{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "import os; import sys; sys.path.append('../')\n",
    "import pandas as pd\n",
    "import tqdm\n",
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=pd.errors.PerformanceWarning)\n",
    "\n",
    "import socceraction.vaep as vaep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Configure file and folder names\n",
    "datafolder = \"../data\"\n",
    "spadl_h5 = os.path.join(datafolder,\"spadl-statsbomb.h5\")\n",
    "predictions_h5 = os.path.join(datafolder,\"predictions.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "nb of games: 64\n"
     ]
    }
   ],
   "source": [
    "games = pd.read_hdf(spadl_h5,\"games\")\n",
    "games = games[games.competition_name == \"FIFA World Cup\"]\n",
    "print(\"nb of games:\", len(games))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 64/64 [00:04<00:00, 15.32it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Index(['game_id', 'period_id', 'time_seconds', 'timestamp', 'team_id',\n",
       "       'player_id', 'start_x', 'start_y', 'end_x', 'end_y', 'type_id',\n",
       "       'result_id', 'bodypart_id', 'type_name', 'result_name', 'bodypart_name',\n",
       "       'player_name', 'player_nickname', 'jersey_number', 'country_id',\n",
       "       'country_name', 'extra', 'team_name', 'scores', 'concedes',\n",
       "       'offensive_value', 'defensive_value', 'vaep_value'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "players = pd.read_hdf(spadl_h5,\"players\")\n",
    "teams = pd.read_hdf(spadl_h5,\"teams\")\n",
    "actiontypes = pd.read_hdf(spadl_h5, \"actiontypes\")\n",
    "bodyparts = pd.read_hdf(spadl_h5, \"bodyparts\")\n",
    "results = pd.read_hdf(spadl_h5, \"results\")\n",
    "\n",
    "A = []\n",
    "for game in tqdm.tqdm(list(games.itertuples())):\n",
    "    actions = pd.read_hdf(spadl_h5,f\"actions/game_{game.game_id}\")\n",
    "    actions = (\n",
    "        actions.merge(actiontypes)\n",
    "        .merge(results)\n",
    "        .merge(bodyparts)\n",
    "        .merge(players,\"left\",on=\"player_id\")\n",
    "        .merge(teams,\"left\",on=\"team_id\")\n",
    "        .sort_values([\"period_id\", \"time_seconds\", \"timestamp\"])\n",
    "        .reset_index(drop=True)\n",
    "    )\n",
    "    preds = pd.read_hdf(predictions_h5,f\"game_{game.game_id}\")\n",
    "    values = vaep.value(actions,preds.scores,preds.concedes)\n",
    "    A.append(pd.concat([actions,preds,values],axis=1))\n",
    "A = pd.concat(A).sort_values([\"game_id\",\"period_id\", \"time_seconds\", \"timestamp\"]).reset_index(drop=True)\n",
    "A.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### (optional) inspect Belgium's top 10 most valuable non-shot actions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-06-28 England 0-1 Belgium 9' dribble Michy Batshuayi Tunga\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-06-28 England 0-1 Belgium 8' pass Marouane Fellaini-Bakkioui\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-07-02 Belgium 3-2 Japan 77' cross Nacer Chadli\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-07-02 Belgium 3-2 Japan 24' cross Dries Mertens\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-07-02 Belgium 3-2 Japan 73' cross Eden Hazard\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-06-23 Belgium 5-2 Tunisia 80' cross Kevin De Bruyne\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-07-10 France 1-0 Belgium 21' corner_crossed Nacer Chadli\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-06-23 Belgium 5-2 Tunisia 89' cross Youri Tielemans\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-06-23 Belgium 5-2 Tunisia 47' pass Thomas Meunier\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2018-06-23 Belgium 5-2 Tunisia 91' dribble Michy Batshuayi Tunga\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotsoccer\n",
    "\n",
    "sorted_A = A.sort_values(\"vaep_value\", ascending=False)\n",
    "sorted_A = sorted_A[sorted_A.team_name == \"Belgium\"] # view only actions from Belgium\n",
    "sorted_A = sorted_A[~sorted_A.type_name.str.contains(\"shot\")] #eliminate shots\n",
    "\n",
    "def get_time(period_id,time_seconds):\n",
    "    m = int((period_id-1)*45 + time_seconds // 60)\n",
    "    s = time_seconds % 60\n",
    "    if s == int(s):\n",
    "        s = int(s)\n",
    "    return f\"{m}m{s}s\"\n",
    "\n",
    "for j in range(0,10):\n",
    "    row = list(sorted_A[j:j+1].itertuples())[0]\n",
    "    i = row.Index\n",
    "    a = A[i - 3 : i+2].copy()\n",
    "    \n",
    "    a[\"player\"] = a[[\"player_nickname\",\"player_name\"]].apply(lambda x: x[0] if x[0] else x[1],axis=1)\n",
    "    \n",
    "    g = list(games[games.game_id == a.game_id.values[0]].itertuples())[0]\n",
    "    game_info = f\"{g.match_date} {g.home_team_name} {g.home_score}-{g.away_score} {g.away_team_name}\"\n",
    "    minute = int((row.period_id-1)*45 + row.time_seconds // 60)\n",
    "    print(f\"{game_info} {minute}' {row.type_name} {row.player_name}\")\n",
    "\n",
    "    a[\"scores\"] = a.scores.apply(lambda x : \"%.3f\" % x )\n",
    "    a[\"vaep_value\"] = a.vaep_value.apply(lambda x : \"%.3f\" % x )\n",
    "    a[\"time\"] = a[[\"period_id\",\"time_seconds\"]].apply(lambda x: get_time(*x),axis=1)\n",
    "    cols = [\"time\",\"type_name\",\"player\",\"team_name\",\"scores\",\"vaep_value\"]\n",
    "    matplotsoccer.actions(a[[\"start_x\",\"start_y\",\"end_x\",\"end_y\"]],\n",
    "                a.type_name,\n",
    "                team=a.team_name,\n",
    "                result = a.result_name == \"success\",\n",
    "                label=a[cols],\n",
    "                labeltitle = cols,\n",
    "                zoom=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "soccer_rating_pressing-3.6.2",
   "language": "python",
   "name": "soccer_rating_pressing-3.6.2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.2"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}