ts_tutorial_intro.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ts_tutorial_intro.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyOkO5GivmYZzeWPLJFtv2tK",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/iosband/ts_tutorial/blob/master/ts_tutorial_intro.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FGwbMwRrRduA",
        "colab_type": "text"
      },
      "source": [
        "# A Tutorial on Thompson Sampling\n",
        "\n",
        "This colab notebook provides a way to play around with the code for [A Tutorial on Thompson Sampling](https://arxiv.org/abs/1707.02038) without needing to install anything.\n",
        "\n",
        "All of the underlying code is taken from the public git repository:\n",
        "https://github.com/iosband/ts_tutorial\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fzTeLfcNRoCH",
        "colab_type": "text"
      },
      "source": [
        "## Cloning the respository and setting up the environment\n",
        "\n",
        "Our code is made to work with Python 3.8 but would likely work with other versions.\n",
        "\n",
        "The commands below allow the package to be run in a default Google colab kernel."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IFr07q33QDWE",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#@title Cloning github repository and installing requied packaged\n",
        "%%capture\n",
        "!git clone https://github.com/iosband/ts_tutorial.git\n",
        "!pip install plotnine\n",
        "!pip install cvxpy"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Uw8CWF9pQST8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "d3436e98-5c01-46c3-eed2-0be270e23524"
      },
      "source": [
        "#@title Code assumes you are running from `ts_tutorial/src` directory\n",
        "cd ts_tutorial/src"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/ts_tutorial/src\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IlqufejeRsbb",
        "colab_type": "text"
      },
      "source": [
        "## Reproducing specific figures\n",
        "\n",
        "We present the exact code used to generate the figures for each plot in `reproduce_figures.py`.\n",
        "\n",
        "If you run this script with appropriate flags, you can reproduce the figures in the paper."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7Qv9kMolRt2R",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 323
        },
        "outputId": "86d9b267-b6c2-4e9c-c3ba-d28d19c0368d"
      },
      "source": [
        "!python reproduce_figures.py --help"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "usage: reproduce_figures.py [-h] [--figure FIGURE] [--run_frac RUN_FRAC]\n",
            "                            [--data_path DATA_PATH] [--plot_path PLOT_PATH]\n",
            "\n",
            "Reproduce figures.\n",
            "\n",
            "optional arguments:\n",
            "  -h, --help            show this help message and exit\n",
            "  --figure FIGURE       Figures to reproduce. Must be one of the following\n",
            "                        options: ['3.1', '3.2a', '3.2b', '4.1a', '4.1b',\n",
            "                        '4.3a', '4.3b', '5.1', '5.2a', '5.2b', '6.2', '6.3',\n",
            "                        '6.4', '7.1', '7.2', '7.3', '7.4', '7.5', 'all']\n",
            "  --run_frac RUN_FRAC   Proportion of paper experiments to run. Must be in [0,\n",
            "                        1].\n",
            "  --data_path DATA_PATH\n",
            "                        Path to store intermediate .csv files of experiment\n",
            "                        results. Must exist in OS.\n",
            "  --plot_path PLOT_PATH\n",
            "                        Path to store output paper plots. Must exist in OS.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4nmdGontT_3c",
        "colab_type": "text"
      },
      "source": [
        "### Example = figure 3.1\n",
        "\n",
        "The code below will reproduces Figure 3.1 with 1% of the seeds of the paper, save output to /tmp/.\n",
        "\n",
        "We choose to run only 1% of the seeds in the interest of computational speed."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RUI3Ee3YSgs6",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "c44dab1f-e04b-4000-f257-41ec304b041b"
      },
      "source": [
        "!python reproduce_figures.py --figure=3.1 --run_frac=0.01 --data_path=/tmp/ --plot_path=/tmp/"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "********************************************************************************\n",
            "Parsing Namespace(data_path='/tmp/', figure='3.1', plot_path='/tmp/', run_frac=0.01)\n",
            "WARNING - this can take a long time on a single machine... you may want to parallelize the jobs.\n",
            "\n",
            "********************************************************************************\n",
            "Reproducing Figure 3.1, from TS Tutorial https://arxiv.org/abs/1707.02038\n",
            "With run_frac 0.01 this will launch 200 jobs, compared to 20000 in the paper.\n",
            "The config file with all necessary details of the underlying experiment is \n",
            "   finite_arm.config_simple.\n",
            "The experiment results are saved to /tmp/ with experiment_id finite_simple.\n",
            "The output plots are saved to /tmp/ with experiment_id finite_simple.\n",
            "\n",
            "\n",
            "********************************************************************************\n",
            "Starting job 0 out of 200\n",
            "Starting job 1 out of 200\n",
            "Starting job 2 out of 200\n",
            "Starting job 3 out of 200\n",
            "Starting job 4 out of 200\n",
            "Starting job 5 out of 200\n",
            "Starting job 6 out of 200\n",
            "Starting job 7 out of 200\n",
            "Starting job 8 out of 200\n",
            "Starting job 9 out of 200\n",
            "Starting job 10 out of 200\n",
            "Starting job 11 out of 200\n",
            "Starting job 12 out of 200\n",
            "Starting job 13 out of 200\n",
            "Starting job 14 out of 200\n",
            "Starting job 15 out of 200\n",
            "Starting job 16 out of 200\n",
            "Starting job 17 out of 200\n",
            "Starting job 18 out of 200\n",
            "Starting job 19 out of 200\n",
            "Starting job 20 out of 200\n",
            "Starting job 21 out of 200\n",
            "Starting job 22 out of 200\n",
            "Starting job 23 out of 200\n",
            "Starting job 24 out of 200\n",
            "Starting job 25 out of 200\n",
            "Starting job 26 out of 200\n",
            "Starting job 27 out of 200\n",
            "Starting job 28 out of 200\n",
            "Starting job 29 out of 200\n",
            "Starting job 30 out of 200\n",
            "Starting job 31 out of 200\n",
            "Starting job 32 out of 200\n",
            "Starting job 33 out of 200\n",
            "Starting job 34 out of 200\n",
            "Starting job 35 out of 200\n",
            "Starting job 36 out of 200\n",
            "Starting job 37 out of 200\n",
            "Starting job 38 out of 200\n",
            "Starting job 39 out of 200\n",
            "Starting job 40 out of 200\n",
            "Starting job 41 out of 200\n",
            "Starting job 42 out of 200\n",
            "Starting job 43 out of 200\n",
            "Starting job 44 out of 200\n",
            "Starting job 45 out of 200\n",
            "Starting job 46 out of 200\n",
            "Starting job 47 out of 200\n",
            "Starting job 48 out of 200\n",
            "Starting job 49 out of 200\n",
            "Starting job 50 out of 200\n",
            "Starting job 51 out of 200\n",
            "Starting job 52 out of 200\n",
            "Starting job 53 out of 200\n",
            "Starting job 54 out of 200\n",
            "Starting job 55 out of 200\n",
            "Starting job 56 out of 200\n",
            "Starting job 57 out of 200\n",
            "Starting job 58 out of 200\n",
            "Starting job 59 out of 200\n",
            "Starting job 60 out of 200\n",
            "Starting job 61 out of 200\n",
            "Starting job 62 out of 200\n",
            "Starting job 63 out of 200\n",
            "Starting job 64 out of 200\n",
            "Starting job 65 out of 200\n",
            "Starting job 66 out of 200\n",
            "Starting job 67 out of 200\n",
            "Starting job 68 out of 200\n",
            "Starting job 69 out of 200\n",
            "Starting job 70 out of 200\n",
            "Starting job 71 out of 200\n",
            "Starting job 72 out of 200\n",
            "Starting job 73 out of 200\n",
            "Starting job 74 out of 200\n",
            "Starting job 75 out of 200\n",
            "Starting job 76 out of 200\n",
            "Starting job 77 out of 200\n",
            "Starting job 78 out of 200\n",
            "Starting job 79 out of 200\n",
            "Starting job 80 out of 200\n",
            "Starting job 81 out of 200\n",
            "Starting job 82 out of 200\n",
            "Starting job 83 out of 200\n",
            "Starting job 84 out of 200\n",
            "Starting job 85 out of 200\n",
            "Starting job 86 out of 200\n",
            "Starting job 87 out of 200\n",
            "Starting job 88 out of 200\n",
            "Starting job 89 out of 200\n",
            "Starting job 90 out of 200\n",
            "Starting job 91 out of 200\n",
            "Starting job 92 out of 200\n",
            "Starting job 93 out of 200\n",
            "Starting job 94 out of 200\n",
            "Starting job 95 out of 200\n",
            "Starting job 96 out of 200\n",
            "Starting job 97 out of 200\n",
            "Starting job 98 out of 200\n",
            "Starting job 99 out of 200\n",
            "Starting job 100 out of 200\n",
            "Starting job 101 out of 200\n",
            "Starting job 102 out of 200\n",
            "Starting job 103 out of 200\n",
            "Starting job 104 out of 200\n",
            "Starting job 105 out of 200\n",
            "Starting job 106 out of 200\n",
            "Starting job 107 out of 200\n",
            "Starting job 108 out of 200\n",
            "Starting job 109 out of 200\n",
            "Starting job 110 out of 200\n",
            "Starting job 111 out of 200\n",
            "Starting job 112 out of 200\n",
            "Starting job 113 out of 200\n",
            "Starting job 114 out of 200\n",
            "Starting job 115 out of 200\n",
            "Starting job 116 out of 200\n",
            "Starting job 117 out of 200\n",
            "Starting job 118 out of 200\n",
            "Starting job 119 out of 200\n",
            "Starting job 120 out of 200\n",
            "Starting job 121 out of 200\n",
            "Starting job 122 out of 200\n",
            "Starting job 123 out of 200\n",
            "Starting job 124 out of 200\n",
            "Starting job 125 out of 200\n",
            "Starting job 126 out of 200\n",
            "Starting job 127 out of 200\n",
            "Starting job 128 out of 200\n",
            "Starting job 129 out of 200\n",
            "Starting job 130 out of 200\n",
            "Starting job 131 out of 200\n",
            "Starting job 132 out of 200\n",
            "Starting job 133 out of 200\n",
            "Starting job 134 out of 200\n",
            "Starting job 135 out of 200\n",
            "Starting job 136 out of 200\n",
            "Starting job 137 out of 200\n",
            "Starting job 138 out of 200\n",
            "Starting job 139 out of 200\n",
            "Starting job 140 out of 200\n",
            "Starting job 141 out of 200\n",
            "Starting job 142 out of 200\n",
            "Starting job 143 out of 200\n",
            "Starting job 144 out of 200\n",
            "Starting job 145 out of 200\n",
            "Starting job 146 out of 200\n",
            "Starting job 147 out of 200\n",
            "Starting job 148 out of 200\n",
            "Starting job 149 out of 200\n",
            "Starting job 150 out of 200\n",
            "Starting job 151 out of 200\n",
            "Starting job 152 out of 200\n",
            "Starting job 153 out of 200\n",
            "Starting job 154 out of 200\n",
            "Starting job 155 out of 200\n",
            "Starting job 156 out of 200\n",
            "Starting job 157 out of 200\n",
            "Starting job 158 out of 200\n",
            "Starting job 159 out of 200\n",
            "Starting job 160 out of 200\n",
            "Starting job 161 out of 200\n",
            "Starting job 162 out of 200\n",
            "Starting job 163 out of 200\n",
            "Starting job 164 out of 200\n",
            "Starting job 165 out of 200\n",
            "Starting job 166 out of 200\n",
            "Starting job 167 out of 200\n",
            "Starting job 168 out of 200\n",
            "Starting job 169 out of 200\n",
            "Starting job 170 out of 200\n",
            "Starting job 171 out of 200\n",
            "Starting job 172 out of 200\n",
            "Starting job 173 out of 200\n",
            "Starting job 174 out of 200\n",
            "Starting job 175 out of 200\n",
            "Starting job 176 out of 200\n",
            "Starting job 177 out of 200\n",
            "Starting job 178 out of 200\n",
            "Starting job 179 out of 200\n",
            "Starting job 180 out of 200\n",
            "Starting job 181 out of 200\n",
            "Starting job 182 out of 200\n",
            "Starting job 183 out of 200\n",
            "Starting job 184 out of 200\n",
            "Starting job 185 out of 200\n",
            "Starting job 186 out of 200\n",
            "Starting job 187 out of 200\n",
            "Starting job 188 out of 200\n",
            "Starting job 189 out of 200\n",
            "Starting job 190 out of 200\n",
            "Starting job 191 out of 200\n",
            "Starting job 192 out of 200\n",
            "Starting job 193 out of 200\n",
            "Starting job 194 out of 200\n",
            "Starting job 195 out of 200\n",
            "Starting job 196 out of 200\n",
            "Starting job 197 out of 200\n",
            "Starting job 198 out of 200\n",
            "Starting job 199 out of 200\n",
            "********************************************************************************\n",
            "Saving final plot to /tmp/finite_simple_ts_action.png\n",
            "This may not *precisely* match the paper due to run_frac 0.01\n",
            "********************************************************************************\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "/usr/local/lib/python3.6/dist-packages/plotnine/ggplot.py:729: PlotnineWarning: Saving 8 x 8 in image.\n",
            "  from_inches(height, units), units), PlotnineWarning)\n",
            "/usr/local/lib/python3.6/dist-packages/plotnine/ggplot.py:730: PlotnineWarning: Filename: /tmp/finite_simple_ts_action.png\n",
            "  warn('Filename: {}'.format(filename), PlotnineWarning)\n",
            "********************************************************************************\n",
            "Saving final plot to /tmp/finite_simple_greedy_action.png\n",
            "This may not *precisely* match the paper due to run_frac 0.01\n",
            "********************************************************************************\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "/usr/local/lib/python3.6/dist-packages/plotnine/ggplot.py:729: PlotnineWarning: Saving 8 x 8 in image.\n",
            "  from_inches(height, units), units), PlotnineWarning)\n",
            "/usr/local/lib/python3.6/dist-packages/plotnine/ggplot.py:730: PlotnineWarning: Filename: /tmp/finite_simple_greedy_action.png\n",
            "  warn('Filename: {}'.format(filename), PlotnineWarning)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mFsINE9SS1oo",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "3bd47c57-e6ff-4229-81fc-a468034179da"
      },
      "source": [
        "#@title All of the experimental data + plots have been saved to /tmp/\n",
        "ls /tmp/"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "'exp=finite_simple|id=0.csv'    'exp=finite_simple|id=192.csv'\n",
            "'exp=finite_simple|id=100.csv'  'exp=finite_simple|id=193.csv'\n",
            "'exp=finite_simple|id=101.csv'  'exp=finite_simple|id=194.csv'\n",
            "'exp=finite_simple|id=102.csv'  'exp=finite_simple|id=195.csv'\n",
            "'exp=finite_simple|id=103.csv'  'exp=finite_simple|id=196.csv'\n",
            "'exp=finite_simple|id=104.csv'  'exp=finite_simple|id=197.csv'\n",
            "'exp=finite_simple|id=105.csv'  'exp=finite_simple|id=198.csv'\n",
            "'exp=finite_simple|id=106.csv'  'exp=finite_simple|id=199.csv'\n",
            "'exp=finite_simple|id=107.csv'  'exp=finite_simple|id=19.csv'\n",
            "'exp=finite_simple|id=108.csv'  'exp=finite_simple|id=1.csv'\n",
            "'exp=finite_simple|id=109.csv'  'exp=finite_simple|id=20.csv'\n",
            "'exp=finite_simple|id=10.csv'   'exp=finite_simple|id=21.csv'\n",
            "'exp=finite_simple|id=110.csv'  'exp=finite_simple|id=22.csv'\n",
            "'exp=finite_simple|id=111.csv'  'exp=finite_simple|id=23.csv'\n",
            "'exp=finite_simple|id=112.csv'  'exp=finite_simple|id=24.csv'\n",
            "'exp=finite_simple|id=113.csv'  'exp=finite_simple|id=25.csv'\n",
            "'exp=finite_simple|id=114.csv'  'exp=finite_simple|id=26.csv'\n",
            "'exp=finite_simple|id=115.csv'  'exp=finite_simple|id=27.csv'\n",
            "'exp=finite_simple|id=116.csv'  'exp=finite_simple|id=28.csv'\n",
            "'exp=finite_simple|id=117.csv'  'exp=finite_simple|id=29.csv'\n",
            "'exp=finite_simple|id=118.csv'  'exp=finite_simple|id=2.csv'\n",
            "'exp=finite_simple|id=119.csv'  'exp=finite_simple|id=30.csv'\n",
            "'exp=finite_simple|id=11.csv'   'exp=finite_simple|id=31.csv'\n",
            "'exp=finite_simple|id=120.csv'  'exp=finite_simple|id=32.csv'\n",
            "'exp=finite_simple|id=121.csv'  'exp=finite_simple|id=33.csv'\n",
            "'exp=finite_simple|id=122.csv'  'exp=finite_simple|id=34.csv'\n",
            "'exp=finite_simple|id=123.csv'  'exp=finite_simple|id=35.csv'\n",
            "'exp=finite_simple|id=124.csv'  'exp=finite_simple|id=36.csv'\n",
            "'exp=finite_simple|id=125.csv'  'exp=finite_simple|id=37.csv'\n",
            "'exp=finite_simple|id=126.csv'  'exp=finite_simple|id=38.csv'\n",
            "'exp=finite_simple|id=127.csv'  'exp=finite_simple|id=39.csv'\n",
            "'exp=finite_simple|id=128.csv'  'exp=finite_simple|id=3.csv'\n",
            "'exp=finite_simple|id=129.csv'  'exp=finite_simple|id=40.csv'\n",
            "'exp=finite_simple|id=12.csv'   'exp=finite_simple|id=41.csv'\n",
            "'exp=finite_simple|id=130.csv'  'exp=finite_simple|id=42.csv'\n",
            "'exp=finite_simple|id=131.csv'  'exp=finite_simple|id=43.csv'\n",
            "'exp=finite_simple|id=132.csv'  'exp=finite_simple|id=44.csv'\n",
            "'exp=finite_simple|id=133.csv'  'exp=finite_simple|id=45.csv'\n",
            "'exp=finite_simple|id=134.csv'  'exp=finite_simple|id=46.csv'\n",
            "'exp=finite_simple|id=135.csv'  'exp=finite_simple|id=47.csv'\n",
            "'exp=finite_simple|id=136.csv'  'exp=finite_simple|id=48.csv'\n",
            "'exp=finite_simple|id=137.csv'  'exp=finite_simple|id=49.csv'\n",
            "'exp=finite_simple|id=138.csv'  'exp=finite_simple|id=4.csv'\n",
            "'exp=finite_simple|id=139.csv'  'exp=finite_simple|id=50.csv'\n",
            "'exp=finite_simple|id=13.csv'   'exp=finite_simple|id=51.csv'\n",
            "'exp=finite_simple|id=140.csv'  'exp=finite_simple|id=52.csv'\n",
            "'exp=finite_simple|id=141.csv'  'exp=finite_simple|id=53.csv'\n",
            "'exp=finite_simple|id=142.csv'  'exp=finite_simple|id=54.csv'\n",
            "'exp=finite_simple|id=143.csv'  'exp=finite_simple|id=55.csv'\n",
            "'exp=finite_simple|id=144.csv'  'exp=finite_simple|id=56.csv'\n",
            "'exp=finite_simple|id=145.csv'  'exp=finite_simple|id=57.csv'\n",
            "'exp=finite_simple|id=146.csv'  'exp=finite_simple|id=58.csv'\n",
            "'exp=finite_simple|id=147.csv'  'exp=finite_simple|id=59.csv'\n",
            "'exp=finite_simple|id=148.csv'  'exp=finite_simple|id=5.csv'\n",
            "'exp=finite_simple|id=149.csv'  'exp=finite_simple|id=60.csv'\n",
            "'exp=finite_simple|id=14.csv'   'exp=finite_simple|id=61.csv'\n",
            "'exp=finite_simple|id=150.csv'  'exp=finite_simple|id=62.csv'\n",
            "'exp=finite_simple|id=151.csv'  'exp=finite_simple|id=63.csv'\n",
            "'exp=finite_simple|id=152.csv'  'exp=finite_simple|id=64.csv'\n",
            "'exp=finite_simple|id=153.csv'  'exp=finite_simple|id=65.csv'\n",
            "'exp=finite_simple|id=154.csv'  'exp=finite_simple|id=66.csv'\n",
            "'exp=finite_simple|id=155.csv'  'exp=finite_simple|id=67.csv'\n",
            "'exp=finite_simple|id=156.csv'  'exp=finite_simple|id=68.csv'\n",
            "'exp=finite_simple|id=157.csv'  'exp=finite_simple|id=69.csv'\n",
            "'exp=finite_simple|id=158.csv'  'exp=finite_simple|id=6.csv'\n",
            "'exp=finite_simple|id=159.csv'  'exp=finite_simple|id=70.csv'\n",
            "'exp=finite_simple|id=15.csv'   'exp=finite_simple|id=71.csv'\n",
            "'exp=finite_simple|id=160.csv'  'exp=finite_simple|id=72.csv'\n",
            "'exp=finite_simple|id=161.csv'  'exp=finite_simple|id=73.csv'\n",
            "'exp=finite_simple|id=162.csv'  'exp=finite_simple|id=74.csv'\n",
            "'exp=finite_simple|id=163.csv'  'exp=finite_simple|id=75.csv'\n",
            "'exp=finite_simple|id=164.csv'  'exp=finite_simple|id=76.csv'\n",
            "'exp=finite_simple|id=165.csv'  'exp=finite_simple|id=77.csv'\n",
            "'exp=finite_simple|id=166.csv'  'exp=finite_simple|id=78.csv'\n",
            "'exp=finite_simple|id=167.csv'  'exp=finite_simple|id=79.csv'\n",
            "'exp=finite_simple|id=168.csv'  'exp=finite_simple|id=7.csv'\n",
            "'exp=finite_simple|id=169.csv'  'exp=finite_simple|id=80.csv'\n",
            "'exp=finite_simple|id=16.csv'   'exp=finite_simple|id=81.csv'\n",
            "'exp=finite_simple|id=170.csv'  'exp=finite_simple|id=82.csv'\n",
            "'exp=finite_simple|id=171.csv'  'exp=finite_simple|id=83.csv'\n",
            "'exp=finite_simple|id=172.csv'  'exp=finite_simple|id=84.csv'\n",
            "'exp=finite_simple|id=173.csv'  'exp=finite_simple|id=85.csv'\n",
            "'exp=finite_simple|id=174.csv'  'exp=finite_simple|id=86.csv'\n",
            "'exp=finite_simple|id=175.csv'  'exp=finite_simple|id=87.csv'\n",
            "'exp=finite_simple|id=176.csv'  'exp=finite_simple|id=88.csv'\n",
            "'exp=finite_simple|id=177.csv'  'exp=finite_simple|id=89.csv'\n",
            "'exp=finite_simple|id=178.csv'  'exp=finite_simple|id=8.csv'\n",
            "'exp=finite_simple|id=179.csv'  'exp=finite_simple|id=90.csv'\n",
            "'exp=finite_simple|id=17.csv'   'exp=finite_simple|id=91.csv'\n",
            "'exp=finite_simple|id=180.csv'  'exp=finite_simple|id=92.csv'\n",
            "'exp=finite_simple|id=181.csv'  'exp=finite_simple|id=93.csv'\n",
            "'exp=finite_simple|id=182.csv'  'exp=finite_simple|id=94.csv'\n",
            "'exp=finite_simple|id=183.csv'  'exp=finite_simple|id=95.csv'\n",
            "'exp=finite_simple|id=184.csv'  'exp=finite_simple|id=96.csv'\n",
            "'exp=finite_simple|id=185.csv'  'exp=finite_simple|id=97.csv'\n",
            "'exp=finite_simple|id=186.csv'  'exp=finite_simple|id=98.csv'\n",
            "'exp=finite_simple|id=187.csv'  'exp=finite_simple|id=99.csv'\n",
            "'exp=finite_simple|id=188.csv'  'exp=finite_simple|id=9.csv'\n",
            "'exp=finite_simple|id=189.csv'  'exp=finite_simple|params.csv'\n",
            "'exp=finite_simple|id=18.csv'    finite_simple_greedy_action.png\n",
            "'exp=finite_simple|id=190.csv'   finite_simple_ts_action.png\n",
            "'exp=finite_simple|id=191.csv'\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cwYE3c1QSlmj",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 711
        },
        "outputId": "6ecd8e75-fdaa-43ab-dbbf-829127fe3d29"
      },
      "source": [
        "#@title Action selections for the greedy agent\n",
        "from IPython import display\n",
        "display.Image('/tmp/finite_simple_greedy_action.png')"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r5RLqr3hT5gZ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 711
        },
        "outputId": "f6aabf2b-2e41-4bc9-82bb-17a62eb13de7"
      },
      "source": [
        "#@title Action selection for the TS agent\n",
        "display.Image('/tmp/finite_simple_ts_action.png')"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GKW4uVk5UwaJ",
        "colab_type": "text"
      },
      "source": [
        "Note that these results do not *exactly* reproduce the paper due to the smaller number of seeds (for speed)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "50peu_XURvZd",
        "colab_type": "text"
      },
      "source": [
        "## Understanding the underlying experiment code\n",
        "\n",
        "This section runs through the `local_runner.py` and shows how the underlying code is used.\n",
        "\n",
        "Agent settings are defined in a `config` file, and then the \"runner\" loops through the different agents and evaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bNgaqVETQZA9",
        "colab_type": "code",
        "colab": {},
        "cellView": "both"
      },
      "source": [
        "#@title Imports (double-click to show/hide)\n",
        "import importlib\n",
        "import os\n",
        "import sys\n",
        "from base import config_lib\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import plotnine as gg\n"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SwH2BbE30ihZ",
        "colab_type": "text"
      },
      "source": [
        "The list of experiment configs in the paper is given:\n",
        "\n",
        "```\n",
        "['finite_arm.config_simple', 'finite_arm.config_simple_rand', 'finite_arm.config_simple_sanity', 'finite_arm.config_misspecified', 'finite_arm.config_drift', 'cascading.config_cascading', 'ensemble_nn.config_nn', 'graph.config_indep', 'graph.config_indep_binary', 'graph.config_correlated', 'graph.config_correlated_sanity']\n",
        "```\n",
        "\n",
        "You can explore each setting in the relevant file of source code."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fEPDW5NQQero",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#@title Set these variables to run from pre-populated config\n",
        "CONFIG_PATH = 'finite_arm.config_simple' #@param\n",
        "N_JOBS = 100 #@param"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FPiI--uXQbgA",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#@title Pulling in that config file and running these agents\n",
        "# Running from a local config file\n",
        "sys.path.append(os.getcwd())\n",
        "\n",
        "# Loading in the experiment config file\n",
        "config_module = importlib.import_module(CONFIG_PATH)\n",
        "config = config_module.get_config()\n",
        "\n",
        "results = []\n",
        "for job_id in range(N_JOBS):\n",
        "  # Running the experiment.\n",
        "  job_config = config_lib.get_job_config(config, job_id)\n",
        "  experiment = job_config['experiment']\n",
        "  experiment.run_experiment()\n",
        "  results.append(experiment.results)\n"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "C2uf5W4MQgUN",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 728
        },
        "outputId": "6e0dea32-81d0-4af1-f44b-cb0ad30642fa"
      },
      "source": [
        "#@title Plotting and analysis (uses plotnine by default)\n",
        "\n",
        "# Collating data from experiments results with Pandas\n",
        "params_df = config_lib.get_params_df(config)\n",
        "df = pd.merge(pd.concat(results), params_df, on='unique_id')\n",
        "plt_df = (df.groupby(['agent', 't'])\n",
        "          .agg({'instant_regret': np.mean})\n",
        "          .reset_index())\n",
        "\n",
        "# Set up a simple plot\n",
        "gg.theme_set(gg.theme_bw(base_size=16, base_family='serif'))\n",
        "gg.theme_update(figure_size=(12, 8))\n",
        "\n",
        "p = (gg.ggplot(plt_df)\n",
        "     + gg.aes('t', 'instant_regret', colour='agent')\n",
        "     + gg.geom_line())\n",
        "print(p)"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1200x800 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "<ggplot: (8749007193769)>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JbjR8adz1Bwb",
        "colab_type": "text"
      },
      "source": [
        "Code for analysing and plotting agents is given in `base/plot.py` and if you use `reproduce_figures.py` will be called automatically."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6qzAdGQyVGEs",
        "colab_type": "text"
      },
      "source": [
        "## Running agents and environments without a config\n",
        "\n",
        "\n",
        "Our code is meant to be as simple as possible, optimizing for ease of use rather than speed.\n",
        "\n",
        "We break the experiments/algorithms into three key parts:\n",
        "\n",
        "- `agent`: handles learning + action selection of a bandit algorithm.\n",
        "- `environment`: evaluates rewards/outcomes from an action.\n",
        "- `experiment`: contains agent, environment and seed."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Dm_BCAq4VSwx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#@title Running a very simple agent on a finite armed bandit\n",
        "from base.experiment import BaseExperiment\n",
        "from finite_arm.agent_finite import FiniteBernoulliBanditTS\n",
        "from finite_arm.env_finite import FiniteArmedBernoulliBandit\n",
        "\n",
        "probs = [0.7, 0.8, 0.9]  #@param\n",
        "n_steps = 1000  #@param\n",
        "seed = 0  #@param\n",
        "\n",
        "agent = FiniteBernoulliBanditTS(n_arm=len(probs))\n",
        "env = FiniteArmedBernoulliBandit(probs)\n",
        "experiment = BaseExperiment(\n",
        "    agent, env, n_steps=n_steps, seed=seed, unique_id='example')\n",
        "\n",
        "experiment.run_experiment()\n"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tiJ7dGuwVXiI",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 830
        },
        "outputId": "deb39e2a-c276-4374-c5d1-816ad7f69cd3"
      },
      "source": [
        "#@title Simple display / plot of results\n",
        "\n",
        "print(experiment.results.head())\n",
        "\n",
        "p = (gg.ggplot(experiment.results)\n",
        "     + gg.aes(x='t', y='cum_regret / t', colour='unique_id')\n",
        "     + gg.geom_line())\n",
        "print(p)\n"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   t  instant_regret  cum_regret  action unique_id\n",
            "0  1             0.2         0.2       0   example\n",
            "1  2             0.1         0.3       1   example\n",
            "2  3             0.2         0.5       0   example\n",
            "3  4             0.2         0.7       0   example\n",
            "4  5             0.2         0.9       0   example\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1200x800 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "<ggplot: (8748996455358)>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OQaltHPXz2-r",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}