NFLDataBowl2023.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/fyng/PassplayNET/blob/main/NFLDataBowl2023.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ig9_iIbNgvUD"
      },
      "source": [
        "# Import Kaggle Challenge\n",
        "\n",
        "Refer to https://www.kaggle.com/general/74235"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4iy70FU5hIkw"
      },
      "source": [
        "1. Create and upload kaggle.json"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_8TBKm-Ngrs4"
      },
      "outputs": [],
      "source": [
        "! pip install -q kaggle\n",
        "! mkdir ~/.kaggle\n",
        "! cp kaggle.json ~/.kaggle/\n",
        "! chmod 600 ~/.kaggle/kaggle.json\n",
        "\n",
        "# # check if connection is working\n",
        "# ! kaggle datasets list"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Zcli264bg_qp",
        "outputId": "77f995fd-2d1d-440c-85ba-1a3d78aa68f0"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "nfl-big-data-bowl-2023.zip: Skipping, found more recently modified local copy (use --force to force download)\n",
            "Archive:  /content/nfl-big-data-bowl-2023.zip\n",
            "  inflating: games.csv               \n",
            "  inflating: pffScoutingData.csv     \n",
            "  inflating: players.csv             \n",
            "  inflating: plays.csv               \n",
            "  inflating: week1.csv               \n",
            "  inflating: week2.csv               \n",
            "  inflating: week3.csv               \n",
            "  inflating: week4.csv               \n",
            "  inflating: week5.csv               \n",
            "  inflating: week6.csv               \n",
            "  inflating: week7.csv               \n",
            "  inflating: week8.csv               \n"
          ]
        }
      ],
      "source": [
        "! kaggle competitions download -c nfl-big-data-bowl-2023\n",
        "! unzip /content/nfl-big-data-bowl-2023.zip"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R2VJgapns7aU"
      },
      "source": [
        "How train is made?\n",
        "\n",
        "Every row is a unique play in a unique game\n",
        "\n",
        "Every row contains relevant data of every player. \n",
        "* each variable spans 1 second, spaced with every 1/10 of a second (only 2 NAs)\n",
        "* every record is the same length\n",
        "\n",
        "Action_time based on different events (Mason and Tad)\n",
        "\n",
        "Pass yards (Feiyang & Eric)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_VFcR8u1h2HF"
      },
      "source": [
        "# Explore Data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "l3epmUqZh8BQ"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn import preprocessing"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Iud8dTC42nCj"
      },
      "source": [
        "### Play Data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qnHzuEei27Qh"
      },
      "source": [
        "Play data\n",
        "* `gameId`: Game identifier, unique (numeric)\n",
        "* `playId`: Play identifier, not unique across games (numeric)\n",
        "* `playDescription`: Description of play (text)\n",
        "* `quarter`: Game quarter (numeric)\n",
        "* `down`: Down (numeric)\n",
        "* `yardsToGo`: Distance needed for a first down (numeric)\n",
        "* `possessionTeam`: Team abbr of team on offense with possession of ball (text)\n",
        "* `defensiveTeam`: Team abbr of team on defense (text)\n",
        "* `yardlineSide`: 3-letter team code corresponding to line-of-scrimmage (text)\n",
        "* `yardlineNumber`: Yard line at line-of-scrimmage (numeric)\n",
        "* `gameClock`: Time on clock of play (MM:SS)\n",
        "* `preSnapHomeScore`: Home score prior to the play (numeric)\n",
        "* `preSnapVisitorScore`: Visiting team score prior to the play (numeric)\n",
        "* `passResult`: Dropback outcome of the play (C: Complete pass, I: Incomplete pass, S: Quarterback sack, IN: Intercepted pass, R: Scramble, text)\n",
        "* `penaltyYards`: yards gained by offense by penalty (numeric)\n",
        "* `prePenaltyPlayResult`: Net yards gained by the offense, before penalty yardage (numeric)\n",
        "* `playResult`: Net yards gained by the offense, including penalty yardage (numeric)\n",
        "* `foulName[i]`: Name of the i-th penalty committed during the play. i ranges between 1 and 3 (text)\n",
        "* `foulNFLId[i]`: nflId of the player who comitted the i-th penalty during the play. i ranges between 1 and 3 (numeric)\n",
        "* `absoluteYardlineNumber`: Distance from end zone for possession team (numeric)\n",
        "* `offenseFormation`: Formation used by possession team (text)\n",
        "* `personnelO`: Personnel used by offensive team (text)\n",
        "* `defendersInTheBox`: Number of defenders in close proximity to line-of-scrimmage (numeric)\n",
        "* `personnelD`: Personnel used by defensive team (text)\n",
        "* `dropbackType`: Dropback categorization of quarterback (text)\n",
        "* `pff_playAction`: indicator for whether offense executes play action fake on the play. Variable provided by PFF (binary)\n",
        "* `pff_passCoverage`: Coverage scheme of defense. Variable provided by PFF (text)\n",
        "* `Bracket`: Recorded in the field and up to the 12 yard line in the red zone – when two offensive players have an in and out bracket by two defenders\n",
        "* `Goal Line`: Calls where a Goal Line defense is used.\n",
        "* `Red Zone`: Calls that are typically specific to the Red Zone and do not occur in the field often\n",
        "* `Prevent`: Special end of half or end of game situations where a Prevent defense is utilized\n",
        "* `Miscellaneous`: Coverage concepts that we feel do not comfortably fit into any of our coverage categories\n",
        "* `pff_passCoverageType`: Whether defense's coverage type was man, zone or other. Variable provided by PFF (text)\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "B9LgBqEp2sPi"
      },
      "outputs": [],
      "source": [
        "pl = pd.read_csv(\"/content/plays.csv\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "id": "OXHwazdt4JeB",
        "outputId": "c98392a2-9776-4fa4-d428-d99bd246a581"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(8557, 6)\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-8453a838-85d1-4025-8723-54bc3dbc0fd8\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>gameId</th>\n",
              "      <th>playId</th>\n",
              "      <th>quarter</th>\n",
              "      <th>down</th>\n",
              "      <th>yardsToGo</th>\n",
              "      <th>playResult</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>97</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>137</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>10</td>\n",
              "      <td>28</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>187</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>282</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>10</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>349</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>15</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8453a838-85d1-4025-8723-54bc3dbc0fd8')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-8453a838-85d1-4025-8723-54bc3dbc0fd8 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-8453a838-85d1-4025-8723-54bc3dbc0fd8');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "       gameId  playId  quarter  down  yardsToGo  playResult\n",
              "0  2021090900      97        1     3          2           0\n",
              "1  2021090900     137        1     1         10          28\n",
              "2  2021090900     187        1     2          6           5\n",
              "3  2021090900     282        1     1         10           0\n",
              "4  2021090900     349        1     3         15           0"
            ]
          },
          "execution_count": 95,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "play = pl[['gameId', 'playId', 'quarter', 'down', 'yardsToGo', 'playResult']]\n",
        "\n",
        "print(play.shape)\n",
        "play.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KP8UXMOB9Mm8"
      },
      "source": [
        "### Tracking Data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f7A4vtRDjAaY"
      },
      "source": [
        "Tracking data\n",
        "Files week[week].csv contains player tracking data from week [week].\n",
        "\n",
        "* `gameId`: Game identifier, unique (numeric)\n",
        "* `playId`: Play identifier, not unique across games (numeric)\n",
        "* `nflId`: Player identification number, unique across players. When value is NA, row corresponds to ball. (numeric)\n",
        "* `frameId`: Frame identifier for each play, starting at 1 (numeric)\n",
        "* `time`: Time stamp of play (time, yyyy-mm-dd, hh:mm:ss)\n",
        "* `jerseyNumber`: Jersey number of player (numeric)\n",
        "* `club`: Team abbrevation of corresponding player (text)\n",
        "* `playDirection`: Direction that the offense is moving (left or right)\n",
        "* `x`: Player position along the long axis of the field, 0 - 120 yards. See Figure 1 below. (numeric)\n",
        "* `y`: Player position along the short axis of the field, 0 - 53.3 yards. See Figure 1 below. (numeric)\n",
        "* `s`: Speed in yards/second (numeric)\n",
        "* `a`: Acceleration in yards/second^2 (numeric)\n",
        "* `dis`: Distance traveled from prior time point, in yards (numeric)\n",
        "* `o`: Player orientation (deg), 0 - 360 degrees (numeric)\n",
        "* `dir`: Angle of player motion (deg), 0 - 360 degrees (numeric)\n",
        "* `event`: Tagged play details, including moment of ball snap, pass release, pass catch, tackle, etc (text)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "Q2NwVcaB7E7K",
        "outputId": "adc8a303-b061-42d6-edb9-113a659b953e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(8071, 387)\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-c2d2ceb3-1440-4edc-82c6-b6299aae2589\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>gameId</th>\n",
              "      <th>playId</th>\n",
              "      <th>x_0_C</th>\n",
              "      <th>x_0_LT</th>\n",
              "      <th>x_0_LG</th>\n",
              "      <th>x_0_RG</th>\n",
              "      <th>x_0_RT</th>\n",
              "      <th>x_0.1_C</th>\n",
              "      <th>x_0.1_LT</th>\n",
              "      <th>x_0.1_LG</th>\n",
              "      <th>...</th>\n",
              "      <th>dir_0.9_C</th>\n",
              "      <th>dir_0.9_LT</th>\n",
              "      <th>dir_0.9_LG</th>\n",
              "      <th>dir_0.9_RG</th>\n",
              "      <th>dir_0.9_RT</th>\n",
              "      <th>dir_1_C</th>\n",
              "      <th>dir_1_LT</th>\n",
              "      <th>dir_1_LG</th>\n",
              "      <th>dir_1_RG</th>\n",
              "      <th>dir_1_RT</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>97</td>\n",
              "      <td>42.10</td>\n",
              "      <td>41.38</td>\n",
              "      <td>41.93</td>\n",
              "      <td>41.41</td>\n",
              "      <td>41.11</td>\n",
              "      <td>42.03</td>\n",
              "      <td>41.28</td>\n",
              "      <td>41.92</td>\n",
              "      <td>...</td>\n",
              "      <td>285.24</td>\n",
              "      <td>287.77</td>\n",
              "      <td>270.50</td>\n",
              "      <td>257.73</td>\n",
              "      <td>265.30</td>\n",
              "      <td>283.12</td>\n",
              "      <td>284.67</td>\n",
              "      <td>269.06</td>\n",
              "      <td>254.98</td>\n",
              "      <td>271.06</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>137</td>\n",
              "      <td>109.22</td>\n",
              "      <td>109.62</td>\n",
              "      <td>109.38</td>\n",
              "      <td>108.74</td>\n",
              "      <td>109.85</td>\n",
              "      <td>109.23</td>\n",
              "      <td>109.62</td>\n",
              "      <td>109.40</td>\n",
              "      <td>...</td>\n",
              "      <td>108.10</td>\n",
              "      <td>78.57</td>\n",
              "      <td>87.15</td>\n",
              "      <td>75.40</td>\n",
              "      <td>52.29</td>\n",
              "      <td>101.57</td>\n",
              "      <td>86.06</td>\n",
              "      <td>87.84</td>\n",
              "      <td>74.85</td>\n",
              "      <td>59.29</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>187</td>\n",
              "      <td>77.67</td>\n",
              "      <td>78.27</td>\n",
              "      <td>78.04</td>\n",
              "      <td>77.89</td>\n",
              "      <td>78.27</td>\n",
              "      <td>77.74</td>\n",
              "      <td>78.30</td>\n",
              "      <td>78.10</td>\n",
              "      <td>...</td>\n",
              "      <td>66.90</td>\n",
              "      <td>114.62</td>\n",
              "      <td>58.75</td>\n",
              "      <td>65.39</td>\n",
              "      <td>66.44</td>\n",
              "      <td>67.82</td>\n",
              "      <td>108.23</td>\n",
              "      <td>58.58</td>\n",
              "      <td>67.31</td>\n",
              "      <td>71.08</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>282</td>\n",
              "      <td>49.82</td>\n",
              "      <td>50.62</td>\n",
              "      <td>50.41</td>\n",
              "      <td>50.51</td>\n",
              "      <td>50.53</td>\n",
              "      <td>49.89</td>\n",
              "      <td>50.63</td>\n",
              "      <td>50.36</td>\n",
              "      <td>...</td>\n",
              "      <td>129.31</td>\n",
              "      <td>67.66</td>\n",
              "      <td>155.38</td>\n",
              "      <td>178.61</td>\n",
              "      <td>193.86</td>\n",
              "      <td>133.71</td>\n",
              "      <td>65.34</td>\n",
              "      <td>133.87</td>\n",
              "      <td>165.07</td>\n",
              "      <td>189.11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>349</td>\n",
              "      <td>54.93</td>\n",
              "      <td>55.78</td>\n",
              "      <td>55.73</td>\n",
              "      <td>55.72</td>\n",
              "      <td>55.78</td>\n",
              "      <td>55.01</td>\n",
              "      <td>55.81</td>\n",
              "      <td>55.74</td>\n",
              "      <td>...</td>\n",
              "      <td>92.44</td>\n",
              "      <td>99.67</td>\n",
              "      <td>99.29</td>\n",
              "      <td>68.57</td>\n",
              "      <td>71.53</td>\n",
              "      <td>93.19</td>\n",
              "      <td>99.64</td>\n",
              "      <td>95.55</td>\n",
              "      <td>69.08</td>\n",
              "      <td>71.77</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 387 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c2d2ceb3-1440-4edc-82c6-b6299aae2589')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-c2d2ceb3-1440-4edc-82c6-b6299aae2589 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-c2d2ceb3-1440-4edc-82c6-b6299aae2589');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "       gameId  playId   x_0_C  x_0_LT  x_0_LG  x_0_RG  x_0_RT  x_0.1_C  \\\n",
              "0  2021090900      97   42.10   41.38   41.93   41.41   41.11    42.03   \n",
              "1  2021090900     137  109.22  109.62  109.38  108.74  109.85   109.23   \n",
              "2  2021090900     187   77.67   78.27   78.04   77.89   78.27    77.74   \n",
              "3  2021090900     282   49.82   50.62   50.41   50.51   50.53    49.89   \n",
              "4  2021090900     349   54.93   55.78   55.73   55.72   55.78    55.01   \n",
              "\n",
              "   x_0.1_LT  x_0.1_LG  ...  dir_0.9_C  dir_0.9_LT  dir_0.9_LG  dir_0.9_RG  \\\n",
              "0     41.28     41.92  ...     285.24      287.77      270.50      257.73   \n",
              "1    109.62    109.40  ...     108.10       78.57       87.15       75.40   \n",
              "2     78.30     78.10  ...      66.90      114.62       58.75       65.39   \n",
              "3     50.63     50.36  ...     129.31       67.66      155.38      178.61   \n",
              "4     55.81     55.74  ...      92.44       99.67       99.29       68.57   \n",
              "\n",
              "   dir_0.9_RT  dir_1_C  dir_1_LT  dir_1_LG  dir_1_RG  dir_1_RT  \n",
              "0      265.30   283.12    284.67    269.06    254.98    271.06  \n",
              "1       52.29   101.57     86.06     87.84     74.85     59.29  \n",
              "2       66.44    67.82    108.23     58.58     67.31     71.08  \n",
              "3      193.86   133.71     65.34    133.87    165.07    189.11  \n",
              "4       71.53    93.19     99.64     95.55     69.08     71.77  \n",
              "\n",
              "[5 rows x 387 columns]"
            ]
          },
          "execution_count": 96,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Position data of all players for the first 1 second of the game, at 0.1 second time resolution\n",
        "# Position data: x-y position, speed, acceleration, distance traveled, orientation, direction\n",
        "\n",
        "field = pd.read_csv(\"https://raw.githubusercontent.com/fyng/PassplayNET/main/training_data.csv\")\n",
        "field = field.drop(['action_time', 'event'], axis = 1)\n",
        "\n",
        "print(field.shape)\n",
        "field.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bL_Hicv39R_y"
      },
      "source": [
        "### Combined Game State Data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 270
        },
        "id": "cWGWY0Tj8HR9",
        "outputId": "bb6e0b66-49ca-44a8-96ba-8352f75cc90f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "(8071, 391)\n",
            "(8069, 391)\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-bce55918-b877-40fe-b9a1-1ef74b1f443d\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>gameId</th>\n",
              "      <th>playId</th>\n",
              "      <th>x_0_C</th>\n",
              "      <th>x_0_LT</th>\n",
              "      <th>x_0_LG</th>\n",
              "      <th>x_0_RG</th>\n",
              "      <th>x_0_RT</th>\n",
              "      <th>x_0.1_C</th>\n",
              "      <th>x_0.1_LT</th>\n",
              "      <th>x_0.1_LG</th>\n",
              "      <th>...</th>\n",
              "      <th>dir_0.9_RT</th>\n",
              "      <th>dir_1_C</th>\n",
              "      <th>dir_1_LT</th>\n",
              "      <th>dir_1_LG</th>\n",
              "      <th>dir_1_RG</th>\n",
              "      <th>dir_1_RT</th>\n",
              "      <th>quarter</th>\n",
              "      <th>down</th>\n",
              "      <th>yardsToGo</th>\n",
              "      <th>playResult</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>97</td>\n",
              "      <td>42.10</td>\n",
              "      <td>41.38</td>\n",
              "      <td>41.93</td>\n",
              "      <td>41.41</td>\n",
              "      <td>41.11</td>\n",
              "      <td>42.03</td>\n",
              "      <td>41.28</td>\n",
              "      <td>41.92</td>\n",
              "      <td>...</td>\n",
              "      <td>265.30</td>\n",
              "      <td>283.12</td>\n",
              "      <td>284.67</td>\n",
              "      <td>269.06</td>\n",
              "      <td>254.98</td>\n",
              "      <td>271.06</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>137</td>\n",
              "      <td>109.22</td>\n",
              "      <td>109.62</td>\n",
              "      <td>109.38</td>\n",
              "      <td>108.74</td>\n",
              "      <td>109.85</td>\n",
              "      <td>109.23</td>\n",
              "      <td>109.62</td>\n",
              "      <td>109.40</td>\n",
              "      <td>...</td>\n",
              "      <td>52.29</td>\n",
              "      <td>101.57</td>\n",
              "      <td>86.06</td>\n",
              "      <td>87.84</td>\n",
              "      <td>74.85</td>\n",
              "      <td>59.29</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>10</td>\n",
              "      <td>28</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>187</td>\n",
              "      <td>77.67</td>\n",
              "      <td>78.27</td>\n",
              "      <td>78.04</td>\n",
              "      <td>77.89</td>\n",
              "      <td>78.27</td>\n",
              "      <td>77.74</td>\n",
              "      <td>78.30</td>\n",
              "      <td>78.10</td>\n",
              "      <td>...</td>\n",
              "      <td>66.44</td>\n",
              "      <td>67.82</td>\n",
              "      <td>108.23</td>\n",
              "      <td>58.58</td>\n",
              "      <td>67.31</td>\n",
              "      <td>71.08</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>282</td>\n",
              "      <td>49.82</td>\n",
              "      <td>50.62</td>\n",
              "      <td>50.41</td>\n",
              "      <td>50.51</td>\n",
              "      <td>50.53</td>\n",
              "      <td>49.89</td>\n",
              "      <td>50.63</td>\n",
              "      <td>50.36</td>\n",
              "      <td>...</td>\n",
              "      <td>193.86</td>\n",
              "      <td>133.71</td>\n",
              "      <td>65.34</td>\n",
              "      <td>133.87</td>\n",
              "      <td>165.07</td>\n",
              "      <td>189.11</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>10</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2021090900</td>\n",
              "      <td>349</td>\n",
              "      <td>54.93</td>\n",
              "      <td>55.78</td>\n",
              "      <td>55.73</td>\n",
              "      <td>55.72</td>\n",
              "      <td>55.78</td>\n",
              "      <td>55.01</td>\n",
              "      <td>55.81</td>\n",
              "      <td>55.74</td>\n",
              "      <td>...</td>\n",
              "      <td>71.53</td>\n",
              "      <td>93.19</td>\n",
              "      <td>99.64</td>\n",
              "      <td>95.55</td>\n",
              "      <td>69.08</td>\n",
              "      <td>71.77</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>15</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 391 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-bce55918-b877-40fe-b9a1-1ef74b1f443d')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-bce55918-b877-40fe-b9a1-1ef74b1f443d button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-bce55918-b877-40fe-b9a1-1ef74b1f443d');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "       gameId  playId   x_0_C  x_0_LT  x_0_LG  x_0_RG  x_0_RT  x_0.1_C  \\\n",
              "0  2021090900      97   42.10   41.38   41.93   41.41   41.11    42.03   \n",
              "1  2021090900     137  109.22  109.62  109.38  108.74  109.85   109.23   \n",
              "2  2021090900     187   77.67   78.27   78.04   77.89   78.27    77.74   \n",
              "3  2021090900     282   49.82   50.62   50.41   50.51   50.53    49.89   \n",
              "4  2021090900     349   54.93   55.78   55.73   55.72   55.78    55.01   \n",
              "\n",
              "   x_0.1_LT  x_0.1_LG  ...  dir_0.9_RT  dir_1_C  dir_1_LT  dir_1_LG  dir_1_RG  \\\n",
              "0     41.28     41.92  ...      265.30   283.12    284.67    269.06    254.98   \n",
              "1    109.62    109.40  ...       52.29   101.57     86.06     87.84     74.85   \n",
              "2     78.30     78.10  ...       66.44    67.82    108.23     58.58     67.31   \n",
              "3     50.63     50.36  ...      193.86   133.71     65.34    133.87    165.07   \n",
              "4     55.81     55.74  ...       71.53    93.19     99.64     95.55     69.08   \n",
              "\n",
              "   dir_1_RT  quarter  down  yardsToGo  playResult  \n",
              "0    271.06        1     3          2           0  \n",
              "1     59.29        1     1         10          28  \n",
              "2     71.08        1     2          6           5  \n",
              "3    189.11        1     1         10           0  \n",
              "4     71.77        1     3         15           0  \n",
              "\n",
              "[5 rows x 391 columns]"
            ]
          },
          "execution_count": 76,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Each row is a unique combination of playId and gameId\n",
        "\n",
        "data = pd.merge(field, play, on=['gameId', 'playId'])\n",
        "\n",
        "# remove NaN and Inf\n",
        "print(data.shape)\n",
        "data = data[~data.isin([np.nan, np.inf, -np.inf]).any(1)]\n",
        "print(data.shape)\n",
        "\n",
        "data.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Tz8adDIbAbGq"
      },
      "source": [
        "### Understanding the Data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "pe6l15CsAlUy",
        "outputId": "536339a9-1279-4816-973e-a717aa437a0a"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.hist(data[data['down'] == 0]['playResult'], bins=[-40, 0, 1, 10, 40])\n",
        "plt.show()\n",
        "\n",
        "plt.hist(data[data['down'] == 1]['playResult'], bins=[-40, 0, 1, 10, 40])\n",
        "plt.show()\n",
        "\n",
        "plt.hist(data[data['down'] == 2]['playResult'], bins=[-40, 0, 1, 10, 40])\n",
        "plt.show()\n",
        "\n",
        "plt.hist(data[data['down'] == 3]['playResult'], bins=[-40, 0, 1, 10, 40])\n",
        "plt.show()\n",
        "# Question for Eric: is it intuitive for most plays to have 0 net yards gained?"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aNO6oNgg-vrq"
      },
      "source": [
        "# Models"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "Go4pUnjN8SxU",
        "outputId": "77ad5979-73f0-4214-a104-8be513a14416"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD4CAYAAAAAczaOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAATm0lEQVR4nO3db4xd9Z3f8fenDiGrJgITpqxjrDXZepuSSgXkArtZVRQa/kVaE3U3JQ8Wb0TlRAUpUVdVna1USFJUtuomEmqWFbu4MVUawibZxQWn1CGsojzgj0kdg2EpAyHClgOzceIERaWFfvtgfk7vmpm5dzx37oz5vV/S1T33e37n3O85Y3/umXPPvZOqQpLUh7+x0g1IkibH0Jekjhj6ktQRQ1+SOmLoS1JH3rLSDSzkzDPPrI0bN650G5J0Unn88cf/qqqm5pq3qkN/48aN7N27d6XbkKSTSpLvzzfP0zuS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpI4a+JHXE0Jekjhj6ktSRVf2JXOmkc/NpS1z+6Hj6kObhkb4kdcTQl6SOGPqS1BFDX5I6MjT0k7wtyaNJvpvkQJJPtfoXknwvyb52O6/Vk+S2JNNJ9ie5YGBdW5M8225bl2+zJElzGeXqnVeBS6vqlSSnAN9O8vU2719W1VeOG38VsKndLgJuBy5KcgZwE7AZKODxJLuq6kfj2BBJ0nBDj/Rr1ivt4SntVgsssgW4qy33MHB6knXAFcCeqjrSgn4PcOXS2pckLcZI5/STrEmyD3iZ2eB+pM26pZ3C+VySU1ttPfDiwOIHW22++vHPtS3J3iR7Z2ZmFrk5kqSFjBT6VfV6VZ0HnA1cmOTvAZ8E3gP8A+AM4F+No6GquqOqNlfV5qmpOf/EoyTpBC3q6p2q+jHwEHBlVR1up3BeBf4TcGEbdgjYMLDY2a02X12SNCGjXL0zleT0Nv0LwPuBv2zn6UkS4BrgybbILuC6dhXPxcDRqjoMPABcnmRtkrXA5a0mSZqQUa7eWQfsTLKG2ReJe6rqviTfTDIFBNgHfKyN3w1cDUwDPwM+AlBVR5J8Bnisjft0VR0Z36ZIkoYZGvpVtR84f476pfOML+CGeebtAHYsskdJ0pj4iVxJ6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpI0NDP8nbkjya5LtJDiT5VKufk+SRJNNJvpzkra1+ans83eZvHFjXJ1v9mSRXLNdGSZLmNvQPowOvApdW1StJTgG+neTrwL8APldVdyf5I+B64PZ2/6Oq+ttJrgV+H/inSc4FrgXeC7wL+EaSX6mq15dhu6ST082njWEdR5e+Dr1pDT3Sr1mvtIentFsBlwJfafWdwDVtekt7TJt/WZK0+t1V9WpVfQ+YBi4cy1ZIkkYy0jn9JGuS7ANeBvYAzwE/rqrX2pCDwPo2vR54EaDNPwq8c7A+xzKSpAkYKfSr6vWqOg84m9mj8/csV0NJtiXZm2TvzMzMcj2NJHVpUVfvVNWPgYeAXwVOT3LsPYGzgUNt+hCwAaDNPw344WB9jmUGn+OOqtpcVZunpqYW054kaYhRrt6ZSnJ6m/4F4P3A08yG/2+2YVuBe9v0rvaYNv+bVVWtfm27uuccYBPw6Lg2RJI03ChX76wDdiZZw+yLxD1VdV+Sp4C7k/xb4H8Ad7bxdwL/Ock0cITZK3aoqgNJ7gGeAl4DbvDKHUmarKGhX1X7gfPnqD/PHFffVNX/An5rnnXdAtyy+DYlSePgJ3IlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOjPKFa+rYxu33j3V9L9z6gbGuT9LieKQvSR0x9CWpI4a+JHXE0Jekjhj6ktQRQ1+SOmLoS1JHDH1J6sjQ0E+yIclDSZ5KciDJx1v95iSHkuxrt6sHlvlkkukkzyS5YqB+ZatNJ9m+PJskSZrPKJ/IfQ343ar6TpJ3AI8n2dPmfa6q/sPg4CTnAtcC7wXeBXwjya+02Z8H3g8cBB5LsquqnhrHhkiShhsa+lV1GDjcpn+a5Glg/QKLbAHurqpXge8lmQYubPOmq+p5gCR3t7GGviRNyKLO6SfZCJwPPNJKNybZn2RHkrWtth54cWCxg602X/3459iWZG+SvTMzM4tpT5I0xMihn+TtwFeBT1TVT4DbgV8GzmP2N4E/GEdDVXVHVW2uqs1TU1PjWKUkqRnpWzaTnMJs4H+xqr4GUFUvDcz/Y+C+9vAQsGFg8bNbjQXqkqQJGOXqnQB3Ak9X1WcH6usGhn0QeLJN7wKuTXJqknOATcCjwGPApiTnJHkrs2/27hrPZkiSRjHKkf77gN8Gnkiyr9V+D/hwkvOAAl4APgpQVQeS3MPsG7SvATdU1esASW4EHgDWADuq6sAYt0WSNMQoV+98G8gcs3YvsMwtwC1z1HcvtJwkaXn5iVxJ6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6MtKfS5TGZeP2+8e6vhdu/cBY1ye92XmkL0kdMfQlqSOGviR1ZGjoJ9mQ5KEkTyU5kOTjrX5Gkj1Jnm33a1s9SW5LMp1kf5ILBta1tY1/NsnW5dssSdJcRjnSfw343ao6F7gYuCHJucB24MGq2gQ82B4DXAVsardtwO0w+yIB3ARcBFwI3HTshUKSNBlDQ7+qDlfVd9r0T4GngfXAFmBnG7YTuKZNbwHuqlkPA6cnWQdcAeypqiNV9SNgD3DlWLdGkrSgRZ3TT7IROB94BDirqg63WT8AzmrT64EXBxY72Grz1Y9/jm1J9ibZOzMzs5j2JElDjBz6Sd4OfBX4RFX9ZHBeVRVQ42ioqu6oqs1VtXlqamocq5QkNSOFfpJTmA38L1bV11r5pXbahnb/cqsfAjYMLH52q81XlyRNyChX7wS4E3i6qj47MGsXcOwKnK3AvQP169pVPBcDR9tpoAeAy5OsbW/gXt5qkqQJGeVrGN4H/DbwRJJ9rfZ7wK3APUmuB74PfKjN2w1cDUwDPwM+AlBVR5J8Bnisjft0VR0Zy1ZIkkYyNPSr6ttA5pl92RzjC7hhnnXtAHYspkFJ0vj4iVxJ6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqyCh/LlEniY3b71/pFiStch7pS1JHDH1J6sjQ0E+yI8nLSZ4cqN2c5FCSfe129cC8TyaZTvJMkisG6le22nSS7ePfFEnSMKMc6X8BuHKO+ueq6rx22w2Q5FzgWuC9bZk/TLImyRrg88BVwLnAh9tYSdIEDX0jt6q+lWTjiOvbAtxdVa8C30syDVzY5k1X1fMASe5uY59adMeSpBO2lHP6NybZ307/rG219cCLA2MOttp89TdIsi3J3iR7Z2ZmltCeJOl4Jxr6twO/DJwHHAb+YFwNVdUdVbW5qjZPTU2Na7WSJE7wOv2qeunYdJI/Bu5rDw8BGwaGnt1qLFCXJE3ICR3pJ1k38PCDwLEre3YB1yY5Nck5wCbgUeAxYFOSc5K8ldk3e3edeNuSpBMx9Eg/yZeAS4AzkxwEbgIuSXIeUMALwEcBqupAknuYfYP2NeCGqnq9redG4AFgDbCjqg6MfWskSQsa5eqdD89RvnOB8bcAt8xR3w3sXlR3kqSx8hO5ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkeGhn6SHUleTvLkQO2MJHuSPNvu17Z6ktyWZDrJ/iQXDCyztY1/NsnW5dkcSdJCRjnS/wJw5XG17cCDVbUJeLA9BrgK2NRu24DbYfZFArgJuAi4ELjp2AuFJGlyhoZ+VX0LOHJceQuws03vBK4ZqN9Vsx4GTk+yDrgC2FNVR6rqR8Ae3vhCIklaZid6Tv+sqjrcpn8AnNWm1wMvDow72Grz1d8gybYke5PsnZmZOcH2JElzWfIbuVVVQI2hl2Pru6OqNlfV5qmpqXGtVpLEiYf+S+20De3+5VY/BGwYGHd2q81XlyRN0ImG/i7g2BU4W4F7B+rXtat4LgaOttNADwCXJ1nb3sC9vNUkSRP0lmEDknwJuAQ4M8lBZq/CuRW4J8n1wPeBD7Xhu4GrgWngZ8BHAKrqSJLPAI+1cZ+uquPfHJYkLbOhoV9VH55n1mVzjC3ghnnWswPYsajuJElj5SdyJakjhr4kdcTQl6SODD2nL+kkc/NpS1z+6Hj60Kpk6OuktnH7/WNd3wu3fmCs65NWG0/vSFJHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOLCn0k7yQ5Ikk+5LsbbUzkuxJ8my7X9vqSXJbkukk+5NcMI4NkCSNbhxH+v+oqs6rqs3t8XbgwaraBDzYHgNcBWxqt23A7WN4bknSIizH6Z0twM42vRO4ZqB+V816GDg9ybpleH5J0jyWGvoF/PckjyfZ1mpnVdXhNv0D4Kw2vR54cWDZg6321yTZlmRvkr0zMzNLbE+SNGipfy7x16vqUJK/BexJ8peDM6uqktRiVlhVdwB3AGzevHlRy0qSFrakI/2qOtTuXwb+DLgQeOnYaZt2/3IbfgjYMLD42a0mSZqQEw79JH8zyTuOTQOXA08Cu4CtbdhW4N42vQu4rl3FczFwdOA0kCRpApZyeucs4M+SHFvPf6mq/5bkMeCeJNcD3wc+1MbvBq4GpoGfAR9ZwnO/KWzcfv9KtyCpMycc+lX1PPD356j/ELhsjnoBN5zo80mSls5P5EpSRwx9SeqIoS9JHTH0Jakjhr4kdWSpn8iV3lxuPm2lO5CWlUf6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xOv0Jf11S/2sws1Hx9OHloVH+pLUEUNfkjpi6EtSRwx9SeqIoS9JHfHqnUXwD5lLOtlN/Eg/yZVJnkkynWT7pJ9fkno20SP9JGuAzwPvBw4CjyXZVVVPTbIPSctoHH+TwGv9l82kT+9cCExX1fMASe4GtgDLEvqejpFOUn5AbNlMOvTXAy8OPD4IXDQ4IMk2YFt7+EqSZybU25nAX03ouZbCPsfv571mhRsZ4mTZpyvf56dG/kmufK+jWWyfvzTfjFX3Rm5V3QHcMennTbK3qjZP+nkXyz7H72Tp1T7H72TpdZx9TvqN3EPAhoHHZ7eaJGkCJh36jwGbkpyT5K3AtcCuCfcgSd2a6OmdqnotyY3AA8AaYEdVHZhkDwuY+CmlE2Sf43ey9Gqf43ey9Dq2PlNV41qXJGmV82sYJKkjhr4kdaTb0E/yW0kOJPm/Sea9FGqlvzYiyRlJ9iR5tt2vnWfc60n2tdvE3hwftn+SnJrky23+I0k2Tqq34/oY1ufvJJkZ2If/bIX63JHk5SRPzjM/SW5r27E/yQWT7rH1MazPS5IcHdif/2bSPQ70siHJQ0meav/nPz7HmBXfryP2ufT9WlVd3oC/C/wd4C+AzfOMWQM8B7wbeCvwXeDcCff574HtbXo78PvzjHtlBfbh0P0D/HPgj9r0tcCXV2mfvwP8x0n3Nkev/xC4AHhynvlXA19n9nNkFwOPrNI+LwHuW+n92XpZB1zQpt8B/M85fv4rvl9H7HPJ+7XbI/2qerqqhn3a9+dfG1FV/xs49rURk7QF2NmmdwLXTPj5FzLK/hns/yvAZUkm/cHX1fBzHElVfQs4ssCQLcBdNeth4PQk6ybT3f83Qp+rRlUdrqrvtOmfAk8z++0Ag1Z8v47Y55J1G/ojmutrI8b+QxjirKo63KZ/AJw1z7i3Jdmb5OEkk3phGGX//HxMVb0GHAXeOZHu5uihme/n+E/ar/ZfSbJhjvmrwWr4NzmqX03y3SRfT/LelW4GoJ1ePB945LhZq2q/LtAnLHG/rrqvYRinJN8AfnGOWf+6qu6ddD/zWajPwQdVVUnmu8b2l6rqUJJ3A99M8kRVPTfuXt/E/ivwpap6NclHmf3t5NIV7ulk9h1m/02+kuRq4M+BTSvZUJK3A18FPlFVP1nJXhYypM8l79c3dehX1T9e4iom8rURC/WZ5KUk66rqcPt18+V51nGo3T+f5C+YPUpY7tAfZf8cG3MwyVuA04AfLnNfxxvaZ1UN9vQnzL6XshqdFF9lMhhWVbU7yR8mObOqVuTLzZKcwmyQfrGqvjbHkFWxX4f1OY796umdha2Gr43YBWxt01uBN/yGkmRtklPb9JnA+1imr6s+zij7Z7D/3wS+We0dqQka2udx529/g9nzqavRLuC6drXJxcDRgdN/q0aSXzz23k2SC5nNmkm/2B/rJcCdwNNV9dl5hq34fh2lz7Hs10m/Q71absAHmT1v9yrwEvBAq78L2D0w7mpm30V/jtnTQpPu853Ag8CzwDeAM1p9M/AnbfrXgCeYvSrlCeD6Cfb3hv0DfBr4jTb9NuBPgWngUeDdK/TzHtbnvwMOtH34EPCeFerzS8Bh4P+0f5/XAx8DPtbmh9k/RPRc+1nPeeXZKujzxoH9+TDwayvRZ+vl14EC9gP72u3q1bZfR+xzyfvVr2GQpI54ekeSOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI78P73B8W5QkLbFAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "X = data.loc[:, 'x_0_C':'yardsToGo']\n",
        "names = X.columns\n",
        "scalar = preprocessing.StandardScaler()\n",
        "X = scalar.fit_transform(X)\n",
        "\n",
        "plt.hist(X.mean(axis = 1))\n",
        "plt.hist(X.std(axis = 1))\n",
        "\n",
        "y = data['playResult']\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cbzbfNPO-yRE",
        "outputId": "2d2978ad-49d2-4f12-e882-4b1cf4ef9d6e"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "-0.05799500618530007"
            ]
          },
          "execution_count": 144,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# try simple linear regression first\n",
        "\n",
        "from sklearn.linear_model import LinearRegression\n",
        "reg = LinearRegression().fit(X_train, y_train)\n",
        "reg.score(X_test, y_test)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NbE8kS4UAQyx"
      },
      "source": [
        "That doesn't look too good.. Let's try something more advanced"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "3FLovmq-_7oD"
      },
      "outputs": [],
      "source": [
        "# how about a decision tree?\n",
        "from sklearn import tree\n",
        "score = []\n",
        "\n",
        "for i in range(5, 20, 5):\n",
        "    clf = tree.DecisionTreeRegressor(min_samples_leaf=5, max_depth=i, random_state=42)\n",
        "    clf = clf.fit(X_train, y_train)\n",
        "    clf_score = clf.score(X_test, y_test)\n",
        "    score.append(clf_score)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "Bi-9NXMcDm8j",
        "outputId": "45511ecb-73f7-4f45-fa9c-3ca7cbaa85e8"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f775e2cc940>]"
            ]
          },
          "execution_count": 118,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.plot(range(5, 20, 5), score)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "whcPxiO-IiQM"
      },
      "source": [
        "Decision tree also does poorly, furthermore deeper trees overfit. Oh-oh.."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 301
        },
        "id": "iAn-islbDrcE",
        "outputId": "3b07c8e2-ceef-4c78-b4ed-6af9c2aafd54"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Text(0, 0.5, 'Principal Component 2')"
            ]
          },
          "execution_count": 149,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Do some PCA\n",
        "from sklearn.decomposition import PCA\n",
        "pca = PCA(n_components=2)\n",
        "principalComponents = pca.fit_transform(X)\n",
        "\n",
        "principalDf = pd.DataFrame(data = principalComponents\n",
        "             , columns = ['pc1', 'pc2'])\n",
        "finalDf = pd.concat([principalDf, y], axis = 1)\n",
        "\n",
        "plt.scatter(finalDf['pc1'], finalDf['pc2'])\n",
        "plt.xlabel('Principal Component 1', fontsize = 15)\n",
        "plt.ylabel('Principal Component 2', fontsize = 15)\n",
        "\n",
        "# fig = plt.figure(figsize = (8,8))\n",
        "# ax = fig.add_subplot(1,1,1) \n",
        "# ax.set_xlabel('Principal Component 1', fontsize = 15)\n",
        "# ax.set_ylabel('Principal Component 2', fontsize = 15)\n",
        "# ax.set_title('2 component PCA', fontsize = 20)\n",
        "# targets = y.unique().tolist()\n",
        "# colors = ['r', 'g', 'b']\n",
        "# for target in targets:\n",
        "#     # rgb = (np.random.rand(), np.random.rand(), np.random.rand())\n",
        "#     indicesToKeep = finalDf['playResult'] == target\n",
        "#     ax.scatter(finalDf.loc[indicesToKeep, 'principal component 1']\n",
        "#                , finalDf.loc[indicesToKeep, 'principal component 2']\n",
        "#             #    , c = rgb\n",
        "#                , s = 50)\n",
        "# ax.legend(targets)\n",
        "# ax.grid()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "lga93768MZ87",
        "outputId": "19653ac3-f7f1-4584-adb7-a7aa5a2e00d2"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'dir_1_C'"
            ]
          },
          "execution_count": 155,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "i = pca.components_[0,:].argmax()\n",
        "names[i]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "dzwtdgVmQ0AF",
        "outputId": "46505c3e-85ea-424a-bf1d-0277752b2789"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "-0.04565543678507833"
            ]
          },
          "execution_count": 166,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Random forest\n",
        "\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "rf = RandomForestRegressor(n_estimators = 1000, random_state = 42)\n",
        "# Train the model on training data\n",
        "rf.fit(X_train, y_train)\n",
        "rf.score(X_test, y_test)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Md2VthsoQzx4"
      },
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyMrLOMRtJ06gEekDBon6bg8",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}