
        "])\n",
        "\n",
        "model.compile(optimizer='adam',\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "Epochs = 10\n",
        "history = model.fit(xtrain, ytrain, epochs=10)\n"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "2448/2448 [==============================] - 6s 3ms/step - loss: 0.2878 - accuracy: 0.8934\n",
            "Epoch 2/10\n",
            "2448/2448 [==============================] - 6s 3ms/step - loss: 0.0964 - accuracy: 0.9732\n",
            "Epoch 3/10\n",
            "2448/2448 [==============================] - 6s 3ms/step - loss: 0.0792 - accuracy: 0.9780\n",
            "Epoch 4/10\n",
            "2448/2448 [==============================] - 6s 3ms/step - loss: 0.0727 - accuracy: 0.9804\n",
            "Epoch 5/10\n",
            "2448/2448 [==============================] - 6s 2ms/step - loss: 0.0682 - accuracy: 0.9820\n",
            "Epoch 6/10\n",
            "2448/2448 [==============================] - 6s 3ms/step - loss: 0.0644 - accuracy: 0.9838\n",
            "Epoch 7/10\n",
            "2448/2448 [==============================] - 6s 3ms/step - loss: 0.0626 - accuracy: 0.9838\n",
            "Epoch 8/10\n",
            "2448/2448 [==============================] - 6s 3ms/step - loss: 0.0604 - accuracy: 0.9845\n",
            "Epoch 9/10\n",
            "2448/2448 [==============================] - 7s 3ms/step - loss: 0.0582 - accuracy: 0.9850\n",
            "Epoch 10/10\n",
            "2448/2448 [==============================] - 6s 3ms/step - loss: 0.0568 - accuracy: 0.9851\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V1U9wepopzmk",
        "colab_type": "text"
      },
      "source": [
        "## Visualize the model performance"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "plc2jrLNDLQR",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "60c5db45-4397-419c-97ac-3e40db482aa0"
      },
      "source": [
        "test_loss, test_acc = model.evaluate(xtest,  ytest, verbose=2)\n",
        "\n",
        "print('\\nTest accuracy:', test_acc)"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1050/1050 - 2s - loss: 0.0503 - accuracy: 0.9887\n",
            "\n",
            "Test accuracy: 0.9887101650238037\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8Rg7tlsBsyyW",
        "colab_type": "text"
      },
      "source": [
        "## Convert model to TF Lite format"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z-p0amAlpl5m",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "80563233-62e0-4a02-e1b6-04bac26d5839"
      },
      "source": [
        "# Convert the model to the TensorFlow Lite format without quantization\n",
        "converter = tf.lite.TFLiteConverter.from_keras_model(model)\n",
        "tflite_model = converter.convert()\n",
        "\n",
        "# Save the model\n",
        "open(\"act_tracker.tflite\", \"wb\").write(tflite_model)"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "234180"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 56
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R-Oo9sV3s6IR",
        "colab_type": "text"
      },
      "source": [
        "## Encode in Header file (as byte array)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pI9NFE3Ds_Mg",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "e0886d6a-1ed3-4677-cb0e-dc800035b85a"
      },
      "source": [
        "!echo \"const unsigned char model[] = {\" > /content/model.h\n",
        "!cat act_tracker.tflite | xxd -i      >> /content/model.h\n",
        "!echo \"};\"                              >> /content/model.h\n",
        "\n",
        "import os\n",
        "model_h_size = os.path.getsize(\"model.h\")\n",
        "print(f\"Header file, model.h, is {model_h_size:,} bytes.\")\n",
        "\n",
        "basic_model_size = os.path.getsize(\"act_tracker.tflite\")\n",
        "print(\"TFLite Model is %d bytes\" % basic_model_size)\n"
      ],
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/bin/bash: xxd: command not found\n",
            "Header file, model.h, is 35 bytes.\n",
            "TFLite Model is 234180 bytes\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}