regression_model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Install Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install -q matplotlib numpy pandas pathlib seaborn\n",
    "!pip install -q tensorflow \n",
    "!pip install -q git+https://github.com/tensorflow/docs\n",
    "\n",
    "# !pip install -r -q requirements.txt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "pycharm": {
     "is_executing": false
    }
   },
   "outputs": [],
   "source": [
    "from __future__ import absolute_import, division, print_function, unicode_literals\n",
    "\n",
    "import itertools\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.metrics import cohen_kappa_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.0.0\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import tensorflow_docs as tfdocs\n",
    "import tensorflow_docs.plots\n",
    "import tensorflow_docs.modeling\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "\n",
    "assert float(tf.__version__.split(\".\", 1)[0]) >= 2.0, \"Please use Tensorflow version 2!\"\n",
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load the transformed data\n",
    "Take the pkl file with extra feature columns into a dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "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>essay_set</th>\n",
       "      <th>essay</th>\n",
       "      <th>domain1_score</th>\n",
       "      <th>essay_length</th>\n",
       "      <th>avg_sentence_length</th>\n",
       "      <th>avg_word_length</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>noun_phrases</th>\n",
       "      <th>syntax_errors</th>\n",
       "      <th>readability_index</th>\n",
       "      <th>difficult_words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8074</th>\n",
       "      <td>5</td>\n",
       "      <td>In the memoir, \"Narciso Rodriguez\" by Narciso ...</td>\n",
       "      <td>4</td>\n",
       "      <td>231.0</td>\n",
       "      <td>16.357143</td>\n",
       "      <td>4.471861</td>\n",
       "      <td>0.082832</td>\n",
       "      <td>12.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>26.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10912</th>\n",
       "      <td>7</td>\n",
       "      <td>The time I was patience was when I was @NUM1 y...</td>\n",
       "      <td>16</td>\n",
       "      <td>23.0</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>4.608696</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "      <td>Did you know that more and more people these d...</td>\n",
       "      <td>10</td>\n",
       "      <td>43.0</td>\n",
       "      <td>14.333333</td>\n",
       "      <td>4.395349</td>\n",
       "      <td>0.027083</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.8</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6767</th>\n",
       "      <td>4</td>\n",
       "      <td>The author concludes the story with he paragra...</td>\n",
       "      <td>1</td>\n",
       "      <td>411.0</td>\n",
       "      <td>21.473684</td>\n",
       "      <td>4.990268</td>\n",
       "      <td>0.250740</td>\n",
       "      <td>48.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>14.3</td>\n",
       "      <td>59.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4441</th>\n",
       "      <td>3</td>\n",
       "      <td>There are many ways that the features of the s...</td>\n",
       "      <td>2</td>\n",
       "      <td>87.0</td>\n",
       "      <td>43.500000</td>\n",
       "      <td>4.022989</td>\n",
       "      <td>-0.152778</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.8</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1762</th>\n",
       "      <td>1</td>\n",
       "      <td>Guaranteed, @NUM1 years from now we will still...</td>\n",
       "      <td>9</td>\n",
       "      <td>189.0</td>\n",
       "      <td>18.600000</td>\n",
       "      <td>4.455026</td>\n",
       "      <td>0.406000</td>\n",
       "      <td>17.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>25.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1522</th>\n",
       "      <td>1</td>\n",
       "      <td>Dear @CAPS1 @CAPS2 @CAPS3, @CAPS4 many people ...</td>\n",
       "      <td>10</td>\n",
       "      <td>73.0</td>\n",
       "      <td>14.600000</td>\n",
       "      <td>4.534247</td>\n",
       "      <td>0.123333</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>7.7</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>296</th>\n",
       "      <td>1</td>\n",
       "      <td>Dear local Newspaper, I assume you've heard ab...</td>\n",
       "      <td>9</td>\n",
       "      <td>30.0</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>3.566667</td>\n",
       "      <td>0.102778</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.9</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1632</th>\n",
       "      <td>1</td>\n",
       "      <td>To @CAPS1 @CAPS2 @ORGANIZATION1, Computers are...</td>\n",
       "      <td>8</td>\n",
       "      <td>166.0</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>4.114458</td>\n",
       "      <td>0.029167</td>\n",
       "      <td>8.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>25.8</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>931</th>\n",
       "      <td>1</td>\n",
       "      <td>Dear @CAPS1, I strongly believe computers have...</td>\n",
       "      <td>8</td>\n",
       "      <td>53.0</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>3.962264</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10.8</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12976 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       essay_set                                              essay  \\\n",
       "8074           5  In the memoir, \"Narciso Rodriguez\" by Narciso ...   \n",
       "10912          7  The time I was patience was when I was @NUM1 y...   \n",
       "6              1  Did you know that more and more people these d...   \n",
       "6767           4  The author concludes the story with he paragra...   \n",
       "4441           3  There are many ways that the features of the s...   \n",
       "...          ...                                                ...   \n",
       "1762           1  Guaranteed, @NUM1 years from now we will still...   \n",
       "1522           1  Dear @CAPS1 @CAPS2 @CAPS3, @CAPS4 many people ...   \n",
       "296            1  Dear local Newspaper, I assume you've heard ab...   \n",
       "1632           1  To @CAPS1 @CAPS2 @ORGANIZATION1, Computers are...   \n",
       "931            1  Dear @CAPS1, I strongly believe computers have...   \n",
       "\n",
       "       domain1_score  essay_length  avg_sentence_length  avg_word_length  \\\n",
       "8074               4         231.0            16.357143         4.471861   \n",
       "10912             16          23.0            23.000000         4.608696   \n",
       "6                 10          43.0            14.333333         4.395349   \n",
       "6767               1         411.0            21.473684         4.990268   \n",
       "4441               2          87.0            43.500000         4.022989   \n",
       "...              ...           ...                  ...              ...   \n",
       "1762               9         189.0            18.600000         4.455026   \n",
       "1522              10          73.0            14.600000         4.534247   \n",
       "296                9          30.0            10.000000         3.566667   \n",
       "1632               8         166.0            27.000000         4.114458   \n",
       "931                8          53.0            26.000000         3.962264   \n",
       "\n",
       "       sentiment  noun_phrases  syntax_errors  readability_index  \\\n",
       "8074    0.082832          12.0           12.0               11.0   \n",
       "10912   0.000000           1.0            0.0               12.0   \n",
       "6       0.027083           2.0            2.0                6.8   \n",
       "6767    0.250740          48.0           14.0               14.3   \n",
       "4441   -0.152778           4.0            4.0               19.8   \n",
       "...          ...           ...            ...                ...   \n",
       "1762    0.406000          17.0           10.0               11.0   \n",
       "1522    0.123333           7.0            4.0                7.7   \n",
       "296     0.102778           2.0            1.0                0.9   \n",
       "1632    0.029167           8.0           18.0               25.8   \n",
       "931     0.000000           2.0            1.0               10.8   \n",
       "\n",
       "       difficult_words  \n",
       "8074              26.0  \n",
       "10912              5.0  \n",
       "6                  5.0  \n",
       "6767              59.0  \n",
       "4441               6.0  \n",
       "...                ...  \n",
       "1762              25.0  \n",
       "1522              12.0  \n",
       "296                2.0  \n",
       "1632              10.0  \n",
       "931                3.0  \n",
       "\n",
       "[12976 rows x 11 columns]"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset = pd.read_pickle('output/training_set_rel3.pkl')\n",
    "dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Inspect Columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "essay_set                int64\n",
       "essay                   object\n",
       "domain1_score            int64\n",
       "essay_length           float64\n",
       "avg_sentence_length    float64\n",
       "avg_word_length        float64\n",
       "sentiment              float64\n",
       "noun_phrases           float64\n",
       "syntax_errors          float64\n",
       "readability_index      float64\n",
       "difficult_words        float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "essay_set              False\n",
       "essay                  False\n",
       "domain1_score          False\n",
       "essay_length           False\n",
       "avg_sentence_length    False\n",
       "avg_word_length        False\n",
       "sentiment              False\n",
       "noun_phrases           False\n",
       "syntax_errors          False\n",
       "readability_index      False\n",
       "difficult_words        False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.isna().sum() > 0 "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">You must extract a minimum of three different types of features.\n",
    "\n",
    "Please see `data_etl.ipynb` for details.\n",
    "```\n",
    "meta_features = ['essay_length', 'avg_sentence_length', 'avg_word_length']\n",
    "grammar_features = ['sentiment', 'noun_phrases', 'syntax_errors']\n",
    "redability_features = ['readability_index', 'difficult_words']\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# dataset = dataset.dropna(axis='columns').drop(columns=['essay', 'essay_set'])\n",
    "dataset = dataset.dropna(axis='columns').drop(columns=['essay'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('essay_set',),\n",
       " ('essay_length',),\n",
       " ('avg_sentence_length',),\n",
       " ('avg_word_length',),\n",
       " ('sentiment',),\n",
       " ('noun_phrases',),\n",
       " ('syntax_errors',),\n",
       " ('readability_index',),\n",
       " ('difficult_words',),\n",
       " ('essay_set', 'essay_length'),\n",
       " ('essay_set', 'avg_sentence_length'),\n",
       " ('essay_set', 'avg_word_length'),\n",
       " ('essay_set', 'sentiment'),\n",
       " ('essay_set', 'noun_phrases'),\n",
       " ('essay_set', 'syntax_errors'),\n",
       " ('essay_set', 'readability_index'),\n",
       " ('essay_set', 'difficult_words'),\n",
       " ('essay_length', 'avg_sentence_length'),\n",
       " ('essay_length', 'avg_word_length'),\n",
       " ('essay_length', 'sentiment'),\n",
       " ('essay_length', 'noun_phrases'),\n",
       " ('essay_length', 'syntax_errors'),\n",
       " ('essay_length', 'readability_index'),\n",
       " ('essay_length', 'difficult_words'),\n",
       " ('avg_sentence_length', 'avg_word_length'),\n",
       " ('avg_sentence_length', 'sentiment'),\n",
       " ('avg_sentence_length', 'noun_phrases'),\n",
       " ('avg_sentence_length', 'syntax_errors'),\n",
       " ('avg_sentence_length', 'readability_index'),\n",
       " ('avg_sentence_length', 'difficult_words'),\n",
       " ('avg_word_length', 'sentiment'),\n",
       " ('avg_word_length', 'noun_phrases'),\n",
       " ('avg_word_length', 'syntax_errors'),\n",
       " ('avg_word_length', 'readability_index'),\n",
       " ('avg_word_length', 'difficult_words'),\n",
       " ('sentiment', 'noun_phrases'),\n",
       " ('sentiment', 'syntax_errors'),\n",
       " ('sentiment', 'readability_index'),\n",
       " ('sentiment', 'difficult_words'),\n",
       " ('noun_phrases', 'syntax_errors'),\n",
       " ('noun_phrases', 'readability_index'),\n",
       " ('noun_phrases', 'difficult_words'),\n",
       " ('syntax_errors', 'readability_index'),\n",
       " ('syntax_errors', 'difficult_words'),\n",
       " ('readability_index', 'difficult_words'),\n",
       " ('essay_set', 'essay_length', 'avg_sentence_length'),\n",
       " ('essay_set', 'essay_length', 'avg_word_length'),\n",
       " ('essay_set', 'essay_length', 'sentiment'),\n",
       " ('essay_set', 'essay_length', 'noun_phrases'),\n",
       " ('essay_set', 'essay_length', 'syntax_errors'),\n",
       " ('essay_set', 'essay_length', 'readability_index'),\n",
       " ('essay_set', 'essay_length', 'difficult_words'),\n",
       " ('essay_set', 'avg_sentence_length', 'avg_word_length'),\n",
       " ('essay_set', 'avg_sentence_length', 'sentiment'),\n",
       " ('essay_set', 'avg_sentence_length', 'noun_phrases'),\n",
       " ('essay_set', 'avg_sentence_length', 'syntax_errors'),\n",
       " ('essay_set', 'avg_sentence_length', 'readability_index'),\n",
       " ('essay_set', 'avg_sentence_length', 'difficult_words'),\n",
       " ('essay_set', 'avg_word_length', 'sentiment'),\n",
       " ('essay_set', 'avg_word_length', 'noun_phrases'),\n",
       " ('essay_set', 'avg_word_length', 'syntax_errors'),\n",
       " ('essay_set', 'avg_word_length', 'readability_index'),\n",
       " ('essay_set', 'avg_word_length', 'difficult_words'),\n",
       " ('essay_set', 'sentiment', 'noun_phrases'),\n",
       " ('essay_set', 'sentiment', 'syntax_errors'),\n",
       " ('essay_set', 'sentiment', 'readability_index'),\n",
       " ('essay_set', 'sentiment', 'difficult_words'),\n",
       " ('essay_set', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_set', 'noun_phrases', 'readability_index'),\n",
       " ('essay_set', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_set', 'syntax_errors', 'readability_index'),\n",
       " ('essay_set', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_set', 'readability_index', 'difficult_words'),\n",
       " ('essay_length', 'avg_sentence_length', 'avg_word_length'),\n",
       " ('essay_length', 'avg_sentence_length', 'sentiment'),\n",
       " ('essay_length', 'avg_sentence_length', 'noun_phrases'),\n",
       " ('essay_length', 'avg_sentence_length', 'syntax_errors'),\n",
       " ('essay_length', 'avg_sentence_length', 'readability_index'),\n",
       " ('essay_length', 'avg_sentence_length', 'difficult_words'),\n",
       " ('essay_length', 'avg_word_length', 'sentiment'),\n",
       " ('essay_length', 'avg_word_length', 'noun_phrases'),\n",
       " ('essay_length', 'avg_word_length', 'syntax_errors'),\n",
       " ('essay_length', 'avg_word_length', 'readability_index'),\n",
       " ('essay_length', 'avg_word_length', 'difficult_words'),\n",
       " ('essay_length', 'sentiment', 'noun_phrases'),\n",
       " ('essay_length', 'sentiment', 'syntax_errors'),\n",
       " ('essay_length', 'sentiment', 'readability_index'),\n",
       " ('essay_length', 'sentiment', 'difficult_words'),\n",
       " ('essay_length', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_length', 'noun_phrases', 'readability_index'),\n",
       " ('essay_length', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_length', 'syntax_errors', 'readability_index'),\n",
       " ('essay_length', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_length', 'readability_index', 'difficult_words'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'sentiment'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'noun_phrases'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'syntax_errors'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'readability_index'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'difficult_words'),\n",
       " ('avg_sentence_length', 'sentiment', 'noun_phrases'),\n",
       " ('avg_sentence_length', 'sentiment', 'syntax_errors'),\n",
       " ('avg_sentence_length', 'sentiment', 'readability_index'),\n",
       " ('avg_sentence_length', 'sentiment', 'difficult_words'),\n",
       " ('avg_sentence_length', 'noun_phrases', 'syntax_errors'),\n",
       " ('avg_sentence_length', 'noun_phrases', 'readability_index'),\n",
       " ('avg_sentence_length', 'noun_phrases', 'difficult_words'),\n",
       " ('avg_sentence_length', 'syntax_errors', 'readability_index'),\n",
       " ('avg_sentence_length', 'syntax_errors', 'difficult_words'),\n",
       " ('avg_sentence_length', 'readability_index', 'difficult_words'),\n",
       " ('avg_word_length', 'sentiment', 'noun_phrases'),\n",
       " ('avg_word_length', 'sentiment', 'syntax_errors'),\n",
       " ('avg_word_length', 'sentiment', 'readability_index'),\n",
       " ('avg_word_length', 'sentiment', 'difficult_words'),\n",
       " ('avg_word_length', 'noun_phrases', 'syntax_errors'),\n",
       " ('avg_word_length', 'noun_phrases', 'readability_index'),\n",
       " ('avg_word_length', 'noun_phrases', 'difficult_words'),\n",
       " ('avg_word_length', 'syntax_errors', 'readability_index'),\n",
       " ('avg_word_length', 'syntax_errors', 'difficult_words'),\n",
       " ('avg_word_length', 'readability_index', 'difficult_words'),\n",
       " ('sentiment', 'noun_phrases', 'syntax_errors'),\n",
       " ('sentiment', 'noun_phrases', 'readability_index'),\n",
       " ('sentiment', 'noun_phrases', 'difficult_words'),\n",
       " ('sentiment', 'syntax_errors', 'readability_index'),\n",
       " ('sentiment', 'syntax_errors', 'difficult_words'),\n",
       " ('sentiment', 'readability_index', 'difficult_words'),\n",
       " ('noun_phrases', 'syntax_errors', 'readability_index'),\n",
       " ('noun_phrases', 'syntax_errors', 'difficult_words'),\n",
       " ('noun_phrases', 'readability_index', 'difficult_words'),\n",
       " ('syntax_errors', 'readability_index', 'difficult_words'),\n",
       " ('essay_set', 'essay_length', 'avg_sentence_length', 'avg_word_length'),\n",
       " ('essay_set', 'essay_length', 'avg_sentence_length', 'sentiment'),\n",
       " ('essay_set', 'essay_length', 'avg_sentence_length', 'noun_phrases'),\n",
       " ('essay_set', 'essay_length', 'avg_sentence_length', 'syntax_errors'),\n",
       " ('essay_set', 'essay_length', 'avg_sentence_length', 'readability_index'),\n",
       " ('essay_set', 'essay_length', 'avg_sentence_length', 'difficult_words'),\n",
       " ('essay_set', 'essay_length', 'avg_word_length', 'sentiment'),\n",
       " ('essay_set', 'essay_length', 'avg_word_length', 'noun_phrases'),\n",
       " ('essay_set', 'essay_length', 'avg_word_length', 'syntax_errors'),\n",
       " ('essay_set', 'essay_length', 'avg_word_length', 'readability_index'),\n",
       " ('essay_set', 'essay_length', 'avg_word_length', 'difficult_words'),\n",
       " ('essay_set', 'essay_length', 'sentiment', 'noun_phrases'),\n",
       " ('essay_set', 'essay_length', 'sentiment', 'syntax_errors'),\n",
       " ('essay_set', 'essay_length', 'sentiment', 'readability_index'),\n",
       " ('essay_set', 'essay_length', 'sentiment', 'difficult_words'),\n",
       " ('essay_set', 'essay_length', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_set', 'essay_length', 'noun_phrases', 'readability_index'),\n",
       " ('essay_set', 'essay_length', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_set', 'essay_length', 'syntax_errors', 'readability_index'),\n",
       " ('essay_set', 'essay_length', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_set', 'essay_length', 'readability_index', 'difficult_words'),\n",
       " ('essay_set', 'avg_sentence_length', 'avg_word_length', 'sentiment'),\n",
       " ('essay_set', 'avg_sentence_length', 'avg_word_length', 'noun_phrases'),\n",
       " ('essay_set', 'avg_sentence_length', 'avg_word_length', 'syntax_errors'),\n",
       " ('essay_set', 'avg_sentence_length', 'avg_word_length', 'readability_index'),\n",
       " ('essay_set', 'avg_sentence_length', 'avg_word_length', 'difficult_words'),\n",
       " ('essay_set', 'avg_sentence_length', 'sentiment', 'noun_phrases'),\n",
       " ('essay_set', 'avg_sentence_length', 'sentiment', 'syntax_errors'),\n",
       " ('essay_set', 'avg_sentence_length', 'sentiment', 'readability_index'),\n",
       " ('essay_set', 'avg_sentence_length', 'sentiment', 'difficult_words'),\n",
       " ('essay_set', 'avg_sentence_length', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_set', 'avg_sentence_length', 'noun_phrases', 'readability_index'),\n",
       " ('essay_set', 'avg_sentence_length', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_set', 'avg_sentence_length', 'syntax_errors', 'readability_index'),\n",
       " ('essay_set', 'avg_sentence_length', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_set', 'avg_sentence_length', 'readability_index', 'difficult_words'),\n",
       " ('essay_set', 'avg_word_length', 'sentiment', 'noun_phrases'),\n",
       " ('essay_set', 'avg_word_length', 'sentiment', 'syntax_errors'),\n",
       " ('essay_set', 'avg_word_length', 'sentiment', 'readability_index'),\n",
       " ('essay_set', 'avg_word_length', 'sentiment', 'difficult_words'),\n",
       " ('essay_set', 'avg_word_length', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_set', 'avg_word_length', 'noun_phrases', 'readability_index'),\n",
       " ('essay_set', 'avg_word_length', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_set', 'avg_word_length', 'syntax_errors', 'readability_index'),\n",
       " ('essay_set', 'avg_word_length', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_set', 'avg_word_length', 'readability_index', 'difficult_words'),\n",
       " ('essay_set', 'sentiment', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_set', 'sentiment', 'noun_phrases', 'readability_index'),\n",
       " ('essay_set', 'sentiment', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_set', 'sentiment', 'syntax_errors', 'readability_index'),\n",
       " ('essay_set', 'sentiment', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_set', 'sentiment', 'readability_index', 'difficult_words'),\n",
       " ('essay_set', 'noun_phrases', 'syntax_errors', 'readability_index'),\n",
       " ('essay_set', 'noun_phrases', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_set', 'noun_phrases', 'readability_index', 'difficult_words'),\n",
       " ('essay_set', 'syntax_errors', 'readability_index', 'difficult_words'),\n",
       " ('essay_length', 'avg_sentence_length', 'avg_word_length', 'sentiment'),\n",
       " ('essay_length', 'avg_sentence_length', 'avg_word_length', 'noun_phrases'),\n",
       " ('essay_length', 'avg_sentence_length', 'avg_word_length', 'syntax_errors'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'readability_index'),\n",
       " ('essay_length', 'avg_sentence_length', 'avg_word_length', 'difficult_words'),\n",
       " ('essay_length', 'avg_sentence_length', 'sentiment', 'noun_phrases'),\n",
       " ('essay_length', 'avg_sentence_length', 'sentiment', 'syntax_errors'),\n",
       " ('essay_length', 'avg_sentence_length', 'sentiment', 'readability_index'),\n",
       " ('essay_length', 'avg_sentence_length', 'sentiment', 'difficult_words'),\n",
       " ('essay_length', 'avg_sentence_length', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_length', 'avg_sentence_length', 'noun_phrases', 'readability_index'),\n",
       " ('essay_length', 'avg_sentence_length', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_length', 'avg_sentence_length', 'syntax_errors', 'readability_index'),\n",
       " ('essay_length', 'avg_sentence_length', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length', 'avg_word_length', 'sentiment', 'noun_phrases'),\n",
       " ('essay_length', 'avg_word_length', 'sentiment', 'syntax_errors'),\n",
       " ('essay_length', 'avg_word_length', 'sentiment', 'readability_index'),\n",
       " ('essay_length', 'avg_word_length', 'sentiment', 'difficult_words'),\n",
       " ('essay_length', 'avg_word_length', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_length', 'avg_word_length', 'noun_phrases', 'readability_index'),\n",
       " ('essay_length', 'avg_word_length', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_length', 'avg_word_length', 'syntax_errors', 'readability_index'),\n",
       " ('essay_length', 'avg_word_length', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_length', 'avg_word_length', 'readability_index', 'difficult_words'),\n",
       " ('essay_length', 'sentiment', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_length', 'sentiment', 'noun_phrases', 'readability_index'),\n",
       " ('essay_length', 'sentiment', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_length', 'sentiment', 'syntax_errors', 'readability_index'),\n",
       " ('essay_length', 'sentiment', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_length', 'sentiment', 'readability_index', 'difficult_words'),\n",
       " ('essay_length', 'noun_phrases', 'syntax_errors', 'readability_index'),\n",
       " ('essay_length', 'noun_phrases', 'syntax_errors', 'difficult_words'),\n",
       " ('essay_length', 'noun_phrases', 'readability_index', 'difficult_words'),\n",
       " ('essay_length', 'syntax_errors', 'readability_index', 'difficult_words'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'sentiment', 'noun_phrases'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'sentiment', 'syntax_errors'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'sentiment', 'readability_index'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'sentiment', 'difficult_words'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'noun_phrases', 'syntax_errors'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('avg_sentence_length', 'avg_word_length', 'noun_phrases', 'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length', 'sentiment', 'noun_phrases', 'syntax_errors'),\n",
       " ('avg_sentence_length', 'sentiment', 'noun_phrases', 'readability_index'),\n",
       " ('avg_sentence_length', 'sentiment', 'noun_phrases', 'difficult_words'),\n",
       " ('avg_sentence_length', 'sentiment', 'syntax_errors', 'readability_index'),\n",
       " ('avg_sentence_length', 'sentiment', 'syntax_errors', 'difficult_words'),\n",
       " ('avg_sentence_length', 'sentiment', 'readability_index', 'difficult_words'),\n",
       " ('avg_sentence_length', 'noun_phrases', 'syntax_errors', 'readability_index'),\n",
       " ('avg_sentence_length', 'noun_phrases', 'syntax_errors', 'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_word_length', 'sentiment', 'noun_phrases', 'syntax_errors'),\n",
       " ('avg_word_length', 'sentiment', 'noun_phrases', 'readability_index'),\n",
       " ('avg_word_length', 'sentiment', 'noun_phrases', 'difficult_words'),\n",
       " ('avg_word_length', 'sentiment', 'syntax_errors', 'readability_index'),\n",
       " ('avg_word_length', 'sentiment', 'syntax_errors', 'difficult_words'),\n",
       " ('avg_word_length', 'sentiment', 'readability_index', 'difficult_words'),\n",
       " ('avg_word_length', 'noun_phrases', 'syntax_errors', 'readability_index'),\n",
       " ('avg_word_length', 'noun_phrases', 'syntax_errors', 'difficult_words'),\n",
       " ('avg_word_length', 'noun_phrases', 'readability_index', 'difficult_words'),\n",
       " ('avg_word_length', 'syntax_errors', 'readability_index', 'difficult_words'),\n",
       " ('sentiment', 'noun_phrases', 'syntax_errors', 'readability_index'),\n",
       " ('sentiment', 'noun_phrases', 'syntax_errors', 'difficult_words'),\n",
       " ('sentiment', 'noun_phrases', 'readability_index', 'difficult_words'),\n",
       " ('sentiment', 'syntax_errors', 'readability_index', 'difficult_words'),\n",
       " ('noun_phrases', 'syntax_errors', 'readability_index', 'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set', 'essay_length', 'avg_word_length', 'sentiment', 'noun_phrases'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set', 'essay_length', 'sentiment', 'noun_phrases', 'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set', 'essay_length', 'sentiment', 'noun_phrases', 'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words'),\n",
       " ('essay_set',\n",
       "  'essay_length',\n",
       "  'avg_sentence_length',\n",
       "  'avg_word_length',\n",
       "  'sentiment',\n",
       "  'noun_phrases',\n",
       "  'syntax_errors',\n",
       "  'readability_index',\n",
       "  'difficult_words')]"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_feature_combinations(dataset):\n",
    "    attributes = list(dataset)\n",
    "    attributes.remove('domain1_score')\n",
    "\n",
    "    attribute_combinations = []\n",
    "    for size in range(len(attributes)):\n",
    "        attribute_combinations = attribute_combinations + list(itertools.combinations(attributes, size + 1))\n",
    "\n",
    "    return attribute_combinations\n",
    "\n",
    "get_feature_combinations(dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "train_dataset = dataset.sample(frac=0.8,random_state=0)\n",
    "test_dataset = dataset.drop(train_dataset.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    10381.000000\n",
       "mean         6.716405\n",
       "std          8.891451\n",
       "min          0.000000\n",
       "25%          2.000000\n",
       "50%          3.000000\n",
       "75%          8.000000\n",
       "max         60.000000\n",
       "Name: domain1_score, dtype: float64"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_stats = train_dataset.describe().pop(\"domain1_score\").transpose()\n",
    "train_stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "pycharm": {
     "is_executing": false
    }
   },
   "outputs": [],
   "source": [
    "train_labels = train_dataset.pop('domain1_score')\n",
    "test_labels = test_dataset.pop('domain1_score')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "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>essay_set</th>\n",
       "      <th>essay_length</th>\n",
       "      <th>avg_sentence_length</th>\n",
       "      <th>avg_word_length</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>noun_phrases</th>\n",
       "      <th>syntax_errors</th>\n",
       "      <th>readability_index</th>\n",
       "      <th>difficult_words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2976</th>\n",
       "      <td>-0.530443</td>\n",
       "      <td>10.828783</td>\n",
       "      <td>3.106009</td>\n",
       "      <td>-0.225797</td>\n",
       "      <td>-0.757252</td>\n",
       "      <td>0.256830</td>\n",
       "      <td>0.144363</td>\n",
       "      <td>1.302779</td>\n",
       "      <td>1.269039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1324</th>\n",
       "      <td>-0.642910</td>\n",
       "      <td>71.111404</td>\n",
       "      <td>1.931348</td>\n",
       "      <td>-0.281923</td>\n",
       "      <td>-0.729901</td>\n",
       "      <td>2.843585</td>\n",
       "      <td>1.718909</td>\n",
       "      <td>0.740441</td>\n",
       "      <td>2.843585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12143</th>\n",
       "      <td>0.031895</td>\n",
       "      <td>53.116591</td>\n",
       "      <td>0.809245</td>\n",
       "      <td>-0.237416</td>\n",
       "      <td>-0.732906</td>\n",
       "      <td>2.731117</td>\n",
       "      <td>0.256830</td>\n",
       "      <td>0.144363</td>\n",
       "      <td>8.017093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3803</th>\n",
       "      <td>-0.417975</td>\n",
       "      <td>14.315277</td>\n",
       "      <td>1.381506</td>\n",
       "      <td>-0.266899</td>\n",
       "      <td>-0.751680</td>\n",
       "      <td>0.594233</td>\n",
       "      <td>0.144363</td>\n",
       "      <td>0.256830</td>\n",
       "      <td>1.156571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>252</th>\n",
       "      <td>-0.642910</td>\n",
       "      <td>93.942322</td>\n",
       "      <td>0.600481</td>\n",
       "      <td>-0.270512</td>\n",
       "      <td>-0.741380</td>\n",
       "      <td>12.065926</td>\n",
       "      <td>3.068520</td>\n",
       "      <td>0.234337</td>\n",
       "      <td>8.129561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3355</th>\n",
       "      <td>-0.530443</td>\n",
       "      <td>39.508015</td>\n",
       "      <td>1.375587</td>\n",
       "      <td>-0.327184</td>\n",
       "      <td>-0.739280</td>\n",
       "      <td>2.731117</td>\n",
       "      <td>3.293455</td>\n",
       "      <td>0.346804</td>\n",
       "      <td>1.269039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6542</th>\n",
       "      <td>-0.305507</td>\n",
       "      <td>9.029301</td>\n",
       "      <td>0.642434</td>\n",
       "      <td>-0.309386</td>\n",
       "      <td>-0.748951</td>\n",
       "      <td>-0.193040</td>\n",
       "      <td>0.481766</td>\n",
       "      <td>-0.260520</td>\n",
       "      <td>0.256830</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4225</th>\n",
       "      <td>-0.417975</td>\n",
       "      <td>6.330080</td>\n",
       "      <td>0.661714</td>\n",
       "      <td>-0.294796</td>\n",
       "      <td>-0.727261</td>\n",
       "      <td>-0.417975</td>\n",
       "      <td>-0.642910</td>\n",
       "      <td>-0.249274</td>\n",
       "      <td>-0.305507</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6479</th>\n",
       "      <td>-0.305507</td>\n",
       "      <td>47.830615</td>\n",
       "      <td>1.056600</td>\n",
       "      <td>-0.234695</td>\n",
       "      <td>-0.739356</td>\n",
       "      <td>2.056312</td>\n",
       "      <td>0.369298</td>\n",
       "      <td>0.403038</td>\n",
       "      <td>4.755533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9469</th>\n",
       "      <td>-0.080572</td>\n",
       "      <td>39.845417</td>\n",
       "      <td>0.710866</td>\n",
       "      <td>-0.260957</td>\n",
       "      <td>-0.745277</td>\n",
       "      <td>1.831376</td>\n",
       "      <td>0.594233</td>\n",
       "      <td>0.144363</td>\n",
       "      <td>2.956052</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10381 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       essay_set  essay_length  avg_sentence_length  avg_word_length  \\\n",
       "2976   -0.530443     10.828783             3.106009        -0.225797   \n",
       "1324   -0.642910     71.111404             1.931348        -0.281923   \n",
       "12143   0.031895     53.116591             0.809245        -0.237416   \n",
       "3803   -0.417975     14.315277             1.381506        -0.266899   \n",
       "252    -0.642910     93.942322             0.600481        -0.270512   \n",
       "...          ...           ...                  ...              ...   \n",
       "3355   -0.530443     39.508015             1.375587        -0.327184   \n",
       "6542   -0.305507      9.029301             0.642434        -0.309386   \n",
       "4225   -0.417975      6.330080             0.661714        -0.294796   \n",
       "6479   -0.305507     47.830615             1.056600        -0.234695   \n",
       "9469   -0.080572     39.845417             0.710866        -0.260957   \n",
       "\n",
       "       sentiment  noun_phrases  syntax_errors  readability_index  \\\n",
       "2976   -0.757252      0.256830       0.144363           1.302779   \n",
       "1324   -0.729901      2.843585       1.718909           0.740441   \n",
       "12143  -0.732906      2.731117       0.256830           0.144363   \n",
       "3803   -0.751680      0.594233       0.144363           0.256830   \n",
       "252    -0.741380     12.065926       3.068520           0.234337   \n",
       "...          ...           ...            ...                ...   \n",
       "3355   -0.739280      2.731117       3.293455           0.346804   \n",
       "6542   -0.748951     -0.193040       0.481766          -0.260520   \n",
       "4225   -0.727261     -0.417975      -0.642910          -0.249274   \n",
       "6479   -0.739356      2.056312       0.369298           0.403038   \n",
       "9469   -0.745277      1.831376       0.594233           0.144363   \n",
       "\n",
       "       difficult_words  \n",
       "2976          1.269039  \n",
       "1324          2.843585  \n",
       "12143         8.017093  \n",
       "3803          1.156571  \n",
       "252           8.129561  \n",
       "...                ...  \n",
       "3355          1.269039  \n",
       "6542          0.256830  \n",
       "4225         -0.305507  \n",
       "6479          4.755533  \n",
       "9469          2.956052  \n",
       "\n",
       "[10381 rows x 9 columns]"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def norm(x):\n",
    "    return (x - train_stats['mean']) / train_stats['std']\n",
    "\n",
    "normed_train_data = norm(train_dataset)\n",
    "normed_test_data = norm(test_dataset)\n",
    "normed_train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def build_model():\n",
    "    model = keras.Sequential([\n",
    "    layers.Dense(64, activation='relu', input_shape=[len(train_dataset.keys())]),\n",
    "    layers.Dense(64, activation='relu'),\n",
    "    layers.Dense(1)\n",
    "  ])\n",
    "\n",
    "    optimizer = tf.keras.optimizers.RMSprop(0.001)\n",
    "\n",
    "    model.compile(loss='mse',\n",
    "                optimizer=optimizer,\n",
    "                metrics=['mae', 'mse'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "model = build_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_6\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_18 (Dense)             (None, 64)                640       \n",
      "_________________________________________________________________\n",
      "dense_19 (Dense)             (None, 64)                4160      \n",
      "_________________________________________________________________\n",
      "dense_20 (Dense)             (None, 1)                 65        \n",
      "=================================================================\n",
      "Total params: 4,865\n",
      "Trainable params: 4,865\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Falling back from v2 loop because of error: Failed to find data adapter that can handle input: <class 'pandas.core.frame.DataFrame'>, <class 'NoneType'>\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[ 2.5179029],\n",
       "       [15.623663 ],\n",
       "       [12.137756 ],\n",
       "       [ 3.2502847],\n",
       "       [20.479496 ],\n",
       "       [ 2.6699939],\n",
       "       [ 3.4960928],\n",
       "       [ 2.5713978],\n",
       "       [ 6.2497005],\n",
       "       [ 3.9603148]], dtype=float32)"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "example_batch = normed_train_data[:10]\n",
    "example_result = model.predict(example_batch)\n",
    "example_result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Falling back from v2 loop because of error: Failed to find data adapter that can handle input: <class 'pandas.core.frame.DataFrame'>, <class 'NoneType'>\n",
      "\n",
      "Epoch: 0, loss:80.5847,  mae:5.6110,  mse:80.5847,  val_loss:74.0108,  val_mae:5.6114,  val_mse:74.0108,  \n",
      "....................................................................................................\n",
      "Epoch: 100, loss:5.9622,  mae:1.4214,  mse:5.9622,  val_loss:6.4593,  val_mae:1.5959,  val_mse:6.4593,  \n",
      "....................................................................................................\n",
      "Epoch: 200, loss:5.3566,  mae:1.3723,  mse:5.3566,  val_loss:5.4534,  val_mae:1.4110,  val_mse:5.4534,  \n",
      "....................................................................................................\n",
      "Epoch: 300, loss:5.0835,  mae:1.3342,  mse:5.0835,  val_loss:5.7658,  val_mae:1.4457,  val_mse:5.7658,  \n",
      "............"
     ]
    }
   ],
   "source": [
    "model = build_model()\n",
    "\n",
    "EPOCHS = 1000\n",
    "# The patience parameter is the amount of epochs to check for improvement\n",
    "early_stop = keras.callbacks.EarlyStopping(monitor='val_loss', patience=200)\n",
    "\n",
    "early_history = model.fit(normed_train_data, train_labels, \n",
    "                    epochs=EPOCHS, validation_split = 0.2, verbose=0, \n",
    "                    callbacks=[early_stop, tfdocs.modeling.EpochDots()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "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>loss</th>\n",
       "      <th>mae</th>\n",
       "      <th>mse</th>\n",
       "      <th>val_loss</th>\n",
       "      <th>val_mae</th>\n",
       "      <th>val_mse</th>\n",
       "      <th>epoch</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>307</th>\n",
       "      <td>5.039297</td>\n",
       "      <td>1.333754</td>\n",
       "      <td>5.039298</td>\n",
       "      <td>6.490976</td>\n",
       "      <td>1.494451</td>\n",
       "      <td>6.490976</td>\n",
       "      <td>307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>308</th>\n",
       "      <td>5.077600</td>\n",
       "      <td>1.341877</td>\n",
       "      <td>5.077598</td>\n",
       "      <td>5.928021</td>\n",
       "      <td>1.441738</td>\n",
       "      <td>5.928021</td>\n",
       "      <td>308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>309</th>\n",
       "      <td>5.063305</td>\n",
       "      <td>1.338954</td>\n",
       "      <td>5.063303</td>\n",
       "      <td>5.888891</td>\n",
       "      <td>1.458616</td>\n",
       "      <td>5.888891</td>\n",
       "      <td>309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>310</th>\n",
       "      <td>5.043942</td>\n",
       "      <td>1.328696</td>\n",
       "      <td>5.043941</td>\n",
       "      <td>6.003290</td>\n",
       "      <td>1.461893</td>\n",
       "      <td>6.003289</td>\n",
       "      <td>310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>311</th>\n",
       "      <td>5.026554</td>\n",
       "      <td>1.331606</td>\n",
       "      <td>5.026555</td>\n",
       "      <td>5.819243</td>\n",
       "      <td>1.457323</td>\n",
       "      <td>5.819243</td>\n",
       "      <td>311</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         loss       mae       mse  val_loss   val_mae   val_mse  epoch\n",
       "307  5.039297  1.333754  5.039298  6.490976  1.494451  6.490976    307\n",
       "308  5.077600  1.341877  5.077598  5.928021  1.441738  5.928021    308\n",
       "309  5.063305  1.338954  5.063303  5.888891  1.458616  5.888891    309\n",
       "310  5.043942  1.328696  5.043941  6.003290  1.461893  6.003289    310\n",
       "311  5.026554  1.331606  5.026555  5.819243  1.457323  5.819243    311"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hist = pd.DataFrame(early_history.history)\n",
    "hist['epoch'] = early_history.epoch\n",
    "hist.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "plotter = tfdocs.plots.HistoryPlotter(smoothing_std=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'MAE [domain1_score]')"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6cAAAJNCAYAAADNtAMgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAgAElEQVR4nOzdeZiU5Z3v/8/d1VXdVdXdVb2xCcgiNhJEECSKMTY6iuNo3DhmPATBK1ExJk6YzPHEOZOM6G/mOBl/ZtEYlyga425EE42JK4lrEASVoIaArexL7/t6nz9qoVm7uruq63mq3q/r6st6qrqqvo3PP5/re9/f21hrBQAAAABAOuWkuwAAAAAAAAinAAAAAIC0I5wCAAAAANKOcAoAAAAASDvCKQAAAAAg7QinAAAAAIC0y013Ab2Fw2F7zDHHpLsMuFRzc7OCwWC6y4BLcf9goLh3MBjcPxgo7h0MRjrvnzVr1uy11pYf6jVHhdPhw4dr9erV6S4DLrVy5UpVVlamuwy4FPcPBop7B4PB/YOB4t7BYKTz/jHGfHa411jWCwAAAABIO8IpAAAAACDtCKcAAAAAgLRz1J5TAAAAAO7Q2dmprVu3qq2tLd2loJ9CoZA++uijlH5Hfn6+Ro8eLa/Xm/B7CKcAAAAA+m3r1q0qLCzUuHHjZIxJdznoh8bGRhUWFqbs8621qq6u1tatWzV+/PiE38eyXgAAAAD91tbWptLSUoIpDmKMUWlpab+76oRTAAAAAANCMMXhDOTeIJwCAAAAcCWPx6Pp06fHf2655ZZ+vf/GG2/UrbfemvDvv/POO/riF7+o6dOn67jjjtONN94oKXJu6FtvvdWv707UnDlzBv0Zf/jDH+L/RgUFBTrxxBM1ffp0XX755Ql/Rnd3t0477bRB13Ik7DkFAAAA4Ep+v1/r1q0b0Hu7urr6/Z5FixbpiSee0AknnKDu7m598sknkiLhtKCgIClB8kDJCL3z5s3TvHnzJEmVlZVatmyZTj/99IN+r6urS7m5h46IHo9Hr7/++qBrORI6pwAAAAAyyk033aSTTjpJU6dO1VVXXSVrraRIMPvOd76jWbNm6Sc/+Un89zdt2qQTTzwxfr1x48b9rmN2796tkSNHSoqEtSlTpqiqqkp33XWXfvSjH2n69Ol6/fXXVVVVpTPOOEPTpk3TmWeeqc8//1yStHjxYi1ZskSzZs3Sscceq+eee06S9MADD+iCCy5QZWWlJk2apGXLlsW/s6CgQFIkAFdWVmr+/PmaPHmyFixYEP+7fve732ny5MmaOXOmrrvuOp133nkJ/1v94he/0IUXXqi5c+dq3rx5amho0BlnnKETTzxR06ZNi9fY1dWlcDgsSXr55Zd15pln6uKLL1ZFRUW/OrBHQjgFAAAA4Eqtra37Let9/PHHJUnf+ta39O6772r9+vVqbW2NByxJ6ujo0OrVq/Xd7343/tzEiRMVCoXiXdjly5friiuuOOj7li5dqoqKCl100UW6++671dbWpnHjxmnJkiVaunSp1q1bp9NOO03f/va3tWjRIn3wwQdasGCBrrvuuvhnVFVVadWqVXr++ee1ZMmS+NCgVatW6de//rU++OADPfnkk1q9evVB37927Vr9+Mc/1oYNG7R582a9+eabamtr09VXX60XXnhBa9as0Z49e/r977h27Vo9/fTTeuWVV+T3+/XMM8/ovffe08svv6ylS5ce8j3vvfee7rjjDm3YsEEfffSR3nnnnX5/74FY1gsAAABgUJb99i/asL0hqZ85ZVSR/v38Lxzxdw63rPe1117TD3/4Q7W0tKimpkZf+MIXdP7550uSvvrVrx7ys77xjW9o+fLluu222/T4449r1apVB/3OD37wAy1YsEAvvviiHnnkET366KNauXLlQb/39ttv6+mnn5YkLVy4UNdff338tUsvvVQ5OTmaNGmSJkyYoI8//liSdNZZZ6m0tFSSdPHFF+uNN97QrFmz9vvc2bNna/To0ZKk6dOnq6qqSgUFBZowYUL8yJbLLrtM99xzzxH/3Q509tlnq7i4WFLkGJjvfe97euONN5STk6MtW7Zo79698a5pzMknn6xRo0btV8vJJ5/cr+89EOEUAAAAQMZoa2vTN7/5Ta1evVpjxozRjTfeuN+RJsFg8JDvu+SSS7Rs2TKdccYZmjlzZjwoHmjixIm65pprdOWVV6q8vFzV1dX9qu/AKbax68M931teXl78scfjGdC+2UPp/W/yy1/+UvX19XrvvfeUm5ur0aNHH/JImFTUQjgFAAAAMCh9dTiHUixIlZWVqampSU899ZTmz5/f5/vy8/M1b948XXPNNbrvvvsO+TvPP/+8zj33XBljtHHjRnk8HoXDYRUWFqqhYV/neM6cOXrssce0cOFCPfzww/tNuX3yySe1aNEiffrpp9q8ebMqKiq0du1avfTSS6qpqYkvq73//vsT+nsrKiq0efNmVVVVady4cfGlzQNVX1+vYcOGKTc3Vy+99JK2bds2qM/rD8IpAAAAAFeK7TmNOeecc3TLLbfoyiuv1NSpUzVixAiddNJJCX/eggULtGLFCp199tmHfP2hhx7S0qVLFQgElJubq4cfflgej0fnn3++5s+fr2effVa33367br/9dl1xxRX67//+b5WXl2v58uXxzxg7dqxmz56thoYG3XXXXcrPz5cUWbJ7ySWXaOvWrfra17520JLew/H7/brzzjt1zjnnKBgM9uvvPZSFCxfq/PPP1/HHH6/Zs2dr0qRJg/q8/jCxCU9OUFFRYWPjmIH+ik0wAwaC+wcDxb2DweD+wUA54d756KOPdNxxx6W1hmS79dZbVV9fr5tvvjkln7948WKdd955B3VyH3jgAa1evVp33HHHgD63qalJBQUFstbq2muv1aRJkw47yEiSGhsbVVhYOKDv6o9D3SPGmDXW2kMmbzqnAAAAALLeRRddpE2bNunVV19Ndyn9du+99+rBBx9UR0eHZsyYoauvvjrdJQ0I4RQAAABA1luxYkXKv+OBBx445POLFy/W4sWLB/y5S5cuPWKn1C045xQAAAAAkHaEUwAAAABA2hFOAQAAAABpRzgFAAAAAKSdo8Lp1sYebd7TlO4yAAAAALiAx+PR9OnT4z+33HJLv95/44036tZbb03499955x198Ytf1PTp03XcccfpxhtvlBQ52uett97q13cnas6cOYP+jJaWFpWWlqqhoWG/5y+88EI9/vjjh33fypUrdd555w36+xPlqGm9XVZ6e3O1JpQXpLsUAAAAAA7n9/u1bt26Ab23q6ur3+9ZtGiRnnjiCZ1wwgnq7u7WJ598IikS4goKCpISJA+UjNAbCAQ0b948rVixQosWLZIk1dfX64033tAjjzwy6M9PFkd1Tj1GentTdbrLAAAAAOBiN910k0466SRNnTpVV111lay1kqTKykp95zvf0axZs/STn/wk/vubNm3SiSeeGL/euHHjftcxu3fv1siRIyVFurZTpkxRVVWV7rrrLv3oRz/S9OnT9frrr6uqqkpnnHGGpk2bpjPPPFOff/65pMiRMUuWLNGsWbN07LHH6rnnnpMUOWLmggsuUGVlpSZNmqRly5bFv7OgINK4W7lypSorKzV//nxNnjxZCxYsiP9dv/vd7zR58mTNnDlT11133SG7nZdddpkee+yx+PWKFSs0b948BQIBrVq1SqeccopmzJihOXPmxEP3UHNUOM3PNXpnc038HxkAAAAADqe1tXW/Zb2xJarf+ta39O6772r9+vVqbW2Nh0BJ6ujo0OrVq/Xd7343/tzEiRMVCoXiXdjly5friiuuOOj7li5dqoqKCl100UW6++671dbWpnHjxmnJkiVaunSp1q1bp9NOO03f/va3tWjRIn3wwQdasGCBrrvuuvhnVFVVadWqVXr++ee1ZMkStbW1SZJWrVqlX//61/rggw/05JNPavXq1Qd9/9q1a/XjH/9YGzZs0ObNm/Xmm2+qra1NV199tV544QWtWbNGe/bsOeS/1bx58/Tee++pujrSDHzsscd02WWXSZImT56s119/XWvXrtVNN92kf/3Xf+3X/4dkcdSy3nyPtLepXZv2NOmYYYXpLgcAAABAgr5699sHPXfetJFaeMo4tXZ0a/HyVQe9Pn/maP2PWWNU09yha361Zr/XHr/6lD6/83DLel977TX98Ic/VEtLi2pqavSFL3xB559/fqTOr371kJ/1jW98Q8uXL9dtt92mxx9/XKtWHVzvD37wAy1YsEAvvviiHnnkET366KNauXLlQb/39ttv6+mnn5YkLVy4UNdff338tUsvvVQ5OTmaNGmSJkyYoI8//liSdNZZZ6m0tFSSdPHFF+uNN97QrFmz9vvc2bNna/To0ZKk6dOnq6qqSgUFBZowYYLGjx8vKdIhveeeew6qyefz6Stf+YqeeuopzZs3T2vXrtW8efMkRZb4Llq0SBs3bpQxRp2dnYf8N0o1R3VO/blGEkt7AQAAAAxMW1ubvvnNb+qpp57Shx9+qCuvvDLenZSkYDB4yPddcskleuGFF/Tcc89p5syZ8aB4oIkTJ+qaa67RK6+8ovfffz/eiUyUMeaQ14d7vre8vLz4Y4/H0+99s7Glvc8884wuuOACeb1eSdL3v/99zZ07V+vXr9dvf/vb/f69hpKjOqe5OdLwUL7e3lythaeMS3c5AAAAABJ0pE6n3+c54uslQV9CndJExIJVWVmZmpqa9NRTT2n+/Pl9vi8/P1/z5s3TNddco/vuu++Qv/P888/r3HPPlTFGGzdulMfjUTgcVmFh4X6TcOfMmaPHHntMCxcu1MMPP6zTTjst/tqTTz6pRYsW6dNPP9XmzZtVUVGhtWvX6qWXXlJNTY38fr+eeeYZ3X///Qn9vRUVFdq8ebOqqqo0bty4I07frays1OWXX657771Xd9xxR/z5+vp6HXXUUZIi+1/TxVGdU0k6eUIp+04BAAAA9OnAPaff+973FA6HdeWVV2rq1KmaN2+eTjrppIQ/b8GCBcrJydHZZ599yNcfeughVVRUaPr06fHg6fF4dP7552vFihXxgUi33367li9frmnTpumhhx7ab/jS2LFjNXv2bP393/+97rrrLuXn50uKLNm95JJLNG3aNF1yySUHLek9HL/frzvvvFPnnHOOZs6cqcLCQoVCoUP+bk5OjubPn6+amhqdfvrp8eevv/563XDDDZoxY8aAphgni3FSCKyoqLA3P/yyrn/qA/3hO19WxQj2nSJxsQlmwEBw/2CguHcwGNw/GCgn3DsfffSRjjvuuLTWkGy33nqr6uvrdfPNN6fk8xcvXqzzzjvvoE7uAw88oNWrV+/XzeyPpqYmFRQUyFqra6+9VpMmTdLSpUsP+/uNjY0qLEx91jrUPWKMWWOtPWTydtSyXkk6ZUJkbffbm/YSTgEAAAAMiYsuukibNm3Sq6++mu5S+u3ee+/Vgw8+qI6ODs2YMUNXX311uksaEEeF05o2q4a2Th0V9uudzTVafOr4dJcEAAAAIAusWLEi5d9xuP2cixcv1uLFiwf8uUuXLj1ip9QtHLXntKHDau3ndTplYqne+bRaPT3OWXIMAAAAAEgdR4VTI2nznmbNHl+iupZObd7blO6SAAAAAByGk+bXwFkGcm84Kpx6c6RNe5o0dVRkutSGHY1prggAAADAoeTn56u6upqAioNYa1VdXR2fRJwoR+059eYYbdrTpGOGFcjrMdqwvUFfOWFUussCAAAAcIDRo0dr69at2rNnT7pLQT+1tbX1Ozj2V35+vkaPHt2v9zgrnHqknh4rY6RJwwq1YUdD328CAAAAMOS8Xq/Gj2eAqRutXLlSM2bMSHcZB3HUst5wntFbN5wprydHU0YVacN2wikAAAAAZANHhdPepows0t6mdu1ubEt3KQAAAACAFHPUsl5JWvLQGp12bJmmjCqSJH20o1HDClO7HhoAAAAAkF6O65x+uK1eqz6t0XEjI+GUpb0AAAAAkPkcF04nDivQpj1NCvm9Gl3sZygSAAAAAGQBx4XTCWVBbdrdrJ4eqykji7Rhe326SwIAAAAApJjjwunEYQVq7ezWzoY2TRlVpM17m9XS0ZXusgAAAAAAKeS4cHrciEJNHxNWU3uXpowskrXSJzsb010WAAAAACCFHDetd9a4Ej1z7amSpIDPI0nasKNBM8YWp7MsAAAAAEAKOa5z2ttRYb+K8nOZ2AsAAAAAGc6R4fRfnnxfVz+0WsYYTRlVxMReAAAAAMhwjgynPT1WH2yNTOk9bmSRPt7RqJ4em+aqAAAAAACp4shwOnFYgXbUt6m5vUsTyyPTe3c3tqe7LAAAAABAijgynB5dGpAkbalt0bjSoCTp073N6SwJAAAAAJBCjgyno4uj4bSmVePKIo+rqgmnAAAAAJCpHBlOjy4J6NzjRygc8GpkyC+fJ4dwCgAAAAAZzHHnnEpScdCnOxfMjF+PLQ2oimW9AAAAAJCxHNk5jeno6pEkjSsNqmpvS5qrAQAAAACkimPD6bcfXasLf/amJGlcaUCf1TRznAwAAAAAZCjHhtPigFdbayPd0nFlQbV19mhXY1uaqwIAAAAApIJjw+noYr8a2rpU39qp8WUcJwMAAAAAmcyx4XRM9DiZrbUt8XNPP6tm3ykAAAAAZCLHhtPR8XDaqlEhv3y5OUzsBQAAAIAM5dhwOrY0oK9/abyOCvuVk2N0dEmAZb0AAAAAkKEcec6pJIX8Xn3/vCnx66NLg6qqJpwCAAAAQCZybOdUkto6u7WzPjKhd3xZQJ9Vt3CcDAAAAABkIEeH0289slaL7l8lKXKcTHtXj3Y2cJwMAAAAAGQaR4fTMSV+ba1tkbVW40ojx8kwFAkAAAAAMo+jw+no4oCaO7pV19KpcdGzTqs4TgYAAAAAMo7Dw6lfkrSltkUji/Ijx8kwFAkAAAAAMo6jw+mYXmedcpwMAAAAAGQuR4fTo0sD+rd/OE6TRxRKigxF+ozOKQAAAABkHMeecypJwbxcfeO0CfHro8J+vb2pOo0VAQAAAABSwdGdU0naWtui9dvqJUkjQ/lqau9SQ1tnmqsCAAAAACST48Ppzc9t0HceXydJGhmODEjaWc9ZpwAAAACQSRwfTkcXB+JnnY4K5UuStte1prkqAAAAAEAyOT6cjgzlq62zR/WtnfHO6Q46pwAAAACQURwfTkdEu6U7G9o0vDBPOUbaQecUAAAAADJKyqf1GmOqJDVK6pbUZa2d1Z/3j4yG0x31bZo8okjDCvO1nc4pAAAAAGSUoTpKZq61du9A3jhpeKHu+tqJmjoqJEkaGc7Xjno6pwAAAACQSRx9zqkkFeV7dc7UkfHrkaF8fbyjMY0VAQAAAACSbSj2nFpJLxpj1hhjrhrIB7xbVaM1n9VIkkaG/Npe3yprbTJrBAAAAACkkUl1yDPGHGWt3WaMGSbpJUnfttb+qdfrV0m6SpLKy8tnPvHEEwd9xo1vtarQZ/TdWfn6Q1WnHv24Q3ecEVCBz6S0drhLU1OTCgoK0l0GXIr7BwPFvYPB4P7BQHHvYDDSef/MnTt3zeHmEKV8Wa+1dlv0v7uNMSskzZb0p16v3yPpHkmqqKiwlZWVB33GpM9X6/PqFlVWflktH+7Qox+/pwlTZ2rKqKJUlw8XWblypQ51/wCJ4P7BQHHvYDC4fzBQ3DsYDKfePyld1muMCRpjCmOPJZ0taX1/P2dEUb52NkQm9O6b3stQJAAAAADIFKnunA6XtMIYE/uuR6y1v+/vh4wI5au+tVOtHd0aFfZLEsfJAAAAAEAGSWk4tdZulnTCYD9nRFGkW7qzoU1jSwLKzTHaUUfnFAAAAAAyxVBM6x200yvK9cy1p2pkKF+eHKPhRfnaQecUAAAAADKG4885laSygjyVFeTFr0eF87WdzikAAAAAZAxXdE6ttXpy9Zb4WacjQn46pwAAAACQQVwRTo0xuvm5DXp23XZJ0qhQvnbWt6mnJ7VntAIAAAAAhoYrwqkUmdi7s37fcTId3T2qbu5Ic1UAAAAAgGRwUTj17zvrNHqczE6W9gIAAABARnBPOC3Ki4fRUaHYWacMRQIAAACATOCecBrya09Tuzq7ezQyHDn3lLNOAQAAACAzuOIoGUlaPGecvvbFscrNMSoN+uTLzWFiLwAAAABkCNeE05Kgb7/rkaF8bSecAgAAAEBGcM2y3vrWTv30lY36cGu9JGl4Ub52EU4BAAAAICO4Jpxaa3XbS3/VqqoaSVJ5YZ72NrWnuSoAAAAAQDK4JpyG/F7l5eZoZ3RCb3lBnvYQTgEAAAAgI7gmnBpjNDKUHx+CVF6Yp8a2LrV1dqe5MgAAAADAYLkmnErSiFC+djVEw2lBniRpTyPdUwAAAABwO1eF0+FF+dodDaPlhdFwytJeAAAAAHA91xwlI0n/9+LjlZfrkSSVRTune+mcAgAAAIDruSqcBnz7yqVzCgAAAACZw1XLetdvq9cNT3+o3Q1tKi3wSWLPKQAAAABkAleF0z2N7Xp01efaUtsqrydHxQEvZ50CAAAAQAZwVTiN7zNt2jcUic4pAAAAALifu8JpYWQpL+EUAAAAADKLq8JpaTA2obdDUqSTurepI50lAQAAAACSwFXh1Jebo/LCPLV3dUuSygsinVNrbZorAwAAAAAMhquOkpGkVf96powxkiLLels7u9Xc0a2CPNf9KQAAAACAKFd1TiXFg6nUa0AS+04BAAAAwNVcF05/9c5nuuHpDyRFOqeStIfjZAAAAADA1VwXTv+2u0nPfbBDUq9wSucUAAAAAFzNdeG0rMCnxrYutXV2H3TuKQAAAADAnVwYTiOBtLq5QyVBn3IMnVMAAAAAcDvXhtO9je3y5BiVRo+TAQAAAAC4l+vC6YhQvkYX+9XR3SMpElZZ1gsAAAAA7ua6w0GnHhXSG//7jPh1eSGdUwAAAABwO9d1Tg9UVuAjnAIAAACAy7kynF7zqzW6+4+bJEU6p3ubOmStTXNVAAAAAICBct2yXkn6aEeDPDlGklRekKeO7h41tHYpFPCmuTIAAAAAwEC4snPaewhSeWFkeu+eprZ0lgQAAAAAGATXhtPqpg5Jkc6pJO1p7EhnSQAAAACAQXBnOC30HaJzylAkAAAAAHArV4bTiuGFmjS8UD09VmXxzinhFAAAAADcypUDkRaeMk4LTxknSQr5vfJ6DOEUAAAAAFzMlZ3T3nJyjEqDeapmWS8AAAAAuJYrw+lfdzXqrNv+qLc27ZUkFQd9qm1hIBIAAAAAuJUrw2lebo427m7S9rrI8TGlQZ+qmwmnAAAAAOBWrgynsSFIsYm9xUGfagmnAAAAAOBargynwbxc+b0e7Y0OQaJzCgAAAADu5spwKu1/1mlxwKfGti51dvekuSoAAAAAwEC4NpyeUTFMk4YXSpJKgl5JYigSAAAAALiUK885laRlF0yNPy4JRvag1jR3aFhhfrpKAgAAAAAMkGs7p70VRzunNew7BQAAAABXcm04vXPl3zT7P16WJJX26pwCAAAAANzHteE0xxjtbmxXa0d3vHPKcTIAAAAA4E6uDaclAZ8kqbq5XcXxx4RTAAAAAHAj14bT4mAkkNY2d8rryVFRfi6dUwAAAABwKdeG09jxMTXR42NKgj7VtHSmsyQAAAAAwAC5NpyOCvt18Yyj4st7S4I+1TS3p7kqAAAAAMBAuPac05Ehv2776vT4dUnQp211bWmsCAAAAAAwUK7tnMZ091hJdE4BAAAAwM1cHU5PveVVff/Z9ZIiA5JqmztlrU1zVQAAAACA/nJ1OA34PPEJvaVBnzq6e9TU3pXmqgAAAAAA/eXqcFoc9KkmGk5jZ53WNjOxFwAAAADcxtXhtCTgU230KJnSgkg4jR0tAwAAAABwD1eH00N1ThmKBAAAAADu49qjZCTp9GPLVRrcd86pJNWwrBcAAAAAXMfV4fScqSN0ztQRknqHUzqnAAAAAOA2rl7Wa61VfUunOrp6VJCXK6/H0DkFAAAAABdydTj941/36ISbXtSH2+pljFFJ0EfnFAAAAABcyNXhNLaUt7bXUCQ6pwAAAADgPhkRTmMTe0sL9h0tAwAAAABwj8wIpy29O6eEUwAAAABwG1eHU7/Xo7zcnPiy3pIg4RQAAAAA3MjVR8kYY/Tds4/VlJEhSZFwWt/aqc7uHnk9rs7dAAAAAJBVXB1OJemqL0+MP44t861r6VR5YV66SgIAAAAA9JPr24t1LR36vLpF0sEDkgAAAAAA7uD6cPqDZ/+iy+//sySpJEA4BQAAAAA3cn047T0EqaQgeu4px8kAAAAAgKu4PpwWB3xqaOtSZ3dPvHNaTecUAAAAAFzF9eG0JOiVFBmCFI6G01rCKQAAAAC4iuvDaXGvIUi+3BwFfR7VtXSmuSoAAAAAQH+4PpyeMDqs/7zoeJVF95uGAz7VtdI5BQAAAAA3cf05p2NKAvqfXxwbvw4HvKqncwoAAAAAruL6zml3j9X6bfXaUd8qKRJO61oJpwAAAADgJhkRTs+7/Q09tXqrJCns93GUDAAAAAC4jOvDqS83R4V5ufHjY1jWCwAAAADu4/pwKknhoFf10aW8sWW91to0VwUAAAAASFRmhFO/T3XRpbxhv0/dPVaN7V1prgoAAAAAkKjMCKe9hiCFA15JYmkvAAAAALiI64+SkaRr5x6jnugy3nAgct5pXUunxpSksyoAAAAAQKIyIpyePKE0/jjWOWViLwAAAAC4R0Ys691W16pXPtqlnh6r4mg45axTAAAAAHCPjAinv1+/U19/cLUa27oU8keW9dbTOQUAAAAA18iIcBr2x7qlHQr5Y8t66ZwCAAAAgFtkRjiNLeVt6ZQvN0cFebmqI5wCAAAAgGtkVjiN7jMN+b2qa2VZLwAAAAC4RYaE09jxMR3Ray+dUwAAAABwkYw4SuaosF8PfX22Jo8okhQLp3ROAQAAAMAtMiKc5ns9Om1Sefw6HPBpR31DGisCAAAAAPRHRizrlaSXN+zSms9qJEWm99azrBcAAAAAXCNjwumy5/6iX73zuaTost7WTllr01wVAAAAACARGRNOw35ffJ9pccCn7gIpRyIAACAASURBVB6rxvauNFcFAAAAAEhE5oTTgFe1LfuOkpHE0l4AAAAAcIkMCqc+1UfPOY0dLVPLxF4AAAAAcIXMCad+b69lvZHOKWedAgAAAIA7ZMRRMpJ09ekTdPkpR0uKLPGVpLpWwikAAAAAuEHGhNPRxYH445A/sqy3jmW9AAAAAOAKGbOs9/PqFi1/81PVNnfs65yyrBcAAAAAXCFjwunG3Y1a9tsN+rymRV5PjgrycgmnAAAAAOASGRNOY93S2ITeUK8BSQAAAAAAZ8ugcBrZZxo7TqY46GUgEgAAAAC4ROaEU//++0zDfh+dUwAAAABwiSEJp8YYjzFmrTHmuVR9R+iAcBoKeNlzCgAAAAAuMVRHyfyTpI8kFaXqC3I9OfrT/5qr0oLI8t7iAMt6AQAAAMAtUt45NcaMlvQPkn6R6u8aWxpQMC+St2PLent6bKq/FgAAAAAwSEOxrPfHkq6X1JPqL3r6va16cvUWSZHpvT1WauroSvXXAgAAAAAGKaXLeo0x50naba1dY4ypPMzvXCXpKkkqLy/XypUrB/x9973bqrYuqbxpk3ZsjSzp/f2rr2tYIGPmPuEImpqaBnX/ILtx/2CguHcwGNw/GCjuHQyGU++fVO85PVXSV4wx50rKl1RkjPmVtfZrsV+w1t4j6R5JqqiosJWVlQP+sqe2v6cN2xtUWVmprg27dN/61Zo87URNGx0e3F8BV1i5cqUGc/8gu3H/YKC4dzAY3D8YKO4dDIZT75+UthSttTdYa0dba8dJ+kdJr/YOpskW7jUEKRzYf3ovAAAAAMC5Mmq9a+8hSLFwWstZpwAAAADgeEN1lIystSslrUzld8SGIDW2dykciBwpU89xMgAAAADgeEMWTofC104+Wl87+Wjlez3q7I4MB2ZZLwAAAAA4X0aF03yvJ/7Y68lRQV4uy3oBAAAAwAUyas/pjvpW3fTbDfp4Z4OkyDLfejqnAAAAAOB4GRVOm9u7df+bn+qTnY2S9p/eCwAAAABwrowKp7EJvbEhSGG/j2W9AAAAAOACGRVOQ/79zzZlWS8AAAAAuENGhdMDhyCxrBcAAAAA3CGjwqkkFQe9auvslhRZ1lvX0qGeHpvmqgAAAAAAR5JRR8lI0h//Za5ycoykSOe0x0qN7V3xJb8AAAAAAOfJuM5pLJhKUjjgkyT2nQIAAACAw2VcOH383c+17Ld/kSSFYwOSWpnYCwAAAABOlnHh9MNt9Xp23XZJ+46WqaVzCgAAAACOlnHhtPcQpNiy3jrOOgUAAAAARzviQCRjzG8S+Iwaa+3i5JQzeL2HIMU6p/UcJwMAAAAAjtbXtN7jJH3jCK8bST9LXjmD13sI0shwviSptplwCgAAAABO1lc4/T/W2j8e6ReMMcuSWM+glQZ9KivIU0tnl7yeHBXm5TIQCQAAAAAc7ojh1Fr7RO9rY0zAWttypN9Jt7mTh2n1v/1d/DoU8HKUDAAAAAA4XEIDkYwxc4wxGyR9HL0+wRhzZ0orS5JwwKtaBiIBAAAAgKMlOq33R5LmSaqWJGvt+5K+nKqiBqOpvUtX/nK1fr9+pySpOOBTHQORAAAAAMDREj5Kxlq75YCnupNcS1L4PDl6acMu/XVXoyQp5GdZLwAAAAA4XV8DkWK2GGPmSLLGGK+kf5L0UerKGjhfbo4K8nLjS3lZ1gsAAAAAzpdo53SJpGslHSVpm6Tp0WtH6t0tLQ74VN/aqZ4em+aqAAAAAACH02fn1BjjkbTQWrtgCOpJiuKgN77PNOT3qsdKje1dCvm9aa4MAAAAAHAofXZOrbXdkv7nENSSNJOGFao44JMkhaP/Zd8pAAAAADhXontO3zDG3CHpcUnNsSette+lpKpB+tFXp8cfFwci3dLalg6NLQ2kqyQAAAAAwBEkGk5jae+mXs9ZSWckt5zkC0fDKcfJAAAAAIBzJRROrbVzU11IMj25eose/vPnWvHNOQr5I8t665jYCwAAAACOldC0XmNMyBhzmzFmdfTn/zfGhFJd3EDVt3Zq3ZY6NbZ37eucsucUAAAAABwr0aNk7pfUKOnS6E+DpOWpKmqwYlN561s6FfYTTgEAAADA6RLdczrRWntJr+tlxph1qSgoGWITeutaOjWmJKDCvFzVtbKsFwAAAACcKtHOaasx5kuxC2PMqZJaU1PS4O0bghQJpKGAl84pAAAAADhYop3TayQ92Gufaa2kxSmpKAmGFebpxLFh+TyR7F0c8DEQCQAAAAAcLNFpvesknWCMKYpeN6S0qkE6ujSop795avw6HPBylAwAAAAAOFii03r/0xgTttY2WGsbjDHFxpj/L9XFJUvIz7JeAAAAAHCyRPec/r21ti52Ya2tlXRuakpKjovvfFM/eXmjJJb1AgAAAIDTJRpOPcaYvNiFMcYvKe8Iv592uxra9VlNs6TIst761k719Ng0VwUAAAAAOJREByI9LOkVY0zsbNMrJD2YmpKSIxzwqj66lDfk96rHSo1tXQpFJ/kCAAAAAJwj0YFI/2WMeV/S30Wfutla+4fUlTV4vYcgFcfOPW3tIJwCAAAAgAMlOhApKOlFa+2/SLpXUp4xxtEpLxzwqTa6zzR27mktQ5EAAAAAwJES3XP6J0n5xpijJP1e0kJJD6SqqGQ4cWyxZh1dLGlfOK3nOBkAAAAAcKRE95waa22LMebrkn5urf2hMWZdKgsbrK9/aXz8ccgfXdbLxF4AAAAAcKREO6fGGHOKpAWSno8+50lNSclH5xQAAAAAnC3RcPpPkm6QtMJa+xdjzARJr6WurMF74cMdmnHTi9pS06KQPxpO2XMKAAAAAI6UUDi11v7JWvsVa+1/Ra83W2uvi71ujLk9VQUOVK4nR7Utnapr6ZTXk6OgzxOf3gsAAAAAcJZEO6d9OTVJn5M0saW8da2xib0+1dE5BQAAAABHSlY4dZywf//jY0J+r+pbGYgEAAAAAE6UueE0EJnQW9/rrFM6pwAAAADgTMkKpyZJn5M0Ib9XF04fpTElAUnRcMqeUwAAAABwpETPOe3LT5L0OUnjy83Rj/9xRvw65GfPKQAAAAA41YA7p8aYe2KPrbUPJKWaFOjpsZIindP61g5Za9NcEQAAAADgQEfsnBpjSg73kqRzk19Ocs3/+Vsq8nt1/+KTFPJ71dlt1drZrYAvWQ1jAAAAAEAy9JXS9kj6TPvvKbXR62GpKipZ8r0e1cUGIkWn99a1dBJOAQAAAMBh+kppmyWdaa39/MAXjDFbUlNS8oQCXm2vb5XU69zTlk6NCvvTWRYAAAAA4AB97Tn9saTiw7z2wyTXknRh/77jY0L+yNEydZx1CgAAAACOc8TOqbX2Z0d47fbkl5NcxQGf6lo61NNj453Teib2AgAAAIDjJLz50hgzR9K43u+x1v4yBTUlzezxJeronqCuXuGUs04BAAAAwHkSCqfGmIckTZS0TlJ39GkrydHh9MvHluvLx5ZLksKxZb10TgEAAADAcRLtnM6SNMW67JBQa62aO7rl8+Qo35sjX24Oe04BAAAAwIH6GogUs17SiFQWkgprPqvV1H//g97eXC1jjEJ+rxpY1gsAAAAAjpNo57RM0gZjzCpJ7bEnrbVfSUlVSRIOxJby7jvrlGW9AAAAAOA8iYbTG1NZRKr0Pts0dk04BQAAAADnSSicWmv/mOpCUiHs3z+chvw+batrTWdJAAAAAIBDOOKeU2PMG9H/NhpjGnr9NBpjGoamxIHL9eSoMC9XtbFlvQGv6lsYiAQAAAAATnPEzqm19kvR/xYOTTnJd92ZkzRpeIGk6J5TBiIBAAAAgOMkuudUkmSMGSYpP3Ztrf086RUl2ZVfnhB/HA541dLRrfaubuXletJYFQAAAACgt4SOkjHGfMUYs1HSp5L+KKlK0gsprCtp6ls6tbW2RZIUik7vrad7CgAAAACOkug5pzdLOlnSX6214yWdKemdlFWVRP/27HotvG+VJCkUHZBUz8ReAAAAAHCURMNpp7W2WlKOMSbHWvuapFkprCtpwn7vvoFIsXBK5xQAAAAAHCXRPad1xpgCSX+S9LAxZrek5tSVlTzFAa/qWzvV02MPOvcUAAAAAOAMiXZOL5DUKmmppN9L2iTp/FQVlUyhgE/WSo1tXQr7I3tOmdgLAAAAAM6SUOfUWtssScaYIkm/TWlFSRZbylvb0qHiYDScctYpAAAAADhKQuHUGHO1pGWS2iT1SDKSrKQJR3qfE5x4dLFuvnCqwgGvCvNylWPYcwoAAAAATpPontN/kTTVWrs3lcWkwviyoMaXBePXIb+XPacAAAAA4DCJ7jndJKkllYWkSmd3jzZsb9CexnZJUjjgY88pAAAAADhMouH0BklvGWPuNsb8NPaTysKSpbGtS+f+9HU9/8F2SVKR38ueUwAAAABwmESX9d4t6VVJHyqy59Q1ivIjf2JtdClvmHAKAAAAAI6TaDj1Wmv/OaWVpEiuJ0dF+bnxIUjhgFdV1a44ohUAAAAAskaiy3pfMMZcZYwZaYwpif2ktLIkCgd8qo12S8MMRAIAAAAAx0m0c3pZ9L839HrOFUfJSFJxYF8gDQV8amjrVHePlSfHpLkyAAAAAICUYDi11o5PdSGp9M9nVygvN9IkDvu9slZqaO1UcdCX5soAAAAAAFKC4dQY45V0jaQvR59aKelua60r1seefmx5/HFx0CtJqm3pIJwCAAAAgEMkuuf055JmSroz+jMz+pwrbKlp0esb90iSigORQFrLvlMAAAAAcIxE95yeZK09odf1q8aY91NRUCo8uWarbn91o/72H+fuC6fNHCcDAAAAAE6RaOe02xgzMXZhjJkgqTs1JSVfcWDfPtOSYKxzSjgFAAAAAKdItHP6vyS9ZozZLMlIOlrSFSmrKsnCgcg+07rWTpUVEE4BAAAAwGkSndb7ijFmkqSK6FOfWGvbU1dWcoX9kUBa09yhcaUBeT2GPacAAAAA4CBHDKfGmIsP89IxxhhZa59OQU1JF5vKW9fSIWOMwgEfe04BAAAAwEH66pyeH/3vMElzJL2iyLLeuZLekuSKcDqxPKgHrjhJxx8VkiSVBHws6wUAAAAABzliOLXWXiFJxpgXJU2x1u6IXo+U9EDKq0uSwnyvKiuGxa/DAa9qm1nWCwAAAABOkei03jGxYBq1S9LYFNSTMi9v2KX3t9RJkkqCdE4BAAAAwEkSndb7ijHmD5IejV5/VdLLqSkpNb739Ic6a8ownTAmHNlzSjgFAAAAAMdIdFrvt4wxF0n6cvSpe6y1K1JXVvKVBL2qiQ5BKgl6VdvSKWutjDFprgwAAAAAkGjnVNEw6qpA2lukWxrZZ1oc8Km7x6qhrUshvzfNlQEAAAAAjrjn1BjzXF8fkMjvOEFJr+NjigORo2U4TgYAAAAAnKGvzumXjDG/OcLrRtKUJNaTMsVBr2o/64w/lqTalg6NUzCdZQEAAAAA1Hc4vSCBz3BF+/Ga04/R4jnjJfXqnDIUCQAAAAAcoa9zTv84VIWk2tjSQPzxvmW9nHUKAAAAAE6Q6DmnrrelpkUPvV2l2uYOFQfpnAIAAACAk2RNON24u1Hff/YvqqpuVlF+rjw5hnAKAAAAAA7R17TeoiO8Njb55aROOLqUt66lU8YYFQe8qmFZLwAAAAA4Ql+d05WxB8aYVw547ZmkV5NCJdFwWhM9PiYc8KmOzikAAAAAOEJf4dT0elxyhNcc78AJvSUBXzyoAgAAAADSq69wag/z+FDXjlZ4wD7TcMCruhaW9QIAAACAE/R1zukwY8w/K9IljT1W9Lo8pZUlWU6O0Sv/fLrKCvMkSSVBn9ZuqUtzVQAAAAAAqe9weq+kwkM8lqRfpKSiFBpXFow/ju05tdbKGFetUAYAAACAjHPEcGqtXXa414wxJyW/nNR6/oMdamzr1D/OHquSoFed3VZN7V0qzPemuzQAAAAAyGr9OufUGDPFGHOzMeZvkn6eoppS5jfvb9PyN6sk7RuQxL5TAAAAAEi/vpb1yhgzTtJl0Z9OSUdLmmWtrUplYalQHPBpbUtd/LEUOVpmTEkgnWUBAAAAQNY7YufUGPO2pOcVCbGXWGtnSmp0YzCVpOKgT7XRfabFwf2PlgEAAAAApE9fy3p3KTIEabj2Ted11REyvZUEfPF9psWByD5TwikAAAAApN8Rw6m19kJJx0taI+lGY8ynkoqNMbOHorhkC8cCaXOnSmKd02b2nAIAAABAuvW559RaWy9puaTlxpjhki6V9CNjzFhr7ZgjvdcYky/pT5Lyot/1lLX23wdf9sCcf8IonXv8SAV8Hlkr5Rg6pwAAAADgBH2G096stbsk3S7pdmPM0Qm8pV3SGdbaJmOMV9IbxpgXrLXvDKDWQcv3euKPjYmcdUo4BQAAAID0O2I4Ncb8po/3f+VIL1prraSm6KU3+pO2Pav1LZ2647WNOmfqCM08ukThgJdlvQAAAADgAH11Tk+RtEXSo5L+LMn09wuMMR5F9qweI+ln1to/9/czksXK6t7XP9WIkF8zjy5RCZ1TAAAAAHAEE2luHubFSLA8S5EzTqcpcqzMo9bav/T7i4wJS1oh6dvW2vW9nr9K0lWSVF5ePvOJJ57o70cnrMdaff0PLTpvoleXTPLpJ++1aW+r1c2n+lP2nRg6TU1NKigoSHcZcCnuHwwU9w4Gg/sHA8W9g8FI5/0zd+7cNdbaWYd67YidU2ttt6TfS/q9MSZPkZC60hizzFp7R3+KsNbWGWNek3SOpPW9nr9H0j2SVFFRYSsrK/vzsf1W/MZLKiobocrK4/W7ve9r+1/3KNXfiaGxcuVK/l9iwLh/MFDcOxgM7h8MFPcOBsOp90+fA5GiofQfFAmm4yT9VJEOaJ+MMeWSOqPB1K9IF/a/BlxtEhQHvPGlvMVBn2qbO2WtlTH9XrEMAAAAAEiSvgYi/VLSVEm/k7Ss93LcBI2U9GB0eXCOpCestc8NqNIkKQn61NLRLUkqL8hTR3ePGtq6FPJ701kWAAAAAGS1vjqnX5PULOmfJF3Xq7toFBnGW3SkN1trP5A0Y7BFJtNjV50iT07k7ygt8EmSqpvaCacAAAAAkEZ97TnNGapChkosmEpSaTBPklTd3KEJ5emqCAAAAACQceGzLy9t2KWlj6+TtVZlBZFwurexPc1VAQAAAEB2y7pwunlPk1as3abmjm6VRZf17m3mrFMAAAAASKesC6clwUggrWnqUHFw355TAAAAAED6ZF04jS/lbW6X15Oj4oBXewmnAAAAAJBWWRdO903o7Yhe58UfAwAAAADSI+vCaVlBnkqCPnV09UiSSoM+wikAAAAApFlf55xmnFFhv977/lnx67LCPH20vSGNFQEAAAAAsq5zeqCyoI89pwAAAACQZlkZTm/8zV/001c2Soos821o64ov8wUAAAAADL2sW9YrSR9srVPAF/nTS6PTe6ub2zUy5E9nWQAAAACQtbKyc1pakBdfynvg9F4AAAAAwNDLynBaVuDT3mgYjZ97yr5TAAAAAEibrAynpcE81TS3q6fHqizaOd1L5xQAAAAA0iYrw+nYkoAqRhSpuaNr355TOqcAAAAAkDZZORDp0pPG6NKTxkiSrLXK9+awrBcAAAAA0igrO6e9GWNUGsxjIBIAAAAApFFWhtNtda268Gdv6rVPdkuSygrztLeZcAoAAAAA6ZKV4TQvN0frttRpS02LJKks6NPeRpb1AgAAAEC6ZGU4LQ74ZMy+Cb2lBT5VNxNOAQAAACBdsjKcenKMSgK++ITesoLInlNrbZorAwAAAIDslJXhVIp2S+Od0zx19VjVt3amuSoAAAAAyE5ZeZSMJJ00rkQFeZE/v6zAJymyzDcc8KWzLAAAAADISlkbTv/jouPjj8sK8iRJ1U3tOmZYQbpKAgAAAICslbXLensr7dU5BQAAAAAMvawNp4+t+lyn/N9X1NHVs69zysReAAAAAEiLrA2nPVbaUd+mmuaOg46WAQAAAAAMrawNp/uW8rbHj5bZ20TnFAAAAADSIWvDaWxCb3VzR/Q6L37uKQAAAABgaGVtOC0N7pvQK0U6qSzrBQAAAID0yNpwWlaYp7OnDI8PQyovzNPuxrY0VwUAAAAA2SlrzzktyMvVPZfPil+PKMrXroZ2WWtljEljZQAAAACQfbK2cxpjrZUkDS/KV0dXj+paOtNcEQAAAABkn6wOpwvv+7Ou/OUaSdKIUL4kaWcDS3sBAAAAYKhldTj15Jj4PtPhRZG9p4RTAAAAABh6WR1OS4N5qo5O6B1eFOmc7qonnAIAAADAUMvqcFpW6NOexsgQpGGFLOsFAAAAgHTJ6nA6oihfHd09qm3plC83R2UFPu0inAIAAADAkMvqcDptdFiXn3K0enpN7N3V0J7mqgAAAAAg+2TtOaeSNPPoYs08ujh+PbwoXzvZcwoAAAAAQy6rO6eS1Nndo9aObkmxzinhFAAAAACGWlaH0+4eq+O+/3v9fOXfJEX2oFY3d6i9qzvNlQEAAABAdsnqcOrJMSot8GlHdCnviFDkrNPd7DsFAAAAgCGV1eFUinRLY8fHxM86ZWkvAAAAAAwpwmlo3xCkfeGUzikAAAAADCXCaa/O6YhoON1J5xQAAAAAhlRWHyUjSX83ZbhGhPzq6bEKB7zy5eawrBcAAAAAhljWh9PTJpXrtEnl8esRnHUKAAAAAEMu65f1dvdYba1tUUNbp6T9l/kCAAAAAIZG1ofTz6qb9aX/ek0vb9glSRpWlMeyXgAAAAAYYlkfTkeE9h+CNKIoX7sa2mStTWdZAAAAAJBVsj6cBny5KsrPje8zHRHKV1tnjxpau9JcGQAAAABkj6wPp9Khzzpl3ykAAAAADB3CqaQRIf++Zb0hwikAAAAADLWsP0pGkq44dZw6unokScMLI+F0F8fJAAAAAMCQIZxKmlsxLP54WFGeJDqnAAAAADCUWNYrqaGtU+9W1ai5vUv5Xo+KA16OkwEAAACAIUQ4lbSmqlb/46639fHORkmRPag7WNYLAAAAAEOGcKp9E3pj3dLRxX5tqWlJZ0kAAAAAkFUIp+o1oTfaLR1THNDW2lZZa9NZFgAAAABkDcKppOKAV77cnPgQpDElfrV2dmtvU0eaKwMAAACA7EA4lWSM0Yii/P06p5K0pZalvQAAAAAwFDhKJuo/LzpeJUGfJGlMSTSc1rToxLHF6SwLAAAAALIC4TTqS5PK4o/HlPglSVtrW9NVDgAAAABkFZb1Rm2ra9Wz67apo6tHAV+uygp8TOwFAAAAgCFCOI1662979U+PrdP2uki3dHRxgD2nAAAAADBECKdRY0v2H4I0piSgLTUs6wUAAACAoUA4jYoNQfo8upR3TLFf2+pa1dXdk86yAAAAACArEE6j/l979x0n11Xfffxzpu7M9t6LerGqLRdZtiX3AlgYCBgTqgMGTEIIgYCf5AkmoRMSeOxADAEMGLAB29jY2Ma25G65yLJ6WdXVarW9t2n3+ePemZ1d7aruanbl7/v12tfM3DvlrHR3Zr73nPM7xVlp+NyuoXCaFyQas2hwlpcRERERERGRiaNw6nC7DBW5gUQRJK11KiIiIiIicvpoKZkkd37g7KS1Tp3lZNr6YUYqWyUiIiIiInLmUzhNMq80K3G9LCeAy6jnVERERERE5HTQsN4ku5t7+OHa3XQPhPG6XZRmB7TWqYiIiIiIyGmgcJpkV2MP33psO/ta4kWRAtS1azkZERERERGRiaZwmuSItU5zg+o5FREREREROQ0UTpPEiyAlLyfT1D3IQDiaymaJiIiIiIic8RROk2SmeckNepPCqVOxV0N7RUREREREJpTC6QhVecEj1zrV0F4REREREZEJpaVkRvjfj5xLdsAL2MN6QcvJiIiIiIiITDSF0xEKMvyJ64UZfvweFwdaFU5FREREREQmkob1jrD9cBf/94+baeoawOUyTC/MoLa5J9XNEhEREREROaMpnI7Q1hPiFy/tp7bJDqSzijLY1ahwKiIiIiIiMpEUTkcYOc90dnEG9R399A5GUtksERERERGRM5rC6Qil2Wl4XCaxnMzMokwAdmtor4iIiIiIyIRROB3B43ZRnhtgv1MEaVZxBoCG9oqIiIiIiEwghdNRzCzMoGvAHsZbnRfE6zbsbOpOcatERERERETOXFpKZhT/88Fz8Ljt3O5xu5hekEGtek5FREREREQmjHpORxEPpnEzizPY1aRwKiIiIiIiMlEUTkfR3D3I39z9Gmu2NwH2cjJ17X30h6IpbpmIiIiIiMiZSeF0FJlpHtbsaGL9gXYAZhdnYlmq2CsiIiIiIjJRFE5HkeZ1U5MfZMdhuwjSrCK7Ym+thvaKiIiIiIhMCIXTMcwpyWRHox1Oq/PT8bgMu1SxV0REREREZEIonI5hdnEmB9r66AtF8Hlc1BSks1MVe0VERERERCaEwukYFlVkc3ZVLu19YcAe2qthvSIiIiIiIhND65yO4bK5xVw2tzhxe1ZRBo9vOcxAOEqa153ClomIiIiIiJx51HN6nGYWZxKzYG9Lb6qbIiIiIiIicsZROD2Kz927gU/84jVgqGLvzkYVRRIRERERERlvCqdHYQxsPNgJwPTCdHxuF1sPdaW4VSIiIiIiImcehdOjmFOcyeGuATr7wvg9buaVZbGhriPVzRIRERERETnjKJwexZySTIDEeqdLKrLZVN9JNGalslkiIiIiIiJnHIXTozginFbl0BeKsqtJ805FRERERETGk8LpUZRkpbF6SRlVeUEAFlfkALDhgIb2ioiIiIiIjCeF06MwxvD9G5eycnYhANMK0slK8/DmQYVTERERERGR8aRwehxaegYZjEQxxrC4MocNdZ2pbpKIiIiIiMgZReH0GF7a3cqyf3+S1/a1A7C0Mocdh7voC0VS3DIREREREZEzh8LpMcwvywJg/X47N2dOyAAAIABJREFUnC6uzCFmweZ6rXcqIiIiIiIyXhROjyE74GVWUQZvOOubLq50iiLVtaeyWSIiIiIiImcUhdPjsLQqhzcOtGNZFgUZfipyA7ypeaciIiIiIiLjRuH0OJxdlUt7X5h9rX0ALKnMYUOdKvaKiIiIiIiMF4XT43DJ7EK+/Z5F5AV9gB1O6zv6ae4eTHHLREREREREzgwTGk6NMZXGmDXGmK3GmC3GmM9O5OtNlLKcAO9dVkl20AvY4RTg9f2adyoiIiIiIjIeJrrnNAJ83rKs+cAFwK3GmPkT/JoToq6tj8c2NwCwqCKHdJ+b53Y1p7hVIiIiIiIiZ4YJDaeWZTVYlrXeud4NbAPKJ/I1J8rvXqvj0/esp3cwgs/jYvmMAp7d1YxlWalumoiIiIiIyJR32uacGmNqgKXAutP1muNpaVUuMQs2HrSr9K6cXUBdW3+iSJKIiIiIiIicPHM6ev6MMRnAM8DXLMu6f8S+TwCfACgsLDznvvvum/D2nIyekMXfPt3H9TO83DDLR1NfjC8+289fz/NxRbU31c0ToKenh4yMjFQ3Q6YoHT9ysnTsyKnQ8SMnS8eOnIpUHj+XXnrp65ZlLRttn2eiX9wY4wX+ANwzMpgCWJZ1F3AXwJw5c6xVq1ZNdJNO2s93v0htf5RVqy4G4M4ta2iwMli16twUt0wA1q5dy2Q+fmRy0/EjJ0vHjpwKHT9ysnTsyKmYrMfPRFfrNcD/Atssy/reRL7W6XDF/GK2NXTR0mMvIXPJrEJe2tNKKBJLcctERERERESmtomec7oC+CBwmTFmg/Nz3QS/5oS58dxKXvk/V1CQ4Qfs9U/7QlFe29+W4paJiIiIiIhMbRM6rNeyrOcBM5GvcTrlBH3Dbi+fkY/HZXh2ZwsXzihIUatERERERESmvtNWrfdM8eq+Nj7801foHgiT4fdwTnUuz+7UeqciIiIiIiKnQuH0BMViFs/sbOa5XS2APbR3a0MXTV0DKW6ZiIiIiIjI1KVweoLOqc4lO+DlyW2NAFw1vxiARzY1pLJZIiIiIiIiU5rC6QnyuF1cNreINdubiERjzCrO5KyyLB58oz7VTRMREREREZmyFE5PwhXzimnvC/PKPrtK7w1Ly3nzYCd7mntS3DIREREREZGpSeH0JFw6t5DzpuVhWfbtdywuwxh4cMOh1DZMRERERERkilI4PQlBn4f7blnOipn28jHFWWmsmFHAg2/UY8UTq4iIiIiIiBw3hdNT0D0QZnN9JwCrl5RxoK2P9Qc6UtwqERERERGRqUfh9BTc+us3+NQ9rxOLWVyzoAS/x8UfN6gwkoiIiIiIyIlSOD0F7zmngrq2fp7d1Uxmmpcr5hfz8JuHGIxEU900ERERERGRKUXh9BRcc1YJ+ek+7ll3AIAbz62kvS/MQyqMJCIiIiIickIUTk+Bz+PivedW8tS2Rura+rhoZgFzSzL53+f3qjCSiIiIiIjICVA4PUUfWl6Nx+XiL1sbMcZw80XT2H64m+d2taS6aSIiIiIiIlOGwukpKs0OsOYLq/jYRdMAuH5JGYWZfn783J4Ut0xERERERGTqUDgdB+U5AQC6BsL4PW4+cmENz+1qYfvhrhS3TEREREREZGpQOB0nD7xxkAu+/hSHOvq56bwq0rwufvLc3lQ3S0REREREZEpQOB0n59bkEY7G+O+1teSm+3jfskoefKOefS29qW6aiIiIiIjIpKdwOk4qcoP81bJK7nv1IPtbe7n10pl43S6++8SOVDdNRERERERk0lM4HUefvXwWHrfh3/60jaKsND5+8TT+tLGBN+s6Ut00ERERERGRSU3hdBwVZ6Xxt5fN4tmdzexv7eXjl0wnL93HN/+8XeueioiIiIiIHIXC6Tj72EU1/OUfLqE6P53MNC9/d9lMXtrTyjM7m1PdNBERERERkUlL4XSc+T1uqvPTAWjo7Oem86upzg/ytUe2EYrEUtw6ERERERGRyUnhdILc8fQurviPZ2jo7Odf3zGfXU093PXs7lQ3S0REREREZFJSOJ0gN5xdgctl+OxvN3DxrELetqiUHzxdy57mnlQ3TUREREREZNJROJ0g5TkBvvmuRWyo6+D7T+7iX98xH7/HxW0PbFJxJBERERERkREUTifQ2xaV8r5lldy5tpY9zb3cdt08Xt7Txr2v1qW6aSIiIiIiIpOKwukE+9fr57OoPJtQJMb7llWyfHo+tz+8ldqm7lQ3TUREREREZNJQOJ1gQZ+HBz69gktmF+JyGf7zfYsJ+tzces8b9IeiqW6eiIiIiIjIpOBJdQPeClwuA8DPXtjLq/va+N77FvORn73Kvz60mW+/Z/EJP180ZrGhroOF5dn4PC5+8dI+7ni6Fq/bxYyiDC6dU8hlc4sSS9qIiIiIiIhMdgqnp1HMgkc3HSbo8/CJi6fxP8/u5eyqXG48r+qYj916qIs/bTzEGwc62Hiwg95QlHv+5nxWzCygOj+dy+cVMRiOsaGug9sf3srXHtnG2i+soiI3eBp+MxERERERkVOjcHoafWxFDR19Ie5cU0t+hp+5JZnc9sAmirPTuHRO0ZiP23iwg+vveAGPyzC/LIt3n1PBspo8FlZkA7BydiErZxcm7r+vpZe69r5EMH1qWyMrZxficWsUt4iIiIiITE4Kp6eRMYbPXzWHq+aX8OUHNrLlUBezizK49Z71/Oivz6Y4K0DXQJjGrgFeqG0hN+jji9fMZWF5Nv/+zgW8Y1EZ2UHvMV+npiCdmgJ7SO/Ggx3cfPdrXDyrgDvef/ZxPV5EREREROR0UzhNgYUV2Tz46RWs3dHMosps3vXfL3Lz3a8Rjg6tf5rp97B6aRlgh9q/vqD65F6rPJtvvmsh//LHzay+83l+8uFlzCzKHJffQ0REREREZLwonKaIx+3iivnFANz9sfNYfccLBLxw23XzWFCezZySTLzjMAzXGMON51UxsyiDT/7qdW6480V+9MFzWDGz4JSfW0REREREZLxoEuIkMKMwg/s/fSF+r5tvP74DYxiXYJpsWU0ef/zMRZTnBmjoHBjX5xYRERERETlVCqeTxOziTO67ZTlpHhfvv+tl1u1pHffXKM8J8NBnLuI951QAcLC9b9xfQ0RERERE5GQonE4i0wrSufeW5RRk+vnAT9bxm1cOjPtr+Dz2f/mWQ51c9h/P8ONn94z7a4iIiIiIiJwohdNJpjIvyAOfXsGFMwv48v2b+MpDWwhFYuP+OrOLM7lyXjFfe3Qb//PM7nF/fhERERERkROhcDoJZQe8/PTDy7j5omn8/MV9vPuHL7KnuWdcX8PrdvH9G5fw9kWlfOPP27lzTS2WZR37gSIiIiIiIhNA4XSS8rhd/Mvb5/Ojvz6buvY+3vaD5/ntKwfGNUB63C7+631LuH5xGd95fAdPbG0ct+cWERERERE5EQqnk9w1C0p57LOXsLQqhy/dv4kb73qZ2qbucXv+eED93nsXc+W84nF7XhERERERkROhcDoFlGSn8aubz+cb71rI9sPdXPv95/jWY9vpGgiPy/O7XIZ3nV2By2U42N7HjXe9RG3T+A4jFhERERERORqF0ynC5TK8/7wqnvr8Sq5fXM4P1+5m1XfW8vMX9o5rwaTGrkF2Nfbwzjtf4PEth8fteUVERERERI5G4XSKKcjw8x/vXczDn7mIuSWZfOXhrVz63bX86uX9DISjp/z851Tn8vDfXsSMwnRu+eXrfP3RbePyvCIiIiIiIkejcDpFLazI5p6/OZ+ff/RcirL8/PODm1n5nTX8zzO76ew/teG+ZTkB7r1lOTedX8Vdz+7hjqdrx6nVIiIiIiIio/OkugFy8owxrJpTxMrZhby4u5U7nq7lG3/ezvef2sVfnVPBB5dXM7Mo86SeO83r5us3LOSas0pYXJkDQG1TDwUZPnKCvvH8NURERERERBROzwTGGFbMLGDFzAI213fy0xf28utXDnD3S/s5pzqX9y2r5JqFJWSleU/4uS+ZXQiAZVl8/r4N7Gvt428vm8kHzq8m4HOP968ixykUifHi7hZqm3rY1dhDS88gxhgunlXAhy+sAWDdnlbmlGTqZIKIiIiITAkKp2eYBeXZfO+9S7jtunncv/4g975axxf/sJF/fnAzF80q4LqFpVw5r5js4IkFVWMM33z3Ir7x5+38+yPb+P6Tu3j74lI+eEEN88uyJui3kWSWZXG4a4DS7AAuA//4u4209AySn+6jJDuNmAVtvSEA+kIRbvzxywAsKs/m4lmFXLOghAXl2an8FURERERExqRweoYqyPDziUtm8PGLp/NGXQePbmzgz5sP8/T2Jrxuu6f12gUlrJxdREl22nE957zSLH7xsfN4ZW8b975ax4NvHGJxRQ7zy7Jo6Oznsc2HWVKZw7zSLNK86lUdL72DER7cUM8vXtxPZ3+YF750GR63i/tuuYCcoI+89CN7Rj0uF/fcfD7r9rbxfG0LP3xmN3esqeXrNyzkpvOrUvBbiIiIiIgcncLpGc4Yw9lVuZxdlcv/eds83jzYyZ83NfDIpgb+6Q+bAJhdnMFFMwtZVmPf71hh9bxpeZw3LY/bV5+Fx2UAeHlPK7c/vBUAj8swqziTmvwgt103j8q8IM3dg/SHouRl+DBJzxX0uTHGYFkWxphRXu2tq6MvxE+f38vPXtxH90CE+aVZfO7KWcQsCzeG6YUZYz7W53Fx4cwCLpxZwOeunE1HX4iH3jzE5fOKAHh6eyPr9rbxsRXTKM46vpMTIiIiIiITSeH0LcQYw5LKHJZU5vCla+ey/XA3z+5s5tldzfxq3X5++sJeAMpzAiytyuHsqlwWV2YzuziTzFHmq2b4hw6fG5ZWcP60fDYe7GTjwQ62NXSxo7Ebn8cuCH3fa3V85/EdRzzHm/96FdkBL996bAc/eW4PAZ+b0uw0qvLSqckP8sVr5uLzuOjsDxP0ufG63zoFptcfaOcHT9dyzVklfPySaZxdlXvSAT4n6ONDy2sStzce7OTHz+7hp8/vZfWScj5xyXRmF59c8SwRERERkfGgcPoWZYxhXmkW80qzuGXlDEKRGFsbunh9fzvrD7Tz+v52/rSxIXH/8pwAc0symVuayZySLKYXpFOdHxwWWstyApTlBLhmQckRr3f1WcUUZPiOWObG74TX5TPycbugdzDKwfZ+DrT1sqGug//ztnkA3P7QFv745iHKctKYXpDB4opsllblcuncoon450mJ3sEIP3thLy6X4dOrZnLpnCKe/IeVzCwau4f0ZP39FbN519IK/vf5Pdz7Wh2/f/0gH15eze2rF4z7a4mIiIiIHA+FUwHsYaDxXtWbmQZAQ2c/W+rtHtDth7vZcbiLZ3Y2E4lZicflp/uoyg9Sk59OVV6Q6vwg1fl2cM1P9yV6+mYWZR51WZuVswtZ6VQGjkse6vuOxWWU5QTY39bHrsZu7ljTzIzCjEQ4/eLv32TXgQGe69lKdX6Qkqw0KvOCzCud/MWaBiNRfr3uAHeuqaWlJ8TqJWWJ330igmlcVX6Q21cv4LNXzOaXL+2nLMce3tsfivKrl/dzxfxiphWkn9JrdPaHWb+/na6BMKuXlANw2XfXcrCjH7BPTuQGfVw0q4Cv37AQgK6B8ElVlpbJacuhTp7b1cKmg510DYQZDMdwuwy//vj5GGPYfriLvHQfRZkaXj5eBsJRXt/fTmtviPbeEP3hKOk+N0urcllQnk0oEmNnYzcl2WkUZPhT3VwRmQJ6BiPUNvVwuHOASCzG2xeVAfBCbQsdfWGmFaQzoygdv0c1R+TUKJzKmEqzA5RmB7hifnFi22Akyp7mXva39rKvtY/9rX3sb+3llb1tPLihHmsotxL0uanIDVCZG6QyL2hfzws6twOjDhVOljyE9dK5RcN6SftCERq7BhO33S5Dc1+Me9btZyAcA2DVnEJ+/tHzAHjPD1/E7TIUZaWRn+6jMNPP2VW5LJ+Rf0r/Rqfqxd0tfOF3G6nv6Gf59Hzu+tAczq7KPa1tyEv38dkrZiVuP7urma89uo2vPbqN6YXprJpdxHnT8rh4VgHp/mO/Zazd0cST2xp5bV87Oxq7sSyoyQ8mwunHL5nO/tY+wP4S3d4XosSZ9xqNWVz8rTVU5we5YWk51y8uI19fnqcUy7LYeLCTRRXZGGO4Z90Bfr3uANX5QXKDPgJee+h+/O/7Xx7czKv72llckc1VZ5Xw7rMrjrtIm9gnEdfv7+CNA+3MLc3iPedUMBCO8oGfrDvivl+4eg4LyrNp7Brg7f/veQBKs9M4qyyb86bl8vZF9klAEXlri8UsXE5Nkf96cie/e+0g9c5JZbC/N8TD6Y+f28PaHc2AXXNkZlEGV8wr5h+vnnP6Gy5nBIVTOSF+jzsxHHikwYg9JHd/ay/7W/uoa+unrr2PurY+1u1to2cwMuz+OUFvIqhW5AapzA1Q4YTXitzAUSv+Bn0ephUMHb7feNci1ua1sXLlSpq7B2nsGsTrsd9YLctiWkE6e1t62XSwg9aeEN2DEW46v4rlM/KJxiyu/q9nmVaQzoKybM6tyWVJVQ5B38T8eRzq6MfCHiqdlealIMPHt969iItmFUzI652oq88q4bkvXspT2xp5antTYj7ymn9cxTS/h1+8tI/fvlKHx22IWRbRGLT3hnj+ny7F43bx5LZGHlhfz9nVuVy3sJRlNbksrRwK3O8/b+xqweFojE+tmsHDbx7i9oe38rVHtrFydiF/d/ksFlfmnIbfXk5WNGbx6KYG7ni6lh2N3fzhUxdyTnUun7l0Jv9w5ewxe+i+unoBT2+3T2h85/Ed/McTO/j4JdP58rXzTvNvMHVYlsV/PrmLJ7YcZvvhbsAehXDzRfaol5ygj/tuWU5eupecoI+gz03vYBS/155GkZfu438+eA51bX1sru9kY30nT25rpCI3SFlOgLq2PjbVd3LB9PxRq4GLnG7dA2E21XdyqGOAho5+WntD9IUifHTFNOaVZrHlUCd3v7iP3HQfFc4Uo9nFmVTkBlRs8TiEozE21Xeybk8br+xt5Y26Dp774qVkpnnJz/BzTnUuN51fxayiDMpzA8PeF77znsW09AxS29TD9sNdbDzYyd7W3sT+rz+6jbPKsrh8XvGwWiUiY9FRIuPG73EzozCDGaNUkbUsi46+sBNWh0JrXXs/2xu6eXJrE6FobNhjijL9Tk/rUI9rRZ7dE1uanYZnlOJIxti9o0VJFWiNMXznrxYPu19/KMpAOArYQ1XmlmSytaGLJ7c1Yll2T+z/fft8PnxhDdGYhWVZo77e8eoaCPPYpsM88EY9L+9t5YYl5XzvfUtYUJ7Ng7eumHQfnpV5QT6yYhofWTGNwUiUzfVd1OQHAcgOeCnLSSMas3AZgzGGheVZDERiZLhdfPGauXzlHWed1L9XmtfNJ1fO4JMrZ7DjcDf3v3GQP75xiMGIfWzsa+mlsWuAc2vyEmd1U6Wpa4DNhzqZVpDBtIJ0djZ284Xfb6S9N0RHXwjLApfL8M13LeTahaXsa+nl7pf2Mac4k7mlWcwpziTgm/rDnyLRGI84oXRXUw+zijL41rsXMrfEHsZ/rJ64+MmuWy+dyf7WXu57rY65JfbJr7beEF++fyOXzy3mwpn5VOQGJ/T3iFngMpzS3/pE2dnYzYa6Dt67rBJjDK/vbyMn6OW26+ZywfR85pZkJQrQgV1VPVnyybZ0v4erzxpeG6C+o5+8oP2F87HNh/nao9sAqMgNMK80i7klmXxq1QyCPg+HOvrpC0XIS/eT4fcQicUIR6wTXj9bZCyHOvp5dmczZ5Vls7Aim831Xdz046HRAFlpHoI+D+9YbPfetfSEeGZnM229IcLRoSFcv/n4BSyfkU9tUw/7W3tZWJE9IVMI6tr6aOkZpKUnRGvPIO19YWYWZXClM/LsO49vx7Lsz7jMNA9ZaV7mlGQm1h6va+sjK81LRpoH9wR+tg2Eo3T1h2npCVGeEyA76OWxzQ187t436Xe+E80oTOfaBaX0h6Jkpnn54AXVfPCC6jGfszDTT2Gmn3mlWYn/D8sZRtc9EOahDYe469k9+DwuVs0u5G2LSk9bUB2MRGnqGqTJWTEiEotRlJnG/DL7M+ZAax95Gb6UhGbLsmjuHmR/mz0Ksb69n8NdA3z+Kvtk7oNv1POjZ3bjdhn8HhdpXjdFmX5ue9s8ijLTqO/op3sgTGl2gKw0z6T7HnkqFE7ltDDGkJvuIzfdx6KKI3vAYjGLpu7BodCaFGBf3dfOQ28eImmqKx6XoTQnjfKcAMVZaRRl+uluCtOZU5+4XZSVNuYbTsDnTgSD7ICXO246G7BD5Ov723l1bxtLq+x2vrynlU/+6nVWzCjgktmFnDctj+kF6ccdjv7lwc3c+1odoUiMmvwgf3/5bG5YWj7s32Yy83vcnFM91PO5ekl5YojuaMZrvuickky+fO08vnj1XOL/1He/tI+fvbCP8pwAF88qYPmMfM6flj8hw0Aty6KzP0zMsnuaugbC/MfjO9jT0sv2w900d9vDyv/5bfP4m4unE/C6yQ54qckPkhPw4nIZYjErEagOtvfz21fqEl8A3C7D/NIsvvXuRcwvyxo2jGoqCUctbn94K/npPu64aSnXLSg96d+jOj+dL1w9N3F7X2svGw928viWRsAeHr6kMoe/u3wW0wsz6OgL0T0QISvgJdPvOeJ1I9EYjd2DHGzr42B7P/Ud/TR09vP1GxbaJ60e384fXq+nayBMX8j+f8nwe9h8+9UA/P1v3+CRTQ1k+D1kpnnJDXqZXZyZONnVOxghOIEnGPa39vLIpgYe2nCI7Yft6ufXLighM83LLz52/rh+iS1POonwkRU1nFOTy7o9bWw51Mm2hi6e2tbIp1fNBOAnz+1NVHeP87gMtV+/DoDbHtjEwxsOEfC5Sfd7KM7yM7s4k9uvPwtjDJ394VH/v+StKxKN8dKeVp7Z0cwzO5vZ1dQD2EPRF1Zks6Qyh1/dfD4VuQFKstOOGFm1cnYh6267gljMoqVnkLr2frY2dLGwwg5/D715iB88tQuAsuw0FlXksKgym4+tmHbc67J39oXZ3dLD7qYetjZ0UZDh59ZL7b+J1Xe+QFtvaNj9Vy8pS4TTX7y0n4FwdFhw/vDyahaUZzMYiXLxt9cktmf4PWSlefjYRdP4m4un094b4iZnmL4B4l8ZPrpiGgXYQf6Tv3o9sd++YvjkJdO5dmEpWw918ZGfvUJnfzhxohfgB+9fyvWLy5hZlMH7zq3kvGl5nFuTR2HmqU+liX+vyUzz8uKXLmP9AbvI5p83N/DE1ka+8a6FvP+8KnoHI0Qt65S/N8RiFnta7EKaPo+L652QfO6/P0nXwPBRe+9aancOAFzxvWcIRWOk+9wUZ6dRnJnG6iVl3HheFZZl8eimwxRn+e3vlVn+E5pPa1kWXQMRmroGaOgcsENoSy83nV/F9MIM7l9fz+d/9+awx+Sn+/jQ8moKMvxkBTxU5QWJWRaDkRh9oSiv7W/H67JPQv5m3QHuWFML2MdMaXYaZTkB/ueD55DmdbPlUCed/WHKc+ye7nTf1HnPVTiVScHlMpRkp1GSnca5NXlH7A9HYxzq6Odge7/T42oH2EMd/bxxoIPGrgEGIzF+u2PDsMel+9zkZfjIDfrICfrIDXrJDdq3c50hb/FtOUEveek+Vs0u5NI5Q/Nb89J9vG1hKc/ubOaxLYcB+43g4b+9iGkF6Tyx5TBrdzYzGI7RH47Q2R/mUMcAT/3DSlwuQ36Gj/efW8k7l5azpDJn0ofRySb5C/gXrp7DoopsHtl4mEc2NfDbV+vIT/fx2j9fgTGG+9cfpHcwQn6Gn7x0X2J+cU5waAhSXyjC4c4BDncNcLhzgIIMP5c4xbj+7jdv0NYboqGzn4bOAfpCUT5yYQ1fuf4s/B4XD244RHV+kItnFbCwPJsF5dmJHsLKvCC/+Nh5Y/4eF80qYMvtV1PX3sf2w91sOtjJa/vbKMiw2/bTF/bysxf2saQyhzklmcwpyWRuSSZVecFJdcz0DEb4zboDPLmtkV9//AICPjd/+NSFVOcFx/2D7+yqXF780mXsaurhhdoWXqht5ZW9bYl/j/vX1/PVP9nrK7sMZAW8pPs8/PEzKyjI8PPdJ3byo2d2D3vOwkw/X7k+ht/jpjovnYtnFZAd8JKZ5sXjNsOOt8vnFVOaE6BnIEL3QJjW3hBN3UNz3W+++1X2NPdSHoiwPrSDuaVZnFWWRXX+yRUSC0djWJZdoO7eVw8k1qI+uyqH268/i+sWlibm6k9k74rX7Uqsj53ctvhSXu87t5LFldm09YboGYjg9bjwuV2JQm4XTM/H53bRH4rSG4pwqKOfDXUdif+3z/72DV7a3eoUULFH2yytzDlt1dfjJ0MPdfbT0DFAe1+IlbMLqcwLUtfWx1+2NpLmdVOY6ac8J0BFXkBF2iZAJBrjUMcAVc6onE/fs57BcIzzpuXx3mWVrJxTyCynKGDA5z6uqS8u19DoqeSTqp9cOZ2LZhaw8WAHbzpL3j23q5lbLpkBwDf/vJ036zoozU4jI81DKBIjO+hNTC/4wE9e5oXa1sTzBbxurl04NPrg6zcsxO9xkZ/hoyDDT27Ql1iNAGDTV+wTXuFojJ6BCF0D4UQoNhi+/Z5FdA9E6OoP25cD4cSoE5fLUJEbcGp62OHWsuzvN2C/F+Sn+4jH3njtj/hIirx0H5fOKSI76CU7YP/kBn0sq7H/fWYWZfKV68865r/tyXK5DMtq8lhWk8f/fft81h9oZ5ZTIPP+9Qf5ysNbWVqZw8WzCllcmc30ggwqcgNjfp4MhKOJf7s7nt7Fc7ta2NrQRbcTQs+tyU2E0y9cPQefx0VRZhpBnxuP20WhM8XEsiy++e6FNHcPcrhrgKauQRq7Buh1TlS294W59dfrh712btDL566czYeW19DaM8h3n9i6goOSAAAbGElEQVRJ0OemPxylsy9Me1+Ij188nUvnFrFubxs33vXysMf7PS6Wz8hnemEGy2py+erqs5xioumU5aQNC7+XzS3msrnFjOU951QwrzSLQx39HOq0vw+39oQSx91Pn9/HH9YfHPaY4iw/6267AoBv/Hkbazb28/O9r5Du85Dud1OaHeBzV84G4Bcv7WN3Uw/hmEUkGiMStSjNSUucQP7u4zs42N6H2+XC7QK3y8WMwnT+5uLpAPzuNftkfJrHjd/rcupNBBInjI7GWMkVbFJszpw51o4dR66FKXIslmXx6JNrmb1oGU3d9htM/LK9N0S786bR3heiozdM94j5r8l8HtewwDoUbD1YFrT1hWnuHuSWS6ZTmOnngTfq+fW6A6R57d7YDL+H6YXpfHX1As2vmEDRmMXWQ10c6uxPDFG89vvPsa2ha9j9zqvJ475PLgfgqv98hp2NPcP2Xzm/mB9/aBlr167l+1u9WBaJM5Cl2WksrcrhnGr7hElyBenx9vT2Rn7/+kE213dxoM0uGOV2GbZ+9Wr8Hjd/eP0gB9r6qMoLUpUfpCovSGGG/7ScCY3FLNYfaOeRTQ3cv76ezv4wF87I5/s3Lh2Xs+wnq7aph/UH2unqD9Pp/PQMRviXt80nN93HGwfa2X64m4rcAOXOPLTj7SU5Hve9WscLu1t4ZVcDjX0WMWt4IbZv/nk7RZl+5pZkUpoTIC/oIzPNPnsdjsbY1tBFXVs/B9r62HKok2d3NvNv71zA6iXl1LX18eS2Rq6YV0xl3sQNZ06FRzY2sKGund3Nvexp7uFAWx/n1uRx7y323+mX799Ius/DvNIsphWmM70gfdgJpuMxEI7S2DVAfUc/+1r6OKc6lzklmby+v433/3gdocjwaSQ/+utzuGZBCX/Z2sjHf/HaEc/3y5vP4+JZhdS19VHf0c/C8uzjKhB3PNauXcuqVavG5bmSWZbF4a4Bgl4P2UEvtU09fP+pXTR3D9A7GKV3MMJgJMbXbljAqjlFbDnUyX/+ZRdFWX579FGmPQrpnOpccsdh7nF/KMq2w12s39/Our1tvLqvjdygj6c/vxJjDJsOdjKjKH3C6j2M1DMYSXxG/7+ndvHMzmYaOgfoDUXwuV1U5QX5/acuBOBXL++ndzDCjMIMphemU52fPqEniI7XRB07p8vWQ108sukQz+9qYWN9ZyJYb/vqNQR8br77+A4e33IYC/szvy8UwWB4+bbLAfjH373JnuYe5pZmJVacmFGYMS7/N+FojN3NdnXieHBt7B7gyvklrJxdyN6WXv7qRy/RF4oQ8LrJCdodHp9aOYMr5hfT0jPIA+vrKcryU5KVRlV+kOLMtNPWe3moo599Lb3Ud/TT0Wd/NrqMSRTA/K8nd/L4+t14g5n0DkboHYxSlOXnoc9cBMDNP3/V7ql1GzwuFx63YW5JJj/58LkA3PLL19h+uJtI1CIas4jEYiypzEnsX/WdNexzil/GXT63iP/9iL3fGPO6ZVnLRmu7wqmcMU7kTTocjdHRF6ajLym4OiG2wwmx7Un745fR2Oh/L8ZATuDIQJsT9JKZ5nGGBdpDAzP8HjLSPGQ6lxl+z5QabjGZhaMx2ntDtPaGaOsN0dIzSGaaJ3H28Ydrd2NhUZJl99LHL4M+z6T6kO8djLCzsZv6jv5ERcTP3bvhiIrY0wrSWfOPqwD45cv7aejoJ93vITvgpTw3wLT8dGpOcjmgpu4BsKAoK40Xa1u46Sfr8LldXD6viE9cMp2lp7mq9GS2du1aLlhxMbucEx8LK+zlWi785tO09AwOu2+8J76jL8SSr/4lsb04y8/K2YXcdH41S95ixb8GwlHaekOU5QSwLIu/+tFLbKzvHBYgP3hBNf/2zgVYlsWNd71MQYY9z83jMoSiMS6eVciV84tp6Oznuu8/R3vf8DW1b7tuLp+4ZAZN3QP85Lm9VOUFKctJoyTLHvKWm+7F73ETicboHYzSF47Q1DVIfYd98uDGcyvJCfr477W1fPuxHbgMzCrKtL8QV+Vww9Lykz7xMV7vPZ39YX697gC1TT3UNtvDT3sGI3zthgV84Pxqth/u4pZfvk5xpt07GPS58XlcfPTCaSysyOblPa185aEtNHUPDhuieu8nLuD86fms2dHEHU/Xkp/uoyDTT0GGn4IMH9cvLiMn6GN/ay+7GnvoC0fpD0Xo6AvT0DnAP79tHh63i3/6/Ubufa0OsIfon1uTx+XzirlqfrE+/07SZPrcOlUdfSF2NvZwqKOfdzpTn3718n5eqG3BGLtnzu9xMb0wnU9eMkPHzDiYyOOnLxSx67tEYok6LwGfO1GXRuFU3hIm+k06Pn8gOdB29IVo7x0Kr20jtnX0D81lO5YMv4eAz02aM/whzesmzeMmzecmzZkMn7zPn3Tb53HhcRlcxh6WGP8ZbZvb2Wb/2HND7G04BY7A43wI+L0uuw1eN36P64z+MJgKH/KDkSj17faX5Tqnd/WDy2sAeNd/v8DGg53D1iFO7jX+0E9foWcgnPhSX5DhZ2F5NlfML8ayLO56dg+Huwaobephx+FumroHuWWlXTU3XvTosrlFx1wC6q3oaMdOc/cgOw5309wzQGtPiPmlWVw4s4BYzOLJbY12sbe8oEZZjBCJxtjX2se+ll72tvRSU5DOlfOL6QtFuPnnr9HYbfdmRGMWfq+Lj188nVsvnclAOMq//WkrpdlplGTbox+q84OUZY89TPBEtPeG2FDXMeynPxRl0+1X4fe4+dXL+6lr62N+mT3EuyY//ZgFto73vScWs3hiayN1bX3sb+vlQJs9zWX1kjL+/orZdPaHWXz7ExRn+ZlZlMHMwgxmFmeywhlGeCJCkRgtPXYhmdnFGQR9Hp7Ycpifv7gvUfin3Sn89twXL6UyL8gP1+7mW49tH/Y8GX4PT//jSooy09hQZ0/BWVKZQ3GWlosaD1Phc0smr1QeP0cLp/o0FDlOxpjEfI3qE1geNRqz6BmM2D8DEXoGw3QNxK/bl92D9py2gXDUOcMUYyBiX+/sD9MUts869YedfeHosMIGp4vP7QRWJ6yODMwBZ2hz4rZzPeg7+v6A156TEA/U9qU9jMTjMpNqzmUq+T1uphdmjPpF8/5Pr8CyLELRGO29YQ6295H8z1aZG2Bfa4z9rX28vr+dtr4Q71hUxhXzizHGcMeaWqIxixmFGVw8q5C5JZmsnGPPxfW4XUctgiVji1eyHMnlMlw1omKuDPG4XXbAKhp+rAd9Hn7ziQvGfFya183Xblg4Ye3KTfcNW3fbsiwaOgcSc8U213fyh/UHE4VvXAbOqc7ld58cGh46EI7icRnCUYtwLEbHoQirnOf/xp+3Ud/en5h32D0Q4aKZBXzl+rMwBj5/3wZ6Q1GyA16q8oLML81imjM6IjvgZeNXrhqX+bE+j4syZyh83FVnlQw7ZiPRGG19IfLT7eP73WeXs2JmPkGfm4DPLuqTfDLrrTYiQEROjsKpyARzu4ZC7XiKxewKbgPhKKFojGjMSvxEYpazBumR2yJRCwsLy7KDc8yyr8cse85c/D6DkWji+ZMDsX0ZZdAJ0PF9PYMRmrsHEyG6zxnGkVyd8GS5nN7ceMEaO8DavcXxADsy0MZvu10Gn8dFmhOq7Z5g17De5zSPmwN1YdrWH0wEbnsS/9B9hwVyj2tSLjdijMHvcVOS7T6igvHIL+yRaGzYCY5XbrvijFjaRuR0MsYMC3DffPcivrp6AbVNPWw51MmBtr5hf1e/XneArSPmxc/LG3oveX1fO229IXu5kYCX0uy0RCVlYwwP3rqCosy0MZftOZ2Fmzxu17BlWUYu4yYicjIUTkWmKJfLDFsSZ7IKR2NDvb6hGP3O9fgchHiQDUdjRGIWUecy4oTqsBO8R73t9Dwk347EYsPuG47G6AtFEr3RI8P2MFveHP2XGIXHZRJB1u8EXq97qPfXldQL7HbCtNuZN+N22WE7fh+P838ZD8HxcJw8zNvvHepxHjb027ntc7tOqIfZ4x4esCf7cSQyVfg8LuaXZSXWUkz2yN9dRFd/hJhl4XEbvG4XLz7/bGJ/vADPWGYVZ457e0VEJhOFUxGZUF63Hdom41zFWMweBjsQjrLm2edZuuz8Yb3B8SA7OCLUjgy6g+EoA5Foompd1Om1jkTt6+FwjGgsOqwnO3GfWIxwxBr2fCfDGIYNr/aPEXDtbS58bjdej8Hv/P94Pfalz/nCHN8Wv+1zlgvxOffze5xto2zXMGyR0Rljjuj1dOnvRUQkQeFURN6yXC5DmssOdLlprpOubDueLGtouHZ8jrFd8S7KQPwyaVt/aGi49dA2OzwPJj1He2/Y2Wb3XociMULR2BFLaowHr9skAqvbmTMcL7gVL7rlMsOLcCXvG37/kbeT7u+KP/44759UAGx4qDaJsO5LCuLeEaE7fntkMO8YjNHZF05smwxLTIiIiExFCqciIpOIMSYxXPd0lA+xrPgQaLsXOT4UOhyxCEWjhCJD20MR+2cwknQ7aXtolG3xucyWZQ2b1zxynrNlWcRiDLu/BUc+3rlPvM3D949y/xGvF41ZiR7zeFvDTo/3KVnzROKqJx5+k4Os2x7+nbx9rH0jHxcPyPaPSQRnb2J/Um+3294Wn5+dHPzdxmBcw08MjDxJoF5vERFJJYVTEZG3MGOcolJuCPDWnXcan6McD63hpEs7jFtJ4Xyo53kwEmPz1u1UT585fF98f3joOeNFxkKRGD2DEVp7hu9Lftx4FBI7GaP1OA8PuYzaG+5yjfK4Eb3c7hG93GPlYMMoOwyJ1x6ttzy5HSYpdLuNwe/MCx+5NNbIquN+Z8i7PzEcPj7P237cmbyUlojIZKFwKiIib3l20So7kJyoop7drLpo2ri2J967OxiJEXHC6lBQHvoJRazht6MWYec+yb3SdmXuoV7kqJVUqdvZF433YCf3Qo98XGxkb7VFNHZkT3VyL/ewbdZQb/1oxlp6Pf58kejI3vjRqo4P/Y7RpKrm8RMDJyseZANJVbwDPvewud124HUPqyTucTuVxZOqiSe2uQ21B8O0vn4Qj3v4GtSuxDrVQ4HbPWLfsPuNOKFwxP2MXajN7ZyQ8rpceD12u7xuLdklIpODwqmIiMgkkzwfWsZPfAmuwciRS2QNhKMMJF9PzOMeMe/bmbcdnxc+GI4NW0ZrMBIbXlncqSJ+1JHjm4+/UvhEiYdlrxOi48PE49c9LjPsttc9FGwTgdgJv+6kYD10fehy2P6k0OyKVzdPbGPY8w57/sR1jvL6QyF95Ha3yySGwCcPh4//bgrrIqmhcCoiIiJvCalcgivmLHkVX+4qHlqfe/5Fzjv/gmG93fH50cPmSif1ZkdHzKMe3qt9ZI9y8r5oDLtKuNMbH++ZH74t3iPvtNeZkx5xlvqylwiLEYlGEnO/41XIY4lLjtgWn/Od2H+UXvRUi4dxrxNY/R4X/qMMCR/qOR9xmXTdN8pyYy6nJ9tlhi8/Ntp93EnB2u0yDEbsNcnj2xWo5UygcCoiIiIywVwug89l8OEatj0/4KIyL5iiVk0OicA6IrTGRg23SfutoeW5YsO2MXx//DmSQr59gmBoOHwkMTzeSsw3j8SsYUXeknvZByMxOvpCScuNxRIV0Qci0TGHqI+7Jx8D7OXEfO7kgmvuoYJrXtewfXbhNffwomxeF16XK1EYzTjPaXCKpcVvOwE4PoR85P3sfcnbndvOdZL2jXw8yfdPvN7Qc7uMOWKt74DPHmofcK5rObOpT+FURERERFLG5TK4RiuENUVZll1NPL5+9WB4qCDaULiOJXqxY87laGthR50e93jPe8yye91jlsXOXbVUT5s+VNDNCdDDi7ENzbeOF2Mbq1BbOBrDYuy531NFICm0pnldSdfdI/a5h83dTg7M8e0YY1ePj8+9x75kWCV4sLCvx+fCW8TnxnP0x8NQhXrncS5jjqzU7kmq1B6v3J607FnyuuSJyu0ee5m0sYauD0Tsky+TbRi7wqmIiIiIyDgxxtjBwOMiK807Ya+zNnqAVatmTshzD4WloaW9rKQQhnM7ltg3dB+Sw9oojyexfcTjk553tMfH54z3O/PB+0fMD+8Px+gPRRLzxPvja30713sGI8Pmh/eHoqMsazY8RFoWR4TWePVww9D62fGeYZdr6H7JvdDJlcwBu7L4KI83xhCzhnrvw06Ru5BTTX7cTxw8+WeAYXPK44E3ft3jzC8fc23xpJ7t411//GgUTkVEREREJCEemJxbqWyKJIn3kg+r3B4ZsU75iCXQItEY4dhQJff49e27dlFdM92+Tyzp/jH7OZPvG4nFxgzyyXPbj3f98aNROBUREREREZnkTmXZs5HWRvZPWM/7sZjPj73PNfYuERERERERkdND4VRERERERERSTuFUREREREREUk7hVERERERERFJO4VRERERERERSTuFUREREREREUk7hVERERERERFJO4VRERERERERSTuFUREREREREUk7hVERERERERFJO4VRERERERERSTuFUREREREREUk7hVERERERERFJO4VRERERERERSTuFUREREREREUk7hVERERERERFJO4VRERERERERSTuFUREREREREUk7hVERERERERFJO4VRERERERERSTuFUREREREREUk7hVERERERERFJO4VRERERERERSzliWleo2JBhjuoEdqW6HTFkFQEuqGyFTlo4fOVk6duRU6PiRk6VjR05FKo+fasuyCkfb4TndLTmGHZZlLUt1I2RqMsa8puNHTpaOHzlZOnbkVOj4kZOlY0dOxWQ9fjSsV0RERERERFJO4VRERERERERSbrKF07tS3QCZ0nT8yKnQ8SMnS8eOnAodP3KydOzIqZiUx8+kKogkIiIiIiIib02TredURERERERE3oImTTg1xlxjjNlhjKk1xnwp1e2Ryc8Ys88Ys8kYs8EY85qzLc8Y8xdjzC7nMjfV7ZTUM8b81BjTZIzZnLRt1GPF2H7gvBdtNMacnbqWy2QwxvHzFWNMvfP+s8EYc13Svi87x88OY8zVqWm1TAbGmEpjzBpjzFZjzBZjzGed7Xr/kaM6yrGj9x45JmNMmjHmFWPMm87xc7uzfZoxZp1znNxrjPE52/3O7Vpnf02q2j4pwqkxxg3cCVwLzAfeb4yZn9pWyRRxqWVZS5JKYX8JeMqyrFnAU85tkZ8D14zYNtaxci0wy/n5BPDD09RGmbx+zpHHD8B/Ou8/SyzLehTA+ey6ETjLecx/O59x8tYUAT5vWdZ84ALgVucY0fuPHMtYxw7ovUeObRC4zLKsxcAS4BpjzAXAt7CPn5lAO3Czc/+bgXZn+38690uJSRFOgfOAWsuy9liWFQJ+C6xOcZtkaloN3O1cvxt4ZwrbIpOEZVnPAm0jNo91rKwGfmHZXgZyjDGlp6elMhmNcfyMZTXwW8uyBi3L2gvUYn/GyVuQZVkNlmWtd653A9uAcvT+I8dwlGNnLHrvkQTnPaTHuel1fizgMuD3zvaR7z3x96TfA5cbY8xpau4wkyWclgN1SbcPcvQ/QBGw/8ieMMa8boz5hLOt2LKsBuf6YaA4NU2TKWCsY0XvR3K8PuMMvfxp0hQCHT8yKmeY3FJgHXr/kRMw4tgBvffIcTDGuI0xG4Am4C/AbqDDsqyIc5fkYyRx/Dj7O4H809ti22QJpyIn4yLLss7GHgZ1qzHmkuSdll2KWuWo5Zh0rMhJ+CEwA3u4VAPwH6ltjkxmxpgM4A/A31uW1ZW8T+8/cjSjHDt675HjYllW1LKsJUAFdi/63BQ36bhMlnBaD1Qm3a5wtomMybKseueyCXgA+w+vMT4EyrlsSl0LZZIb61jR+5Eck2VZjc4Hfwz4MUPD53T8yDDGGC92uLjHsqz7nc16/5FjGu3Y0XuPnCjLsjqANcBy7KkCHmdX8jGSOH6c/dlA62luKjB5wumrwCyngpQPe0L3Qyluk0xixph0Y0xm/DpwFbAZ+7j5sHO3DwN/TE0LZQoY61h5CPiQUzXzAqAzafidCJAIFHE3YL//gH383OhUPpyGXdjmldPdPpkcnDlb/wtssyzre0m79P4jRzXWsaP3HjkexphCY0yOcz0AXIk9b3kN8B7nbiPfe+LvSe8BnnZGdZx2nmPfZeJZlhUxxnwGeBxwAz+1LGtLipslk1sx8IAzV9sD/NqyrMeMMa8C9xljbgb2A+9NYRtlkjDG/AZYBRQYYw4C/wp8k9GPlUeB67CLSfQBHz3tDZZJZYzjZ5UxZgn2cMx9wC0AlmVtMcbcB2zFrrZ5q2VZ0VS0WyaFFcAHgU3O3C+A29D7jxzbWMfO+/XeI8ehFLjbqdjsAu6zLOtPxpitwG+NMf8OvIF9AgTn8pfGmFrsAoA3pqLRACZFoVhEREREREQkYbIM6xUREREREZG3MIVTERERERERSTmFUxEREREREUk5hVMRERERERFJOYVTERERERERSTmFUxERkZNkjIkaYzYk/XxpHJ+7xhiz+dj3FBEROTNMinVORUREpqh+y7KWpLoRIiIiZwL1nIqIiIwzY8w+Y8y3jTGbjDGvGGNmOttrjDFPG2M2GmOeMsZUOduLjTEPGGPedH4udJ7KbYz5sTFmizHmCWNMwLn/3xljtjrP89sU/ZoiIiLjSuFURETk5AVGDOt9X9K+TsuyFgJ3AP/lbPt/wN2WZS0C7gF+4Gz/AfCMZVmLgbOBLc72WcCdlmWdBXQA73a2fwlY6jzPJyfqlxMRETmdjGVZqW6DiIjIlGSM6bEsK2OU7fuAyyzL2mOM8QKHLcvKN8a0AKWWZYWd7Q2WZRUYY5qBCsuyBpOeowb4i2VZs5zb/wR4Lcv6d2PMY0AP8CDwoGVZPRP8q4qIiEw49ZyKiIhMDGuM6ydiMOl6lKFaEW8D7sTuZX3VGKMaEiIiMuUpnIqIiEyM9yVdvuRcfxG40bn+AeA55/pTwKcAjDFuY0z2WE9qjHEBlZZlrQH+CcgGjui9FRERmWp0plVEROTkBYwxG5JuP2ZZVnw5mVxjzEbs3s/3O9v+FviZMeYLQDPwUWf7Z4G7jDE3Y/eQfgpoGOM13cCvnABrgB9YltUxbr+RiIhIimjOqYiIyDhz5pwusyyrJdVtERERmSo0rFdERERERERSTj2nIiIiIiIiknLqORUREREREZGUUzgVERERERGRlFM4FRERERERkZRTOBUREREREZGUUzgVERERERGRlFM4FRERERERkZT7/0t4NINoiPeCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1152x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plotter.plot({'Early Stopping': early_history}, metric = \"mae\")\n",
    "plt.ylabel('MAE [domain1_score]')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'MSE [domain1_score^2]')"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plotter.plot({'Early Stopping': early_history}, metric = \"mse\")\n",
    "plt.ylabel('MSE [domain1_score^2]')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Falling back from v2 loop because of error: Failed to find data adapter that can handle input: <class 'pandas.core.frame.DataFrame'>, <class 'NoneType'>\n",
      "2595/2595 - 0s - loss: 5.8124 - mae: 1.4965 - mse: 5.8124\n",
      "Testing set Mean Abs Error:  1.50 domain1_score\n"
     ]
    }
   ],
   "source": [
    "loss, mae, mse = model.evaluate(normed_test_data, test_labels, verbose=2)\n",
    "\n",
    "print(\"Testing set Mean Abs Error: {:5.2f} domain1_score\".format(mae))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Falling back from v2 loop because of error: Failed to find data adapter that can handle input: <class 'pandas.core.frame.DataFrame'>, <class 'NoneType'>\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_predictions = model.predict(normed_test_data).flatten()\n",
    "\n",
    "a = plt.axes(aspect='equal')\n",
    "plt.scatter(test_labels, test_predictions)\n",
    "plt.xlabel('True Values [domain1_score]')\n",
    "plt.ylabel('Predictions [domain1_score]')\n",
    "lims = [0, 60]\n",
    "plt.xlim(lims)\n",
    "plt.ylim(lims)\n",
    "_ = plt.plot(lims, lims)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">You should evaluate your system’s performance overall and for each subset of test essays using quadratic weighted kappa (https://www.kaggle.com/c/asap-aes/overview/evaluation).\n",
    "\n",
    "A weighted Kappa cab be used to calculate the similarity between predicted and actual score. A perfect score of close to 1.0 is granted when both the predictions and actuals are the same.\n",
    "Whereas, the least possible score is -1 which is given when the predictions are furthest away from actuals. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9639417746039599"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = cohen_kappa_score(test_labels.values, test_predictions.astype(int), weights='quadratic')\n",
    "result"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">You should compare the performance of your model to (at least) a baseline that predicts a random class for each test essay."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.016571317210290926"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "random_predictions = np.random.uniform(low=0, high=test_labels.values.max(), size=test_predictions.size)\n",
    "baseline = cohen_kappa_score(test_labels.values, random_predictions.astype(int), weights='quadratic')\n",
    "baseline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model performed 5716.92910932478 better than the baseline (random scoring).\n"
     ]
    }
   ],
   "source": [
    "pct = (result-baseline)/baseline * 100\n",
    "print(\"Model performed {0}% better than the baseline (random scoring).\".format(pct))"
   ]
  }
 ],
 "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": "3.7.5"
  },
  "pycharm": {
   "stem_cell": {
    "cell_type": "raw",
    "metadata": {
     "collapsed": false
    },
    "source": []
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}