columns that did not need to be aggregated were being aggregated. I

src/pre_processing.ipynb

@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 55,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -15,7 +15,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 56,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
@@ -85,10 +85,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 57,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
+    "# here we are making it such that each game has two rows, where each one is a team view of the game with\n",
+    "# opposing metrics.\n",
     "\n",
     "detailed_metrics = {\n",
     "    \"Score\",\n",
@@ -128,7 +130,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 58,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -326,7 +328,7 @@
        "[5 rows x 36 columns]"
       ]
      },
-     "execution_count": 58,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -337,7 +339,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 59,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -349,7 +351,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 60,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -547,7 +549,7 @@
        "[5 rows x 37 columns]"
       ]
      },
-     "execution_count": 60,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -558,7 +560,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 61,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
@@ -572,7 +574,7 @@
        "Name: Win, dtype: int64"
       ]
      },
-     "execution_count": 61,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -595,8 +597,36 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 95,
+   "execution_count": 15,
    "metadata": {},
+   "outputs": [],
+   "source": [
+    "exclude_agg_cols = {\n",
+    "    \"TeamID\",\n",
+    "    \"Season\",\n",
+    "    \"League\",\n",
+    "    \"GameResult\",\n",
+    "    \"OppLoc\",\n",
+    "    \"TeamLoc\",\n",
+    "    \"Season\",\n",
+    "    \"DayNum\",\n",
+    "}\n",
+    "\n",
+    "agg_funcs = [\n",
+    "    np.min,\n",
+    "    np.max,\n",
+    "    np.std,\n",
+    "    np.median,\n",
+    "    np.mean,\n",
+    "]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {
+    "tags": []
+   },
    "outputs": [
     {
      "data": {
@@ -622,13 +652,13 @@
        "      <th>TeamID</th>\n",
        "      <th>Season</th>\n",
        "      <th>League</th>\n",
-       "      <th>Season min</th>\n",
-       "      <th>Season max</th>\n",
-       "      <th>Season std</th>\n",
-       "      <th>Season median</th>\n",
-       "      <th>Season mean</th>\n",
-       "      <th>DayNum min</th>\n",
-       "      <th>DayNum max</th>\n",
+       "      <th>TeamScore min</th>\n",
+       "      <th>TeamScore max</th>\n",
+       "      <th>TeamScore std</th>\n",
+       "      <th>TeamScore median</th>\n",
+       "      <th>TeamScore mean</th>\n",
+       "      <th>OppScore min</th>\n",
+       "      <th>OppScore max</th>\n",
        "      <th>...</th>\n",
        "      <th>ScoreDiff min</th>\n",
        "      <th>ScoreDiff max</th>\n",
@@ -648,13 +678,13 @@
        "      <td>3430</td>\n",
        "      <td>2012</td>\n",
        "      <td>W</td>\n",
-       "      <td>2012</td>\n",
-       "      <td>2012</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2012.0</td>\n",
-       "      <td>2012.0</td>\n",
-       "      <td>11</td>\n",
-       "      <td>131</td>\n",
+       "      <td>41</td>\n",
+       "      <td>78</td>\n",
+       "      <td>10.808339</td>\n",
+       "      <td>61.0</td>\n",
+       "      <td>58.965517</td>\n",
+       "      <td>36</td>\n",
+       "      <td>85</td>\n",
        "      <td>...</td>\n",
        "      <td>-32</td>\n",
        "      <td>35</td>\n",
@@ -672,13 +702,13 @@
        "      <td>1431</td>\n",
        "      <td>2018</td>\n",
        "      <td>M</td>\n",
-       "      <td>2018</td>\n",
-       "      <td>2018</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2018.0</td>\n",
-       "      <td>2018.0</td>\n",
-       "      <td>17</td>\n",
-       "      <td>128</td>\n",
+       "      <td>33</td>\n",
+       "      <td>88</td>\n",
+       "      <td>12.283247</td>\n",
+       "      <td>67.0</td>\n",
+       "      <td>66.466667</td>\n",
+       "      <td>44</td>\n",
+       "      <td>97</td>\n",
        "      <td>...</td>\n",
        "      <td>-49</td>\n",
        "      <td>29</td>\n",
@@ -696,13 +726,13 @@
        "      <td>1315</td>\n",
        "      <td>2014</td>\n",
        "      <td>M</td>\n",
-       "      <td>2014</td>\n",
-       "      <td>2014</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2014.0</td>\n",
-       "      <td>2014.0</td>\n",
-       "      <td>15</td>\n",
-       "      <td>131</td>\n",
+       "      <td>43</td>\n",
+       "      <td>95</td>\n",
+       "      <td>10.019980</td>\n",
+       "      <td>72.0</td>\n",
+       "      <td>73.000000</td>\n",
+       "      <td>61</td>\n",
+       "      <td>103</td>\n",
        "      <td>...</td>\n",
        "      <td>-27</td>\n",
        "      <td>18</td>\n",
@@ -720,13 +750,13 @@
        "      <td>1307</td>\n",
        "      <td>2005</td>\n",
        "      <td>M</td>\n",
-       "      <td>2005</td>\n",
-       "      <td>2005</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2005.0</td>\n",
-       "      <td>2005.0</td>\n",
-       "      <td>11</td>\n",
-       "      <td>131</td>\n",
+       "      <td>53</td>\n",
+       "      <td>101</td>\n",
+       "      <td>12.911860</td>\n",
+       "      <td>77.0</td>\n",
+       "      <td>75.870968</td>\n",
+       "      <td>47</td>\n",
+       "      <td>81</td>\n",
        "      <td>...</td>\n",
        "      <td>-17</td>\n",
        "      <td>34</td>\n",
@@ -744,13 +774,13 @@
        "      <td>1266</td>\n",
        "      <td>2008</td>\n",
        "      <td>M</td>\n",
-       "      <td>2008</td>\n",
-       "      <td>2008</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2008.0</td>\n",
-       "      <td>2008.0</td>\n",
-       "      <td>5</td>\n",
-       "      <td>130</td>\n",
+       "      <td>51</td>\n",
+       "      <td>100</td>\n",
+       "      <td>11.841315</td>\n",
+       "      <td>75.5</td>\n",
+       "      <td>75.906250</td>\n",
+       "      <td>37</td>\n",
+       "      <td>89</td>\n",
        "      <td>...</td>\n",
        "      <td>-20</td>\n",
        "      <td>47</td>\n",
@@ -768,13 +798,13 @@
        "      <td>1352</td>\n",
        "      <td>2016</td>\n",
        "      <td>M</td>\n",
-       "      <td>2016</td>\n",
-       "      <td>2016</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2016.0</td>\n",
-       "      <td>2016.0</td>\n",
-       "      <td>11</td>\n",
-       "      <td>121</td>\n",
+       "      <td>44</td>\n",
+       "      <td>89</td>\n",
+       "      <td>10.298567</td>\n",
+       "      <td>67.0</td>\n",
+       "      <td>65.062500</td>\n",
+       "      <td>45</td>\n",
+       "      <td>106</td>\n",
        "      <td>...</td>\n",
        "      <td>-62</td>\n",
        "      <td>18</td>\n",
@@ -792,13 +822,13 @@
        "      <td>1194</td>\n",
        "      <td>2005</td>\n",
        "      <td>M</td>\n",
-       "      <td>2005</td>\n",
-       "      <td>2005</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2005.0</td>\n",
-       "      <td>2005.0</td>\n",
-       "      <td>18</td>\n",
-       "      <td>117</td>\n",
+       "      <td>45</td>\n",
+       "      <td>104</td>\n",
+       "      <td>14.194618</td>\n",
+       "      <td>76.0</td>\n",
+       "      <td>76.777778</td>\n",
+       "      <td>59</td>\n",
+       "      <td>107</td>\n",
        "      <td>...</td>\n",
        "      <td>-45</td>\n",
        "      <td>27</td>\n",
@@ -816,13 +846,13 @@
        "      <td>3270</td>\n",
        "      <td>2021</td>\n",
        "      <td>W</td>\n",
-       "      <td>2021</td>\n",
-       "      <td>2021</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2021.0</td>\n",
-       "      <td>2021.0</td>\n",
-       "      <td>23</td>\n",
-       "      <td>128</td>\n",
+       "      <td>24</td>\n",
+       "      <td>80</td>\n",
+       "      <td>13.385137</td>\n",
+       "      <td>53.0</td>\n",
+       "      <td>55.476190</td>\n",
+       "      <td>41</td>\n",
+       "      <td>117</td>\n",
        "      <td>...</td>\n",
        "      <td>-93</td>\n",
        "      <td>24</td>\n",
@@ -840,13 +870,13 @@
        "      <td>3240</td>\n",
        "      <td>2014</td>\n",
        "      <td>W</td>\n",
-       "      <td>2014</td>\n",
-       "      <td>2014</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2014.0</td>\n",
-       "      <td>2014.0</td>\n",
-       "      <td>4</td>\n",
-       "      <td>123</td>\n",
+       "      <td>43</td>\n",
+       "      <td>84</td>\n",
+       "      <td>11.319009</td>\n",
+       "      <td>62.5</td>\n",
+       "      <td>63.593750</td>\n",
+       "      <td>45</td>\n",
+       "      <td>100</td>\n",
        "      <td>...</td>\n",
        "      <td>-42</td>\n",
        "      <td>17</td>\n",
@@ -864,13 +894,13 @@
        "      <td>3452</td>\n",
        "      <td>2011</td>\n",
        "      <td>W</td>\n",
-       "      <td>2011</td>\n",
-       "      <td>2011</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>2011.0</td>\n",
-       "      <td>2011.0</td>\n",
-       "      <td>11</td>\n",
-       "      <td>124</td>\n",
+       "      <td>39</td>\n",
+       "      <td>90</td>\n",
+       "      <td>12.518374</td>\n",
+       "      <td>65.0</td>\n",
+       "      <td>65.750000</td>\n",
+       "      <td>21</td>\n",
+       "      <td>79</td>\n",
        "      <td>...</td>\n",
        "      <td>-23</td>\n",
        "      <td>57</td>\n",
@@ -885,89 +915,70 @@
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "<p>10 rows × 173 columns</p>\n",
+       "<p>10 rows × 158 columns</p>\n",
        "</div>"
       ],
       "text/plain": [
-       "       TeamID  Season League  Season min  Season max  Season std  \\\n",
-       "12348    3430    2012      W        2012        2012         0.0   \n",
-       "6900     1431    2018      M        2018        2018         0.0   \n",
-       "4406     1315    2014      M        2014        2014         0.0   \n",
-       "4233     1307    2005      M        2005        2005         0.0   \n",
-       "3407     1266    2008      M        2008        2008         0.0   \n",
-       "5190     1352    2016      M        2016        2016         0.0   \n",
-       "1892     1194    2005      M        2005        2005         0.0   \n",
-       "10020    3270    2021      W        2021        2021         0.0   \n",
-       "9567     3240    2014      W        2014        2014         0.0   \n",
-       "12617    3452    2011      W        2011        2011         0.0   \n",
+       "       TeamID  Season League  TeamScore min  TeamScore max  TeamScore std  \\\n",
+       "12348    3430    2012      W             41             78      10.808339   \n",
+       "6900     1431    2018      M             33             88      12.283247   \n",
+       "4406     1315    2014      M             43             95      10.019980   \n",
+       "4233     1307    2005      M             53            101      12.911860   \n",
+       "3407     1266    2008      M             51            100      11.841315   \n",
+       "5190     1352    2016      M             44             89      10.298567   \n",
+       "1892     1194    2005      M             45            104      14.194618   \n",
+       "10020    3270    2021      W             24             80      13.385137   \n",
+       "9567     3240    2014      W             43             84      11.319009   \n",
+       "12617    3452    2011      W             39             90      12.518374   \n",
        "\n",
-       "       Season median  Season mean  DayNum min  DayNum max  ...  ScoreDiff min  \\\n",
-       "12348         2012.0       2012.0          11         131  ...            -32   \n",
-       "6900          2018.0       2018.0          17         128  ...            -49   \n",
-       "4406          2014.0       2014.0          15         131  ...            -27   \n",
-       "4233          2005.0       2005.0          11         131  ...            -17   \n",
-       "3407          2008.0       2008.0           5         130  ...            -20   \n",
-       "5190          2016.0       2016.0          11         121  ...            -62   \n",
-       "1892          2005.0       2005.0          18         117  ...            -45   \n",
-       "10020         2021.0       2021.0          23         128  ...            -93   \n",
-       "9567          2014.0       2014.0           4         123  ...            -42   \n",
-       "12617         2011.0       2011.0          11         124  ...            -23   \n",
+       "       TeamScore median  TeamScore mean  OppScore min  OppScore max  ...  \\\n",
+       "12348              61.0       58.965517            36            85  ...   \n",
+       "6900               67.0       66.466667            44            97  ...   \n",
+       "4406               72.0       73.000000            61           103  ...   \n",
+       "4233               77.0       75.870968            47            81  ...   \n",
+       "3407               75.5       75.906250            37            89  ...   \n",
+       "5190               67.0       65.062500            45           106  ...   \n",
+       "1892               76.0       76.777778            59           107  ...   \n",
+       "10020              53.0       55.476190            41           117  ...   \n",
+       "9567               62.5       63.593750            45           100  ...   \n",
+       "12617              65.0       65.750000            21            79  ...   \n",
        "\n",
-       "       ScoreDiff max  ScoreDiff std  ScoreDiff median  ScoreDiff mean  \\\n",
-       "12348             35      16.997102              -1.0       -2.517241   \n",
-       "6900              29      14.772645              -5.0       -5.100000   \n",
-       "4406              18      12.316786              -2.0       -2.645161   \n",
-       "4233              34      13.022891              11.0       10.935484   \n",
-       "3407              47      17.828682              10.0       11.593750   \n",
-       "5190              18      14.365582              -7.0       -5.781250   \n",
-       "1892              27      14.449736              -3.0       -1.888889   \n",
-       "10020             24      27.245445             -15.0      -14.285714   \n",
-       "9567              17      13.277095              -2.0       -4.093750   \n",
-       "12617             57      18.777131              13.5       13.500000   \n",
+       "       ScoreDiff min  ScoreDiff max  ScoreDiff std  ScoreDiff median  \\\n",
+       "12348            -32             35      16.997102              -1.0   \n",
+       "6900             -49             29      14.772645              -5.0   \n",
+       "4406             -27             18      12.316786              -2.0   \n",
+       "4233             -17             34      13.022891              11.0   \n",
+       "3407             -20             47      17.828682              10.0   \n",
+       "5190             -62             18      14.365582              -7.0   \n",
+       "1892             -45             27      14.449736              -3.0   \n",
+       "10020            -93             24      27.245445             -15.0   \n",
+       "9567             -42             17      13.277095              -2.0   \n",
+       "12617            -23             57      18.777131              13.5   \n",
        "\n",
-       "       Win min  Win max   Win std  Win median  Win mean  \n",
-       "12348        0        1  0.508548         0.0  0.482759  \n",
-       "6900         0        1  0.479463         0.0  0.333333  \n",
-       "4406         0        1  0.508001         0.0  0.483871  \n",
-       "4233         0        1  0.401610         1.0  0.806452  \n",
-       "3407         0        1  0.456803         1.0  0.718750  \n",
-       "5190         0        1  0.470929         0.0  0.312500  \n",
-       "1892         0        1  0.492103         0.0  0.370370  \n",
-       "10020        0        1  0.462910         0.0  0.285714  \n",
-       "9567         0        1  0.504016         0.0  0.437500  \n",
-       "12617        0        1  0.456803         1.0  0.718750  \n",
+       "       ScoreDiff mean  Win min  Win max   Win std  Win median  Win mean  \n",
+       "12348       -2.517241        0        1  0.508548         0.0  0.482759  \n",
+       "6900        -5.100000        0        1  0.479463         0.0  0.333333  \n",
+       "4406        -2.645161        0        1  0.508001         0.0  0.483871  \n",
+       "4233        10.935484        0        1  0.401610         1.0  0.806452  \n",
+       "3407        11.593750        0        1  0.456803         1.0  0.718750  \n",
+       "5190        -5.781250        0        1  0.470929         0.0  0.312500  \n",
+       "1892        -1.888889        0        1  0.492103         0.0  0.370370  \n",
+       "10020      -14.285714        0        1  0.462910         0.0  0.285714  \n",
+       "9567        -4.093750        0        1  0.504016         0.0  0.437500  \n",
+       "12617       13.500000        0        1  0.456803         1.0  0.718750  \n",
        "\n",
-       "[10 rows x 173 columns]"
+       "[10 rows x 158 columns]"
       ]
      },
-     "execution_count": 95,
+     "execution_count": 16,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "exclude_agg_cols = {\n",
-    "    \"TeamID\",\n",
-    "    \"Season\",\n",
-    "    \"League\",\n",
-    "    \"GameResult\",\n",
-    "    \"OppLoc\",\n",
-    "    \"TeamLoc\",\n",
-    "}\n",
-    "\n",
-    "agg_funcs = [\n",
-    "    np.min,\n",
-    "    np.max,\n",
-    "    np.std,\n",
-    "    np.median,\n",
-    "    np.mean,\n",
-    "]\n",
-    "\n",
-    "# numeric_detailed_cols = detailed_team_results_df.select_dtypes(\"number\").columns\n",
-    "\n",
     "team_reg_agg = (\n",
     "    detailed_team_results_df.groupby([\"TeamID\", \"Season\", \"League\"])\n",
-    "    .agg({col: agg_funcs for col in detailed_team_results_df.select_dtypes(\"number\").columns})\n",
+    "    .agg({col: agg_funcs for col in detailed_team_results_df.select_dtypes(\"number\").columns if col not in exclude_agg_cols})\n",
     "    .reset_index()\n",
     ")\n",
     "\n",
@@ -975,13 +986,6 @@
     "\n",
     "team_reg_agg.sample(10, random_state=1)"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {