Coursework.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c92e394a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns; sns.set()\n",
    "import numpy as np\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "#Andreya's path\n",
    "#path = \"C:\\\\Users\\\\andre\\\\OneDrive\\\\Desktop\\\\University\\\\CITY - 3 YEAR\\\\Intro to AI\\\\Coursework\"\n",
    "#Mousuf's path\n",
    "path = \"C:\\\\Users\\\\mousu\\\\Documents\\\\GitHub\\\\Ai-Coursework-T3\"\n",
    "#STEFAN PATH DONT DELETE PLS\n",
    "#path = \"C:\\\\Users\\\\stefa\\\\OneDrive\\\\Documents\\\\Ai-Coursework-T3\"\n",
    "    \n",
    "#load from file\n",
    "filename_read = os.path.join(path, \"Dataset_Kmeans_Clustering_csv.csv\")\n",
    "df = pd.read_csv(filename_read)\n",
    "df = pd.read_csv(filename_read, na_values=['NA', '?'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "5113b3a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#divide the dataset into individual columns\n",
    "column_t1_summoner1_role = df[\"t1_summoner1_role\"]\n",
    "column_t1_summoner2_role = df[\"t1_summoner2_role\"]\n",
    "column_t1_summoner3_role = df[\"t1_summoner3_role\"]\n",
    "column_t1_summoner4_role = df[\"t1_summoner4_role\"]\n",
    "column_t1_summoner5_role = df[\"t1_summoner5_role\"]\n",
    "\n",
    "column_t1_summoner1_mastery_points = df[\"t1_mast1\"]\n",
    "column_t1_summoner2_mastery_points = df[\"t1_mast2\"]\n",
    "column_t1_summoner3_mastery_points = df[\"t1_mast3\"]\n",
    "column_t1_summoner4_mastery_points = df[\"t1_mast4\"]\n",
    "column_t1_summoner5_mastery_points = df[\"t1_mast5\"]\n",
    "\n",
    "column_t2_summoner1_role = df[\"t2_summoner1_role\"]\n",
    "column_t2_summoner2_role = df[\"t2_summoner2_role\"]\n",
    "column_t2_summoner3_role = df[\"t2_summoner3_role\"]\n",
    "column_t2_summoner4_role = df[\"t2_summoner4_role\"]\n",
    "column_t2_summoner5_role = df[\"t2_summoner5_role\"]\n",
    "\n",
    "column_t2_summoner1_mastery_points = df[\"t2_mast1\"]\n",
    "column_t2_summoner2_mastery_points = df[\"t2_mast2\"]\n",
    "column_t2_summoner3_mastery_points = df[\"t2_mast3\"]\n",
    "column_t2_summoner4_mastery_points = df[\"t2_mast4\"]\n",
    "column_t2_summoner5_mastery_points = df[\"t2_mast5\"]\n",
    "\n",
    "column_winner = df[\"winner\"]\n",
    "\n",
    "#merge each team 1 player's column with his mastery points\n",
    "summoner1_frame = [column_t1_summoner1_role, column_t1_summoner1_mastery_points]\n",
    "summoner2_frame = [column_t1_summoner2_role, column_t1_summoner2_mastery_points]\n",
    "summoner3_frame = [column_t1_summoner3_role, column_t1_summoner3_mastery_points]\n",
    "summoner4_frame = [column_t1_summoner4_role, column_t1_summoner4_mastery_points]\n",
    "summoner5_frame = [column_t1_summoner5_role, column_t1_summoner5_mastery_points]\n",
    "\n",
    "\n",
    "#merge each team 2 player's column with his mastery points\n",
    "summoner1_frame = [column_t2_summoner1_role, column_t2_summoner1_mastery_points]\n",
    "summoner2_frame = [column_t2_summoner2_role, column_t2_summoner2_mastery_points]\n",
    "summoner3_frame = [column_t2_summoner3_role, column_t2_summoner3_mastery_points]\n",
    "summoner4_frame = [column_t2_summoner4_role, column_t2_summoner4_mastery_points]\n",
    "summoner5_frame = [column_t2_summoner5_role, column_t2_summoner5_mastery_points]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "8499046e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#merging each summoner 1 from both teams's position column with his mastery point column\n",
    "t2_s1_merge = pd.merge(column_t2_summoner1_role , column_t2_summoner1_mastery_points, right_index=True, left_index=True)\n",
    "t1_s1_merge = pd.merge(column_t1_summoner1_role , column_t1_summoner1_mastery_points, right_index=True, left_index=True)\n",
    "s1_result = pd.merge(t2_s1_merge, t1_s1_merge, right_index=True, left_index=True) \n",
    "\n",
    "#locating where t2 summoner 1 player has greater amount of mastery points\n",
    "t2_greater_mp = s1_result.loc[s1_result['t2_mast1'] > s1_result['t1_mast1']]\n",
    "\n",
    "#comparing with original data(winner)\n",
    "t2_greater_mp_winner = pd.merge(t2_greater_mp, column_winner, right_index=True, left_index=True)\n",
    "\n",
    "#Only when Team 2 are winners(no matter mastery points difference)\n",
    "only_t2_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t2\"]\n",
    "\n",
    "#Only when Team 1 are winners(no matter mastery points difference)\n",
    "only_t1_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t1\"]\n",
    "\n",
    "\n",
    "fig = plt.figure()\n",
    "ax1 = fig.add_subplot(111)\n",
    "#Team 2 wins when they have more mastery points\n",
    "ax1.scatter(only_t2_winner['t2_summoner1_role'],only_t2_winner['t2_mast1'], c='r')\n",
    "#Team 1 wins despite that they have more mastery points\n",
    "ax1.scatter(only_t1_winner['t1_summoner1_role'],only_t1_winner['t1_mast1'], c='b')\n",
    "leg = plt.legend(loc='upper left', labels=[\"Mastery points T2 > Mastery points T1 [winner = T2]\", \"Mastery points T2 > Mastery points T1 [winner = T1]\"])\n",
    "leg.get_frame().set_alpha(0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "bed7c162",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#merging each summoner 2 from both teams's position column with his mastery point column\n",
    "t2_s2_merge = pd.merge(column_t2_summoner2_role , column_t2_summoner2_mastery_points, right_index=True, left_index=True)\n",
    "t1_s2_merge = pd.merge(column_t1_summoner2_role , column_t1_summoner2_mastery_points, right_index=True, left_index=True)\n",
    "s2_result = pd.merge(t2_s2_merge, t1_s2_merge, right_index=True, left_index=True) \n",
    "\n",
    "#locating where t2 summoner 2 player has greater amount of mastery points\n",
    "t2_greater_mp = s2_result.loc[s2_result['t2_mast2'] > s2_result['t1_mast2']]\n",
    "\n",
    "#comparing with original data(winner)\n",
    "t2_greater_mp_winner = pd.merge(t2_greater_mp, column_winner, right_index=True, left_index=True)\n",
    "\n",
    "#Only when Team 2 are winners(no matter mastery points difference)\n",
    "only_t2_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t2\"]\n",
    "\n",
    "#Only when Team 1 are winners(no matter mastery points difference)\n",
    "only_t1_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t1\"]\n",
    "\n",
    "\n",
    "fig = plt.figure()\n",
    "ax1 = fig.add_subplot(111)\n",
    "#Team 2 wins when they have more mastery points\n",
    "ax1.scatter(only_t2_winner['t2_summoner2_role'],only_t2_winner['t2_mast2'], c='r')\n",
    "#Team 1 wins despite that they have more mastery points\n",
    "ax1.scatter(only_t1_winner['t1_summoner2_role'],only_t1_winner['t1_mast2'], c='b')\n",
    "leg = plt.legend(loc='upper left', labels=[\"Mastery points T2 > Mastery points T1 [winner = T2]\", \"Mastery points T2 > Mastery points T1 [winner = T1]\"])\n",
    "leg.get_frame().set_alpha(0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "2f2fe8ae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#merging each summoner 3 from both teams's position column with his mastery point column\n",
    "t2_s3_merge = pd.merge(column_t2_summoner3_role , column_t2_summoner3_mastery_points, right_index=True, left_index=True)\n",
    "t1_s3_merge = pd.merge(column_t1_summoner3_role , column_t1_summoner3_mastery_points, right_index=True, left_index=True)\n",
    "s3_result = pd.merge(t2_s3_merge, t1_s3_merge, right_index=True, left_index=True) \n",
    "\n",
    "#locating where t2 summoner 3 player has greater amount of mastery points\n",
    "t2_greater_mp = s3_result.loc[s3_result['t2_mast3'] > s3_result['t1_mast3']]\n",
    "\n",
    "#comparing with original data(winner)\n",
    "t2_greater_mp_winner = pd.merge(t2_greater_mp, column_winner, right_index=True, left_index=True)\n",
    "\n",
    "#Only when Team 2 are winners(no matter mastery points difference)\n",
    "only_t2_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t2\"]\n",
    "\n",
    "#Only when Team 1 are winners(no matter mastery points difference)\n",
    "only_t1_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t1\"]\n",
    "\n",
    "\n",
    "fig = plt.figure()\n",
    "ax1 = fig.add_subplot(111)\n",
    "#Team 2 wins when they have more mastery points\n",
    "ax1.scatter(only_t2_winner['t2_summoner3_role'],only_t2_winner['t2_mast3'], c='r')\n",
    "#Team 1 wins despite that they have more mastery points\n",
    "ax1.scatter(only_t1_winner['t1_summoner3_role'],only_t1_winner['t1_mast3'], c='b')\n",
    "leg = plt.legend(loc='upper left', labels=[\"Mastery points T2 > Mastery points T1 [winner = T2]\", \"Mastery points T2 > Mastery points T1 [winner = T1]\"])\n",
    "leg.get_frame().set_alpha(0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "28b31dfc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#merging each summoner 4 from both teams's position column with his mastery point column\n",
    "t2_s4_merge = pd.merge(column_t2_summoner4_role , column_t2_summoner4_mastery_points, right_index=True, left_index=True)\n",
    "t1_s4_merge = pd.merge(column_t1_summoner4_role , column_t1_summoner4_mastery_points, right_index=True, left_index=True)\n",
    "s4_result = pd.merge(t2_s4_merge, t1_s4_merge, right_index=True, left_index=True) \n",
    "\n",
    "#locating where t2 summoner 4 player has greater amount of mastery points\n",
    "t2_greater_mp = s4_result.loc[s4_result['t2_mast4'] > s4_result['t1_mast4']]\n",
    "\n",
    "#comparing with original data(winner)\n",
    "t2_greater_mp_winner = pd.merge(t2_greater_mp, column_winner, right_index=True, left_index=True)\n",
    "\n",
    "#Only when Team 2 are winners(no matter mastery points difference)\n",
    "only_t2_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t2\"]\n",
    "\n",
    "#Only when Team 1 are winners(no matter mastery points difference)\n",
    "only_t1_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t1\"]\n",
    "\n",
    "\n",
    "fig = plt.figure()\n",
    "ax1 = fig.add_subplot(111)\n",
    "#Team 2 wins when they have more mastery points\n",
    "ax1.scatter(only_t2_winner['t2_summoner4_role'],only_t2_winner['t2_mast4'], c='r')\n",
    "#Team 1 wins despite that they have more mastery points\n",
    "ax1.scatter(only_t1_winner['t1_summoner4_role'],only_t1_winner['t1_mast4'], c='b')\n",
    "leg = plt.legend(loc='upper left', labels=[\"Mastery points T2 > Mastery points T1 [winner = T2]\", \"Mastery points T2 > Mastery points T1 [winner = T1]\"])\n",
    "leg.get_frame().set_alpha(0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "be303b67",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWwAAAEGCAYAAABB8K+FAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAyqElEQVR4nO3dd3gUVd/G8e/uJptCCC3EoKBAqIIoSAkBCVIEQvMFBakiNSCgKE2Iht4EQaooHZSIECkailKlBKT3gCAK0kICxJC65f0jT1aWbEgmZJMd8vtc13M95uzuzJnD7L1nzsyc0ZjNZjNCCCEcnjavKyCEECJrJLCFEEIlJLCFEEIlJLCFEEIlJLCFEEIlJLCFEEIlciWw4+LiaNWqFdeuXXvs+y5fvky3bt1o06YNvXr14v79+7lRPSGEUAW7B/aJEyfo1KkTV65ceez7zGYz/fv3p0+fPmzcuJHKlSvz9ddf27t6QgihGk72XsGaNWsICQlh+PDhlrL169ezfPlyTCYTVapUISQkhIsXL+Lu7k6DBg0ACAoKIjY21t7VE0II1dDk1p2OjRo1YsWKFSQkJBASEsLSpUtxcXFhxowZuLm5Ubp0aX788UeKFi3K2bNnqVChAp9++imFCxfOjeoJIYTDy/WTjgcPHuSvv/6iQ4cOtG3blu3bt3P58mUMBgOHDh2ia9eubNq0iVKlSjFlypTcrp4QQjgsuw+JPMpoNNKiRQuCg4MBePDgAUajkTNnzvDCCy/w0ksvAdCqVSsGDx6c29UTQgiHles97Dp16vDLL78QHR2N2WxmzJgxLF++nOrVqxMTE8P58+cB2LFjB1WqVMnt6gkhhMPK9R52pUqVGDhwIO+++y4mk4nKlSvTt29fXFxcmDdvHsHBwSQkJODj48O0adNyu3pCCOGwcu2koxBCiCcjdzoKIYRKSGALIYRKSGALIYRK2P2k4927DzCZlA+TFyvmQXR0nB1q9HSS9lJG2ksZaS/lsttmWq2GIkUK2HzN7oFtMpmzFdhpnxVZJ+2ljLSXMtJeyuV0m8mQiBBCqIQEthBCqESu3zgDkJycRGJiPGaziYwuA09MvE9iYkou10y9pL2UkfZS5mltL41Gg5OTM+7uBdFoNHldnUzlemCnhvUD3N0LotM5ZdhI7u564uOTc7l26iXtpYy0lzJPa3uZzWbi4/8lKSkBV1f3vK5OpnI9sBMT43F3L4iTk3Nur1oIoVDStavEnztLdEICWjc33Cu/iEvJUnldrRyj0WhwdS3Agwf3JbBtMZtN6HR5MhIjhFAg6dpV4k4cB6MRAFNCQurf8FSFtlarzXBo1tHk+klHs9msirEiIfK7+HNnLWFtYTSmlj9FNBqNBLZa3Lhxnfr1azJt2kSr8osXI6lfvybh4ZsUL/Ps2dPMnz87p6qYbVOmjOf8+cd/uQYN6pelZR08eIAePTrTo0dnmjZ9jY4d36RHj8588slQADZsCKNbtw50796RSZPGkpKSvRNUixcvpH79mpw+fdKq/MsvZ1C/fs1sL/PEiWPZ+mxOuXMniqFDHz+/u5L9ZsaMqfTo0ZmuXd+mYUM/y7/Nzz9vtLzn0qU/6Nq1g83Ph4dvokWLRowf/1mG6zAlJABw8cZ15mz+KV15XurT51169OhMu3YtadGikWX7L136gytX/mTAgN706NGZfv3e4+LFSP755xo9enTO9j7kKGRsAihUqBAHDx7AaDSi0+kA2L79FwoXLpKt5V258id378bkZBWzZeTITzN9z7FjR7K0rDp16lKnTl0ABg7sS8+efalRI3Xn//vvv1i9eiWLF6/E3b0AEyeOISxsDR07drF8/u7dGMLDN9G69f/h6en52HV5ez/Dzp3bqVq1GgAmk4njx7NWT1uOHTtC9eqvZvvzOcHLqzjTpz8+jJXsNx9/PAJI7XAMGtSPZcu+s3p98+afWLhwHk5OGX/F69dvwOjRYzJ8XevmhikhgfIlnqV8iWetyvPaN98sB1J/eI4dO2K1Hf3796Jr1x7Uq/caR478zoQJY1i+fDXLln0ngZ1bYiP2cydsHYaYaJyKFsOrXXs8/fxzZNlubu6UL1+BEyeOWULo0KEIatasbXnPunXfs2VLOImJCWi1WsaOnUzp0mWYO3cWv/9+EJ1OS/36Abz9dicWLfqKhIQEli9fTNeuPZg//0uOHTuC0WgiMLAVHTt24ejRwyxYMBuj0USZMmU4ceI4X3wxl+eff4GEhAS6dHmL1avDcHFxsdShVasm+Pu/RmTkOdzd3fnsswmUKPEsp0+fYu7cGSQkJFG4cGGGDRtFyZKlLMEKsHLlUlxdXbly5U98fcsREjKRefNmAam9lQULFjN58lguX74EwP/939u0afN/WWo/vV7Pxx+PpEABDwDKli3HrVs3rd7j7l6ABw8e0KtXV+rU8adDh3d4/vnSNpdXv34D9u3bw6BBQwA4efI4VapU4+LFCwA8eBDH5MnjiYq6zZ07UbzySnWCg8cRFXWbceM+JSEhAa1WwwcfDOPq1b+IjDzH1KkTmDRpOi4uLkyfPpm4uFicnV0YMmQYFSpUYuLEMdy/f59//rlKUNBAvv12BV99tQRIDb8zZ04xdOgnljqGh29i9+4dxMbGcvduDPXqvcbAgUPQaDSsWLGEbds2o9VqqVXLjwEDBnP79i0GDerH2rWbmDhxDAUKeBAZeY6oqNu8914fGjR43Wq/qVevAdOmTcRoNKLX6xk1KoRSpZ7P0r9HXFwce/fuZsyYiUyYEJLp+7t378i4cVMoXboMY8cGU6BAAYYO/YQrTjqWrQ3l/2r7EbpvDxM7dWd06EpefLk6Z9aGcu/eXT78cBh169azuU0tW7YhPj6eL76YyuXLlzCZTHTp0p2mTZsTHr6JzZt/4v79e9Sr14B+/d631GfSpLFcuBBpVcfOnbvxxhstsrT9rVu/id//ssHXt3y6fVHNVBHYsRH7ubViGebk1MuKDDHR3FqxDCDHQvv115uyc+d2atSoyblzZyhXrrxlXOvBgzj27NnN3LkLcXFxZdGir/jxxx/o1KkbERH7WbVqDUlJSUydOgG9Xk/v3kEcO3aEd9/txfr1awFYsuRbkpOT+eijgVSq9CIAV6/+zdq1P+Hh4cHixQvZtm0zvXsHsWvXdvz961uFNcC9e/eoXv1VRo0KYe3aUGbN+pwJE6YxZswoPv98BmXKVGDHjl8ZM2Y0ixatsPrs6dMn+fbbtXh5Fadfvx4cPHiADz8cxtq13/PNN8s5duwIsbGxLF36Hffv32Pu3FlZDmwfnxL4+JQA4O7du4SFrWHUKOugcHFxoW/fAfTo0Zvt27cxfnwIhQoVYtKk6ej1eqv3FipUmBIlnuXcuTNUrlyF7dt/oXHjppa23L9/L+XLV2DChKmkpKTQtevbREaeZ//+3/D3r0/nzt05evQwJ08ep3Pnbvz880Z69uyLr285+vfvyZAhw3nllWqcOXOeUaOGsnp12P/WW4hp02ZiNpuZPXsm//xzjeeeK8nmzT/Rr9/AdNt9/vxZli79joIFPRk0qB979uxEr9ezd+8eFi9eiU7nRHDwcNavX4e/f32rz96+fYv58xdx+fIlBg3qR8uWbaz2m0mTxvLOO11p1KgJ27dv48yZU1kObA8PDyZO/JwbN65n6f1169bnyJFDlC5dhkuXLlrKj16+RL2GjdD+bxxb6+aGroAHZlc3Fi5cyt69e/jmmwXUrVsvw21avnwxFStWJjh4LA8exBEU1JMXX6wKQFTUbVat+iHdUcCj+45SgYGtLf+9aNFXNGjQ8ImW50hUEdh3wtZZwjqNOTmZO2Hrciyw69d/jW++WYDJZGL79l9o1Kgp27dvA6BAAQ/GjJnAr79u4+rVvzl4cD/ly1fEy6s4Li4u9O/fE3//1+jTp3+6kD18+BAXL17gyJHDACQkxHPp0h+ULl2GUqVewMMjtVcaGNiaDz8cQO/eQWzZ8jN9+77Po/R6F5o3bwlAixatWLhwHlev/kXBggWpUqUq8fHJNGrUhGnTJhIXZz3pTJkyvnh7PwPACy+U4d9/Y61eL1vWl7///ouPPhqIn189+vcfpLgNo6JuM3ToYFq1ams5UnmURqNBq9Va/j8jjRql/oBWqFCJ06dPMGTIMMtrTZs25+zZ06xZ8x1XrvzJ/fv3SUiIp2bN2owePZwLFyLx969P+/bW47fx8fGcO3eWSZPGodVqMJnMJCQkcP/+PQBLkGg0Glq0aMnWreEEBrYhJiaGKlWqpqtjvXoNKFq0GACNG7/BkSO/o9e70KRJM1xcXAFo2bINmzf/nC6wa9eug0ajoWxZX2Jj76dbdt269fjii2kcPLgff//XaNiwcYZt9aT8/esTGvotNWrUokyZsvz111/cvRtDRMQ+xo+fxvXr13A5fYIiTZuh2bDOMjRWtqyv1X5ka5sOHz5EUlKiZWw9MTGRP/+8DECFCpVsDtk8aQ8bUi9umDfvS86ePcXs2QuVNYgDU0VgG2KiFZVnh7t7AcqVK8/Jk8c5evR3goIGWgL71q2bDBrUj/btO+Dn50/RosW4eDESJycnvv56GcePH+XAgX0EBb3HnDlfWy3XaDQxYMBgAgIaAam9ZDc3V86cOW0V7iVKPIuPTwl2795BTEy0zYDQajWWK2xMJjM6nS6DyWXMmEzWZ/cf7sXaOiteqFBhVq5cw++/H+TAgX307NmVlSvXULBgwSy1319/XeHjjwfRvn1HOnXqmu71pKQkVq1axtat4dSuXZfg4LE8//wLGS6vQYOG9O/fi9q1/Xj55RpW4b52bSi7du2gTZv/4623avPnn5cwm81Uq/YKq1atYf/+vWzfvo3w8E3MmjXf8jmTyYRe78KyZd9ZbgS5ffsWnp6FAKz+PQIDW/Pxx4PQ6/U0bx5os44PX56adrmq2Wyyeo/ZDEajId1n9frUdWV0xdTrrzehatVq7Nv3Gz/8sJqIiH2MGBGcYXs9iapVq/HHHyEcPnyI6tVfpUiRYuzcuZ2UFAM+Pj5cv37tkbrrLXV/eD+ytU0mk5FPPx1PxYqVAIiJicbTsxDbtm1O17lJ86Q9bIPBwIQJIdy5E8Xs2QstnaKngSquEnH6Xy8mq+XZ1ahRE776ai4VK75o9ct//vxZSpYsRceOXXjxxapEROzHZDJy4cJ5Bg7sy8svV2fgwA8pXbosf//9FzqdDuP/DiNffbUmGzeux2AwEB8fz4ABvThz5rTN9bds2YZZs6bTrJntgEhMTGTv3j0AhIdvpE4df55//gXu379vWeb27b/wzDMlLCGUGZ1Oh8FgYO/e3Ywb9yn+/vX58MOhuLm5cfv2rSwtIz7+AUOGvE+fPv1thjWkDivp9XoWLVrB0KEjHxvWkPoD4uNTgkWLvqJx46ZWr/3++0HatGn3vx6XhosXL2AymZg//0u2bg2nRYtWDBkywtJL0+mcMBqNeHh4ULJkKbZuDf/fciJ4//2+Ntfv41OC4sW9Wb9+neWo5lEREfuJi4sjKSmJX37Zip+fPzVq1OLXX7eSlJSIwWAgPHxjhkcbj3p4v/nss084e/YMb77Znt69g4iMPJ+lZWSHTqfjxRersHZtKNWr1+TVV2uyYsUSy1DHk6hRo5ZlKOvOnTu8+24nu48pz5v3JfHxD/jii7lPVViDSnrYXu3aW41hA2j0erzatc/R9dSr14ApU8bTu3eQVXmtWn78+ONaunZ9G2dnZ158sSqXL1+iQoVKVK1aje7dO+Lq6kr58hXx8/Pn+vV/WLLkaxYsmEOfPv25du0q773XGaPRSGBga2rUqMnRo4fTrT8g4HWmTp2QYUAA7Nz5K998M59ixYoTHDwGvV7PuHGTmTp1Eg8exOPpWYhx4yZneZvr129Ajx6d+frrpezcuZ1u3Tqg1+sJCGiEr2+5LC1j06b13L0bw+rVq1i9epVluQ+3Y9GixejW7b0s1wtSf0CXLPnGcrVImg4dOjN9+mRCQ1OvSqlatRo3blynffuOjB0bTHj4T2i1Wj7+eCSQeoXL9OmTCQ4eS0jIBD7/fBKrV6eOMY8bNynDXm6TJm+wa9cOvLyK23y9SJEiDBs2mHv37tGsWaBlqODixUh69eqO0WigTp26tG/fkaio25lub+XKVSz7Tbdu7zF16gSWL1+ETqeznIC1l7p163P8+FFeeKE0RYsW4+7dmHTDONnRs2cfZsyYSrduHTCZUo82n3uupN0us0w7h1KixLP07dvDUv7oVTRqZfeH8EZHx1kdtt+7d4fChb0y/dyjcxfY8yoRR2A2m4mI2Mf69euYOnWmzffUr1+TvXvTBz08vXM92Etm7WUwGBg//jMaNWpiGc56mK3LydREaf2flv0ro+9QVnNJieLFCxIV9a/iz2m1GooVs31koIoeNqReDfI0BfSjZs/+gn379mR6ra6wP7PZzJtvtqBWrTq89lrDvK6O3ezdu4fx4z/j00/H5XVV7O6ff64xevTwvK7GE1NND1s8nrSXMtJeyjzt7SU9bCGEyIfShm8v3I3BqUjRHB2+lcAWQogcYu+b/FRxWZ8QQqjB427yywlZDuypU6cycuTIHFmpEEI8jex9k1+WAvvAgQP8+OOPObJCIYR4Wtn7Jr9MA/vevXvMnDmToKCgzN6qSjIftsyHnRvUOB92mvPnzzJu3JPdLp7THjcfdpqfftrAxIljAHJtPmyvdu3RPDKZWU7e5JfpScfPPvuMIUOGcOPGjRxZYXYdOHOTsN2XiI5NopinC+0CfKlbxSdHli3zYWdO5sN+MmqcDztNpUovUqPGKw51Wd/j5sNOSkpiyZKvCQv7gYYNU296eu65krkyH3baicU7Yesw5PZVIj/88AMlSpSgbt26hIWFZWsFj15PmJh4H3d3fQbvtpb2vr0nrrN8y3mSU1In1omOTWL5lvO46J2o//Kzj1tEptzcnHF3L0DFihU5f/4UtWqlzoF9+PBB/Pz80OudcHfXExq6mvDwTSQkJKDRaJky5XPKli3LzJnTiYiIQKfTEhDwOp07d2HJkoXEx8fz3XfLeO+9Xsya9QVHjvyO0Wiideu2dO3ajcOHf+fLL2diNBopW9aXY8eOMn/+V7zwQmkSEuJp164t69f/ZDVBTqNGAbz2WgPOnTtLgQIFmDhxMs8++xwnT55g+vSpJCUlU7hwYUaP/oznn3+ePn160q9ffwCWLFmEq6srf/55mXLlyjNp0lRmzpwBQFBQDxYvXs7YsSGWHsrbb3egXbu3bLaZTqfF1dXZ8u/j6VmA0aM/pXjxogBUqlSR27dvW/0763RFSE5OpHfvrtSrV59OnbpQunSZdMt2dtbRsOHr7N//GyNGpIbSkSOHeeWVV7h48QLu7nri4uIYNy6EW7duERUVRY0arzJ+/ERu377F6NGf/G8+bC3Dho3g779T58OeNm0iM2bMxMXFlcmTJ3D//j1cXd0YPnwklSpVJiQkmHv37nH16lUGD/6QZcuWsGzZSgA2bdrAqVMnGTXqvx/AjRs3sGPHr8TGxhIdHU2DBgF89NFQNBoNixd/Q3j4z+h0Wvz8/PnggyHcunWTPn168fPPWwgJCcbDw4Nz585x69Yt+vYNolGjRlb7TYMGDZkwYSxGoxEXFxfGjBlnc/4VNzdnNBqNVVv/+++/HDjwG5MnT+Wzz4Jtft/0eiecnLS4u+vp0KEdU6ZMp2zZsowePZICBQowatSnnDx5gm++Wci7777HwoUL+OabJfTp05MqVapy7NhR7t69y4gRI6lX7zWb29S27ZvEx8czZcpE/vjjD0wmEz16vEfz5oFs3LiBn37ayL17d3nttQAGDfrAUreQkE/TzZ3SvXsPAgPTT9nw8HakOXHiMDqdhg8/HMLp06fSbb+t9khMdKZ48axNdJaZ4q2b4du6WY4s61GPDezw8HCioqJo27Yt9+/fJz4+nkmTJjFq1Kgsr+DRG2cSE1Oy9Ev98IX6q3+JtIR1muQUE6t/iaRG+Se72D0hIQWz2UyDBo3ZsmUrVaq8wrlzZyhbthxms5nkZANRUTFs376dL7/8yjIf9urV39GpUzd++22v1XzYBoOGnj37cezYETp37sH333+PwWBk0aJVlvmwfX0rYDQa+euvK1bzYW/YsJHevYPYvHkLdevWx2jUWLXVvXt3qVatOiNGfMrataFMnjyJCROmMXLkcKv5sEeOHM6iRSswGk0kJqYOTZw4cdxqPuydO3czcOBHhIZ+x1dfLePQocPcvXuXxYtXWebDbt68jc02S1tuWt0KF/aicGEv4uOTuXv3LqGhqxk1KuSRf2cN773Xjy5d3mP79m2MHj3K5nzYKSlGChQoiI9PCY4cOUblylUID99Mo0ZN+OGHNcTHJ7N9+w7KlCnHmDGTLfNhHzt2kv37f8PPr55lPuxDhw7TuXM3wsLC6NmzL889VzrdfNgjRqTOh20wmPDw8GTVqh8wm818/vk0Ll68zHPPlWTDhg306zfQanuSkw2cOXPaaj7sLVu2otfr2blzJ4sWrbDMh/3dd6vx96+P2WwmPj4Zg8HEP//cYM6cry1zRzdtGmi130yaNJYOHbpY5sM+fPgoXl4lMtx/H66bTufCuHFTuXHjerrXHq6/wWAiPj6ZOnXqsW/fPnx8ShIZmTphVnx8Mrt376FOnXqWfSg+Phmj0URCQhILFixh7949zJkzh+rV62S4TQsWLMDXtyIjR4ZY5sP29a1EcrKBmzdvWubDfriOI0bYPjLMbDvSvPxyTV5+OXU489HXMlpOYmJKtm5yeZxcv3Fm6dKllv8OCwvj0KFDisI6p0THJikqzw6ZD1vmw5b5sDOeD/thapgP+2mlihtninm62AznYp6259PNDpkPW+bDlvmwn475sJ9mWb4Ou127dkyZMsWedcl43QG+6J2sq6p30tIuwDdH1yPzYct82GlkPmz1zof9NFNFDzvtahB7XSWSRubDlvmwHybzYatvPuynnczW5yBkPuzcJfNhy3zYD1PLbH0yl4iDmD37C2bO/Jz33/8wr6uS76XNh63VavPFfNj5QdqNM2onPeynhLSXMtJeyjzt7SU9bCGEEDkq1wPb1iVlQgiRV8xmc4Ynnh1Nrge2k5Mz8fH/YjQaJbiFEHnKbDaTlJRgmUPI0eX6ZX3u7gVJSkrgwYP7mM3mDEM7MdHZckusyJy0lzLSXso8re2l0WjQ6XS4u+fMPCL2luuBrdFocHV1x9XV/bHvy+6AfX4l7aWMtJcy0l6OQU46CiGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESkhgCyGESuT6U9OFEOpxc9VyYvfs5oLJBFotng0C8On6bl5XK9+SwBZC2HRz1XJid+38r8BksvwtoZ03ZEhECGFT7J7disqF/UlgCyFsM5mUlQu7k8AWQtimzSAeMioXdictL4SwybNBgKJyYX9ZCuwvv/ySwMBAWrZsydKlS+1dJyGEA/Dp+i6eDV//r0et1eLZ8HU54ZiHMr1K5NChQ0RERLBx40YMBgOBgYEEBARQtmzZ3KifECIP+XR9F5+u71K8eEGiov7N6+rke5n2sGvXrs2KFStwcnIiOjoao9GIu7t7btRNCCHEQ7I0JOLs7Mzs2bNp2bIldevW5ZlnnrF3vYQQQjxCYzabzVl9c0JCAkFBQQQGBtKxY0d71ksIIcQjMh3DvnTpEsnJyVSuXBk3NzfeeOMNIiMjs7yC6Og4TKYs/yZYyJiZMtJeykh7KSPtpVx220yr1VCsmIft1zL78LVr1wgODiY5OZnk5GS2b9/Oq6++qrgSQgghnkymPeyAgABOnDjBm2++iU6n44033qBly5a5UTchhBAPydLkT4MHD2bw4MH2rosQQojHkDsdhRBCJSSwhRBCJSSwhRBCJSSwhRBCJeSJMyoXG7GfO2HruHA3BqciRfFq1x5PP/+8rpYQwg4ksFUsNmI/t1Ysw5ycDIAhJppbK5YBSGgL8RSSIREVuxO2zhLWaczJydwJW5dHNRJC2JMEtooZYqIVlQsh1E0CW8WcihZTVC6EUDcJbBXzatcejV5vVabR6/Fq1z6PaiSEsCc56ahiaScW74StwyBXiQjx1JPAVjlPP388/fxl+ksh8gEZEhFCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJWQwBZCCJVwysqb5s6dy+bNmwEICAhg+PDhdq2UEEKI9DIN7P3797N3715+/PFHNBoNvXv35pdffqFp06a5UT8hRB6KjdjPnbB1XLgbg1ORoni1a4+nn39eVyvfyjSwixcvzsiRI9Hr9QD4+vpy/fp1u1dMCJG3YiP2c2vFMszJyQAYYqK5tWIZgIR2Hsl0DLt8+fK88sorAFy5coXw8HACAgLsXS8hRB67E7bOEtZpzMnJ3Albl0c1Elkawwa4ePEi/fr1Y8SIEZQuXTrLKyhWzCM79QKgePGC2f5sfiTtpYy01+NdiIm2WW6IiZa2y6KcbqcsBfaRI0cYPHgwo0aNomXLlopWEB0dh8lkVlyx4sULEhX1r+LP5VfSXspIe2WBVgsmk81yabvMZXcf02o1GXZ0Mw3sGzdu8P777zNz5kzq1q2reOVCCJWyFdaPKxd2l2lgL168mKSkJKZMmWIpe+edd+jUqZNdKyaEyFtORYthsDEs4lS0WB7URgBozGaz8vEKBWRIJHdIeykj7ZW5R68SAdDo9TzTvYdcJZIFeTIkIoTIn9JC+U7YOgxyHbZDkMAWQmTI088fTz9/OSJxEDKXiBBCqIQEthBCqIQEthBCqIQEthBCqIQEthBCqIRcJSLyFZkuVKiZBLbIN2S6UKF2MiQi8g2ZLlSonQS2yDdszYvxuHIhHI0MiYh8QyYzUk7G/B2LBLbKyRcq67zatefm0iVgNPxXqHPCq137vKuUA5Mxf8cjQyIqlvaFMsREg9ls+ULFRuzP66o5sEdnjrTrZJWqJmP+jkcCW8XkC6XMnbB1YDRaFxqN0l4ZkDF/x+NwQyJyiJ918oVSRtpLGRnzdzwO1cOOjdjPzaWLrQ7xby5dLIf4GcjoiyNfKNukvZTxatcejV5vVabR62XMPw85VGDfWv2tzUPWW6u/zZsKOTj5Qikj7aWMp58/z3TvkfqDptHgVLSYPG0mjznUkIj5wQNF5fmdPBFEGWkv5eQBBo7FoQJbKCdfKGWkvYSaOdSQCBqNsnIhhMhHHKuHndED3O37YHdVk6tqhMg/HCuwtVowmWyXi3TkTjQh8hfHCmxbYf248nzucTfOSGDbJkckwt7suY85VGDLhfrKyI0gysgRiXLyA6eMvfcxhxprkOtklZEbQZSRW/mVkblqlLP3PuZQge3p509B/3r/jVlrtRT0rye/6BmQHzhl5IhEGfmBU87e+5hDBXZsxH7+3b/vvzFrk4l/9++TX/QMyJ1oymg9PBSV53fyA6ecvY96HWoMW06iKSc3gmSdOYPLQzMqz+/knJJyXu3aW41hQ84e9TpUD1t+0YU9ydQHyni1aw86nXWhTidDbo9h76Neh+phawoUsPnl0RQokAe1EU8b6TFmx6N3Gctdx5mx51GvQ/WwMRiUlQuhgJykVSb1gQ+PfPeMBjnpmIccqodtTkpSVC6EEjJbnzIyROl4HCqwhbA3OUmbdVoPD0xxcTbLRd5wqMCWMWzlbq5aTuye3VwwmUCrxbNBAD5d383raomngFxV43gcKrCf6dSFm0sXWz91RqfjmU5d8q5SDuzmquXE7tr5X4HJZPlbQts2udU66+SqGseT5ZOOcXFxtGrVimvXrtmtMp5+/vi818vqkhif93rJFyoDsbt3KSrP71KfGbrkkWeGLpEbszIgUx84niz1sE+cOEFwcDBXrlyxc3VkjFERmT9ckduh39m86uF26HfSKbDBvVo16yO4h8pF3shSD3vNmjWEhITg7e1t7/oIYTe2TqA9rjy/iz95UlG5sL8s9bAnTpxo73qI7NDr4ZFb+S3lQjwhuazP8dj9pGOxYtm/BKh48YI5WJOnj3lgfy5+8WW68vID+0vb2XDhMa9Je6V34TFPgJL2ypqcbie7B3Z0dBwmk/IxVRnDztyfoT9kWK55sXou10bdZF+z4TFPgJL2yljalUjZvTlLq9Vk2NF1qMv6QC67UsJw/bqiciGUkPsilMtXT5yJjdjPzSWLrC+7WrJILrsSOcPFRVl5fidz+yjmUE+c2bFjByVLlsyRFdtya+Xy9IdhJlNquRBPSOvsrKg8v5O5fZSz94lahxoSkR1EGadnn7U5/OH07LN5UBvHJ5f1KXfGozS7i9Ug1qkAnoYHBEQfpUrclbyulsOy9zCSQw2JCGW8AlspKhdCiTMepdns7U+sswdoNMQ6e7DZ258zHqXzumoOy96dTglsFctoXEzmKxY5YXexGhi01gfhBq0Tu4vVyKMaqYCdx/0lsFVMbmwQ9hTrZPswPqNyYX8S2Comk/MIe/I02J6VL6NyYX8S2CqW0SQ8MjmPyAkB0UdxMlkfyjuZDAREH82jGgmHukpEKCOT8wh7SrsaRK4ScRwS2ComY9jC3qrEXZGAdiAyJCKEECohgS2EEDlFp1NWrpAEthBC5BSNRlm5QhLYQgiRU+TGGSGEECCBLYQQqiGBLYQQKiGBLYQQKiGBLYQQKiF3OgohMiQPMHAsEthCCJvOeJTmJ+96mLWpN33EOnvwk3c9ACrkZcXyMRkSEULY9ItXbUtYpzFrdfziVTuPaiQksIUQNiXqbD9NPqNyYX8S2EIIoRIyhi3yldUlGvOX+39PlX8h/jqdbmzPwxo5LjdjEglOrjbLRd5wuMCWs9LCXixh/dBEPH+5P8vqEo0JycN6Oaomdw4R7l0P40Pj2DqTkSZ3DgGBeVexfMyhAvuMR2k2e/tbntQc6+zBZm9/QM5Kiyf3aFgDoNFY9bjFf6rEXeFkQV+r9imZcFM6UJnY6lWL44UqYkaDBjOv3I+k2Z3fc2TZDhXYu4vVsIR1GoPWid3FavB/eVQnRydHJMJetnrVsnlEstWrlnSgMrDVqxbHClWytJkZTerf5Eyn06ECO9apgKLy/E6OSIQ9PRw8Fpr/Akikd7xQRZttdrxQxRxZvkNdJeJsSlFUnt897ohE2GA2KysXQiEzth9UkFG5Ug4V2Cla2x3+jMrzOzkiUcjOTwMRQoPtH/+MypVyqMAmw18h+ULZltFOID1GIfKCzmT7yTIZlSvlYIEtlJEfOCEciUHrrKhcKQlsIYRQCQlsIYRQCQlsIYRQCQlsIYRQCQlsIYRQiSwF9qZNmwgMDKRp06Z8++239q6TEEIIGzK9I+XWrVvMnDmTsLAw9Ho977zzDnXq1KFcuXK5UT8hhBD/k2kPe//+/fj5+VG4cGHc3d1p1qwZW7ZsyY26CSGEeEimPezbt29TvHhxy9/e3t6cPHkyyysoVswjezV7RPHiBXNkOfmFtJcy0l7KSHsplxNtlmlgm21MjKNRMPdCdHQcJtOT3yodFfXvEy8jP5H2UkbaSxlpL+Wy2mZarSbDjm6mQyLPPPMMd+7csfx9+/ZtvL29s1hFhczm9DOn2SoTqaS9lJH2UkbaSzk7t1mmge3v78+BAweIiYkhISGBbdu20aBBgxxZ+aOWjGry38Y99L8lo5rYZX1qJ+2ljLSXMtJeytm7zTRmW2Mej9i0aRMLFy4kJSWFt956iz59+mR5BdkdEilevKAcdikg7aWMtJcy0l7KZbfNHjckkqWJplu3bk3r1q0Vr1gIIUTOkTsdhRBCJSSwhRBCJSSwhRBCJez+sEStNvtPP3mSz+ZH0l7KSHspI+2lXHba7HGfydJVIkIIIfKeDIkIIYRKSGALIYRKSGALIYRKSGALIYRKSGALIYRKSGALIYRKSGALIYRKSGALIYRKSGALIYRK2P3WdFsuXLhA69atmT17Ns2aNQOgW7du3Lx5E3d3d4xGI3q9ng8++ICAgADL59avX8/KlSsxGAyYTCbefvttunfvnhebkGPCwsI4dOgQALVr16Zdu3aW1+bMmQPAoEGD6NatG97e3syYMcPm6wAHDhxg3rx5REVFYTKZqFy5MqNGjcLHx4eDBw8yd+5cVq5cabX+a9eu0bx5c3x9fa3KO3ToQJcuXXJ+g3PBtWvXaNy4MR07dmTcuHGW8nPnzvHmm28yefJkfvzxRwYOHEidOnVo1KgRrq6uODs7k5KSgqenJ6NGjaJatWp5uBVZt2XLFr7++msMBgNms5m2bdvSu3dvunXrZtnGNCNHjrTsZw9vt9lsxsnJieHDh+Pn58ecOXMIDQ3Fy8sLgMTERJo3b86QIUMAePDgAdOnT2fv3r24ubnh4eHBoEGDqFu3rmU9ERERFCpUCICEhAQKFy7M5MmTOX78OCtWrADg0qVLPP/88zg7O1OjRg1CQkJys+kUGzt2LEePHiUlJYW///7b8r3p3r073t7ezJ49m7i4OLRaLfXq1WPIkCG4ublZfc80Gg0pKSl4e3szefJkfHx8srz+PAnssLAwmjVrRmhoqCWwASZMmGDZuU6dOkXv3r359ttvKVeuHN9//z2hoaEsXLgQb29vYmNj6dmzJ25ubrz99tt5sRm5buvWrbRo0YImTdI/veLw4cMMGzaMuXPn8sorrwDw7bff8v7777Nu3brHLtfb25sNGzbYo8p5pnDhwvz2228YjUZ0Oh0A4eHhFC1a1Ob7v/76a0qWLAnArl276NOnD5s3b87w/Y7i1q1bTJ06lbCwMIoUKcKDBw/o1q0bZcqUydLnH97u3377jSFDhrBnzx4A3nnnHUtnID4+nrZt2/LSSy/RuHFjgoKCqFy5Mj///DN6vZ6zZ8/St29fZsyYYfkODx482KoDMnHiRObMmcOsWbNo3749AI0aNbKqg6NL+0G5du0a3bt3t3xvDhw4wOjRo5kzZw5VqlQhOTmZKVOmMGDAAJYsWQKk/57NmDGD8ePHM2/evCyvP9eHRAwGAxs3bmTIkCGcPXuWv//+2+b7XnrpJVq0aMEPP/wAwIIFCxg5cqTleZKenp5MnTqVChUq5Frd81r//v0ZO3Ys9+7dS/fa/Pnz6d+/vyWsAbp06UJgYCDJycm5V0kHUaBAASpXrszvv/9uKdu3bx/+/v6ZfrZhw4ZUq1aNn376yZ5VzBF3794lJSWFxMREIHW7p0yZQrly5RQvq1atWsTExPDvv+mfkuLu7k6VKlW4cuUKhw4d4vr163zyySfo9XoAXnzxRfr378/8+fNtLjs5OZmoqChLj/tpM3/+fAYOHEiVKlUA0Ov1fPLJJ/zxxx8cOXLE5mdq1qzJlStXFK0n1wN7165dPPvss5QpU4YmTZoQGhqa4XvLly/P5cuXiYmJ4caNG7z88stWr/v6+qYre5rVrFmT5s2bM378+HSvHT9+nFq1aqUr79Wrl+VLlZHbt2/Ttm1bq/9FRkbmWL3zSosWLdi6dSsAJ0+epGLFijg7O2fps2n7nqOrVKkSjRs3pkmTJrz11lt8/vnnmEwmXnjhBcXL2rBhA6VLl7Z5VPHPP/9w+PBhqlevzqlTp6hatSoajfWscrVq1eLUqVOWv2fPnk2bNm1o0KABLVu2pESJEgwbNkz5RqrAqVOn0g2hOTs7W9rrUSkpKWzevJkaNWooWk+uD4mEhYXRqlUrAAIDAxk6dCgffvihzfdqNBpcXV3RalN/V57miQUf3fkhdXvTtj3NRx99RNu2bfn1118zXEZycrJlmOj+/ft88cUXj1330zgkAvD6668za9YsTCYTmzdvpkWLFoSHh2fps2n7nhqMHTuWAQMGsHfvXvbu3UuHDh2YPn16lvapvn37WsbuS5QowaxZsyyvhYaG8uuvv2IymdDpdAQFBfHqq69y/PhxjEZjumWnpKRY/Z02JHL58mV69uxJnTp18PCw/axCtdNoNBgMhnTlDx/dpnWM0sqrVavGxx9/rGg9uRrY0dHR7Nmzh9OnT7NixQrMZjOxsbFs27bN5vsjIyPx9fWlcOHClCpVitOnT1v1Ig8dOsSePXsYOnRobm1Cjjh8+DClSpXimWeewWw2o9PpKFiwILGxsVbvi46OTndo6+bmxqRJkxgyZAjNmjWzHGK+9NJLHD16lPLly6PX6y0B3K1bt3RfpPzCw8ODSpUqceTIESIiIvj444+zHNiRkZFW51cc1a5du4iPjycwMJD27dvTvn171qxZw9q1aylUqFC64Y3o6Gg8PT0tfz9u/PjhMeyHvfzyy6xcuZKUlBSrI5bjx4/z0ksvpXt/2bJlGTp0KKNGjWLr1q0ULFgwu5vrsKpVq8bx48epVKmSpSw5OZmzZ8/Su3dvIGc6Rrk6JLJx40b8/PzYs2cPO3bsYOfOnQQFBfH999+ne+/JkyfZunUrb731FpB6aD9lyhSioqIAiImJYcqUKdk69Mtr69ats/SQIyMjKVWqFHXr1iU8PJz4+HgAoqKi2LVrF35+fuk+nzY08vBw0qBBg5g3bx4nTpywlJ0/f56rV69aTrrlRy1atGDGjBlUrVoVJ6es9U927NjBuXPnaNGihZ1r9+RcXV2ZMWMG165dA1J70H/88QeVK1fGz8+P9evXW3p+ly5d4vTp01bnObKjZs2alCtXjkmTJlk6A6dPn2bBggUMGDDA5mdatWpFqVKlMhzjVrtBgwaxYMECzpw5A6QebUyYMIGyZcvy6quv5th6crWHHRYWZrksKE3nzp1ZtGgRHh4eBAcH4+7ujkajwc3NjZkzZ1p+/Tt16kRKSgo9e/ZEo9FgNpvp2LGjKq8Q6du3L8OHD2fVqlX4+Pgwa9YsChUqxPnz5+nQoQMajQatVsuwYcMoX768zWV89NFH7N692/J3zZo1mTlzJrNmzeLOnTuYzWYKFy7MiBEjqFmzJgcPHrSMQaZp3bo1ffv2tTpUS1OrVi2Cg4Pt0wC56PXXX2f06NF88MEHj31f2tAAQJEiRSz7pKPz8/Nj4MCBBAUFWcLztdde4/3330er1XL16lXatm2LVqvFxcWFGTNm5MiVL3PnzmXmzJm0atUKnU5HoUKF+Pzzz60uIXzU8OHD6dGjB507d6ZUqVJPXAdHUrNmTaZOncrEiRO5f/8+BoOBBg0aMH/+fJtDU9klT5wRQgiVkDsdhRBCJSSwhRBCJSSwhRBCJSSwhRBCJSSwhRBCJSSwhRBCJSSwhRBCJSSwhRBCJf4f35mYCh+9UcsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#merging each summoner 5 from both teams's position column with his mastery point column\n",
    "t2_s5_merge = pd.merge(column_t2_summoner5_role , column_t2_summoner5_mastery_points, right_index=True, left_index=True)\n",
    "t1_s5_merge = pd.merge(column_t1_summoner5_role , column_t1_summoner5_mastery_points, right_index=True, left_index=True)\n",
    "s5_result = pd.merge(t2_s5_merge, t1_s5_merge, right_index=True, left_index=True) \n",
    "\n",
    "#locating where t2 summoner 5 player has greater amount of mastery points\n",
    "t2_greater_mp = s5_result.loc[s5_result['t2_mast5'] > s5_result['t1_mast5']]\n",
    "\n",
    "#comparing with original data(winner)\n",
    "t2_greater_mp_winner = pd.merge(t2_greater_mp, column_winner, right_index=True, left_index=True)\n",
    "\n",
    "#Only when Team 2 are winners(no matter mastery points difference)\n",
    "only_t2_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t2\"]\n",
    "\n",
    "#Only when Team 1 are winners(no matter mastery points difference)\n",
    "only_t1_winner = t2_greater_mp_winner.loc[t2_greater_mp_winner['winner'] == \"t1\"]\n",
    "\n",
    "\n",
    "fig = plt.figure()\n",
    "ax1 = fig.add_subplot(111)\n",
    "#Team 2 wins when they have more mastery points\n",
    "ax1.scatter(only_t2_winner['t2_summoner5_role'],only_t2_winner['t2_mast5'], c='r')\n",
    "#Team 1 wins despite that they have more mastery points\n",
    "ax1.scatter(only_t1_winner['t1_summoner5_role'],only_t1_winner['t1_mast5'], c='b')\n",
    "leg = plt.legend(loc='upper left', labels=[\"Mastery points T2 > Mastery points T1 [winner = T2]\", \"Mastery points T2 > Mastery points T1 [winner = T1]\"])\n",
    "leg.get_frame().set_alpha(0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "48d216b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "#merging the required features into the desired dataset\n",
    "X = pd.merge(s1_result, s2_result, right_index=True, left_index=True)\n",
    "X = pd.merge(X, s3_result, right_index=True, left_index=True)\n",
    "X = pd.merge(X, s4_result, right_index=True, left_index=True)\n",
    "X = pd.merge(X, s5_result, right_index=True, left_index=True)\n",
    "\n",
    "y = column_winner"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "03377670",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Indexing non numeric values\n",
    "#Team: 1; Position: TOP\n",
    "X['t1_summoner1_role'] = X['t1_summoner1_role'].replace('TOP', 0)\n",
    "X['t1_summoner2_role'] = X['t1_summoner2_role'].replace('TOP', 0)\n",
    "X['t1_summoner3_role'] = X['t1_summoner3_role'].replace('TOP', 0)\n",
    "X['t1_summoner4_role'] = X['t1_summoner4_role'].replace('TOP', 0)\n",
    "X['t1_summoner5_role'] = X['t1_summoner5_role'].replace('TOP', 0)\n",
    "\n",
    "#Team: 2; Position: TOP\n",
    "X['t2_summoner1_role'] = X['t2_summoner1_role'].replace('TOP', 0)\n",
    "X['t2_summoner2_role'] = X['t2_summoner2_role'].replace('TOP', 0)\n",
    "X['t2_summoner3_role'] = X['t2_summoner3_role'].replace('TOP', 0)\n",
    "X['t2_summoner4_role'] = X['t2_summoner4_role'].replace('TOP', 0)\n",
    "X['t2_summoner5_role'] = X['t2_summoner5_role'].replace('TOP', 0)\n",
    "\n",
    "\n",
    "\n",
    "#Team: 1; Position: JUNGLE\n",
    "X['t1_summoner1_role'] = X['t1_summoner1_role'].replace('JUNGLE', 1)\n",
    "X['t1_summoner2_role'] = X['t1_summoner2_role'].replace('JUNGLE', 1)\n",
    "X['t1_summoner3_role'] = X['t1_summoner3_role'].replace('JUNGLE', 1)\n",
    "X['t1_summoner4_role'] = X['t1_summoner4_role'].replace('JUNGLE', 1)\n",
    "X['t1_summoner5_role'] = X['t1_summoner5_role'].replace('JUNGLE', 1)\n",
    "\n",
    "#Team: 2; Position: JUNGLE\n",
    "X['t2_summoner1_role'] = X['t2_summoner1_role'].replace('JUNGLE', 1)\n",
    "X['t2_summoner2_role'] = X['t2_summoner2_role'].replace('JUNGLE', 1)\n",
    "X['t2_summoner3_role'] = X['t2_summoner3_role'].replace('JUNGLE', 1)\n",
    "X['t2_summoner4_role'] = X['t2_summoner4_role'].replace('JUNGLE', 1)\n",
    "X['t2_summoner5_role'] = X['t2_summoner5_role'].replace('JUNGLE', 1)\n",
    "\n",
    "\n",
    "\n",
    "#Team: 1; Position: MID\n",
    "X['t1_summoner1_role'] = X['t1_summoner1_role'].replace('MID', 2)\n",
    "X['t1_summoner2_role'] = X['t1_summoner2_role'].replace('MID', 2)\n",
    "X['t1_summoner3_role'] = X['t1_summoner3_role'].replace('MID', 2)\n",
    "X['t1_summoner4_role'] = X['t1_summoner4_role'].replace('MID', 2)\n",
    "X['t1_summoner5_role'] = X['t1_summoner5_role'].replace('MID', 2)\n",
    "\n",
    "#Team: 2; Position: MID\n",
    "X['t2_summoner1_role'] = X['t2_summoner1_role'].replace('MID', 2)\n",
    "X['t2_summoner2_role'] = X['t2_summoner2_role'].replace('MID', 2)\n",
    "X['t2_summoner3_role'] = X['t2_summoner3_role'].replace('MID', 2)\n",
    "X['t2_summoner4_role'] = X['t2_summoner4_role'].replace('MID', 2)\n",
    "X['t2_summoner5_role'] = X['t2_summoner5_role'].replace('MID', 2)\n",
    "\n",
    "\n",
    "\n",
    "#Team: 1; Position: ADC\n",
    "X['t1_summoner1_role'] = X['t1_summoner1_role'].replace('ADC', 3)\n",
    "X['t1_summoner2_role'] = X['t1_summoner2_role'].replace('ADC', 3)\n",
    "X['t1_summoner3_role'] = X['t1_summoner3_role'].replace('ADC', 3)\n",
    "X['t1_summoner4_role'] = X['t1_summoner4_role'].replace('ADC', 3)\n",
    "X['t1_summoner5_role'] = X['t1_summoner5_role'].replace('ADC', 3)\n",
    "\n",
    "#Team: 2; Position: ADC\n",
    "X['t2_summoner1_role'] = X['t2_summoner1_role'].replace('ADC', 3)\n",
    "X['t2_summoner2_role'] = X['t2_summoner2_role'].replace('ADC', 3)\n",
    "X['t2_summoner3_role'] = X['t2_summoner3_role'].replace('ADC', 3)\n",
    "X['t2_summoner4_role'] = X['t2_summoner4_role'].replace('ADC', 3)\n",
    "X['t2_summoner5_role'] = X['t2_summoner5_role'].replace('ADC', 3)\n",
    "\n",
    "\n",
    "\n",
    "#Team: 1; Position: SUPPORT\n",
    "X['t1_summoner1_role'] = X['t1_summoner1_role'].replace('SUPPORT', 4)\n",
    "X['t1_summoner2_role'] = X['t1_summoner2_role'].replace('SUPPORT', 4)\n",
    "X['t1_summoner3_role'] = X['t1_summoner3_role'].replace('SUPPORT', 4)\n",
    "X['t1_summoner4_role'] = X['t1_summoner4_role'].replace('SUPPORT', 4)\n",
    "X['t1_summoner5_role'] = X['t1_summoner5_role'].replace('SUPPORT', 4)\n",
    "\n",
    "#Team: 2; Position: SUPPORT\n",
    "X['t2_summoner1_role'] = X['t2_summoner1_role'].replace('SUPPORT', 4)\n",
    "X['t2_summoner2_role'] = X['t2_summoner2_role'].replace('SUPPORT', 4)\n",
    "X['t2_summoner3_role'] = X['t2_summoner3_role'].replace('SUPPORT', 4)\n",
    "X['t2_summoner4_role'] = X['t2_summoner4_role'].replace('SUPPORT', 4)\n",
    "X['t2_summoner5_role'] = X['t2_summoner5_role'].replace('SUPPORT', 4)\n",
    "\n",
    "\n",
    "\n",
    "#Winner column\n",
    "y = y.replace('t1', 1)\n",
    "y = y.replace('t2', 2)\n",
    "\n",
    "X_c = X\n",
    "\n",
    "#Converting String values to Integers\n",
    "imp = SimpleImputer(strategy=\"most_frequent\")\n",
    "X = imp.fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "6e0d5726",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6363636363636364"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#RandomForestClassifier\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "rf = RandomForestClassifier(random_state=200,max_depth=100,min_samples_split=30,n_estimators=120)\n",
    "rf.fit(X_train,y_train)\n",
    "\n",
    "#Success rate(predict who wins the game)\n",
    "rf.score(X_test,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "e702bb65",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "t2_mast4       0.094\n",
      "t2_mast3       0.093\n",
      "t1_mast5       0.090\n",
      "t2_mast5       0.087\n",
      "t1_mast1       0.087\n",
      "t2_mast1       0.085\n",
      "t1_mast4       0.083\n",
      "t1_mast3       0.082\n",
      "t2_mast2       0.080\n",
      "t1_mast2       0.079\n",
      "t2_summoner4_role     0.015\n",
      "t1_summoner3_role     0.015\n",
      "t1_summoner4_role     0.015\n",
      "t1_summoner1_role     0.014\n",
      "t1_summoner5_role     0.014\n",
      "t2_summoner5_role     0.014\n",
      "t2_summoner1_role     0.014\n",
      "t2_summoner3_role     0.014\n",
      "t1_summoner2_role     0.013\n",
      "t2_summoner2_role     0.013\n"
     ]
    }
   ],
   "source": [
    "#What have the model learned from the dataset\n",
    "#How much each feature(column) determines the output\n",
    "for col,val in sorted(\n",
    "           zip(X_c.columns, rf.feature_importances_),\n",
    "           key=lambda x: x[1],\n",
    "           reverse=True,\n",
    "   )[:]:\n",
    "           print(f\"{col:10}{val:10.3f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "2914fae9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5471458773784356"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#LogisticRegression classifier: Attempted to improve the success rate\n",
    "lr = LogisticRegression(random_state=42,tol=0.001)\n",
    "lr.fit(X_train, y_train)\n",
    "lr.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "e24d0f06",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5217758985200845"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#KNeighbors classifier : Another attempt to increase the success rate\n",
    "knc = KNeighborsClassifier(weights='distance')\n",
    "knc.fit(X_train, y_train)\n",
    "knc.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "0feb38c4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-104164872460033.97"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Kmeans cluster : A third attempt to increase the success rate\n",
    "kmeans = KMeans(n_clusters=118,random_state=42,max_iter=600)\n",
    "kmeans.fit(X_train, y_train)\n",
    "kmeans.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bcca2046",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}