{ "cells": [ { "cell_type": "code", "execution_count": 479, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "\n", "# Assume you've already read all files into a list of DataFrames\n", "dataframes = [\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_torn_mcl.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_torn_hamstring.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_sprained_mcl.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_shoulder_sprain.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_shoulder_labrum.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_rotator_cuff.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_hip_labrum.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_hip_flexor_surgery.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_hip_flexor_strain.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_calf_strain.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_dislocated_shoulder.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_back_surgery.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_back_spasm.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_quad.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_meniscus.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_leg_fractured.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_hand_finger_fractured.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_hamstring.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_foot_sprain.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_foot_fracture.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_bone_spurs.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_arm.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_ankle_sprain.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_ankle_fracture (1).csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_acl (1).csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_achilles.csv\"),\n", " pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_ankle_fracture.csv\"),\n", "]\n", "\n", "# Concatenate all DataFrames\n", "combined_df = pd.concat(dataframes, ignore_index=True)\n", "\n", "\n", "\n" ] }, { "cell_type": "code", "execution_count": 480, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0.1Unnamed: 0InjuredNotesNameActivatedClassificationsTop ClassificationTop ScoreSpecific Injurydays_injured
00130482015-10-27placed on IL with torn left AchillesBrandon Jenningsnone{'sequence': 'placed on IL with torn left Achi...achilles tear0.475043left achilles tear injury250
11171642017-01-07placed on IL with Achilles injury / sprained a...Avery Bradleynone{'sequence': 'placed on IL with Achilles injur...achilles tear0.474971achilles tear injury250
22163362016-11-21placed on IL with sore/strained left AchillesTreveon Grahamnone{'sequence': 'placed on IL with sore/strained ...achilles tear0.472060left achilles tear injury250
33117242015-01-27placed on IL with torn left Achilles (out for ...Brandon Jenningsnone{'sequence': 'placed on IL with torn left Achi...achilles tear0.471717left achilles tear injury250
44169142016-12-23placed on IL with sore left AchillesJose Bareanone{'sequence': 'placed on IL with sore left Achi...achilles tear0.471639left achilles tear injury250
....................................
258258203802017-12-30strained right Achilles tendon (DTD)Austin Riversnone{'sequence': 'strained right Achilles tendon (...achilles tear0.456959right achilles tear injury250
259259150532016-03-12strained right Achilles tendon (DTD)Richaun Holmesnone{'sequence': 'strained right Achilles tendon (...achilles tear0.456959right achilles tear injury250
26026011322011-01-12Achilles tendon injury (DTD)Reggie Williamsnone{'sequence': 'Achilles tendon injury (DTD)', '...achilles tear0.456949achilles tear injury250
26126152010-10-08surgery to repair torn right Achilles tendonJonas Jerebkonone{'sequence': 'surgery to repair torn right Ach...achilles tear0.456810right achilles tear injury250
26226220502011-03-19right Achilles tendon injury (DTD)Al Harringtonnone{'sequence': 'right Achilles tendon injury (DT...achilles tear0.455742right achilles tear injury250
\n", "

263 rows × 11 columns

\n", "
" ], "text/plain": [ " Unnamed: 0.1 Unnamed: 0 Injured \\\n", "0 0 13048 2015-10-27 \n", "1 1 17164 2017-01-07 \n", "2 2 16336 2016-11-21 \n", "3 3 11724 2015-01-27 \n", "4 4 16914 2016-12-23 \n", ".. ... ... ... \n", "258 258 20380 2017-12-30 \n", "259 259 15053 2016-03-12 \n", "260 260 1132 2011-01-12 \n", "261 261 5 2010-10-08 \n", "262 262 2050 2011-03-19 \n", "\n", " Notes Name \\\n", "0 placed on IL with torn left Achilles Brandon Jennings \n", "1 placed on IL with Achilles injury / sprained a... Avery Bradley \n", "2 placed on IL with sore/strained left Achilles Treveon Graham \n", "3 placed on IL with torn left Achilles (out for ... Brandon Jennings \n", "4 placed on IL with sore left Achilles Jose Barea \n", ".. ... ... \n", "258 strained right Achilles tendon (DTD) Austin Rivers \n", "259 strained right Achilles tendon (DTD) Richaun Holmes \n", "260 Achilles tendon injury (DTD) Reggie Williams \n", "261 surgery to repair torn right Achilles tendon Jonas Jerebko \n", "262 right Achilles tendon injury (DTD) Al Harrington \n", "\n", " Activated Classifications \\\n", "0 none {'sequence': 'placed on IL with torn left Achi... \n", "1 none {'sequence': 'placed on IL with Achilles injur... \n", "2 none {'sequence': 'placed on IL with sore/strained ... \n", "3 none {'sequence': 'placed on IL with torn left Achi... \n", "4 none {'sequence': 'placed on IL with sore left Achi... \n", ".. ... ... \n", "258 none {'sequence': 'strained right Achilles tendon (... \n", "259 none {'sequence': 'strained right Achilles tendon (... \n", "260 none {'sequence': 'Achilles tendon injury (DTD)', '... \n", "261 none {'sequence': 'surgery to repair torn right Ach... \n", "262 none {'sequence': 'right Achilles tendon injury (DT... \n", "\n", " Top Classification Top Score Specific Injury days_injured \n", "0 achilles tear 0.475043 left achilles tear injury 250 \n", "1 achilles tear 0.474971 achilles tear injury 250 \n", "2 achilles tear 0.472060 left achilles tear injury 250 \n", "3 achilles tear 0.471717 left achilles tear injury 250 \n", "4 achilles tear 0.471639 left achilles tear injury 250 \n", ".. ... ... ... ... \n", "258 achilles tear 0.456959 right achilles tear injury 250 \n", "259 achilles tear 0.456959 right achilles tear injury 250 \n", "260 achilles tear 0.456949 achilles tear injury 250 \n", "261 achilles tear 0.456810 right achilles tear injury 250 \n", "262 achilles tear 0.455742 right achilles tear injury 250 \n", "\n", "[263 rows x 11 columns]" ] }, "execution_count": 480, "metadata": {}, "output_type": "execute_result" } ], "source": [ "achilles = pd.read_csv(\"/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_achilles.csv\")\n", "achilles" ] }, { "cell_type": "code", "execution_count": 481, "metadata": {}, "outputs": [ { "ename": "KeyError", "evalue": "\"['Unnamed: 0.2'] not found in axis\"", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mKeyError\u001b[0m Traceback (most recent call last)", "Cell \u001b[0;32mIn[481], line 2\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mnumpy\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mnp\u001b[39;00m\n\u001b[0;32m----> 2\u001b[0m achilles \u001b[38;5;241m=\u001b[39m achilles\u001b[38;5;241m.\u001b[39mdrop(columns\u001b[38;5;241m=\u001b[39m{\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mUnnamed: 0.2\u001b[39m\u001b[38;5;124m'\u001b[39m,\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mUnnamed: 0.1\u001b[39m\u001b[38;5;124m'\u001b[39m})\n", "File \u001b[0;32m/opt/anaconda3/lib/python3.12/site-packages/pandas/core/frame.py:5581\u001b[0m, in \u001b[0;36mDataFrame.drop\u001b[0;34m(self, labels, axis, index, columns, level, inplace, errors)\u001b[0m\n\u001b[1;32m 5433\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mdrop\u001b[39m(\n\u001b[1;32m 5434\u001b[0m \u001b[38;5;28mself\u001b[39m,\n\u001b[1;32m 5435\u001b[0m labels: IndexLabel \u001b[38;5;241m|\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 5442\u001b[0m errors: IgnoreRaise \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mraise\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[1;32m 5443\u001b[0m ) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m DataFrame \u001b[38;5;241m|\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m 5444\u001b[0m \u001b[38;5;250m \u001b[39m\u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m 5445\u001b[0m \u001b[38;5;124;03m Drop specified labels from rows or columns.\u001b[39;00m\n\u001b[1;32m 5446\u001b[0m \n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 5579\u001b[0m \u001b[38;5;124;03m weight 1.0 0.8\u001b[39;00m\n\u001b[1;32m 5580\u001b[0m \u001b[38;5;124;03m \"\"\"\u001b[39;00m\n\u001b[0;32m-> 5581\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28msuper\u001b[39m()\u001b[38;5;241m.\u001b[39mdrop(\n\u001b[1;32m 5582\u001b[0m labels\u001b[38;5;241m=\u001b[39mlabels,\n\u001b[1;32m 5583\u001b[0m axis\u001b[38;5;241m=\u001b[39maxis,\n\u001b[1;32m 5584\u001b[0m index\u001b[38;5;241m=\u001b[39mindex,\n\u001b[1;32m 5585\u001b[0m columns\u001b[38;5;241m=\u001b[39mcolumns,\n\u001b[1;32m 5586\u001b[0m level\u001b[38;5;241m=\u001b[39mlevel,\n\u001b[1;32m 5587\u001b[0m inplace\u001b[38;5;241m=\u001b[39minplace,\n\u001b[1;32m 5588\u001b[0m errors\u001b[38;5;241m=\u001b[39merrors,\n\u001b[1;32m 5589\u001b[0m )\n", "File \u001b[0;32m/opt/anaconda3/lib/python3.12/site-packages/pandas/core/generic.py:4788\u001b[0m, in \u001b[0;36mNDFrame.drop\u001b[0;34m(self, labels, axis, index, columns, level, inplace, errors)\u001b[0m\n\u001b[1;32m 4786\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m axis, labels \u001b[38;5;129;01min\u001b[39;00m axes\u001b[38;5;241m.\u001b[39mitems():\n\u001b[1;32m 4787\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m labels \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m-> 4788\u001b[0m obj \u001b[38;5;241m=\u001b[39m obj\u001b[38;5;241m.\u001b[39m_drop_axis(labels, axis, level\u001b[38;5;241m=\u001b[39mlevel, errors\u001b[38;5;241m=\u001b[39merrors)\n\u001b[1;32m 4790\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m inplace:\n\u001b[1;32m 4791\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_update_inplace(obj)\n", "File \u001b[0;32m/opt/anaconda3/lib/python3.12/site-packages/pandas/core/generic.py:4830\u001b[0m, in \u001b[0;36mNDFrame._drop_axis\u001b[0;34m(self, labels, axis, level, errors, only_slice)\u001b[0m\n\u001b[1;32m 4828\u001b[0m new_axis \u001b[38;5;241m=\u001b[39m axis\u001b[38;5;241m.\u001b[39mdrop(labels, level\u001b[38;5;241m=\u001b[39mlevel, errors\u001b[38;5;241m=\u001b[39merrors)\n\u001b[1;32m 4829\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 4830\u001b[0m new_axis \u001b[38;5;241m=\u001b[39m axis\u001b[38;5;241m.\u001b[39mdrop(labels, errors\u001b[38;5;241m=\u001b[39merrors)\n\u001b[1;32m 4831\u001b[0m indexer \u001b[38;5;241m=\u001b[39m axis\u001b[38;5;241m.\u001b[39mget_indexer(new_axis)\n\u001b[1;32m 4833\u001b[0m \u001b[38;5;66;03m# Case for non-unique axis\u001b[39;00m\n\u001b[1;32m 4834\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n", "File \u001b[0;32m/opt/anaconda3/lib/python3.12/site-packages/pandas/core/indexes/base.py:7070\u001b[0m, in \u001b[0;36mIndex.drop\u001b[0;34m(self, labels, errors)\u001b[0m\n\u001b[1;32m 7068\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m mask\u001b[38;5;241m.\u001b[39many():\n\u001b[1;32m 7069\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m errors \u001b[38;5;241m!=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mignore\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[0;32m-> 7070\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mlabels[mask]\u001b[38;5;241m.\u001b[39mtolist()\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m not found in axis\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m 7071\u001b[0m indexer \u001b[38;5;241m=\u001b[39m indexer[\u001b[38;5;241m~\u001b[39mmask]\n\u001b[1;32m 7072\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdelete(indexer)\n", "\u001b[0;31mKeyError\u001b[0m: \"['Unnamed: 0.2'] not found in axis\"" ] } ], "source": [ "import numpy as np\n", "achilles = achilles.drop(columns={'Unnamed: 0.2','Unnamed: 0.1'})\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "ename": "KeyError", "evalue": "\"['Out_For_The_Season', 'Has_Injury'] not found in axis\"", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mKeyError\u001b[0m Traceback (most recent call last)", "Cell \u001b[0;32mIn[455], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m achilles \u001b[38;5;241m=\u001b[39m achilles\u001b[38;5;241m.\u001b[39mdrop(columns\u001b[38;5;241m=\u001b[39m{\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mOut_For_The_Season\u001b[39m\u001b[38;5;124m'\u001b[39m,\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mHas_Injury\u001b[39m\u001b[38;5;124m'\u001b[39m})\n\u001b[1;32m 3\u001b[0m achilles\u001b[38;5;241m.\u001b[39mto_csv(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_achilles.csv\u001b[39m\u001b[38;5;124m'\u001b[39m)\n", "File \u001b[0;32m/opt/anaconda3/lib/python3.12/site-packages/pandas/core/frame.py:5581\u001b[0m, in \u001b[0;36mDataFrame.drop\u001b[0;34m(self, labels, axis, index, columns, level, inplace, errors)\u001b[0m\n\u001b[1;32m 5433\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mdrop\u001b[39m(\n\u001b[1;32m 5434\u001b[0m \u001b[38;5;28mself\u001b[39m,\n\u001b[1;32m 5435\u001b[0m labels: IndexLabel \u001b[38;5;241m|\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 5442\u001b[0m errors: IgnoreRaise \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mraise\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[1;32m 5443\u001b[0m ) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m DataFrame \u001b[38;5;241m|\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m 5444\u001b[0m \u001b[38;5;250m \u001b[39m\u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m 5445\u001b[0m \u001b[38;5;124;03m Drop specified labels from rows or columns.\u001b[39;00m\n\u001b[1;32m 5446\u001b[0m \n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 5579\u001b[0m \u001b[38;5;124;03m weight 1.0 0.8\u001b[39;00m\n\u001b[1;32m 5580\u001b[0m \u001b[38;5;124;03m \"\"\"\u001b[39;00m\n\u001b[0;32m-> 5581\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28msuper\u001b[39m()\u001b[38;5;241m.\u001b[39mdrop(\n\u001b[1;32m 5582\u001b[0m labels\u001b[38;5;241m=\u001b[39mlabels,\n\u001b[1;32m 5583\u001b[0m axis\u001b[38;5;241m=\u001b[39maxis,\n\u001b[1;32m 5584\u001b[0m index\u001b[38;5;241m=\u001b[39mindex,\n\u001b[1;32m 5585\u001b[0m columns\u001b[38;5;241m=\u001b[39mcolumns,\n\u001b[1;32m 5586\u001b[0m level\u001b[38;5;241m=\u001b[39mlevel,\n\u001b[1;32m 5587\u001b[0m inplace\u001b[38;5;241m=\u001b[39minplace,\n\u001b[1;32m 5588\u001b[0m errors\u001b[38;5;241m=\u001b[39merrors,\n\u001b[1;32m 5589\u001b[0m )\n", "File \u001b[0;32m/opt/anaconda3/lib/python3.12/site-packages/pandas/core/generic.py:4788\u001b[0m, in \u001b[0;36mNDFrame.drop\u001b[0;34m(self, labels, axis, index, columns, level, inplace, errors)\u001b[0m\n\u001b[1;32m 4786\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m axis, labels \u001b[38;5;129;01min\u001b[39;00m axes\u001b[38;5;241m.\u001b[39mitems():\n\u001b[1;32m 4787\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m labels \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m-> 4788\u001b[0m obj \u001b[38;5;241m=\u001b[39m obj\u001b[38;5;241m.\u001b[39m_drop_axis(labels, axis, level\u001b[38;5;241m=\u001b[39mlevel, errors\u001b[38;5;241m=\u001b[39merrors)\n\u001b[1;32m 4790\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m inplace:\n\u001b[1;32m 4791\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_update_inplace(obj)\n", "File \u001b[0;32m/opt/anaconda3/lib/python3.12/site-packages/pandas/core/generic.py:4830\u001b[0m, in \u001b[0;36mNDFrame._drop_axis\u001b[0;34m(self, labels, axis, level, errors, only_slice)\u001b[0m\n\u001b[1;32m 4828\u001b[0m new_axis \u001b[38;5;241m=\u001b[39m axis\u001b[38;5;241m.\u001b[39mdrop(labels, level\u001b[38;5;241m=\u001b[39mlevel, errors\u001b[38;5;241m=\u001b[39merrors)\n\u001b[1;32m 4829\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 4830\u001b[0m new_axis \u001b[38;5;241m=\u001b[39m axis\u001b[38;5;241m.\u001b[39mdrop(labels, errors\u001b[38;5;241m=\u001b[39merrors)\n\u001b[1;32m 4831\u001b[0m indexer \u001b[38;5;241m=\u001b[39m axis\u001b[38;5;241m.\u001b[39mget_indexer(new_axis)\n\u001b[1;32m 4833\u001b[0m \u001b[38;5;66;03m# Case for non-unique axis\u001b[39;00m\n\u001b[1;32m 4834\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n", "File \u001b[0;32m/opt/anaconda3/lib/python3.12/site-packages/pandas/core/indexes/base.py:7070\u001b[0m, in \u001b[0;36mIndex.drop\u001b[0;34m(self, labels, errors)\u001b[0m\n\u001b[1;32m 7068\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m mask\u001b[38;5;241m.\u001b[39many():\n\u001b[1;32m 7069\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m errors \u001b[38;5;241m!=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mignore\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[0;32m-> 7070\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mlabels[mask]\u001b[38;5;241m.\u001b[39mtolist()\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m not found in axis\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m 7071\u001b[0m indexer \u001b[38;5;241m=\u001b[39m indexer[\u001b[38;5;241m~\u001b[39mmask]\n\u001b[1;32m 7072\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdelete(indexer)\n", "\u001b[0;31mKeyError\u001b[0m: \"['Out_For_The_Season', 'Has_Injury'] not found in axis\"" ] } ], "source": [ "achilles = achilles.drop(columns={'Out_For_The_Season','Has_Injury'})\n", "\n", "achilles.to_csv('/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_achilles.csv')\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "achilles = achilles.rename(columns={'injured':'Injured','Activated_From_IL':'Activated'})\n", "achilles['Activated'] = 'none'\n", "achilles['days_injured'] = 250\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0.1Unnamed: 0InjuredNotesNameActivatedClassificationsTop ClassificationTop ScoreSpecific Injurydays_injured
00130482015-10-27placed on IL with torn left AchillesBrandon Jenningsnone{'sequence': 'placed on IL with torn left Achi...achilles tear0.475043left achilles tear injury250
11171642017-01-07placed on IL with Achilles injury / sprained a...Avery Bradleynone{'sequence': 'placed on IL with Achilles injur...achilles tear0.474971achilles tear injury250
22163362016-11-21placed on IL with sore/strained left AchillesTreveon Grahamnone{'sequence': 'placed on IL with sore/strained ...achilles tear0.472060left achilles tear injury250
33117242015-01-27placed on IL with torn left Achilles (out for ...Brandon Jenningsnone{'sequence': 'placed on IL with torn left Achi...achilles tear0.471717left achilles tear injury250
44169142016-12-23placed on IL with sore left AchillesJose Bareanone{'sequence': 'placed on IL with sore left Achi...achilles tear0.471639left achilles tear injury250
....................................
258258203802017-12-30strained right Achilles tendon (DTD)Austin Riversnone{'sequence': 'strained right Achilles tendon (...achilles tear0.456959right achilles tear injury250
259259150532016-03-12strained right Achilles tendon (DTD)Richaun Holmesnone{'sequence': 'strained right Achilles tendon (...achilles tear0.456959right achilles tear injury250
26026011322011-01-12Achilles tendon injury (DTD)Reggie Williamsnone{'sequence': 'Achilles tendon injury (DTD)', '...achilles tear0.456949achilles tear injury250
26126152010-10-08surgery to repair torn right Achilles tendonJonas Jerebkonone{'sequence': 'surgery to repair torn right Ach...achilles tear0.456810right achilles tear injury250
26226220502011-03-19right Achilles tendon injury (DTD)Al Harringtonnone{'sequence': 'right Achilles tendon injury (DT...achilles tear0.455742right achilles tear injury250
\n", "

263 rows × 11 columns

\n", "
" ], "text/plain": [ " Unnamed: 0.1 Unnamed: 0 Injured \\\n", "0 0 13048 2015-10-27 \n", "1 1 17164 2017-01-07 \n", "2 2 16336 2016-11-21 \n", "3 3 11724 2015-01-27 \n", "4 4 16914 2016-12-23 \n", ".. ... ... ... \n", "258 258 20380 2017-12-30 \n", "259 259 15053 2016-03-12 \n", "260 260 1132 2011-01-12 \n", "261 261 5 2010-10-08 \n", "262 262 2050 2011-03-19 \n", "\n", " Notes Name \\\n", "0 placed on IL with torn left Achilles Brandon Jennings \n", "1 placed on IL with Achilles injury / sprained a... Avery Bradley \n", "2 placed on IL with sore/strained left Achilles Treveon Graham \n", "3 placed on IL with torn left Achilles (out for ... Brandon Jennings \n", "4 placed on IL with sore left Achilles Jose Barea \n", ".. ... ... \n", "258 strained right Achilles tendon (DTD) Austin Rivers \n", "259 strained right Achilles tendon (DTD) Richaun Holmes \n", "260 Achilles tendon injury (DTD) Reggie Williams \n", "261 surgery to repair torn right Achilles tendon Jonas Jerebko \n", "262 right Achilles tendon injury (DTD) Al Harrington \n", "\n", " Activated Classifications \\\n", "0 none {'sequence': 'placed on IL with torn left Achi... \n", "1 none {'sequence': 'placed on IL with Achilles injur... \n", "2 none {'sequence': 'placed on IL with sore/strained ... \n", "3 none {'sequence': 'placed on IL with torn left Achi... \n", "4 none {'sequence': 'placed on IL with sore left Achi... \n", ".. ... ... \n", "258 none {'sequence': 'strained right Achilles tendon (... \n", "259 none {'sequence': 'strained right Achilles tendon (... \n", "260 none {'sequence': 'Achilles tendon injury (DTD)', '... \n", "261 none {'sequence': 'surgery to repair torn right Ach... \n", "262 none {'sequence': 'right Achilles tendon injury (DT... \n", "\n", " Top Classification Top Score Specific Injury days_injured \n", "0 achilles tear 0.475043 left achilles tear injury 250 \n", "1 achilles tear 0.474971 achilles tear injury 250 \n", "2 achilles tear 0.472060 left achilles tear injury 250 \n", "3 achilles tear 0.471717 left achilles tear injury 250 \n", "4 achilles tear 0.471639 left achilles tear injury 250 \n", ".. ... ... ... ... \n", "258 achilles tear 0.456959 right achilles tear injury 250 \n", "259 achilles tear 0.456959 right achilles tear injury 250 \n", "260 achilles tear 0.456949 achilles tear injury 250 \n", "261 achilles tear 0.456810 right achilles tear injury 250 \n", "262 achilles tear 0.455742 right achilles tear injury 250 \n", "\n", "[263 rows x 11 columns]" ] }, "execution_count": 457, "metadata": {}, "output_type": "execute_result" } ], "source": [ "achilles" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "new_order = ['Unnamed: 0','Name','Notes','Injured','Activated','days_injured','Classifications','Top Classification','Top Score','Specific Injury']\n", "achilles = achilles[new_order]" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/var/folders/8t/t11lp0b952n0xtfmnwbzxzvw0000gn/T/ipykernel_11952/3813116858.py:1: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", " achilles['Injured'] = pd.to_datetime(achilles['Injured'], errors='coerce')\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0NameNotesInjuredActivateddays_injuredClassificationsTop ClassificationTop ScoreSpecific Injury
013048Brandon Jenningsplaced on IL with torn left Achilles2015-10-27none250{'sequence': 'placed on IL with torn left Achi...achilles tear0.475043left achilles tear injury
117164Avery Bradleyplaced on IL with Achilles injury / sprained a...2017-01-07none250{'sequence': 'placed on IL with Achilles injur...achilles tear0.474971achilles tear injury
216336Treveon Grahamplaced on IL with sore/strained left Achilles2016-11-21none250{'sequence': 'placed on IL with sore/strained ...achilles tear0.472060left achilles tear injury
311724Brandon Jenningsplaced on IL with torn left Achilles (out for ...2015-01-27none250{'sequence': 'placed on IL with torn left Achi...achilles tear0.471717left achilles tear injury
416914Jose Bareaplaced on IL with sore left Achilles2016-12-23none250{'sequence': 'placed on IL with sore left Achi...achilles tear0.471639left achilles tear injury
.................................
25820380Austin Riversstrained right Achilles tendon (DTD)2017-12-30none250{'sequence': 'strained right Achilles tendon (...achilles tear0.456959right achilles tear injury
25915053Richaun Holmesstrained right Achilles tendon (DTD)2016-03-12none250{'sequence': 'strained right Achilles tendon (...achilles tear0.456959right achilles tear injury
2601132Reggie WilliamsAchilles tendon injury (DTD)2011-01-12none250{'sequence': 'Achilles tendon injury (DTD)', '...achilles tear0.456949achilles tear injury
2615Jonas Jerebkosurgery to repair torn right Achilles tendon2010-10-08none250{'sequence': 'surgery to repair torn right Ach...achilles tear0.456810right achilles tear injury
2622050Al Harringtonright Achilles tendon injury (DTD)2011-03-19none250{'sequence': 'right Achilles tendon injury (DT...achilles tear0.455742right achilles tear injury
\n", "

263 rows × 10 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 Name \\\n", "0 13048 Brandon Jennings \n", "1 17164 Avery Bradley \n", "2 16336 Treveon Graham \n", "3 11724 Brandon Jennings \n", "4 16914 Jose Barea \n", ".. ... ... \n", "258 20380 Austin Rivers \n", "259 15053 Richaun Holmes \n", "260 1132 Reggie Williams \n", "261 5 Jonas Jerebko \n", "262 2050 Al Harrington \n", "\n", " Notes Injured Activated \\\n", "0 placed on IL with torn left Achilles 2015-10-27 none \n", "1 placed on IL with Achilles injury / sprained a... 2017-01-07 none \n", "2 placed on IL with sore/strained left Achilles 2016-11-21 none \n", "3 placed on IL with torn left Achilles (out for ... 2015-01-27 none \n", "4 placed on IL with sore left Achilles 2016-12-23 none \n", ".. ... ... ... \n", "258 strained right Achilles tendon (DTD) 2017-12-30 none \n", "259 strained right Achilles tendon (DTD) 2016-03-12 none \n", "260 Achilles tendon injury (DTD) 2011-01-12 none \n", "261 surgery to repair torn right Achilles tendon 2010-10-08 none \n", "262 right Achilles tendon injury (DTD) 2011-03-19 none \n", "\n", " days_injured Classifications \\\n", "0 250 {'sequence': 'placed on IL with torn left Achi... \n", "1 250 {'sequence': 'placed on IL with Achilles injur... \n", "2 250 {'sequence': 'placed on IL with sore/strained ... \n", "3 250 {'sequence': 'placed on IL with torn left Achi... \n", "4 250 {'sequence': 'placed on IL with sore left Achi... \n", ".. ... ... \n", "258 250 {'sequence': 'strained right Achilles tendon (... \n", "259 250 {'sequence': 'strained right Achilles tendon (... \n", "260 250 {'sequence': 'Achilles tendon injury (DTD)', '... \n", "261 250 {'sequence': 'surgery to repair torn right Ach... \n", "262 250 {'sequence': 'right Achilles tendon injury (DT... \n", "\n", " Top Classification Top Score Specific Injury \n", "0 achilles tear 0.475043 left achilles tear injury \n", "1 achilles tear 0.474971 achilles tear injury \n", "2 achilles tear 0.472060 left achilles tear injury \n", "3 achilles tear 0.471717 left achilles tear injury \n", "4 achilles tear 0.471639 left achilles tear injury \n", ".. ... ... ... \n", "258 achilles tear 0.456959 right achilles tear injury \n", "259 achilles tear 0.456959 right achilles tear injury \n", "260 achilles tear 0.456949 achilles tear injury \n", "261 achilles tear 0.456810 right achilles tear injury \n", "262 achilles tear 0.455742 right achilles tear injury \n", "\n", "[263 rows x 10 columns]" ] }, "execution_count": 459, "metadata": {}, "output_type": "execute_result" } ], "source": [ "achilles['Injured'] = pd.to_datetime(achilles['Injured'], errors='coerce')\n", "achilles" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "achilles.to_csv('/Users/laraschuman/Desktop/CTP-Project/combined_sorted_notes/sorted_notes_with_injury_achilles.csv')\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0injuredNotesNameActivatedClassificationsTop ClassificationTop ScoreSpecific Injury
01304827/10/2015placed on IL with torn left AchillesBrandon JenningsFalse{'sequence': 'placed on IL with torn left Achi...achilles tear0.475043left achilles tear injury
11716407/01/2017placed on IL with Achilles injury / sprained a...Avery BradleyFalse{'sequence': 'placed on IL with Achilles injur...achilles tear0.474971achilles tear injury
21633621/11/2016placed on IL with sore/strained left AchillesTreveon GrahamFalse{'sequence': 'placed on IL with sore/strained ...achilles tear0.472060left achilles tear injury
31172427/01/2015placed on IL with torn left Achilles (out for ...Brandon JenningsFalse{'sequence': 'placed on IL with torn left Achi...achilles tear0.471717left achilles tear injury
41691423/12/2016placed on IL with sore left AchillesJose BareaFalse{'sequence': 'placed on IL with sore left Achi...achilles tear0.471639left achilles tear injury
..............................
2582038030/12/2017strained right Achilles tendon (DTD)Austin RiversFalse{'sequence': 'strained right Achilles tendon (...achilles tear0.456959right achilles tear injury
2591505312/03/2016strained right Achilles tendon (DTD)Richaun HolmesFalse{'sequence': 'strained right Achilles tendon (...achilles tear0.456959right achilles tear injury
260113212/01/2011Achilles tendon injury (DTD)Reggie WilliamsFalse{'sequence': 'Achilles tendon injury (DTD)', '...achilles tear0.456949achilles tear injury
261508/10/2010surgery to repair torn right Achilles tendonJonas JerebkoFalse{'sequence': 'surgery to repair torn right Ach...achilles tear0.456810right achilles tear injury
262205019/03/2011right Achilles tendon injury (DTD)Al HarringtonFalse{'sequence': 'right Achilles tendon injury (DT...achilles tear0.455742right achilles tear injury
\n", "

263 rows × 9 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 injured \\\n", "0 13048 27/10/2015 \n", "1 17164 07/01/2017 \n", "2 16336 21/11/2016 \n", "3 11724 27/01/2015 \n", "4 16914 23/12/2016 \n", ".. ... ... \n", "258 20380 30/12/2017 \n", "259 15053 12/03/2016 \n", "260 1132 12/01/2011 \n", "261 5 08/10/2010 \n", "262 2050 19/03/2011 \n", "\n", " Notes Name \\\n", "0 placed on IL with torn left Achilles Brandon Jennings \n", "1 placed on IL with Achilles injury / sprained a... Avery Bradley \n", "2 placed on IL with sore/strained left Achilles Treveon Graham \n", "3 placed on IL with torn left Achilles (out for ... Brandon Jennings \n", "4 placed on IL with sore left Achilles Jose Barea \n", ".. ... ... \n", "258 strained right Achilles tendon (DTD) Austin Rivers \n", "259 strained right Achilles tendon (DTD) Richaun Holmes \n", "260 Achilles tendon injury (DTD) Reggie Williams \n", "261 surgery to repair torn right Achilles tendon Jonas Jerebko \n", "262 right Achilles tendon injury (DTD) Al Harrington \n", "\n", " Activated Classifications \\\n", "0 False {'sequence': 'placed on IL with torn left Achi... \n", "1 False {'sequence': 'placed on IL with Achilles injur... \n", "2 False {'sequence': 'placed on IL with sore/strained ... \n", "3 False {'sequence': 'placed on IL with torn left Achi... \n", "4 False {'sequence': 'placed on IL with sore left Achi... \n", ".. ... ... \n", "258 False {'sequence': 'strained right Achilles tendon (... \n", "259 False {'sequence': 'strained right Achilles tendon (... \n", "260 False {'sequence': 'Achilles tendon injury (DTD)', '... \n", "261 False {'sequence': 'surgery to repair torn right Ach... \n", "262 False {'sequence': 'right Achilles tendon injury (DT... \n", "\n", " Top Classification Top Score Specific Injury \n", "0 achilles tear 0.475043 left achilles tear injury \n", "1 achilles tear 0.474971 achilles tear injury \n", "2 achilles tear 0.472060 left achilles tear injury \n", "3 achilles tear 0.471717 left achilles tear injury \n", "4 achilles tear 0.471639 left achilles tear injury \n", ".. ... ... ... \n", "258 achilles tear 0.456959 right achilles tear injury \n", "259 achilles tear 0.456959 right achilles tear injury \n", "260 achilles tear 0.456949 achilles tear injury \n", "261 achilles tear 0.456810 right achilles tear injury \n", "262 achilles tear 0.455742 right achilles tear injury \n", "\n", "[263 rows x 9 columns]" ] }, "execution_count": 396, "metadata": {}, "output_type": "execute_result" } ], "source": [ "achilles" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0NameNotesInjuredActivateddays_injuredClassificationsTop ClassificationTop ScoreSpecific InjuryUnnamed: 0.1
045518Sergey Karasevsurgery on right knee to repair torn MCL, disl...2015-03-122015-12-08271.0{'sequence': 'surgery on right knee to repair ...torn mcl0.467165right torn mcl injuryNaN
146401Sergey Karasevsurgery on right knee to repair torn MCL, disl...2015-03-122015-12-08271.0{'sequence': 'surgery on right knee to repair ...torn mcl0.467165right torn mcl injuryNaN
250960Sergey Karasevsurgery on right knee to repair torn MCL, disl...2015-03-122015-12-08271.0{'sequence': 'surgery on right knee to repair ...torn mcl0.467165right torn mcl injuryNaN
354657Sergey Karasevsurgery on right knee to repair torn MCL, disl...2015-03-122015-12-08271.0{'sequence': 'surgery on right knee to repair ...torn mcl0.467165right torn mcl injuryNaN
455315Sergey Karasevsurgery on right knee to repair torn MCL, disl...2015-03-122015-12-08271.0{'sequence': 'surgery on right knee to repair ...torn mcl0.467165right torn mcl injuryNaN
....................................
328746878Shane Larkinfractured right ankle (out indefinitely)2013-07-122013-11-18129.0{'sequence': 'fractured right ankle (out indef...ankle fracture0.455315right ankle fracture injuryNaN
3288203837Shane Larkinfractured right ankle (out indefinitely)2013-07-122013-11-18129.0{'sequence': 'fractured right ankle (out indef...ankle fracture0.455315right ankle fracture injuryNaN
328983193Shane Larkinfractured right ankle (out indefinitely)2013-07-122013-11-18129.0{'sequence': 'fractured right ankle (out indef...ankle fracture0.455315right ankle fracture injuryNaN
3290107300Shane Larkinfractured right ankle (out indefinitely)2013-07-122013-11-18129.0{'sequence': 'fractured right ankle (out indef...ankle fracture0.455315right ankle fracture injuryNaN
3291184549Shane Larkinfractured right ankle (out indefinitely)2013-07-122013-11-18129.0{'sequence': 'fractured right ankle (out indef...ankle fracture0.455315right ankle fracture injuryNaN
\n", "

3292 rows × 11 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 Name \\\n", "0 45518 Sergey Karasev \n", "1 46401 Sergey Karasev \n", "2 50960 Sergey Karasev \n", "3 54657 Sergey Karasev \n", "4 55315 Sergey Karasev \n", "... ... ... \n", "3287 46878 Shane Larkin \n", "3288 203837 Shane Larkin \n", "3289 83193 Shane Larkin \n", "3290 107300 Shane Larkin \n", "3291 184549 Shane Larkin \n", "\n", " Notes Injured \\\n", "0 surgery on right knee to repair torn MCL, disl... 2015-03-12 \n", "1 surgery on right knee to repair torn MCL, disl... 2015-03-12 \n", "2 surgery on right knee to repair torn MCL, disl... 2015-03-12 \n", "3 surgery on right knee to repair torn MCL, disl... 2015-03-12 \n", "4 surgery on right knee to repair torn MCL, disl... 2015-03-12 \n", "... ... ... \n", "3287 fractured right ankle (out indefinitely) 2013-07-12 \n", "3288 fractured right ankle (out indefinitely) 2013-07-12 \n", "3289 fractured right ankle (out indefinitely) 2013-07-12 \n", "3290 fractured right ankle (out indefinitely) 2013-07-12 \n", "3291 fractured right ankle (out indefinitely) 2013-07-12 \n", "\n", " Activated days_injured \\\n", "0 2015-12-08 271.0 \n", "1 2015-12-08 271.0 \n", "2 2015-12-08 271.0 \n", "3 2015-12-08 271.0 \n", "4 2015-12-08 271.0 \n", "... ... ... \n", "3287 2013-11-18 129.0 \n", "3288 2013-11-18 129.0 \n", "3289 2013-11-18 129.0 \n", "3290 2013-11-18 129.0 \n", "3291 2013-11-18 129.0 \n", "\n", " Classifications Top Classification \\\n", "0 {'sequence': 'surgery on right knee to repair ... torn mcl \n", "1 {'sequence': 'surgery on right knee to repair ... torn mcl \n", "2 {'sequence': 'surgery on right knee to repair ... torn mcl \n", "3 {'sequence': 'surgery on right knee to repair ... torn mcl \n", "4 {'sequence': 'surgery on right knee to repair ... torn mcl \n", "... ... ... \n", "3287 {'sequence': 'fractured right ankle (out indef... ankle fracture \n", "3288 {'sequence': 'fractured right ankle (out indef... ankle fracture \n", "3289 {'sequence': 'fractured right ankle (out indef... ankle fracture \n", "3290 {'sequence': 'fractured right ankle (out indef... ankle fracture \n", "3291 {'sequence': 'fractured right ankle (out indef... ankle fracture \n", "\n", " Top Score Specific Injury Unnamed: 0.1 \n", "0 0.467165 right torn mcl injury NaN \n", "1 0.467165 right torn mcl injury NaN \n", "2 0.467165 right torn mcl injury NaN \n", "3 0.467165 right torn mcl injury NaN \n", "4 0.467165 right torn mcl injury NaN \n", "... ... ... ... \n", "3287 0.455315 right ankle fracture injury NaN \n", "3288 0.455315 right ankle fracture injury NaN \n", "3289 0.455315 right ankle fracture injury NaN \n", "3290 0.455315 right ankle fracture injury NaN \n", "3291 0.455315 right ankle fracture injury NaN \n", "\n", "[3292 rows x 11 columns]" ] }, "execution_count": 460, "metadata": {}, "output_type": "execute_result" } ], "source": [ "combined_df" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "combined_df = combined_df.drop(columns=['Unnamed: 0','Classifications','Notes','Top Classification','Top Score'])\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Name Injured Activated days_injured Specific Injury \n", "Andrew Bogut 2012-01-26 2012-11-05 284.0 left ankle fracture injury 58\n", " Tony Parker 2017-05-04 2017-11-27 207.0 left quad injury injury 49\n", "Derrick Rose 2017-04-05 2017-10-29 207.0 left meniscus tear injury 39\n", "Chandler Parsons 2017-03-13 2017-12-01 263.0 left meniscus tear injury 39\n", " 2016-03-25 2016-11-06 226.0 right meniscus tear injury 39\n", " ..\n", "Alan Williams 2017-09-25 2018-03-26 182.0 right meniscus tear injury 3\n", "Michael Frazier II 2020-03-10 2020-07-31 143.0 left arm injury injury 2\n", "Jason Thompson 2011-06-15 2012-03-28 287.0 right foot fracture injury 2\n", "Ronnie Brewer 2012-09-07 2013-02-27 173.0 left meniscus tear injury 1\n", "Justin Patton 2017-07-04 2018-02-24 235.0 left foot fracture injury 1\n", "Name: count, Length: 187, dtype: int64" ] }, "execution_count": 39, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
NameInjuredActivateddays_injuredSpecific InjuryUnnamed: 0.1
0Sergey Karasev2015-03-122015-12-08271.0right torn mcl injuryNaN
1Sergey Karasev2015-03-122015-12-08271.0right torn mcl injuryNaN
2Sergey Karasev2015-03-122015-12-08271.0right torn mcl injuryNaN
3Sergey Karasev2015-03-122015-12-08271.0right torn mcl injuryNaN
4Sergey Karasev2015-03-122015-12-08271.0right torn mcl injuryNaN
.....................
3287Shane Larkin2013-07-122013-11-18129.0right ankle fracture injuryNaN
3288Shane Larkin2013-07-122013-11-18129.0right ankle fracture injuryNaN
3289Shane Larkin2013-07-122013-11-18129.0right ankle fracture injuryNaN
3290Shane Larkin2013-07-122013-11-18129.0right ankle fracture injuryNaN
3291Shane Larkin2013-07-122013-11-18129.0right ankle fracture injuryNaN
\n", "

3292 rows × 6 columns

\n", "
" ], "text/plain": [ " Name Injured Activated days_injured \\\n", "0 Sergey Karasev 2015-03-12 2015-12-08 271.0 \n", "1 Sergey Karasev 2015-03-12 2015-12-08 271.0 \n", "2 Sergey Karasev 2015-03-12 2015-12-08 271.0 \n", "3 Sergey Karasev 2015-03-12 2015-12-08 271.0 \n", "4 Sergey Karasev 2015-03-12 2015-12-08 271.0 \n", "... ... ... ... ... \n", "3287 Shane Larkin 2013-07-12 2013-11-18 129.0 \n", "3288 Shane Larkin 2013-07-12 2013-11-18 129.0 \n", "3289 Shane Larkin 2013-07-12 2013-11-18 129.0 \n", "3290 Shane Larkin 2013-07-12 2013-11-18 129.0 \n", "3291 Shane Larkin 2013-07-12 2013-11-18 129.0 \n", "\n", " Specific Injury Unnamed: 0.1 \n", "0 right torn mcl injury NaN \n", "1 right torn mcl injury NaN \n", "2 right torn mcl injury NaN \n", "3 right torn mcl injury NaN \n", "4 right torn mcl injury NaN \n", "... ... ... \n", "3287 right ankle fracture injury NaN \n", "3288 right ankle fracture injury NaN \n", "3289 right ankle fracture injury NaN \n", "3290 right ankle fracture injury NaN \n", "3291 right ankle fracture injury NaN \n", "\n", "[3292 rows x 6 columns]" ] }, "execution_count": 463, "metadata": {}, "output_type": "execute_result" } ], "source": [ "combined_df = combined_df.rename(columns={'Specfic Injury': 'Injury'})\n", "combined_df\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "combined_df = combined_df.drop(columns='Unnamed: 0.1')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
NameInjuredActivateddays_injuredSpecific Injury
0Sergey Karasev2015-03-122015-12-08271.0right torn mcl injury
14Jordan Farmar2013-12-022014-04-06125.0left torn hamstring injury
16Khris Middleton2016-09-212017-02-08140.0left torn hamstring injury
34Khris Middleton2016-09-292017-02-08132.0left torn hamstring injury
41Solomon Hill2017-08-282018-03-18202.0left torn hamstring injury
..................
3216Jerome Robinson2020-03-08none250.0left achilles tear injury
3219Eric Bledsoe2019-02-02none250.0left achilles tear injury
3223Isaiah Hartenstein2019-03-11none250.0right achilles tear injury
3227Rudy Fernandez2012-01-21none250.0right achilles tear injury
3230Reggie Williams2011-01-12none250.0achilles tear injury
\n", "

296 rows × 5 columns

\n", "
" ], "text/plain": [ " Name Injured Activated days_injured \\\n", "0 Sergey Karasev 2015-03-12 2015-12-08 271.0 \n", "14 Jordan Farmar 2013-12-02 2014-04-06 125.0 \n", "16 Khris Middleton 2016-09-21 2017-02-08 140.0 \n", "34 Khris Middleton 2016-09-29 2017-02-08 132.0 \n", "41 Solomon Hill 2017-08-28 2018-03-18 202.0 \n", "... ... ... ... ... \n", "3216 Jerome Robinson 2020-03-08 none 250.0 \n", "3219 Eric Bledsoe 2019-02-02 none 250.0 \n", "3223 Isaiah Hartenstein 2019-03-11 none 250.0 \n", "3227 Rudy Fernandez 2012-01-21 none 250.0 \n", "3230 Reggie Williams 2011-01-12 none 250.0 \n", "\n", " Specific Injury \n", "0 right torn mcl injury \n", "14 left torn hamstring injury \n", "16 left torn hamstring injury \n", "34 left torn hamstring injury \n", "41 left torn hamstring injury \n", "... ... \n", "3216 left achilles tear injury \n", "3219 left achilles tear injury \n", "3223 right achilles tear injury \n", "3227 right achilles tear injury \n", "3230 achilles tear injury \n", "\n", "[296 rows x 5 columns]" ] }, "execution_count": 468, "metadata": {}, "output_type": "execute_result" } ], "source": [ "combined_df_test = combined_df.drop_duplicates(subset=['Name', 'Specific Injury', 'days_injured'], keep='first')\n", "combined_df_test\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/var/folders/8t/t11lp0b952n0xtfmnwbzxzvw0000gn/T/ipykernel_11952/3471448739.py:9: FutureWarning: \n", "\n", "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n", "\n", " sns.barplot(x=injury_counts.index, y=injury_counts.values, palette='viridis')\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "\n", "# Count the occurrences of each injury type\n", "injury_counts = combined_df_test['Specific Injury'].value_counts()\n", "\n", "# Create a bar chart\n", "plt.figure(figsize=(10, 6))\n", "sns.barplot(x=injury_counts.index, y=injury_counts.values, palette='viridis')\n", "\n", "# Add labels and title\n", "plt.xlabel('Injury Type')\n", "plt.ylabel('Count')\n", "plt.title('Frequency of Injuries')\n", "\n", "# Rotate x-axis labels if needed\n", "plt.xticks(rotation=45, ha='right')\n", "\n", "# Show the plot\n", "plt.tight_layout()\n", "plt.show()\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Specific Injury\n", "left achilles tear injury 46\n", "right achilles tear injury 45\n", "achilles tear injury 18\n", "left ankle sprain injury 15\n", "right foot fracture injury 14\n", "right meniscus tear injury 13\n", "left meniscus tear injury 12\n", "left acl tear injury 11\n", "right fractured hand injury 9\n", "back surgery injury 7\n", "right ankle sprain injury 7\n", "left hamstring injury injury 6\n", "left fractured leg injury 6\n", "left hip labrum injury 6\n", "right fractured leg injury 5\n", "left torn hamstring injury 4\n", "left foot fracture injury 4\n", "left fractured hand injury 4\n", "right hip labrum injury 4\n", "right torn shoulder labrum injury 4\n", "right sprained mcl injury 4\n", "left dislocated shoulder injury 4\n", "left sprained mcl injury 4\n", "left quad injury injury 3\n", "right quad injury injury 3\n", "right acl tear injury 3\n", "right ankle fracture injury 3\n", "left shoulder sprain injury 3\n", "right hamstring injury injury 3\n", "right shoulder sprain injury 2\n", "left bone spurs injury 2\n", "right bone spurs injury 2\n", "right hip flexor strain injury 2\n", "right foot sprain injury 2\n", "right torn rotator cuff injury injury 2\n", "right hip flexor surgery injury 2\n", "right calf strain injury 2\n", "left calf strain injury 1\n", "left arm injury injury 1\n", "right arm injury injury 1\n", "left torn shoulder labrum injury 1\n", "fractured leg injury 1\n", "left ankle fracture injury 1\n", "left hip flexor surgery injury 1\n", "acl tear injury 1\n", "lower back spasm injury 1\n", "right torn mcl injury 1\n", "Name: count, dtype: int64" ] }, "execution_count": 470, "metadata": {}, "output_type": "execute_result" } ], "source": [ "combined_df_test['Specific Injury'].value_counts()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "False" ] }, "execution_count": 72, "metadata": {}, "output_type": "execute_result" } ], "source": [ "acl_tear_injuries = combined_df_test[combined_df_test['Specific Injury'].str.contains('achilles tear', case=False, na=False)]\n", "acl_tear_injuries\n", "player_name = \"Klay Thompson\"\n", "exists = player_name in acl_tear_injuries['Name'].values\n", "exists" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Specific Injury\n", "achilles tear injury NaN\n", "acl tear injury 178.000000\n", "back surgery injury 201.285714\n", "fractured leg injury 205.000000\n", "left achilles tear injury NaN\n", "left acl tear injury 283.909091\n", "left ankle fracture injury 284.000000\n", "left ankle sprain injury 250.266667\n", "left arm injury injury 143.000000\n", "left bone spurs injury 194.500000\n", "left calf strain injury 337.000000\n", "left dislocated shoulder injury 269.000000\n", "left foot fracture injury 182.250000\n", "left fractured hand injury 240.000000\n", "left fractured leg injury 213.666667\n", "left hamstring injury injury 187.666667\n", "left hip flexor surgery injury 96.000000\n", "left hip labrum injury 243.166667\n", "left meniscus tear injury 209.666667\n", "left quad injury injury 251.666667\n", "left shoulder sprain injury 286.000000\n", "left sprained mcl injury 243.000000\n", "left torn hamstring injury 149.750000\n", "left torn shoulder labrum injury 246.000000\n", "lower back spasm injury 234.000000\n", "right achilles tear injury NaN\n", "right acl tear injury 254.000000\n", "right ankle fracture injury 119.333333\n", "right ankle sprain injury 234.714286\n", "right arm injury injury 291.000000\n", "right bone spurs injury 151.500000\n", "right calf strain injury 236.000000\n", "right foot fracture injury 244.285714\n", "right foot sprain injury 294.000000\n", "right fractured hand injury 175.666667\n", "right fractured leg injury 232.000000\n", "right hamstring injury injury 209.666667\n", "right hip flexor strain injury 249.500000\n", "right hip flexor surgery injury 258.000000\n", "right hip labrum injury 296.500000\n", "right meniscus tear injury 217.076923\n", "right quad injury injury 283.000000\n", "right shoulder sprain injury 259.500000\n", "right sprained mcl injury 203.750000\n", "right torn mcl injury 271.000000\n", "right torn rotator cuff injury injury 251.500000\n", "right torn shoulder labrum injury 194.500000\n", "Name: days_injured, dtype: float64" ] }, "execution_count": 290, "metadata": {}, "output_type": "execute_result" } ], "source": [ "avg_days_injured = combined_df_test.groupby('Specific Injury')['days_injured'].mean()\n", "avg_days_injured" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Nameteam_abbreviationageplayer_heightplayer_weightcountrydraft_yeardraft_rounddraft_numbergpptsrebastnet_ratingoreb_pctdreb_pctusg_pctts_pctast_pctseason
0JR SmithDenver Nuggets25.0198.1299.790240USA20041187912.34.12.25.80.0280.1620.2200.5500.1412010
1JaVale McGeeWashington Wizards23.0213.36114.305184USA20081187910.18.00.5-4.40.1160.2140.1650.5660.0272010
2Jamaal MagloireMiami Heat33.0210.82115.665960Canada2000119181.93.40.27.20.1310.3520.1050.5850.0312010
3James PoseyIndiana Pacers34.0203.2098.429464USA1999118494.93.00.7-3.70.0130.1930.1460.4850.0682010
4Jamario MoonLos Angeles Clippers31.0203.2092.986360USAUndraftedUndraftedUndrafted594.32.80.9-6.50.0250.1540.1200.4990.0742010
\n", "
" ], "text/plain": [ " Name team_abbreviation age player_height player_weight \\\n", "0 JR Smith Denver Nuggets 25.0 198.12 99.790240 \n", "1 JaVale McGee Washington Wizards 23.0 213.36 114.305184 \n", "2 Jamaal Magloire Miami Heat 33.0 210.82 115.665960 \n", "3 James Posey Indiana Pacers 34.0 203.20 98.429464 \n", "4 Jamario Moon Los Angeles Clippers 31.0 203.20 92.986360 \n", "\n", " country draft_year draft_round draft_number gp pts reb ast net_rating \\\n", "0 USA 2004 1 18 79 12.3 4.1 2.2 5.8 \n", "1 USA 2008 1 18 79 10.1 8.0 0.5 -4.4 \n", "2 Canada 2000 1 19 18 1.9 3.4 0.2 7.2 \n", "3 USA 1999 1 18 49 4.9 3.0 0.7 -3.7 \n", "4 USA Undrafted Undrafted Undrafted 59 4.3 2.8 0.9 -6.5 \n", "\n", " oreb_pct dreb_pct usg_pct ts_pct ast_pct season \n", "0 0.028 0.162 0.220 0.550 0.141 2010 \n", "1 0.116 0.214 0.165 0.566 0.027 2010 \n", "2 0.131 0.352 0.105 0.585 0.031 2010 \n", "3 0.013 0.193 0.146 0.485 0.068 2010 \n", "4 0.025 0.154 0.120 0.499 0.074 2010 " ] }, "execution_count": 210, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_players = pd.read_csv('/Users/laraschuman/Desktop/CTP-Project/player_data.csv')\n", "df_players.head()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/var/folders/8t/t11lp0b952n0xtfmnwbzxzvw0000gn/T/ipykernel_11952/3076390773.py:2: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", " combined_df_test['season'] = pd.to_datetime(combined_df_test['Injured']).dt.year\n" ] }, { "data": { "text/plain": [ "Index(['Name', 'team_abbreviation', 'age', 'player_height', 'player_weight',\n", " 'country', 'draft_year', 'draft_round', 'draft_number', 'gp', 'pts',\n", " 'reb', 'ast', 'net_rating', 'oreb_pct', 'dreb_pct', 'usg_pct', 'ts_pct',\n", " 'ast_pct', 'season', 'Injured', 'Activated', 'days_injured',\n", " 'Specific Injury'],\n", " dtype='object')" ] }, "execution_count": 212, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Extract the year from the 'Date' column in new_df\n", "combined_df_test['season'] = pd.to_datetime(combined_df_test['Injured']).dt.year\n", "\n", "# Merge the two dataframes on 'Name' and 'Year'\n", "result_df = pd.merge(df_players, combined_df_test, how=\"left\", on=[\"Name\", \"season\"])\n", "\n", "\n", "# Show the first few rows of the result\n", "result_df.columns\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Index(['Name', 'team_abbreviation', 'age', 'player_height', 'player_weight',\n", " 'country', 'draft_year', 'draft_round', 'draft_number', 'gp', 'pts',\n", " 'reb', 'ast', 'net_rating', 'oreb_pct', 'dreb_pct', 'usg_pct', 'ts_pct',\n", " 'ast_pct', 'season', 'Injured', 'Activated', 'days_injured',\n", " 'Specific Injury'],\n", " dtype='object')" ] }, "execution_count": 213, "metadata": {}, "output_type": "execute_result" } ], "source": [ "result_df.columns\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "result_df['days_injured'] = result_df['days_injured'].fillna(0)\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 12.3\n", "1 10.1\n", "2 1.9\n", "3 4.9\n", "4 4.3\n", " ... \n", "5494 3.1\n", "5495 3.9\n", "5496 10.4\n", "5497 5.3\n", "5498 14.1\n", "Name: pts, Length: 5499, dtype: float64" ] }, "execution_count": 215, "metadata": {}, "output_type": "execute_result" } ], "source": [ "result_df['pts']" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# Ensure that 'Activated' is in datetime format if it's not already\n", "result_df['Activated'] = pd.to_datetime(result_df['Activated'])\n", "\n", "# Define the injury season (assuming 'Activated' gives the injury date)\n", "result_df['injury_season'] = result_df['Activated'].dt.year\n", "\n", "# Mark rows as 'before injury' or 'after injury' based on the injury season\n", "result_df['before_after_injury'] = result_df.apply(\n", " lambda row: 'before_injury' if row['season'] < row['injury_season'] else ('after_injury' if row['season'] > row['injury_season'] else 'injury_season'),\n", " axis=1\n", ")\n", "\n", "# Separate the data into before injury, after injury, and injury season\n", "before_injury = result_df[result_df['before_after_injury'] == 'before_injury']\n", "after_injury = result_df[result_df['before_after_injury'] == 'after_injury']\n", "injury_season = result_df[result_df['before_after_injury'] == 'injury_season']\n", "\n", "# Check the resulting datasets\n", "\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Nameteam_abbreviationageplayer_heightplayer_weightcountrydraft_yeardraft_rounddraft_numbergp...usg_pctts_pctast_pctseasonInjuredActivateddays_injuredSpecific Injuryinjury_seasonbefore_after_injury
390Reggie EvansToronto Raptors31.0203.20111.130040USA20103030...0.1020.4660.07020102010-11-262011-03-09103.0right foot fracture injury2011.0before_injury
489Jason ThompsonSacramento Kings25.0210.82113.398000USA200811264...0.1530.5580.07020112011-06-152012-03-28287.0right foot fracture injury2012.0before_injury
662Eric BledsoeLos Angeles Clippers22.0185.4288.450440USA201011840...0.1970.4540.22820112011-10-072012-01-26111.0right meniscus tear injury2012.0before_injury
683Dominique JonesDallas Mavericks23.0195.5897.522280USA201012533...0.2010.4930.29820112011-02-112012-01-23346.0right foot fracture injury2012.0before_injury
684Dominique JonesDallas Mavericks23.0195.5897.522280USA201012533...0.2010.4930.29820112011-02-082012-01-23349.0right foot fracture injury2012.0before_injury
..................................................................
4759Stephen CurryGolden State Warriors32.0190.5083.914520USA2009175...0.2860.5570.35520192019-10-312020-03-04125.0left fractured hand injury2020.0before_injury
4762Svi MykhailiukDetroit Pistons23.0200.6692.986360Ukraine201824756...0.1650.5780.12020192019-04-052020-02-12313.0left fractured hand injury2020.0before_injury
4827Zion WilliamsonNew Orleans Pelicans19.0198.12128.820128USA20191124...0.2910.6160.11920192019-10-182020-01-2296.0right meniscus tear injury2020.0before_injury
4828Zion WilliamsonNew Orleans Pelicans19.0198.12128.820128USA20191124...0.2910.6160.11920192019-10-212020-01-2293.0right meniscus tear injury2020.0before_injury
4928Richaun HolmesSacramento Kings26.0208.28106.594120USA201523744...0.1560.6810.05020192019-04-052020-03-07337.0left ankle sprain injury2020.0before_injury
\n", "

74 rows × 26 columns

\n", "
" ], "text/plain": [ " Name team_abbreviation age player_height \\\n", "390 Reggie Evans Toronto Raptors 31.0 203.20 \n", "489 Jason Thompson Sacramento Kings 25.0 210.82 \n", "662 Eric Bledsoe Los Angeles Clippers 22.0 185.42 \n", "683 Dominique Jones Dallas Mavericks 23.0 195.58 \n", "684 Dominique Jones Dallas Mavericks 23.0 195.58 \n", "... ... ... ... ... \n", "4759 Stephen Curry Golden State Warriors 32.0 190.50 \n", "4762 Svi Mykhailiuk Detroit Pistons 23.0 200.66 \n", "4827 Zion Williamson New Orleans Pelicans 19.0 198.12 \n", "4828 Zion Williamson New Orleans Pelicans 19.0 198.12 \n", "4928 Richaun Holmes Sacramento Kings 26.0 208.28 \n", "\n", " player_weight country draft_year draft_round draft_number gp ... \\\n", "390 111.130040 USA 2010 3 0 30 ... \n", "489 113.398000 USA 2008 1 12 64 ... \n", "662 88.450440 USA 2010 1 18 40 ... \n", "683 97.522280 USA 2010 1 25 33 ... \n", "684 97.522280 USA 2010 1 25 33 ... \n", "... ... ... ... ... ... .. ... \n", "4759 83.914520 USA 2009 1 7 5 ... \n", "4762 92.986360 Ukraine 2018 2 47 56 ... \n", "4827 128.820128 USA 2019 1 1 24 ... \n", "4828 128.820128 USA 2019 1 1 24 ... \n", "4928 106.594120 USA 2015 2 37 44 ... \n", "\n", " usg_pct ts_pct ast_pct season Injured Activated days_injured \\\n", "390 0.102 0.466 0.070 2010 2010-11-26 2011-03-09 103.0 \n", "489 0.153 0.558 0.070 2011 2011-06-15 2012-03-28 287.0 \n", "662 0.197 0.454 0.228 2011 2011-10-07 2012-01-26 111.0 \n", "683 0.201 0.493 0.298 2011 2011-02-11 2012-01-23 346.0 \n", "684 0.201 0.493 0.298 2011 2011-02-08 2012-01-23 349.0 \n", "... ... ... ... ... ... ... ... \n", "4759 0.286 0.557 0.355 2019 2019-10-31 2020-03-04 125.0 \n", "4762 0.165 0.578 0.120 2019 2019-04-05 2020-02-12 313.0 \n", "4827 0.291 0.616 0.119 2019 2019-10-18 2020-01-22 96.0 \n", "4828 0.291 0.616 0.119 2019 2019-10-21 2020-01-22 93.0 \n", "4928 0.156 0.681 0.050 2019 2019-04-05 2020-03-07 337.0 \n", "\n", " Specific Injury injury_season before_after_injury \n", "390 right foot fracture injury 2011.0 before_injury \n", "489 right foot fracture injury 2012.0 before_injury \n", "662 right meniscus tear injury 2012.0 before_injury \n", "683 right foot fracture injury 2012.0 before_injury \n", "684 right foot fracture injury 2012.0 before_injury \n", "... ... ... ... \n", "4759 left fractured hand injury 2020.0 before_injury \n", "4762 left fractured hand injury 2020.0 before_injury \n", "4827 right meniscus tear injury 2020.0 before_injury \n", "4828 right meniscus tear injury 2020.0 before_injury \n", "4928 left ankle sprain injury 2020.0 before_injury \n", "\n", "[74 rows x 26 columns]" ] }, "execution_count": 364, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Step 1: Filter before_injury DataFrame to only keep rows with players in after_injury\n", "before_injury_filtered = before_injury[before_injury['Name'].isin(after_injury['Name'])]\n", "\n", "# Step 2: Concatenate the filtered before_injury and after_injury data\n", "injury_data_cleaned = pd.concat([before_injury_filtered, after_injury])\n", "\n", "# Step 3: Combine the original before_injury and after_injury into one final dataset\n", "injury_data_cleaned = pd.concat([before_injury, after_injury])\n", "\n", "# Step 4: Replace 'Undrafted' in the 'draft_round' and 'draft_number' columns\n", "injury_data_cleaned['draft_round'] = injury_data_cleaned['draft_round'].replace('Undrafted', 3)\n", "injury_data_cleaned['draft_number'] = injury_data_cleaned['draft_number'].replace('Undrafted', 0)\n", "\n", "# Step 5: Replace 'Undrafted' in the 'draft_year' column with the value from 'season'\n", "injury_data_cleaned.loc[injury_data_cleaned['draft_year'] == 'Undrafted', 'draft_year'] = injury_data_cleaned['season']\n", "\n", "# Display the final dataset\n", "injury_data_cleaned\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/var/folders/8t/t11lp0b952n0xtfmnwbzxzvw0000gn/T/ipykernel_11952/4162268832.py:14: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", " before_injury['Injury_Status'] = 'Before'\n", "/var/folders/8t/t11lp0b952n0xtfmnwbzxzvw0000gn/T/ipykernel_11952/4162268832.py:15: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", " after_injury['Injury_Status'] = 'After'\n" ] } ], "source": [ "import pandas as pd\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.linear_model import LinearRegression\n", "from sklearn.ensemble import RandomForestRegressor\n", "from sklearn.metrics import mean_absolute_error, r2_score\n", "\n", "# Assuming 'result_df' has the required data\n", "# Split the dataset into before and after injury\n", "before_injury = result_df[result_df['days_injured'] == 0] # Before injury (assuming 0 means no injury)\n", "after_injury = result_df[result_df['days_injured'] > 0] # After injury\n", "\n", "# Combine the before and after data into one dataset\n", "# Add a column to indicate before/after injury\n", "before_injury['Injury_Status'] = 'Before'\n", "after_injury['Injury_Status'] = 'After'\n", "\n", "# Get the list of players that exist in both before_injury and after_injury\n", "common_players = after_injury['Name'].isin(before_injury['Name'])\n", "\n", "# Filter the before_injury DataFrame to only keep rows with players that exist in after_injury\n", "before_injury_filtered = before_injury[before_injury['Name'].isin(after_injury['Name'])]\n", "\n", "# Concatenate the filtered before_injury and after_injury data\n", "injury_data = pd.concat([before_injury_filtered, after_injury])\n", "\n", "# Combine into a single dataset\n", "injury_data = pd.concat([before_injury, after_injury])\n", "injury_data['draft_round'] = injury_data['draft_round'].replace('Undrafted', 3)\n", "injury_data['draft_number'] = injury_data['draft_number'].replace('Undrafted', 0)\n", "injury_data.loc[injury_data['draft_year'] == 'Undrafted', 'draft_year'] = injury_data['season']\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Namegpptsrebastnet_ratingoreb_pctdreb_pctusg_pctts_pct
\n", "
" ], "text/plain": [ "Empty DataFrame\n", "Columns: [Name, gp, pts, reb, ast, net_rating, oreb_pct, dreb_pct, usg_pct, ts_pct]\n", "Index: []" ] }, "execution_count": 366, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# List of tangible stats you want to aggregate\n", "tangible_stats = ['gp', 'pts', 'reb', 'ast', 'net_rating', 'oreb_pct', 'dreb_pct', 'usg_pct', 'ts_pct']\n", "\n", "# Group by player Name and calculate the mean for all tangible stats\n", "before_injury_avg = before_injury_filtered.groupby('Name')[tangible_stats].mean().round(2).reset_index()\n", "\n", "# Display the result\n", "before_injury_avg.head()\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Namegpptsrebastnet_ratingoreb_pctdreb_pctusg_pctts_pct
0Al Harrington10.05.12.71.0-16.00.060.210.250.43
1Alan Williams5.04.04.41.60.10.040.260.180.46
2Alec Burks31.013.33.52.0-2.00.020.140.250.52
3Alex Len69.06.36.60.50.50.100.220.140.54
4Alonzo Gee13.00.81.20.51.40.050.140.110.31
.................................
121Toney Douglas38.06.21.92.0-0.30.030.100.240.39
122Tony Allen22.04.72.10.4-2.00.070.090.180.51
123Wilson Chandler43.013.05.11.39.30.050.170.230.56
124Zach LaVine24.016.73.93.0-12.10.010.140.290.50
125Zion Williamson24.022.56.32.15.10.090.110.290.62
\n", "

126 rows × 10 columns

\n", "
" ], "text/plain": [ " Name gp pts reb ast net_rating oreb_pct dreb_pct \\\n", "0 Al Harrington 10.0 5.1 2.7 1.0 -16.0 0.06 0.21 \n", "1 Alan Williams 5.0 4.0 4.4 1.6 0.1 0.04 0.26 \n", "2 Alec Burks 31.0 13.3 3.5 2.0 -2.0 0.02 0.14 \n", "3 Alex Len 69.0 6.3 6.6 0.5 0.5 0.10 0.22 \n", "4 Alonzo Gee 13.0 0.8 1.2 0.5 1.4 0.05 0.14 \n", ".. ... ... ... ... ... ... ... ... \n", "121 Toney Douglas 38.0 6.2 1.9 2.0 -0.3 0.03 0.10 \n", "122 Tony Allen 22.0 4.7 2.1 0.4 -2.0 0.07 0.09 \n", "123 Wilson Chandler 43.0 13.0 5.1 1.3 9.3 0.05 0.17 \n", "124 Zach LaVine 24.0 16.7 3.9 3.0 -12.1 0.01 0.14 \n", "125 Zion Williamson 24.0 22.5 6.3 2.1 5.1 0.09 0.11 \n", "\n", " usg_pct ts_pct \n", "0 0.25 0.43 \n", "1 0.18 0.46 \n", "2 0.25 0.52 \n", "3 0.14 0.54 \n", "4 0.11 0.31 \n", ".. ... ... \n", "121 0.24 0.39 \n", "122 0.18 0.51 \n", "123 0.23 0.56 \n", "124 0.29 0.50 \n", "125 0.29 0.62 \n", "\n", "[126 rows x 10 columns]" ] }, "execution_count": 218, "metadata": {}, "output_type": "execute_result" } ], "source": [ "after_injury_avg = after_injury.groupby('Name')[tangible_stats].mean().round(2).reset_index()\n", "after_injury_avg" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Namegp_beforepts_beforereb_beforeast_beforenet_rating_beforeoreb_pct_beforedreb_pct_beforeusg_pct_beforets_pct_beforegp_afterpts_afterreb_afterast_afternet_rating_afteroreb_pct_afterdreb_pct_afterusg_pct_afterts_pct_after
0Al Harrington57.0010.434.331.204.200.050.180.230.5210.05.12.71.0-16.00.060.210.250.43
1Alan Williams20.674.634.600.538.870.150.360.210.555.04.04.41.60.10.040.260.180.46
2Alec Burks57.0010.333.391.87-2.960.030.140.210.5331.013.33.52.0-2.00.020.140.250.52
3Alex Len66.007.605.640.84-6.710.100.210.180.5669.06.36.60.50.50.100.220.140.54
4Alonzo Gee65.506.633.281.05-7.050.050.120.150.5313.00.81.20.51.40.050.140.110.31
............................................................
121Toney Douglas50.006.672.202.221.070.030.120.180.5038.06.21.92.0-0.30.030.100.240.39
122Tony Allen66.008.964.231.374.560.070.120.180.5222.04.72.10.4-2.00.070.090.180.51
123Wilson Chandler56.3811.225.221.76-1.450.040.160.180.5243.013.05.11.39.30.050.170.230.56
124Zach LaVine64.5019.933.923.88-5.880.020.110.260.5724.016.73.93.0-12.10.010.140.290.50
125Zion Williamson61.0027.007.203.702.100.080.130.290.6524.022.56.32.15.10.090.110.290.62
\n", "

126 rows × 19 columns

\n", "
" ], "text/plain": [ " Name gp_before pts_before reb_before ast_before \\\n", "0 Al Harrington 57.00 10.43 4.33 1.20 \n", "1 Alan Williams 20.67 4.63 4.60 0.53 \n", "2 Alec Burks 57.00 10.33 3.39 1.87 \n", "3 Alex Len 66.00 7.60 5.64 0.84 \n", "4 Alonzo Gee 65.50 6.63 3.28 1.05 \n", ".. ... ... ... ... ... \n", "121 Toney Douglas 50.00 6.67 2.20 2.22 \n", "122 Tony Allen 66.00 8.96 4.23 1.37 \n", "123 Wilson Chandler 56.38 11.22 5.22 1.76 \n", "124 Zach LaVine 64.50 19.93 3.92 3.88 \n", "125 Zion Williamson 61.00 27.00 7.20 3.70 \n", "\n", " net_rating_before oreb_pct_before dreb_pct_before usg_pct_before \\\n", "0 4.20 0.05 0.18 0.23 \n", "1 8.87 0.15 0.36 0.21 \n", "2 -2.96 0.03 0.14 0.21 \n", "3 -6.71 0.10 0.21 0.18 \n", "4 -7.05 0.05 0.12 0.15 \n", ".. ... ... ... ... \n", "121 1.07 0.03 0.12 0.18 \n", "122 4.56 0.07 0.12 0.18 \n", "123 -1.45 0.04 0.16 0.18 \n", "124 -5.88 0.02 0.11 0.26 \n", "125 2.10 0.08 0.13 0.29 \n", "\n", " ts_pct_before gp_after pts_after reb_after ast_after \\\n", "0 0.52 10.0 5.1 2.7 1.0 \n", "1 0.55 5.0 4.0 4.4 1.6 \n", "2 0.53 31.0 13.3 3.5 2.0 \n", "3 0.56 69.0 6.3 6.6 0.5 \n", "4 0.53 13.0 0.8 1.2 0.5 \n", ".. ... ... ... ... ... \n", "121 0.50 38.0 6.2 1.9 2.0 \n", "122 0.52 22.0 4.7 2.1 0.4 \n", "123 0.52 43.0 13.0 5.1 1.3 \n", "124 0.57 24.0 16.7 3.9 3.0 \n", "125 0.65 24.0 22.5 6.3 2.1 \n", "\n", " net_rating_after oreb_pct_after dreb_pct_after usg_pct_after \\\n", "0 -16.0 0.06 0.21 0.25 \n", "1 0.1 0.04 0.26 0.18 \n", "2 -2.0 0.02 0.14 0.25 \n", "3 0.5 0.10 0.22 0.14 \n", "4 1.4 0.05 0.14 0.11 \n", ".. ... ... ... ... \n", "121 -0.3 0.03 0.10 0.24 \n", "122 -2.0 0.07 0.09 0.18 \n", "123 9.3 0.05 0.17 0.23 \n", "124 -12.1 0.01 0.14 0.29 \n", "125 5.1 0.09 0.11 0.29 \n", "\n", " ts_pct_after \n", "0 0.43 \n", "1 0.46 \n", "2 0.52 \n", "3 0.54 \n", "4 0.31 \n", ".. ... \n", "121 0.39 \n", "122 0.51 \n", "123 0.56 \n", "124 0.50 \n", "125 0.62 \n", "\n", "[126 rows x 19 columns]" ] }, "execution_count": 219, "metadata": {}, "output_type": "execute_result" } ], "source": [ "performance_data = pd.merge(before_injury_avg, after_injury_avg, on='Name', suffixes=('_before', '_after'))\n", "performance_data\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "performance_data['points_change'] = performance_data['pts_after'] - performance_data['pts_before']\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Namegp_beforepts_beforereb_beforeast_beforenet_rating_beforeoreb_pct_beforedreb_pct_beforeusg_pct_beforets_pct_before...ts_pct_changegp_changepts_changereb_changeast_changenet_rating_changeoreb_pct_changedreb_pct_changeusg_pct_changets_pct_change
0Al Harrington57.0010.434.331.204.200.050.180.230.52...-0.09-47.00-5.33-1.63-0.20-20.200.010.030.02-0.09
1Alan Williams20.674.634.600.538.870.150.360.210.55...-0.09-15.67-0.63-0.201.07-8.77-0.11-0.10-0.03-0.09
2Alec Burks57.0010.333.391.87-2.960.030.140.210.53...-0.01-26.002.970.110.130.96-0.010.000.04-0.01
3Alex Len66.007.605.640.84-6.710.100.210.180.56...-0.023.00-1.300.96-0.347.210.000.01-0.04-0.02
4Alonzo Gee65.506.633.281.05-7.050.050.120.150.53...-0.22-52.50-5.83-2.08-0.558.450.000.02-0.04-0.22
..................................................................
121Toney Douglas50.006.672.202.221.070.030.120.180.50...-0.11-12.00-0.47-0.30-0.22-1.370.00-0.020.06-0.11
122Tony Allen66.008.964.231.374.560.070.120.180.52...-0.01-44.00-4.26-2.13-0.97-6.560.00-0.030.00-0.01
123Wilson Chandler56.3811.225.221.76-1.450.040.160.180.52...0.04-13.381.78-0.12-0.4610.750.010.010.050.04
124Zach LaVine64.5019.933.923.88-5.880.020.110.260.57...-0.07-40.50-3.23-0.02-0.88-6.22-0.010.030.03-0.07
125Zion Williamson61.0027.007.203.702.100.080.130.290.65...-0.03-37.00-4.50-0.90-1.603.000.01-0.020.00-0.03
\n", "

126 rows × 38 columns

\n", "
" ], "text/plain": [ " Name gp_before pts_before reb_before ast_before \\\n", "0 Al Harrington 57.00 10.43 4.33 1.20 \n", "1 Alan Williams 20.67 4.63 4.60 0.53 \n", "2 Alec Burks 57.00 10.33 3.39 1.87 \n", "3 Alex Len 66.00 7.60 5.64 0.84 \n", "4 Alonzo Gee 65.50 6.63 3.28 1.05 \n", ".. ... ... ... ... ... \n", "121 Toney Douglas 50.00 6.67 2.20 2.22 \n", "122 Tony Allen 66.00 8.96 4.23 1.37 \n", "123 Wilson Chandler 56.38 11.22 5.22 1.76 \n", "124 Zach LaVine 64.50 19.93 3.92 3.88 \n", "125 Zion Williamson 61.00 27.00 7.20 3.70 \n", "\n", " net_rating_before oreb_pct_before dreb_pct_before usg_pct_before \\\n", "0 4.20 0.05 0.18 0.23 \n", "1 8.87 0.15 0.36 0.21 \n", "2 -2.96 0.03 0.14 0.21 \n", "3 -6.71 0.10 0.21 0.18 \n", "4 -7.05 0.05 0.12 0.15 \n", ".. ... ... ... ... \n", "121 1.07 0.03 0.12 0.18 \n", "122 4.56 0.07 0.12 0.18 \n", "123 -1.45 0.04 0.16 0.18 \n", "124 -5.88 0.02 0.11 0.26 \n", "125 2.10 0.08 0.13 0.29 \n", "\n", " ts_pct_before ... ts_pct_change gp_change pts_change reb_change \\\n", "0 0.52 ... -0.09 -47.00 -5.33 -1.63 \n", "1 0.55 ... -0.09 -15.67 -0.63 -0.20 \n", "2 0.53 ... -0.01 -26.00 2.97 0.11 \n", "3 0.56 ... -0.02 3.00 -1.30 0.96 \n", "4 0.53 ... -0.22 -52.50 -5.83 -2.08 \n", ".. ... ... ... ... ... ... \n", "121 0.50 ... -0.11 -12.00 -0.47 -0.30 \n", "122 0.52 ... -0.01 -44.00 -4.26 -2.13 \n", "123 0.52 ... 0.04 -13.38 1.78 -0.12 \n", "124 0.57 ... -0.07 -40.50 -3.23 -0.02 \n", "125 0.65 ... -0.03 -37.00 -4.50 -0.90 \n", "\n", " ast_change net_rating_change oreb_pct_change dreb_pct_change \\\n", "0 -0.20 -20.20 0.01 0.03 \n", "1 1.07 -8.77 -0.11 -0.10 \n", "2 0.13 0.96 -0.01 0.00 \n", "3 -0.34 7.21 0.00 0.01 \n", "4 -0.55 8.45 0.00 0.02 \n", ".. ... ... ... ... \n", "121 -0.22 -1.37 0.00 -0.02 \n", "122 -0.97 -6.56 0.00 -0.03 \n", "123 -0.46 10.75 0.01 0.01 \n", "124 -0.88 -6.22 -0.01 0.03 \n", "125 -1.60 3.00 0.01 -0.02 \n", "\n", " usg_pct_change ts_pct_change \n", "0 0.02 -0.09 \n", "1 -0.03 -0.09 \n", "2 0.04 -0.01 \n", "3 -0.04 -0.02 \n", "4 -0.04 -0.22 \n", ".. ... ... \n", "121 0.06 -0.11 \n", "122 0.00 -0.01 \n", "123 0.05 0.04 \n", "124 0.03 -0.07 \n", "125 0.00 -0.03 \n", "\n", "[126 rows x 38 columns]" ] }, "execution_count": 222, "metadata": {}, "output_type": "execute_result" } ], "source": [ "performance_changes = pd.DataFrame()\n", "\n", "# Calculate the change in each stat\n", "for stat in tangible_stats:\n", " performance_changes[f'{stat}_change'] = performance_data[f'{stat}_after'] - performance_data[f'{stat}_before']\n", "\n", "# Combine the performance_changes DataFrame with the original performance_data\n", "performance_data = pd.concat([performance_data, performance_changes], axis=1)\n", "performance_data" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Index(['Name', 'gp_before', 'pts_before', 'reb_before', 'ast_before',\n", " 'net_rating_before', 'oreb_pct_before', 'dreb_pct_before',\n", " 'usg_pct_before', 'ts_pct_before', 'gp_after', 'pts_after', 'reb_after',\n", " 'ast_after', 'net_rating_after', 'oreb_pct_after', 'dreb_pct_after',\n", " 'usg_pct_after', 'ts_pct_after', 'points_change', 'gp_change',\n", " 'pts_change', 'reb_change', 'ast_change', 'net_rating_change',\n", " 'oreb_pct_change', 'dreb_pct_change', 'usg_pct_change', 'ts_pct_change',\n", " 'gp_change', 'pts_change', 'reb_change', 'ast_change',\n", " 'net_rating_change', 'oreb_pct_change', 'dreb_pct_change',\n", " 'usg_pct_change', 'ts_pct_change'],\n", " dtype='object')" ] }, "execution_count": 191, "metadata": {}, "output_type": "execute_result" } ], "source": [ "performance_data.columns" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "injury_data_merged= pd.merge(injury_data, performance_data[['Name', 'points_change', 'gp_change',\n", " 'pts_change', 'reb_change', 'ast_change', 'net_rating_change',\n", " 'oreb_pct_change', 'dreb_pct_change', 'usg_pct_change', 'ts_pct_change',\n", " 'gp_change', 'pts_change', 'reb_change', 'ast_change',\n", " 'net_rating_change', 'oreb_pct_change', 'dreb_pct_change',\n", " 'usg_pct_change', 'ts_pct_change']], on='Name')\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Index(['Name', 'team_abbreviation', 'age', 'player_height', 'player_weight',\n", " 'country', 'draft_year', 'draft_round', 'draft_number', 'gp', 'pts',\n", " 'reb', 'ast', 'net_rating', 'oreb_pct', 'dreb_pct', 'usg_pct', 'ts_pct',\n", " 'ast_pct', 'season', 'Injured', 'Activated', 'days_injured',\n", " 'Specific Injury', 'Injury_Status', 'points_change', 'gp_change',\n", " 'gp_change', 'pts_change', 'pts_change', 'reb_change', 'reb_change',\n", " 'ast_change', 'ast_change', 'net_rating_change', 'net_rating_change',\n", " 'oreb_pct_change', 'oreb_pct_change', 'dreb_pct_change',\n", " 'dreb_pct_change', 'usg_pct_change', 'usg_pct_change', 'ts_pct_change',\n", " 'ts_pct_change', 'gp_change', 'gp_change', 'pts_change', 'pts_change',\n", " 'reb_change', 'reb_change', 'ast_change', 'ast_change',\n", " 'net_rating_change', 'net_rating_change', 'oreb_pct_change',\n", " 'oreb_pct_change', 'dreb_pct_change', 'dreb_pct_change',\n", " 'usg_pct_change', 'usg_pct_change', 'ts_pct_change', 'ts_pct_change'],\n", " dtype='object')" ] }, "execution_count": 224, "metadata": {}, "output_type": "execute_result" } ], "source": [ "injury_data_merged.columns" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "injury_data_merged = injury_data_merged.loc[:, ~injury_data_merged.columns.duplicated()]\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Nameteam_abbreviationageplayer_heightplayer_weightcountrydraft_yeardraft_rounddraft_numbergp...points_changegp_changepts_changereb_changeast_changenet_rating_changeoreb_pct_changedreb_pct_changeusg_pct_changets_pct_change
0Jameer NelsonOrlando Magic29.0182.8886.182480USA200412076...-6.10-12.00-6.10-0.91-2.13-4.79-0.01-0.01-0.06-0.02
1Jason ThompsonSacramento Kings24.0210.82113.398000USA200811275...1.90-9.401.901.180.32-1.100.03-0.01-0.010.03
2Gary NealSan Antonio Spurs26.0193.0495.254320USA20103080...1.17-12.001.170.22-0.33-6.070.010.01-0.010.08
3Glen DavisBoston Celtics25.0205.74131.088088USA200723578...-0.626.25-0.620.12-0.05-5.080.01-0.01-0.020.00
4Greivis VasquezMemphis Grizzlies24.0198.1295.707912Venezuela201012870...-2.27-40.00-2.27-0.23-0.35-5.120.000.000.00-0.05
..................................................................
932Svi MykhailiukDetroit Pistons23.0200.6692.986360Ukraine201824756...3.152.003.150.200.603.75-0.01-0.01-0.010.09
933Zion WilliamsonNew Orleans Pelicans19.0198.12128.820128USA20191124...-4.50-37.00-4.50-0.90-1.603.000.01-0.020.00-0.03
934Zion WilliamsonNew Orleans Pelicans19.0198.12128.820128USA20191124...-4.50-37.00-4.50-0.90-1.603.000.01-0.020.00-0.03
935Richaun HolmesSacramento Kings26.0208.28106.594120USA201523744...3.44-13.403.443.00-0.104.560.010.02-0.010.06
936Jakob PoeltlSan Antonio Spurs25.0215.90111.130040Austria20161969...3.32-0.753.323.180.92-1.35-0.010.000.00-0.02
\n", "

937 rows × 35 columns

\n", "
" ], "text/plain": [ " Name team_abbreviation age player_height \\\n", "0 Jameer Nelson Orlando Magic 29.0 182.88 \n", "1 Jason Thompson Sacramento Kings 24.0 210.82 \n", "2 Gary Neal San Antonio Spurs 26.0 193.04 \n", "3 Glen Davis Boston Celtics 25.0 205.74 \n", "4 Greivis Vasquez Memphis Grizzlies 24.0 198.12 \n", ".. ... ... ... ... \n", "932 Svi Mykhailiuk Detroit Pistons 23.0 200.66 \n", "933 Zion Williamson New Orleans Pelicans 19.0 198.12 \n", "934 Zion Williamson New Orleans Pelicans 19.0 198.12 \n", "935 Richaun Holmes Sacramento Kings 26.0 208.28 \n", "936 Jakob Poeltl San Antonio Spurs 25.0 215.90 \n", "\n", " player_weight country draft_year draft_round draft_number gp ... \\\n", "0 86.182480 USA 2004 1 20 76 ... \n", "1 113.398000 USA 2008 1 12 75 ... \n", "2 95.254320 USA 2010 3 0 80 ... \n", "3 131.088088 USA 2007 2 35 78 ... \n", "4 95.707912 Venezuela 2010 1 28 70 ... \n", ".. ... ... ... ... ... .. ... \n", "932 92.986360 Ukraine 2018 2 47 56 ... \n", "933 128.820128 USA 2019 1 1 24 ... \n", "934 128.820128 USA 2019 1 1 24 ... \n", "935 106.594120 USA 2015 2 37 44 ... \n", "936 111.130040 Austria 2016 1 9 69 ... \n", "\n", " points_change gp_change pts_change reb_change ast_change \\\n", "0 -6.10 -12.00 -6.10 -0.91 -2.13 \n", "1 1.90 -9.40 1.90 1.18 0.32 \n", "2 1.17 -12.00 1.17 0.22 -0.33 \n", "3 -0.62 6.25 -0.62 0.12 -0.05 \n", "4 -2.27 -40.00 -2.27 -0.23 -0.35 \n", ".. ... ... ... ... ... \n", "932 3.15 2.00 3.15 0.20 0.60 \n", "933 -4.50 -37.00 -4.50 -0.90 -1.60 \n", "934 -4.50 -37.00 -4.50 -0.90 -1.60 \n", "935 3.44 -13.40 3.44 3.00 -0.10 \n", "936 3.32 -0.75 3.32 3.18 0.92 \n", "\n", " net_rating_change oreb_pct_change dreb_pct_change usg_pct_change \\\n", "0 -4.79 -0.01 -0.01 -0.06 \n", "1 -1.10 0.03 -0.01 -0.01 \n", "2 -6.07 0.01 0.01 -0.01 \n", "3 -5.08 0.01 -0.01 -0.02 \n", "4 -5.12 0.00 0.00 0.00 \n", ".. ... ... ... ... \n", "932 3.75 -0.01 -0.01 -0.01 \n", "933 3.00 0.01 -0.02 0.00 \n", "934 3.00 0.01 -0.02 0.00 \n", "935 4.56 0.01 0.02 -0.01 \n", "936 -1.35 -0.01 0.00 0.00 \n", "\n", " ts_pct_change \n", "0 -0.02 \n", "1 0.03 \n", "2 0.08 \n", "3 0.00 \n", "4 -0.05 \n", ".. ... \n", "932 0.09 \n", "933 -0.03 \n", "934 -0.03 \n", "935 0.06 \n", "936 -0.02 \n", "\n", "[937 rows x 35 columns]" ] }, "execution_count": 256, "metadata": {}, "output_type": "execute_result" } ], "source": [ "injury_data_merged" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "grouped_data = injury_data_merged.groupby(['Name','player_height','player_weight','age']).agg({\n", " 'pts_change': 'mean',\n", " 'gp_change': 'mean',\n", " 'reb_change': 'mean',\n", " 'ast_change': 'mean',\n", " 'net_rating_change': 'mean',\n", " 'oreb_pct_change': 'mean',\n", " 'dreb_pct_change': 'mean',\n", " 'usg_pct_change': 'mean',\n", " 'ts_pct_change': 'mean'\n", "}).round(2).reset_index()\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "injury_grouped_data = injury_data_merged.groupby(['Specific Injury','Name']).agg({\n", " 'pts_change': 'mean',\n", " 'gp_change': 'mean',\n", " 'reb_change': 'mean',\n", " 'ast_change': 'mean',\n", " 'net_rating_change': 'mean',\n", " 'oreb_pct_change': 'mean',\n", " 'dreb_pct_change': 'mean',\n", " 'usg_pct_change': 'mean',\n", " 'ts_pct_change': 'mean'\n", "}).round(2).reset_index()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Specific InjuryNamepts_changegp_changereb_changeast_changenet_rating_changeoreb_pct_changedreb_pct_changeusg_pct_changets_pct_change
0back surgery injuryDonatas Motiejunas0.28-5.80-0.220.2617.24-0.010.000.000.00
1back surgery injuryMarquis Daniels-2.30-16.00-0.700.001.700.000.010.02-0.08
2back surgery injuryMartell Webster-5.20-27.17-0.80-0.73-6.550.010.00-0.01-0.10
3back surgery injuryMarvin Williams1.03-15.220.23-0.044.260.020.010.03-0.02
4fractured leg injuryThabo Sefolosha0.7918.780.880.22-2.98-0.010.020.000.03
....................................
145right torn rotator cuff injury injuryKobe Bryant-10.00-27.25-1.45-0.67-13.48-0.01-0.01-0.03-0.04
146right torn shoulder labrum injuryDelon Wright-1.92-34.40-1.38-0.906.680.00-0.040.00-0.01
147right torn shoulder labrum injuryMichael Carter-Williams-2.43-7.57-0.94-0.862.45-0.010.020.01-0.03
148right torn shoulder labrum injuryMichael Kidd-Gilchrist4.90-55.711.230.2014.69-0.01-0.010.020.09
149right torn shoulder labrum injuryToney Douglas-0.47-12.00-0.30-0.22-1.370.00-0.020.06-0.11
\n", "

150 rows × 11 columns

\n", "
" ], "text/plain": [ " Specific Injury Name \\\n", "0 back surgery injury Donatas Motiejunas \n", "1 back surgery injury Marquis Daniels \n", "2 back surgery injury Martell Webster \n", "3 back surgery injury Marvin Williams \n", "4 fractured leg injury Thabo Sefolosha \n", ".. ... ... \n", "145 right torn rotator cuff injury injury Kobe Bryant \n", "146 right torn shoulder labrum injury Delon Wright \n", "147 right torn shoulder labrum injury Michael Carter-Williams \n", "148 right torn shoulder labrum injury Michael Kidd-Gilchrist \n", "149 right torn shoulder labrum injury Toney Douglas \n", "\n", " pts_change gp_change reb_change ast_change net_rating_change \\\n", "0 0.28 -5.80 -0.22 0.26 17.24 \n", "1 -2.30 -16.00 -0.70 0.00 1.70 \n", "2 -5.20 -27.17 -0.80 -0.73 -6.55 \n", "3 1.03 -15.22 0.23 -0.04 4.26 \n", "4 0.79 18.78 0.88 0.22 -2.98 \n", ".. ... ... ... ... ... \n", "145 -10.00 -27.25 -1.45 -0.67 -13.48 \n", "146 -1.92 -34.40 -1.38 -0.90 6.68 \n", "147 -2.43 -7.57 -0.94 -0.86 2.45 \n", "148 4.90 -55.71 1.23 0.20 14.69 \n", "149 -0.47 -12.00 -0.30 -0.22 -1.37 \n", "\n", " oreb_pct_change dreb_pct_change usg_pct_change ts_pct_change \n", "0 -0.01 0.00 0.00 0.00 \n", "1 0.00 0.01 0.02 -0.08 \n", "2 0.01 0.00 -0.01 -0.10 \n", "3 0.02 0.01 0.03 -0.02 \n", "4 -0.01 0.02 0.00 0.03 \n", ".. ... ... ... ... \n", "145 -0.01 -0.01 -0.03 -0.04 \n", "146 0.00 -0.04 0.00 -0.01 \n", "147 -0.01 0.02 0.01 -0.03 \n", "148 -0.01 -0.01 0.02 0.09 \n", "149 0.00 -0.02 0.06 -0.11 \n", "\n", "[150 rows x 11 columns]" ] }, "execution_count": 308, "metadata": {}, "output_type": "execute_result" } ], "source": [ "injury_grouped_data" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Name player_height player_weight age pts_change gp_change reb_change ast_change net_rating_change oreb_pct_change dreb_pct_change usg_pct_change ts_pct_change\n", "Al Harrington 205.74 111.130040 33.0 -5.33 -47.00 -1.63 -0.20 -20.20 0.01 0.03 0.02 -0.09 1\n", "Luc Mbah a Moute 203.20 104.326160 30.0 -2.99 24.00 -1.01 -0.33 11.28 0.00 -0.02 -0.04 -0.01 1\n", " 32.0 -2.99 24.00 -1.01 -0.33 11.28 0.00 -0.02 -0.04 -0.01 1\n", " 33.0 -2.99 24.00 -1.01 -0.33 11.28 0.00 -0.02 -0.04 -0.01 1\n", "Luke Babbitt 205.74 102.058200 22.0 0.53 2.71 -0.41 -0.01 1.23 -0.02 -0.05 -0.01 0.06 1\n", " ..\n", "Gary Neal 193.04 95.254320 32.0 1.17 -12.00 0.22 -0.33 -6.07 0.01 0.01 -0.01 0.08 1\n", "Glen Davis 205.74 131.088088 25.0 -0.62 6.25 0.12 -0.05 -5.08 0.01 -0.01 -0.02 0.00 1\n", " 26.0 -0.62 6.25 0.12 -0.05 -5.08 0.01 -0.01 -0.02 0.00 1\n", " 27.0 -0.62 6.25 0.12 -0.05 -5.08 0.01 -0.01 -0.02 0.00 1\n", "Zion Williamson 200.66 128.820128 20.0 -4.50 -37.00 -0.90 -1.60 3.00 0.01 -0.02 0.00 -0.03 1\n", "Name: count, Length: 912, dtype: int64" ] }, "execution_count": 298, "metadata": {}, "output_type": "execute_result" } ], "source": [ "grouped_data.value_counts()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "merged_data_group = pd.merge(injury_data, grouped_data, on=['Name'], suffixes=('', '_grouped'))\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "merged_data_group_injury = pd.merge(injury_data, injury_grouped_data, on=['Name'], suffixes=('', '_grouped'))\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Nameteam_abbreviationageplayer_heightplayer_weightcountrydraft_yeardraft_rounddraft_numbergp...Injury_Statuspts_changegp_changereb_changeast_changenet_rating_changeoreb_pct_changedreb_pct_changeusg_pct_changets_pct_change
0Jameer NelsonOrlando Magic29.0182.8886.182480USA200412076...Before-6.10-12.00-0.91-2.13-4.79-0.01-0.01-0.06-0.02
1Jason ThompsonSacramento Kings24.0210.82113.398000USA200811275...Before1.90-9.401.180.32-1.100.03-0.01-0.010.03
2Gary NealSan Antonio Spurs26.0193.0495.254320USA20103080...Before1.17-12.000.22-0.33-6.070.010.01-0.010.08
3Glen DavisBoston Celtics25.0205.74131.088088USA200723578...Before-0.626.250.12-0.05-5.080.01-0.01-0.020.00
4Greivis VasquezMemphis Grizzlies24.0198.1295.707912Venezuela201012870...Before-2.27-40.00-0.23-0.35-5.120.000.000.00-0.05
..................................................................
932Svi MykhailiukDetroit Pistons23.0200.6692.986360Ukraine201824756...After3.152.000.200.603.75-0.01-0.01-0.010.09
933Zion WilliamsonNew Orleans Pelicans19.0198.12128.820128USA20191124...After-4.50-37.00-0.90-1.603.000.01-0.020.00-0.03
934Zion WilliamsonNew Orleans Pelicans19.0198.12128.820128USA20191124...After-4.50-37.00-0.90-1.603.000.01-0.020.00-0.03
935Richaun HolmesSacramento Kings26.0208.28106.594120USA201523744...After3.44-13.403.00-0.104.560.010.02-0.010.06
936Jakob PoeltlSan Antonio Spurs25.0215.90111.130040Austria20161969...After3.32-0.753.180.92-1.35-0.010.000.00-0.02
\n", "

937 rows × 34 columns

\n", "
" ], "text/plain": [ " Name team_abbreviation age player_height \\\n", "0 Jameer Nelson Orlando Magic 29.0 182.88 \n", "1 Jason Thompson Sacramento Kings 24.0 210.82 \n", "2 Gary Neal San Antonio Spurs 26.0 193.04 \n", "3 Glen Davis Boston Celtics 25.0 205.74 \n", "4 Greivis Vasquez Memphis Grizzlies 24.0 198.12 \n", ".. ... ... ... ... \n", "932 Svi Mykhailiuk Detroit Pistons 23.0 200.66 \n", "933 Zion Williamson New Orleans Pelicans 19.0 198.12 \n", "934 Zion Williamson New Orleans Pelicans 19.0 198.12 \n", "935 Richaun Holmes Sacramento Kings 26.0 208.28 \n", "936 Jakob Poeltl San Antonio Spurs 25.0 215.90 \n", "\n", " player_weight country draft_year draft_round draft_number gp ... \\\n", "0 86.182480 USA 2004 1 20 76 ... \n", "1 113.398000 USA 2008 1 12 75 ... \n", "2 95.254320 USA 2010 3 0 80 ... \n", "3 131.088088 USA 2007 2 35 78 ... \n", "4 95.707912 Venezuela 2010 1 28 70 ... \n", ".. ... ... ... ... ... .. ... \n", "932 92.986360 Ukraine 2018 2 47 56 ... \n", "933 128.820128 USA 2019 1 1 24 ... \n", "934 128.820128 USA 2019 1 1 24 ... \n", "935 106.594120 USA 2015 2 37 44 ... \n", "936 111.130040 Austria 2016 1 9 69 ... \n", "\n", " Injury_Status pts_change gp_change reb_change ast_change \\\n", "0 Before -6.10 -12.00 -0.91 -2.13 \n", "1 Before 1.90 -9.40 1.18 0.32 \n", "2 Before 1.17 -12.00 0.22 -0.33 \n", "3 Before -0.62 6.25 0.12 -0.05 \n", "4 Before -2.27 -40.00 -0.23 -0.35 \n", ".. ... ... ... ... ... \n", "932 After 3.15 2.00 0.20 0.60 \n", "933 After -4.50 -37.00 -0.90 -1.60 \n", "934 After -4.50 -37.00 -0.90 -1.60 \n", "935 After 3.44 -13.40 3.00 -0.10 \n", "936 After 3.32 -0.75 3.18 0.92 \n", "\n", " net_rating_change oreb_pct_change dreb_pct_change usg_pct_change \\\n", "0 -4.79 -0.01 -0.01 -0.06 \n", "1 -1.10 0.03 -0.01 -0.01 \n", "2 -6.07 0.01 0.01 -0.01 \n", "3 -5.08 0.01 -0.01 -0.02 \n", "4 -5.12 0.00 0.00 0.00 \n", ".. ... ... ... ... \n", "932 3.75 -0.01 -0.01 -0.01 \n", "933 3.00 0.01 -0.02 0.00 \n", "934 3.00 0.01 -0.02 0.00 \n", "935 4.56 0.01 0.02 -0.01 \n", "936 -1.35 -0.01 0.00 0.00 \n", "\n", " ts_pct_change \n", "0 -0.02 \n", "1 0.03 \n", "2 0.08 \n", "3 0.00 \n", "4 -0.05 \n", ".. ... \n", "932 0.09 \n", "933 -0.03 \n", "934 -0.03 \n", "935 0.06 \n", "936 -0.02 \n", "\n", "[937 rows x 34 columns]" ] }, "execution_count": 294, "metadata": {}, "output_type": "execute_result" } ], "source": [ "merged_data_group" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "drop_duplicated = merged_data_group.drop_duplicates(subset=['Specific Injury', 'Name', 'season'])\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "drop_duplicated_injury = merged_data_group_injury.drop_duplicates(subset=['Specific Injury', 'Name', 'season'])\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "drop_duplicated_injury.to_csv('/Users/laraschuman/Desktop/CTP-Project/drop_duplicated.csv')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Nameteam_abbreviationageplayer_heightplayer_weightcountrydraft_yeardraft_rounddraft_numbergp...Specific Injury_groupedpts_changegp_changereb_changeast_changenet_rating_changeoreb_pct_changedreb_pct_changeusg_pct_changets_pct_change
0Jameer NelsonOrlando Magic29.0182.8886.182480USA200412076...right calf strain injury-6.10-12.00-0.91-2.13-4.79-0.01-0.01-0.06-0.02
1Jason ThompsonSacramento Kings24.0210.82113.398000USA200811275...right foot fracture injury1.90-9.401.180.32-1.100.03-0.01-0.010.03
2Gary NealSan Antonio Spurs26.0193.0495.254320USA20103080...left ankle sprain injury1.17-12.000.22-0.33-6.070.010.01-0.010.08
3Glen DavisBoston Celtics25.0205.74131.088088USA200723578...left foot fracture injury-0.626.250.12-0.05-5.080.01-0.01-0.020.00
4Greivis VasquezMemphis Grizzlies24.0198.1295.707912Venezuela201012870...right bone spurs injury-2.27-40.00-0.23-0.35-5.120.000.000.00-0.05
..................................................................
1166Stephen CurryGolden State Warriors32.0190.5083.914520USA2009175...left fractured hand injury-3.71-62.700.630.14-25.340.000.040.01-0.07
1167Svi MykhailiukDetroit Pistons23.0200.6692.986360Ukraine201824756...left fractured hand injury3.152.000.200.603.75-0.01-0.01-0.010.09
1168Zion WilliamsonNew Orleans Pelicans19.0198.12128.820128USA20191124...right meniscus tear injury-4.50-37.00-0.90-1.603.000.01-0.020.00-0.03
1170Richaun HolmesSacramento Kings26.0208.28106.594120USA201523744...left ankle sprain injury3.44-13.403.00-0.104.560.010.02-0.010.06
1171Jakob PoeltlSan Antonio Spurs25.0215.90111.130040Austria20161969...right sprained mcl injury3.32-0.753.180.92-1.35-0.010.000.00-0.02
\n", "

920 rows × 35 columns

\n", "
" ], "text/plain": [ " Name team_abbreviation age player_height \\\n", "0 Jameer Nelson Orlando Magic 29.0 182.88 \n", "1 Jason Thompson Sacramento Kings 24.0 210.82 \n", "2 Gary Neal San Antonio Spurs 26.0 193.04 \n", "3 Glen Davis Boston Celtics 25.0 205.74 \n", "4 Greivis Vasquez Memphis Grizzlies 24.0 198.12 \n", "... ... ... ... ... \n", "1166 Stephen Curry Golden State Warriors 32.0 190.50 \n", "1167 Svi Mykhailiuk Detroit Pistons 23.0 200.66 \n", "1168 Zion Williamson New Orleans Pelicans 19.0 198.12 \n", "1170 Richaun Holmes Sacramento Kings 26.0 208.28 \n", "1171 Jakob Poeltl San Antonio Spurs 25.0 215.90 \n", "\n", " player_weight country draft_year draft_round draft_number gp ... \\\n", "0 86.182480 USA 2004 1 20 76 ... \n", "1 113.398000 USA 2008 1 12 75 ... \n", "2 95.254320 USA 2010 3 0 80 ... \n", "3 131.088088 USA 2007 2 35 78 ... \n", "4 95.707912 Venezuela 2010 1 28 70 ... \n", "... ... ... ... ... ... .. ... \n", "1166 83.914520 USA 2009 1 7 5 ... \n", "1167 92.986360 Ukraine 2018 2 47 56 ... \n", "1168 128.820128 USA 2019 1 1 24 ... \n", "1170 106.594120 USA 2015 2 37 44 ... \n", "1171 111.130040 Austria 2016 1 9 69 ... \n", "\n", " Specific Injury_grouped pts_change gp_change reb_change \\\n", "0 right calf strain injury -6.10 -12.00 -0.91 \n", "1 right foot fracture injury 1.90 -9.40 1.18 \n", "2 left ankle sprain injury 1.17 -12.00 0.22 \n", "3 left foot fracture injury -0.62 6.25 0.12 \n", "4 right bone spurs injury -2.27 -40.00 -0.23 \n", "... ... ... ... ... \n", "1166 left fractured hand injury -3.71 -62.70 0.63 \n", "1167 left fractured hand injury 3.15 2.00 0.20 \n", "1168 right meniscus tear injury -4.50 -37.00 -0.90 \n", "1170 left ankle sprain injury 3.44 -13.40 3.00 \n", "1171 right sprained mcl injury 3.32 -0.75 3.18 \n", "\n", " ast_change net_rating_change oreb_pct_change dreb_pct_change \\\n", "0 -2.13 -4.79 -0.01 -0.01 \n", "1 0.32 -1.10 0.03 -0.01 \n", "2 -0.33 -6.07 0.01 0.01 \n", "3 -0.05 -5.08 0.01 -0.01 \n", "4 -0.35 -5.12 0.00 0.00 \n", "... ... ... ... ... \n", "1166 0.14 -25.34 0.00 0.04 \n", "1167 0.60 3.75 -0.01 -0.01 \n", "1168 -1.60 3.00 0.01 -0.02 \n", "1170 -0.10 4.56 0.01 0.02 \n", "1171 0.92 -1.35 -0.01 0.00 \n", "\n", " usg_pct_change ts_pct_change \n", "0 -0.06 -0.02 \n", "1 -0.01 0.03 \n", "2 -0.01 0.08 \n", "3 -0.02 0.00 \n", "4 0.00 -0.05 \n", "... ... ... \n", "1166 0.01 -0.07 \n", "1167 -0.01 0.09 \n", "1168 0.00 -0.03 \n", "1170 -0.01 0.06 \n", "1171 0.00 -0.02 \n", "\n", "[920 rows x 35 columns]" ] }, "execution_count": 345, "metadata": {}, "output_type": "execute_result" } ], "source": [ "drop_duplicated_injury" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
ptspts_changeSpecific Injuryseasondays_injured
72127.0-4.5NaN20200.0
93322.5-4.5right meniscus tear injury201996.0
\n", "
" ], "text/plain": [ " pts pts_change Specific Injury season days_injured\n", "721 27.0 -4.5 NaN 2020 0.0\n", "933 22.5 -4.5 right meniscus tear injury 2019 96.0" ] }, "execution_count": 371, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Filter the data for 'Name' equal to 'Zion'\n", "zion_data = drop_duplicated[drop_duplicated['Name'] == 'Zion Williamson']\n", "\n", "# Drop duplicates within the 'zion_data'\n", "zion_data_unique = zion_data.drop_duplicates()\n", "\n", "# Display the unique data for Zion\n", "# Select 'pts' and 'pts_change' columns\n", "zion_pts_data = zion_data_unique[['pts', 'pts_change','Specific Injury','season','days_injured']]\n", "\n", "# Display the selected columns\n", "zion_pts_data" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# One-hot encode the 'Specific Injury' column on the merged data\n", "merged_data_encoded = pd.get_dummies(drop_duplicated_injury, columns=['Specific Injury_grouped'], drop_first=True)\n", "\n", "\n", "# Define the features for training (including the one-hot encoded 'Specific Injury' columns)\n", "features = ['age', 'player_height', 'player_weight', 'season'] + [col for col in merged_data_encoded.columns if col.startswith('Specific Injury_grouped')]\n", "\n", "# Define the multiple target columns\n", "targets = ['pts_change', 'ast_change', 'reb_change']\n", "\n", "# Set up the features (X) and target (y) for training\n", "X_merged = merged_data_encoded[features]\n", "y_merged = merged_data_encoded[targets]" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# One-hot encode the 'Specific Injury' column on the merged data\n", "merged_data_encoded = pd.get_dummies(drop_duplicated, columns=['Specific Injury'], drop_first=True)\n", "\n", "\n", "# Define the features for training (including the one-hot encoded 'Specific Injury' columns)\n", "features = ['age', 'player_height', 'player_weight', 'season'] + [col for col in merged_data_encoded.columns if col.startswith('Specific Injury')]\n", "\n", "# Define the multiple target columns\n", "targets = ['pts_change', 'ast_change', 'reb_change']\n", "\n", "# Set up the features (X) and target (y) for training\n", "X_merged = merged_data_encoded[features]\n", "y_merged = merged_data_encoded[targets]" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Mean Absolute Error: 0.47548278985507225\n", "R²: 0.8035047608572542\n", "Model saved successfully!\n" ] } ], "source": [ "from sklearn.model_selection import train_test_split\n", "from sklearn.linear_model import LinearRegression\n", "from sklearn.ensemble import RandomForestRegressor\n", "from sklearn.metrics import mean_absolute_error, r2_score\n", "import pickle\n", "\n", "\n", "# Split into training and testing sets\n", "X_train, X_test, y_train, y_test = train_test_split(X_merged, y_merged, test_size=0.2, random_state=42)\n", "\n", "# Example: Using Random Forest Regressor\n", "model = RandomForestRegressor(random_state=42)\n", "model.fit(X_train, y_train)\n", "\n", "# Make predictions\n", "y_pred = model.predict(X_test)\n", "\n", "# Evaluate the model\n", "mae = mean_absolute_error(y_test, y_pred)\n", "r2 = r2_score(y_test, y_pred)\n", "\n", "print(f'Mean Absolute Error: {mae}')\n", "print(f'R²: {r2}')\n", "\n", "with open('/Users/laraschuman/Desktop/CTP-Project/injury_model.pkl', 'wb') as f:\n", " pickle.dump(model, f)\n", "\n", "print(\"Model saved successfully!\")\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model loaded successfully!\n", "Predicted values: [[-5.8681 -2.0426 -0.8884]\n", " [ 1.5225 0.2629 1.013 ]\n", " [ 0.9691 -0.26 0.1798]\n", " ...\n", " [-3.653 -1.1847 -0.7626]\n", " [ 2.9848 -0.1239 2.6627]\n", " [ 2.4768 0.6983 2.3908]]\n" ] } ], "source": [ "import pandas as pd\n", "\n", "# Load the pre-trained model\n", "with open('/Users/laraschuman/Desktop/CTP-Project/injury_model.pkl', 'rb') as f:\n", " loaded_model = pickle.load(f)\n", "\n", "print(\"Model loaded successfully!\")\n", "\n", "\n", "# Ensure the input data matches the training feature set\n", "# Reorder or drop columns as necessary to align with X_train during training\n", "training_features = X_train.columns # Use the features from your training set\n", "new_data = merged_data_encoded[training_features] # Select only the relevant columns\n", "\n", "# Apply any preprocessing (e.g., scaling or encoding) used during training, if applicable\n", "\n", "# Make predictions\n", "predictions = loaded_model.predict(new_data)\n", "\n", "print(f\"Predicted values: {predictions}\")\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model saved successfully!\n", "Mean Absolute Error: 0.9993055525362317\n", "R²: 0.4734527115674096\n", "Model loaded successfully!\n", "Predicted values:\n", " pts_change ast_change reb_change\n", "0 -6.1649 -2.0992 -0.9387\n", "1 0.4630 0.1794 0.1823\n", "2 0.2398 -0.3065 0.0635\n", "3 -1.0060 -0.0507 -0.2680\n", "4 -2.1998 -0.3124 -0.3197\n", ".. ... ... ...\n", "915 -1.6661 -0.1137 0.2257\n", "916 2.9123 0.5944 0.2251\n", "917 -3.2647 -1.0888 -0.5737\n", "918 2.2471 -0.0985 1.8709\n", "919 2.1491 0.6155 2.1662\n", "\n", "[920 rows x 3 columns]\n" ] } ], "source": [ "import pandas as pd\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.ensemble import RandomForestRegressor\n", "from sklearn.metrics import mean_absolute_error, r2_score\n", "import pickle\n", "\n", "# One-hot encode the 'Specific Injury' column\n", "merged_data_encoded = pd.get_dummies(drop_duplicated, columns=['Specific Injury'], drop_first=True)\n", "\n", "# Define features and targets\n", "features = ['age', 'player_height', 'player_weight', 'season'] + \\\n", " [col for col in merged_data_encoded.columns if col.startswith('Specific Injury')]\n", "targets = ['pts_change', 'ast_change', 'reb_change']\n", "\n", "X_merged = merged_data_encoded[features]\n", "y_merged = merged_data_encoded[targets]\n", "\n", "# Split into training and testing sets\n", "X_train, X_test, y_train, y_test = train_test_split(X_merged, y_merged, test_size=0.2, random_state=42)\n", "\n", "# Train the Random Forest Regressor\n", "model = RandomForestRegressor(random_state=42)\n", "model.fit(X_train, y_train)\n", "\n", "# Save the trained model\n", "model_path = '/Users/laraschuman/Desktop/CTP-Project/injury_model.pkl'\n", "with open(model_path, 'wb') as f:\n", " pickle.dump(model, f)\n", "print(\"Model saved successfully!\")\n", "\n", "# Evaluate the model\n", "y_pred = model.predict(X_test)\n", "mae = mean_absolute_error(y_test, y_pred)\n", "r2 = r2_score(y_test, y_pred)\n", "\n", "print(f\"Mean Absolute Error: {mae}\")\n", "print(f\"R²: {r2}\")\n", "\n", "# Load the saved model\n", "with open(model_path, 'rb') as f:\n", " loaded_model = pickle.load(f)\n", "print(\"Model loaded successfully!\")\n", "\n", "# Align new data with training features\n", "training_features = X_train.columns\n", "new_data = merged_data_encoded[training_features] # Ensure the columns match exactly\n", "\n", "# Reindex `new_data` to align with `training_features`\n", "new_data = new_data.reindex(columns=training_features, fill_value=0)\n", "\n", "# Make predictions\n", "predictions = loaded_model.predict(new_data)\n", "\n", "# Format predictions\n", "predicted_df = pd.DataFrame(predictions, columns=targets)\n", "print(\"Predicted values:\")\n", "print(predicted_df)\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Specific Injury\n", "left ankle sprain injury 14\n", "right foot fracture injury 11\n", "right meniscus tear injury 10\n", "left meniscus tear injury 8\n", "left acl tear injury 8\n", "right ankle sprain injury 7\n", "right fractured hand injury 7\n", "left fractured leg injury 6\n", "back surgery injury 5\n", "left hamstring injury injury 5\n", "right fractured leg injury 5\n", "right torn shoulder labrum injury 4\n", "left fractured hand injury 4\n", "right sprained mcl injury 4\n", "left foot fracture injury 4\n", "left torn hamstring injury 3\n", "left sprained mcl injury 3\n", "right hamstring injury injury 3\n", "left shoulder sprain injury 3\n", "left dislocated shoulder injury 3\n", "right quad injury injury 3\n", "left hip labrum injury 3\n", "right shoulder sprain injury 2\n", "right torn rotator cuff injury injury 2\n", "right calf strain injury 2\n", "right foot sprain injury 2\n", "left quad injury injury 2\n", "right ankle fracture injury 2\n", "right hip flexor strain injury 2\n", "left bone spurs injury 2\n", "right hip flexor surgery injury 2\n", "right hip labrum injury 2\n", "right bone spurs injury 2\n", "right torn mcl injury 1\n", "fractured leg injury 1\n", "right acl tear injury 1\n", "left hip flexor surgery injury 1\n", "left ankle fracture injury 1\n", "right arm injury injury 1\n", "left calf strain injury 1\n", "left torn shoulder labrum injury 1\n", "lower back spasm injury 1\n", "Name: count, dtype: int64" ] }, "execution_count": 440, "metadata": {}, "output_type": "execute_result" } ], "source": [ "drop_duplicated['Specific Injury'].value_counts()" ] }, { "cell_type": "code", "execution_count": 489, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/var/folders/8t/t11lp0b952n0xtfmnwbzxzvw0000gn/T/ipykernel_11952/613144370.py:1: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", " drop_duplicated_injury['specific_injury'] = drop_duplicated_injury['Specific Injury_grouped'].str.replace(\"left \", \"\", regex=False)\n", "/var/folders/8t/t11lp0b952n0xtfmnwbzxzvw0000gn/T/ipykernel_11952/613144370.py:2: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", " drop_duplicated_injury['specific_injury'] = drop_duplicated_injury['Specific Injury_grouped'].str.replace(\"right \", \"\", regex=False)\n" ] }, { "data": { "text/plain": [ "0 calf strain injury\n", "1 foot fracture injury\n", "2 left ankle sprain injury\n", "3 left foot fracture injury\n", "4 bone spurs injury\n", " ... \n", "1166 left fractured hand injury\n", "1167 left fractured hand injury\n", "1168 meniscus tear injury\n", "1170 left ankle sprain injury\n", "1171 sprained mcl injury\n", "Name: specific_injury, Length: 920, dtype: object" ] }, "execution_count": 489, "metadata": {}, "output_type": "execute_result" } ], "source": [ "drop_duplicated_injury['specific_injury'] = drop_duplicated_injury['Specific Injury_grouped'].str.replace(\"left \", \"\", regex=False)\n", "drop_duplicated_injury['specific_injury'] = drop_duplicated_injury['Specific Injury_grouped'].str.replace(\"right \", \"\", regex=False)\n", "\n", "# Drop duplicate rows based on 'specific_injury' (if needed)\n", "drop_duplicated_injury['specific_injury']" ] }, { "cell_type": "code", "execution_count": 485, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Nameteam_abbreviationageplayer_heightplayer_weightcountrydraft_yeardraft_rounddraft_numbergp...Specific Injury_groupedpts_changegp_changereb_changeast_changenet_rating_changeoreb_pct_changedreb_pct_changeusg_pct_changets_pct_change
0Jameer NelsonOrlando Magic29.0182.8886.182480USA200412076...right calf strain injury-6.10-12.00-0.91-2.13-4.79-0.01-0.01-0.06-0.02
1Jason ThompsonSacramento Kings24.0210.82113.398000USA200811275...right foot fracture injury1.90-9.401.180.32-1.100.03-0.01-0.010.03
2Gary NealSan Antonio Spurs26.0193.0495.254320USA20103080...left ankle sprain injury1.17-12.000.22-0.33-6.070.010.01-0.010.08
3Glen DavisBoston Celtics25.0205.74131.088088USA200723578...left foot fracture injury-0.626.250.12-0.05-5.080.01-0.01-0.020.00
4Greivis VasquezMemphis Grizzlies24.0198.1295.707912Venezuela201012870...right bone spurs injury-2.27-40.00-0.23-0.35-5.120.000.000.00-0.05
..................................................................
1166Stephen CurryGolden State Warriors32.0190.5083.914520USA2009175...left fractured hand injury-3.71-62.700.630.14-25.340.000.040.01-0.07
1167Svi MykhailiukDetroit Pistons23.0200.6692.986360Ukraine201824756...left fractured hand injury3.152.000.200.603.75-0.01-0.01-0.010.09
1168Zion WilliamsonNew Orleans Pelicans19.0198.12128.820128USA20191124...right meniscus tear injury-4.50-37.00-0.90-1.603.000.01-0.020.00-0.03
1170Richaun HolmesSacramento Kings26.0208.28106.594120USA201523744...left ankle sprain injury3.44-13.403.00-0.104.560.010.02-0.010.06
1171Jakob PoeltlSan Antonio Spurs25.0215.90111.130040Austria20161969...right sprained mcl injury3.32-0.753.180.92-1.35-0.010.000.00-0.02
\n", "

920 rows × 35 columns

\n", "
" ], "text/plain": [ " Name team_abbreviation age player_height \\\n", "0 Jameer Nelson Orlando Magic 29.0 182.88 \n", "1 Jason Thompson Sacramento Kings 24.0 210.82 \n", "2 Gary Neal San Antonio Spurs 26.0 193.04 \n", "3 Glen Davis Boston Celtics 25.0 205.74 \n", "4 Greivis Vasquez Memphis Grizzlies 24.0 198.12 \n", "... ... ... ... ... \n", "1166 Stephen Curry Golden State Warriors 32.0 190.50 \n", "1167 Svi Mykhailiuk Detroit Pistons 23.0 200.66 \n", "1168 Zion Williamson New Orleans Pelicans 19.0 198.12 \n", "1170 Richaun Holmes Sacramento Kings 26.0 208.28 \n", "1171 Jakob Poeltl San Antonio Spurs 25.0 215.90 \n", "\n", " player_weight country draft_year draft_round draft_number gp ... \\\n", "0 86.182480 USA 2004 1 20 76 ... \n", "1 113.398000 USA 2008 1 12 75 ... \n", "2 95.254320 USA 2010 3 0 80 ... \n", "3 131.088088 USA 2007 2 35 78 ... \n", "4 95.707912 Venezuela 2010 1 28 70 ... \n", "... ... ... ... ... ... .. ... \n", "1166 83.914520 USA 2009 1 7 5 ... \n", "1167 92.986360 Ukraine 2018 2 47 56 ... \n", "1168 128.820128 USA 2019 1 1 24 ... \n", "1170 106.594120 USA 2015 2 37 44 ... \n", "1171 111.130040 Austria 2016 1 9 69 ... \n", "\n", " Specific Injury_grouped pts_change gp_change reb_change \\\n", "0 right calf strain injury -6.10 -12.00 -0.91 \n", "1 right foot fracture injury 1.90 -9.40 1.18 \n", "2 left ankle sprain injury 1.17 -12.00 0.22 \n", "3 left foot fracture injury -0.62 6.25 0.12 \n", "4 right bone spurs injury -2.27 -40.00 -0.23 \n", "... ... ... ... ... \n", "1166 left fractured hand injury -3.71 -62.70 0.63 \n", "1167 left fractured hand injury 3.15 2.00 0.20 \n", "1168 right meniscus tear injury -4.50 -37.00 -0.90 \n", "1170 left ankle sprain injury 3.44 -13.40 3.00 \n", "1171 right sprained mcl injury 3.32 -0.75 3.18 \n", "\n", " ast_change net_rating_change oreb_pct_change dreb_pct_change \\\n", "0 -2.13 -4.79 -0.01 -0.01 \n", "1 0.32 -1.10 0.03 -0.01 \n", "2 -0.33 -6.07 0.01 0.01 \n", "3 -0.05 -5.08 0.01 -0.01 \n", "4 -0.35 -5.12 0.00 0.00 \n", "... ... ... ... ... \n", "1166 0.14 -25.34 0.00 0.04 \n", "1167 0.60 3.75 -0.01 -0.01 \n", "1168 -1.60 3.00 0.01 -0.02 \n", "1170 -0.10 4.56 0.01 0.02 \n", "1171 0.92 -1.35 -0.01 0.00 \n", "\n", " usg_pct_change ts_pct_change \n", "0 -0.06 -0.02 \n", "1 -0.01 0.03 \n", "2 -0.01 0.08 \n", "3 -0.02 0.00 \n", "4 0.00 -0.05 \n", "... ... ... \n", "1166 0.01 -0.07 \n", "1167 -0.01 0.09 \n", "1168 0.00 -0.03 \n", "1170 -0.01 0.06 \n", "1171 0.00 -0.02 \n", "\n", "[920 rows x 35 columns]" ] }, "execution_count": 485, "metadata": {}, "output_type": "execute_result" } ], "source": [ "drop_duplicated_injury" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Specific Injury\n", "foot fracture injury 1\n", "hip flexor surgery injury 1\n", "calf strain injury 1\n", "quad injury injury 1\n", "shoulder sprain injury 1\n", "foot sprain injury 1\n", "torn rotator cuff injury injury 1\n", "torn mcl injury 1\n", "hip flexor strain injury 1\n", "fractured leg injury 1\n", "sprained mcl injury 1\n", "ankle sprain injury 1\n", "hamstring injury injury 1\n", "meniscus tear injury 1\n", "torn hamstring injury 1\n", "dislocated shoulder injury 1\n", "ankle fracture injury 1\n", "fractured hand injury 1\n", "bone spurs injury 1\n", "acl tear injury 1\n", "hip labrum injury 1\n", "back surgery injury 1\n", "arm injury injury 1\n", "torn shoulder labrum injury 1\n", "lower back spasm injury 1\n", "Name: count, dtype: int64" ] }, "execution_count": 448, "metadata": {}, "output_type": "execute_result" } ], "source": [ "drop_duplicated['Specific Injury'].value_counts()" ] }, { "cell_type": "code", "execution_count": 492, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Evaluation Metrics by Target:\n", "pts_change: MAE = 1.273210045289855, R² = 0.6798948156812019\n", "ast_change: MAE = 0.31011294038992415, R² = 0.7708349254437783\n", "reb_change: MAE = 0.43051706521739136, R² = 0.6194001893032414\n", "Top Features by Importance:\n", " Feature Importance\n", "2 player_weight 0.341395\n", "1 player_height 0.223940\n", "0 age 0.117367\n", "12 specific_injury_fractured leg injury 0.053004\n", "13 specific_injury_sprained mcl injury 0.031181\n", "Model saved successfully to /Users/laraschuman/Desktop/CTP-Project/rf_injury_change_model.pkl!\n" ] } ], "source": [ "import pandas as pd\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.ensemble import RandomForestRegressor\n", "from sklearn.multioutput import MultiOutputRegressor\n", "from sklearn.metrics import mean_absolute_error, r2_score\n", "from sklearn.impute import SimpleImputer\n", "import pickle\n", "\n", "# Step 1: Preprocess the Data\n", "# One-hot encode 'Specific Injury' column\n", "merged_data_encoded = pd.get_dummies(drop_duplicated_injury, columns=['specific_injury'], drop_first=True)\n", "\n", "# Features for training (including one-hot encoded 'Specific Injury')\n", "features = ['age', 'player_height', 'player_weight', 'days_injured'] + [\n", " col for col in merged_data_encoded.columns if col.startswith('specific_injury_')\n", "]\n", "\n", "# Targets\n", "targets = ['pts_change', 'ast_change', 'reb_change']\n", "\n", "# Set up features (X) and targets (y)\n", "X_merged = merged_data_encoded[features]\n", "y_merged = merged_data_encoded[targets]\n", "\n", "# Handle missing values\n", "imputer = SimpleImputer(strategy=\"median\")\n", "X_merged = pd.DataFrame(imputer.fit_transform(X_merged), columns=X_merged.columns)\n", "\n", "# Step 2: Train-Test Split\n", "X_train, X_test, y_train, y_test = train_test_split(X_merged, y_merged, test_size=0.2, random_state=42)\n", "\n", "# Step 3: Train the Model\n", "rf_model = MultiOutputRegressor(RandomForestRegressor(random_state=42))\n", "rf_model.fit(X_train, y_train)\n", "\n", "# Step 4: Evaluate the Model\n", "y_pred = rf_model.predict(X_test)\n", "\n", "# Calculate Mean Absolute Error (MAE) and R² for each target\n", "evaluation_metrics = {\n", " target: {\n", " 'MAE': mean_absolute_error(y_test[target], y_pred[:, idx]),\n", " 'R²': r2_score(y_test[target], y_pred[:, idx])\n", " }\n", " for idx, target in enumerate(targets)\n", "}\n", "\n", "print(\"Evaluation Metrics by Target:\")\n", "for target, metrics in evaluation_metrics.items():\n", " print(f\"{target}: MAE = {metrics['MAE']}, R² = {metrics['R²']}\")\n", "\n", "# Step 5: Feature Importance\n", "# Get feature importance from the first Random Forest Regressor as a proxy\n", "feature_importances = pd.DataFrame({\n", " 'Feature': X_merged.columns,\n", " 'Importance': rf_model.estimators_[0].feature_importances_\n", "}).sort_values(by='Importance', ascending=False)\n", "\n", "print(\"Top Features by Importance:\")\n", "print(feature_importances.head())\n", "\n", "# Step 6: Save the Model\n", "model_path = '/Users/laraschuman/Desktop/CTP-Project/rf_injury_change_model.pkl'\n", "with open(model_path, 'wb') as f:\n", " pickle.dump(rf_model, f)\n", "\n", "print(f\"Model saved successfully to {model_path}!\")\n" ] }, { "cell_type": "code", "execution_count": 493, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/var/folders/8t/t11lp0b952n0xtfmnwbzxzvw0000gn/T/ipykernel_11952/1065402976.py:42: SettingWithCopyWarning: \n", "A value is trying to be set on a copy of a slice from a DataFrame.\n", "Try using .loc[row_indexer,col_indexer] = value instead\n", "\n", "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", " drop_duplicated_injury['specific_injury'] = pd.Categorical(drop_duplicated_injury['specific_injury'], categories=injury_list, ordered=False)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Evaluation Metrics by Target:\n", "pts_change: MAE = 1.225611992753623, R² = 0.6870166424159725\n", "ast_change: MAE = 0.30058603260869576, R² = 0.7719661450788522\n", "reb_change: MAE = 0.41204285196687374, R² = 0.6485851920112147\n", "Top Injury Features by Importance:\n", " Feature Importance\n", "13 fractured leg injury 0.053007\n", "14 sprained mcl injury 0.030865\n", "7 quad injury injury 0.026076\n", "15 ankle sprain injury 0.025429\n", "4 foot fracture injury 0.023935\n", "17 meniscus tear injury 0.022673\n", "16 hamstring injury injury 0.022511\n", "5 hip flexor surgery injury 0.021272\n", "6 calf strain injury 0.017689\n", "27 torn shoulder labrum injury 0.012570\n" ] }, { "data": { "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Model saved successfully to /Users/laraschuman/Desktop/CTP-Project/rf_injury_change_model.pkl!\n" ] } ], "source": [ "import pandas as pd\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.ensemble import RandomForestRegressor\n", "from sklearn.multioutput import MultiOutputRegressor\n", "from sklearn.metrics import mean_absolute_error, r2_score\n", "from sklearn.impute import SimpleImputer\n", "import pickle\n", "import matplotlib.pyplot as plt\n", "\n", "# List of all possible injuries to ensure consistent encoding\n", "injury_list = [\n", " \"foot fracture injury\",\n", " \"hip flexor surgery injury\",\n", " \"calf strain injury\",\n", " \"quad injury injury\",\n", " \"shoulder sprain injury\",\n", " \"foot sprain injury\",\n", " \"torn rotator cuff injury injury\",\n", " \"torn mcl injury\",\n", " \"hip flexor strain injury\",\n", " \"fractured leg injury\",\n", " \"sprained mcl injury\",\n", " \"ankle sprain injury\",\n", " \"hamstring injury injury\",\n", " \"meniscus tear injury\",\n", " \"torn hamstring injury\",\n", " \"dislocated shoulder injury\",\n", " \"ankle fracture injury\",\n", " \"fractured hand injury\",\n", " \"bone spurs injury\",\n", " \"acl tear injury\",\n", " \"hip labrum injury\",\n", " \"back surgery injury\",\n", " \"arm injury injury\",\n", " \"torn shoulder labrum injury\",\n", " \"lower back spasm injury\"\n", "]\n", "\n", "\n", "# Step 1: Preprocess the Data\n", "# One-hot encode 'specific_injury' column using the injury_list\n", "drop_duplicated_injury['specific_injury'] = pd.Categorical(drop_duplicated_injury['specific_injury'], categories=injury_list, ordered=False)\n", "merged_data_encoded = pd.get_dummies(drop_duplicated_injury, columns=['specific_injury'], prefix='', prefix_sep='')\n", "\n", "# Features for training (ensuring all injuries are included)\n", "features = ['age', 'player_height', 'player_weight', 'days_injured'] + [\n", " injury for injury in injury_list if injury in merged_data_encoded.columns\n", "]\n", "\n", "# Targets\n", "targets = ['pts_change', 'ast_change', 'reb_change']\n", "\n", "# Set up features (X) and targets (y)\n", "X_merged = merged_data_encoded[features]\n", "y_merged = merged_data_encoded[targets]\n", "\n", "# Handle missing values\n", "imputer = SimpleImputer(strategy=\"median\")\n", "X_merged = pd.DataFrame(imputer.fit_transform(X_merged), columns=X_merged.columns)\n", "\n", "# Step 2: Train-Test Split\n", "X_train, X_test, y_train, y_test = train_test_split(X_merged, y_merged, test_size=0.2, random_state=42)\n", "\n", "# Step 3: Train the Model\n", "rf_model = MultiOutputRegressor(RandomForestRegressor(random_state=42))\n", "rf_model.fit(X_train, y_train)\n", "\n", "# Step 4: Evaluate the Model\n", "y_pred = rf_model.predict(X_test)\n", "\n", "# Calculate Mean Absolute Error (MAE) and R² for each target\n", "evaluation_metrics = {\n", " target: {\n", " 'MAE': mean_absolute_error(y_test[target], y_pred[:, idx]),\n", " 'R²': r2_score(y_test[target], y_pred[:, idx])\n", " }\n", " for idx, target in enumerate(targets)\n", "}\n", "\n", "print(\"Evaluation Metrics by Target:\")\n", "for target, metrics in evaluation_metrics.items():\n", " print(f\"{target}: MAE = {metrics['MAE']}, R² = {metrics['R²']}\")\n", "\n", "# Step 5: Feature Importance\n", "# Get feature importance from the first Random Forest Regressor as a proxy\n", "feature_importances = pd.DataFrame({\n", " 'Feature': X_merged.columns,\n", " 'Importance': rf_model.estimators_[0].feature_importances_\n", "}).sort_values(by='Importance', ascending=False)\n", "\n", "# Display and analyze specific injury impact\n", "injury_importances = feature_importances[feature_importances['Feature'].isin(injury_list)]\n", "\n", "print(\"Top Injury Features by Importance:\")\n", "print(injury_importances.head(10))\n", "\n", "# Plot injury feature importances\n", "plt.figure(figsize=(12, 6))\n", "injury_importances.head(10).plot(kind='barh', x='Feature', y='Importance', legend=False, title=\"Top 10 Injury Features by Importance\")\n", "plt.xlabel(\"Importance\")\n", "plt.ylabel(\"Injury Type\")\n", "plt.show()\n", "\n", "# Save feature importance\n", "feature_importances.to_csv('/Users/laraschuman/Desktop/CTP-Project/feature_importances.csv', index=False)\n", "\n", "# Step 6: Save the Model\n", "model_path = '/Users/laraschuman/Desktop/CTP-Project/rf_injury_change_model.pkl'\n", "with open(model_path, 'wb') as f:\n", " pickle.dump(rf_model, f)\n", "\n", "print(f\"Model saved successfully to {model_path}!\")\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "threshold = 0" ] }, { "cell_type": "code", "execution_count": 498, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "days_injured\n", "0.0 766\n", "260.0 5\n", "125.0 4\n", "129.0 3\n", "246.0 3\n", " ... \n", "242.0 1\n", "136.0 1\n", "184.0 1\n", "191.0 1\n", "152.0 1\n", "Name: count, Length: 115, dtype: int64" ] }, "execution_count": 498, "metadata": {}, "output_type": "execute_result" } ], "source": [ "drop_duplicated_injury['days_injured'].value_counts()" ] }, { "cell_type": "code", "execution_count": 500, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Nameteam_abbreviationageplayer_heightplayer_weightcountrydraft_yeardraft_rounddraft_numbergp...pts_changegp_changereb_changeast_changenet_rating_changeoreb_pct_changedreb_pct_changeusg_pct_changets_pct_changespecific_injury
934Reggie EvansToronto Raptors31.0203.20111.130040USA20103030...0.87-29.254.380.82-6.350.030.01-0.02-0.03foot fracture injury
935Jason ThompsonSacramento Kings25.0210.82113.398000USA200811264...1.90-9.401.180.32-1.100.03-0.01-0.010.03foot fracture injury
936Eric BledsoeLos Angeles Clippers22.0185.4288.450440USA201011840...-11.91-26.20-2.49-3.342.060.03-0.01-0.03-0.10meniscus tear injury
937Dominique JonesDallas Mavericks23.0195.5897.522280USA201012533...-0.459.50-0.20-0.70-6.35-0.020.01-0.030.07foot fracture injury
939Toney DouglasNew York Knicks26.0187.9683.914520USA200912938...-0.47-12.00-0.30-0.22-1.370.00-0.020.06-0.11torn shoulder labrum injury
..................................................................
1166Stephen CurryGolden State Warriors32.0190.5083.914520USA2009175...-3.71-62.700.630.14-25.340.000.040.01-0.07NaN
1167Svi MykhailiukDetroit Pistons23.0200.6692.986360Ukraine201824756...3.152.000.200.603.75-0.01-0.01-0.010.09NaN
1168Zion WilliamsonNew Orleans Pelicans19.0198.12128.820128USA20191124...-4.50-37.00-0.90-1.603.000.01-0.020.00-0.03meniscus tear injury
1170Richaun HolmesSacramento Kings26.0208.28106.594120USA201523744...3.44-13.403.00-0.104.560.010.02-0.010.06NaN
1171Jakob PoeltlSan Antonio Spurs25.0215.90111.130040Austria20161969...3.32-0.753.180.92-1.35-0.010.000.00-0.02sprained mcl injury
\n", "

154 rows × 36 columns

\n", "
" ], "text/plain": [ " Name team_abbreviation age player_height \\\n", "934 Reggie Evans Toronto Raptors 31.0 203.20 \n", "935 Jason Thompson Sacramento Kings 25.0 210.82 \n", "936 Eric Bledsoe Los Angeles Clippers 22.0 185.42 \n", "937 Dominique Jones Dallas Mavericks 23.0 195.58 \n", "939 Toney Douglas New York Knicks 26.0 187.96 \n", "... ... ... ... ... \n", "1166 Stephen Curry Golden State Warriors 32.0 190.50 \n", "1167 Svi Mykhailiuk Detroit Pistons 23.0 200.66 \n", "1168 Zion Williamson New Orleans Pelicans 19.0 198.12 \n", "1170 Richaun Holmes Sacramento Kings 26.0 208.28 \n", "1171 Jakob Poeltl San Antonio Spurs 25.0 215.90 \n", "\n", " player_weight country draft_year draft_round draft_number gp ... \\\n", "934 111.130040 USA 2010 3 0 30 ... \n", "935 113.398000 USA 2008 1 12 64 ... \n", "936 88.450440 USA 2010 1 18 40 ... \n", "937 97.522280 USA 2010 1 25 33 ... \n", "939 83.914520 USA 2009 1 29 38 ... \n", "... ... ... ... ... ... .. ... \n", "1166 83.914520 USA 2009 1 7 5 ... \n", "1167 92.986360 Ukraine 2018 2 47 56 ... \n", "1168 128.820128 USA 2019 1 1 24 ... \n", "1170 106.594120 USA 2015 2 37 44 ... \n", "1171 111.130040 Austria 2016 1 9 69 ... \n", "\n", " pts_change gp_change reb_change ast_change net_rating_change \\\n", "934 0.87 -29.25 4.38 0.82 -6.35 \n", "935 1.90 -9.40 1.18 0.32 -1.10 \n", "936 -11.91 -26.20 -2.49 -3.34 2.06 \n", "937 -0.45 9.50 -0.20 -0.70 -6.35 \n", "939 -0.47 -12.00 -0.30 -0.22 -1.37 \n", "... ... ... ... ... ... \n", "1166 -3.71 -62.70 0.63 0.14 -25.34 \n", "1167 3.15 2.00 0.20 0.60 3.75 \n", "1168 -4.50 -37.00 -0.90 -1.60 3.00 \n", "1170 3.44 -13.40 3.00 -0.10 4.56 \n", "1171 3.32 -0.75 3.18 0.92 -1.35 \n", "\n", " oreb_pct_change dreb_pct_change usg_pct_change ts_pct_change \\\n", "934 0.03 0.01 -0.02 -0.03 \n", "935 0.03 -0.01 -0.01 0.03 \n", "936 0.03 -0.01 -0.03 -0.10 \n", "937 -0.02 0.01 -0.03 0.07 \n", "939 0.00 -0.02 0.06 -0.11 \n", "... ... ... ... ... \n", "1166 0.00 0.04 0.01 -0.07 \n", "1167 -0.01 -0.01 -0.01 0.09 \n", "1168 0.01 -0.02 0.00 -0.03 \n", "1170 0.01 0.02 -0.01 0.06 \n", "1171 -0.01 0.00 0.00 -0.02 \n", "\n", " specific_injury \n", "934 foot fracture injury \n", "935 foot fracture injury \n", "936 meniscus tear injury \n", "937 foot fracture injury \n", "939 torn shoulder labrum injury \n", "... ... \n", "1166 NaN \n", "1167 NaN \n", "1168 meniscus tear injury \n", "1170 NaN \n", "1171 sprained mcl injury \n", "\n", "[154 rows x 36 columns]" ] }, "execution_count": 500, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Filter out rows where 'days_injured' is equal to zero\n", "filtered_injury_data = drop_duplicated_injury[drop_duplicated_injury['days_injured'] != 0]\n", "\n", "# Display the value counts for the filtered data\n", "filtered_injury_data" ] }, { "cell_type": "code", "execution_count": 501, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/var/folders/8t/t11lp0b952n0xtfmnwbzxzvw0000gn/T/ipykernel_11952/1296445815.py:1: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.\n", " avg_days_injured = filtered_injury_data.groupby('specific_injury')['days_injured'].mean()\n" ] }, { "data": { "text/plain": [ "specific_injury\n", "foot fracture injury 207.666667\n", "hip flexor surgery injury 256.000000\n", "calf strain injury 236.000000\n", "quad injury injury 283.000000\n", "shoulder sprain injury 259.500000\n", "foot sprain injury 294.000000\n", "torn rotator cuff injury injury NaN\n", "torn mcl injury 271.000000\n", "hip flexor strain injury 253.000000\n", "fractured leg injury 250.250000\n", "sprained mcl injury 228.666667\n", "ankle sprain injury 231.333333\n", "hamstring injury injury 220.000000\n", "meniscus tear injury 201.250000\n", "torn hamstring injury NaN\n", "dislocated shoulder injury NaN\n", "ankle fracture injury 114.500000\n", "fractured hand injury 169.142857\n", "bone spurs injury 151.500000\n", "acl tear injury 268.000000\n", "hip labrum injury 247.500000\n", "back surgery injury 215.800000\n", "arm injury injury 303.666667\n", "torn shoulder labrum injury 195.666667\n", "lower back spasm injury NaN\n", "Name: days_injured, dtype: float64" ] }, "execution_count": 501, "metadata": {}, "output_type": "execute_result" } ], "source": [ "avg_days_injured = filtered_injury_data.groupby('specific_injury')['days_injured'].mean()\n", "avg_days_injured" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "y_test_class = (y_test['pts_change'] > threshold).astype(int)\n", "\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "if y_pred.ndim == 1: # For single-output regression\n", " y_pred_class = (y_pred > threshold).astype(int)\n", "else: # Multi-output regression\n", " y_pred_class = (y_pred[:, 0] > threshold).astype(int)\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 0.91\n", "Recall: 0.86\n", "F1-Score: 0.89\n" ] } ], "source": [ "from sklearn.metrics import accuracy_score, recall_score, f1_score\n", "\n", "# Classification Metrics\n", "accuracy = accuracy_score(y_test_class, y_pred_class)\n", "recall = recall_score(y_test_class, y_pred_class)\n", "f1 = f1_score(y_test_class, y_pred_class)\n", "\n", "print(f'Accuracy: {accuracy:.2f}')\n", "print(f'Recall: {recall:.2f}')\n", "print(f'F1-Score: {f1:.2f}')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "\n", "sns.barplot(data=injury_grouped_data, x='Specific Injury', y='pts_change')\n", "plt.xticks(rotation=90)\n", "plt.title('Performance Change by Injury Type')\n", "plt.xlabel('Injury Type')\n", "plt.ylabel('Average Points Change')\n", "plt.show()\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "corr_matrix = merged_data_encoded[['pts_change', 'gp_change', 'reb_change', 'ast_change', 'net_rating_change']].corr()\n", "plt.figure(figsize=(8,6))\n", "sns.heatmap(corr_matrix, annot=True, cmap='coolwarm', fmt='.2f', cbar=True)\n", "plt.title('Correlation Matrix of Performance Changes')\n", "plt.show()\n" ] } ], "metadata": { "kernelspec": { "display_name": "base", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.4" } }, "nbformat": 4, "nbformat_minor": 2 }