014_Tensor_Sim.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from lib.data_utils import *\n",
    "from lib.model_utils import *\n",
    "from tqdm import tqdm\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import time\n",
    "from torch.nn.functional import one_hot"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load data & models"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train, df_train_y, df_optimization, df_optimization_y, df_test, df_test_y, complete_feature_set, features_model = load_model_training_data_template(train_sets = ['data/wyscout/csv/events/Italy.csv', 'data/wyscout/csv/events/Germany.csv', 'data/wyscout/csv/events/France.csv'], optimization_sets = ['data/wyscout/csv/events/hack_train.csv',], test_sets = ['data/wyscout/csv/events/Spain.csv', 'data/wyscout/csv/events/England.csv'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['subtype_id_1', 'subtype_id_2', 'subtype_id_3', 'subtype_id_4', 'subtype_id_5', 'subtype_id_6', 'subtype_id_7', 'subtype_id_8', 'subtype_id_9', 'subtype_id_10', 'subtype_id_11', 'subtype_id_12', 'subtype_id_13', 'subtype_id_14', 'subtype_id_15', 'subtype_id_16', 'subtype_id_17', 'subtype_id_18', 'subtype_id_19', 'subtype_id_20', 'subtype_id_21', 'subtype_id_22', 'subtype_id_23', 'subtype_id_24', 'subtype_id_25', 'subtype_id_26', 'subtype_id_27', 'subtype_id_28', 'subtype_id_29', 'subtype_id_30', 'subtype_id_31', 'subtype_id_32', 'subtype_id_33', 'period', 'minute', 'x', 'y', 'is_home_team', 'accurate', 'goal', 'home_score', 'away_score', 'next_action_x', 'next_action_y', 'next_action_type_1', 'next_action_type_2', 'next_action_type_3', 'next_action_type_4', 'next_action_type_5', 'next_action_type_6', 'next_action_type_7', 'next_action_type_8', 'next_action_type_9', 'next_action_type_10', 'next_action_type_11', 'next_action_type_12', 'next_action_type_13', 'next_action_type_14', 'next_action_type_15', 'next_action_type_16', 'next_action_type_17', 'next_action_type_18', 'next_action_type_19', 'next_action_type_20', 'next_action_type_21', 'next_action_type_22', 'next_action_type_23', 'next_action_type_24', 'next_action_type_25', 'next_action_type_26', 'next_action_type_27', 'next_action_type_28', 'next_action_type_29', 'next_action_type_30', 'next_action_type_31', 'next_action_type_32', 'next_action_type_33', 'next_action_accurate', 'next_action_goal']\n"
     ]
    }
   ],
   "source": [
    "print(complete_feature_set)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Simulation class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from lib.simulator import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\jjate\\OneDrive\\Desktop\\LargeEventsModel\\lib\\simulator.py:14: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
      "  self.model_type = torch.load(model_type_path).to(self.device)\n",
      "c:\\Users\\jjate\\OneDrive\\Desktop\\LargeEventsModel\\lib\\simulator.py:15: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
      "  self.model_acc = torch.load(model_acc_path).to(self.device)\n",
      "c:\\Users\\jjate\\OneDrive\\Desktop\\LargeEventsModel\\lib\\simulator.py:16: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
      "  self.model_data = torch.load(model_data_path).to(self.device)\n"
     ]
    }
   ],
   "source": [
    "sim = Simulator(\n",
    "    model_type_path='models/lem/LEMv3_MODEL_TYPE_TORCH.pth',\n",
    "    model_acc_path='models/lem/LEMv4_MODEL_ACC_TORCH.pth',\n",
    "    model_data_path='models/lem/LEMv3_MODEL_DATA_TORCH.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 91%|█████████ | 1822/2000 [00:51<00:04, 35.65it/s]\n"
     ]
    }
   ],
   "source": [
    "feature_tensor, all_sims_data = sim.simulate([1 if i == 27 else 0 for i in range(33)] + [0, 0, 0.5, 0.5, 1, 1, 0, 0, 0], store_full_sim=True, n_sims=10000)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Post-simulation analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>period</th>\n",
       "      <th>minute</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>is_home_team</th>\n",
       "      <th>accurate</th>\n",
       "      <th>goal</th>\n",
       "      <th>home_score</th>\n",
       "      <th>away_score</th>\n",
       "      <th>type_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.256667</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.049722</td>\n",
       "      <td>0.79</td>\n",
       "      <td>0.24</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.175833</td>\n",
       "      <td>0.72</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.403333</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.31</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.146944</td>\n",
       "      <td>0.76</td>\n",
       "      <td>0.41</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9995</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.210278</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9996</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.189167</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.33</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9997</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.308333</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.76</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9998</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.157778</td>\n",
       "      <td>0.68</td>\n",
       "      <td>0.74</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9999</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.159167</td>\n",
       "      <td>0.74</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10000 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      period    minute     x     y  is_home_team  accurate  goal  home_score  \\\n",
       "0        2.0  0.256667  0.59  0.57           0.0       0.0   0.0           1   \n",
       "1        2.0  0.049722  0.79  0.24           1.0       0.0   0.0           1   \n",
       "2        2.0  0.175833  0.72  0.81           0.0       1.0   0.0           0   \n",
       "3        2.0  0.403333  0.93  0.31           1.0       0.0   0.0           2   \n",
       "4        2.0  0.146944  0.76  0.41           0.0       0.0   0.0           1   \n",
       "...      ...       ...   ...   ...           ...       ...   ...         ...   \n",
       "9995     2.0  0.210278  0.00  1.00           1.0       1.0   0.0           0   \n",
       "9996     2.0  0.189167  0.35  0.33           1.0       1.0   0.0           0   \n",
       "9997     2.0  0.308333  0.22  0.76           1.0       1.0   0.0           1   \n",
       "9998     2.0  0.157778  0.68  0.74           1.0       1.0   0.0           4   \n",
       "9999     2.0  0.159167  0.74  0.50           0.0       1.0   0.0           3   \n",
       "\n",
       "      away_score  type_id  \n",
       "0              0       28  \n",
       "1              1       23  \n",
       "2              1        1  \n",
       "3              2        1  \n",
       "4              1        2  \n",
       "...          ...      ...  \n",
       "9995           2       31  \n",
       "9996           1        1  \n",
       "9997           2        4  \n",
       "9998           0       28  \n",
       "9999           0       28  \n",
       "\n",
       "[10000 rows x 10 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.DataFrame(feature_tensor)\n",
    "data['type_id'] = data[data.columns[:33]].idxmax(axis=1) + 1\n",
    "data = data.drop(data.columns[:33], axis=1)\n",
    "data.columns = ['period', 'minute', 'x', 'y', 'is_home_team', 'accurate', 'goal', 'home_score', 'away_score', 'type_id']\n",
    "data.home_score = (data.home_score * 10).astype(int)\n",
    "data.away_score = (data.away_score * 10).astype(int)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "period\n",
       "2.0    10000\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.period.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x1500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize'] = (9, 15)  # Adjusted height for 3 plots\n",
    "\n",
    "# Simulated data setup for the pivot_table and bar plots\n",
    "\n",
    "# First plot: Distribution of Results in Simulated Matches\n",
    "plt.subplot(3, 1, 1)  # First of 3 rows\n",
    "plt.title('Distribution of Results in Simulated Matches')\n",
    "plt.imshow(data.pivot_table(index='home_score', columns='away_score', values='x', aggfunc='count'), cmap='hot_r')\n",
    "plt.ylim(8.5, -0.5)\n",
    "plt.xlim(-0.5, 8.5)\n",
    "plt.ylabel('Home Goals')\n",
    "plt.xlabel('Away Goals')\n",
    "plt.gca().set_aspect('equal', adjustable='box')\n",
    "\n",
    "# Second plot: Distribution of Home Goals\n",
    "df_results = pd.DataFrame(feature_tensor[:, 40:42] * 10, columns=['HG', 'AG'])\n",
    "plt.subplot(3, 1, 2)  # Second of 3 rows\n",
    "plt.bar(df_results.groupby('HG').HG.count().index, \n",
    "        df_results.groupby('HG').HG.count() / df_results.groupby('HG').HG.count().sum(), \n",
    "        alpha=0.5, label=f'Simulation (Avg: {df_results[\"HG\"].mean():.2f})')\n",
    "plt.legend()\n",
    "plt.title('Distribution of Home Goals')\n",
    "plt.xlabel('Home Goals')\n",
    "plt.ylabel('Frequency')\n",
    "plt.xlim(-0.5, 8.5)\n",
    "plt.ylim(0, 0.5)\n",
    "plt.gca().set_aspect(20, adjustable='box')\n",
    "\n",
    "# Third plot: Distribution of Away Goals\n",
    "plt.subplot(3, 1, 3)  # Third of 3 rows\n",
    "plt.bar(df_results.groupby('AG').AG.count().index, \n",
    "        df_results.groupby('AG').AG.count() / df_results.groupby('AG').AG.count().sum(), \n",
    "        alpha=0.5, label=f'Simulation (Avg: {df_results[\"AG\"].mean():.2f})')\n",
    "plt.legend()\n",
    "plt.xlim(-0.5, 8.5)\n",
    "plt.ylim(0, 0.5)\n",
    "plt.title('Distribution of Away Goals')\n",
    "plt.xlabel('Away Goals')\n",
    "plt.ylabel('Frequency')\n",
    "plt.gca().set_aspect(20, adjustable='box')\n",
    "\n",
    "# Adjust layout\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x1500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_results = pd.DataFrame(feature_tensor[:, 40:42] * 10, columns=['HG', 'AG'])\n",
    "plt.subplot(121)\n",
    "plt.bar(df_results.groupby('HG').HG.count().index, df_results.groupby('HG').HG.count()/df_results.groupby('HG').HG.count().max(), alpha=0.5, label=f'Simulation (Avg: {df_results[\"HG\"].mean():.2f})')\n",
    "plt.legend();\n",
    "plt.subplot(122)\n",
    "plt.bar(df_results.groupby('AG').AG.count().index, df_results.groupby('AG').AG.count()/df_results.groupby('AG').AG.count().max(), alpha=0.5, label=f'Simulation (Avg: {df_results[\"AG\"].mean():.2f})')\n",
    "plt.legend();\n",
    "\n",
    "# plt.savefig('figures/paper_lem/distribution_scores.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = pd.DataFrame(torch.concat(all_sims_data))\n",
    "data[data.columns[0:33]].sum().plot.bar(figsize=(20, 10));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_all_data = []\n",
    "for i in range(len(all_sims_data)):\n",
    "    data = pd.DataFrame(all_sims_data[i])\n",
    "    data['type_id'] = data[data.columns[:33]].idxmax(axis=1) + 1\n",
    "    data = data.drop(data.columns[:33], axis=1)\n",
    "    data.columns = ['period', 'minute', 'x', 'y', 'is_home_team', 'accurate', 'goal', 'home_score', 'away_score', 'type_id']\n",
    "    data['event_id'] = i\n",
    "    data['game_id'] = [i for i in range(len(all_sims_data[i]))]\n",
    "    df_all_data.append(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_all_data = pd.concat(df_all_data)\n",
    "df_all_data.minute *= 60\n",
    "df_all_data.x *= 100\n",
    "df_all_data.y *= 100\n",
    "df_all_data.home_score = (df_all_data.home_score * 10).astype(int)\n",
    "df_all_data.away_score = (df_all_data.away_score * 10).astype(int)\n",
    "df_all_data['type_name'] = df_all_data.type_id.map({v: k for k, v in SUBEVENT_TYPE_MAP.items()})\n",
    "df_all_data = df_all_data.reset_index(drop=True)"
   ]
  },
  {
   "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>period</th>\n",
       "      <th>minute</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>is_home_team</th>\n",
       "      <th>accurate</th>\n",
       "      <th>goal</th>\n",
       "      <th>home_score</th>\n",
       "      <th>away_score</th>\n",
       "      <th>type_id</th>\n",
       "      <th>event_id</th>\n",
       "      <th>game_id</th>\n",
       "      <th>type_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12882614</th>\n",
       "      <td>1.0</td>\n",
       "      <td>36.999962</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>28</td>\n",
       "      <td>1288</td>\n",
       "      <td>2614</td>\n",
       "      <td>simple_pass</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15462159</th>\n",
       "      <td>1.0</td>\n",
       "      <td>43.816635</td>\n",
       "      <td>88.000000</td>\n",
       "      <td>66.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>23</td>\n",
       "      <td>1546</td>\n",
       "      <td>2159</td>\n",
       "      <td>cross</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8391026</th>\n",
       "      <td>1.0</td>\n",
       "      <td>6.000001</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>56.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>839</td>\n",
       "      <td>1026</td>\n",
       "      <td>shot</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8301108</th>\n",
       "      <td>0.0</td>\n",
       "      <td>42.883293</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>58.999996</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>22</td>\n",
       "      <td>830</td>\n",
       "      <td>1108</td>\n",
       "      <td>touch</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12423044</th>\n",
       "      <td>1.0</td>\n",
       "      <td>29.383339</td>\n",
       "      <td>76.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>23</td>\n",
       "      <td>1242</td>\n",
       "      <td>3044</td>\n",
       "      <td>cross</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2193160</th>\n",
       "      <td>0.0</td>\n",
       "      <td>10.183331</td>\n",
       "      <td>46.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>28</td>\n",
       "      <td>219</td>\n",
       "      <td>3160</td>\n",
       "      <td>simple_pass</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20241</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>88.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>241</td>\n",
       "      <td>ground_attacking_duel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16816457</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.816667</td>\n",
       "      <td>60.000004</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "      <td>1681</td>\n",
       "      <td>6457</td>\n",
       "      <td>simple_pass</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1252376</th>\n",
       "      <td>0.0</td>\n",
       "      <td>6.950000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>92.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "      <td>125</td>\n",
       "      <td>2376</td>\n",
       "      <td>simple_pass</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1740923</th>\n",
       "      <td>0.0</td>\n",
       "      <td>10.483331</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>28</td>\n",
       "      <td>174</td>\n",
       "      <td>923</td>\n",
       "      <td>simple_pass</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          period     minute          x          y  is_home_team  accurate  \\\n",
       "12882614     1.0  36.999962  75.000000   7.000000           1.0       1.0   \n",
       "15462159     1.0  43.816635  88.000000  66.000000           1.0       0.0   \n",
       "8391026      1.0   6.000001  89.000000  56.000000           0.0       1.0   \n",
       "8301108      0.0  42.883293  13.000000  58.999996           1.0       0.0   \n",
       "12423044     1.0  29.383339  76.000000   9.000000           0.0       0.0   \n",
       "2193160      0.0  10.183331  46.000000  25.000000           0.0       1.0   \n",
       "20241        0.0   0.100000  81.000000  88.000000           1.0       1.0   \n",
       "16816457     2.0   3.816667  60.000004  29.000000           0.0       1.0   \n",
       "1252376      0.0   6.950000  40.000000  92.000000           1.0       1.0   \n",
       "1740923      0.0  10.483331  32.000000   9.000000           0.0       1.0   \n",
       "\n",
       "          goal  home_score  away_score  type_id  event_id  game_id  \\\n",
       "12882614   0.0           1           1       28      1288     2614   \n",
       "15462159   0.0           1           0       23      1546     2159   \n",
       "8391026    0.0           1           1       33       839     1026   \n",
       "8301108    0.0           2           0       22       830     1108   \n",
       "12423044   0.0           0           3       23      1242     3044   \n",
       "2193160    0.0           0           1       28       219     3160   \n",
       "20241      0.0           0           0        2         2      241   \n",
       "16816457   0.0           3           0       28      1681     6457   \n",
       "1252376    0.0           0           0       28       125     2376   \n",
       "1740923    0.0           0           0       28       174      923   \n",
       "\n",
       "                      type_name  \n",
       "12882614            simple_pass  \n",
       "15462159                  cross  \n",
       "8391026                    shot  \n",
       "8301108                   touch  \n",
       "12423044                  cross  \n",
       "2193160             simple_pass  \n",
       "20241     ground_attacking_duel  \n",
       "16816457            simple_pass  \n",
       "1252376             simple_pass  \n",
       "1740923             simple_pass  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_all_data.sample(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1574.553, 1612.232645403377)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_all_data[df_all_data.period <= 1].groupby('game_id').minute.count().mean(), df_train.groupby('match_id').minute.count().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "diffs = []\n",
    "for game_id in df_all_data.game_id.unique():\n",
    "    diffs.append(df_all_data[df_all_data.game_id == game_id].minute.diff().dropna().values)\n",
    "diffs = np.concatenate(diffs)\n",
    "diffs = diffs[diffs > -10]\n",
    "plt.hist(diffs, bins=100);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1832000, 6125.0, 3091.0, 1359.0000915527344, 1286.0000610351562)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_all_data), df_all_data.goal.sum(), df_all_data[df_all_data.type_name.isin(['shot', 'penalty', 'free_kick_shot'])].goal.sum(), data.home_score.sum()*10, data.away_score.sum()*10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[26782, 6125.0, 2821.0, 206.0, 0.0, 0.014618995633187772, 0.10533193936225824]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[\n",
    "    len(df_all_data[df_all_data.type_name == 'shot']), \n",
    "    df_all_data.goal.sum(), \n",
    "    df_all_data[df_all_data.type_name == 'shot'].goal.sum(), \n",
    "    df_all_data[df_all_data.type_name == 'penalty'].goal.sum(), \n",
    "    df_all_data[df_all_data.type_name == 'direct_free_kick'].goal.sum(), \n",
    "    len(df_all_data[df_all_data.type_name == 'shot']) / len(df_all_data), \n",
    "    df_all_data[df_all_data.type_name == 'shot'].goal.sum()/len(df_all_data[df_all_data.type_name == 'shot']),\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>goal</th>\n",
       "      <th>goal/shot</th>\n",
       "      <th>shot</th>\n",
       "      <th>simple_pass</th>\n",
       "      <th>ground_attacking_duel</th>\n",
       "      <th>ground_defending_duel</th>\n",
       "      <th>touch</th>\n",
       "      <th>air_duel</th>\n",
       "      <th>high_pass</th>\n",
       "      <th>throw_in</th>\n",
       "      <th>cross</th>\n",
       "      <th>free_kick</th>\n",
       "      <th>foul</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.334</td>\n",
       "      <td>10.533</td>\n",
       "      <td>1.46</td>\n",
       "      <td>42.8</td>\n",
       "      <td>8.1</td>\n",
       "      <td>8.3</td>\n",
       "      <td>5.2</td>\n",
       "      <td>6.2</td>\n",
       "      <td>4.1</td>\n",
       "      <td>2.5</td>\n",
       "      <td>2.1</td>\n",
       "      <td>1.4</td>\n",
       "      <td>1.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.333</td>\n",
       "      <td>10.291</td>\n",
       "      <td>1.40</td>\n",
       "      <td>41.4</td>\n",
       "      <td>9.1</td>\n",
       "      <td>9.0</td>\n",
       "      <td>5.7</td>\n",
       "      <td>5.2</td>\n",
       "      <td>4.2</td>\n",
       "      <td>2.8</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>1.6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    goal  goal/shot  shot  simple_pass  ground_attacking_duel  \\\n",
       "0  0.334     10.533  1.46         42.8                    8.1   \n",
       "1  0.333     10.291  1.40         41.4                    9.1   \n",
       "\n",
       "   ground_defending_duel  touch  air_duel  high_pass  throw_in  cross  \\\n",
       "0                    8.3    5.2       6.2        4.1       2.5    2.1   \n",
       "1                    9.0    5.7       5.2        4.2       2.8    2.0   \n",
       "\n",
       "   free_kick  foul  \n",
       "0        1.4   1.4  \n",
       "1        1.5   1.6  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame([{\n",
    "    'goal': round(df_all_data['goal'].sum() / len(df_all_data) * 100,3),\n",
    "    'goal/shot': round(df_all_data[df_all_data.type_name == 'shot'].goal.sum() / len(df_all_data[df_all_data.type_name == 'shot']) * 100,3),\n",
    "    'shot': round(len(df_all_data[df_all_data.type_name == 'shot']) / len(df_all_data) * 100,2), \n",
    "    'simple_pass': round(len(df_all_data[df_all_data.type_name == 'simple_pass']) / len(df_all_data) * 100,1), \n",
    "    'ground_attacking_duel': round(len(df_all_data[df_all_data.type_name == 'ground_attacking_duel']) / len(df_all_data) * 100,1), \n",
    "    'ground_defending_duel': round(len(df_all_data[df_all_data.type_name == 'ground_defending_duel']) / len(df_all_data) * 100,1), \n",
    "    'touch': round(len(df_all_data[df_all_data.type_name == 'touch']) / len(df_all_data) * 100,1), \n",
    "    'air_duel': round(len(df_all_data[df_all_data.type_name == 'air_duel']) / len(df_all_data) * 100,1), \n",
    "    'high_pass': round(len(df_all_data[df_all_data.type_name == 'high_pass']) / len(df_all_data) * 100,1), \n",
    "    'throw_in': round(len(df_all_data[df_all_data.type_name == 'throw_in']) / len(df_all_data) * 100,1), \n",
    "    'cross': round(len(df_all_data[df_all_data.type_name == 'cross']) / len(df_all_data) * 100,1), \n",
    "    'free_kick': round(len(df_all_data[df_all_data.type_name == 'free_kick']) / len(df_all_data) * 100,1), \n",
    "    'foul': round(len(df_all_data[df_all_data.type_name == 'foul']) / len(df_all_data) * 100,1), \n",
    "},\n",
    "{\n",
    "    'goal': round(df_train['goal'].sum() / len(df_train) * 100,3),\n",
    "    'goal/shot': round(df_train[df_train.subtype_name == 'shot'].goal.sum() / len(df_train[df_train.subtype_name == 'shot']) * 100,3),\n",
    "    'shot': round(len(df_train[df_train.subtype_name == 'shot']) / len(df_train) * 100,2), \n",
    "    'simple_pass': round(len(df_train[df_train.subtype_name == 'simple_pass']) / len(df_train) * 100,1), \n",
    "    'ground_attacking_duel': round(len(df_train[df_train.subtype_name == 'ground_attacking_duel']) / len(df_train) * 100,1), \n",
    "    'ground_defending_duel': round(len(df_train[df_train.subtype_name == 'ground_defending_duel']) / len(df_train) * 100,1), \n",
    "    'touch': round(len(df_train[df_train.subtype_name == 'touch']) / len(df_train) * 100,1), \n",
    "    'air_duel': round(len(df_train[df_train.subtype_name == 'air_duel']) / len(df_train) * 100,1), \n",
    "    'high_pass': round(len(df_train[df_train.subtype_name == 'high_pass']) / len(df_train) * 100,1), \n",
    "    'throw_in': round(len(df_train[df_train.subtype_name == 'throw_in']) / len(df_train) * 100,1), \n",
    "    'cross': round(len(df_train[df_train.subtype_name == 'cross']) / len(df_train) * 100,1), \n",
    "    'free_kick': round(len(df_train[df_train.subtype_name == 'free_kick']) / len(df_train) * 100,1), \n",
    "    'foul': round(len(df_train[df_train.subtype_name == 'foul']) / len(df_train) * 100,1), \n",
    "}])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "goal  type_name              \n",
       "0.0   simple_pass                784870\n",
       "      ground_defending_duel      152692\n",
       "      ground_attacking_duel      148491\n",
       "      air_duel                   112901\n",
       "      touch                       95073\n",
       "      ground_loose_ball_duel      90964\n",
       "      high_pass                   75596\n",
       "      head_pass                   54464\n",
       "      throw_in                    45610\n",
       "      cross                       38416\n",
       "      clearance                   31803\n",
       "      foul                        26541\n",
       "      free_kick                   25933\n",
       "      shot                        23961\n",
       "      launch                      23424\n",
       "      goal_kick                   18197\n",
       "      smart_pass                  16384\n",
       "      acceleration                15629\n",
       "      corner                      13428\n",
       "      hand_pass                    9688\n",
       "      free_kick_cross              5145\n",
       "      save_attempt                 4925\n",
       "      reflexes                     4550\n",
       "      goalkeeper_leaving_line      4286\n",
       "1.0   shot                         2821\n",
       "      reflexes                     2406\n",
       "0.0   free_kick_shot               1159\n",
       "      hand_foul                    1083\n",
       "1.0   save_attempt                  606\n",
       "      penalty                       206\n",
       "0.0   out_of_game_foul              189\n",
       "      protest                       166\n",
       "      late_card_foul                132\n",
       "      penalty                        89\n",
       "1.0   free_kick_shot                 64\n",
       "0.0   time_lost_foul                 47\n",
       "      simulation                     35\n",
       "1.0   clearance                      10\n",
       "0.0   violent_foul                    4\n",
       "1.0   ground_defending_duel           4\n",
       "      cross                           3\n",
       "      launch                          3\n",
       "      smart_pass                      1\n",
       "      touch                           1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_all_data[['goal', 'type_name']].value_counts()"
   ]
  },
  {
   "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": "3.8.19"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}