milion_song_classificaiton.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "pycharm": {
     "is_executing": false
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import sklearn\n",
    "import os\n",
    "import pickle\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "from scipy.stats import norm\n",
    "from scipy import stats\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from tensorflow.keras import layers, regularizers\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this file only the final model is shown, other models are described in assignment report."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C:\\Users\\USAID\\Desktop\\TU GRAZ\\WS2020\\DL\\KU\\ASS3\\song-decades-train.pickle\n",
      "C:\\Users\\USAID\\Desktop\\TU GRAZ\\WS2020\\DL\\KU\\ASS3\\song-decades-test.pickle\n"
     ]
    }
   ],
   "source": [
    "path_train = \"\"\n",
    "path_test = \"\"\n",
    "for dirname, _, filenames in os.walk(os.getcwd()):\n",
    "    for filename in filenames:\n",
    "        if filename == \"song-decades-train.pickle\":\n",
    "            path_train = os.path.join(dirname, filename)\n",
    "        if filename == \"song-decades-test.pickle\":\n",
    "            path_test = os.path.join(dirname, filename)\n",
    "\n",
    "print(path_train)\n",
    "print(path_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a"
   },
   "outputs": [],
   "source": [
    "dict1 = pickle.load(open(path_train, 'rb'))\n",
    "x_train, y_train = dict1['x_train'], dict1['y_train']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "dict2 = pickle.load(open(path_test, 'rb'))\n",
    "x_test, y_test = dict2['x_test'], dict2['y_test']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get pd.DataFrame of training dataset to better inspect features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>80</th>\n",
       "      <th>81</th>\n",
       "      <th>82</th>\n",
       "      <th>83</th>\n",
       "      <th>84</th>\n",
       "      <th>85</th>\n",
       "      <th>86</th>\n",
       "      <th>87</th>\n",
       "      <th>88</th>\n",
       "      <th>89</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>26.875080</td>\n",
       "      <td>1243.586426</td>\n",
       "      <td>2178.185791</td>\n",
       "      <td>1184.748535</td>\n",
       "      <td>945.931641</td>\n",
       "      <td>863.210815</td>\n",
       "      <td>538.291138</td>\n",
       "      <td>315.642029</td>\n",
       "      <td>295.795990</td>\n",
       "      <td>287.132355</td>\n",
       "      <td>...</td>\n",
       "      <td>11.103380</td>\n",
       "      <td>10.297690</td>\n",
       "      <td>1.499990</td>\n",
       "      <td>-12.45978</td>\n",
       "      <td>-6.92671</td>\n",
       "      <td>-4.70198</td>\n",
       "      <td>12.03861</td>\n",
       "      <td>3.30228</td>\n",
       "      <td>5.93018</td>\n",
       "      <td>7.82323</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20.692801</td>\n",
       "      <td>1820.812378</td>\n",
       "      <td>2511.977295</td>\n",
       "      <td>1460.752075</td>\n",
       "      <td>1270.285889</td>\n",
       "      <td>942.201050</td>\n",
       "      <td>1194.021240</td>\n",
       "      <td>659.212769</td>\n",
       "      <td>453.929443</td>\n",
       "      <td>408.896027</td>\n",
       "      <td>...</td>\n",
       "      <td>30.698589</td>\n",
       "      <td>10.317030</td>\n",
       "      <td>-8.083160</td>\n",
       "      <td>8.30787</td>\n",
       "      <td>-11.70716</td>\n",
       "      <td>-0.48728</td>\n",
       "      <td>6.65790</td>\n",
       "      <td>13.57977</td>\n",
       "      <td>-4.43494</td>\n",
       "      <td>1.68251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>69.203201</td>\n",
       "      <td>3601.314453</td>\n",
       "      <td>5130.273926</td>\n",
       "      <td>2433.981689</td>\n",
       "      <td>1348.701416</td>\n",
       "      <td>702.506226</td>\n",
       "      <td>804.283630</td>\n",
       "      <td>334.737091</td>\n",
       "      <td>806.163147</td>\n",
       "      <td>211.870300</td>\n",
       "      <td>...</td>\n",
       "      <td>0.314660</td>\n",
       "      <td>-32.814411</td>\n",
       "      <td>4.651490</td>\n",
       "      <td>-12.62334</td>\n",
       "      <td>-30.28458</td>\n",
       "      <td>-5.80660</td>\n",
       "      <td>8.66985</td>\n",
       "      <td>-3.31120</td>\n",
       "      <td>1.52785</td>\n",
       "      <td>18.10549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>20.014610</td>\n",
       "      <td>1155.213257</td>\n",
       "      <td>1404.434204</td>\n",
       "      <td>1346.863525</td>\n",
       "      <td>919.088318</td>\n",
       "      <td>624.909485</td>\n",
       "      <td>562.027527</td>\n",
       "      <td>425.893799</td>\n",
       "      <td>268.491577</td>\n",
       "      <td>271.148346</td>\n",
       "      <td>...</td>\n",
       "      <td>21.436501</td>\n",
       "      <td>-1.225280</td>\n",
       "      <td>-11.430400</td>\n",
       "      <td>-12.31766</td>\n",
       "      <td>-0.44003</td>\n",
       "      <td>-0.05229</td>\n",
       "      <td>16.70611</td>\n",
       "      <td>10.61933</td>\n",
       "      <td>-3.15288</td>\n",
       "      <td>-2.11476</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>51.705200</td>\n",
       "      <td>3586.935547</td>\n",
       "      <td>1946.638306</td>\n",
       "      <td>1884.391968</td>\n",
       "      <td>1166.562012</td>\n",
       "      <td>888.135925</td>\n",
       "      <td>602.705933</td>\n",
       "      <td>412.445007</td>\n",
       "      <td>467.966064</td>\n",
       "      <td>325.744019</td>\n",
       "      <td>...</td>\n",
       "      <td>30.519779</td>\n",
       "      <td>-2.196240</td>\n",
       "      <td>-22.376381</td>\n",
       "      <td>4.19173</td>\n",
       "      <td>-13.30331</td>\n",
       "      <td>0.13214</td>\n",
       "      <td>9.47696</td>\n",
       "      <td>-1.21018</td>\n",
       "      <td>-4.49580</td>\n",
       "      <td>6.40991</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 90 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          0            1            2            3            4           5   \\\n",
       "0  26.875080  1243.586426  2178.185791  1184.748535   945.931641  863.210815   \n",
       "1  20.692801  1820.812378  2511.977295  1460.752075  1270.285889  942.201050   \n",
       "2  69.203201  3601.314453  5130.273926  2433.981689  1348.701416  702.506226   \n",
       "3  20.014610  1155.213257  1404.434204  1346.863525   919.088318  624.909485   \n",
       "4  51.705200  3586.935547  1946.638306  1884.391968  1166.562012  888.135925   \n",
       "\n",
       "            6           7           8           9   ...         80         81  \\\n",
       "0   538.291138  315.642029  295.795990  287.132355  ...  11.103380  10.297690   \n",
       "1  1194.021240  659.212769  453.929443  408.896027  ...  30.698589  10.317030   \n",
       "2   804.283630  334.737091  806.163147  211.870300  ...   0.314660 -32.814411   \n",
       "3   562.027527  425.893799  268.491577  271.148346  ...  21.436501  -1.225280   \n",
       "4   602.705933  412.445007  467.966064  325.744019  ...  30.519779  -2.196240   \n",
       "\n",
       "          82        83        84       85        86        87       88  \\\n",
       "0   1.499990 -12.45978  -6.92671 -4.70198  12.03861   3.30228  5.93018   \n",
       "1  -8.083160   8.30787 -11.70716 -0.48728   6.65790  13.57977 -4.43494   \n",
       "2   4.651490 -12.62334 -30.28458 -5.80660   8.66985  -3.31120  1.52785   \n",
       "3 -11.430400 -12.31766  -0.44003 -0.05229  16.70611  10.61933 -3.15288   \n",
       "4 -22.376381   4.19173 -13.30331  0.13214   9.47696  -1.21018 -4.49580   \n",
       "\n",
       "         89  \n",
       "0   7.82323  \n",
       "1   1.68251  \n",
       "2  18.10549  \n",
       "3  -2.11476  \n",
       "4   6.40991  \n",
       "\n",
       "[5 rows x 90 columns]"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train = pd.DataFrame(data=x_train)\n",
    "df_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Features are inspected using pd.DataFrame.describe() where it shows their max and min values, mean, standard deviation..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>80</th>\n",
       "      <th>81</th>\n",
       "      <th>82</th>\n",
       "      <th>83</th>\n",
       "      <th>84</th>\n",
       "      <th>85</th>\n",
       "      <th>86</th>\n",
       "      <th>87</th>\n",
       "      <th>88</th>\n",
       "      <th>89</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "      <td>40000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>33.514835</td>\n",
       "      <td>2278.871582</td>\n",
       "      <td>1946.816772</td>\n",
       "      <td>1441.489746</td>\n",
       "      <td>890.089417</td>\n",
       "      <td>860.854126</td>\n",
       "      <td>588.293213</td>\n",
       "      <td>487.343506</td>\n",
       "      <td>392.929962</td>\n",
       "      <td>317.177490</td>\n",
       "      <td>...</td>\n",
       "      <td>13.741096</td>\n",
       "      <td>1.729082</td>\n",
       "      <td>-7.857687</td>\n",
       "      <td>-7.844602</td>\n",
       "      <td>-2.702416</td>\n",
       "      <td>-1.560551</td>\n",
       "      <td>3.977398</td>\n",
       "      <td>1.463833</td>\n",
       "      <td>-0.106070</td>\n",
       "      <td>3.140625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>21.122564</td>\n",
       "      <td>1543.175659</td>\n",
       "      <td>1168.658569</td>\n",
       "      <td>1015.415527</td>\n",
       "      <td>440.580597</td>\n",
       "      <td>540.267334</td>\n",
       "      <td>295.005493</td>\n",
       "      <td>269.585175</td>\n",
       "      <td>198.789169</td>\n",
       "      <td>151.849960</td>\n",
       "      <td>...</td>\n",
       "      <td>34.148834</td>\n",
       "      <td>15.420406</td>\n",
       "      <td>22.621651</td>\n",
       "      <td>12.446113</td>\n",
       "      <td>14.373337</td>\n",
       "      <td>7.600089</td>\n",
       "      <td>10.610641</td>\n",
       "      <td>6.272420</td>\n",
       "      <td>4.213229</td>\n",
       "      <td>8.278407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.036930</td>\n",
       "      <td>112.444641</td>\n",
       "      <td>121.719467</td>\n",
       "      <td>89.312119</td>\n",
       "      <td>83.807739</td>\n",
       "      <td>27.581730</td>\n",
       "      <td>47.160439</td>\n",
       "      <td>36.352909</td>\n",
       "      <td>37.709881</td>\n",
       "      <td>36.054668</td>\n",
       "      <td>...</td>\n",
       "      <td>-221.158722</td>\n",
       "      <td>-104.084618</td>\n",
       "      <td>-165.326096</td>\n",
       "      <td>-69.499222</td>\n",
       "      <td>-117.810760</td>\n",
       "      <td>-48.965302</td>\n",
       "      <td>-87.522530</td>\n",
       "      <td>-37.189442</td>\n",
       "      <td>-33.086590</td>\n",
       "      <td>-60.526249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>18.622054</td>\n",
       "      <td>1290.966553</td>\n",
       "      <td>1160.160431</td>\n",
       "      <td>812.603470</td>\n",
       "      <td>611.054443</td>\n",
       "      <td>498.419724</td>\n",
       "      <td>394.283630</td>\n",
       "      <td>317.546623</td>\n",
       "      <td>265.666290</td>\n",
       "      <td>217.493771</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.781860</td>\n",
       "      <td>-7.612370</td>\n",
       "      <td>-21.451480</td>\n",
       "      <td>-16.386513</td>\n",
       "      <td>-11.086965</td>\n",
       "      <td>-6.046705</td>\n",
       "      <td>-2.317632</td>\n",
       "      <td>-2.656647</td>\n",
       "      <td>-2.535740</td>\n",
       "      <td>-1.944260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>29.377995</td>\n",
       "      <td>1902.085205</td>\n",
       "      <td>1688.840210</td>\n",
       "      <td>1204.094482</td>\n",
       "      <td>803.469757</td>\n",
       "      <td>736.162262</td>\n",
       "      <td>532.649933</td>\n",
       "      <td>431.805389</td>\n",
       "      <td>352.099960</td>\n",
       "      <td>289.435516</td>\n",
       "      <td>...</td>\n",
       "      <td>15.068390</td>\n",
       "      <td>0.519170</td>\n",
       "      <td>-7.087320</td>\n",
       "      <td>-9.199850</td>\n",
       "      <td>-2.331365</td>\n",
       "      <td>-1.494425</td>\n",
       "      <td>3.968800</td>\n",
       "      <td>1.265975</td>\n",
       "      <td>0.009405</td>\n",
       "      <td>2.954020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>42.972118</td>\n",
       "      <td>2796.437317</td>\n",
       "      <td>2430.511902</td>\n",
       "      <td>1783.188446</td>\n",
       "      <td>1065.835754</td>\n",
       "      <td>1073.828400</td>\n",
       "      <td>713.739182</td>\n",
       "      <td>584.575684</td>\n",
       "      <td>469.336227</td>\n",
       "      <td>381.478683</td>\n",
       "      <td>...</td>\n",
       "      <td>34.436212</td>\n",
       "      <td>9.404370</td>\n",
       "      <td>6.272870</td>\n",
       "      <td>-0.968955</td>\n",
       "      <td>6.270848</td>\n",
       "      <td>2.962065</td>\n",
       "      <td>10.290805</td>\n",
       "      <td>5.381975</td>\n",
       "      <td>2.435333</td>\n",
       "      <td>7.980932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>282.059784</td>\n",
       "      <td>30846.714844</td>\n",
       "      <td>16037.863281</td>\n",
       "      <td>31849.486328</td>\n",
       "      <td>19865.931641</td>\n",
       "      <td>10921.137695</td>\n",
       "      <td>10667.728516</td>\n",
       "      <td>7355.391113</td>\n",
       "      <td>4447.481445</td>\n",
       "      <td>3721.873291</td>\n",
       "      <td>...</td>\n",
       "      <td>291.865997</td>\n",
       "      <td>335.771820</td>\n",
       "      <td>204.414627</td>\n",
       "      <td>83.948952</td>\n",
       "      <td>133.663635</td>\n",
       "      <td>126.741272</td>\n",
       "      <td>77.996002</td>\n",
       "      <td>44.710110</td>\n",
       "      <td>88.020821</td>\n",
       "      <td>55.956692</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 90 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 0             1             2             3             4   \\\n",
       "count  40000.000000  40000.000000  40000.000000  40000.000000  40000.000000   \n",
       "mean      33.514835   2278.871582   1946.816772   1441.489746    890.089417   \n",
       "std       21.122564   1543.175659   1168.658569   1015.415527    440.580597   \n",
       "min        1.036930    112.444641    121.719467     89.312119     83.807739   \n",
       "25%       18.622054   1290.966553   1160.160431    812.603470    611.054443   \n",
       "50%       29.377995   1902.085205   1688.840210   1204.094482    803.469757   \n",
       "75%       42.972118   2796.437317   2430.511902   1783.188446   1065.835754   \n",
       "max      282.059784  30846.714844  16037.863281  31849.486328  19865.931641   \n",
       "\n",
       "                 5             6             7             8             9   \\\n",
       "count  40000.000000  40000.000000  40000.000000  40000.000000  40000.000000   \n",
       "mean     860.854126    588.293213    487.343506    392.929962    317.177490   \n",
       "std      540.267334    295.005493    269.585175    198.789169    151.849960   \n",
       "min       27.581730     47.160439     36.352909     37.709881     36.054668   \n",
       "25%      498.419724    394.283630    317.546623    265.666290    217.493771   \n",
       "50%      736.162262    532.649933    431.805389    352.099960    289.435516   \n",
       "75%     1073.828400    713.739182    584.575684    469.336227    381.478683   \n",
       "max    10921.137695  10667.728516   7355.391113   4447.481445   3721.873291   \n",
       "\n",
       "       ...            80            81            82            83  \\\n",
       "count  ...  40000.000000  40000.000000  40000.000000  40000.000000   \n",
       "mean   ...     13.741096      1.729082     -7.857687     -7.844602   \n",
       "std    ...     34.148834     15.420406     22.621651     12.446113   \n",
       "min    ...   -221.158722   -104.084618   -165.326096    -69.499222   \n",
       "25%    ...     -5.781860     -7.612370    -21.451480    -16.386513   \n",
       "50%    ...     15.068390      0.519170     -7.087320     -9.199850   \n",
       "75%    ...     34.436212      9.404370      6.272870     -0.968955   \n",
       "max    ...    291.865997    335.771820    204.414627     83.948952   \n",
       "\n",
       "                 84            85            86            87            88  \\\n",
       "count  40000.000000  40000.000000  40000.000000  40000.000000  40000.000000   \n",
       "mean      -2.702416     -1.560551      3.977398      1.463833     -0.106070   \n",
       "std       14.373337      7.600089     10.610641      6.272420      4.213229   \n",
       "min     -117.810760    -48.965302    -87.522530    -37.189442    -33.086590   \n",
       "25%      -11.086965     -6.046705     -2.317632     -2.656647     -2.535740   \n",
       "50%       -2.331365     -1.494425      3.968800      1.265975      0.009405   \n",
       "75%        6.270848      2.962065     10.290805      5.381975      2.435333   \n",
       "max      133.663635    126.741272     77.996002     44.710110     88.020821   \n",
       "\n",
       "                 89  \n",
       "count  40000.000000  \n",
       "mean       3.140625  \n",
       "std        8.278407  \n",
       "min      -60.526249  \n",
       "25%       -1.944260  \n",
       "50%        2.954020  \n",
       "75%        7.980932  \n",
       "max       55.956692  \n",
       "\n",
       "[8 rows x 90 columns]"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#correlation matrix\n",
    "corrmat = df_train.corr()\n",
    "f, ax = plt.subplots(figsize=(7, 7))\n",
    "sns.heatmap(corrmat, vmax=.8, square=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "columns = df_train.columns[0:12]\n",
    "df_train_partial = df_train[columns]\n",
    "corrmat = df_train_partial.corr()\n",
    "f, ax = plt.subplots(figsize=(7, 7))\n",
    "sns.set(font_scale=0.75)\n",
    "sns.heatmap(corrmat, vmax=.8, square=True,annot=True, fmt='.2f');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Normal probability plot')"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIgAAAHxCAYAAAAGOYdtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdeXiU9b3//9esmcwkARICARIghC3sKC7YIhZbPWhLBa2iFBEpX0pbPRXrV/2ppz3WWqFaLLTuVZGKy89i5Xig2hZFpamSSgQDISEsCUICBkL22b9/hIyEbJNklizPx3V5Se65c8977oGL4ZX35/0x+P1+vwAAAAAAANBrGaNdAAAAAAAAAKKLgAgAAAAAAKCXIyACAAAAAADo5QiIAAAAAAAAejkCIgAAAAAAgF6OgAgAAAAAAKCXIyACoIULF+q+++5r9rFbbrlF99xzj44cOaIxY8YoOzs7qGu+//772r9/fyjLBAAACNqsWbP0rW99S7W1tU0ea+2zTzSNGTNGb731VkSfc9asWXriiSc6/P0ff/yxxowZo5KSkqCeY+3atfrWt74lSU0+X5aXl+uNN97ocC1tueeee3TLLbcEff7+/fv1/vvvh60eoKshIAIQlEGDBumjjz7S5MmT2zy3tLRUy5YtU1lZWQQqAwAAaF5RUZF++9vfRruMXu+NN95oNpg59/Plo48+GvGArDU/+tGPtHv37miXAUQMARGAoJhMJiUnJ8tisbR5rt/vj0BFAAAArUtLS9Of/vQnffrpp9EupVdLTEyU3W5vcvzcz5dd7TNkV6sHCDcCIgBBObcFOCcnR/Pnz9eUKVN00UUX6a677lJ5ebkkaebMmZKkm2++Wffcc48k6ejRo7rjjjs0ffp0TZ06VT/60Y9UXFwcuH51dbXuu+8+XXjhhbrwwgu1cuVKLVy4UGvXrpVU3468cOFC3X777TrvvPO0evVq+Xw+PfHEE7riiis0YcIETZs2TbfddptOnjwpqb7leeLEifrggw905ZVXatKkSbrllltUUlKiBx98UOeff74uueQSPfPMMxG7jwAAIHLmzp2rqVOn6r777pPT6WzxvLY+p8yaNUsrV67UlVdeqYsvvli5ubmaNWuWXn75ZS1btkyTJk3S5Zdfrq1bt+rdd9/VFVdcoalTp2rp0qWBzyWS9M477+jaa6/VpEmTNHnyZM2fP1+7du0K6rWsXbtWt9xyix5//HFdcMEFuuiii/TQQw/J5XJJ+uqz2lNPPaXp06dr9uzZcrlcbb42qb77e/HixZo4caKuvPJKvffee4HH2vq8dfZrmzVrliZNmqQf/vCHOn78eKP719wytrM/X65du1ZvvPGGPvnkE40ZM0Z79uzRmDFjmtyfm266SQ8//HCz92jMmDF69dVXNW/ePE2aNEnz5s3Tjh07Wryn+fn5Wrp0qS644AJdeOGF+r//9/8GXtfChQtVVFSk3//+95o1a1aL1wB6EgIiAO3m9Xq1fPlyTZ8+XW+//baeeeYZ7d69WytXrpQkvfnmm5LqP8jcd999qqqq0o033qjTp0/rueee0/r161VZWanvf//7qqyslFS/Jjw7O1u///3vtW7dOu3Zs6fJX+iffPKJ0tLS9Oabb+q6667TCy+8oJdeekn333+/3nnnHT322GP697//rSeffDLwPW63W2vWrNGjjz4auO6cOXNkt9v1xhtv6IYbbtBjjz3GvCQAAHogg8Gghx9+WEePHg380OlcwXxOkaRXXnlFv/zlL/X0008rMzNTUv2SqNmzZ+vtt9/WmDFj9LOf/UzPPfecHnvsMT355JP67LPP9Mc//lGStGvXLv30pz/VvHnztHnzZq1fv16S9MADDwT9erKzs/Xpp59q/fr1WrVqlf7617/qoYceanTO//7v/+pPf/qTHn30UblcrqBe2+uvv64ZM2Zo06ZNuuaaa7R8+fJAMBPM5y1Jeumll/SrX/1Kr732miorK7VkyZJ2deDceuut+va3v62pU6fqo48+0pgxYzR27Fht2rQpcM6RI0f06aefau7cuS1eZ9WqVZo/f77efPNNTZgwQUuWLGkSiDVc68Ybb1SfPn308ssv64knnlBeXp5uvfVWeb1erV27VkOGDNGtt94a1rlIQFdCQARAkvSXv/xFU6dObfLfJ5980uTcyspKnTp1Sv3799eQIUM0efJk/eEPf9CiRYsk1bcRS1KfPn0UHx+vt956SxUVFfrtb3+r8ePHa8KECfrd736n06dPa9OmTSouLta7776rX/ziF7rwwguVmZmp1atXKyYmptHzGgwG3XbbbRo2bJjS0tKUnp6ulStX6tJLL9WQIUM0c+ZMzZgxQ/n5+YHv8fv9uuOOOzRx4kRNnTpVF198seLi4nTnnXcqPT1dy5YtkyQVFBSE69YCAIAoGj58uG677TY9//zz+vzzz5s83tbnlAazZs3ShRdeqMmTJ8toNAaOXXPNNRo6dKiuv/56VVdXa8WKFZo4caIuvvhiXXLJJYHPGBaLRT//+c+1YMECpaamatKkSfre977X6HNLW0wmk1avXq2xY8dq5syZ+ulPf6qNGzeqqqoqcM6CBQuUkZGh8ePHB/3a/uM//kO33nqr0tPTtXz5cl1wwQV66aWXJCmoz1uS9POf/1zTp09XZmamVq5cqfz8fGVlZQX92hwOh2w2mywWi5KTk2UymTR37lxt3rxZHo9HkrRp0yaNHj06ENA15/rrr9f111+vjIwM/fznP1dycrJef/31Judt2LBBCQkJ+vWvf63Ro0dr2rRpWr16tfbu3asPP/xQffv2lclkkt1uD3y2BXo6c7QLANA1fPOb39SKFSuaHL/77rubHOvbt68WL16sBx98UGvXrtXXvvY1feMb39Ds2bObvXZBQYFGjBihvn37Bo4lJiYqIyND+fn56t+/vyRpypQpjR4fPnx4o+skJyfLZrMFvp41a5Z27typ1atX6+DBgzpw4IAKCws1bdq0Rt83dOjQwK/tdrtSU1NlMBgkKXC9hvZsAADQ8yxevFjvvPOO7r33Xm3cuLHRY219TmmQlpbW5LrDhg0L/Do2NlZS488dNpstsAQ/MzNT8fHxevrpp7V//34dPnxYe/fulc/nC/p1jBgxQklJSYGvp0yZIrfbrYMHD6pfv35N6gz2tU2dOrXR80ycOFHbt2+XFPznrfPOOy/w69TUVCUmJio/P1+XXHJJ0K/vXHPmzNGjjz6q7du3a+bMmdq0aZPmz5/f6vdccMEFgV+bTCZNmDCh2RCuoKBAEydObDRfMyMjQ/369VN+fr4uu+yyDtcNdFd0EAGQJMXFxWnYsGFN/js7kDnb3XffrX/84x+67bbbVFNTo3vvvVfLly9v9txzO4Ea+Hw+WSwWmUwmSW0PAjy3lieffFK33nqrqqqqNGPGDK1cuVJz5sxp8n3nDtZu+KkfAADoHUwmkx5++GEdPHhQTz31VKPH2vqc0tp5ZnPTn7e39DnjX//6l2bPnq29e/dq4sSJWrFihe677772vIwmz+f1eps859l1BvvaGj6LNfD7/bJarZKC/7x17uv2+XyBa3RUYmKiZsyYobffflu7du1ScXGxvvOd77T6PefeI5/PF/jB4NmCvTdAb8K/kgC0W1FRUaBld8GCBXryySe1cuVKbdu2TWVlZU3+Eh45cqQOHDgQ+AmaJJ08eVIHDx5URkaGxowZI4PBoM8++yzweHl5uQ4fPtxqHevWrdPtt9+uBx54QN/73vc0fvx4HT58mB0nAABAE6NGjdLy5cv19NNPq6ioKHC8rc8pobJhwwZ97Wtf0+OPP66bb75ZF198sb744gtJwe+WdfDgQdXU1AS+/uyzz2Sz2TRixIhmzw/2te3Zs6fR9+3cuVMjR46UFPznrbOv0fCco0aNCup1NWguyJk3b57ef/99bdmyRTNmzGjUQdWc3NzcwK89Ho9yc3M1bty4JueNHDlSu3fvltvtDhzbv3+/Tp8+Hbg3zdUD9GQERADarV+/ftqyZYt+8YtfqLCwUIWFhdqyZYuGDh2qfv36yeFwSJL27dunU6dOac6cOUpMTNSKFSu0Z88e5ebmasWKFUpISNDVV1+ttLQ0XXHFFXrwwQe1Y8cO7du3T3fddZdqa2tb/Ys5MTFRH330kQoLC1VQUKAHH3xQO3fuZLkYAABo1rJlyzRy5EiVlJQEjrX1OSVUEhMTtW/fPuXk5Ki4uFjr16/XunXrJAW/1L2yslL33nuv9u/fr/fee0+PP/64brrppsDytnMF+9reeust/elPf9KBAwf02GOPaffu3frBD34QqDuYz1sPPPCAsrOztWvXLt11112aNm1ao+VewXA4HCotLVVxcXFg7tBll10mk8mkDRs2aN68eW1e4/nnn9eWLVtUWFio//qv/9Lp06d1ww03NDmvYVD3vffeq4KCAmVnZ+tnP/uZxo4dq+nTpwfqOXTokEpLS9v1OoDuioAIQLvFx8fr2WefVXFxsa6//npdd911cjqdeuaZZ2Q0GhUXF6eFCxfq0Ucf1f3336+YmBj98Y9/lNVq1YIFC7Ro0SLFx8fr5ZdfVkJCgiTpoYce0vjx47Vs2TItXLhQ48aN05AhQ1pt8V25cqUqKio0d+5cLV68WOXl5brzzju1f/9+1dbWRup2AACAbsJsNuvhhx9utAwpmM8poXD77bcrMzNTS5Ys0bXXXqt3331XjzzyiCRp9+7dQV0jNTU1MBD7gQce0A033KCf/exnLZ4f7GtbsmSJNm/erDlz5uj999/XU089FeiiCfbz1tKlS7VixQrdfPPNGjJkiNasWdPuezRv3jx5vV5dddVVgY4ki8Wiq6++WjabLai5QNdff72eeuopzZ07V0VFRXrppZc0cODAJuf1799fzz//vEpLS3Xttdfqxz/+sTIzM/XCCy8EPn/ecsst+uCDDzRnzpx2zYoCuiuDn7UYAKLM6XTqww8/1CWXXCK73S6pfnv6iy66SP/1X/+la665JsoVAgAARNfatWu1adMm/e1vf4t2KRF3++23a8CAAbr//vtbPW/MmDFatWqVvvvd70aoMqBnYRczAFFntVr1i1/8QpdeeqmWLl0qn88X+OnNpZdeGu3yAAAAEAUfffSR8vLytHXrVr311lvRLgfo8QiIAESdwWDQ008/rVWrVum6666Tz+fT1KlT9eKLLyoxMTHa5QEAACAKXn/9dW3fvl333HNPSAeGA2geS8wAAAAAAAB6OYZUAwAAAAAA9HIERAAAAAAAAL1cl51BdOpUtXy+6K5+S0qKU1lZVVRr6I64b+3HPesY7lvHcN86hvvWMS3dN6PRoH79HFGoCG2J1mcw/ox1X7x33RvvX/fFe9e9ReP9a+vzV5cNiHw+f9QDooY60H7ct/bjnnUM961juG8dw33rGO5b9xLNz2D8Xum+eO+6N96/7ov3rnvrau8fS8wAAAAAAAB6OQIiAAAAAACAXo6ACAAAAAAAoJcjIAIAAAAAAOjlCIgAAAAAAAB6OQIiAAAAAACAXo6ACAAAAAAAoJcjIAIAAAAAAOjlCIgAAAAAAAB6OQIiAAAAAACAXo6ACAAAAAAAoJcjIAIAAAAAAOjlCIgAAAAAAAB6OQIiAAAAAACAXs4c7QIAAAAAAAB6u6zcEm3cVqiyCqeSEmI0b2aGpo9PidjzExABAAAAAABEUVZuidZtyZPL45MklVU4tW5LniRFLCRiiRkAAAAAAEAUbdxWGAiHGrg8Pm3cVhixGgiIAAAAAAAAoqiswtmu4+HAErMuxOOTnG5Pi4/HWMwyE+kBAAAAANCjJCXENBsGJSXERKwGAqIuxOn2aMfe0hYfvyBzoMwxvGUAAAAAAPQk82ZmNJpBJElWs1HzZmZErAb6UboBj9en7LzjWrclTxU1rmiXAwAAAAAAQmj6+BQtmj020DGUlBCjRbPHsosZvvLl6Vp9tKtEFdUumYwGFRSX6wffHqfx6YnRLg0AAAAAAITI9PEpEQ2EzkUHURfl8/n12f4vteVfRfJ4ffrWBan62U1T5Yi16LHXcvTa1gK5z5lwDgAAAAAA0BF0EHVBdS6Ptv77C315uk7pg+J10biBslpMSk2O0wOLpun1rfv1zifF2nvolG67dpKS+tiiXTIAAAAAAOjG6CDqgnYXnlRZRZ1mTB6kGZMHy2oxSZIMRoM8Pr/mXZah/zNnvEpO1mjjhwdU7fQE/qOpCAAAAAAAtBcdRF2M0+VVwZFyDU+JV/qghMaPub36LP9E4OtB/R3amX9CGUMSZDQYJLHTGQAAAAAAaD+ShC5mX9Epebx+TRiR1Oa5QwfE6XBJpb4sr9OAfrERqA4AAAAAAAQrK7dEG7cVqqzCqaSEGM2bmRHVQdStYYlZF+Jye7X3cLmGJDvULz6mzfOHJDtkMEjFx6siUB0AAAAAAAhWVm6J1m3JU1mFU5JUVuHUui15ysotiXJlzSMg6kL+lVsqp9urCUFuYW+1mJSSaFdxaWWYKwMAAAAAAO2xcVuhXOcMCnZ5fNq4rTBKFbWOgKiL8Pp8+se/i5Xc19au5WJpA+NUUePW6SpnGKsDAAAAAADt0dA5FOzxaCMg6iJ25B3XyQqnxqcnynBm4HQw0gbESZKKWGYGAAAAAECXkZTQ/OiYlo5HGwFRF+D3+7XlX0VKSbQHAp9gOWwWJSXYVFxKQAQAAAAAQFcxb2aGrObGsYvVbNS8mRlRqqh1BERdwOcHT6r4eJUun5baru6hBmkD4/Tl6TrV1HnCUB0AAAAAAGiv6eNTtGj22EDHUFJCjBbNHttldzFjm/suYMu/DqtffIymjR2gnfkn2v39aQPilFPwpY6coIsIAAAAAICuYvr4lC4bCJ2LDqIoq6nzKK+oXJdOHiyzqWNvR984q+JiLSwzAwAAAAAAHUJAFGUNXT/pg+I7fA2DwaChA+N0rKxGdS6WmQEAAAAAgPYhIIqy4jO7j6Umt2849bnSBsTJ5/dr76FToSgLAAAAAAD0IgREUVZ8vEoOm1n94ju3zV1yv1jFWEzaVVgWosoAAAAAAEBvQUAUZcXHq5Q2IK5Du5edzWgwKHWAQ7kHT8rj9YWoOgAAAAAA0BsQEEWRz+fXF19WKW1Ax+cPnW1IcpxqnZ7AsjUAAAAAAIBgEBBF0fHyWrncPqUOcITkev3i6pepHf2yOiTXAwAAAAAAvQMBURQ1dPoMDVEHUbzdIrPJQEAEAAAAAADahYAoioqPV8poMGhwf3tIrmc0GjSgn52ACAAAAAAAtAsBURQVl1ZpUJJdFrMpZNdMSbTraBkBEQAAAAAACB4BURQdOVG/g1koDUqy68vyOjnd3pBeFwAAAAAA9FwERFFSXedWWYVTqSEOiFKS7PJLKimrCel1AQAAAABAz2WOdgG91ZEzA6pD3UGUkli/I9rRL6s1LCU0w68BAAAAAEDzsnJLtHFbocoqnEpKiNG8mRmaPj4l2mW1GwFRlBSFKSBK7muTyWjQFwyqBgAAAAAgrLJyS7RuS55cHp8kqazCqXVb8iSp24VELDGLEI9PqnZ6Av8dKqlUXKxFZrMxcMzn7/zzmEzG+kHVBEQAAAAAAITVxm2FgXCogcvj08ZthVGqqOPoIIoQp9ujHXtLA1/nF5Urzm5Rdt7xwLHJo5ND8lyD+jtUVFoZkmsBAAAAAIDmlVU423W8K6ODKAp8Pr/Kq5xKjI8Jy/WH9HfoxKlaudjJDAAAAACAsElKaP7f9S0d78oIiKKgosYlr8+vfmEKiAb3d9TvZHaSncwAAAAAAAiXeTMzZDU3jlasZqPmzcyIUkUdR0AUBacq61vNwhYQJdkliTlEAAAAAACE0fTxKVo0e2ygYygpIUaLZo/tdgOqJWYQRcWpSqcMBqlPnDUs1x+YaGcnMwAAeqjjx4/rrrvuktPp1KxZs9SnTx+98sorio+P18yZM/WDH/xAq1ev1scff6xRo0bpl7/8ZbRLBgCgR5s+PqVbBkLnIiCKglOVTvVxWGUyhqeBy2wyakC/WDqIAADogV5++WUtW7ZMl1xyiRYtWqQhQ4Zo7dq1SktLkyQdPXpURUVFevXVV7Vq1Srl5ORoypQpUa4aAAB0dSwxi4JTFU4lJtjC+hyD+zt0tIwZRAAA9DRLlizRhRdeKEny+/0qLS3VI488osWLF6uoqEh79uzReeedJ0maNm2acnJyolkuAADoJuggirA6l1c1Tk/Y5g81GNLfoU/zT8jt8cpiNoX1uQAAQOQkJCRIkt544w1lZmYqNTVVc+fO1ZEjR/Sb3/xGl19+uez2+nmEsbGxqqlp3w+MkpLiQl5zsJKT46P23Ogc3rvujfev++K969662vtHQBRh5WEeUN1gcH+H/H6p5GSt0gZE74MeAAAIvc2bN+vdd9/V73//e7lcLsXFxWns2LEqLy+Xw+FQaWmpJKm2tjYQFgWrrKxKPp8/HGW3Kjk5XidOVEb8edF5vHfdG+9f98V7Fz1ZuSXauK1QZRVOJSXEaN7MjHbPIIrG+2c0Glr9QRBLzCLsZGWdpAgEREkOSexkBgBAT7Nr1y699tprWrNmjaxWqxYtWiSXy6Xi4mIlJSUpMzNT2dnZkqTs7GxNmDAhyhUDANBzZOWWaN2WPJVV1Dd/lFU4tW5LnrJyS6JcWefRQRRh5ZUu2awmxcaE99YPTLTLaGAnMwAAeprnnntOZWVlWrp0qSRp4cKFWrBggWw2mx566CGlpqZq2LBhmj9/vtLT0zVt2rQoVwwAQM+xcVuhXB5fo2Muj08btxV2+53MCIgirMbplsNmCfvzWMz1O5kdIyACAKBHWbNmTZNj11xzTaOv77jjjkiVAwBAr9LQORTs8e6EJWYRVuv0KjYmMkOj63cyIyACAAAAACAUkhKaHxfT0vHuhIAowupcHtnCvLysweD+DpWerJX7nPY3AAAAAADQfvNmZshqbhylWM1GzZuZEaWKQoeAKIL8fr/qXF7FWiPVQWSXz+9X6an2bW8LAAAAAACamj4+RYtmjw10DCUlxGjR7LHdfv6QxAyiiHK6vfL7JZs1fLfdYDSo2umRJPVLsEmSDh6rDPxakmIsZpmJBgEAAAAAaLfp41N6RCB0LgKiCKp1eiUprDOInG6vPss/IUnyen0ySPr3vuPy+r5aZnZB5kCZI7TMDQAAAAAAdH30kURQ7ZnOnkjNIDKZjIqzW3S6qvtPUwcAAAAAAOFDQBRBda4zHUQRmkEkSX3jYlRe7YrY8wEAAAAAgO6HgCiC6iLcQSRJCQ6LKmvc8vv9EXtOAAAAAADQvRAQRVCtyyujwdBkS7xwirdb5fP5VV3nidhzAgAAAACA7oVJxRFU6/TIFmOSwWCI2HMm2K2SpIpql+JiLRF7XgAAAAAAeoqs3BJt3FaosgqnkhJiNG9mRo/byYwOogiqc3kUG8Yt7puT4DgTENUwhwgAAAAAgPbKyi3Rui15Kquo3wCqrMKpdVvylJVbEuXKQouAKIJqnV7ZwrjFfXNiY0wymwyqrHZH9HkBAAAAAOgJNm4rlMvja3TM5fFp47bCKFUUHgREERSNDiKDwaB4u5UOIgAAAAAAOqChcyjY490VAVGE+Px+1bm8io1wB5EkJdgtqmCrewAAAAAA2i0pIaZdx7srAqIIqa51y++XbBHuIJLq5xBV1brl87HVPQAAAAAA7TFvZkaT3citZqPmzcyIUkXhwS5mEVJZUz8DKNIziKT6re79fqmq1h0YWg0AAAAAANrWsFtZT9/FjIAoQirPzACK9AwiSUpw1G9vX1HjIiACAAAAACAIvWFr+7O1ucTM5XJp+fLlmj9/vl577bXA8VOnTunmm2/WDTfcoK1bt0qSXnzxRc2fP18LFy7UsWPHJEn33nuvbrzxRq1ZsyZML6F7aJgBFJUZRGdCIXYyAwAAAACgbb1la/uztRkQbd68WTNmzNCGDRv07rvvyuWqDzo2bNigJUuW6KWXXtL69etVW1urrVu36tVXX9WPf/xjrV+/XtnZ2UpMTNQrr7yigwcPqrS0NOwvqKv6aolZ5DuIYiwmWcxGdjIDAAAAACAIvWVr+7O1mVbk5ubquuuuk9Fo1KhRo1RYWKjMzEzl5uZq0aJFiomJUVxcnJxOZ6BLyOfzyWq16vPPP9f5558vSTrvvPO0e/duDRw4MKjCkpLiOvGyQic5OT4k16l1+2Q0GpTU1y6DwdDsORaLWfFxthav0dbjrZ3TLz5GNXUexcfZZLfHKDnR3r4X0E6hum+9CfesY7hvHcN96xjuW8dw3wAAQHfTW7a2P1ubAVF1dbXs9vowITY2VtXV1YHjDocjcLyurk4pKSmqqqrSmjVr9Pjjj+vPf/5zs98bjLKyqqjvupWcHK8TJypDcq2Tp2tls5pUVd3ybya326PKqroOP97aOQ6bWSfK61RZVaeaGqdOeL3BF99OobxvvQX3rGO4bx3DfesY7lvHtHTfjEZDl/lhEAAAwLmSEmKaDYN62tb2Z2tziZndbldtba0kqba2NhAK2e121dTUBI7b7XZVVVVp+fLl+slPfqKUlBQ5HI5mv7c3qqx2RWVAdYN4e/1W916vr+2TAQAAAADoxXrL1vZnazMgGj9+vHbs2CG/36+9e/dqxIgRgePZ2dlyOp0qLy9XQkKC7rnnHt1yyy36+te/LkkaN26cduzYIUn697//rXHjxoXxpXRtlTWuqGxx3yAwqLqWQdUAAAAAALRm+vgULZo9NtAxlJQQo0Wzx/boXczabGm56qqrtGLFCm3cuFHXXnut1q1bp29+85u66aabdOedd+p3v/udfvSjHyk/P1/bt2/X6dOn9eKLL2rGjBlaunSp3nzzTd1www26+OKLNXjw4Ei8pi6posatAX1jo/b8CfYzW91XM6gaAAAAAIC2TB+f0qMDoXO1GRDFxMToD3/4Q7OPvfDCC42+3rlzZ5Nzfv3rX3ewtJ7D5/erqsaloQOjN2shvqGDqIYOIgAAAAAA0Fj0huL0IlW1bvn8iuoMohiLSTEWEx1EAAAAAAC0Iiu3RBu3FaqswqmkhBjNm5nRKzqJCIgioCGUieYMIklKcFhUUUNABAAAAADAubJyS6Bij74AACAASURBVLThb/tUXffVrt9lFU6t25InST0+JGpzSDU6ryEgimYHkSQl2K2qrGaJGQAAAAAAZ8vKLdG6LXmNwqEGLo9PG7cVRqGqyCIgioCu0kEU77CqxumR0930NzwAAAAAAL3Vxm2Fcnl8LT5eVuGMYDXRQUAUAae7TAdR/U5mJ8pro1oHAAAAAABdSVsBUMN29z0ZAVEEVFS7ZDIaZLVE93Y37GR24hQBEQAAAAAADVoLgKxmo+bNzIhgNdFBQBQBFdUuxdstMhgMUa0jwV4fEB2ngwgAAAAAgIB5MzNkNTeNSOJizVo0e2yPH1AtsYtZRJyucSn+TDgTTRazUbExJjqIAAAAAAA4S0MA1Bu3t29AQBQBFVWuwPKuaEuwW5lBBAAAAADAOaaPT+lVgdC5WGIWAadrXIEB0dEW77DqOB1EAAAAAADgLAREYebz+1VZ7e4SS8yk+p3MqmrdqqnzRLsUAAAAAADQRRAQhVl1rVs+vz8wIDraEs4sdSs9VRPlSgAAAAAAQFfBDKIwO13tkiTFOyzy+fxRruarncxKT9YofVBClKsBAAAAACC6snJLevVw6gYERGFWcSYgSrBbVV7ljPjzV1eeVt7nO7Rv9yfy+f2acuEsyddHJSfpIAIAAAAA9G5ZuSVatyVPLo9PklRW4dS6LXmS1OtCIgKiMGsIiOKjEBAVHcjT+icflNNZqz79kuX3+fR67ipZ7f20L+VOacaIiNYDAAAAAEBXsnFbYSAcauDy+LRxWyEBEUKr4qwlZpFUfHCf1j/5oOIS+mrxLb/UoNQR8vl8ys/N1saXn9Df///HdOt3pig5eUBE6wIAAAAAoKsoq2i+kaOl4z0ZQ6rD7HS1SyajQfaYyGVxJ0qK9dKTD8oR30eLb/ulBqdlyGAwyGQyKXPSRbrqprvkrqvWXXf9VJWVlRGrCwAAAACAriQpIaZdx3syAqIwq6h2KcFhlcFgiMjz+f1+vf3GszIajVp82y+V0DepyTkTxo9TyvkLdfjwQT355JqI1AUAAAAAQFczb2aGrObG0YjVbNS8mRlRqih6WGIWZqdrXIGt5SMha/s2HczfrW9/7/+oT7/+zZ4zMNEue/9RuvTy72jz5k269toblJExMmI1AgAAAAAQTVm5Jdrwt32qrvNKkgwGye9Xr97FjA6iMKuodqlPhAIil8up559Zq4GDh+v8S65o8byURLskacrXviu73aGnnlobkfoAAAAAAIi2rNwSPf/2nkA4JNWHQ2aTodeGQxIBUdidrnYpwR6ZgOif723SieOluuraJTKZTC2e54i1KC7WotN1Ri1adKs+/jhLn3zyr4jUCAAAAABANG3cViivv+lxj9evjdsKI19QF0FAFEY+v1+V1W71iQt/QORxu/XxB/+raRdeovRRE9o8f3CSXUfLajR37vc0ePAQPffcU2GvEQAAAACAaGtth7LeuHtZAwKiMKqudcvn90ekgyg355+qrjytb3/3uqDOH9TfoZKyGlmtVl1//U3Ky9ujvLw9Ya4SAAAAAIDoycotkbGVPaR64+5lDQiIwqii2iVJERlS/fGHm5U0YLAmT50W1PmDEu2qqnWrosalK66YLZvNpk2b3gxzlQAAAAAAREdWbonWbcmTr5nlZdJXM4h6KwKiMDodoYDoi6L9OnIoXxfNmC2jMbi3dFB/hySppKxGcXFxuvzyK/T3v7+jqqqqcJYKAAAAAEBUbNxWKJfH1+xjcbFmLb4qs9cOqJYIiMKqps4jSXLYzGF9nk8+3CKr1aYpF34j6O8ZdGYns6Nl1ZKk7353nurq6vTuu1vCUiMAAAAAANHU2nyhNf95aa8OhyQCorCqddUHRLEx4QuInM5a7f70I0264FLZYh1Bf19iH5usZqNKymokSWPHjtOYMWO1adOb8vtb6LcDAAAAAKCbamm+UG+eO3Q2AqIwqnN5JUk2a8tbzndW/ufZ8rhdmjRtZru+z2gwKCXRHuggkqTvfGeuDhzYr/z8faEuEwAAhMjx48e1aNEizZ8/X88884xycnL0ve99TwsWLFBxcbEkafXq1Zo/f74eeOCBKFcLAEDXMW9mhqzmxjGI1Wzs1XOHzkZAFEZfBUTh6yD6fOd2xSf009D0se3+3oadzBpceuk3ZDKZ9MEH74WyRAAAEEIvv/yyli1bpldffVXbt2/X6tWr9eyzz+q///u/9eyzz+ro0aMqKirSq6++qvj4eOXk5ES7ZAAAuoTp41O0aPbYQMdQUkKMFs0e2+uXljUI73CcXq7O6ZHZZJDFbJTL2/wgrM5wOmtVsHenzp/+zaCHU59tUKJdn+wpldPtVYzFpL59+2ry5Knatu09LV26POT1AgCAzluyZIns9vpZgn6/X36/X3379lXfvn11+PBh7dmzR+edd54kadq0acrJydGUKVOCvn5SUlxY6g5GcnJ81J4bncN7173x/nVfvHftN+eyeM25bFS0y5DU9d4/AqIwqnN5w9o91LC8bPzUr3Xo+wf1d8gvqfRkjYYOrP+NOXPmLK1evUqHDh3Q8OEjQlgtAAAIhYSEBEnSG2+8oczMTO3duzfwmN/vV1VVVSBAio2NVU1NTbPXaUlZWZV8Le3/G0bJyfE6caIy4s+LzuO96954/7ov3rvuLRrvn9FoaPUHQSwxC6M6lyes84dyc/7Z4eVlUtOdzCRpxozLZDAYtG0by8wAAOiqNm/erHfffVd33nlno+MGg0EOh0O1tbWSpNra2kBYBAAA0BoCojCq7yAKT0DkdNYqf8+nGjdleruXlxmMBlU7PYpzWGUwSEWlVap2elTt9Cg2vq8yx03U++9vDUvdAACgc3bt2qXXXntNa9askdVqlc1m08mTJ7V//34NGTJEmZmZys7OliRlZ2drwoQJUa4YAICuISu3RHc9sV23PrJVdz2xXVm5JdEuqUthiVkY1To9soVpi/v9e3Pkcbs0bvL0dn+v0+3VZ/knJElxsRbtPXxKA/rFBh4fOvp8/fXNF/TFF0c0ZEhqyGoGAACd99xzz6msrExLly6VJN1555364Q9/KIPBoFWrVik1NVXDhg3T/PnzlZ6ermnTpkW5YgAAoi8rt0TrtuTJ5amfD1xW4dS6LXmSxJDqMwiIwqjO5VWc3RKWa+/P26kYm11DR3RseVmDPg6rTlc5Gx0bN/li/fXNF/TRR9t0ww0LOnV9AAAQWmvWrGly7PXXX2/09R133BGpcgAA6BY2bisMhEMNXB6fNm4rJCA6gyVmYRSuIdV+v1/79+7UiNETZTJ17voJDqsqqt3y+b8aRtk3cYDShg7Xjh0fd7ZUAAAAAACirqzC2a7jvREBURiFa0j1l6VHdPrUlxqVeV6nr9UnLkY+v19VNe5Gx887/0Ll5OyU08kfFgAAAABA9xYX23xzRVJCTIQr6boIiMIoXEOqC/bmSJJGZk7p9LX6OKySpIpqV6Pj551/gVwupz7//LNOPwcAAAAAANGSlVui2jpPk+Nmk0HzZmZEoaKuiYAoTHx+v+pcXsWGYYnZ/r071X/AEPVNHNDpazUEROXnBEQTJ02V2WzWJ5+wzAwAAAAA0H1t3FYor7/p8RiLkflDZyEgChOnyytJssWEtoPI7XLqUGFuSLqHJCnGapLNalJFVeOAKDbWrokTJys7m4AIAAAAANB9tTRnqLrOG+FKujYCojCpawiIQtxBdLhwrzxul0ZmTg3ZNfs4rDpd3fQPzLRpF6qgIF+nTp0M2XMBAAAAABApWbklLT7G/KHGCIjCpM5Vv74x1DOI9u/Lkclk1vCRE0J2zT5xVp2udsnvb9xzd8EFF0mSsrM/CdlzAQAAAAAQKRu3Fbb4GPOHGiMgCpOGDqJQzyA6tD9XaeljZLWGLuns44iRy+0L1Nxg1Kgx6tOnD9vdAwAAAAC6pda2sWf+UGMERGFS5wx9B1FNTY2OFR/QsIxxIbumJMXbLZKkqtrGW92bTCZNmXK+cnI+DenzAQAAAAAQCS0tI2N5WVMERGFSF4Yh1Xl7dsvv92n4yPEhu6YkOWKbD4gkadKkKSopOabS0tKQPicAAAAAAOGUlVuiyhpXk+NWs5HlZc0gIAqTcAypzv08R0ajSWnDx4TsmpLkiK2vsbqFgEiSdu3aGdLnBAAAAAAgXLJyS/T823vk8jTd3/5rE1NYXtaM0A7IQUDDkOrYEC4x27P7Mw1Oy5A1xhaya0qS1WyS1WJUVa2nyWMjR46S3e7Qrl05+ta3/iOkzwsAAAAAQChl5ZZow9/2tbqF/a7CsghW1H3QQRQmtSHuIHK7nMrP36thI0M7f6hBXKyl2Q4ik8mkCRMmateunLA8LwAAAAAAoZCVW6Ln3t7TajgktT64ujcjIAqTOpdHBoNktYTmFh85XCCP263hIR5Q3cBhs6iqrmlAJNUvMzt48IBOny4Py3MDAAAAANBZr/w9X/6mK8qaYEB18wiIwqTO6ZXNapLBYAjJ9Q4X7pHBYNDQEZkhud65GjqI/M38aWqYQ/T557vD8twAAAAAAHRWc2NTzmU2GRhQ3QICojCpc3lDOqD6UOEeDU/PUKw9LmTXPJsj1iyP1y+nu2krXmbmOJnNZgZVAwAAAAC6pKzckjbPMUhafFUmA6pbQEAUJrUuj2whGlDt83l15NA+ZY6fFJLrNSfuzFb31c0krjExNo0dO445RAAAAACALumVv+e3+rjZZNAPvjOOcKgVBERhEsoOouPHiuVy1mlM5oSQXK85jjMBUVUzg6ql+mVmeXl7VVdXF7YaAAAAAABor6zcklaXl8VYTHQOBYGAKEzqQthBdORwfRI6Zmx4BlRLUpytoYOo+YBo/PiJ8nq9KijYF7YaAAAAAABor43bClt8zGEz6ck7ZxIOBYGAKEzqXF7FxoSmg+jIoXzZHfFKGTQkJNdrjtVilMVkbLGDaNy48ZKkPXs+D1sNAAAAAAC0V2vb1t/0rTERrKR7IyAKkzpnaDuIhgwbHbId0ZpjMBjkiDWrqq75trykpP4aODBFe/bkhq0GAAAAAADaq6Vt6x02E51D7UBAFCb1M4g6HxDV1dboRMkRpQ4bFYKqWtew1X1Lxo0br717CYgAAAAAAF3HpIykJsesZiPdQ+1EQBQGfr8/ZEOqvyjaL7/fr7Th4f+N7Yi1tLjETJIyMyeopOSYTp4sC3stAAAAAAC0JSu3RNt3N93i/msTU+geaicCojDweH3y+vyKjel8B9GRQ/VDoYdEoIPIEWuR2+OTy+1t9vGGOUR79+4Jey0AAAAAALTllb/ny+XxNTm+q5DGhvYiIAqDWld9wBKKDqIjhwvUf+AQxdodnb5WW+La2Op+9OixMplMDKoGAAAAAETd+nfyWtzevrXB1WgeAVEY1Dnrf4N2dgaR3+/XkUP7lDpsdCjKalOcrT7Qqm5hULXNZtOIERnMIQIAAAAARFVWbone23m0xceN4dvjqcciIAqDukAHUecCovKTx1VdVaG04ZEJiBxtdBBJUmbmeOXl7ZHP17SFDwAAAACASHjl7/mtPu7zR6iQHoSAKAzqQrTE7IvD+yVJQ4aGf/6QVB9omYyGNnYym6CqqioVFxdFpCYAAAAAAM7V0tKyBkkJMRGqpOcgIAqDOteZJWadHFJ9tLhQJpNZAwYNDUVZbTIYDIprYyezrwZVs8wMAAAAABB569/Ja/Vxs8mgeTMzIlRNz0FAFAa1ztB0EB0tLtSAQUNltlhCUVZQHLHmVjuI0tKGyWazad++1v9AAgAAAAAQDu/ntDx7SJIWX5XJFvcdQEAUBg0dRLGdmEHk9/t17MgBDR4a2dTTYbO02qpnMpk0cuRoFRQQEAEAAAAAIisrt0T+VuYLLf3OOMKhDiIgCoNQDKkuP3lctTVVGpwW2YAoLtYip9srp9vb4jljxoxVQUE+g6oBAAAAABHV1nBqwqGOIyAKg1AMqT5aVChJEQ+IGnYyO1lR1+I5o0ePVW1tLYOqAQAAAAAR1dqKl29MHRzBSnoeAqIwqHN5ZLUYZTQaOnyNhgHVAwcNC2FlbYsLBETOFs8ZPXqMJCk/n2VmAAAAAICuYeGVY6NdQrdGQBQGtU5vtxxQLUlxsfV1t9ZBNGxYuqxWK4OqAQAAAAARk5Vb0uJjDlvndhGH1LkUA5Ikj09yur9qc6uqcyvGYlS186tjvlaGaJ3L7/fraHGhxk2+OJRlBiU2xiyjofUOIrPZrIyMUSoo2BfBygAAAAAAvVlr84du+taYCFbSMxEQhYDT7dGOvaWBr0tP1sjr8zc6Nnl0ctDXKz95IioDqiXJYDDIEWtptYNIql9m9ve/vyOfzyejkUY0AAAAAED4ZOWWtDp/iOHUnce/7MPA4/HJYur4rT1aXD+gelDaiFCV1C4Om0UnK9sKiMaqurpaR49+EaGqAAAAAAC91cZthS0+1onxvzgLAVEYuDw+Wcwdv7XHjhTKaDRp4ODhoSuqHeJiLa0uMZPqAyKJQdUAAAAAgPAra+XfqO0Z6YKWERCFgcfrk7lTAdEhJaekymKxhrCq4Dlizaqodsnt8bZ4Tnr6CJnNZuXnM4cIAAAAABA+699pvTEhKSEmQpX0bAREYeD2+GTtREBU+sUhpQxJD2FF7RPMVvdWq1Xp6RkERAAAAACAsMnKLdF7O4+2es68mZGf39sTERCFgdvjk7mDM4iqK0+r4nSZUoYMD21R7eA4ExB9ebr1OUQjR45SYWFBJEoCAAAAAPRCL/219aYEh83EgOoQISAKMZ/PL6/P3+EZRCVHD0lSl+ggKmtjJ7OMjFE6deqkysq+jERZAAAAAIBexuluefSJxPb2oURAFGJur0+SOh4QfXFIkqLaQWSPMctoCK6DSBJdRAAAAACAkGtr9pDVbKB7KIQIiELM7elkQHTkoBL6JMkRlxDKstrFaDSoT1yMyk7XtnpeQ0C0fz8BEQAAAAAgdIKZPbRodmaEqukd2kwxXC6Xli9frvnz5+u1114LHD916pRuvvlm3XDDDdq6dWvgeG5uru6///7A11dffbUWLlyohQsX6uTJkyEuv+vxNAREHZxBVHL0kAZGsXuoQVKCrc0OooSEPhowYKD278+PUFUAAAAAgJ4uK7dEz/7PnlbPyRzWl+6hEGszxdi8ebNmzJihDRs26N1335XL5ZIkbdiwQUuWLNFLL72k9evXS5K2b9+ue++9Vx6PR5JUUVGhiRMnav369Vq/fr0SExPD+FK6hs4sMXO7XDpRckSDUoeHuKr265cQ0+YMIqm+i4gOIgAAAABAKAQTDknSXTeeF4FqehdzWyfk5ubquuuuk9Fo1KhRo1RYWKjMzEzl5uZq0aJFiomJUVxcnCoqKhQTE6NHH31Uzz//vCSpoKBA+/bt04IFC3TppZdq2bJlQReWlBTX8VcVQsnJ8W2e4z9Zo/g4myTpVJVbkpQQbwsckySLxdzo6+YcO1osn8+r4SNGN3tuMNcI1TkpSQ79O++4+iU6Wt2RbeLE8fr44ywlJFgVExMTOB7MfUNj3LOO4b51DPetY7hvHcN9AwAAwXph8942z3HYTBGopPdpMyCqrq6W3W6XJMXGxqq6ujpw3OFwBI7X1NRo2rRpOnLkSOB74+Li9OMf/1iXX365/vM//1N5eXkaO3ZsUIWVlVXJ5/O3+wWFUnJyvE6cqGzzvBqnR5VV9d02FWf+73F/dUyS3Od83ZwTh+oHcPXtP6TZc4O5RqjOiY81y+eXCg58qf59Y1s8b/DgYfJ6vdqx4zONGVO//jPY+4avcM86hvvWMdy3juG+dUxL981oNHSZHwYBAICuYf07efJ4284B2LksPNpcB2W321VbWz+suLa2NhAK2e121dTUBI43hEhnS09P16WXXiqDwaDp06frwIEDoay9S2oYUt1a101LDhYWyGKNUWL/6K+j7Bdf32EU7E5mLDMDAAAAAHRUMEOpJWYPhVObKcb48eO1Y8cO+f1+7d27VyNGjAgcz87OltPpVHl5uRISmu669eabb+r111+XJO3atUsjR44McfldT2dmEB06sF8DBw+T0Rj9drmkPvUBUVtziAYPTlVsbCwBEQAAAACgQ4KdO9TXYWH2UBi1mWJcddVV+uc//6nrrrtOV155pdatW6cDBw7opptu0osvvqgbb7xRixYtavZ758yZow8++EDf//73lZKSotGjR4f8BXQ1ng5uc+/3+3Xw4H6lDB4ehqrar29c/TyhtjqITCaTRozIUGEhAREAAJF222236ciRI3rttdd0zTXXaOHChXruueckSatXr9b8+fP1wAMPRLlKAABa99Jf97V5jkHSb2+bEf5ierE2ZxDFxMToD3/4Q7OPvfDCC02Opaam6pFHHpFUP5vomWee6WSJ3YvL45PRYJDJ2L6AqKK8TNVVVRo4eFiYKmsfi9movnFWlbUREElSRsYobd36d/n9fhkMhghUBwBA7+Z2u7VixQp99tlnkqT9+/dr7dq1SktLkyQdPXpURUVFevXVV7Vq1Srl5ORoypQp0SwZAIAWOd3eNs/5wXfGRaCS3q3966DQKo/X16HlZaXHDktSlwmIpPplZsFsdZ+RMUpVVZU6ceJ4BKoCAABut1s333yzLrnkEknSgQMH9Mgjj2jx4sUqKirSnj17dN559S3406ZNU05OTjTLBQCgUxw2E3OHIqDNDiK0j9vTwYDoaJGkLhYQJdh08FhFm+elp9fPpTp48IAGDBgY7rIAAOj17Ha7LrjgAv35z3+WJF122WWaO3eujhw5ot/85je6/PLLG+1C27CxSDCiubtccnJ81J4bncN7173x/nVfPeG9m///vd3q4waD9MN5k3vEaz1XV3tNBEQh5vb4ZDa1f5nV8aOHldQ/WbH2rrPlb/8+sfr3vhPy+fwyGlt+TQ0B0YEDhbrooumRKg8AAJwxd+5cxcXFaezYsSovL5fD4VBpaamklnebbUlZWZV8vra3GA615OR4nThRGfHnRefx3nVvvH/dV094737zyqeqdra+vOwH3x6n8UP7dvvXeq5ovH9Go6HVHwSxxCzEOtxBdOywhqdnhKGijkvqY5PX51d5lbPV8/r06aukpP46eLAwQpUBAICzLVq0SC6XS8XFxUpKSlJmZqays7MlSdnZ2ZowYUKUKwQAoKm9h8tbfXzpd8axtCyC6CAKMbfXpxhL+7ap93o9OlFyJDBHoKvof9ZW94kJtlbPTU/PICACACBKlixZogULFshms+mhhx5Samqqhg0bpvnz5ys9PV3Tpk2LdokAADSSlVvS5jmEQ5FFQBRiHo9PcbGWdn1P2fGj8no9Gja8i3UQnQmFvjxdp1GprZ87YkSG3nrrz/J6254+DwAAQqNh59jU1FRdddVVjR674447olESAABBeWHz3lYf78DkFnQSS8xCzO3xyWJq320tPVq/g9mwM7N8uoqkhg6iILa6HzEiQ06nU8eOHQ13WQAAAACAbiwrt0Qeb+vz7m79NtvaRxoBUYi5O7DNfemxwzIajUpN7To7mElSjMWkeLslqK3u08/MTzpwYH+4ywIAAAAAdFNZuSV69n/2tHrON6YOZnlZFBAQhZDf7+/QkOrSo0VKGjBEFqs1TJV1XFKCTV8G0UE0fHi6pPqt7gEAAAAAOFcw4ZAkLbxybASqwbkIiEKooUXO3O6A6LBSBnet7qEG/fvYglpiFhsbq8GDh+jAAQZVAwAAAACaeu7ttsMhq5nhQ9FCQBRCbo9Pkto1g8hZV6vyk8c1oIsGREl9bCqrqJPf3/r6UImdzAAA6Ay/36+CgoKg/s4FAKC7Wf9OnoL5K27R7MzwF4NmsYtZCHm8ZwKidnQQHT9WJEkaOGhoWGrqrP59YuX2+FRR41YfR+tL4EaMGKF//Wu7XC5XhKoDAKD7W7VqlUaNGqXCwkIdPnxYsbGxWrVqVbTLAgAgpN7b2faGRoOTYpk9FEV0EIWQy9P+gOhESbEkaUBKWlhq6qyGre6DWWaWnp4hr9erQ4cOhbkqAAB6jpycHM2dO1e7d+/W2rVrdezYsWiXBABASC37zdY2zzEapIeWTo9ANWgJAVEIeTqwxOx4SbHMFqv6Jg0IV1md0rDV/Zena9s8t2Ens4KCgrDWBABAT+Lz+fTHP/5RkyZNUk5Ojmpr2/47FwCA7uIHK7fK7W39HKNBeu7uWZEpCC0iIAohd0eWmJUUK3lgqoxGU7jK6pRAB1EQW90PHTpMJpOJgAgAgHb41a9+JZvNpp/85Cc6evQoy8sAAD3Gj3/7vnxBzB0iHOoaCIhCyN2hJWZHlJySGq6SOs1uM8seYw5qq3uLxaK0tKEERAAAtENKSooqKyv16KOPKjY2VlZr6zP/AADoDta/k6dal6/N89i1rOsgIAqh9gZEzrpanT51QslddP5Qg2C3upfql5kREAEAELy7775baWlpys3NVVJSku6+++5olwQAQKcFM5RaYteyroSAKITcnvqFleYgZxCdKO3aA6obJLUjIBoxIkPFxcWqqakJc1UAAPQMFRUVuvrqq2U2mzVp0iQZjXw8AwB0b+vfyQvqPHYt61r4BBJCbq9fBklmU3AtcidKjkjqBgFRgk1fVtTJ72978eiIEfWDqg8dOhjusgAA6BH69++vl19+WdXV1dq0aZP69esX7ZIAAOiwrNySoLe0Z9eyroWAKIQ8Hp/MZqMMhuACouPHimQ2W9Sv/8AwV9Y5/fvY5HR5VV3nafPchp3MDh4sDHdZAAD0CL/+9a/l8/k0adIknTx5Uo888ki0SwIAoMOe/Z89bZ6TOawv4VAXREAUQm6Pr507mB1R/4FDuuQOZgajQdVOj6qdHjns9cMyj5yoChzztDBrbNCgwbLZbAREAAAE6aOPPtLgwYP19a9/XWlpacrKyop2SQAAdMitj2wN6ry7bjwvzJWgI8zRLqAncXu8sgQ5f0iSTpQUKy19TBgrRv8b+AAAIABJREFU6jin26vP8k9IUmD+0Cd7S1Vysn620AWZA2WOafrbx2QyKSMjQwcOHIhcsQAAdGN79+6VJPn9fhUUFMhkMunyyy+PclUAALTPirUfBnVe5rC+Ya4EHUVAFEJub/AdRE5nrcpPHtd5078Z5qo6zxFrkSRV1bqDOn/UqFH68MOPwlkSAAA9xk9+8pNGX996661RqgQAgI7Jyi1ReXXb/140Guge6soIiELIfWYGUTC+LP1CUtcfUC1JMRajzCaDqmvbnkEk1QdEf/nLX3T6dLn69CEdBgCgNf/4xz8Cv/7yyy9VXl4exWoAAGi/54KYOyRJz909K8yVoDMIiELI7fEptpllV805UdI9triXJIPBoLhYS7s6iCTp4MEDmjKFdBgAgNY0LDGTJKvVqt/97ndRrAbA/2PvzuPjPst7739mX6SRRsto8SbJuyzb8ZKEGHAW26FtQqEmgQSyEEhzEmghD+T0AU45rxdtoaSBnDxpH15wCAUCbVICTc4TSkKCLMchiZ14ixfZ8iJbsrzIWme0zGjW3/OHLDuObc1ImtFo+b7/AHvmnvldmlFsz1f3fV0iMjJb61tJPu8aHvjzJRmvRcZGAVEaxeIG1hR7ELWdacFisVJQXJbhqtIjRwGRiIhIWjU0NACwYcPFx837+/uzUY6IiMio/Oylg0nX3LRyBmtqJsdn3+lMAVEaRWLxlHsQtbW2UFw6E4tl4k0wu5xcl412fyiltaWlpeTm5nLs2NEMVyUiIjJ5Pf3001e877vf/e44ViIiIjJ6sfjw+4e8OTbu+ZPF41SNjIUCojSKxVLfQdTe2sLMigUZrih9cpxWItEEkVgcu3X4UMtkMlFZOZempuPjVJ2IiMjk894QKBgMkkgkMAyD9vb2LFYlIiKSuq31rUnX/K8vrR2HSiQdFBClSSJhkDAMbBZT0rWRSBh/Vxsrrr1pHCpLj9xzk8z6QzHsnuS7nqqq5vL665sxDAOTKflrIiIiMl1985vf5MCBAwQCAWKxGGVlZfzqV7/KdlkiIiJJPZWkOfVNK2eMUyWSDqltd5GkYvEEQEo7iDrOnsQwDErK52S6rLQZ6aj7iooqAoEAfn93JssSERGZ9BobG3n++ee57rrr+N3vfofNZst2SSIiIkl9/tG6pGt0tGxyUUCUJkPnLlMZc39hgtmsjNaUTrkjDIiqquYCg42qRURE5MrsdjvhcJhgMIjb7aavry/bJYmIiAzrL/8peTg0o8g1DpVIOikgSpOR7CAammBW6CvPdFlp47RbsFpM9AVTC4gqK6sA1IdIREQkifvvv5/f/OY33Hzzzaxfv54VK1ZkuyQREZEr2lrfSiKFufbffmBN5ouRtFIPojS5EBAl77fT3tpCUUk5FsvkeflNJhO5Lhu9wUhK64uLfeTm5tLUpB1EIiIil/PCCy/wkY98hOuvv/78bbfccksWKxIREUkuWd8hALPa0E5KkyehmOCiI9lBdPYk5TOrMl1S2nncdnpSDIhMJhMVFVXaQSQiInIFra2t3HfffcyZM4ePfexjrF27FrNZm7tFRGTi+t6zu1Ja95OvrctwJZIJ+ldImsRi53oQJQmIopEw3R2t+Mpmj0dZaeVx2+gLRjGMFPYTMtiHSD2IRERELu8LX/gCv/71r/niF7/I3r17uffee/nOd77D/v37s12aiIjIZR1s9idd89OvKxyarBQQpUmqR8w62k6fm2A2+QKiXLeNeMIgFI6ltL6ysopAwK9JZiIiIsOYN28eX/rSl/jnf/5nLBYLn/3sZ7NdkoiIyCVSmVqmsfaTm46YpUmqTaovTDCbfAGRx2UHoDflRtUXJpmtXLk6Y3WJiIhMVp2dnbzyyivU1tZisVi49dZb2bJlS7bLEhERucgvX2lIaZ3G2k9uCojS5PyY+yQBUVvrCcxmy6SaYDbE4x4cdZ96QHRhkpkCIhERkYvdc889hMNhbr31Vv7pn/4Jn8+X7ZJEREQua/Pu00nXVFd4x6ESySQFRGlyfgeRdfgjZu2tJynylWO12sajrLTKddkwmUh5kpnPV0JOTo76EImIiFzG3//931NVNfmGVoiIyPTy1X/5Y0rr/ubTqzJciWSaehClSapHzNpaWyZlg2oAs9lEjtOW8g4ik8lEZWUVzc2aZCYiIvJ+CodERGSi++ZTW/H3J//8p95DU4MCojSJxQ3MZhNm05V3EEWjEbraW/GVzRrHytLL47bRG0otIILBPkTaQSQiIiIiIjK5fO/ZXZzuDCVd582xqffQFKEjZmkSiyeSTjDrbDuNYSQoKZ8zTlWln8dto6m1N+X1lZVV/O53L+L3d+P1FmSwMhERkcnlF7/4xRXvu/fee8exEhERkUulMtIe4H99aW2GK5leYgE/kbNncS9cNO7X1g6iNBkMiJIfL4PJOcFsSK7bTiSaIDiQ2qj7qqrBSWZNTTpmJiIi8l4ejwePx8OWLVs4ffo0xcXFdHZ2snXr1myXJiIi01wqI+1BR8vSKXSskTNP/W+O/d+PcOrJJ0hEUuv9m07aQZQmsbiBLYUR92azmSLf5P2PyOMabK7dEQjhy3cmXV9RMRQQHWPFCjUtExERGbJx40YAXnzxRb7+9a8DcMstt3DfffdlsSoREZnuUg2HQGPtxyoRjdK3/R2662oJNx3H7HTiveEmvOs2YLbbx70eBURpEosl30HU3tpCYXE5Vtvkm2A2ZGjUfYd/AFI4KVdSUoLbncPx49pBJCIicjkmk4lf//rXLFq0iH379pGTk5PtkkREZJr65lOp72L96dfXZbCSqS3a3U3zK7/lzO9fJd7bi62sDN9n7iZvzYewuFxZq0sBUZqk0oOo7UzLpG5QDeBxD6aYHYHkzcrgwiSzpiY1qhYREbmcJ598khdeeIEXX3yRqqoqnnjiiWyXJCIi01QqTalB4dBoGIZB6Mhh/HW19O3aCYZBzrLleNffjLt6CSZz9jsAKSBKk1g8gdNx5ZczGonQ1XGGJSvWjGNV6WezmnHaLXQEBlJ+TGVlFVu3vpnBqkRERCYvs9lMMBjEZDIxY8YM2tramDVrcv9ASUREJp9Uj5Z5cybviZhsSEQi9L69FX9dLeGWFsxuNwUbPsLc2/6cXsvE2jWc/YhqiojFjWGPmJ0+fZJEYnJPMBvicdvo8KeWLMPgqPvu7i78/tS64IuIiEwnX/va15g9ezb19fUUFRXxta99LdsliYjINDOSvkOaWpaaaEc77b95jmN/8xXOPv0zjIRByT33Mfd7T+D71J04y8qyXeIltIMoTaJJjpidaB7swVMyyY+YweAxs5HsILowyUyNqkVERN6vp6eHW2+9lV/96lcsX74c8wTYYi4iItPHSMIhHS0bnmEYhBoO0r3pD/TveReA3JWr8K7bgGvRYkym4dvSZJsCojRJNua+5UQTJpOZopKZ41hVZuS6bBw/3UM0lsBmTf6P2MrKKgCam48rIBIREXmf4uJi/v3f/53+/n5efPFFCgoKsl2SiIhMEwqH0iMxMEDP1rfwb64lcvo05txcCv70Frw3rsNWVJTt8lKmgChNkh0xa2k+TmFxKTbb+I+qSzeP24bBYKPq8qLkZyZLSkpxudwcP65G1SIiIu/3D//wDzz//PMsX76crq4uHn300WyXJCIi04DCobGLnD2Lf/Mmet78I4lQCMecCko/dz+eaz6QlTH1Y6WAKA0SCYNEwhj+iNmJJnxls8exqswZmmTW1p1aQHRhkplG3YuIiLzfAw88wDPPPJPtMkREZBr5y39KPRx64M+XZLCSycdIJAge2I9/Uy39+/eB2Yxn9dV4123AOW/+hD9GNhwFRGkQjsYBrriDKBaLcuZUCx9ctHo8y8oYj3uwa33bCBpVV1XNZds2TTITERF5P4vFwre+9S0qKyvP9x+69957s1yViIhMVQ9+r46Ekdpal93MmpqJ10w5G+LBID1vvYF/8yaiZ89iycuj8KMfw3vDjVi9U+N4uAKiNIgkCYi62luJx+OUTJEdRE67BbvNTHt36gFRRUUVL730WwIBP/n53gxWJyIiMrl84hOfyHYJIiIyTfzylQbOfXxNyQ++emPGapkswqdP499cS89bb2GEB3DOnUfRX34cz9XXYrJOrUhlan01WRKJJQCueMSsrfUEwJQJiEwmE8X5rhHvIAJoajrOVVetzFRpIiIik87HPvYxXnrpJdra2li5ciXz5s3LdkkiIjJFbd59OuW107nvkJFI0L93D/5NtQQP1mOyWvFc8wG86zfgPDeEaSpSQJQGyY6YtbeexGQyUVQ6+SeYDSn2Oke0g2hokpkCIhERkYv97d/+LXPnzmXTpk0sXLiQRx55hJ/85CfZLktERKYYNaVOLt7XR+CN1/G/VkesowNrQQFFf/EJ8q+/EWteXrbLyzgFRGkQiQwGRFca+d7e2kJp2Qzsdsd4lpVRvnwXB453kzAMzCk04SotLcPlctPUpElmIiIi73X69GkeffRR3njjDdauXcuPfvSjbJckIiJTjMKh4YVbWuiu+wO9b2/DiERwLViI7/ZPkbti1ZQ7Rjac6fOVZlA4NrSD6MpHzCrmVI5jRZlXnO8kFk/g7w1TmOdMut5kMlFRUalJZiIiIu/jdrvZsmULsViMnTt3kpOTfEKoiIhIqhQOXZ4Rj9O3exf+ulpChw9hstnwfGANBes34Jg9J9vlZcXlt7zIiESiQz2ILn054/EYnW1nmF1ROc5VZVax1wUMjrpPVVXVXI4f1w4iERGR9/rHf/xHtm7dSm5uLi+//DLf/va3U37sl770JU6ePMm7777LJz/5Se666y5aWloAeOKJJ7jzzjv5n//zf2aqdBERmeC+9+yulNdOl3Ao1ttD5+9+y/Gv/w1nfvQDop0dFN/+KeZ+7wnK7vv8tA2HQDuI0mK4KWZdHa3E4zHmzJlajayK8wd3DbX5QyyuSG2kX2XlXF5++b/o6QmQl5efyfJEREQmvIaGhvO//ou/+Ivzv+7q6qKkpGTYx0ajUb761a+yZ88eAH74wx/y1FNP0dHRwVNPPcVDDz3EiRMn+I//+A8ee+wx3n33XVasWJGZL0RERCakX77SwMFmf0prH/jzJRmuJvsGmo7jr6ul9523MWIx3NVLKPnM3eRctQKTWXtnQAFRWgzXpLrtzOBP8WZXVNJvjGtZGVWQ58RiNtE+gklm721UvXy5/pEqIiLT29NPPw3A0aNHAVi6dCkNDQ1YLBb+7d/+bdjHRqNR7r33Xv7zP/8TgFAohNfrxev10tzczIEDB1i1ahUAV199tQIiEZFpKNWJZdUVXtbUlGW4muwwYjF6d27Hv6mWgWONmBwO8j58Pd5163HMmDpDpNJFAVEaXDhidmkPovbWwYBo1uwKDp3oG9e6MsliNlGU5xzxETNQQCQiIgLw3e9+F4DPf/7z/Ou//ismkwnDMLj//vuTPtbtdnPNNdecD4jeyzAM+vr6cLvdALhcLoLBYMp1FRXlprw23Xw+T9auLWOj925y0/s3eV3pvfvzR/6/lJ/jsS/fkK5yJoxIVzetr7xK6yuvEu324ywvo+ovP0fJupuwTqBefxPtvz0FRGkw3BGz9tYWCopKcTpdwNQJiAB8BS7aRrCDqKSkFJfLpT5EIiIi79HT00NLSwtz5syhsbGR/v7+MT2fyWQiJyeHs2fPAoO7i4bColR0dvaRSIz/tmefz0N7e++4X1fGTu/d5Kb3b/K60ns30qbUU+X9NwyDgWON+DfV0rtzO8TjuJcup+Tez+OuWYrJbKY7mIDgxPh6s/HfntlsGvYHQQqI0iAcTWA2mTCbL91B1Nbagq9sVhaqyrwSr4vjp3tSXm82m6moqKK5WZPMREREhvzd3/0djz32GK2trcycOZPHH398xM/hdDrp6uqiq6uLmTNnUl1dze9+9zvuvvtuduzYwbp106PxqIjIdDcdJ5YlohF633kHf10t4eYmzC4X3pvW4b1pPfbSqXl0LlMUEKVBJBbHar00HIrH43ScPcX86pVZqCrzfF4XwXCMvlCUXJctpcdUVlaxffvbGa5MRERk8njiiSf4yU9+Mqbn+OIXv8hDDz2EyWTiscceY9asWVRUVHDnnXdSVVXF1VdfnaZqRURkoppu4VC0q5PAa5sJvL6FeF8v9vIZlNx1D3lrPojZ6cp2eZOSAqI0iETjlz1e1t05OMGspGx2FqrKvLKiwe3qrZ1B5s9KbSpZVdVcfv/739Hb24PHk5fJ8kRERCYFwzB46qmnqKqqwmQa/IHT+vXrU3rso48+CsCsWbN47rnnLrrvK1/5SnoLFRGRCWsk4dBknlhmGAahw4fw19XSt3sXGAY5V63Au24D7uol5/8eldFRQJQG4Whi2AlmUzUgmnEuIDrT2Z9yQPTeSWbLll2VsdpEREQmi5UrVxIOhy8ae59qQCQiIjKScAiYlBPLEuEwPW9vxb+plsipk5jdORTc/Cd4b1qHrdiX7fKmDAVEaRCOxoedYFY8RXsQFee7sFrMnOlMfTJKZeWFSWYKiEREZLrr7e1l1qxZtLW1UVpayrp16/B4JtZEExERmbhGGg5NtqNl0fZ2/Js3EXjjdRLBIPZZsym993N4PnAdZocj2+VNOQqI0iASjWO77ASzk3gLfTgcU/P8o9lsoqzQxenO1KetlJaW4XQ6NclMRESmvQMHDvDNb36TW2+9lVmzZnHy5Enuuecevv3tb7N06dJslyciIhPcnf/jv0a0frKEQ4ZhEDxQj7+ulv69e8BkInfVarzrNuBasFDHyDJIAVEaXKkHUVvrCXxT9HjZkPKiHJpaRzbJrLKyiqYmTTITEZHp7fvf/z4//OEPKS0tPX/bRz/6Ub7+9a/zi1/8IouViYjIRLe1vpX+cDzl9ZMhHEoMhOh56038dZuItJ7B4vFQeMtHyb/hJmyFhdkub1pQQJQGkWgCh+3igCiRGJxgNnfR1D5GVV7kZkdDG9FYHJvVktJjKiqq2Llze4YrExERmdhisdhF4RDAjBkziMdT/we/iIhMT0/99kDKayd6OBRpbcW/eRM9b71BIhTCUVlF2ecfIPeaazDb7Nkub1pRQJQG4WicHOfFL2V3ZxuxWHTKNqgeUl6UgwG0doWYXZKb0mOqqubyyisv0dvbqz4LIiIyrb3/78JAIIBhGFmsSEREJrqpMM7eSCTo378Pf10twf37wGLBc/U1eNdtwDl3no6RZYkCojSIRONYrRfvIGo7cwJgGhwxuzDJLNWA6OJJZsszVpuIiMhE9td//dc89NBD3HXXXcyaNYvTp0/z85//nC996UvZLk1ERCaoyR4OxYP99LzxBv7Nm4i2t2HJz6foY39B/g03Ys33Zru8aU8BURpEoolLppi1t54EwFc6NSeYDSkrdGMCTnek3qh6aJJZc/MxBUQiIjJtXXvttTz++OP89re/5Z133sHn8/HYY48xZ86cbJcmIiIT0GQOh8KnTuGvq6Vn21sY4TDOefMp2vgJPKuuxmRVLDFR6J0Yo0TCIBpPXNKkuq31BPkFxThd7ixVNj7sNgvFXietXamPui8rK9ckMxEREaCsrIwHHngg22WIiMgENxnDISORoH/Pbro31RJqOIjJasVz7XV412/AWVGZ7fLkMhQQjVE4OthI8v0BUXvrSXylU/t42ZDyohxOd6QeEJnNZioqKjXJTEREREREJImRhEMTQbyvj8AfX8f/2iZinZ1YCwsp/sTt5K29HqsnL9vlyTAUEI1R5HxAdOGIWSIRp/3sSSrn12SrrHFVXuTmQFM3iYSB2ZxaM7HKyip27dqZ4cpEREREREQmr5GGQ9ncPTRwohl/XS29b2/DiEZxLVyE71N3krtiFSZLahOvJbsUEI3R5XYQ+TvbiEUjlJRPjx4C5UU5xOIJOgIhSgpSO1JXWTmXV155mb6+PnJzU2tuLSIiIiIiMl1MhnDIiMXo270Lf10toSOHMdnt5K35EN6b1uOYPT1O1EwlCojGKBxNABcHRG1DDarLpnaD6iEzinIAON0ZHFFABNDUdIylS9WoWkREREREZMhED4diPT0EXn+NwJbNxLq7sRX7KP7kHeR/+HosOTnjWoukjwKiMQpHLt1B1N7aAjBtehCVnRt139oZhPmpPaaqaiggOq6ASERERERE5JyJHA6Fjh3DX/cH+nZsx4jFcC+poeSue8lZfhUmszn5E8iEljQgikQiPPzww3R3d7Nx40buuOMOALq7u3n44YcJh8M8+OCDrFs3+E1ZX1/Ps88+y7e//W0AvvGNb9DU1MSaNWv48pe/nMEvJTvCl+lB1NbaQl5+ES739EhOc1028tw2TnemPuq+rKwch8NBU5MmmYmIiIiIiMDEDIcS0Sh9O7bjr6tl4PgxTA4n+dffgPem9djLZ2T8+jJ+kkZ8L730EmvXruWZZ57h1VdfJRKJAPDMM89w//3384tf/IJf/vKXALz55pt84xvfIBaLAbBjxw4KCwt59tlnOX78OGfPns3gl5Id5wMi68U7iKbL8bIh5UU5nBlBQDQ4yaxKk8xEREREREQYeTj028c/nqFKBsX83XT8n+c5/rVHaP3XHxMPBvF9+i7mfv8JSj5zj8KhKSjpDqL6+npuv/12zGYzCxYsoLGxkerqaurr6/nsZz+Lw+EgNzeXnp4eHA4H3//+9/npT38KwP79+1m9ejUAq1atYt++fZSWlqZUWFHRxGhc7PN5hr3fccIPgNfjwpPrIJFI0H72JGtu+DM8uc7z62w260W/v5xka9LxHOla43Y78BVe6Dc0d5aX1989RXHx4PuW7HUDWLRoAdu3b09p7XSg12F09LqNjl630dHrNjp63URERIY3UXYOGYbBwNGj+Ov+QO+unZBIkLNsOd51G3AvqdExsikuaUDU39+P2z0YBLhcLvr7+8/fnnOu+ZTL5SIYDHL11Vdz8uTJpI9NRWdnH4mEkfpXkgE+n4f29t5h17R3DX5N4XCEXpNBV/sZopEwBcUz6e0bOL8uGo1d9PvLSbYmHc+RrjWhgQhNJ8Pnf+92WOgPRdl3uI0yn4dgMIzDZsU6zJ8f5eVzaG39LU1NZ8jJmRiBYLak8r0ml9LrNjp63UZHr9voXOl1M5tNE+aHQSIiItk0EcKhRCRC7ztv46+rJXyiGbPLhXfdhsFjZCUlab+eTExJAyK3200oFAIgFAqdD4XcbjfBYJCcnBxCodD5IOi9hu4beqzX601n7RNC5H1NqtvONaguKZvaDarD0Th7Dref/32gbzAs2rL7FAsqCuntG+Ca6lKsjit/i1VWVgGDjaprapZltmAREREREZEJJtvhULSzE//mTQTeeJ1EXx/2GTMpufte8q77IGbn8KdOZOpJuj+spqaG7du3YxgGBw8eZO7cuedv37FjB+FwGL/fT15e3iWPXbJkCdu3bwdg586dLFmyJM3lZ9+FJtWDL+XZMycA8E3xgOj98nPsAAT6w0lWXvDeSWYiIiIiIiLTSbbCIcMwCDYc5PQP/oXjX//vdL/yMu4Fi5j1379Gxd99G++N6xQOTVNJdxDdcsstfPWrX+X555/ntttu4+mnn2bDhg185jOf4ZFHHuHJJ5/ki1/84mUfe8011/DCCy9wxx13cN111zFjxtRrYhWOJrCYTZjNg1PM2s+0kF9QjNN16Y6qqczttGK1mAj0RVJ+TFlZOXa7g+PHNclMRERERESmj2yEQ4lwmJ5tb+Gv20Tk1EnMOTkU/Mmf4b1pHbai4jE/v0x+SQMih8PBD37wg8ve97Of/eyS22bNmsWjjz4KgMlk4rvf/e4YS5zYwtE4dpvl/O/bWk9QUj4nixVlh8lkIj/HQaA/9YDIYrFQUVFJc7N2EImIiIiIyPQw3uFQpL2NQN0mAm/+kUQwiGP2HErv+zyea6/DbLeP6bllakkaEMnwwtE4Dtvg8bJ4PE7H2VPMW3RVlqvKjvxcO61dwRE9prKyij17dmeoIhERERERkYljvMIhI5EgeKAef10t/fv2gtmMZ9VqvOtuxjl/PiaTaVTPK1ObAqIxirxnB1F3RyuxWHRa7iCCwT5Ex073EDnXlykVVVVz+cMffk9/f9+0n2QmIiIiIiJT13iEQ/FQiJ633sBft4no2VYsnjwKb/1z8m+4CVtBwYifT6YXBURjNBC5EBC1tQ42qJ62AVHu4PbE7t4wbnvS/ucAVFQMTjJrbm5iyZKlGatNREREREQkWzIdDkVaz+CvqyXw5psY4QGcVXMpuv+/kXv1NZhtthE9l0xfCojGKBKNY7eeG3F/ZnDEva90VjZLypqhSWbdvQO4i1Jr0v3eSWYKiEREREREZKrJVDhkxOP0vbsb/+ZNBOv3g8WC55pr8a67Gde56eMiI6GAaIzC0ThO++DL2HbmBAVFpdgd03MkoMdtx2SC7p4wM1MMiMrLZ2iSmYiIiIiITEmZCIfi/f0E3nidE398jYHWs1i8Xoo+vpH862/Emp8/2lJFFBCNVTiaIC/nwhGz6Xq8DMBsNpHnttPdO5DyYwYnmVXQ1KRJZiIiIiIiMjWMNBiC5OFQ+NRJ/Jtq6dn2FkYkQt6Sago+fhu5K1dhsuqjvYydvovGKBwZnGIWj8fobDvDopprsl1SVnk9DjoDqQdEMNiHaN++PRmqSEREREREZPykMxw6f4ysrpbQoQZMNhueD1yHd90GZq9eSnt771jLFTlPAdEYhc9NMetsO0M8HqOkfHa2S8qqojwHza29hCMjm2RWW/sKwWA/bndOBqsTERERERHJnHSFQ/HeXgJ/3IL/tTpiXV1YC4sovu2T5K+9AUuupj9LZiggGqNINI7DZqGtZWiCWUWWK8quwrzB/kudPanvIqqsHJxk1tTUxJIlNRmpS0REREREJJPSEQ4NNDfh31RL7zvbMGIxXIurKfn0XeRctRKTObVJ0SKjpYAs+imNAAAgAElEQVRoDBKGQSSWwG4z09Z6ApPJTHHpzGyXlVVF5wKirhEFRIMd9pubjysgEhERERGRSWcs4ZARi9G7awf+TbUMNB7FZLeT96G1eNdtwDFzen++lPGlgGgMItHBY1R2m4W2My0UFpdis9mzXFV2OewW8nLsdPaEU37MjBkzsdvtmmQmIiIiIiKTymiCIRgMh2IBP4HXt+B/bTPxgB+brwTfpz5N3oc/jEWtNyQLFBCNQTiaAIYCouk9wey9fF4XZ7uCKa+3WCzMmaNJZiIiIiIiMnmMJhyyWeD/ua2CM0/9iN4d2yEex12zFO9n7yNn6XIdI5OsUkA0BuFzO4gsxOnqOEPNyg9muaKJwVfgovFUgOBAlBxHat9ilZVV7N+/L8OViYiIiIiIjN1IwyGLEefxG9x019XS8o+/wOx04r1xHd6b1mEvK89QlSIjo4BoDIYmdQW6WkkkEpSUTe8JZkNKCtwAtLT14ct3pfSYysq51Na+SjAYxO12Z7I8ERERERGRURlpMOSJ9bMicJjrY020/rQXe1k5JZ+5m7wPfgizM7XPSiLjRQHRGAztIOo82wKgI2bn+LyDf9CdONvHqgW+lB4zNMmsubmJ6uolGatNRERERERkNFIOhwyDWQNtrA40sKjvBGYTOJdfhXfdBtxLajCZTJktVGSUFBCNwVBA1HG2BbPZQlHJjCxXNDE4HVZyXTZazvam/JiqqsFJZk1NxxQQiYiIiIjIhJJKOGRNxFjSe5zVgQZKI90MmO0UfeRPyL9pHXZfyThUKTI2CojGIHLuiNnZMyco8pVjtdqyXNHEUZjnoKWtL+X15eUzsdlsNDVpkpmIiIiIiEwMqQRD+dE+VgYOcVXPEVyJCG12L1tmfYj7v3EvZodjHKoUSQ8FRGMwtIPozClNMHu/ojwnJ8520D8QJceZPDizWq2aZCYiIiIiIhPGsOGQYVARamV1oIH5/ScBOJwzm53exXzrW5/mwzpGJpOQAqIxCEfjJOIR2s+epmbl2myXM6EU5TsBaG7tZUllYUqPqaycS329JpmJiIiIiEj2DBcM2RJRlvYeY1WgAV8kQNDsYFtBDbvzFtFry+GnX183jpWKpJcCojEIRxNE+toxDEMTzN6nMG80AVEVmza9SigUwuVSR38RERERERlfVwqHvJEeVgcOsaz3KM5ElFZHIb8r+SAHcquImy0ACodk0lNANAbhaJxI31kASsoVEL2X026hMM9BU2vqjaorKwcbVTc3H2fxYjWqFhERERGR8XHZYMgwmBs8zepAA/OCp4hjoiG3gp351Zx2FsO5Y2QKhmSqUEA0BpFonGhvKzabnUKfJpi935wSD80jCIguTDJTQCQiIiIiIuPj/eGQIx5hWW8jqwINFEZ76bO4eKNgObvzF9JvdV+0VuGQTCUKiMYgHIkT7WulorIKi8WS7XImnNmlubx7tIPgQBR3Co2qZ8wYnGR2/HjjOFQnIiIiIiLT2fuDoaKIn1WBQyzracRuxDjlKObF0qtoyK0gYbr0857CIZlqFBCNQTgaZ6DnDFUr1KD6cmaX5gKDfYiqU+hDZLVaqays4tgxBUQiIiIiIpI5Q+GQyUgwv/8kqwMNVIZaiWHmoKeSnfmLaXUWX/axCoZkqlJANAZ+fxexgV6q5i3IdikT0pwSDwBNZ1MLiADmzp3Pjh3vZLIsERERERGZpoaCIWc8zPKeI6wKHMYb66PH6mZL4Ur25M0naL3ywByFQzKVKSAag/YzJwCoqppHNMu1TEQ5LhtFec4R9SGaP38Br7zyEn5/N15vQQarExERERGR6WIoGPKFu1kdOEhN73FsRpwTzlI2F6/mcM5sDJP5io9XMCTTgQKiMehqawGgau58Dp8OZ7maiamyzDOiSWbz5w/uxmpsPMLq1ddmqiwREREREZkm7v9uLYv6W1jtb2DOwFmiJgv1nip25S+mzZH8pIPCIZkuFBCNQXd7C84cL/n5Xjh9NtvlTEgVZR52Hm4nOBDD7Uz+7TZv3lBAdFQBkYiIiIiIjNpffeclVgQO84Wew+TFgvitudQVrWZv3nwGLI6kj1cwJNONAqIx6O08hbd4VrbLmNAqywb7EDWf7aW6IvmRMa+3gKKiYo4ePZLp0kREREREZAr6H996jtWBBv6qtwkrCY67ynnV9wEa3TOHPUb2XgqHZDpSQDRKsViM/sAZKhYsy3YpE1rFuYCo6UxPSgERwLx582lsVEAkIiIiIiKpMWIxenduZ+cv/pP7wh1ETFb25C9gV/4iOu3elJ9HwZBMZwqIRunEiWaMRBxfWUW2S5nQPG47Pq+TY2d6Un7MvHkL+M1v/oNYLIbVqm9RERGRZMLhMDfffDMVFRWYzWa+8pWv8J3vfAe73c6jjz7K7Nmzs12iiEhGxPx+/Fs2c+KlV8mNh3DZPNQWX8M+zzzCFnvKz2OzwP/+G4VDMr3p0/coDe1wKZ1Zmd1CJoGq8jyOngqkvH7evAVEo1FOnGhm7tx5GaxMRERkajh27Bi33347X/7ylwF48MEHeeqpp+jo6OCpp57i7//+77NcoYhI+hiGwcCxRvybavG/8zYWDM66Z/JS/mKOuWeAyTSi59OuIZFBCohG6WjjUTBZKCtXD6Jk5s7I552Dbfj7wnhzkzeDe+8kMwVEIiIiyR05coS33nqLbdu2cdtttxEKhfB6vXi9Xpqbm7NdnohIWiSiEXrfeRv/plrCJ5oZMNvYl7+YnfmL8NvzRvx8CoZELqaAaJSOHj2CPbcEtzN54DHdzS0f/MP6+OkeVi70JV0/Z04FNpuNo0ePcPPNf5rp8kRERCa98vJyHnnkEa666iruv//+i+4zDGNEz1VUlJvO0kbE5/Nk7doyNnrvJreJ/v6F2zto/f0rtL5aS6ynh3Z7Prt8H2C/Zy5Rs21Uz/nbxz+e5iqzY6K/dzK8ifb+KSAapcbGIzjyZmO3WbJdyoQ3pzQXi9nEsTOpBURWq5XKyiqOHTs6DtWJiIhMfkuXLsVut2OxWLjqqqvYu3fv+ftMIzxq0dnZRyIxslApHXw+D+3tveN+XRk7vXeT20R9/wzDIHT4EP66Wvp27wLD4LB7FjtnfIBmV9mIj5ENGdo1NBG/5pGaqO+dpCYb75/ZbBr2B0EKiEbB7/fT1dlB0eJrcSggSspuszDLl8ux06k3qp47dz47dryTwapERESmjh/+8IesXLmSG2+8kYaGBux2O11dXXR1dTFz5sxslycikrJEOEzPtq3462qJnDqJOSeHbXnV7M5fRMA2+h2OOk4mkpwColEY2tni8JTjsCsgSsXcGXlsO9BKwjAwp5D2z5+/gFdeeQm/vxuvt2AcKhQREZm87r77bh555BF+/OMf8/GPf5zFixfz0EMPYTKZeOyxx7JdnohIUtH2dvybNxF443USwSCO2bN5qWQNB3KriJlH/7F1RpGLbz+wJo2VikxdCohGobFxMCCy55XhsJmzXM3EZTKb6A/HAJjhyyEUjnPsTA/lRTnn1zhsVqyXeQnnzbvQqHr16mvHpV4REZHJqqSkhF/+8pcX3fbcc89lqRoRkdQYhkHwQD3+ulr69+4Bk4ncVav50ZlCTtpLwDG6Y2RDtGtIZGQUEI1CY+MR8vILsDo8OmI2jHA0zp7D7QD0BiMAvLbrFPNn5Z9fc011KVbHpd+GFyaZHVVAJCIiIiIyhSQGQvS89Sb+uk1EWs9g8XgovOWj/EO9nV5/DrjG9vwKhkRGRwHRKDQ2HqF8ZiVxUJPqFOXn2LFZzXQEQhcFRFfi9RZQVFTM0aNHxqE6ERERERHJtEhrK/66WnreeoPEwACOyirK7n+ARzYHiR+yjPnTqYIhkbFRQDRCsViM48ePc931f8Zp0A6iFJlMJorynXQEBlJ+zLx5888f5xMRERERkcnHSCTo378X/6ZagvX7wWLBc/W1eNdv4K+ea4YtYTCP7TOVgiGR9FBANEKnTp0kEgnjK6vgdJcCopEozndSf7yLWDyB1ZK8d9O8eQv4zW/+g1gshtWqb1URERERkckiHuyn54038G/eRLS9DUu+l6KPbyT/+hv4bz/cBc81p+U6CodE0kefukdo6MhTYekc6IpqitkIFOc7MQzo6glTUpD8YPG8eQuIRqOcONHM3LnzxqFCEREREREZi/CpU4PHyLa9hREO45w3n6KNn+Crm3pJ1FugfldarqNgSCT9FBCN0LFjR7BYLHgKyoAWHDYzkbiR7bImheL8wVCoIxBKKSC60Kj6iAIiEREREZEJykgk6N+zm+5NtYQaDmKyWvFcex3e9Rv44rPHoC4IpvT8YF3BkEjmKCAaocbGo8yZU0ncMGMygdViJhKPZ7usScHttOJ2Wunwp9aHaM6cCmw2G0ePHuHmm/80w9WJiIiIiMhIxPv6CPxxC/7NdcS6OrEWFlL8idvJX3sDD/xgOzx7LG3XUjAkknkKiEbo6NEjLF++gnA0jtNuwWQyZbukSaV4BI2qrVYrlZVVHDumRtUiIiIiIhPFwIlm/HW19L69DSMaxbVoMb47Ps3/9aofY68Z9m5P27UUDImMHwVEI9Db20Nb21nmz19AJBrXiPtRKPa6OHG2j4FIDKc9+bff3Lnz2bHjnXGoTERERERErsSIxejbvQt/XS2hI4cx2e3krfkQ3nXr+cK/HYE/9IAp+SCaVCkYEhl/CohG4PDhQwAsWLCI7afiuFIIOORivnwnAB3+AWaV5CZdP3/+Al555SX8/m683oJMlyciIiIiIu8RCwQIvP4a/i2bifv92Hw+fJ+6k7wPreWBf3kb/u1IWq+nYEgke5RwjMB7A6LXjzXhcujlG6nCPCcmoCOQakC0EBg82nf11ddmuDoREREREQEIHTuGv+4P9O3YjhGL4a5Zivee+8hZtpz7H3sNdr2d1uspGBLJPiUcI3DkSAMlJaV4vV5C4Rhuh46YjZTNasbrcdARCKW0fuHCRQAcPtyggEhEREREJIMS0Sh9O7bjr6tl4PgxTA4n+dffgHfdBh76+UF4uQtefi1t11MoJDKxKCAagcOHD58PLELhGAUeR5YrmpyK8p2cONuLYRhJ13o8eZSXzzi/e0tERERERNIr5u/G/9pmAlteI97bg620DN+n7yLvgx/mgSe3ws8PpvV6CoZEJiYFRCkKBoO0tDSzfv3NwGBApCNmo+PzOjl6MkCgP5LS+oULF3P4cEOGqxIRERERmT4Mw2Dg6BEO/XwLHVu3QSJBzrLleNdt4Eu/bYPtJti+Na3XVDAkMrEp4UhRY+MRDMNg4cLFAITCcdwKiEaltMANQFt36sfMtmypo7+/j5yc5H2LRERERETk8hKRCL3vbMO/qZZwywksOW4K1m0g/6b1PPTT/fBf7WAype16CoVEJg8lHCm60KB6IfFEgnA0rh1Eo+Rx23DaLZztCqa0fiiUO3LkMCtWrMpkaSIiIiIiU1K0sxP/5k0E3nidRF8f9hkzKbnns3ztzQjR4zY4vj+t11MwJDL5KOFI0ZEjh/B6C/D5SugfiAEoIBolk8lEaaE75R1ECxYMTjI7fPiQAiIRERERkRQZhkHoUAP+TbX0vbsLgNyVq/Cu28Bfv3AathpgtqX1mgqGRCYvJRwpOnLkEAsXLsJkMhEKDwVEmmI2WiUFLppbe+nqGSDHN/yxscLCIny+Eo4cUR8iEREREZFkEuEwPdvewl+3icipk5hzcyn401v4zgEHPT258H/OpPUYmdkEP/magiGRyU4BUQoikQjHjjVy551rAM4HROpBNHqlBS4Ajp4MMDtJQASwYMEiTTITERERERlGpK0N/+ZN9LzxOolQCMecCn5X8kEO5lYSO2KF9G4W0m4hkSlGCUcKjh9vJB6PXzTiHnTEbCy8Hgd2q5mjpwLctHJm0vULFy5i27Y3GRgYwOl0jkOFIiIiIiITn5FIEDxQj7+ulv59e8Fs5oBrNjtnLuaUzQf29O0UGqJgSGRqUsKRggsNqgcDoqACojEzm0yUFLhoPBVIaf3ChYtJJBI0Nh6hpmZZhqsTEREREZnY4qEQPW+9gb9uE9GzrVg8ebzpXcbu/IX0Wd1pv55CIZGpTwlHChoaDpCb62HmzFmAjpilS0mhm12H2gn0R8jPsQ+7dmj31qFDDQqIRERERGTaipw5TXfdJnreehMjPMApRzE7Sz/ModwK4qb090hVMCQyfSjhSEFDw0EWL67GdK6RWygcB7SDaKyG+hAdafFz9eKSYdf6fCUUFBRy6NDB8ShNRERERGTCMBIJ+vfuwb95E8H6/cQw0+CpZIdvMa3O4rRfT6GQyPSkhCOJcDjMsWNH+fSn7zl/m46YpUdhnhO71cyhFAIik8nEokXVNDQcGKfqRERERESyK97fT+CN1wm8Vke0vZ1ei4vdhSt4N28BQasr7ddTMCQyvSnhSKKx8QjxeJxFi6rP3xYKx7BaTNis5ixWNvlZzCYqy/M43OJPaf3ixdW8885WgsEgbnf6z1WLiIiIiEwE4ZMt+Os20f7HN7AbMVqcJewou54jOXNImNL7GeS3j3+c9vbetD6niExOCoiSOHhwcMfK4sVLzt82EI5p91CazJ+Vz8tbmwkORHE7h5+7WV29hEQiweHDDaxYsWqcKhQRERERyTwjHqfv3V1s+/lvqAidJWqycNBTxc78xbQ5CtN6Le0UEpHLUcqRxKFDBykoKKSk5MIRqKACorSZPzMfAzhyMsBV84c/Pz0U0jU0HFRAJCIiIiJTQry3l3/9h5+ysucQ+bEgXmsOm4tWsSdvPgMWZ9quo1BIRJJRypHEoUMXN6iGwSbVCojSo6Lcg8Vs4nCLP2lAVFBQSGlpmfoQiYiIiMik941v/ZrVgQaW9B3nRiNBk6uM2uJrOZozCyONx8gUDIlIqpRyDKO/v5/m5iZuuOHiP1RD4ZhG3KeJ3WqhagR9iNSoWkREREQmq7/8bi2L+ppZHWjgcwPtRExW9nrmsyt/MR0Ob9quo1BIREZDKccwDh48SCKRoLp6yUW3h8Ix8nLUJDldFs728so7JwhH4zhslmHXLl68hNdf30xPT4C8vPxxqlBEREREZOQ+/2gdADmxECt6DvOFwGE88RDdNg+1xVezzzOfsMWelmspFBKRsVJANIz9+/cDXDTBDIZ6EA0fZEjqFs728tK2Zo6dClBdOXwDvurqwfeioeEg11573XiUJyIiIiKSsqFQCMNgRriD1f4GFvc1YyFBo3sGL+ev4Zh7JrynhcVoKRQSkXRSQDSMffv24fOVUFhYdNHtITWpTqv5M/MxmeBQiz9pQLRw4VBAdEABkYiIiIhk3flA6ByLEae6t4nVgQbKw52ETTZ25y9kZ/5iuu15Y76eQiERyRSlHMPYu3cvNTVLL7otkTAYiMTVgyiN3E4rc0o8HDqRvA+Rx+Nh9uw56kMkIiIiIlnx/kBoiCfWz4rAYVb0HCEnPkCHLZ9Xi69lf948ImbbmK6pUEhExoNSjivo6urk5MmTfOxjt110+0AkBqAdRGlWXVnAH7a3EI7EcdiHP75XXV3Djh3vYBjGRdPlRERERETS7UqBEACGwayBNq72N7Cw/wQmDI7mzGJH/mKaXeVjOkamUEhExptSjis4cKAe4JIdRMGwAqJMqKkq5Pdvn+BQSzfL5w0/7n7JkqW8+urLnD3bSllZ+ThVKCIiIiLTwbCB0DnWRIwlvcdZHWigNNJNyGxnu7eaXfmLCNg8o762QiERySalHFdw4MB+rFYrCxcuuuj2UDgOoCNmabZwVj42q5n9x7uSBkRDoV19/T4FRCIiIiIyJqkEQkPyo32sDBziqp4juBIR2uxeXvZdR71nLjHz6D4fKBQSkYlCKccV1NfvY9GiRTgczotuD2kHUUbYrBYWzfZSf7wr6dp58xbgcDior9/P+vUfGYfqRERERGQqGEkYdJ5hUBFqZXWggfn9JwE4nDOHnd7FtDhLRnyMTIGQiExUSjkuIx6P09BwkI0b/+KS+4aOmDk15j7taqoK+VXdUbp6BijMc15xndVqZdGiaurr941jdSIiIiIymYwqDHoPWyLK0t5GVvsPURwNEDQ72FawlN15C+m15YzouRQKichkoIDoMpqajhMKBbnqqqsuuW/gXECkI2bpYTKb6D/3ms6dmQ/AriMdrFladn6Nw2bFar74cTU1y/j1r58lHA7jcDjGrV4RERERmXjGGga9V0Gkh1WBQyzrPYozEeWMo4j/KvkQB3MriZtT+yGxAiERmYyUclzGgQP7AS4bEOmIWXqFo3H2HG4HwDAMXA4Lb+07g9VyYavuNdWlWN/3etfULOXZZ2McOXKIpUuXj2vNIiIiIpId6QyCLmIYzA2eZnWggXnBU8Qx0ZBbwc78ak47i5MeI1MgJCJTgVKOy6iv30d+fj5z5syho6Pvovs0xSxzTCYTM4pyaGnvI2EYmIf5i3jJkmXA4HulgEhERERk6shYCHQZjniEZb2NrAo0UBjtpc/i4o2C5ezOX0i/1X3FxykQEpGpSCnHZRw4sJ8lS5ZiukxAEQrHsZhN2N9/5knSorw4h8bTPXT1hCnOv3IfouLiYsrKys/v9hIRERGRyWM8Q6DLKYr4B4+R9TRiN2KcdPr4Y+EKDuXOIWG69BiZAiERmQ6SBkSRSISHH36Y7u5uNm7cyB133AFAd3c3Dz/8MOFwmAcffJB169ZRW1vLj370I/Lz83n88cfxer3ceuutFBYWAvDkk0+e//VE1dvbQ3Nz0xWnY4XCMVwO62XDIxm78qLBn9Sc6egfNiACWLJkKfv27RmPskRERERkBLIdAF2OyUgwv/8kqwMNVIZaiZnMHMytYkf+Ys46iy5aq0BIRKajpAHRSy+9xNq1a7nzzjt54IEH2LhxI3a7nWeeeYb777+f6667joceeoh169bx9NNP88wzz7B161aee+457rzzTpYtW8ajjz46Hl9LWuzfvw/DMFi+fMVl7x8MiDTBLFNcDiuFeQ5Od/SzbF7RsGtrapZRV/cH2trOUlJSOk4VioiIiMhEDICuxBkPs7znCKsCh/HG+uixutlSuJJ38xcQsjgVBomInJM0IKqvr+f222/HbDazYMECGhsbqa6upr6+ns9+9rM4HA5yc3Pp6enB6/Vit9tZvXo1v/rVr1i9ejWHDh3irrvu4vrrr+fBBx9MubCiotwxfWGj1dh4EKvVytq1HwDA5/NcdH/MAE+O46Lbja4gntzhd7vYbNYxr0nHc4xXLZ5c56ivU1mex7uHO3A4bNhtFtxuB77CS8+AX3/9Gv7lX6Cp6RA1NfOHvc5k8P7vNUmNXrfR0es2OnrdRkevm8jkM5kCoCvxhbtZHThITe9xbEacE85Sqh/4LAtWrOJqi37gKyLyfkkDov7+ftzuwQ/nLpeL/v7+87fn5ORcdPvQOrfbTTAYJDc3l7/6q79i/fr1PPzwwzQ0NLB48eKUCuvs7CORMEb1RY3Ftm3vsHDhYvr6Yrhc0N7ee9H9gd4B7BbTRbcHwzF6+waGfd5odOxr0vEc41GLJ9dJb9/AqK9TlOcgYRg0tnQzqySXYDBMezx+yWOLimbicrl48823ufba64e9zkTn83ku+V6T5PS6jY5et9HR6zY6V3rdzGZT1n4YJDLdTYXw50pMRoKF/S181NKM7VQTJrudvLUfxnvTBhbOnp3t8kREJrSkAZHb7SYUCgEQCoXOh0JDIVBOTg6hUOiidcFgELfbTVVVFVVVVZhMJtasWcOxY8dSDoiyIRKJ0NBwgI0bP3nFNaFwDJ/XNY5VTT8lBS4sZhOnO/qZVXLlDw9Wq5WammXs3bt7HKsTERERmVh++UoDm3efznYZWXHTyhnc8yeLifX2ENjyGoEtrxHr7sJaXIz3k3eQ/6G1WHIVRouIpCJpQFRTU8P27dtZsGABBw8e5Ctf+cr523fs2MF1112H3+8nPz+frq4uwuEwO3fupKamhhdeeIFoNMrdd9/N3r17+dznPpfxL2gsDh1qIBKJsHz5VVdcM9SkWjLHYjZTVujmdGcw6drly1fws589RW9vLx6PjjCIiIjI5Pe9Z3dxsNmf7TImlCv1CRpoOk7rvz5F7/a3MWIx3NU1lNx1DznLr8Jk1tRhEZGRSJp03HLLLXz1q1/l+eef57bbbuPpp59mw4YNfOYzn+GRRx7hySef5Itf/CIA9913H3fffTdOp5N//ud/xul08vDDD/P73/+ea665hoULF2b8CxqLffveBWDZsisHRMFwXAHROCgvdnOqoZ2+UHTYdcuWXYVhGOzfv5c1az40TtWJiIiIDG867+oZrVSbRRuxGL07t+PfVMvAsUZMDid5a2/Ae9N6HDNmZLhKEZGpK2nS4XA4+MEPfnDZ+372s59d9PsNGzawYcOGi2778Y9/PIbyxtfevXuYM6cCr7fgsvcnDIMBTTEbFzOKc4B2Trf3D7tuyZKlWCwW9u3bo4BIREREUra1vpVn/nCI/oFL+xxKZox1WljM78e/ZTOBLZuJ9/RgKy3Fd+dd5H3wQ1jclw40ERGRkdFWmHMSiQT79+9h7dobr7gmHIljgHYQjYP8HDset43ms8M3hHW5XCxcuFh9iERERCaorfWtPL+lkc6eMEV5Dj5xwzzW1JSNeP3lbgfO35bjtGAymegLxS76tdkECYPz/y+Zk4lx8YZhMNB4FH9dLb07d0AiQc7SZXjXb8C9ZKmOkYmIpJGSjnNOnGiip6eH5ctXXHFNKBwDFBCNB5PJREWph/qmLvpDUXKGec2XL1/B888/RzgcxuFwjGOVIiIiMpyt9a08/XIDkVgCgM6eME+/3ABw2ZDoSuuPnvTz5r7Wi27/6X8dwGQ2EYsPpj7v3Qn03l8PhUIKh8buSgFQJqY8JqIRet95G/+mWsInmjG7XHjXbcB74zrspaVpvZaIiAxS0nHOu+/uAobvP3zVlwcAACAASURBVDQUELkVEI2LijIP+493sbexkw2rZ11x3fLlK/jVr/6dQ4cODhvwiYiITBeRSISHH36Y7u5uNm7cyB133JGVOp7f0ng+1DlfWyzB81saLxsQXWn9lndPXxLwxI2h/5GxysTOn9GKdnUSeG0zgde3EO/rxT5jBiV330vedR/E7HRmuzwRkSlNScc5u3fvxOcrYebMKwcRofDgT6O0g2h8FOY5yHXZePdI+7AB0VCot2fPbgVEIiIiwEsvvcTatWu58847eeCBB9i4cSN2u33c6+jsCafldu3+GbmJFPokYxgGocOH8NfV0rd7FxgGOStWUrBuA67F1ZhMpmyXKCIyLSjpYPAvpd27d/GBD1w37F9AQR0xG1cmk4k5pbkcOuGnfyBKjtN22XVer5e5c+eze/dO7rnnc+NcpYiIyMRTX1/P7bffjtlsZsGCBTQ2NlJdXT3udRTlOS4b+hTlXf5I+JXWT/f+QdUVXv7m06uyXUbaJcJherZtxV9XS+TUScw5ORR85E/x3ngTtmJftssTEZl2lHQATU3H8Pu7Wbny6mHXqQfR+Kso83CgqZt3j3TwoWXlV1y3atXVvPjiC0Qikaz8hFRERGQi6e/vx31uqpPL5aK/f/ipoO9VVJSbtjru+2gN/++v9xCOXugJ5LBZuO+jNfh8npTXr796Fpt2nLzodovZhMnE+R5EE5nNaubLn1rBjatnZ7uUjLnc+3klA62tnHn5Fc7+YRPx/n5yqiqZ89dfoPj6tVjUTzIrRvL+ycSi925ym2jvn5IOYNeunQCsWDH8T2bUg2j8Fec7KfA42HmofdiAaOXK1fzmN//BgQP7k76PIiIiU53b7SYUCgEQCoXIyclJ+bGdnX0k0rRdp2aOl3v/dNEl08dq5ngvaWrs83muuH5NTRkzi3PSPsUs12Xl0xsWDjtVLZ3S3ch5okilSbVhGAQP1OOv+//bu+/4qOp8/+OvMzOZJDMpk4QUAqEHCE0IRYpSAl5sq7KyIl1F1FVZZbm7lt31t3vXtazreoW1y1rwImBBXRsYQJGeoBSRHpAeAmHSM8nMnN8fgUgJElQ4CXk/H488ksycc+Z9vufAnPnk+/2eTErWrQXDICK9BzGDhxDWJhXDMMgvrAAqzk9oqXYuJhmX80PHrn6z4vjZbMYP/iFIlQ6q5h9KSmpMcnKTH1zu+x5E9vMRS6gaZnZRm0YsWbePMp//tL23unbthmEYfP31ahWIRESkwevYsSNZWVmkpqayceNGJk+ebFmWPh2TzqoAc7rlf+hxqbuC5WUULlvKkYWZVB44gD0yitirriZ6QAYhMTFWxxMRkePYrA5gtWAwyJo1q+nWrfsZly31+bEZBqEhKhCdT91SG+EPmKzddui0y0RGRtG2bTu++ir7PCYTERGpm6688kqWLVvG8OHDGTp0KKEatiPnWcWBAxyc+QY5/z2ZgzPfwBYWTtKEibT8+5M0uu56FYdEROqgBt+DaMeO7RQWFtaqQFTVg8WuOymcZy2So/BEOMnenEfvH/grYbduPXjnndmUl5cTptugiohIAxYaGsozzzxjdQxpYMxgkJJv1uFdkEnphm/AbieyRy88g4cQ3qq11fFEROQMGnyB6Nj8Q7UvEDX4JjvvbIZB93YJLF67j/IKP2HOmo9BenoPZs16g2++WUePHr3Oc0oRERGRhslfXMKR+fPwLlpAZd5B7NEe4q4dRnT/ATiiPVbHExGRWmrw1Y7Vq7No0qQpiYlnHr9e5guoQGSRHu3iWbB6D+u2H6ZXWmKNy3TpchF2u52vv85WgUhERETkHPPt3Yt3YSbbViwj6PMR1iaVRsOuJyK9O4ZD18wiIvVNg/6fu7Kykq+/Xs3QoVfUavkyn59wp+YfskJqUw9RbifZmw6etkDkcrlp374D2dlZTJx4ngOKiIiINABmIEDx2jV4F2ZStmkjhsNB/ID+hPUbQFiz5lbHExGRn6BBF4g2bFhPWVkpPXteXKvly3x+YiI1yaMVbDaD7m3jWbp+/w8O9evZ82Jef/3fFBYWEBUVfZ5TioiIiFyYAsXFFHz5Bd5FC/HnH8YRG0ej639F9CX9SWqVrFtti4hcABr0XcyyslZgt9tJT+9Rq+VLfX7Cwxp0Tc1SfTolUeEPsvLb3NMu06tXb4LBINnZq85jMhEREZELU/mu7zjw6nRyfjeZQ++8RUhCAo3vnETLR/9O7BVXYY+MtDqiiIj8TBp0tSMrayUdO3bC7Y6o1fKapNparZOjSEmIYNHXexnQNbnGu8m1b9+ByMgoVq1aQUbGZRakFBEREanfTL+f4q+/qhpGtnULhtNJVN9+eDKGENqkqdXxRETkHGmw1Q6v18vmzZu45ZbbarW8aZqU+QK4VCA67wybQYnPD0DfzknMXrCNDTuP0DI5CoDQEAeOo33hHA4HPXr0ZNWqFZimWWMRSURERERO5S8ooGDx53i/WETA6yUkPp74G24kqt+l2N1uq+OJiMg51mCrHatXr8I0TXr27P2Dy/mD4Kv046sMEDRN7HZbdbHimKB5LpOKrzLA2i15QNUt70PsNt77ModLujQGoGdaIo7jCnc9e/Zm0aIF7NiRQ6tWrS3JLCIiIlJflOXk4F34GcXZWZh+P66OnfCMvQl35y4YtgY9I4WISIPSYAtEWVkriYyMol279j+4nK/ST9bGXErLq4pCufmlZG08cQ6ci9rGn7OccqIQh41WTaLYuqeAHu0TCKvhrnK9elUV/VauXK4CkYiIiEgNgpWVFGdn4V2YSfmOHGxhYUT3H4gnYzDOpMZWxxMREQs0yAKRaZqsWrWCHj16YrfX7rb1Ff4AAE6H/opitbYpHjbv8rJ9bwEdW8ae8nxCQiItW7YiK2sFI0eOsSChiIiISN3k9x7B+/kiCr74nEBRISFJScSPGkNUn37Yw8OtjiciIhZqkAWiLVs2c+hQHr1796v1OpX+IAAhISoQWS0mMpSEmHC27PbSoUVMjcv07NmbuXPfoqysjHBd7IiIiEgDZpom5du24l2YSdFXqyEYxN25C57Bl+FK66BhZCIiAjTQAtGyZV9iGAa9e/et9ToVlVUFIvUgqhvapnhYsm4/+w+X1vh8nz79mDNnJtnZq7j00gHnOZ2IiIiI9YIVFRStWoF3QSa+3buwuVzEDL6M6EEZOOMTrI4nIiJ1TIMtEHXo0ImYmFOHJ51O5dEhZiGO2g1Jk3OreVIEWRvtbNntrfH5iy7qRkREBEuXLlaBSERERBqUysOH8C5aSMGSxQSLi3E2aUrC2JuI6t0HW2io1fFERKSOanAFory8g2zevInbbrvrrNar8KsHUV1it9lo0zSab3fm4y324Q498VR2OBz07t2XZcuWEAgEaj3XlIiIiEh9ZJomZZs24l24gOI1XwEQ0S0dT8YQwtu1xzAMixOKiEhd1+AKRMuWLQGgX79Lzmq96jmIVCCqM9qmRLNhRz5L1+3nhkFtTnm+X7/+ZGbOZ+PGDXTq1MWChCIiIiLnVtDno3D5UrwLF1Cxby+2iAhiLr8Sz8AMQuLirI4nIiL1SAMsEH1J48bJtGjR6qzWU4Go7ol0OWkS72bJuv1c068FYc4TT+devfpgt9tZuvRLFYhERETkglJx8CDeRQsoXLKYYFkZoc2ak3jzBCJ7XozN6bQ6noiI1EMNqkBUXl7O6tXZ/OIX1511N9sKf4AQh03dc+uYLq3j+GTFLhZ9vZcrLm5+wnORkZF07ZrO0qWLuf32sxtSKCIiIlLXmMEgpd9uwLswk5L168BmI7J7DzwZQwhr3UbXqSIi8pM0qAJRfv5hAgE/GRlDznrdysqgeg/VQfGecNo3j+HTlbvISG9KaMiJcw1dckl/nn76Sfbs2U3TpikWpRQRERH58QJlZRQuW4J34QIqcw9gj4oi9upr8AwYiMMTY3U8ERG5QDSoikdychM++iiTzp0vOut1K/xBTVBdR11xcTOKSiv54uu9pzzXt++lAHz55efnOZWIiIjIT1Oxfx+5/zeDnP+eTN6b/4fd7Sbp1tto9fd/0ujaYSoOiYjIz6pB9SACcLncP2q9Sn9Qt7ivo1o1iSateQyfrNzFwG5NcB7Xi6hx42Tat09j0aIFjBw51sKUIiIiImdmBoOUrFuLd2Empd9uwHA4iOx5MZ7BQwhr0dLqeCIicgFrcAWiH6vCHyA8VM1VV13TrwWPz/yaxWv3MaTHiUPJBg0awnPPTWPfvr0kJzexKKGIiIjI6QVKSihYspiCRQupPJSHIyaGuOt+SXT/gTiioqyOJyIiDYDGTNVSVQ8iNVdd1a5ZDG1TPHy84jsq/YETnhs0qGrOqc8/X2BFNBEREZHT8u3ZTe7rr5Lzu8kcems2jpgYGt9xJy0ffYK4q69RcUhERM4bdYmppYpKzUFU113TrwX/mLWGL9ftJyO9afXjSUmN6dChE4sWZTJq1DgLE4qIiIiAGQhQvOYrvAsXULZ5E0ZICJEX9yFm8BBCU5pZHU9ERBooFYhqwTRNKv0BzUFUx6U1j6FNk2g+Wv4dl3ZJPqHH16BBg3nmmad1NzMRERGxTKCoiIIvv8D7+UL8+fk44uJoNPwGoi/pjz0iwup4IiLSwKlLTC0EgiZBE/UgquMMw+Cafi04UuRj4Vd7Tnhu4MDBgIaZiYiIyPlX/t1ODvz75aphZO++jTMxieS7fkPLR58g9vIrVRwSEZE6QT2IaqHSHwQgJEQForquY8tYLmodx9zFOXRLbURCjAuAxMQkOnbszGefzWP06PEYhmFxUhEREbmQmX4/Rauz8S7MpHz7NozQUKIu6Y8nYzChummGiIjUQap41EJFZVWByKkhZnWSYTMo8fkp8fkprQgwPKMNNpvB9I83UlxeSYnPjz8Il19+JTt2bGfLls1WRxYREZELlL/Ay+EP3iPnvv/mwEvPEygqIn7ESFo98U8Sx4xTcUhEROos9SCqhcpA1V2xdBezuslXGWDtlrwTHuua2ogVG3KZ+dkW2qZ46JmWSEbGZUyb9hSffvoh7dq1tyitiIiIXGhM06Q8ZzvehZkUZWdBIICrUxc842/G3akzhk3XkCIiUvepQFQL3/cg0pt7fZHaNJqd+4tYvSmPJvFuACIjo+jXrz+ZmfO48857CAkJsTiliIiI1GfBykqKs1ZxZGEmvp07sIWH4xmUgWfQYJyJSVbHExEROSsqENVCcVklAO5wFRTqC8Mw6NMpkQ+W7GTFhlwGdK3qzn355VexaFEmK1Ys5dJLB1obUkREROqlyvx8Cr5YRMHizwkUFeFsnEzC6LFE9emLLSzc6ngiIiI/igpEtVBcWolhgCtMzVWfRLqcdGvbiOxNeWRvOsjArk3o2fNiYmPj+OSTD1UgEhERkVozTZOyrVvwLviM4q+/AtPEfVFXPBlDcKV10A0wRESk3lPFoxaKyyqJCA/Bpjf+eqd98xh27i/i7c+30y01nmi3k//6ryt46603OXz4EHFxjayOKCIiInVY0OejaOUKvIsy8e3ejc3lJuayoXgGZRDSKN7qeCIiIj8bTapTC0VHC0RS/9gMg76dk6ioDPDG/Kq7l/3iF9cSCAT48MP3LU4nIiIidVXloTzy3ppNzu9+S+7rr2CakDjuZlo98U/ifzVCxSEREbngqAdRLRSXVtI8KcLqGPIjeSJCuaJ3c/6zdCdZmw7Ss31zevS4mA8+mMvo0eNxOPTPQERERKqGkZVu/BbvwkxK1q4BwyCiWzqewZcRntpWw8hEROSCpk/GZ1Dm8+OrDKgHUT03uEcK67cfZsa8zbRr5mHYsOH84Q+/Y+nSLxkwYJDV8URERMRCwfJyCpcvxbtwARX792GPiCT2iquIHjiIkNg4q+OJiIicFyoQncHhwnKgasJjqb/sNoNbrkrjf17NYuZnW7j1qn4kJCTy3ntvq0AkIiLSQFXkHsC7aAGFS5cQLCsjtEVLkm6ZSETPnthCdO0nIiINiwpEZ3C4oKpApB5E9V/T+Ah+0a8lcxfn0LN9Atde+0teeuk5du7cQYsWLa2OJyIiIueBGQxSuuEbjizIpPSbdWC3E9mjJ56MIYS1aq1hZCIi0mCpQHQGh44ViFwqEF0Irri4GV9tzmPGvM3896+u4rXXpvPWW2/yu989aHU0EREROYcCpaUULluCd9ECKnNzsUdHE3fNdUQPGIgj2mN1PBEREcupQHQGhwvKcDpshIbYrY4iP4FhMyjx+QEYeVkqT8z8mreX7GfIf13JJ59+xIhRN5PcuDEO3ddPRETkguLbtxfvwgUULl+K6fMR1roNcdcOIzK9B4ZuVCEiIlJN74pncLigXL2HLgC+ygBrt+RV/96tbSOyN+XRIrYHAf8HPPfyK/zx/t/jCNU/CRERkfrODAYpWbsG78JMSjd+i+FwENmrN57BQwhr3sLqeCIiInWSPg2fwaGCcs0/dAFKax5DYUkFW3YX0LxdT7KXzKO4+Ne4Q2OsjiYiIiI/UqC4mIIli/F+vhD/oUM4YmJp9MvhRF3aH0dklNXxRERE6jQViH5AMGiSX1hOu2Yal36hMQyDXmmJFJdVsqOwF75NK/nwg3eZcPMEq6OJiIjIWfLt3s2RhZ9RtHIFZkUF4W3bEf+rEUR0Tcewa5oAERGR2lCB6AccKSrHHzCJCNdtTi9ENptB/67JlJb7yU9oz9tzZnLD8BuIjIy0OpqIiIicgen3U7zmK7wLMinbugXD6SSqd188gwYTmpJidTwREZF6RwWiH3DgcCmgW9xfyJwOOxndm3Lk4OVs//x/efmVV5j8m99YHUtEREROw19YSMHizyn4YhH+I0cIaRRPo1+NIPqS/tjdbqvjiYiI1FsqEP2AA4dLAIjUJNUXtIjwEK7K6MVrWy/ivblzCE++mBFDuxITGWp1NBERETmqfOcOvAsyKcpaien34+rQkYTR43B3uQjDptuQioiI/FQqEP2AA4dLMQC3ehBd8Bp5wvmfP/6Oe+8cx8cfvMmanWUM6ZHClb2b4QrT8RcREbGC6fdTlL0K78JMynNyMELDiO4/AM+gwTgbJ1sdT0RE5IKiAtEPyM0vwRMZit1mWB1FzoN2bVtz+ZW/YP6nH9Fi4FV8suI7vlizl5GXteWiNo0ACA1x4NAfKUVERM4pv9eL94tFFHyxiEBhISGJScSPHE1U30uwh4dbHU9EROSCpALRDzhwuJS46DCrY8h54qsM0LnPdSxa+Bmbls/hF2PuY/mGXF7+z7e0aRJNz7QE+nZujCNU/2xERER+bqZpUrZtK96FmRStzoZgEHfnLngyhuDq0FHDyERERM4xfdL9Abn5JbRN0S3uG5KIKA+DrxrNR2+/xIGcr7midx/WbTvENzn5HMgvpXGcm04tY62OKSIicsEIVlZQtGolexcvomR7DrbwcDwZQ/AMzMCZmGh1PBERkQZDBaLTqKgMkF/oUw+iBqhHv6GsXp7JJ3P/TZsO3ejWNp7keDdL1x3gqTlruO6SllzVtwU2Q0MPRUREfqzKw4cp+GIRBYu/IFBchKtZCgljxhHVuy+2MF1/iYiInG8qEJ3GoYJyABpFh2OapsVp5Hyy2+1c/avbmP70g3z2/utcfcPtJMa4uLpfc7btKWTulzvYeaCIW6/uQLiGm4mIiNSaaZqUbd6Ed2EmxV9/BUBE13Q8g4fQ7JKeHDpUbHFCERGRhkufbk8jz1sGQFxUGIcKyixOI+dbs1bt6T3gapZ//h/ad76YNmldcTrsjLu8HalNopm9cBsPv57NpOu7kBTrsjquiIhInWb6/RQsXYJ3YSYVe/dgc7uJGXoFnkEZhMRV3QjCUM9cERERS2m2v9OoLhBpiFmDNeTq0cQnNuW9mf+irLTqL5qGYXBZzxSm3NiVotJK/vpaNmu3HbI4qYiINGQ7duwgIyODsWPHMmXKFAAyMzMZPnw4EyZMwOv1WpwQCleu4OCMVzFsNhJvuoVWTzxF/PAbqotDIiIiYj0ViE7jUEE5oU47ka4Qq6OIRUKcoVw/9l6Ki7x8MOtZTNPEsBmU+Pw0S4rkd6O6ERcdxtS31/Hyh9+yP78Uf9Dq1CIi0tBs376du+66ixkzZvDkk08C8NprrzFz5kzGjRvHnDlzLE4IkT170fx/HqHZQ38h+pL+2JxOqyOJiIjISTTE7DTyvGUkxbrU3bmBS27WmiG/GM38919n6cL36NruNtZuyat+vv9Fjfl6yyGWbzhA1saDXNGnOVf1bkaIw25hahERaUi2bt3K8uXLeffdd7ntttvo1q0bHo8Hp9NJ9+7dmT179lltLy4u4twEbRJ3xkXi4yPPzWvLOadjV7/p+NVfOnb1W107fioQnUaet4zkOnawxBr9Mq5j73fb+OyDN7ikdzq4m1c/57Db6JmWQLtmHrI3HeSDJTtYtn4/113akq5t4nGF6Z+YiIicW6mpqQwePJjk5GTGjx/P1KlTcbmq5sdzuVyUlpae1fYOHy4mGDz/N+iIj48kL6/ovL+u/HQ6dvWbjl/9pWNXv1lx/Gw24wf/EKRPrzUwTZO8gnLS2ydaHUXqAMMwuG703Rw8sJsnHnmImyY9THxSygnLRLmdZHRvSpTbyXuLc3j5w43YbZtIbRpN59ZxdGkVR3Ijt3qkiYjITzZ79mw+/PDD6t9vvvlm2rZtC0CzZs0IDQ2lrKxqLsXS0tLqYpGIiIjID1GBqAZFZZX4KgIkxumCSqqEhoYz+rYHeXXaH3j9uf/h1nsfJTrm1Ik127eI5XdNPeTsK+DbnUfYsCOftxZt561F20mMDadPp8Zc2iWZmAjNvSAiIj/OiBEjGDFiRPXvU6ZMoWnTprRo0YLc3FxiY2PJz8/H5/OxevVqOnbsaGFaERERqS9UIKrBsTuYJcW5LU4idUlsoyT+38NPcv+Uu3j92b9wy28exh0ZfcIyvspA9RxFjeNcNI5zUVJeyd68ErbvLeC9xTn8Z8kO0tvGM6BrMmnNY9SrSEREfpI77riDP/3pTwBMnDgRm83GTTfdxJgxYwgLC2Pq1KkWJxQREZH6QAWiGhwrECXGujD9AYvTSF3SqnUqoyY+wBvP/5XpU//A+Dv/XGNPouO5w0Jom+KhbYqHI0U+ikorydqYS9amg6QkRHBl7+Zccek5mhBUREQueKmpqcyaNeuEx4YMGcKQIUMsSiQiIiL1kW5zX4ND3nKgqkAkcrKWqZ0Ye+f/o8ibz/SnH+TwwX21XjcmMpTrB7bmn3f34+Yr2+MPBHnhgw38+vEFfLFmL5X+4DlMLiIiIiIiIlIzFYhqkOctI9rtJMypDlZSsxatO3DzpL9S4SvnxX/ex7aNa2q9rmEzqAiYpLdL4P6x3ZlwdQecDhuvfbqZydOW8NRba/lk5S52HigkEFTBSERERERERM49VUBqkOctI94TbnUMqeOSm7Xmtil/582XH2PG839l8NWj6NR64hnXO36eomOu6teSLd/ls2N/ITv3F7J++2EAwpx2midG0iTeTZNGbpIbuWkSH0FEeMg52ScRERERERFpmFQgOknQNMk9Uka7Zh6ro0g9ENsoiYmTH+O9mf8i8z9vsH/nN1w27HZiGyWd1XYMwyD5aAEIILVZDNt2e9m+t4A9ecUs++YA5RXfz4fVOM5FWotYOjaPoV0zD64wFYxERERERETkx1OB6CRL1+/nSJGPzi3jrI4i9YQzNIxf3TSF1A7pzJv7b559bDIDhv6KPgN/gSPkxxVuXGEOTNOkVXIUrZKjME2TknI/BcU+8gt9HMgv5cu1+1i4eg+GAS2SoujYMpZOLWNplRyFw67RoyIiIiIiIlJ7KhAdp7iskrcWbadN02gu7phodRypRwzDoNvFGVw1dAD/+PvjfPafGWQvm8/gq0fTqVtfbDb7T95+RHgIEeEhNImPoHPrOLq1S2DnvgI27/KyeZeXj5bv5MNlOwlz2klN8dCxRSxpzWNoEu/GZhg/z46KiIiIiIjIBUkFouO888V2Ssv9jP2vdvpALT9KfEISo257kG2b1vDp3Fd4+7V/sujjN7lk8DA69+iP0xn6s71W0DQ5UuQjISachJhwLu6QwP7Dpew/XMKOfd/PY+QOc5Da1EO7Zh7aNImmWWIkIQ71MBIREREREZHvqUB01PZ9BSxes4/LeqaQkhBhdRyp59q078qd9z3FxnUr+WLeW7w/61nmvf8qF/UcSOh11xC0xWOz/bxFGmeIneZJkTRPiqzKkOJhyy4v2/Z42bangDXbDgHgsBukJETSMjmKVsnRJMeFk+AJ1zxGIiIiIiIiDZgKREAwaPLGvC1ERzi59pKWVseRC4TNZqNj1z50uKg33+VsJHvJPLKXzmfl4o+JjmlEalo6LVM70SK1E5ERjX/213eHh2AzoG2Kh7YpHkrL/eR5y6q/Pv96LwtX76lePiI8hISYcGIiQ4lyO4l2Oau+u51ERVR9j3aHqveRiIiIiIjIBUgFImDR13v5LreIO67tSHiomkR+XoZh0KJ1B1q07sDVN9xG2aHNfPLJp6z/agnZy+YDkNg4hWatOpCY3IKExim0apz+s+dwhTlO6GEUCAZplhRF3pGqgtEhbxl53nL25BVTtLOSUp+/xu24wxyEOR0EgkECQZNAwCRgmphBEwATME0wDAgNsRMeaifM6SDcaScs1IErzIGr+nsIka6QqiLU0WJUpMuJzaYhniIiIiIiIufTGashFRUV3HPPPRw5coRhw4YxYsQIAI4cOcI999yDz+fj9ttvJyMjg8zMTJ5//nmio6N58skn8Xg8PPDAA+zcuZM+ffrwm9/85pzv0Nk6UuTj3cU5dGgRQ8/2CVbHkQtcWLibiwcPJSYlnUAgwIE9OeRs/YbdOd+ybvWX+JbOA+DfUyEiKoa4+MZEexoR5YkjOqYRkdGxuCKiCA93kxcTxOfz4XSGYfyIObPsNhuNPOHs+i8N/gAAEs1JREFUPVh8tHeQk9ZNoqufDwRNfBV+WjX1UFDko6i0goKSCgpLKvFV+LHbbdhtBnabgcNup6o0VMUwqopEvsoA5RUBKvxBynyV5Bf52HeohFKfnzKfH9M8NZcBRLhCiHQ5iTr63RXmwGGz4XAYOOw2HHYbpmkSFu6kuNhHMGgSPLoxwwADAwxw2G2Eh9oJdzoIO/bdWVWwCnXaCQ05+uW0Yf+Zh/yJiIiIiIjUJ2csEH388cdceuml3HjjjUycOJFhw4bhdDqZOXMmEyZMoHfv3txxxx1kZGTw2muvMXPmTJYvX86cOXNIT08nNjaWRx99lMmTJ5Obm0tiYu3uDnYuehAEg0GWb8hl76ESDheUc7iwnMKSCjwRTm66Mg17DbcGd9htZ5yb5XwtU1+yhIc6CPhD6kSW873M2W0jhMh2HUht14EIdyhFxeUUFx7h8MF92AJH2Lh5OwX5eRQVevluy04CgRN79Mw9+t1msxMa5iI0PByHPQR7SAgOu50vIlyUV5rYHSHY7Q7sDjt2ewh2mw3DZgPDYNvXLo4U+jAM4/svm4Fh2AAbhgGFu93kecurii7HOT5NQoyLg97S0+5zcoyLvLIyogCqdh0zwiQYNPEHTPyBIJGuUCoqA/j8ASorg1R4AxzJC3Cw0qQyGCAYBNM0CZpV36GqmFSV++gLHfvhaOEpGDSpoQZVI5vtaPHJVtUGNqNq2zYDbIZBEBPTrJocPBjk+HpYdaHLsIHdVjW80GZUFc+OTXh/fA0vEDSPfgWPFrdOzXOs+GazGThsNmx2o7r4ZTu2veP3/Wh7HHvMOJa/+rmqopkBOELsVFT4q9vSNME8un/m0TY71gvMAAzb8ftgfP+48f3Pxz9+3CE42j4mQY5u++hrHZeqOpdxdAO26r35fiu1OY7Hn6GG8f22j2U7WdVxq8qEWfNrHJ/P6bRTURmsXrh6efPUjNVtccrrn7jPJ+Th+BzmCRkxOW3bHZ/zxJb4PufpirHVOYyaMzqdTkYOu4oYT2QNrVN7Nb2nq6dg3WXlsdF5UX/p2NVvOn71l45d/Xa+j9+ZXu+MBaINGzYwfPhwbDYbqampbN++nbS0NDZs2MD48eMJDQ0lIiKCwsJCPB4PTqeT7t27M3v27OqfAdLT01m/fn2tC0QxMe5aLXe2rhkYdVbLN20cTdPG0WdcrlXTmPOyzPl6HWWxPouIyIUiLk43f6hPztU1WG3oXKm/dOzqNx2/+kvHrn6ra8fvjGMqSkpKcLlcAISHh1NSUlL9uNvtPuHxY8u5XC5KS0tPu66IiIiIiIiIiNQdZywQuVwuysrKACgrK6suCh0rAh17/PjlSktLcblcuN3uGtcVEREREREREZG644wFoo4dO5KVlYVpmmzcuJFWrVpVP56dnY3P58Pr9RIdHU1+fj4+n4/Vq1fTsWNHOnToQFZWFgCrV6+mQ4cO53ZvRERERERERETkrBmmWdPUld/z+Xz89re/5cCBA1x//fUUFxczZMgQPB4PU6ZMoaCggDvvvJMhQ4aQmZnJCy+8QFhYGFOnTsXj8fDggw+Sk5ND7969mTx58vnaLxERERERERERqaUzFohEREREREREROTCdsYhZiIiIiIiIiIicmFTgUhEREREREREpIFTgUhEREREREREpIFTgUhEREREREREpIFTgUhEREREREREpIGz//nPf/6z1SHqmoqKCu6++27eeOMNAoEAnTp1sjpSneXz+cjIyGDhwoW8//77tGzZkkmTJvH+++/Tq1cvoqOjrY5YJ02aNIkuXbqQk5NzSns99dRTPP3006xdu5aMjAyro9Ypx9rtk08+4Y9//CMfffQR+fn5pKenq91OcvDgQe68805mz55Nfn4+drtd51otnNxu27Zt07lWC3l5edxxxx3MmjWLlJQUNm3axO9//3vmzZvHgAEDCAsL44EHHmD69Ons27ePiy++2OrIUofdf//9REZG0rRpU6ujSC1t27aNu+++m1mzZuF0Omnfvr3VkaQW9Jmn/jr5eqV79+5WR5KzNG3aNPbu3UtaWprVUU6gHkQ1+Pjjj7n00kuZOXMm8+fPp6KiwupIdVZOTg7Dhw9nxowZvPbaazz33HO89NJL/OUvf+Gll16yOl6dU1lZyaRJk1i7di3AKe21b98+du3axaxZs4iMjGTNmjUWJ64bTm63bdu2MW3aNGbMmMGtt96qdqvB//3f/3H77bcza9Ysli5dylNPPaVzrRZObre1a9fqXKuF2bNn8+tf/5oZM2Ywffp0XnvtNWbOnMm4ceOYM2cO2dnZxMbG8uabb7Jjxw5yc3Otjix11LZt28jMzLQ6hpyl559/nkcffZRZs2Yxc+ZMq+NILekzT/118vVKcXGx1ZHkLOTn5zNnzhyrY9RIBaIabNiwge7du2Oz2UhNTWX79u1WR6qztm7dyrJlyxg1ahTvvPMOZWVleDwe2rRpw3fffWd1vDqnsrKScePG0bdvX4BT2uvbb78lPT0dgB49eujD51Ent1tOTg6PPfYYN998M7t27VK71WDChAn06tULANM0CQaDOtdq4eR2y83N1blWC3fddRd9+vThyJEj2O12PB4PTqeT7t27s2bNGr755pvqv26mp6ezfv16ixNLXTV9+nSuu+46q2PIWXrwwQdp3rw5hmFgs+njRX2hzzz118nXKw6Hw+JEcjZefvnlOvtep//Ba1BSUoLL5QIgPDyckpISixPVXY0bN2bKlCm8+uqrvPfee5imWf3c8T9LFZfLRc+ePWt8zjRNiouLTzj3SktLz2e8Ouvkdhs4cCCPP/449913H0888YTarQZRUVE4HA7efvtt0tLSMAyj+jmda6d3crvpXKsdwzDYtGkTw4cPp3PnztVt5HK5KC0t1fuq1Mq6deto0qSJhqfXQ7GxsUDVkIkrr7zS4jRSW/q/uf46+XolLCzM6khSS/v376ekpISWLVtaHaVGKjXWwOVyUVZWBlT18HC73RYnqrs6deqE0+nEbrdz0UUXsW7duurnjv9AKmdmGAZut7t66EVZWVn1m7acaNiwYURERNC+fXu8Xq/a7TQ+/vhj5s+fz7/+9S9uvfXW6sd1rv2w49utoqJC51otpaWlsWjRIq677jpatWoFQGlpKS6XC7fbfcL7qsfjsTKq1BGzZ8/mww8/rP59+/btzJ8/n1deecXCVFIbJx+7iRMnsm3bNnJzc7nnnnssTCZnQ5956rfjr1ek/njhhReYOHEiq1atsjpKjdSDqAYdO3YkKysL0zTZuHFj9YWunOq5555j8eLFmKbJpk2bcDqd1RO7NmnSxOp4dV5YWNgJ7ZWWlkZ2djYA2dnZmizwNMaPH09FRQW7d+8mLi5O7VaDdevWMXv2bKZOnYrT6dS5Vksnt5vOtdr53//9X7Zs2YLD4SA2NpZ9+/bh8/lYvXo1HTt2pEOHDmRlZQGwevVqOnToYHFiqQtGjBjBjBkzmDFjBs8//zwej4df//rXzJ07l0ceeYTy8nKrI8ppHH/sZsyYgc/nY926dTz88MNWR5OzoM889dfJ1ytSf6xfv54HHniAF198kRdffJHdu3dbHekE6kFUgyuvvJLf/va3vPvuu1x//fWEhoZaHanOGjNmDFOmTOHFF1/k2muvpX379txxxx0YhsHf//53q+PVeXfeeecJ7dW0aVOaN2/OjTfeSMuWLenRo4fVEeukCRMmMHr0aMLCwnj44YfVbjV4+eWXOXz4MBMnTgRgypQpOtdq4eR2Gzt2rM61Wrjmmmv405/+RCAQ4LrrrqNRo0aMGTOGsLAwpk6disfjYe7cuYwYMYLevXuTnJxsdWSpY9xuNx9//DFQNUypV69eGjJRjzz33HNA1R9wAGbMmGFlHKklfeapv06+XvnnP/9JfHy8xamkNt555x0A3n33XQBSUlKsjHMKw9REMSIiIiIiIiIiDZqGmImIiIiIiIiINHAqEImIiIiIiIiINHAqEImIiIiIiIiINHAqEImIiIiIiIiINHAqEImIiIiIiIiINHAqEInIj7Zy5Ur69+/P2LFjq79KSkpqvf6mTZv49ttvf3KOyspK7rnnHkaNGsVDDz2Ebs4oIiIiVvvb3/7G2LFj6devH7/85S8ZO3Ysq1at4tprrz1nr3ns2iovL4+nnnqq1utNmzaNzMzMc5YLYM+ePaxYsQKAP/zhDwDntC1E5Ow5rA4gIvXb0KFDq9/kz9Znn31GWloaHTp0+EkZ5s+fT2pqKk8//TQPP/wwy5cvp2/fvj9pmyIiIiI/xbHro/vvv5/x48eTlpZ2zl/z+GuryZMnn/PXOxurVq2isLCQ3r1787e//c3qOCJSAxWIRORn9/nnn/P8889jmibjxo3jqquu4osvvmD69OlUVFTQqlUrHnjgAebOncu8efNISEjg+eef59lnn2XPnj088sgjPPvsswwdOpSEhARuvPFGSktLeeeddwC499576d27d/XrrV27lssuuwyAvn37kp2drQKRiIiI1El+v58pU6awa9currjiCm655ZYar53mz5/Piy++iM1mY9iwYYwcOZLx48cTHR1N06ZNufzyy3nssccIBoMMHTqU4cOH13ht9fjjj5OdnY1hGPzjH/+goKCAJ598Er/fj9vt5tlnn60x5zvvvMMbb7xBSkoK+/fv56233mLs2LE888wzREVFVf/83XffnbK9Y9d0ubm5BAIBXnzxRV5//XVKSkro06cPv//973n//ferX6um/X/00UdZv349wWCQv/71r6Smpp6vQyTSYKlAJCI/ybx589i0aRMAGRkZjB8/nqeffpo333wTu93O2LFjGTx4MLt37+aFF14gLCyMX/7ylxiGwbBhw0hLSyM2NrbGbR86dIj//Oc/FBcXM2nSJGbOnElpaSkTJkxg9uzZ1csVFxfjdrsBCA8Pp7S09NzvuIiIiMiPUF5ezr333ktSUhLXXHMNN9100ynXToMGDWLq1KnMmTMHp9PJ6NGjueyyywgGg9xyyy107dqVUaNGMW3aNGJjY7n99tsZOnToKddWmzZtYu/evbz11lusXr2azZs3U1payuOPP05iYiJ33XUX27ZtOyWjaZq8+uqrvP3223i9Xq6//vrT7k9OTk6N22vevDmPP/44Dz30EFlZWYwbN47CwkLatWt3wvrBYLDGa8dly5bx+uuvc/jwYYqLi3/GIyAip6MCkYj8JCcPMTt8+DD79u1j4sSJABQVFZGXl0dMTAz3338/brebgoICgsFgjds7fv6g5ORknE4ne/bsYdeuXYwfPx6AgoICKioqcDqdAERERFQXhUpLS4mIiDgn+yoiIiLyU0VERJCSkgKA0+nkyJEjp1w7bd26lcTERFwuFwDt2rVjz549ALRo0QKA7du3c++99wJV10Z79+495bV27NhBp06dAOjevTsAy5cv55FHHsHlcrFr164ar8ny8/NJTEwkNDSUxMTE6rzHO3bNlpCQUOP2jvX4iY+Pp6Ki4rTtUdP+5+Xl8eCDD/LHP/6RiooK7r777tOuLyI/HxWIRORn5fF4aN68Oa+88gp2u50XXniBhIQEnnjiCTIzMykvL+eKK67ANE0Mw8A0TUJDQzl06BAAW7durd6WzVY1j35ycjJt27atHqL24osvVheHADp37szKlSvp0aMHy5cvZ8CAAed3p0VERER+pJqunVJTU8nNzaWsrIyQkBA2btxIkyZNgO+vj9q0acNzzz1HREQEb7zxBi1atGDlypUn/LEtJSWFzz77DICvvvqKlStX8umnnzJ9+nSio6O54YYbary5R0xMDHl5eZSXl+Pz+di/fz9QVdDKy8sjJCSE3bt3A/DYY4/VuD3DME7Y5sm//9D+x8TEMHPmTJ555hk2bdrEM888w7Rp035KM4tILahAJCI/K7vdzq233sqYMWOoqKhgyJAhhIaG0r9/f66//noiIiJITEzk0KFDtG/fnqeffpoXXniBli1bMmLECLp06XLKNhs1akRGRgYjR46krKyMMWPGnPD85Zdfzn333ceNN95Iy5YtueSSS87X7oqIiIj8JDVdO7lcLiZNmsS4ceMIBoOMHDmS+Pj4E9abPHkyt99+Oz6fj06dOjF69Ojqa6uHHnoIgE6dOpGUlMSoUaOw2+089thjAIwfP56oqCgiIiLIy8s7JZPNZmPSpEmMHTuW+Ph4QkJCABg5ciT33HMPKSkp1QWryy+//Izbg6qC1v333096evoZ9z8iIgK73c6wYcMIDw9n0qRJP62RRaRWDFP3gxYREREREZHTuPbaa0+YVFpELkw2qwOIiIiIiIiIiIi11INIRERERERERKSBUw8iEREREREREZEGTgUiEREREREREZEGTgUiEREREREREZEGTgUiEREREREREZEGTgUiEREREREREZEG7v8D+avcEh5g4P8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(nrows = 1, ncols = 2, figsize = (20,8))\n",
    "\n",
    "sns.distplot(df_train[0], fit=norm, ax = ax1);\n",
    "ax1.set_title('Histogram', fontsize=15)\n",
    "ax1.set_xlabel('Feature 0')\n",
    "\n",
    "stats.probplot(df_train[0], plot = ax2)\n",
    "ax2.set_title('Normal probability plot', fontsize=15)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Normal probability plot')"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(nrows = 1, ncols = 2, figsize = (20,8))\n",
    "\n",
    "sns.distplot(df_train[1], fit=norm, ax = ax1);\n",
    "ax1.set_title('Histogram', fontsize=15)\n",
    "ax1.set_xlabel('Feature 1')\n",
    "\n",
    "stats.probplot(df_train[1], plot = ax2)\n",
    "ax2.set_title('Normal probability plot', fontsize=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Normal probability plot')"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(nrows = 1, ncols = 2, figsize = (20,8))\n",
    "\n",
    "sns.distplot(df_train[14], fit=norm, ax = ax1);\n",
    "ax1.set_title('Histogram', fontsize=15)\n",
    "ax1.set_xlabel('Feature 14')\n",
    "\n",
    "stats.probplot(df_train[14], plot = ax2)\n",
    "ax2.set_title('Normal probability plot', fontsize=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Normal probability plot')"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(nrows = 1, ncols = 2, figsize = (20,8))\n",
    "\n",
    "sns.distplot(df_train[3], fit=norm, ax = ax1);\n",
    "ax1.set_title('Histogram', fontsize=15)\n",
    "\n",
    "stats.probplot(df_train[3], plot = ax2)\n",
    "ax2.set_title('Normal probability plot', fontsize=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Normal probability plot')"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(nrows = 1, ncols = 2, figsize = (20,8))\n",
    "\n",
    "sns.distplot(df_train[89], fit=norm, ax = ax1);\n",
    "ax1.set_title('Histogram', fontsize=15)\n",
    "\n",
    "stats.probplot(df_train[89], plot = ax2)\n",
    "ax2.set_title('Normal probability plot', fontsize=15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Analyzing feature skewness"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features with positive skewness: [ 0.  1.  2.  3.  4.  5.  6.  7.  8.  9. 10. 11. 12. 13. 17. 20. 21. 23.\n",
      " 24. 25. 33. 35. 38. 39. 43. 44. 45. 46. 47. 48. 49. 50. 51. 53. 54. 55.\n",
      " 56. 59. 61. 64. 68. 70. 71. 75. 76. 77. 81. 83. 85. 87. 89.]\n",
      "Features with negative skewness: [14. 15. 16. 18. 19. 22. 26. 27. 28. 29. 30. 31. 32. 34. 36. 37. 40. 41.\n",
      " 42. 52. 57. 58. 60. 62. 63. 65. 66. 67. 69. 72. 73. 74. 78. 79. 80. 82.\n",
      " 84. 86. 88.]\n"
     ]
    }
   ],
   "source": [
    "positive_skewness = np.array([])\n",
    "negative_skewness = np.array([])\n",
    "for i in range(0,90):\n",
    "    if df_train[i].skew()>0:\n",
    "        positive_skewness = np.append(positive_skewness, i)\n",
    "    else:\n",
    "        negative_skewness = np.append(negative_skewness, i)\n",
    "print(\"Features with positive skewness:\", positive_skewness)\n",
    "print(\"Features with negative skewness:\", negative_skewness)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Validation set will be 20% of the train set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_tr, X_val, y_tr, y_val = train_test_split(x_train, y_train, test_size=0.2, random_state=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Solution was tested using StandardScaler and MinMaxScaler but StandardScaler showed much better results with all models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "x_train_p = sc.fit_transform(X_tr)\n",
    "x_val_p = sc.fit_transform(X_val)\n",
    "x_test_p = sc.fit_transform(x_test)\n",
    "\n",
    "# m_scaler = MinMaxScaler()\n",
    "# x_train_p = m_scaler.fit_transform(df_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>80</th>\n",
       "      <th>81</th>\n",
       "      <th>82</th>\n",
       "      <th>83</th>\n",
       "      <th>84</th>\n",
       "      <th>85</th>\n",
       "      <th>86</th>\n",
       "      <th>87</th>\n",
       "      <th>88</th>\n",
       "      <th>89</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>...</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "      <td>3.200000e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-1.192093e-09</td>\n",
       "      <td>-2.145767e-09</td>\n",
       "      <td>-6.437302e-09</td>\n",
       "      <td>1.096725e-08</td>\n",
       "      <td>-8.106232e-09</td>\n",
       "      <td>4.768371e-09</td>\n",
       "      <td>-2.861023e-09</td>\n",
       "      <td>8.225441e-09</td>\n",
       "      <td>-4.410744e-09</td>\n",
       "      <td>-4.827976e-09</td>\n",
       "      <td>...</td>\n",
       "      <td>-9.298325e-09</td>\n",
       "      <td>-3.039837e-09</td>\n",
       "      <td>5.960465e-09</td>\n",
       "      <td>-4.410744e-09</td>\n",
       "      <td>-4.529953e-09</td>\n",
       "      <td>8.702278e-09</td>\n",
       "      <td>-1.907349e-09</td>\n",
       "      <td>-5.245209e-09</td>\n",
       "      <td>8.642673e-09</td>\n",
       "      <td>9.179115e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "      <td>1.000016e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-1.533383e+00</td>\n",
       "      <td>-1.401599e+00</td>\n",
       "      <td>-1.569830e+00</td>\n",
       "      <td>-1.329826e+00</td>\n",
       "      <td>-1.866314e+00</td>\n",
       "      <td>-1.551972e+00</td>\n",
       "      <td>-1.860693e+00</td>\n",
       "      <td>-1.673190e+00</td>\n",
       "      <td>-1.789662e+00</td>\n",
       "      <td>-1.852210e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>-6.853447e+00</td>\n",
       "      <td>-6.810141e+00</td>\n",
       "      <td>-6.950956e+00</td>\n",
       "      <td>-4.950410e+00</td>\n",
       "      <td>-8.019403e+00</td>\n",
       "      <td>-6.206920e+00</td>\n",
       "      <td>-8.579938e+00</td>\n",
       "      <td>-6.151473e+00</td>\n",
       "      <td>-7.425687e+00</td>\n",
       "      <td>-7.694020e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-7.047455e-01</td>\n",
       "      <td>-6.382330e-01</td>\n",
       "      <td>-6.761670e-01</td>\n",
       "      <td>-6.190048e-01</td>\n",
       "      <td>-6.438559e-01</td>\n",
       "      <td>-6.764369e-01</td>\n",
       "      <td>-6.653898e-01</td>\n",
       "      <td>-6.329759e-01</td>\n",
       "      <td>-6.381733e-01</td>\n",
       "      <td>-6.567249e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.745384e-01</td>\n",
       "      <td>-6.061778e-01</td>\n",
       "      <td>-6.032310e-01</td>\n",
       "      <td>-6.885052e-01</td>\n",
       "      <td>-5.867357e-01</td>\n",
       "      <td>-5.901520e-01</td>\n",
       "      <td>-5.892177e-01</td>\n",
       "      <td>-6.590019e-01</td>\n",
       "      <td>-5.798183e-01</td>\n",
       "      <td>-6.155681e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-1.961701e-01</td>\n",
       "      <td>-2.441718e-01</td>\n",
       "      <td>-2.211864e-01</td>\n",
       "      <td>-2.333873e-01</td>\n",
       "      <td>-1.988320e-01</td>\n",
       "      <td>-2.322921e-01</td>\n",
       "      <td>-1.900618e-01</td>\n",
       "      <td>-2.084275e-01</td>\n",
       "      <td>-2.063692e-01</td>\n",
       "      <td>-1.829548e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>3.685155e-02</td>\n",
       "      <td>-7.960736e-02</td>\n",
       "      <td>3.410846e-02</td>\n",
       "      <td>-1.088998e-01</td>\n",
       "      <td>2.392795e-02</td>\n",
       "      <td>9.298738e-03</td>\n",
       "      <td>-1.699594e-03</td>\n",
       "      <td>-3.137111e-02</td>\n",
       "      <td>2.972183e-02</td>\n",
       "      <td>-2.421560e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4.448358e-01</td>\n",
       "      <td>3.312985e-01</td>\n",
       "      <td>4.158741e-01</td>\n",
       "      <td>3.399055e-01</td>\n",
       "      <td>4.041258e-01</td>\n",
       "      <td>4.034996e-01</td>\n",
       "      <td>4.350799e-01</td>\n",
       "      <td>3.594742e-01</td>\n",
       "      <td>3.814051e-01</td>\n",
       "      <td>4.252786e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>6.044971e-01</td>\n",
       "      <td>4.929237e-01</td>\n",
       "      <td>6.246895e-01</td>\n",
       "      <td>5.551307e-01</td>\n",
       "      <td>6.237212e-01</td>\n",
       "      <td>5.973359e-01</td>\n",
       "      <td>5.948678e-01</td>\n",
       "      <td>6.269127e-01</td>\n",
       "      <td>6.061147e-01</td>\n",
       "      <td>5.923088e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.169545e+01</td>\n",
       "      <td>1.844036e+01</td>\n",
       "      <td>1.209160e+01</td>\n",
       "      <td>2.976400e+01</td>\n",
       "      <td>1.602910e+01</td>\n",
       "      <td>1.347296e+01</td>\n",
       "      <td>1.839395e+01</td>\n",
       "      <td>2.538504e+01</td>\n",
       "      <td>2.037012e+01</td>\n",
       "      <td>2.235022e+01</td>\n",
       "      <td>...</td>\n",
       "      <td>6.970065e+00</td>\n",
       "      <td>2.147675e+01</td>\n",
       "      <td>9.362054e+00</td>\n",
       "      <td>7.358592e+00</td>\n",
       "      <td>9.491427e+00</td>\n",
       "      <td>1.678839e+01</td>\n",
       "      <td>6.952916e+00</td>\n",
       "      <td>6.886868e+00</td>\n",
       "      <td>2.091715e+01</td>\n",
       "      <td>6.212358e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 90 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 0             1             2             3             4   \\\n",
       "count  3.200000e+04  3.200000e+04  3.200000e+04  3.200000e+04  3.200000e+04   \n",
       "mean  -1.192093e-09 -2.145767e-09 -6.437302e-09  1.096725e-08 -8.106232e-09   \n",
       "std    1.000016e+00  1.000016e+00  1.000016e+00  1.000016e+00  1.000016e+00   \n",
       "min   -1.533383e+00 -1.401599e+00 -1.569830e+00 -1.329826e+00 -1.866314e+00   \n",
       "25%   -7.047455e-01 -6.382330e-01 -6.761670e-01 -6.190048e-01 -6.438559e-01   \n",
       "50%   -1.961701e-01 -2.441718e-01 -2.211864e-01 -2.333873e-01 -1.988320e-01   \n",
       "75%    4.448358e-01  3.312985e-01  4.158741e-01  3.399055e-01  4.041258e-01   \n",
       "max    1.169545e+01  1.844036e+01  1.209160e+01  2.976400e+01  1.602910e+01   \n",
       "\n",
       "                 5             6             7             8             9   \\\n",
       "count  3.200000e+04  3.200000e+04  3.200000e+04  3.200000e+04  3.200000e+04   \n",
       "mean   4.768371e-09 -2.861023e-09  8.225441e-09 -4.410744e-09 -4.827976e-09   \n",
       "std    1.000016e+00  1.000016e+00  1.000016e+00  1.000016e+00  1.000016e+00   \n",
       "min   -1.551972e+00 -1.860693e+00 -1.673190e+00 -1.789662e+00 -1.852210e+00   \n",
       "25%   -6.764369e-01 -6.653898e-01 -6.329759e-01 -6.381733e-01 -6.567249e-01   \n",
       "50%   -2.322921e-01 -1.900618e-01 -2.084275e-01 -2.063692e-01 -1.829548e-01   \n",
       "75%    4.034996e-01  4.350799e-01  3.594742e-01  3.814051e-01  4.252786e-01   \n",
       "max    1.347296e+01  1.839395e+01  2.538504e+01  2.037012e+01  2.235022e+01   \n",
       "\n",
       "       ...            80            81            82            83  \\\n",
       "count  ...  3.200000e+04  3.200000e+04  3.200000e+04  3.200000e+04   \n",
       "mean   ... -9.298325e-09 -3.039837e-09  5.960465e-09 -4.410744e-09   \n",
       "std    ...  1.000016e+00  1.000016e+00  1.000016e+00  1.000016e+00   \n",
       "min    ... -6.853447e+00 -6.810141e+00 -6.950956e+00 -4.950410e+00   \n",
       "25%    ... -5.745384e-01 -6.061778e-01 -6.032310e-01 -6.885052e-01   \n",
       "50%    ...  3.685155e-02 -7.960736e-02  3.410846e-02 -1.088998e-01   \n",
       "75%    ...  6.044971e-01  4.929237e-01  6.246895e-01  5.551307e-01   \n",
       "max    ...  6.970065e+00  2.147675e+01  9.362054e+00  7.358592e+00   \n",
       "\n",
       "                 84            85            86            87            88  \\\n",
       "count  3.200000e+04  3.200000e+04  3.200000e+04  3.200000e+04  3.200000e+04   \n",
       "mean  -4.529953e-09  8.702278e-09 -1.907349e-09 -5.245209e-09  8.642673e-09   \n",
       "std    1.000016e+00  1.000016e+00  1.000016e+00  1.000016e+00  1.000016e+00   \n",
       "min   -8.019403e+00 -6.206920e+00 -8.579938e+00 -6.151473e+00 -7.425687e+00   \n",
       "25%   -5.867357e-01 -5.901520e-01 -5.892177e-01 -6.590019e-01 -5.798183e-01   \n",
       "50%    2.392795e-02  9.298738e-03 -1.699594e-03 -3.137111e-02  2.972183e-02   \n",
       "75%    6.237212e-01  5.973359e-01  5.948678e-01  6.269127e-01  6.061147e-01   \n",
       "max    9.491427e+00  1.678839e+01  6.952916e+00  6.886868e+00  2.091715e+01   \n",
       "\n",
       "                 89  \n",
       "count  3.200000e+04  \n",
       "mean   9.179115e-09  \n",
       "std    1.000016e+00  \n",
       "min   -7.694020e+00  \n",
       "25%   -6.155681e-01  \n",
       "50%   -2.421560e-02  \n",
       "75%    5.923088e-01  \n",
       "max    6.212358e+00  \n",
       "\n",
       "[8 rows x 90 columns]"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train_p = pd.DataFrame(x_train_p)\n",
    "df_train_p.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "le = LabelEncoder()\n",
    "le_train = le.fit_transform(y_tr)\n",
    "le_val = le.fit_transform(y_val)\n",
    "le_test = le.fit_transform(y_test)\n",
    "\n",
    "oh = OneHotEncoder(sparse=False)\n",
    "\n",
    "le_train = le_train.reshape(len(le_train), 1)\n",
    "y_train_p = oh.fit_transform(le_train)\n",
    "\n",
    "\n",
    "le_val = le_val.reshape(len(le_val), 1)\n",
    "y_val_p = oh.fit_transform(le_val)\n",
    "\n",
    "le_test = le_test.reshape(len(le_test), 1)\n",
    "y_test_p = oh.fit_transform(le_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = pd.DataFrame({})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 1.1410 - accuracy: 0.5033 - val_loss: 1.0141 - val_accuracy: 0.5730\n",
      "Epoch 2/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 1.0221 - accuracy: 0.5718 - val_loss: 0.9910 - val_accuracy: 0.5828\n",
      "Epoch 3/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.9870 - accuracy: 0.5883 - val_loss: 0.9669 - val_accuracy: 0.6000\n",
      "Epoch 4/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.9671 - accuracy: 0.5975 - val_loss: 0.9543 - val_accuracy: 0.6047\n",
      "Epoch 5/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.9506 - accuracy: 0.6040 - val_loss: 0.9451 - val_accuracy: 0.6122\n",
      "Epoch 6/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.9385 - accuracy: 0.6128 - val_loss: 0.9354 - val_accuracy: 0.6145\n",
      "Epoch 7/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.9272 - accuracy: 0.6166 - val_loss: 0.9362 - val_accuracy: 0.6191\n",
      "Epoch 8/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.9199 - accuracy: 0.6197 - val_loss: 0.9263 - val_accuracy: 0.6202\n",
      "Epoch 9/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.9084 - accuracy: 0.6272 - val_loss: 0.9230 - val_accuracy: 0.6223\n",
      "Epoch 10/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.9039 - accuracy: 0.6306 - val_loss: 0.9254 - val_accuracy: 0.6204\n",
      "Epoch 11/60\n",
      "1000/1000 [==============================] - 4s 4ms/step - loss: 0.8935 - accuracy: 0.6300 - val_loss: 0.9231 - val_accuracy: 0.6236\n",
      "Epoch 12/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.8859 - accuracy: 0.6379 - val_loss: 0.9204 - val_accuracy: 0.6270\n",
      "Epoch 13/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.8821 - accuracy: 0.6362 - val_loss: 0.9210 - val_accuracy: 0.6269\n",
      "Epoch 14/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.8805 - accuracy: 0.6387 - val_loss: 0.9221 - val_accuracy: 0.6261\n",
      "Epoch 15/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.8665 - accuracy: 0.6437 - val_loss: 0.9109 - val_accuracy: 0.6298\n",
      "Epoch 16/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.8616 - accuracy: 0.6460 - val_loss: 0.9092 - val_accuracy: 0.6321\n",
      "Epoch 17/60\n",
      "1000/1000 [==============================] - 4s 4ms/step - loss: 0.8604 - accuracy: 0.6483 - val_loss: 0.9090 - val_accuracy: 0.6315\n",
      "Epoch 18/60\n",
      "1000/1000 [==============================] - 4s 4ms/step - loss: 0.8563 - accuracy: 0.6510 - val_loss: 0.9125 - val_accuracy: 0.6299\n",
      "Epoch 19/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.8493 - accuracy: 0.6542 - val_loss: 0.9138 - val_accuracy: 0.6329\n",
      "Epoch 20/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.8457 - accuracy: 0.6562 - val_loss: 0.9129 - val_accuracy: 0.6310\n",
      "Epoch 21/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.8426 - accuracy: 0.6579 - val_loss: 0.9106 - val_accuracy: 0.6344\n",
      "Epoch 22/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8386 - accuracy: 0.6605 - val_loss: 0.9080 - val_accuracy: 0.6319\n",
      "Epoch 23/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8319 - accuracy: 0.6592 - val_loss: 0.9148 - val_accuracy: 0.6325\n",
      "Epoch 24/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8281 - accuracy: 0.6645 - val_loss: 0.9111 - val_accuracy: 0.6280\n",
      "Epoch 25/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8289 - accuracy: 0.6622 - val_loss: 0.9131 - val_accuracy: 0.6270\n",
      "Epoch 26/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8206 - accuracy: 0.6669 - val_loss: 0.9044 - val_accuracy: 0.6323\n",
      "Epoch 27/60\n",
      "1000/1000 [==============================] - 3s 3ms/step - loss: 0.8191 - accuracy: 0.6685 - val_loss: 0.9114 - val_accuracy: 0.6296\n",
      "Epoch 28/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8167 - accuracy: 0.6686 - val_loss: 0.9150 - val_accuracy: 0.6235\n",
      "Epoch 29/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8098 - accuracy: 0.6698 - val_loss: 0.9060 - val_accuracy: 0.6331\n",
      "Epoch 30/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8069 - accuracy: 0.6746 - val_loss: 0.9014 - val_accuracy: 0.6355\n",
      "Epoch 31/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8085 - accuracy: 0.6744 - val_loss: 0.9107 - val_accuracy: 0.6357\n",
      "Epoch 32/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8107 - accuracy: 0.6700 - val_loss: 0.9055 - val_accuracy: 0.6317\n",
      "Epoch 33/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.8024 - accuracy: 0.6751 - val_loss: 0.9038 - val_accuracy: 0.6330\n",
      "Epoch 34/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.7991 - accuracy: 0.6763 - val_loss: 0.9141 - val_accuracy: 0.6326\n",
      "Epoch 35/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.7950 - accuracy: 0.6772 - val_loss: 0.9089 - val_accuracy: 0.6309\n",
      "Epoch 36/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.7892 - accuracy: 0.6828 - val_loss: 0.9061 - val_accuracy: 0.6316\n",
      "Epoch 37/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.7910 - accuracy: 0.6801 - val_loss: 0.9153 - val_accuracy: 0.6308\n",
      "Epoch 38/60\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.7919 - accuracy: 0.6793 - val_loss: 0.9097 - val_accuracy: 0.6361\n",
      "Epoch 39/60\n",
      " 967/1000 [============================>.] - ETA: 0s - loss: 0.7802 - accuracy: 0.6837Restoring model weights from the end of the best epoch.\n",
      "1000/1000 [==============================] - 2s 2ms/step - loss: 0.7823 - accuracy: 0.6832 - val_loss: 0.9092 - val_accuracy: 0.6309\n",
      "Epoch 00039: early stopping\n"
     ]
    }
   ],
   "source": [
    "model = tf.keras.models.Sequential()\n",
    "model.add(tf.keras.layers.Dense(256, input_shape=(90,), activation='relu'))\n",
    "model.add(tf.keras.layers.Dropout(0.5))\n",
    "model.add(tf.keras.layers.Dense(64, activation='relu'))\n",
    "model.add(tf.keras.layers.Dense(16, activation='relu', \n",
    "                                bias_regularizer=regularizers.l2(l2=0.1)))\n",
    "model.add(tf.keras.layers.Dense(4, activation='softmax'))\n",
    "\n",
    "\n",
    "model.compile(optimizer= tf.keras.optimizers.Adam(learning_rate = 0.001),\n",
    "              loss=tf.keras.losses.CategoricalCrossentropy(),\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "#Early stopping with monitoring validation loss, and restoring best weights\n",
    "earlyStop = tf.keras.callbacks.EarlyStopping(\n",
    "    monitor='val_loss', \n",
    "    min_delta=0.005, \n",
    "    patience=13, \n",
    "    verbose=1,\n",
    "    restore_best_weights=True)\n",
    "\n",
    "train_history = model.fit(\n",
    "    x_train_p, y_train_p, \n",
    "    epochs = 60,\n",
    "    batch_size = 32, \n",
    "    validation_data=(x_val_p, y_val_p), \n",
    "    callbacks=[earlyStop])\n",
    "\n",
    "val_loss_arr = np.array(train_history.history['val_loss'])\n",
    "idx = np.argmin(val_loss_arr)\n",
    "\n",
    "temp = pd.DataFrame({\n",
    "                'model': ['M4'],\n",
    "                'train acc': [train_history.history['accuracy'][idx]],\n",
    "                'train loss': [train_history.history['loss'][idx]],\n",
    "                'val acc': [train_history.history['val_accuracy'][idx]],\n",
    "                'val loss': [train_history.history['val_loss'][idx]]\n",
    "              })\n",
    "\n",
    "results = pd.concat([results, temp])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "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>model</th>\n",
       "      <th>train acc</th>\n",
       "      <th>train loss</th>\n",
       "      <th>val acc</th>\n",
       "      <th>val loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>M4</td>\n",
       "      <td>0.674625</td>\n",
       "      <td>0.806931</td>\n",
       "      <td>0.6355</td>\n",
       "      <td>0.901354</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  model  train acc  train loss  val acc  val loss\n",
       "0    M4   0.674625    0.806931   0.6355  0.901354"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(train_history.history['loss'])\n",
    "plt.plot(train_history.history['val_loss'])\n",
    "plt.title('Model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train loss', 'val loss'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "63/63 [==============================] - 0s 1ms/step - loss: 0.8839 - accuracy: 0.6400\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.8839498162269592, 0.6399999856948853]"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(x_test_p, y_test_p, batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = model.predict(x_test_p)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[3 0 0 ... 0 1 3]\n",
      "[2 0 0 ... 3 1 2]\n"
     ]
    }
   ],
   "source": [
    "y = tf.argmax(y_pred, axis=-1)\n",
    "print(np.array(le_test[:,0]))\n",
    "print(np.array(y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "cf = sklearn.metrics.multilabel_confusion_matrix(np.array(le_test[:,0]), np.array(y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[2728,  272],\n",
       "        [ 237,  763]],\n",
       "\n",
       "       [[2545,  455],\n",
       "        [ 321,  679]],\n",
       "\n",
       "       [[2658,  342],\n",
       "        [ 436,  564]],\n",
       "\n",
       "       [[2629,  371],\n",
       "        [ 446,  554]]], dtype=int64)"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "c = pd.DataFrame({})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "1\n",
      "2\n",
      "3\n"
     ]
    }
   ],
   "source": [
    "for i in range(0,4):\n",
    "    c = pd.concat([c, pd.DataFrame(cf[i])])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2728</td>\n",
       "      <td>272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>237</td>\n",
       "      <td>763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2545</td>\n",
       "      <td>455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>321</td>\n",
       "      <td>679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2658</td>\n",
       "      <td>342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>436</td>\n",
       "      <td>564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2629</td>\n",
       "      <td>371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>446</td>\n",
       "      <td>554</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      0    1\n",
       "0  2728  272\n",
       "1   237  763\n",
       "0  2545  455\n",
       "1   321  679\n",
       "0  2658  342\n",
       "1   436  564\n",
       "0  2629  371\n",
       "1   446  554"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c"
   ]
  }
 ],
 "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": "3.7.4"
  },
  "pycharm": {
   "stem_cell": {
    "cell_type": "raw",
    "metadata": {
     "collapsed": false
    },
    "source": []
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}