dota2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy import stats\n",
    "import math\n",
    "from sklearn import preprocessing\n",
    "from sklearn.utils import resample\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.feature_selection import VarianceThreshold\n",
    "from sklearn.feature_selection import SelectKBest\n",
    "from sklearn.feature_selection import chi2\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Učitavanje podataka i imenovanje kolona\n",
    "U datasetu ne postoje labele za kolone, pa je potrebno ručno uneti. Pored kolona za pobedu, region, mod i tip igre, postoji još 113 kolona, koje predstavljaju po jednog dota2 heroja, i one su numerisane 0-112."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "col_names = ['win', 'region', 'mode', 'type']\n",
    "num_arr_str = np.arange(113).astype(str)\n",
    "all_names = np.concatenate((col_names, num_arr_str), axis = 0)\n",
    "\n",
    "dota_data = pd.read_csv(\"dota2.csv\", header=None, names=all_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>win</th>\n",
       "      <th>region</th>\n",
       "      <th>mode</th>\n",
       "      <th>type</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>...</th>\n",
       "      <th>103</th>\n",
       "      <th>104</th>\n",
       "      <th>105</th>\n",
       "      <th>106</th>\n",
       "      <th>107</th>\n",
       "      <th>108</th>\n",
       "      <th>109</th>\n",
       "      <th>110</th>\n",
       "      <th>111</th>\n",
       "      <th>112</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1</td>\n",
       "      <td>223</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>227</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1</td>\n",
       "      <td>136</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>227</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>184</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 117 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   win  region  mode  type  0  1  2  3  4  5  ...  103  104  105  106  107  \\\n",
       "0   -1     223     8     2  0 -1  0  0  0  0  ...   -1    0    0    0    0   \n",
       "1    1     227     8     2  0  0  0  0  0  0  ...   -1    0    0    0    0   \n",
       "2   -1     136     2     2  1  0  0  0 -1  0  ...    0    0    0    0    0   \n",
       "3    1     227     2     2 -1  0  0  0  0  0  ...    0    0    0    0    0   \n",
       "4    1     184     2     3  0  0  0 -1  0  0  ...    0    0    0    0    0   \n",
       "\n",
       "   108  109  110  111  112  \n",
       "0    0    0    0    0    0  \n",
       "1    0    0    0    0    0  \n",
       "2    0    0    0    0    0  \n",
       "3    0    0    0    0    0  \n",
       "4    0    0    0    0    0  \n",
       "\n",
       "[5 rows x 117 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dota_data.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predstavljanje podataka pomoću grafikona"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Naredni grafikon pokazuje zastupljenost regiona u mečevima"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{227: 8781, 154: 7751, 156: 7668, 151: 7612, 153: 7513, 152: 7460, 155: 7325, 224: 6625, 231: 4040, 223: 3892, 225: 2947, 232: 1880, 204: 1646, 182: 1519, 187: 1408, 145: 1374, 144: 1348, 188: 1335, 181: 1315, 171: 1224, 186: 1204, 184: 1096, 185: 1080, 183: 1045, 191: 1038, 111: 918, 192: 913, 112: 901, 132: 845, 133: 843, 131: 820, 121: 788, 138: 775, 123: 748, 134: 737, 122: 736, 124: 723, 136: 688, 135: 688, 137: 687, 261: 300, 161: 271, 251: 168, 241: 142, 213: 97, 211: 19, 212: 11}\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "regi = dota_data.iloc[:,1]\n",
    "abc = dict(regi.value_counts())\n",
    "\n",
    "x = np.fromiter(abc.keys(), dtype=int)\n",
    "y = np.fromiter(abc.values(), dtype=int)\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.bar(x, y, align='center', color='b')\n",
    "\n",
    "ax.set_title('Zastupljenost regiona')\n",
    "ax.set_ylabel('Broj pojavljivanja')\n",
    "ax.set_xlabel('ID regiona')\n",
    "print(abc)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Naredni grafikon pokazuje zastupljenost tipa igre u mečevima"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1: Practice 2: Tournament 3: Tutorial\n",
      "{2: 63272, 3: 39652, 1: 20}\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "type_match = dota_data.iloc[:,3]\n",
    "abc = dict(type_match.value_counts())\n",
    "\n",
    "x = np.fromiter(abc.keys(), dtype=int)\n",
    "y = np.fromiter(abc.values(), dtype=int)\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.bar(x, y, align='center', color='r')\n",
    "\n",
    "ax.set_title('Zastupljenost tipa')\n",
    "ax.set_ylabel('Broj pojavljivanja')\n",
    "ax.set_xlabel('ID tipa')\n",
    "\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels(x)\n",
    "\n",
    "print(\"1: Practice\", \"2: Tournament\", \"3: Tutorial\")\n",
    "print(abc)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Provera tipova podataka i postojanje null vrednosti\n",
    "Iz naredne linije možemo videti da je dataset većih dimenzija, pa je to uzeto u obzir prilikom treniranja."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(102944, 117)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dota_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Brisanje duplikata\n",
    "dota_data = dota_data.drop_duplicates()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Izvršene su provere za NULL podatke i da li su svi podaci brojevne vrednosti."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "win       0\n",
       "region    0\n",
       "mode      0\n",
       "type      0\n",
       "0         0\n",
       "         ..\n",
       "108       0\n",
       "109       0\n",
       "110       0\n",
       "111       0\n",
       "112       0\n",
       "Length: 117, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Broj null podataka u svakoj koloni\n",
    "dota_data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "win       int64\n",
       "region    int64\n",
       "mode      int64\n",
       "type      int64\n",
       "0         int64\n",
       "          ...  \n",
       "108       int64\n",
       "109       int64\n",
       "110       int64\n",
       "111       int64\n",
       "112       int64\n",
       "Length: 117, dtype: object"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#provera tipova podataka\n",
    "dota_data.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocesiranje podataka\n",
    "Bilo je potrebno grupisani podatke u odnosu na region, kako bi se dalo na značaju.\n",
    "Takođe, izbačeni su i tipovi i modovi igre koji se odnose na Tutorial, jer nemaju značaj i predstavljaju trivijalne mečeve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    6\n",
       "1    6\n",
       "2    3\n",
       "3    6\n",
       "4    8\n",
       "Name: region, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Grupisanje regiona\n",
    "dota_data.loc[dota_data[\"region\"].isin([111,112,113,114]), \"region\"] = 1 #US West\n",
    "dota_data.loc[dota_data[\"region\"].isin([121,122,123,124]), \"region\"] = 2 #US East\n",
    "dota_data.loc[dota_data[\"region\"].isin([131,132,133,134,135,136,137,138]), \"region\"] = 3 #Europe West\n",
    "dota_data.loc[dota_data[\"region\"].isin([142,143,144,145]), \"region\"] = 4 #South Korea\n",
    "dota_data.loc[dota_data[\"region\"].isin([151,152,153,154,155,156]), \"region\"] = 5 #Southeast Asia\n",
    "dota_data.loc[dota_data[\"region\"].isin([161,163, 221,222,223,224,225,227,231,232]), \"region\"] = 6 #China\n",
    "dota_data.loc[dota_data[\"region\"].isin([171]), \"region\"] = 7 #Australia\n",
    "dota_data.loc[dota_data[\"region\"].isin([181,182,183,184, 185,186,187,188]), \"region\"] = 8 #Russia\n",
    "dota_data.loc[dota_data[\"region\"].isin([191,192]), \"region\"] = 9 #Europe East\n",
    "dota_data.loc[dota_data[\"region\"].isin([200,202,203,204]), \"region\"] = 10 #South America\n",
    "dota_data.loc[dota_data[\"region\"].isin([211,212,213]), \"region\"] = 11 #South Africa\n",
    "dota_data.loc[dota_data[\"region\"].isin([241,242]), \"region\"] = 12 #Chile\n",
    "dota_data.loc[dota_data[\"region\"].isin([251]), \"region\"] = 13 #Peru\n",
    "dota_data.loc[dota_data[\"region\"].isin([261]), \"region\"] = 14 #India\n",
    "dota_data['region'].head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Uklanjanje instanci ciji je tip igre 'Tutorial' ili 'Practice'\n",
    "df_f1 = dota_data[(dota_data['type'] != 3) & (dota_data['type'] != 1)]\n",
    "\n",
    "#Uklanjanje instanci ciji je mod igre 'Tutorial' \n",
    "df_f2 = df_f1[df_f1['mode'] != 10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    33388\n",
       "-1    29884\n",
       "Name: win, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_f2['win'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Balansiranje pobeda tima 1 i tima 2 za bolje rezultate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-1    29884\n",
       " 1    29884\n",
       "Name: win, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#balansiranje dataseta\n",
    "g = df_f2.groupby('win')\n",
    "g = pd.DataFrame(g.apply(lambda x: x.sample(g.size().min()).reset_index(drop=True)))\n",
    "g['win'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predstavljanje preprocesiranih podataka pomoću grafikona"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Naredni grafikoni pokazuju učestalost korišćenja heroja u zasebnim mečevima"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DescribeResult(nobs=113, minmax=(0, 20679), mean=5289.203539823009, variance=16079401.556415929, skewness=1.5729488508509197, kurtosis=2.4452506007096204)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x200 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x200 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "heroes = g.iloc[:,4:]\n",
    "len_matches = len(heroes);\n",
    "num_arr = np.arange(113)\n",
    "\n",
    "heroes_counts=[]\n",
    "pos=[]\n",
    "neg=[]\n",
    "\n",
    "for x in num_arr_str:\n",
    "    hero = dict(heroes[x].value_counts())\n",
    "    if hero[0]!=len_matches:\n",
    "        pos.append(hero[1])\n",
    "        neg.append(hero[-1])\n",
    "    else:\n",
    "        pos.append(0)\n",
    "        neg.append(0)\n",
    "    heroes_counts.append(hero)\n",
    "        \n",
    "sum = np.add(pos, neg)\n",
    "\n",
    "print(stats.describe(sum)) \n",
    "\n",
    "plt.rcParams['figure.figsize'] = [6, 2]\n",
    "plt.rcParams['figure.dpi'] = 100\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.bar(num_arr, sum, align='center', color='b')\n",
    "ax.set_title('Ukupno pojavljivanje heroja')\n",
    "ax.set_ylabel('Broj pojavljivanja')\n",
    "ax.set_xlabel('ID heroja')\n",
    "plt.show()\n",
    "\n",
    "#prvi zajednicki\n",
    "fig1, ax1 = plt.subplots()\n",
    "ax1.bar(num_arr, pos, align='center', color='r')\n",
    "ax1.bar(num_arr, neg, align='center', color='b')\n",
    "ax1.set_title('Broj pojavljivanja heroja sa oznakama 1 i -1')\n",
    "ax1.set_ylabel('Broj pojavljivanja 1 / -1')\n",
    "ax1.set_xlabel('ID heroja')\n",
    "\n",
    "plt.show()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Naredni grafikon pokazuje zastupljenost regiona u mečevima"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1: US West 2: US East 3: Europe West 4: South Korea 5: Southeast Asia 6: China 7: Australia 8: Russia 9: Europe East 10: South America 11: South Africa 12: Chile 13: Peru 14: India\n",
      "{5: 26635, 6: 15973, 8: 6463, 3: 3159, 4: 1663, 2: 1495, 1: 1076, 9: 1071, 10: 912, 7: 722, 14: 225, 12: 129, 13: 125, 11: 120}\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x200 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "regi = g.iloc[:,1]\n",
    "abc = dict(regi.value_counts())\n",
    "\n",
    "x = np.fromiter(abc.keys(), dtype=int)\n",
    "y = np.fromiter(abc.values(), dtype=int)\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "ax.bar(x, y, align='center', color='b')\n",
    "\n",
    "ax.set_title('Zastupljenost regiona')\n",
    "ax.set_ylabel('Broj pojavljivanja')\n",
    "ax.set_xlabel('ID regiona')\n",
    "\n",
    "print(\"1: US West\", \"2: US East\", \"3: Europe West\", \"4: South Korea\", \"5: Southeast Asia\", \"6: China\", \"7: Australia\", \"8: Russia\", \"9: Europe East\", \"10: South America\",\"11: South Africa\",\"12: Chile\",\"13: Peru\",\"14: India\")\n",
    "print(abc)\n",
    "\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels(x)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Naredni grafikon pokazuje zastupljenost moda igre u mečevima"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1: All Pick 2: Captains Mode 3: Random Draft 4: Single Draft 5: All Random 6: INTRO/DEATH 7: The Diretide 8: Reverse Captains Mode 9: Greevilin\n",
      "{2: 43196, 9: 10559, 8: 3698, 1: 1295, 3: 414, 6: 298, 4: 204, 7: 91, 5: 13}\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x200 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mode = g.iloc[:,2]\n",
    "abc = dict(mode.value_counts())\n",
    "\n",
    "x = np.fromiter(abc.keys(), dtype=int)\n",
    "y = np.fromiter(abc.values(), dtype=int)\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "ax.bar(x, y, align='center', color='g')\n",
    "\n",
    "ax.set_title('Zastupljenost moda')\n",
    "ax.set_ylabel('Broj pojavljivanja')\n",
    "ax.set_xlabel('ID moda')\n",
    "\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels(x)\n",
    "\n",
    "print(\"1: All Pick\", \"2: Captains Mode\", \"3: Random Draft\", \"4: Single Draft\", \"5: All Random\", \"6: INTRO/DEATH\", \"7: The Diretide\", \"8: Reverse Captains Mode\", \"9: Greevilin\")\n",
    "print(abc)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Naredni grafikon pokazuje zastupljenost tipa igre u mečevima"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2: Tournament\n",
      "{2: 59768}\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x200 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "type_match = g.iloc[:,3]\n",
    "abc = dict(type_match.value_counts())\n",
    "\n",
    "x = np.fromiter(abc.keys(), dtype=int)\n",
    "y = np.fromiter(abc.values(), dtype=int)\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.bar(x, y, align='center', color='r')\n",
    "\n",
    "ax.set_title('Zastupljenost tipa')\n",
    "ax.set_ylabel('Broj pojavljivanja')\n",
    "ax.set_xlabel('ID tipa')\n",
    "\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels(x)\n",
    "\n",
    "print(\"2: Tournament\")\n",
    "print(abc)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Razdvajanje feature i target kolona"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y = g.iloc[:,0]\n",
    "X = g.iloc[:,1:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Razdvajanje train i test setova 80:20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "#random state = 1, kako bi se koristio int seed za RNG\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def draw_conf_matrix(confusion_mat):\n",
    "    plt.rcParams['figure.figsize'] = [3, 2.5]\n",
    "    plt.rcParams['figure.dpi'] = 100\n",
    "    df_conf = pd.DataFrame(confusion_mat, index=['-1', '1'], columns=['-1', '1'])\n",
    "    cmap = sns.cubehelix_palette(light=1, as_cmap=True)\n",
    "    res = sns.heatmap(df_conf, annot=True, vmin=2346.0/2, vmax=3670.0, fmt='.2f', cmap=cmap)\n",
    "    res.invert_yaxis()\n",
    "    res.invert_xaxis()\n",
    "    plt.title('Matrica konfuzije')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Treniranje modela korišćenjem Linear Support Vector klasifikacije"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearSVC accuracy :  0.5992136523339467\n",
      "Confusion matrix:\n",
      "[[3602 2336]\n",
      " [2455 3561]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 300x250 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.59      0.61      0.60      5938\n",
      "           1       0.60      0.59      0.60      6016\n",
      "\n",
      "    accuracy                           0.60     11954\n",
      "   macro avg       0.60      0.60      0.60     11954\n",
      "weighted avg       0.60      0.60      0.60     11954\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.metrics import accuracy_score\n",
    "svc_model = LinearSVC(random_state=0,max_iter = 100000, dual = False, penalty='l2')\n",
    "\n",
    "pred = svc_model.fit(X_train, Y_train).predict(X_test)\n",
    "print(\"LinearSVC accuracy : \",accuracy_score(Y_test, pred, normalize = True))\n",
    "print(\"Confusion matrix:\")\n",
    "confusion_mat=confusion_matrix(Y_test, pred)\n",
    "print(confusion_mat)\n",
    "\n",
    "draw_conf_matrix(confusion_mat)\n",
    "\n",
    "print(\"Report:\")\n",
    "print(classification_report(Y_test, pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Treniranje modela korišćenjem logističke regresije"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of Logistic regression classifier on training set:  0.6016229556196929\n",
      "Accuracy of Logistic regression classifier on test set:  0.5992973063409738 \n",
      "\n",
      "Confusion matrix:\n",
      "[[3607 2331]\n",
      " [2459 3557]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 300x250 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.59      0.61      0.60      5938\n",
      "           1       0.60      0.59      0.60      6016\n",
      "\n",
      "    accuracy                           0.60     11954\n",
      "   macro avg       0.60      0.60      0.60     11954\n",
      "weighted avg       0.60      0.60      0.60     11954\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "logreg = LogisticRegression(max_iter=1000, dual=False)\n",
    "logreg.fit(X_train,Y_train)\n",
    "pred = logreg.predict(X_test)\n",
    "\n",
    "print('Accuracy of Logistic regression classifier on training set: ',logreg.score(X_train, Y_train))\n",
    "print('Accuracy of Logistic regression classifier on test set: ', accuracy_score(Y_test, pred),'\\n')\n",
    "print(\"Confusion matrix:\")\n",
    "confusion_mat=confusion_matrix(Y_test, pred)\n",
    "print(confusion_mat)\n",
    "draw_conf_matrix(confusion_mat)\n",
    "print(\"Report:\")\n",
    "print(classification_report(Y_test, pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Treniranje modela korišćenjem Random Forest klasifikatora"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of Random forest classifier on test set:  0.5834866990128827\n",
      "Confusion matrix:\n",
      "[[3503 2435]\n",
      " [2544 3472]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 300x250 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.58      0.59      0.58      5938\n",
      "           1       0.59      0.58      0.58      6016\n",
      "\n",
      "    accuracy                           0.58     11954\n",
      "   macro avg       0.58      0.58      0.58     11954\n",
      "weighted avg       0.58      0.58      0.58     11954\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "clf = RandomForestClassifier(max_depth=100, random_state=0, n_estimators=200, criterion='entropy', min_samples_split=2\n",
    "                            ,n_jobs=-1,bootstrap=True,min_samples_leaf=20)\n",
    "clf.fit(X_train, Y_train)\n",
    "pred = clf.predict(X_test)\n",
    "print('Accuracy of Random forest classifier on test set: ',accuracy_score(Y_test, pred))\n",
    "print(\"Confusion matrix:\")\n",
    "confusion_mat=confusion_matrix(Y_test, pred)\n",
    "print(confusion_mat)\n",
    "\n",
    "draw_conf_matrix(confusion_mat)\n",
    "print(\"Report:\")\n",
    "print(classification_report(Y_test, pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Treniranje modela korišćenjem SGD (stochastic gradient descent) klasifikatora"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of SGD classifier on test set:  0.5972059561653004\n",
      "Confusion matrix:\n",
      "[[3593 2345]\n",
      " [2470 3546]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 300x250 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.59      0.61      0.60      5938\n",
      "           1       0.60      0.59      0.60      6016\n",
      "\n",
      "    accuracy                           0.60     11954\n",
      "   macro avg       0.60      0.60      0.60     11954\n",
      "weighted avg       0.60      0.60      0.60     11954\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import SGDClassifier\n",
    "clf = SGDClassifier(loss='log', penalty='l1',fit_intercept=False, learning_rate='adaptive',eta0=1\n",
    "                   ,power_t=0.5, max_iter=1000)\n",
    "clf.fit(X_train, Y_train)\n",
    "pred = clf.predict(X_test)\n",
    "print('Accuracy of SGD classifier on test set: ',accuracy_score(Y_test, pred))\n",
    "print(\"Confusion matrix:\")\n",
    "confusion_mat=confusion_matrix(Y_test, pred)\n",
    "print(confusion_mat)\n",
    "\n",
    "draw_conf_matrix(confusion_mat)\n",
    "print(\"Report:\")\n",
    "print(classification_report(Y_test, pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Zaključak:\n",
    "Dati dataset ni u jednom slučaju klasifikacije nema rezultate koji se mogu smatrati odličnim. Ovo možemo pripisati tome da ishod pobede u meču zavisi u najvećem slučaju od ljudskog faktora, od \"skill-a\" igrača, koji ne možemo izmeriti. Svi primenjeni algoritmi imaju preciznost veću od 50%, pa možemo smatrati da je donekle klasifikacija bila uspešna, iako daleko od idealne."
   ]
  }
 ],
 "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
}