{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from mplsoccer.pitch import Pitch\n",
    "\n",
    "from sklearn.cluster import KMeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import data\n",
    "df = pd.read_csv('kmeanstutorial.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>ball_receipt_outcome</th>\n",
       "      <th>ball_recovery_recovery_failure</th>\n",
       "      <th>block_deflection</th>\n",
       "      <th>carry_end_location</th>\n",
       "      <th>clearance_aerial_won</th>\n",
       "      <th>counterpress</th>\n",
       "      <th>dribble_outcome</th>\n",
       "      <th>dribble_overrun</th>\n",
       "      <th>duel_outcome</th>\n",
       "      <th>...</th>\n",
       "      <th>shot_statsbomb_xg</th>\n",
       "      <th>shot_technique</th>\n",
       "      <th>shot_type</th>\n",
       "      <th>substitution_outcome</th>\n",
       "      <th>substitution_replacement</th>\n",
       "      <th>tactics</th>\n",
       "      <th>team</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>type</th>\n",
       "      <th>under_pressure</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'formation': 442, 'lineup': [{'player': {'id'...</td>\n",
       "      <td>France</td>\n",
       "      <td>00:00:00.000</td>\n",
       "      <td>Starting XI</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'formation': 433, 'lineup': [{'player': {'id'...</td>\n",
       "      <td>Croatia</td>\n",
       "      <td>00:00:00.000</td>\n",
       "      <td>Starting XI</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>France</td>\n",
       "      <td>00:00:00.000</td>\n",
       "      <td>Half Start</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Croatia</td>\n",
       "      <td>00:00:00.000</td>\n",
       "      <td>Half Start</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Croatia</td>\n",
       "      <td>00:00:00.000</td>\n",
       "      <td>Half Start</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 74 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0 ball_receipt_outcome ball_recovery_recovery_failure  \\\n",
       "0           0                  NaN                            NaN   \n",
       "1           1                  NaN                            NaN   \n",
       "2           2                  NaN                            NaN   \n",
       "3           3                  NaN                            NaN   \n",
       "4           4                  NaN                            NaN   \n",
       "\n",
       "  block_deflection carry_end_location clearance_aerial_won counterpress  \\\n",
       "0              NaN                NaN                  NaN          NaN   \n",
       "1              NaN                NaN                  NaN          NaN   \n",
       "2              NaN                NaN                  NaN          NaN   \n",
       "3              NaN                NaN                  NaN          NaN   \n",
       "4              NaN                NaN                  NaN          NaN   \n",
       "\n",
       "  dribble_outcome dribble_overrun duel_outcome  ... shot_statsbomb_xg  \\\n",
       "0             NaN             NaN          NaN  ...               NaN   \n",
       "1             NaN             NaN          NaN  ...               NaN   \n",
       "2             NaN             NaN          NaN  ...               NaN   \n",
       "3             NaN             NaN          NaN  ...               NaN   \n",
       "4             NaN             NaN          NaN  ...               NaN   \n",
       "\n",
       "   shot_technique shot_type substitution_outcome substitution_replacement  \\\n",
       "0             NaN       NaN                  NaN                      NaN   \n",
       "1             NaN       NaN                  NaN                      NaN   \n",
       "2             NaN       NaN                  NaN                      NaN   \n",
       "3             NaN       NaN                  NaN                      NaN   \n",
       "4             NaN       NaN                  NaN                      NaN   \n",
       "\n",
       "                                             tactics     team     timestamp  \\\n",
       "0  {'formation': 442, 'lineup': [{'player': {'id'...   France  00:00:00.000   \n",
       "1  {'formation': 433, 'lineup': [{'player': {'id'...  Croatia  00:00:00.000   \n",
       "2                                                NaN   France  00:00:00.000   \n",
       "3                                                NaN  Croatia  00:00:00.000   \n",
       "4                                                NaN  Croatia  00:00:00.000   \n",
       "\n",
       "          type under_pressure  \n",
       "0  Starting XI            NaN  \n",
       "1  Starting XI            NaN  \n",
       "2   Half Start            NaN  \n",
       "3   Half Start            NaN  \n",
       "4   Half Start            NaN  \n",
       "\n",
       "[5 rows x 74 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Unnamed: 0', 'ball_receipt_outcome', 'ball_recovery_recovery_failure',\n",
       "       'block_deflection', 'carry_end_location', 'clearance_aerial_won',\n",
       "       'counterpress', 'dribble_outcome', 'dribble_overrun', 'duel_outcome',\n",
       "       'duel_type', 'duration', 'foul_committed_advantage',\n",
       "       'foul_committed_card', 'foul_committed_penalty', 'foul_committed_type',\n",
       "       'foul_won_advantage', 'foul_won_defensive', 'goalkeeper_body_part',\n",
       "       'goalkeeper_end_location', 'goalkeeper_outcome', 'goalkeeper_position',\n",
       "       'goalkeeper_technique', 'goalkeeper_type', 'id', 'index',\n",
       "       'injury_stoppage_in_chain', 'interception_outcome', 'location',\n",
       "       'match_id', 'minute', 'pass_aerial_won', 'pass_angle',\n",
       "       'pass_assisted_shot_id', 'pass_backheel', 'pass_body_part',\n",
       "       'pass_cross', 'pass_cut_back', 'pass_deflected', 'pass_end_location',\n",
       "       'pass_goal_assist', 'pass_height', 'pass_length', 'pass_outcome',\n",
       "       'pass_recipient', 'pass_shot_assist', 'pass_switch', 'pass_type',\n",
       "       'period', 'play_pattern', 'player', 'position', 'possession',\n",
       "       'possession_team', 'related_events', 'second', 'shot_aerial_won',\n",
       "       'shot_body_part', 'shot_deflected', 'shot_end_location',\n",
       "       'shot_first_time', 'shot_freeze_frame', 'shot_key_pass_id',\n",
       "       'shot_outcome', 'shot_statsbomb_xg', 'shot_technique', 'shot_type',\n",
       "       'substitution_outcome', 'substitution_replacement', 'tactics', 'team',\n",
       "       'timestamp', 'type', 'under_pressure'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[['team','type','location','pass_end_location']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>team</th>\n",
       "      <th>type</th>\n",
       "      <th>location</th>\n",
       "      <th>pass_end_location</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>France</td>\n",
       "      <td>Starting XI</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Croatia</td>\n",
       "      <td>Starting XI</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>France</td>\n",
       "      <td>Half Start</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Croatia</td>\n",
       "      <td>Half Start</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Croatia</td>\n",
       "      <td>Half Start</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      team         type location pass_end_location\n",
       "0   France  Starting XI      NaN               NaN\n",
       "1  Croatia  Starting XI      NaN               NaN\n",
       "2   France   Half Start      NaN               NaN\n",
       "3  Croatia   Half Start      NaN               NaN\n",
       "4  Croatia   Half Start      NaN               NaN"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[(df['team']=='France')&(df['type']=='Pass')].reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>team</th>\n",
       "      <th>type</th>\n",
       "      <th>location</th>\n",
       "      <th>pass_end_location</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[48.0, 50.0]</td>\n",
       "      <td>[48.0, 60.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>24</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[49.0, 80.0]</td>\n",
       "      <td>[46.0, 61.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[65.0, 64.0]</td>\n",
       "      <td>[66.0, 69.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>28</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[63.0, 73.0]</td>\n",
       "      <td>[65.0, 79.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>29</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[58.0, 79.0]</td>\n",
       "      <td>[26.0, 69.0]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   index    team  type      location pass_end_location\n",
       "0     11  France  Pass  [48.0, 50.0]      [48.0, 60.0]\n",
       "1     24  France  Pass  [49.0, 80.0]      [46.0, 61.0]\n",
       "2     25  France  Pass  [65.0, 64.0]      [66.0, 69.0]\n",
       "3     28  France  Pass  [63.0, 73.0]      [65.0, 79.0]\n",
       "4     29  France  Pass  [58.0, 79.0]      [26.0, 69.0]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('O')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.location.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[['x','y']] = df.location.str.split(expand=True)\n",
    "df[['endX','endY']] = df.pass_end_location.str.split(expand=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>team</th>\n",
       "      <th>type</th>\n",
       "      <th>location</th>\n",
       "      <th>pass_end_location</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>endX</th>\n",
       "      <th>endY</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[48.0, 50.0]</td>\n",
       "      <td>[48.0, 60.0]</td>\n",
       "      <td>[48.0,</td>\n",
       "      <td>50.0]</td>\n",
       "      <td>[48.0,</td>\n",
       "      <td>60.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>24</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[49.0, 80.0]</td>\n",
       "      <td>[46.0, 61.0]</td>\n",
       "      <td>[49.0,</td>\n",
       "      <td>80.0]</td>\n",
       "      <td>[46.0,</td>\n",
       "      <td>61.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[65.0, 64.0]</td>\n",
       "      <td>[66.0, 69.0]</td>\n",
       "      <td>[65.0,</td>\n",
       "      <td>64.0]</td>\n",
       "      <td>[66.0,</td>\n",
       "      <td>69.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>28</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[63.0, 73.0]</td>\n",
       "      <td>[65.0, 79.0]</td>\n",
       "      <td>[63.0,</td>\n",
       "      <td>73.0]</td>\n",
       "      <td>[65.0,</td>\n",
       "      <td>79.0]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>29</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>[58.0, 79.0]</td>\n",
       "      <td>[26.0, 69.0]</td>\n",
       "      <td>[58.0,</td>\n",
       "      <td>79.0]</td>\n",
       "      <td>[26.0,</td>\n",
       "      <td>69.0]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   index    team  type      location pass_end_location       x      y    endX  \\\n",
       "0     11  France  Pass  [48.0, 50.0]      [48.0, 60.0]  [48.0,  50.0]  [48.0,   \n",
       "1     24  France  Pass  [49.0, 80.0]      [46.0, 61.0]  [49.0,  80.0]  [46.0,   \n",
       "2     25  France  Pass  [65.0, 64.0]      [66.0, 69.0]  [65.0,  64.0]  [66.0,   \n",
       "3     28  France  Pass  [63.0, 73.0]      [65.0, 79.0]  [63.0,  73.0]  [65.0,   \n",
       "4     29  France  Pass  [58.0, 79.0]      [26.0, 69.0]  [58.0,  79.0]  [26.0,   \n",
       "\n",
       "    endY  \n",
       "0  60.0]  \n",
       "1  61.0]  \n",
       "2  69.0]  \n",
       "3  79.0]  \n",
       "4  69.0]  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['x'] = df.loc[:,'x'] = df.x.map(lambda x: x[1:-1]).astype(float)\n",
    "df['y'] = df.loc[:,'y'] = df.y.map(lambda x: x[0:-1]).astype(float)\n",
    "df['endX'] = df.loc[:,'endX'] = df.endX.map(lambda x: x[1:-1]).astype(float)\n",
    "df['endY'] = df.loc[:,'endY'] = df.endY.map(lambda x: x[0:-1]).astype(float)\n",
    "df = df.drop(['location','pass_end_location'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>team</th>\n",
       "      <th>type</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>endX</th>\n",
       "      <th>endY</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>48.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>48.0</td>\n",
       "      <td>60.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>24</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>49.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>61.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>65.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>69.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>28</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>63.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>65.0</td>\n",
       "      <td>79.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>29</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>58.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>69.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   index    team  type     x     y  endX  endY\n",
       "0     11  France  Pass  48.0  50.0  48.0  60.0\n",
       "1     24  France  Pass  49.0  80.0  46.0  61.0\n",
       "2     25  France  Pass  65.0  64.0  66.0  69.0\n",
       "3     28  France  Pass  63.0  73.0  65.0  79.0\n",
       "4     29  France  Pass  58.0  79.0  26.0  69.0"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#implement the kmeans "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.array(df[['x','y','endX','endY']])\n",
    "kmeans = KMeans(n_clusters = 10,random_state=100)\n",
    "kmeans.fit(X)\n",
    "df['cluster'] = kmeans.predict(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>team</th>\n",
       "      <th>type</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>endX</th>\n",
       "      <th>endY</th>\n",
       "      <th>cluster</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>48.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>48.0</td>\n",
       "      <td>60.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>24</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>49.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>65.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>69.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>28</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>63.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>65.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>29</td>\n",
       "      <td>France</td>\n",
       "      <td>Pass</td>\n",
       "      <td>58.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>69.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   index    team  type     x     y  endX  endY  cluster\n",
       "0     11  France  Pass  48.0  50.0  48.0  60.0        0\n",
       "1     24  France  Pass  49.0  80.0  46.0  61.0        0\n",
       "2     25  France  Pass  65.0  64.0  66.0  69.0        0\n",
       "3     28  France  Pass  63.0  73.0  65.0  79.0        0\n",
       "4     29  France  Pass  58.0  79.0  26.0  69.0        0"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    41\n",
       "1    40\n",
       "6    38\n",
       "8    31\n",
       "3    31\n",
       "7    28\n",
       "2    26\n",
       "9    25\n",
       "4    18\n",
       "5    14\n",
       "Name: cluster, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.cluster.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10,10))\n",
    "fig.set_facecolor('#38383b')\n",
    "ax.patch.set_facecolor('#38383b')\n",
    "\n",
    "pitch = Pitch(pitch_type='statsbomb',orientation='horizontal',\n",
    "             pitch_color='#38383b',line_color='white',figsize=(10,10),\n",
    "             constrained_layout=False,tight_layout=True,view='full')\n",
    "\n",
    "pitch.draw(ax=ax)\n",
    "\n",
    "for x in range(len(df['cluster'])):\n",
    "    \n",
    "    if df['cluster'][x] ==0:\n",
    "        pitch.lines(xstart=df['x'][x],ystart=df['y'][x],xend=df['endX'][x],yend=df['endY'][x],\n",
    "                   color='#74c69d',lw=3,zorder=2,comet=True,ax=ax)\n",
    "        \n",
    "    if df['cluster'][x] ==5:\n",
    "        pitch.lines(xstart=df['x'][x],ystart=df['y'][x],xend=df['endX'][x],yend=df['endY'][x],\n",
    "                   color='#add8e6',lw=3,zorder=2,comet=True,ax=ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}