Otto_Group_Product_PyCaret_Classification.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "import pandas as pd \n",
    "from pycaret.classification import * \n",
    "from pycaret.datasets import get_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = pd.read_csv('Data/otto_group_Product_train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = df_train.drop(['id'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>feat_1</th>\n",
       "      <th>feat_2</th>\n",
       "      <th>feat_3</th>\n",
       "      <th>feat_4</th>\n",
       "      <th>feat_5</th>\n",
       "      <th>feat_6</th>\n",
       "      <th>feat_7</th>\n",
       "      <th>feat_8</th>\n",
       "      <th>feat_9</th>\n",
       "      <th>feat_10</th>\n",
       "      <th>feat_11</th>\n",
       "      <th>feat_12</th>\n",
       "      <th>feat_13</th>\n",
       "      <th>feat_14</th>\n",
       "      <th>feat_15</th>\n",
       "      <th>feat_16</th>\n",
       "      <th>feat_17</th>\n",
       "      <th>feat_18</th>\n",
       "      <th>feat_19</th>\n",
       "      <th>feat_20</th>\n",
       "      <th>feat_21</th>\n",
       "      <th>feat_22</th>\n",
       "      <th>feat_23</th>\n",
       "      <th>feat_24</th>\n",
       "      <th>feat_25</th>\n",
       "      <th>feat_26</th>\n",
       "      <th>feat_27</th>\n",
       "      <th>feat_28</th>\n",
       "      <th>feat_29</th>\n",
       "      <th>feat_30</th>\n",
       "      <th>feat_31</th>\n",
       "      <th>feat_32</th>\n",
       "      <th>feat_33</th>\n",
       "      <th>feat_34</th>\n",
       "      <th>feat_35</th>\n",
       "      <th>feat_36</th>\n",
       "      <th>feat_37</th>\n",
       "      <th>feat_38</th>\n",
       "      <th>feat_39</th>\n",
       "      <th>feat_40</th>\n",
       "      <th>feat_41</th>\n",
       "      <th>feat_42</th>\n",
       "      <th>feat_43</th>\n",
       "      <th>feat_44</th>\n",
       "      <th>feat_45</th>\n",
       "      <th>feat_46</th>\n",
       "      <th>feat_47</th>\n",
       "      <th>feat_48</th>\n",
       "      <th>feat_49</th>\n",
       "      <th>feat_50</th>\n",
       "      <th>feat_51</th>\n",
       "      <th>feat_52</th>\n",
       "      <th>feat_53</th>\n",
       "      <th>feat_54</th>\n",
       "      <th>feat_55</th>\n",
       "      <th>feat_56</th>\n",
       "      <th>feat_57</th>\n",
       "      <th>feat_58</th>\n",
       "      <th>feat_59</th>\n",
       "      <th>feat_60</th>\n",
       "      <th>feat_61</th>\n",
       "      <th>feat_62</th>\n",
       "      <th>feat_63</th>\n",
       "      <th>feat_64</th>\n",
       "      <th>feat_65</th>\n",
       "      <th>feat_66</th>\n",
       "      <th>feat_67</th>\n",
       "      <th>feat_68</th>\n",
       "      <th>feat_69</th>\n",
       "      <th>feat_70</th>\n",
       "      <th>feat_71</th>\n",
       "      <th>feat_72</th>\n",
       "      <th>feat_73</th>\n",
       "      <th>feat_74</th>\n",
       "      <th>feat_75</th>\n",
       "      <th>feat_76</th>\n",
       "      <th>feat_77</th>\n",
       "      <th>feat_78</th>\n",
       "      <th>feat_79</th>\n",
       "      <th>feat_80</th>\n",
       "      <th>feat_81</th>\n",
       "      <th>feat_82</th>\n",
       "      <th>feat_83</th>\n",
       "      <th>feat_84</th>\n",
       "      <th>feat_85</th>\n",
       "      <th>feat_86</th>\n",
       "      <th>feat_87</th>\n",
       "      <th>feat_88</th>\n",
       "      <th>feat_89</th>\n",
       "      <th>feat_90</th>\n",
       "      <th>feat_91</th>\n",
       "      <th>feat_92</th>\n",
       "      <th>feat_93</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Class_1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Class_1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Class_1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Class_1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Class_1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   feat_1  feat_2  feat_3  feat_4  feat_5  feat_6  feat_7  feat_8  feat_9  \\\n",
       "0       1       0       0       0       0       0       0       0       0   \n",
       "1       0       0       0       0       0       0       0       1       0   \n",
       "2       0       0       0       0       0       0       0       1       0   \n",
       "3       1       0       0       1       6       1       5       0       0   \n",
       "4       0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "   feat_10  feat_11  feat_12  feat_13  feat_14  feat_15  feat_16  feat_17  \\\n",
       "0        0        1        0        0        0        0        0        2   \n",
       "1        0        0        0        0        0        0        0        0   \n",
       "2        0        0        0        0        0        0        0        1   \n",
       "3        1        1        0        1        0        0        1        1   \n",
       "4        0        0        0        0        0        0        0        4   \n",
       "\n",
       "   feat_18  feat_19  feat_20  feat_21  feat_22  feat_23  feat_24  feat_25  \\\n",
       "0        0        0        0        0        1        0        4        1   \n",
       "1        2        0        0        0        0        0        0        0   \n",
       "2        0        0        0        0        0        0        0        0   \n",
       "3        0        0        0        0        0        0        7        2   \n",
       "4        0        0        0        0        0        0        1        0   \n",
       "\n",
       "   feat_26  feat_27  feat_28  feat_29  feat_30  feat_31  feat_32  feat_33  \\\n",
       "0        1        0        0        2        0        0        0        0   \n",
       "1        0        0        0        0        0        0        0        0   \n",
       "2        0        0        0        0        0        0        0        1   \n",
       "3        2        0        0        0       58        0       10        0   \n",
       "4        0        0        0        0        0        0        0        0   \n",
       "\n",
       "   feat_34  feat_35  feat_36  feat_37  feat_38  feat_39  feat_40  feat_41  \\\n",
       "0        0        1        0        0        0        0        1        0   \n",
       "1        0        0        0        1        0        0        0        0   \n",
       "2        0        0        0        0        0        0        0        0   \n",
       "3        0        0        0        0        3        0        0        0   \n",
       "4        0        0        0        0        0        0        0        1   \n",
       "\n",
       "   feat_42  feat_43  feat_44  feat_45  feat_46  feat_47  feat_48  feat_49  \\\n",
       "0        5        0        0        0        0        0        2        0   \n",
       "1        0        0        0        0        0        0        0        0   \n",
       "2        0        0        0        0        0        0        1        0   \n",
       "3        0        0        2        0        2        0        1        2   \n",
       "4        0        0        0        0        0        0        0        0   \n",
       "\n",
       "   feat_50  feat_51  feat_52  feat_53  feat_54  feat_55  feat_56  feat_57  \\\n",
       "0        0        0        0        0        1        0        0        2   \n",
       "1        0        0        0        0        0        0        0        0   \n",
       "2        0        0        0        0        0        1        0        0   \n",
       "3        1        3        0        0        3        1        0        0   \n",
       "4        2        0        0        0        0        0        0        0   \n",
       "\n",
       "   feat_58  feat_59  feat_60  feat_61  feat_62  feat_63  feat_64  feat_65  \\\n",
       "0        0        0       11        0        1        1        0        1   \n",
       "1        1        0        0        0        0        0        1        0   \n",
       "2        0        0        0        0        0        0        0        0   \n",
       "3        0        0        0        0        0        0        0        2   \n",
       "4        0        0        0        0        1        0        0        0   \n",
       "\n",
       "   feat_66  feat_67  feat_68  feat_69  feat_70  feat_71  feat_72  feat_73  \\\n",
       "0        0        7        0        0        0        1        0        0   \n",
       "1        0        1        0        0        0        0        0        2   \n",
       "2        0        6        0        0        2        0        0        0   \n",
       "3        1        5        0        0        4        0        0        2   \n",
       "4        0        0        1        0        0        3        0        0   \n",
       "\n",
       "   feat_74  feat_75  feat_76  feat_77  feat_78  feat_79  feat_80  feat_81  \\\n",
       "0        0        0        0        0        0        2        1        0   \n",
       "1        1        0        1        0        1        0        0        0   \n",
       "2        0        0        0        0        0        0        0        1   \n",
       "3        1        0        1        0        0        1        1        2   \n",
       "4        0        0        0        0        0        4        0        1   \n",
       "\n",
       "   feat_82  feat_83  feat_84  feat_85  feat_86  feat_87  feat_88  feat_89  \\\n",
       "0        0        0        0        1        0        0        0        0   \n",
       "1        0        0        0        0        0        0        0        0   \n",
       "2        0        0        0        0        0        0        0        0   \n",
       "3        2        0       22        0        1        2        0        0   \n",
       "4        0        0        0        1        0        0        0        0   \n",
       "\n",
       "   feat_90  feat_91  feat_92  feat_93   target  \n",
       "0        0        0        0        0  Class_1  \n",
       "1        0        0        0        0  Class_1  \n",
       "2        0        0        0        0  Class_1  \n",
       "3        0        0        0        0  Class_1  \n",
       "4        1        0        0        0  Class_1  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Remove correlated features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "X = df_train.drop(['target'],axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "#### Identify Highly Correlated Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "           feat_1    feat_2    feat_3        feat_4    feat_5    feat_6  \\\n",
      "feat_1   1.000000  0.031332  0.027807  2.752941e-02  0.042973  0.043603   \n",
      "feat_2   0.031332  1.000000  0.082573  1.349870e-01  0.020926  0.041343   \n",
      "feat_3   0.027807  0.082573  1.000000  5.835232e-01  0.010880  0.004288   \n",
      "feat_4   0.027529  0.134987  0.583523  1.000000e+00  0.017290  0.014059   \n",
      "feat_5   0.042973  0.020926  0.010880  1.729026e-02  1.000000  0.145355   \n",
      "feat_6   0.043603  0.041343  0.004288  1.405895e-02  0.145355  1.000000   \n",
      "feat_7   0.298952  0.222386  0.001294  1.448981e-02  0.075047  0.088014   \n",
      "feat_8   0.056321  0.019815  0.053462  4.618407e-02  0.035861  0.012867   \n",
      "feat_9   0.032285  0.025630  0.063551  4.624977e-02  0.024708  0.009373   \n",
      "feat_10  0.097776  0.051925  0.036944  5.951396e-02  0.091324  0.041940   \n",
      "feat_11  0.042928  0.118534  0.596243  3.894092e-01  0.004882  0.014504   \n",
      "feat_12  0.056934  0.090153  0.050037  5.743356e-02  0.036668  0.028588   \n",
      "feat_13  0.139254  0.157467  0.013870  2.897317e-02  0.059081  0.036293   \n",
      "feat_14  0.063517  0.070057  0.111105  9.921490e-02  0.037607  0.027350   \n",
      "feat_15  0.045738  0.048798  0.065285  5.122155e-02  0.007000  0.018328   \n",
      "feat_16  0.027086  0.108046  0.221426  2.110780e-01  0.062877  0.021934   \n",
      "feat_17  0.053004  0.074902  0.023093  7.553867e-03  0.062197  0.015488   \n",
      "feat_18  0.084856  0.242716  0.115655  2.148952e-01  0.052186  0.048710   \n",
      "feat_19  0.002302  0.176655  0.012228  3.519107e-07  0.008556  0.038493   \n",
      "feat_20  0.070511  0.449160  0.011069  4.465657e-02  0.046200  0.057813   \n",
      "feat_21  0.027026  0.014113  0.354925  2.329227e-01  0.003288  0.008046   \n",
      "feat_22  0.063283  0.215106  0.251082  2.477378e-01  0.075161  0.038939   \n",
      "feat_23  0.048686  0.162065  0.002427  3.062225e-02  0.017281  0.043651   \n",
      "feat_24  0.067255  0.253684  0.031596  3.727726e-03  0.075222  0.082124   \n",
      "feat_25  0.187237  0.096366  0.157459  1.342306e-01  0.003610  0.023319   \n",
      "feat_26  0.022813  0.064856  0.268112  3.657567e-01  0.025116  0.004680   \n",
      "feat_27  0.038826  0.037841  0.508370  3.086287e-01  0.002098  0.001943   \n",
      "feat_28  0.030257  0.072494  0.551398  4.864171e-01  0.047688  0.017132   \n",
      "feat_29  0.069266  0.025689  0.004141  1.427066e-02  0.065957  0.002389   \n",
      "feat_30  0.033108  0.026896  0.007667  8.733991e-04  0.318117  0.196493   \n",
      "feat_31  0.011210  0.193216  0.138548  3.512627e-01  0.013188  0.017112   \n",
      "feat_32  0.061361  0.087699  0.053753  3.673037e-02  0.122957  0.103699   \n",
      "feat_33  0.049454  0.033927  0.078520  5.481916e-02  0.006046  0.019617   \n",
      "feat_34  0.042756  0.026717  0.039765  2.908729e-02  0.001032  0.010199   \n",
      "feat_35  0.010667  0.293374  0.233923  5.544741e-01  0.027013  0.015739   \n",
      "feat_36  0.095475  0.026988  0.034305  2.965968e-02  0.043504  0.014673   \n",
      "feat_37  0.082306  0.124475  0.029425  1.064590e-02  0.035320  0.043213   \n",
      "feat_38  0.104666  0.373022  0.046097  9.248764e-02  0.070100  0.078088   \n",
      "feat_39  0.006898  0.020626  0.016279  2.699561e-03  0.000710  0.033470   \n",
      "feat_40  0.053603  0.057890  0.094511  7.680452e-02  0.025652  0.018090   \n",
      "feat_41  0.080960  0.072120  0.028920  9.976832e-03  0.038764  0.028930   \n",
      "feat_42  0.073611  0.043863  0.036605  7.139722e-02  0.023272  0.020709   \n",
      "feat_43  0.008884  0.017600  0.014068  8.930611e-04  0.048320  0.015515   \n",
      "feat_44  0.090439  0.208450  0.242901  2.386655e-01  0.177137  0.065082   \n",
      "feat_45  0.008601  0.018259  0.007984  2.809149e-03  0.004808  0.025133   \n",
      "feat_46  0.040647  0.042491  0.777517  4.362735e-01  0.003583  0.001892   \n",
      "feat_47  0.026463  0.014574  0.009560  1.291450e-02  0.111428  0.018736   \n",
      "feat_48  0.130564  0.023310  0.111987  8.578789e-02  0.013228  0.013193   \n",
      "feat_49  0.011179  0.311278  0.242692  3.000109e-01  0.054311  0.022864   \n",
      "feat_50  0.040114  0.071664  0.016765  5.343432e-03  0.024087  0.027150   \n",
      "feat_51  0.000902  0.085345  0.031979  4.164659e-02  0.060318  0.006902   \n",
      "feat_52  0.013082  0.105528  0.169111  2.095170e-01  0.003930  0.030929   \n",
      "feat_53  0.156400  0.008015  0.024222  1.630815e-02  0.040202  0.005914   \n",
      "feat_54  0.001399  0.179173  0.694048  5.254557e-01  0.055299  0.031379   \n",
      "feat_55  0.165044  0.134617  0.055445  7.007713e-02  0.142221  0.062765   \n",
      "feat_56  0.015738  0.018908  0.016288  1.961792e-02  0.001560  0.004014   \n",
      "feat_57  0.014598  0.104263  0.018887  6.946199e-02  0.037162  0.036655   \n",
      "feat_58  0.028696  0.371146  0.034506  3.910286e-02  0.001546  0.110743   \n",
      "feat_59  0.139364  0.013283  0.021717  1.580168e-02  0.038893  0.001285   \n",
      "feat_60  0.020267  0.018219  0.011369  6.890047e-02  0.008796  0.022289   \n",
      "feat_61  0.038124  0.063873  0.494881  3.544023e-01  0.001559  0.002227   \n",
      "feat_62  0.091636  0.029040  0.010733  2.522773e-02  0.073215  0.022874   \n",
      "feat_63  0.069799  0.037020  0.031241  5.990274e-02  0.082383  0.017050   \n",
      "feat_64  0.010499  0.005354  0.065105  4.728813e-02  0.021017  0.002764   \n",
      "feat_65  0.110041  0.078801  0.065492  6.228472e-02  0.228349  0.066867   \n",
      "feat_66  0.053010  0.175620  0.088017  1.296545e-01  0.048364  0.033285   \n",
      "feat_67  0.154301  0.068667  0.110081  8.045694e-02  0.061964  0.038289   \n",
      "feat_68  0.014674  0.012802  0.030992  2.009191e-02  0.107405  0.021619   \n",
      "feat_69  0.007544  0.307406  0.032748  1.446082e-02  0.003294  0.074836   \n",
      "feat_70  0.165442  0.112968  0.018774  2.079779e-02  0.118510  0.052401   \n",
      "feat_71  0.013712  0.002336  0.053020  4.241268e-02  0.056428  0.011901   \n",
      "feat_72  0.029983  0.023267  0.045339  2.979578e-02  0.005177  0.011090   \n",
      "feat_73  0.140815  0.039192  0.013972  1.128547e-02  0.001609  0.025023   \n",
      "feat_74  0.051365  0.070724  0.041559  4.909735e-02  0.017265  0.043160   \n",
      "feat_75  0.011596  0.093689  0.044724  3.145389e-02  0.015279  0.006951   \n",
      "feat_76  0.153808  0.259360  0.028670  1.379188e-02  0.035570  0.073867   \n",
      "feat_77  0.123752  0.014911  0.001584  1.531773e-02  0.030462  0.006501   \n",
      "feat_78  0.279202  0.094256  0.021979  1.449856e-02  0.070709  0.061250   \n",
      "feat_79  0.228912  0.033668  0.020566  1.083473e-02  0.055115  0.009942   \n",
      "feat_80  0.013303  0.155768  0.442036  4.057725e-01  0.026223  0.017648   \n",
      "feat_81  0.032427  0.052101  0.013089  2.828377e-02  0.129333  0.044136   \n",
      "feat_82  0.026085  0.119109  0.438458  4.365413e-01  0.057400  0.014907   \n",
      "feat_83  0.059165  0.371691  0.019914  1.051874e-03  0.008006  0.035145   \n",
      "feat_84  0.049634  0.009845  0.011159  5.684499e-03  0.467329  0.177777   \n",
      "feat_85  0.008739  0.006764  0.048626  3.315343e-02  0.034062  0.004290   \n",
      "feat_86  0.107947  0.039090  0.096093  7.102916e-02  0.013879  0.010455   \n",
      "feat_87  0.089374  0.047451  0.009838  5.054728e-03  0.013999  0.015256   \n",
      "feat_88  0.020830  0.047035  0.082336  6.748367e-02  0.019201  0.015437   \n",
      "feat_89  0.096851  0.105527  0.174781  1.837145e-01  0.119951  0.035042   \n",
      "feat_90  0.010310  0.515022  0.015068  9.454061e-03  0.004842  0.054034   \n",
      "feat_91  0.037264  0.026383  0.012417  1.031241e-02  0.012012  0.012465   \n",
      "feat_92  0.054777  0.008219  0.066921  8.763105e-02  0.065331  0.015479   \n",
      "feat_93  0.081783  0.054593  0.006814  1.574563e-02  0.002038  0.008521   \n",
      "\n",
      "           feat_7    feat_8    feat_9   feat_10   feat_11   feat_12   feat_13  \\\n",
      "feat_1   0.298952  0.056321  0.032285  0.097776  0.042928  0.056934  0.139254   \n",
      "feat_2   0.222386  0.019815  0.025630  0.051925  0.118534  0.090153  0.157467   \n",
      "feat_3   0.001294  0.053462  0.063551  0.036944  0.596243  0.050037  0.013870   \n",
      "feat_4   0.014490  0.046184  0.046250  0.059514  0.389409  0.057434  0.028973   \n",
      "feat_5   0.075047  0.035861  0.024708  0.091324  0.004882  0.036668  0.059081   \n",
      "feat_6   0.088014  0.012867  0.009373  0.041940  0.014504  0.028588  0.036293   \n",
      "feat_7   1.000000  0.038121  0.027146  0.194258  0.012418  0.056230  0.199142   \n",
      "feat_8   0.038121  1.000000  0.039281  0.000023  0.065923  0.091424  0.095365   \n",
      "feat_9   0.027146  0.039281  1.000000  0.024323  0.075820  0.021885  0.040164   \n",
      "feat_10  0.194258  0.000023  0.024323  1.000000  0.006010  0.048969  0.086682   \n",
      "feat_11  0.012418  0.065923  0.075820  0.006010  1.000000  0.053391  0.005500   \n",
      "feat_12  0.056230  0.091424  0.021885  0.048969  0.053391  1.000000  0.115288   \n",
      "feat_13  0.199142  0.095365  0.040164  0.086682  0.005500  0.115288  1.000000   \n",
      "feat_14  0.044671  0.061799  0.110188  0.029598  0.151635  0.021695  0.021555   \n",
      "feat_15  0.035721  0.056960  0.009858  0.021700  0.100255  0.028407  0.032725   \n",
      "feat_16  0.043957  0.004659  0.082664  0.063997  0.134924  0.113804  0.087992   \n",
      "feat_17  0.127245  0.173912  0.028709  0.092959  0.031807  0.031434  0.319352   \n",
      "feat_18  0.098972  0.087777  0.043642  0.071635  0.079620  0.100066  0.177439   \n",
      "feat_19  0.058071  0.019387  0.000167  0.009015  0.033679  0.034748  0.042954   \n",
      "feat_20  0.364972  0.062595  0.023397  0.176373  0.005641  0.071224  0.194816   \n",
      "feat_21  0.022908  0.041095  0.028409  0.005134  0.324080  0.048215  0.005092   \n",
      "feat_22  0.162620  0.029032  0.062348  0.141405  0.154861  0.119511  0.209427   \n",
      "feat_23  0.186462  0.012774  0.006940  0.096666  0.026500  0.070561  0.056463   \n",
      "feat_24  0.244813  0.161848  0.073618  0.081684  0.064213  0.085558  0.192249   \n",
      "feat_25  0.048820  0.036939  0.025279  0.009792  0.206207  0.012281  0.006019   \n",
      "feat_26  0.008782  0.041599  0.066414  0.003721  0.191095  0.042363  0.016335   \n",
      "feat_27  0.015429  0.050272  0.042531  0.001551  0.599484  0.041377  0.012657   \n",
      "feat_28  0.000998  0.036668  0.055545  0.022349  0.422271  0.053406  0.015961   \n",
      "feat_29  0.046231  0.104985  0.021328  0.068243  0.021024  0.063105  0.177028   \n",
      "feat_30  0.050535  0.009574  0.015830  0.012623  0.008832  0.013299  0.042503   \n",
      "feat_31  0.032291  0.018342  0.016227  0.038503  0.109761  0.056052  0.024556   \n",
      "feat_32  0.149537  0.017910  0.568417  0.028812  0.002610  0.017584  0.072300   \n",
      "feat_33  0.008628  0.045822  0.045937  0.003292  0.130168  0.048145  0.048620   \n",
      "feat_34  0.022144  0.015314  0.050045  0.030941  0.068741  0.018258  0.004043   \n",
      "feat_35  0.043001  0.024043  0.026879  0.051793  0.108392  0.077834  0.050881   \n",
      "feat_36  0.079760  0.606707  0.035874  0.007687  0.047595  0.063250  0.128211   \n",
      "feat_37  0.141960  0.119417  0.024868  0.187069  0.011404  0.080605  0.077847   \n",
      "feat_38  0.315177  0.034287  0.028766  0.177524  0.055483  0.097602  0.191271   \n",
      "feat_39  0.053230  0.071908  0.013354  0.042223  0.013556  0.008417  0.015840   \n",
      "feat_40  0.039712  0.044847  0.085491  0.038610  0.133424  0.016455  0.036971   \n",
      "feat_41  0.097971  0.075854  0.023446  0.078533  0.038465  0.064854  0.123191   \n",
      "feat_42  0.001220  0.006578  0.055850  0.024055  0.002356  0.028966  0.036274   \n",
      "feat_43  0.010259  0.004214  0.046033  0.111706  0.027301  0.007790  0.026543   \n",
      "feat_44  0.213066  0.001733  0.062023  0.222353  0.172523  0.091486  0.162659   \n",
      "feat_45  0.036262  0.076982  0.013405  0.009960  0.006148  0.004838  0.011546   \n",
      "feat_46  0.012643  0.050246  0.053175  0.013546  0.579272  0.042124  0.008579   \n",
      "feat_47  0.177082  0.015016  0.027254  0.046108  0.004880  0.018930  0.083185   \n",
      "feat_48  0.056163  0.104416  0.077385  0.008297  0.125103  0.035591  0.089556   \n",
      "feat_49  0.124320  0.019114  0.042280  0.079449  0.250991  0.073520  0.075534   \n",
      "feat_50  0.140433  0.001231  0.009925  0.252350  0.019520  0.041978  0.043774   \n",
      "feat_51  0.036493  0.009513  0.013181  0.014012  0.040465  0.016439  0.021930   \n",
      "feat_52  0.006189  0.028689  0.009792  0.007879  0.272738  0.068520  0.009553   \n",
      "feat_53  0.048544  0.018674  0.030075  0.082810  0.052811  0.018344  0.161234   \n",
      "feat_54  0.079607  0.025421  0.078499  0.074637  0.494180  0.102486  0.119941   \n",
      "feat_55  0.343752  0.063415  0.042138  0.213003  0.022723  0.081504  0.223364   \n",
      "feat_56  0.003145  0.002285  0.022639  0.009786  0.033221  0.007665  0.002066   \n",
      "feat_57  0.073969  0.011968  0.049335  0.017525  0.007302  0.042518  0.075397   \n",
      "feat_58  0.106203  0.025674  0.028290  0.020933  0.179361  0.082574  0.132569   \n",
      "feat_59  0.009624  0.118223  0.021451  0.051189  0.039646  0.026684  0.174173   \n",
      "feat_60  0.012336  0.020328  0.066333  0.000619  0.025990  0.027027  0.046147   \n",
      "feat_61  0.006284  0.045203  0.044343  0.004397  0.543858  0.050080  0.009951   \n",
      "feat_62  0.025485  0.214717  0.021138  0.054733  0.051554  0.126326  0.164920   \n",
      "feat_63  0.039585  0.148673  0.013012  0.090421  0.013436  0.038777  0.070109   \n",
      "feat_64  0.011165  0.003194  0.702951  0.022536  0.082868  0.012603  0.003286   \n",
      "feat_65  0.202346  0.025544  0.038163  0.182756  0.041014  0.121550  0.108567   \n",
      "feat_66  0.122660  0.115175  0.001778  0.100722  0.050764  0.084591  0.122272   \n",
      "feat_67  0.148598  0.320949  0.176921  0.043117  0.114944  0.079377  0.211335   \n",
      "feat_68  0.040309  0.075384  0.012192  0.001693  0.040588  0.014873  0.026881   \n",
      "feat_69  0.131430  0.046258  0.029335  0.077354  0.010697  0.057074  0.086611   \n",
      "feat_70  0.237907  0.023089  0.056205  0.322857  0.050540  0.071630  0.145294   \n",
      "feat_71  0.115813  0.081664  0.043286  0.104834  0.071717  0.028629  0.052337   \n",
      "feat_72  0.014921  0.029868  0.058147  0.004225  0.086496  0.031888  0.008543   \n",
      "feat_73  0.022819  0.028999  0.022679  0.000240  0.006893  0.181833  0.030927   \n",
      "feat_74  0.053059  0.000431  0.007594  0.008912  0.186561  0.087012  0.021072   \n",
      "feat_75  0.039865  0.031466  0.027313  0.003828  0.034513  0.052536  0.022496   \n",
      "feat_76  0.375114  0.081682  0.027424  0.106752  0.021357  0.113921  0.189751   \n",
      "feat_77  0.005769  0.027486  0.020185  0.019069  0.008064  0.021418  0.178263   \n",
      "feat_78  0.567084  0.079623  0.015922  0.091760  0.009866  0.026525  0.224472   \n",
      "feat_79  0.066753  0.083714  0.036116  0.113659  0.024491  0.036666  0.233604   \n",
      "feat_80  0.028860  0.038382  0.046721  0.019042  0.561843  0.064162  0.010431   \n",
      "feat_81  0.144308  0.035102  0.005847  0.135928  0.004795  0.052954  0.065423   \n",
      "feat_82  0.022059  0.034409  0.039806  0.029741  0.420361  0.054571  0.010074   \n",
      "feat_83  0.282069  0.033479  0.032875  0.052025  0.000190  0.036529  0.115747   \n",
      "feat_84  0.062634  0.005064  0.013569  0.017939  0.017724  0.009807  0.023221   \n",
      "feat_85  0.037874  0.003416  0.031462  0.086758  0.074293  0.019283  0.002594   \n",
      "feat_86  0.009169  0.029395  0.019144  0.159447  0.123339  0.007214  0.004850   \n",
      "feat_87  0.089574  0.059929  0.016925  0.077421  0.032969  0.016089  0.093870   \n",
      "feat_88  0.033646  0.050931  0.001160  0.054635  0.114491  0.024324  0.036259   \n",
      "feat_89  0.063511  0.007974  0.019147  0.061498  0.137374  0.082220  0.062990   \n",
      "feat_90  0.129578  0.026807  0.020698  0.049908  0.045074  0.062721  0.107722   \n",
      "feat_91  0.068506  0.095990  0.014742  0.024025  0.029511  0.063965  0.044338   \n",
      "feat_92  0.032261  0.013608  0.069707  0.006869  0.013179  0.063922  0.071953   \n",
      "feat_93  0.034912  0.005131  0.006038  0.041316  0.003326  0.012722  0.038989   \n",
      "\n",
      "          feat_14   feat_15   feat_16   feat_17   feat_18       feat_19  \\\n",
      "feat_1   0.063517  0.045738  0.027086  0.053004  0.084856  2.302499e-03   \n",
      "feat_2   0.070057  0.048798  0.108046  0.074902  0.242716  1.766549e-01   \n",
      "feat_3   0.111105  0.065285  0.221426  0.023093  0.115655  1.222845e-02   \n",
      "feat_4   0.099215  0.051222  0.211078  0.007554  0.214895  3.519107e-07   \n",
      "feat_5   0.037607  0.007000  0.062877  0.062197  0.052186  8.555966e-03   \n",
      "feat_6   0.027350  0.018328  0.021934  0.015488  0.048710  3.849262e-02   \n",
      "feat_7   0.044671  0.035721  0.043957  0.127245  0.098972  5.807104e-02   \n",
      "feat_8   0.061799  0.056960  0.004659  0.173912  0.087777  1.938742e-02   \n",
      "feat_9   0.110188  0.009858  0.082664  0.028709  0.043642  1.671680e-04   \n",
      "feat_10  0.029598  0.021700  0.063997  0.092959  0.071635  9.015307e-03   \n",
      "feat_11  0.151635  0.100255  0.134924  0.031807  0.079620  3.367902e-02   \n",
      "feat_12  0.021695  0.028407  0.113804  0.031434  0.100066  3.474800e-02   \n",
      "feat_13  0.021555  0.032725  0.087992  0.319352  0.177439  4.295357e-02   \n",
      "feat_14  1.000000  0.039565  0.018640  0.040148  0.076035  4.274776e-02   \n",
      "feat_15  0.039565  1.000000  0.344157  0.045991  0.044733  3.202887e-02   \n",
      "feat_16  0.018640  0.344157  1.000000  0.014567  0.244849  1.011014e-03   \n",
      "feat_17  0.040148  0.045991  0.014567  1.000000  0.154259  3.731379e-02   \n",
      "feat_18  0.076035  0.044733  0.244849  0.154259  1.000000  3.652979e-02   \n",
      "feat_19  0.042748  0.032029  0.001011  0.037314  0.036530  1.000000e+00   \n",
      "feat_20  0.060775  0.034214  0.088638  0.261954  0.204010  1.054279e-01   \n",
      "feat_21  0.005645  0.086686  0.199338  0.040070  0.043272  2.412224e-02   \n",
      "feat_22  0.063087  0.056194  0.433154  0.157098  0.364125  1.908832e-02   \n",
      "feat_23  0.046284  0.029100  0.037233  0.133157  0.098466  1.000276e-01   \n",
      "feat_24  0.118920  0.062806  0.016491  0.257106  0.134752  2.904751e-01   \n",
      "feat_25  0.557929  0.008228  0.039487  0.034996  0.063139  6.511212e-02   \n",
      "feat_26  0.101571  0.056035  0.187627  0.004430  0.234748  2.116921e-02   \n",
      "feat_27  0.093146  0.045859  0.118580  0.036484  0.043358  1.895432e-02   \n",
      "feat_28  0.097257  0.044554  0.235443  0.020020  0.163886  1.885817e-02   \n",
      "feat_29  0.027883  0.022266  0.142485  0.135787  0.118086  1.078641e-02   \n",
      "feat_30  0.028994  0.007801  0.025419  0.023838  0.018357  6.245008e-03   \n",
      "feat_31  0.052896  0.029179  0.125445  0.028863  0.234090  3.255077e-02   \n",
      "feat_32  0.141330  0.029320  0.025200  0.100666  0.018512  1.155059e-01   \n",
      "feat_33  0.258664  0.273124  0.239344  0.049150  0.033070  4.092405e-02   \n",
      "feat_34  0.094934  0.049157  0.047032  0.009384  0.020361  2.231114e-02   \n",
      "feat_35  0.063796  0.033329  0.208736  0.026671  0.380595  4.061698e-02   \n",
      "feat_36  0.042550  0.058118  0.012829  0.199284  0.076773  1.099318e-02   \n",
      "feat_37  0.038171  0.040783  0.040773  0.079853  0.062521  6.762914e-02   \n",
      "feat_38  0.061308  0.001730  0.193696  0.136263  0.248949  2.468113e-01   \n",
      "feat_39  0.001582  0.003620  0.043227  0.061983  0.058047  6.468167e-03   \n",
      "feat_40  0.487804  0.069358  0.287290  0.071674  0.024125  4.258767e-02   \n",
      "feat_41  0.029650  0.026326  0.079613  0.153653  0.104176  2.565826e-02   \n",
      "feat_42  0.054232  0.045392  0.142924  0.045206  0.229794  2.476852e-02   \n",
      "feat_43  0.073237  0.010929  0.046106  0.018660  0.007073  1.446902e-02   \n",
      "feat_44  0.061226  0.043123  0.353285  0.085796  0.265708  4.202058e-02   \n",
      "feat_45  0.015501  0.018245  0.041837  0.028681  0.038576  6.415927e-03   \n",
      "feat_46  0.099296  0.054633  0.179348  0.033466  0.059749  1.880024e-02   \n",
      "feat_47  0.059736  0.043214  0.038439  0.017601  0.072757  1.206439e-02   \n",
      "feat_48  0.252006  0.051239  0.030415  0.102710  0.019632  3.839369e-02   \n",
      "feat_49  0.070134  0.040979  0.227799  0.049887  0.346609  1.960682e-02   \n",
      "feat_50  0.014922  0.020619  0.058954  0.081685  0.074758  5.629922e-02   \n",
      "feat_51  0.041262  0.024504  0.024675  0.076780  0.114754  4.867697e-03   \n",
      "feat_52  0.060335  0.022071  0.093180  0.004599  0.107743  1.712624e-02   \n",
      "feat_53  0.006808  0.006163  0.086820  0.018729  0.042412  1.375631e-02   \n",
      "feat_54  0.112305  0.034175  0.468766  0.033721  0.272898  2.603274e-04   \n",
      "feat_55  0.055732  0.040297  0.131218  0.163377  0.183218  1.804592e-02   \n",
      "feat_56  0.004436  0.022519  0.021978  0.012085  0.014345  8.823711e-03   \n",
      "feat_57  0.101382  0.069910  0.049373  0.173169  0.284435  2.619428e-02   \n",
      "feat_58  0.059363  0.044920  0.018927  0.040370  0.048980  2.345026e-01   \n",
      "feat_59  0.026693  0.017773  0.046633  0.058797  0.099073  8.377876e-03   \n",
      "feat_60  0.118618  0.079152  0.043111  0.090963  0.234885  2.039351e-03   \n",
      "feat_61  0.093809  0.056169  0.150952  0.030037  0.084745  1.463294e-02   \n",
      "feat_62  0.026822  0.097920  0.337177  0.105504  0.197339  1.471355e-02   \n",
      "feat_63  0.020837  0.018947  0.104440  0.032391  0.119569  1.220845e-02   \n",
      "feat_64  0.066800  0.003848  0.043076  0.021551  0.001922  1.698588e-02   \n",
      "feat_65  0.011517  0.034954  0.182735  0.072733  0.114910  1.579718e-02   \n",
      "feat_66  0.081709  0.018487  0.289082  0.165143  0.285667  3.078749e-02   \n",
      "feat_67  0.086404  0.052486  0.027728  0.306513  0.108521  3.587552e-02   \n",
      "feat_68  0.076979  0.049076  0.015351  0.038816  0.078124  5.128506e-03   \n",
      "feat_69  0.058352  0.052924  0.021699  0.100656  0.071926  3.976207e-01   \n",
      "feat_70  0.054226  0.118795  0.255740  0.108789  0.169238  3.117130e-02   \n",
      "feat_71  0.076935  0.033807  0.001594  0.150286  0.074981  2.087111e-02   \n",
      "feat_72  0.022358  0.764664  0.394998  0.023103  0.002190  2.258364e-02   \n",
      "feat_73  0.040175  0.019194  0.026285  0.002826  0.008668  5.683692e-02   \n",
      "feat_74  0.033498  0.013632  0.034125  0.001770  0.038278  2.261396e-02   \n",
      "feat_75  0.079099  0.048894  0.059511  0.025506  0.011790  3.635667e-02   \n",
      "feat_76  0.058841  0.061486  0.015473  0.134841  0.059279  1.795475e-01   \n",
      "feat_77  0.007714  0.016673  0.129884  0.099665  0.102684  7.950547e-03   \n",
      "feat_78  0.041530  0.027200  0.003965  0.069539  0.042958  6.203551e-03   \n",
      "feat_79  0.045473  0.039033  0.045703  0.114670  0.080390  1.249856e-02   \n",
      "feat_80  0.104507  0.057948  0.205393  0.015193  0.182025  2.391456e-03   \n",
      "feat_81  0.037389  0.019557  0.076330  0.123831  0.135914  8.124693e-04   \n",
      "feat_82  0.084089  0.050702  0.168008  0.012734  0.156176  6.879326e-03   \n",
      "feat_83  0.070643  0.052090  0.017558  0.113987  0.085116  5.672093e-02   \n",
      "feat_84  0.027058  0.009311  0.035211  0.000334  0.028752  2.846741e-03   \n",
      "feat_85  0.021455  0.246847  0.110850  0.015559  0.001555  8.292391e-03   \n",
      "feat_86  0.145787  0.002529  0.003610  0.049102  0.029295  1.455986e-02   \n",
      "feat_87  0.020229  0.023191  0.077770  0.214221  0.126886  4.116324e-04   \n",
      "feat_88  0.323089  0.010840  0.007257  0.034139  0.035981  1.848491e-02   \n",
      "feat_89  0.038881  0.029547  0.248364  0.035390  0.247462  1.111595e-02   \n",
      "feat_90  0.060240  0.046616  0.016863  0.045218  0.094336  4.509254e-01   \n",
      "feat_91  0.038444  0.034402  0.048494  0.088508  0.037275  4.085393e-03   \n",
      "feat_92  0.040133  0.018206  0.210499  0.006538  0.126640  2.766153e-02   \n",
      "feat_93  0.018127  0.020369  0.031467  0.056695  0.058100  1.424267e-02   \n",
      "\n",
      "          feat_20   feat_21   feat_22   feat_23   feat_24   feat_25   feat_26  \\\n",
      "feat_1   0.070511  0.027026  0.063283  0.048686  0.067255  0.187237  0.022813   \n",
      "feat_2   0.449160  0.014113  0.215106  0.162065  0.253684  0.096366  0.064856   \n",
      "feat_3   0.011069  0.354925  0.251082  0.002427  0.031596  0.157459  0.268112   \n",
      "feat_4   0.044657  0.232923  0.247738  0.030622  0.003728  0.134231  0.365757   \n",
      "feat_5   0.046200  0.003288  0.075161  0.017281  0.075222  0.003610  0.025116   \n",
      "feat_6   0.057813  0.008046  0.038939  0.043651  0.082124  0.023319  0.004680   \n",
      "feat_7   0.364972  0.022908  0.162620  0.186462  0.244813  0.048820  0.008782   \n",
      "feat_8   0.062595  0.041095  0.029032  0.012774  0.161848  0.036939  0.041599   \n",
      "feat_9   0.023397  0.028409  0.062348  0.006940  0.073618  0.025279  0.066414   \n",
      "feat_10  0.176373  0.005134  0.141405  0.096666  0.081684  0.009792  0.003721   \n",
      "feat_11  0.005641  0.324080  0.154861  0.026500  0.064213  0.206207  0.191095   \n",
      "feat_12  0.071224  0.048215  0.119511  0.070561  0.085558  0.012281  0.042363   \n",
      "feat_13  0.194816  0.005092  0.209427  0.056463  0.192249  0.006019  0.016335   \n",
      "feat_14  0.060775  0.005645  0.063087  0.046284  0.118920  0.557929  0.101571   \n",
      "feat_15  0.034214  0.086686  0.056194  0.029100  0.062806  0.008228  0.056035   \n",
      "feat_16  0.088638  0.199338  0.433154  0.037233  0.016491  0.039487  0.187627   \n",
      "feat_17  0.261954  0.040070  0.157098  0.133157  0.257106  0.034996  0.004430   \n",
      "feat_18  0.204010  0.043272  0.364125  0.098466  0.134752  0.063139  0.234748   \n",
      "feat_19  0.105428  0.024122  0.019088  0.100028  0.290475  0.065112  0.021169   \n",
      "feat_20  1.000000  0.037885  0.244537  0.225089  0.344270  0.070439  0.004728   \n",
      "feat_21  0.037885  1.000000  0.118657  0.006174  0.041367  0.046066  0.113374   \n",
      "feat_22  0.244537  0.118657  1.000000  0.084268  0.088580  0.047433  0.234156   \n",
      "feat_23  0.225089  0.006174  0.084268  1.000000  0.211602  0.033920  0.006505   \n",
      "feat_24  0.344270  0.041367  0.088580  0.211602  1.000000  0.127708  0.037106   \n",
      "feat_25  0.070439  0.046066  0.047433  0.033920  0.127708  1.000000  0.124594   \n",
      "feat_26  0.004728  0.113374  0.234156  0.006505  0.037106  0.124594  1.000000   \n",
      "feat_27  0.038583  0.376264  0.112099  0.002618  0.017972  0.130455  0.141360   \n",
      "feat_28  0.003058  0.255093  0.253812  0.003651  0.021632  0.126680  0.348072   \n",
      "feat_29  0.059671  0.014585  0.235942  0.006970  0.039872  0.033676  0.028692   \n",
      "feat_30  0.015289  0.004026  0.015912  0.042710  0.133829  0.021807  0.025203   \n",
      "feat_31  0.123438  0.090587  0.148829  0.123996  0.060356  0.067072  0.163779   \n",
      "feat_32  0.162073  0.041779  0.023853  0.133619  0.406661  0.062620  0.042307   \n",
      "feat_33  0.043688  0.027701  0.083295  0.020804  0.042170  0.228878  0.061111   \n",
      "feat_34  0.038096  0.026739  0.043086  0.017839  0.017694  0.116002  0.034864   \n",
      "feat_35  0.153165  0.093071  0.255825  0.075955  0.049722  0.083226  0.270053   \n",
      "feat_36  0.079299  0.036644  0.044453  0.018374  0.193335  0.034228  0.030216   \n",
      "feat_37  0.196200  0.006567  0.065120  0.117174  0.186706  0.030234  0.036926   \n",
      "feat_38  0.489655  0.014056  0.253535  0.229581  0.298310  0.054273  0.021670   \n",
      "feat_39  0.085462  0.015944  0.052726  0.097809  0.181969  0.024830  0.015289   \n",
      "feat_40  0.062804  0.041815  0.018002  0.032760  0.130764  0.417171  0.067956   \n",
      "feat_41  0.190383  0.018627  0.128977  0.021231  0.110562  0.023498  0.010296   \n",
      "feat_42  0.039543  0.009244  0.238647  0.001777  0.015382  0.012705  0.378864   \n",
      "feat_43  0.043772  0.001126  0.009844  0.039410  0.047241  0.084195  0.023106   \n",
      "feat_44  0.239186  0.094507  0.416059  0.115325  0.086296  0.027719  0.177793   \n",
      "feat_45  0.046258  0.011902  0.034295  0.079246  0.150948  0.014604  0.007335   \n",
      "feat_46  0.032060  0.381266  0.171096  0.011156  0.034435  0.142185  0.219206   \n",
      "feat_47  0.103850  0.019792  0.135443  0.059817  0.079616  0.029371  0.002630   \n",
      "feat_48  0.049880  0.050541  0.002619  0.061294  0.192535  0.255795  0.078122   \n",
      "feat_49  0.194854  0.093235  0.307542  0.090079  0.045497  0.088119  0.186126   \n",
      "feat_50  0.240263  0.020847  0.122040  0.152904  0.098027  0.001294  0.020735   \n",
      "feat_51  0.073081  0.020342  0.056819  0.085160  0.053253  0.043517  0.040346   \n",
      "feat_52  0.033363  0.230007  0.081117  0.071561  0.064581  0.073940  0.110586   \n",
      "feat_53  0.028131  0.002909  0.089161  0.003534  0.021068  0.149136  0.015321   \n",
      "feat_54  0.096978  0.318332  0.502168  0.036249  0.016601  0.136167  0.375083   \n",
      "feat_55  0.254142  0.003326  0.339893  0.087014  0.149115  0.010211  0.066164   \n",
      "feat_56  0.012953  0.026285  0.003457  0.013547  0.016847  0.014930  0.007013   \n",
      "feat_57  0.136076  0.022876  0.185503  0.105362  0.208961  0.110428  0.264678   \n",
      "feat_58  0.310481  0.004761  0.057172  0.110441  0.370922  0.091991  0.004414   \n",
      "feat_59  0.020746  0.024209  0.043127  0.004395  0.041787  0.089081  0.024493   \n",
      "feat_60  0.036649  0.032751  0.166049  0.045739  0.108238  0.131640  0.344495   \n",
      "feat_61  0.027135  0.298307  0.136120  0.005186  0.018301  0.132584  0.165298   \n",
      "feat_62  0.082477  0.037588  0.319551  0.005936  0.009847  0.077595  0.005080   \n",
      "feat_63  0.047287  0.015895  0.105657  0.018242  0.035863  0.023574  0.025983   \n",
      "feat_64  0.058862  0.035863  0.017547  0.029366  0.099968  0.009117  0.060043   \n",
      "feat_65  0.143265  0.044702  0.180556  0.066825  0.074939  0.027847  0.047372   \n",
      "feat_66  0.215576  0.022228  0.382385  0.114657  0.116615  0.035200  0.119656   \n",
      "feat_67  0.199391  0.075966  0.061315  0.115559  0.435017  0.001504  0.059968   \n",
      "feat_68  0.011770  0.016256  0.036537  0.008228  0.096420  0.032516  0.016605   \n",
      "feat_69  0.348327  0.044378  0.045054  0.143339  0.419143  0.091005  0.042414   \n",
      "feat_70  0.316436  0.007589  0.231848  0.137402  0.098407  0.126664  0.012608   \n",
      "feat_71  0.098633  0.040361  0.072454  0.056965  0.123829  0.012005  0.031770   \n",
      "feat_72  0.012246  0.072693  0.113100  0.012941  0.050412  0.015754  0.046341   \n",
      "feat_73  0.029960  0.009870  0.012546  0.098657  0.048055  0.041218  0.018681   \n",
      "feat_74  0.033726  0.123215  0.040243  0.085407  0.113057  0.040263  0.036063   \n",
      "feat_75  0.053898  0.047657  0.020839  0.026762  0.156134  0.059513  0.022976   \n",
      "feat_76  0.359596  0.041637  0.060522  0.208441  0.490795  0.080761  0.039677   \n",
      "feat_77  0.017077  0.000132  0.132590  0.010125  0.033078  0.093776  0.001871   \n",
      "feat_78  0.196400  0.027742  0.084217  0.091219  0.223999  0.038806  0.019367   \n",
      "feat_79  0.074068  0.016071  0.093992  0.039133  0.083032  0.092507  0.012442   \n",
      "feat_80  0.025698  0.279618  0.189565  0.037434  0.024981  0.138356  0.206267   \n",
      "feat_81  0.144703  0.004052  0.155515  0.116575  0.084692  0.021283  0.028376   \n",
      "feat_82  0.023469  0.221444  0.177255  0.028166  0.016468  0.114821  0.154009   \n",
      "feat_83  0.364803  0.031940  0.083639  0.181739  0.300629  0.092796  0.001433   \n",
      "feat_84  0.001723  0.004702  0.041519  0.011832  0.091092  0.018320  0.048907   \n",
      "feat_85  0.084570  0.006180  0.044396  0.056994  0.018990  0.021119  0.048889   \n",
      "feat_86  0.016850  0.045562  0.018347  0.121170  0.015444  0.263924  0.072464   \n",
      "feat_87  0.220475  0.016862  0.219974  0.111837  0.123298  0.011294  0.015937   \n",
      "feat_88  0.004081  0.030401  0.045439  0.014039  0.043479  0.207974  0.078470   \n",
      "feat_89  0.111231  0.105392  0.244779  0.059743  0.023581  0.012866  0.094521   \n",
      "feat_90  0.370282  0.033193  0.098595  0.141869  0.357270  0.088187  0.021565   \n",
      "feat_91  0.079181  0.019779  0.104921  0.010438  0.090833  0.045759  0.018447   \n",
      "feat_92  0.018715  0.058008  0.200593  0.031837  0.024375  0.030135  0.199974   \n",
      "feat_93  0.110054  0.007677  0.113276  0.084945  0.089200  0.015708  0.016709   \n",
      "\n",
      "          feat_27   feat_28   feat_29   feat_30   feat_31   feat_32   feat_33  \\\n",
      "feat_1   0.038826  0.030257  0.069266  0.033108  0.011210  0.061361  0.049454   \n",
      "feat_2   0.037841  0.072494  0.025689  0.026896  0.193216  0.087699  0.033927   \n",
      "feat_3   0.508370  0.551398  0.004141  0.007667  0.138548  0.053753  0.078520   \n",
      "feat_4   0.308629  0.486417  0.014271  0.000873  0.351263  0.036730  0.054819   \n",
      "feat_5   0.002098  0.047688  0.065957  0.318117  0.013188  0.122957  0.006046   \n",
      "feat_6   0.001943  0.017132  0.002389  0.196493  0.017112  0.103699  0.019617   \n",
      "feat_7   0.015429  0.000998  0.046231  0.050535  0.032291  0.149537  0.008628   \n",
      "feat_8   0.050272  0.036668  0.104985  0.009574  0.018342  0.017910  0.045822   \n",
      "feat_9   0.042531  0.055545  0.021328  0.015830  0.016227  0.568417  0.045937   \n",
      "feat_10  0.001551  0.022349  0.068243  0.012623  0.038503  0.028812  0.003292   \n",
      "feat_11  0.599484  0.422271  0.021024  0.008832  0.109761  0.002610  0.130168   \n",
      "feat_12  0.041377  0.053406  0.063105  0.013299  0.056052  0.017584  0.048145   \n",
      "feat_13  0.012657  0.015961  0.177028  0.042503  0.024556  0.072300  0.048620   \n",
      "feat_14  0.093146  0.097257  0.027883  0.028994  0.052896  0.141330  0.258664   \n",
      "feat_15  0.045859  0.044554  0.022266  0.007801  0.029179  0.029320  0.273124   \n",
      "feat_16  0.118580  0.235443  0.142485  0.025419  0.125445  0.025200  0.239344   \n",
      "feat_17  0.036484  0.020020  0.135787  0.023838  0.028863  0.100666  0.049150   \n",
      "feat_18  0.043358  0.163886  0.118086  0.018357  0.234090  0.018512  0.033070   \n",
      "feat_19  0.018954  0.018858  0.010786  0.006245  0.032551  0.115506  0.040924   \n",
      "feat_20  0.038583  0.003058  0.059671  0.015289  0.123438  0.162073  0.043688   \n",
      "feat_21  0.376264  0.255093  0.014585  0.004026  0.090587  0.041779  0.027701   \n",
      "feat_22  0.112099  0.253812  0.235942  0.015912  0.148829  0.023853  0.083295   \n",
      "feat_23  0.002618  0.003651  0.006970  0.042710  0.123996  0.133619  0.020804   \n",
      "feat_24  0.017972  0.021632  0.039872  0.133829  0.060356  0.406661  0.042170   \n",
      "feat_25  0.130455  0.126680  0.033676  0.021807  0.067072  0.062620  0.228878   \n",
      "feat_26  0.141360  0.348072  0.028692  0.025203  0.163779  0.042307  0.061111   \n",
      "feat_27  1.000000  0.335252  0.018366  0.005345  0.094907  0.034932  0.076713   \n",
      "feat_28  0.335252  1.000000  0.004385  0.006272  0.158226  0.037948  0.051453   \n",
      "feat_29  0.018366  0.004385  1.000000  0.023337  0.003476  0.052170  0.041168   \n",
      "feat_30  0.005345  0.006272  0.023337  1.000000  0.007509  0.357488  0.019171   \n",
      "feat_31  0.094907  0.158226  0.003476  0.007509  1.000000  0.006080  0.024233   \n",
      "feat_32  0.034932  0.037948  0.052170  0.357488  0.006080  1.000000  0.069877   \n",
      "feat_33  0.076713  0.051453  0.041168  0.019171  0.024233  0.069877  1.000000   \n",
      "feat_34  0.040586  0.031705  0.011341  0.011542  0.017199  0.057669  0.011244   \n",
      "feat_35  0.082270  0.270286  0.012444  0.005045  0.556203  0.020074  0.018519   \n",
      "feat_36  0.035222  0.029080  0.082159  0.033858  0.006242  0.055867  0.020444   \n",
      "feat_37  0.008916  0.017669  0.026360  0.047492  0.033776  0.096995  0.006702   \n",
      "feat_38  0.000735  0.043109  0.041284  0.036073  0.192466  0.141960  0.044232   \n",
      "feat_39  0.018293  0.009859  0.002196  0.006347  0.037898  0.117487  0.020577   \n",
      "feat_40  0.072274  0.068621  0.022948  0.029433  0.019174  0.132667  0.283048   \n",
      "feat_41  0.032546  0.010145  0.154367  0.020869  0.009391  0.039299  0.138983   \n",
      "feat_42  0.011435  0.103324  0.065757  0.006329  0.037239  0.030480  0.012912   \n",
      "feat_43  0.017084  0.019070  0.017998  0.001860  0.055828  0.033969  0.011619   \n",
      "feat_44  0.113467  0.229184  0.103415  0.110357  0.147659  0.076621  0.025225   \n",
      "feat_45  0.008254  0.002959  0.001680  0.007040  0.022507  0.083419  0.015520   \n",
      "feat_46  0.572678  0.467904  0.014711  0.008163  0.088787  0.050622  0.076527   \n",
      "feat_47  0.012920  0.008369  0.021081  0.028166  0.001308  0.049702  0.026461   \n",
      "feat_48  0.095155  0.087888  0.073798  0.022090  0.025759  0.158067  0.083435   \n",
      "feat_49  0.130413  0.262104  0.024018  0.039086  0.234752  0.011588  0.026104   \n",
      "feat_50  0.027077  0.012332  0.011039  0.002592  0.052848  0.125057  0.008562   \n",
      "feat_51  0.037532  0.052675  0.012631  0.038955  0.058908  0.024916  0.025569   \n",
      "feat_52  0.254923  0.152144  0.007411  0.017820  0.243705  0.032964  0.040093   \n",
      "feat_53  0.033497  0.016824  0.147231  0.003433  0.010286  0.058391  0.077197   \n",
      "feat_54  0.399163  0.540238  0.084520  0.007768  0.201311  0.025977  0.002876   \n",
      "feat_55  0.004804  0.061357  0.273112  0.068682  0.042039  0.083579  0.014856   \n",
      "feat_56  0.022714  0.009810  0.001742  0.000284  0.045396  0.016834  0.015381   \n",
      "feat_57  0.015006  0.062168  0.042796  0.021861  0.116651  0.012744  0.036934   \n",
      "feat_58  0.046311  0.015941  0.003237  0.010901  0.045329  0.121410  0.024569   \n",
      "feat_59  0.028375  0.009859  0.024546  0.004690  0.014443  0.007351  0.042400   \n",
      "feat_60  0.026060  0.070262  0.039505  0.006984  0.104792  0.018205  0.072889   \n",
      "feat_61  0.500029  0.393108  0.013307  0.002622  0.120430  0.028499  0.071736   \n",
      "feat_62  0.029421  0.030228  0.252822  0.002875  0.025281  0.006367  0.238590   \n",
      "feat_63  0.010165  0.037190  0.043699  0.013605  0.042732  0.034355  0.027324   \n",
      "feat_64  0.056230  0.056711  0.017184  0.016485  0.018383  0.457911  0.030341   \n",
      "feat_65  0.022504  0.067139  0.158624  0.116697  0.046102  0.057767  0.065155   \n",
      "feat_66  0.025894  0.117781  0.201489  0.019762  0.087548  0.083371  0.056500   \n",
      "feat_67  0.097900  0.080253  0.065628  0.082248  0.018839  0.378091  0.040761   \n",
      "feat_68  0.020866  0.000733  0.000229  0.022408  0.008982  0.022240  0.031524   \n",
      "feat_69  0.037023  0.037170  0.002000  0.009822  0.039048  0.112176  0.038998   \n",
      "feat_70  0.048459  0.001360  0.079299  0.053508  0.076045  0.076867  0.119803   \n",
      "feat_71  0.050763  0.038653  0.050342  0.026446  0.005662  0.073278  0.002155   \n",
      "feat_72  0.043237  0.023556  0.010273  0.006538  0.016424  0.060626  0.303326   \n",
      "feat_73  0.005898  0.013072  0.001577  0.073560  0.003859  0.049874  0.031310   \n",
      "feat_74  0.134597  0.043981  0.005190  0.044089  0.128573  0.069656  0.024346   \n",
      "feat_75  0.040042  0.039447  0.016475  0.121927  0.008119  0.100877  0.047442   \n",
      "feat_76  0.028521  0.029717  0.012754  0.073491  0.027571  0.227310  0.031347   \n",
      "feat_77  0.007386  0.002422  0.612847  0.015577  0.016217  0.053867  0.063717   \n",
      "feat_78  0.020715  0.020470  0.026352  0.025439  0.002431  0.153970  0.027899   \n",
      "feat_79  0.027651  0.013202  0.050629  0.003858  0.006050  0.056263  0.035588   \n",
      "feat_80  0.430435  0.375019  0.008607  0.004286  0.241754  0.014739  0.068258   \n",
      "feat_81  0.004720  0.048691  0.030520  0.095881  0.045104  0.079886  0.019694   \n",
      "feat_82  0.321354  0.404849  0.003870  0.019710  0.191724  0.021769  0.048967   \n",
      "feat_83  0.024884  0.020382  0.015374  0.015350  0.049387  0.112246  0.034326   \n",
      "feat_84  0.002502  0.029148  0.007335  0.716862  0.004130  0.286689  0.005488   \n",
      "feat_85  0.046053  0.039784  0.013104  0.011743  0.012365  0.000020  0.082534   \n",
      "feat_86  0.082510  0.080806  0.011960  0.003346  0.030215  0.055348  0.038480   \n",
      "feat_87  0.028097  0.002941  0.038800  0.024016  0.028640  0.172447  0.030021   \n",
      "feat_88  0.070194  0.074442  0.032585  0.024413  0.030582  0.044493  0.098167   \n",
      "feat_89  0.099536  0.169794  0.055398  0.028606  0.147243  0.014314  0.055363   \n",
      "feat_90  0.025263  0.021330  0.000185  0.016571  0.068371  0.122931  0.048377   \n",
      "feat_91  0.018778  0.015242  0.040526  0.016378  0.008808  0.008934  0.003487   \n",
      "feat_92  0.023790  0.122653  0.084445  0.002712  0.022346  0.066435  0.099836   \n",
      "feat_93  0.000318  0.005275  0.008301  0.006260  0.030656  0.130842  0.012546   \n",
      "\n",
      "          feat_34   feat_35   feat_36   feat_37   feat_38   feat_39   feat_40  \\\n",
      "feat_1   0.042756  0.010667  0.095475  0.082306  0.104666  0.006898  0.053603   \n",
      "feat_2   0.026717  0.293374  0.026988  0.124475  0.373022  0.020626  0.057890   \n",
      "feat_3   0.039765  0.233923  0.034305  0.029425  0.046097  0.016279  0.094511   \n",
      "feat_4   0.029087  0.554474  0.029660  0.010646  0.092488  0.002700  0.076805   \n",
      "feat_5   0.001032  0.027013  0.043504  0.035320  0.070100  0.000710  0.025652   \n",
      "feat_6   0.010199  0.015739  0.014673  0.043213  0.078088  0.033470  0.018090   \n",
      "feat_7   0.022144  0.043001  0.079760  0.141960  0.315177  0.053230  0.039712   \n",
      "feat_8   0.015314  0.024043  0.606707  0.119417  0.034287  0.071908  0.044847   \n",
      "feat_9   0.050045  0.026879  0.035874  0.024868  0.028766  0.013354  0.085491   \n",
      "feat_10  0.030941  0.051793  0.007687  0.187069  0.177524  0.042223  0.038610   \n",
      "feat_11  0.068741  0.108392  0.047595  0.011404  0.055483  0.013556  0.133424   \n",
      "feat_12  0.018258  0.077834  0.063250  0.080605  0.097602  0.008417  0.016455   \n",
      "feat_13  0.004043  0.050881  0.128211  0.077847  0.191271  0.015840  0.036971   \n",
      "feat_14  0.094934  0.063796  0.042550  0.038171  0.061308  0.001582  0.487804   \n",
      "feat_15  0.049157  0.033329  0.058118  0.040783  0.001730  0.003620  0.069358   \n",
      "feat_16  0.047032  0.208736  0.012829  0.040773  0.193696  0.043227  0.287290   \n",
      "feat_17  0.009384  0.026671  0.199284  0.079853  0.136263  0.061983  0.071674   \n",
      "feat_18  0.020361  0.380595  0.076773  0.062521  0.248949  0.058047  0.024125   \n",
      "feat_19  0.022311  0.040617  0.010993  0.067629  0.246811  0.006468  0.042588   \n",
      "feat_20  0.038096  0.153165  0.079299  0.196200  0.489655  0.085462  0.062804   \n",
      "feat_21  0.026739  0.093071  0.036644  0.006567  0.014056  0.015944  0.041815   \n",
      "feat_22  0.043086  0.255825  0.044453  0.065120  0.253535  0.052726  0.018002   \n",
      "feat_23  0.017839  0.075955  0.018374  0.117174  0.229581  0.097809  0.032760   \n",
      "feat_24  0.017694  0.049722  0.193335  0.186706  0.298310  0.181969  0.130764   \n",
      "feat_25  0.116002  0.083226  0.034228  0.030234  0.054273  0.024830  0.417171   \n",
      "feat_26  0.034864  0.270053  0.030216  0.036926  0.021670  0.015289  0.067956   \n",
      "feat_27  0.040586  0.082270  0.035222  0.008916  0.000735  0.018293  0.072274   \n",
      "feat_28  0.031705  0.270286  0.029080  0.017669  0.043109  0.009859  0.068621   \n",
      "feat_29  0.011341  0.012444  0.082159  0.026360  0.041284  0.002196  0.022948   \n",
      "feat_30  0.011542  0.005045  0.033858  0.047492  0.036073  0.006347  0.029433   \n",
      "feat_31  0.017199  0.556203  0.006242  0.033776  0.192466  0.037898  0.019174   \n",
      "feat_32  0.057669  0.020074  0.055867  0.096995  0.141960  0.117487  0.132667   \n",
      "feat_33  0.011244  0.018519  0.020444  0.006702  0.044232  0.020577  0.283048   \n",
      "feat_34  1.000000  0.022267  0.036960  0.030082  0.041901  0.020456  0.063950   \n",
      "feat_35  0.022267  1.000000  0.006591  0.022481  0.213707  0.014495  0.032130   \n",
      "feat_36  0.036960  0.006591  1.000000  0.048980  0.033062  0.000932  0.047362   \n",
      "feat_37  0.030082  0.022481  0.048980  1.000000  0.208764  0.221596  0.015906   \n",
      "feat_38  0.041901  0.213707  0.033062  0.208764  1.000000  0.158922  0.021407   \n",
      "feat_39  0.020456  0.014495  0.000932  0.221596  0.158922  1.000000  0.021638   \n",
      "feat_40  0.063950  0.032130  0.047362  0.015906  0.021407  0.021638  1.000000   \n",
      "feat_41  0.017078  0.013647  0.080966  0.075154  0.145597  0.012608  0.018424   \n",
      "feat_42  0.040423  0.108742  0.004834  0.001908  0.043037  0.003142  0.049079   \n",
      "feat_43  0.036566  0.057303  0.038372  0.210933  0.116811  0.016880  0.068687   \n",
      "feat_44  0.051081  0.240699  0.014352  0.094774  0.287322  0.068093  0.010021   \n",
      "feat_45  0.019361  0.006098  0.001001  0.226044  0.121745  0.824146  0.012797   \n",
      "feat_46  0.037194  0.125971  0.037652  0.029782  0.005029  0.017705  0.079144   \n",
      "feat_47  0.016706  0.003474  0.014814  0.033234  0.073738  0.009785  0.042989   \n",
      "feat_48  0.019349  0.044776  0.140863  0.035257  0.037572  0.069612  0.115672   \n",
      "feat_49  0.033517  0.400371  0.001935  0.028267  0.232639  0.040913  0.050584   \n",
      "feat_50  0.024418  0.037632  0.002601  0.099837  0.290446  0.220654  0.013975   \n",
      "feat_51  0.009592  0.054189  0.012007  0.028112  0.038608  0.005479  0.033740   \n",
      "feat_52  0.029361  0.212755  0.014003  0.052793  0.082314  0.012768  0.020696   \n",
      "feat_53  0.028083  0.010598  0.012986  0.003284  0.024398  0.007054  0.018717   \n",
      "feat_54  0.057204  0.338959  0.004917  0.007759  0.177940  0.018700  0.027177   \n",
      "feat_55  0.034814  0.069446  0.067892  0.087968  0.194682  0.045411  0.046788   \n",
      "feat_56  0.030780  0.045052  0.018979  0.065605  0.035271  0.011425  0.013689   \n",
      "feat_57  0.017135  0.133482  0.021359  0.027198  0.086578  0.007114  0.088847   \n",
      "feat_58  0.030054  0.066752  0.017621  0.110513  0.279557  0.025101  0.057899   \n",
      "feat_59  0.021977  0.007550  0.085643  0.047061  0.028466  0.007327  0.028204   \n",
      "feat_60  0.043871  0.099420  0.007885  0.016753  0.019733  0.002745  0.099869   \n",
      "feat_61  0.036210  0.143230  0.032847  0.020982  0.023578  0.018494  0.079754   \n",
      "feat_62  0.002139  0.046613  0.161397  0.028921  0.140945  0.000659  0.084960   \n",
      "feat_63  0.022741  0.055321  0.108435  0.063563  0.101778  0.011183  0.007991   \n",
      "feat_64  0.052562  0.025843  0.028963  0.012852  0.019928  0.012954  0.073709   \n",
      "feat_65  0.029832  0.058910  0.031601  0.075276  0.160865  0.025084  0.034486   \n",
      "feat_66  0.052581  0.158498  0.119379  0.077615  0.244178  0.038623  0.019922   \n",
      "feat_67  0.046257  0.038446  0.429543  0.124339  0.128729  0.108151  0.110168   \n",
      "feat_68  0.015014  0.005791  0.057658  0.002614  0.030065  0.005995  0.053138   \n",
      "feat_69  0.026018  0.032932  0.017015  0.122514  0.301431  0.030387  0.065651   \n",
      "feat_70  0.065941  0.090731  0.021627  0.184344  0.409175  0.105113  0.090232   \n",
      "feat_71  0.003020  0.024095  0.060492  0.042454  0.060743  0.058650  0.068913   \n",
      "feat_72  0.056553  0.014162  0.035706  0.022233  0.060632  0.000943  0.102698   \n",
      "feat_73  0.011981  0.001981  0.018678  0.169925  0.023959  0.012927  0.034949   \n",
      "feat_74  0.020616  0.073560  0.021245  0.086901  0.048949  0.022546  0.002428   \n",
      "feat_75  0.041175  0.009315  0.051641  0.145731  0.046876  0.008305  0.082554   \n",
      "feat_76  0.033004  0.025874  0.076780  0.201531  0.344171  0.077277  0.070334   \n",
      "feat_77  0.010891  0.014190  0.032473  0.045286  0.025131  0.058445  0.017066   \n",
      "feat_78  0.020402  0.000374  0.152192  0.136333  0.135069  0.018723  0.022609   \n",
      "feat_79  0.012131  0.002152  0.073767  0.045819  0.067772  0.035878  0.043747   \n",
      "feat_80  0.030723  0.274259  0.023800  0.014978  0.109521  0.001440  0.065813   \n",
      "feat_81  0.012386  0.052736  0.021000  0.048283  0.179604  0.081091  0.031681   \n",
      "feat_82  0.031212  0.253594  0.023039  0.007138  0.096153  0.006898  0.068693   \n",
      "feat_83  0.012072  0.035014  0.036176  0.115052  0.215912  0.050781  0.063668   \n",
      "feat_84  0.007639  0.003819  0.011182  0.013236  0.033801  0.003668  0.022284   \n",
      "feat_85  0.047765  0.005873  0.027171  0.160492  0.181822  0.042419  0.002764   \n",
      "feat_86  0.080038  0.041072  0.029148  0.005594  0.047868  0.001503  0.050871   \n",
      "feat_87  0.019620  0.023173  0.109157  0.095706  0.166591  0.070500  0.030951   \n",
      "feat_88  0.084706  0.034481  0.037274  0.015467  0.010355  0.024618  0.162950   \n",
      "feat_89  0.029953  0.192518  0.010805  0.031177  0.263962  0.017853  0.056494   \n",
      "feat_90  0.028500  0.099522  0.008579  0.174533  0.356354  0.019496  0.057816   \n",
      "feat_91  0.006878  0.009463  0.066842  0.031573  0.027772  0.033654  0.036275   \n",
      "feat_92  0.076567  0.064610  0.000525  0.011842  0.005320  0.028631  0.031490   \n",
      "feat_93  0.015377  0.024180  0.032804  0.073010  0.083854  0.016089  0.023675   \n",
      "\n",
      "          feat_41   feat_42   feat_43   feat_44   feat_45   feat_46   feat_47  \\\n",
      "feat_1   0.080960  0.073611  0.008884  0.090439  0.008601  0.040647  0.026463   \n",
      "feat_2   0.072120  0.043863  0.017600  0.208450  0.018259  0.042491  0.014574   \n",
      "feat_3   0.028920  0.036605  0.014068  0.242901  0.007984  0.777517  0.009560   \n",
      "feat_4   0.009977  0.071397  0.000893  0.238665  0.002809  0.436274  0.012914   \n",
      "feat_5   0.038764  0.023272  0.048320  0.177137  0.004808  0.003583  0.111428   \n",
      "feat_6   0.028930  0.020709  0.015515  0.065082  0.025133  0.001892  0.018736   \n",
      "feat_7   0.097971  0.001220  0.010259  0.213066  0.036262  0.012643  0.177082   \n",
      "feat_8   0.075854  0.006578  0.004214  0.001733  0.076982  0.050246  0.015016   \n",
      "feat_9   0.023446  0.055850  0.046033  0.062023  0.013405  0.053175  0.027254   \n",
      "feat_10  0.078533  0.024055  0.111706  0.222353  0.009960  0.013546  0.046108   \n",
      "feat_11  0.038465  0.002356  0.027301  0.172523  0.006148  0.579272  0.004880   \n",
      "feat_12  0.064854  0.028966  0.007790  0.091486  0.004838  0.042124  0.018930   \n",
      "feat_13  0.123191  0.036274  0.026543  0.162659  0.011546  0.008579  0.083185   \n",
      "feat_14  0.029650  0.054232  0.073237  0.061226  0.015501  0.099296  0.059736   \n",
      "feat_15  0.026326  0.045392  0.010929  0.043123  0.018245  0.054633  0.043214   \n",
      "feat_16  0.079613  0.142924  0.046106  0.353285  0.041837  0.179348  0.038439   \n",
      "feat_17  0.153653  0.045206  0.018660  0.085796  0.028681  0.033466  0.017601   \n",
      "feat_18  0.104176  0.229794  0.007073  0.265708  0.038576  0.059749  0.072757   \n",
      "feat_19  0.025658  0.024769  0.014469  0.042021  0.006416  0.018800  0.012064   \n",
      "feat_20  0.190383  0.039543  0.043772  0.239186  0.046258  0.032060  0.103850   \n",
      "feat_21  0.018627  0.009244  0.001126  0.094507  0.011902  0.381266  0.019792   \n",
      "feat_22  0.128977  0.238647  0.009844  0.416059  0.034295  0.171096  0.135443   \n",
      "feat_23  0.021231  0.001777  0.039410  0.115325  0.079246  0.011156  0.059817   \n",
      "feat_24  0.110562  0.015382  0.047241  0.086296  0.150948  0.034435  0.079616   \n",
      "feat_25  0.023498  0.012705  0.084195  0.027719  0.014604  0.142185  0.029371   \n",
      "feat_26  0.010296  0.378864  0.023106  0.177793  0.007335  0.219206  0.002630   \n",
      "feat_27  0.032546  0.011435  0.017084  0.113467  0.008254  0.572678  0.012920   \n",
      "feat_28  0.010145  0.103324  0.019070  0.229184  0.002959  0.467904  0.008369   \n",
      "feat_29  0.154367  0.065757  0.017998  0.103415  0.001680  0.014711  0.021081   \n",
      "feat_30  0.020869  0.006329  0.001860  0.110357  0.007040  0.008163  0.028166   \n",
      "feat_31  0.009391  0.037239  0.055828  0.147659  0.022507  0.088787  0.001308   \n",
      "feat_32  0.039299  0.030480  0.033969  0.076621  0.083419  0.050622  0.049702   \n",
      "feat_33  0.138983  0.012912  0.011619  0.025225  0.015520  0.076527  0.026461   \n",
      "feat_34  0.017078  0.040423  0.036566  0.051081  0.019361  0.037194  0.016706   \n",
      "feat_35  0.013647  0.108742  0.057303  0.240699  0.006098  0.125971  0.003474   \n",
      "feat_36  0.080966  0.004834  0.038372  0.014352  0.001001  0.037652  0.014814   \n",
      "feat_37  0.075154  0.001908  0.210933  0.094774  0.226044  0.029782  0.033234   \n",
      "feat_38  0.145597  0.043037  0.116811  0.287322  0.121745  0.005029  0.073738   \n",
      "feat_39  0.012608  0.003142  0.016880  0.068093  0.824146  0.017705  0.009785   \n",
      "feat_40  0.018424  0.049079  0.068687  0.010021  0.012797  0.079144  0.042989   \n",
      "feat_41  1.000000  0.051125  0.024384  0.064104  0.004404  0.033962  0.022343   \n",
      "feat_42  0.051125  1.000000  0.005319  0.211003  0.008854  0.017923  0.017745   \n",
      "feat_43  0.024384  0.005319  1.000000  0.080740  0.017203  0.014968  0.029769   \n",
      "feat_44  0.064104  0.211003  0.080740  1.000000  0.042672  0.168826  0.154684   \n",
      "feat_45  0.004404  0.008854  0.017203  0.042672  1.000000  0.007301  0.001732   \n",
      "feat_46  0.033962  0.017923  0.014968  0.168826  0.007301  1.000000  0.017981   \n",
      "feat_47  0.022343  0.017745  0.029769  0.154684  0.001732  0.017981  1.000000   \n",
      "feat_48  0.031844  0.012537  0.051727  0.012963  0.071787  0.109526  0.015034   \n",
      "feat_49  0.033257  0.117745  0.026975  0.388665  0.031328  0.146562  0.032821   \n",
      "feat_50  0.044417  0.003906  0.056477  0.188333  0.075324  0.028279  0.033893   \n",
      "feat_51  0.009971  0.013406  0.007465  0.047933  0.001455  0.053718  0.039878   \n",
      "feat_52  0.005032  0.017937  0.026134  0.085201  0.009299  0.166742  0.004416   \n",
      "feat_53  0.130334  0.037325  0.022839  0.071061  0.007686  0.029967  0.009543   \n",
      "feat_54  0.037902  0.162022  0.001902  0.436212  0.018264  0.569484  0.044086   \n",
      "feat_55  0.115234  0.110436  0.016999  0.284885  0.034931  0.020171  0.263623   \n",
      "feat_56  0.009312  0.003905  0.176251  0.010393  0.008955  0.014555  0.013103   \n",
      "feat_57  0.029363  0.271951  0.034859  0.111068  0.003581  0.005103  0.053890   \n",
      "feat_58  0.139469  0.022257  0.014288  0.044159  0.028467  0.031832  0.013860   \n",
      "feat_59  0.069765  0.038613  0.020581  0.109311  0.004403  0.028600  0.010021   \n",
      "feat_60  0.026557  0.332167  0.039480  0.081152  0.000943  0.003954  0.016341   \n",
      "feat_61  0.026785  0.001546  0.016680  0.128809  0.009329  0.502794  0.012562   \n",
      "feat_62  0.241020  0.081785  0.016113  0.116861  0.001492  0.015610  0.046582   \n",
      "feat_63  0.059253  0.035641  0.011440  0.110515  0.005855  0.015763  0.023698   \n",
      "feat_64  0.005579  0.033987  0.023057  0.021038  0.009361  0.059347  0.003333   \n",
      "feat_65  0.071459  0.062684  0.064798  0.295100  0.017275  0.035316  0.079686   \n",
      "feat_66  0.116906  0.147385  0.000381  0.245185  0.028869  0.046867  0.120077   \n",
      "feat_67  0.166986  0.025730  0.080169  0.052369  0.081525  0.107314  0.053932   \n",
      "feat_68  0.066465  0.014355  0.037935  0.037033  0.010629  0.030122  0.244780   \n",
      "feat_69  0.084362  0.035006  0.025677  0.033513  0.014862  0.036424  0.007611   \n",
      "feat_70  0.126442  0.072523  0.321611  0.384838  0.066091  0.040795  0.054501   \n",
      "feat_71  0.053532  0.004829  0.032762  0.073183  0.022953  0.049396  0.392560   \n",
      "feat_72  0.003314  0.024687  0.026895  0.098485  0.008630  0.042077  0.031692   \n",
      "feat_73  0.004248  0.022221  0.012752  0.005774  0.018758  0.010842  0.013389   \n",
      "feat_74  0.007097  0.007210  0.022302  0.036903  0.022145  0.047543  0.002744   \n",
      "feat_75  0.071319  0.032157  0.034776  0.014398  0.015586  0.043645  0.004454   \n",
      "feat_76  0.121279  0.029785  0.022225  0.074762  0.077474  0.033789  0.066191   \n",
      "feat_77  0.072372  0.033537  0.005309  0.072331  0.063837  0.008934  0.008686   \n",
      "feat_78  0.011902  0.016144  0.015939  0.093156  0.025537  0.022669  0.193238   \n",
      "feat_79  0.127535  0.104605  0.024274  0.110654  0.031569  0.031722  0.023517   \n",
      "feat_80  0.003674  0.028310  0.003795  0.179266  0.000147  0.403677  0.010398   \n",
      "feat_81  0.070796  0.044863  0.002574  0.202801  0.026733  0.000241  0.151613   \n",
      "feat_82  0.005659  0.003465  0.009240  0.182162  0.002928  0.363145  0.007708   \n",
      "feat_83  0.056999  0.032126  0.008312  0.084283  0.045234  0.027402  0.037423   \n",
      "feat_84  0.002849  0.011315  0.002255  0.118820  0.002054  0.001093  0.105327   \n",
      "feat_85  0.004431  0.001051  0.415152  0.104923  0.026840  0.049848  0.024970   \n",
      "feat_86  0.000603  0.029038  0.109338  0.049617  0.007039  0.092302  0.033694   \n",
      "feat_87  0.045929  0.145685  0.007965  0.165350  0.051954  0.019250  0.030387   \n",
      "feat_88  0.031026  0.045588  0.158878  0.010699  0.018612  0.077723  0.042558   \n",
      "feat_89  0.100392  0.066354  0.049449  0.205417  0.007262  0.114977  0.139668   \n",
      "feat_90  0.064128  0.011508  0.014281  0.080785  0.017729  0.026419  0.013074   \n",
      "feat_91  0.071538  0.020277  0.019450  0.001806  0.033762  0.022626  0.008806   \n",
      "feat_92  0.159939  0.197315  0.040629  0.073727  0.019553  0.049284  0.016421   \n",
      "feat_93  0.007970  0.095720  0.000640  0.105498  0.014994  0.004403  0.000629   \n",
      "\n",
      "          feat_48   feat_49   feat_50   feat_51   feat_52   feat_53   feat_54  \\\n",
      "feat_1   0.130564  0.011179  0.040114  0.000902  0.013082  0.156400  0.001399   \n",
      "feat_2   0.023310  0.311278  0.071664  0.085345  0.105528  0.008015  0.179173   \n",
      "feat_3   0.111987  0.242692  0.016765  0.031979  0.169111  0.024222  0.694048   \n",
      "feat_4   0.085788  0.300011  0.005343  0.041647  0.209517  0.016308  0.525456   \n",
      "feat_5   0.013228  0.054311  0.024087  0.060318  0.003930  0.040202  0.055299   \n",
      "feat_6   0.013193  0.022864  0.027150  0.006902  0.030929  0.005914  0.031379   \n",
      "feat_7   0.056163  0.124320  0.140433  0.036493  0.006189  0.048544  0.079607   \n",
      "feat_8   0.104416  0.019114  0.001231  0.009513  0.028689  0.018674  0.025421   \n",
      "feat_9   0.077385  0.042280  0.009925  0.013181  0.009792  0.030075  0.078499   \n",
      "feat_10  0.008297  0.079449  0.252350  0.014012  0.007879  0.082810  0.074637   \n",
      "feat_11  0.125103  0.250991  0.019520  0.040465  0.272738  0.052811  0.494180   \n",
      "feat_12  0.035591  0.073520  0.041978  0.016439  0.068520  0.018344  0.102486   \n",
      "feat_13  0.089556  0.075534  0.043774  0.021930  0.009553  0.161234  0.119941   \n",
      "feat_14  0.252006  0.070134  0.014922  0.041262  0.060335  0.006808  0.112305   \n",
      "feat_15  0.051239  0.040979  0.020619  0.024504  0.022071  0.006163  0.034175   \n",
      "feat_16  0.030415  0.227799  0.058954  0.024675  0.093180  0.086820  0.468766   \n",
      "feat_17  0.102710  0.049887  0.081685  0.076780  0.004599  0.018729  0.033721   \n",
      "feat_18  0.019632  0.346609  0.074758  0.114754  0.107743  0.042412  0.272898   \n",
      "feat_19  0.038394  0.019607  0.056299  0.004868  0.017126  0.013756  0.000260   \n",
      "feat_20  0.049880  0.194854  0.240263  0.073081  0.033363  0.028131  0.096978   \n",
      "feat_21  0.050541  0.093235  0.020847  0.020342  0.230007  0.002909  0.318332   \n",
      "feat_22  0.002619  0.307542  0.122040  0.056819  0.081117  0.089161  0.502168   \n",
      "feat_23  0.061294  0.090079  0.152904  0.085160  0.071561  0.003534  0.036249   \n",
      "feat_24  0.192535  0.045497  0.098027  0.053253  0.064581  0.021068  0.016601   \n",
      "feat_25  0.255795  0.088119  0.001294  0.043517  0.073940  0.149136  0.136167   \n",
      "feat_26  0.078122  0.186126  0.020735  0.040346  0.110586  0.015321  0.375083   \n",
      "feat_27  0.095155  0.130413  0.027077  0.037532  0.254923  0.033497  0.399163   \n",
      "feat_28  0.087888  0.262104  0.012332  0.052675  0.152144  0.016824  0.540238   \n",
      "feat_29  0.073798  0.024018  0.011039  0.012631  0.007411  0.147231  0.084520   \n",
      "feat_30  0.022090  0.039086  0.002592  0.038955  0.017820  0.003433  0.007768   \n",
      "feat_31  0.025759  0.234752  0.052848  0.058908  0.243705  0.010286  0.201311   \n",
      "feat_32  0.158067  0.011588  0.125057  0.024916  0.032964  0.058391  0.025977   \n",
      "feat_33  0.083435  0.026104  0.008562  0.025569  0.040093  0.077197  0.002876   \n",
      "feat_34  0.019349  0.033517  0.024418  0.009592  0.029361  0.028083  0.057204   \n",
      "feat_35  0.044776  0.400371  0.037632  0.054189  0.212755  0.010598  0.338959   \n",
      "feat_36  0.140863  0.001935  0.002601  0.012007  0.014003  0.012986  0.004917   \n",
      "feat_37  0.035257  0.028267  0.099837  0.028112  0.052793  0.003284  0.007759   \n",
      "feat_38  0.037572  0.232639  0.290446  0.038608  0.082314  0.024398  0.177940   \n",
      "feat_39  0.069612  0.040913  0.220654  0.005479  0.012768  0.007054  0.018700   \n",
      "feat_40  0.115672  0.050584  0.013975  0.033740  0.020696  0.018717  0.027177   \n",
      "feat_41  0.031844  0.033257  0.044417  0.009971  0.005032  0.130334  0.037902   \n",
      "feat_42  0.012537  0.117745  0.003906  0.013406  0.017937  0.037325  0.162022   \n",
      "feat_43  0.051727  0.026975  0.056477  0.007465  0.026134  0.022839  0.001902   \n",
      "feat_44  0.012963  0.388665  0.188333  0.047933  0.085201  0.071061  0.436212   \n",
      "feat_45  0.071787  0.031328  0.075324  0.001455  0.009299  0.007686  0.018264   \n",
      "feat_46  0.109526  0.146562  0.028279  0.053718  0.166742  0.029967  0.569484   \n",
      "feat_47  0.015034  0.032821  0.033893  0.039878  0.004416  0.009543  0.044086   \n",
      "feat_48  1.000000  0.028316  0.067493  0.009996  0.034030  0.088261  0.085181   \n",
      "feat_49  0.028316  1.000000  0.072909  0.095188  0.105964  0.002183  0.377186   \n",
      "feat_50  0.067493  0.072909  1.000000  0.009406  0.005567  0.023457  0.038062   \n",
      "feat_51  0.009996  0.095188  0.009406  1.000000  0.052590  0.008589  0.041812   \n",
      "feat_52  0.034030  0.105964  0.005567  0.052590  1.000000  0.013411  0.186552   \n",
      "feat_53  0.088261  0.002183  0.023457  0.008589  0.013411  1.000000  0.051162   \n",
      "feat_54  0.085181  0.377186  0.038062  0.041812  0.186552  0.051162  1.000000   \n",
      "feat_55  0.045665  0.134802  0.105915  0.077797  0.014103  0.172746  0.189662   \n",
      "feat_56  0.030145  0.019829  0.011528  0.005882  0.045495  0.008639  0.024859   \n",
      "feat_57  0.000509  0.125088  0.018667  0.197049  0.028884  0.020143  0.108515   \n",
      "feat_58  0.042172  0.057051  0.051836  0.014615  0.063267  0.010057  0.055111   \n",
      "feat_59  0.086275  0.001289  0.024369  0.003985  0.018623  0.113257  0.009751   \n",
      "feat_60  0.023404  0.073149  0.005157  0.083940  0.010198  0.013433  0.097616   \n",
      "feat_61  0.086675  0.189569  0.024572  0.054041  0.253560  0.028828  0.418198   \n",
      "feat_62  0.049943  0.050396  0.008057  0.001125  0.002664  0.254328  0.144185   \n",
      "feat_63  0.036264  0.076030  0.064145  0.034871  0.025127  0.059414  0.077873   \n",
      "feat_64  0.102866  0.024424  0.081781  0.010992  0.015069  0.030776  0.058250   \n",
      "feat_65  0.019454  0.150720  0.105297  0.037801  0.049985  0.098366  0.159901   \n",
      "feat_66  0.015508  0.218145  0.061823  0.039897  0.042590  0.062345  0.252159   \n",
      "feat_67  0.407350  0.021548  0.082293  0.023045  0.027666  0.036349  0.059267   \n",
      "feat_68  0.004573  0.001992  0.001275  0.030563  0.004855  0.028684  0.006572   \n",
      "feat_69  0.041509  0.031680  0.072500  0.012447  0.010622  0.012123  0.007706   \n",
      "feat_70  0.049905  0.139793  0.347168  0.016843  0.001866  0.104217  0.127314   \n",
      "feat_71  0.030908  0.007692  0.104623  0.027463  0.018361  0.013776  0.019349   \n",
      "feat_72  0.036547  0.009323  0.016511  0.019303  0.024320  0.017736  0.064084   \n",
      "feat_73  0.084806  0.001989  0.009222  0.005856  0.007225  0.008826  0.016801   \n",
      "feat_74  0.016403  0.034674  0.016972  0.022998  0.554737  0.000711  0.066009   \n",
      "feat_75  0.050548  0.006816  0.015501  0.013396  0.012264  0.011755  0.050655   \n",
      "feat_76  0.125008  0.047059  0.088080  0.026127  0.012416  0.011055  0.008459   \n",
      "feat_77  0.090005  0.023213  0.005459  0.015799  0.001000  0.111723  0.058522   \n",
      "feat_78  0.074560  0.023846  0.051061  0.028184  0.006928  0.051046  0.019843   \n",
      "feat_79  0.057616  0.041352  0.073377  0.026622  0.006210  0.349737  0.025234   \n",
      "feat_80  0.079256  0.265545  0.001354  0.093025  0.366544  0.020713  0.444238   \n",
      "feat_81  0.006612  0.135250  0.173281  0.107963  0.019771  0.029944  0.082381   \n",
      "feat_82  0.073208  0.343092  0.001742  0.118212  0.189014  0.018108  0.394613   \n",
      "feat_83  0.049315  0.079155  0.065188  0.043710  0.005840  0.006772  0.026663   \n",
      "feat_84  0.006009  0.036612  0.003742  0.079998  0.008737  0.005698  0.026109   \n",
      "feat_85  0.040097  0.006069  0.184132  0.014491  0.010815  0.010533  0.004352   \n",
      "feat_86  0.109048  0.043704  0.132234  0.025348  0.040474  0.076241  0.081509   \n",
      "feat_87  0.099603  0.059635  0.125120  0.081376  0.003709  0.004038  0.057848   \n",
      "feat_88  0.074914  0.045283  0.042334  0.026219  0.041361  0.015673  0.083348   \n",
      "feat_89  0.009900  0.217931  0.076471  0.063843  0.115156  0.081352  0.256305   \n",
      "feat_90  0.070611  0.075120  0.080146  0.017031  0.043089  0.016199  0.034802   \n",
      "feat_91  0.047508  0.006273  0.008383  0.000449  0.012082  0.004092  0.018972   \n",
      "feat_92  0.023148  0.037485  0.040728  0.013779  0.026573  0.135595  0.153724   \n",
      "feat_93  0.053749  0.029427  0.052934  0.017484  0.014812  0.006926  0.029959   \n",
      "\n",
      "          feat_55   feat_56   feat_57   feat_58   feat_59   feat_60   feat_61  \\\n",
      "feat_1   0.165044  0.015738  0.014598  0.028696  0.139364  0.020267  0.038124   \n",
      "feat_2   0.134617  0.018908  0.104263  0.371146  0.013283  0.018219  0.063873   \n",
      "feat_3   0.055445  0.016288  0.018887  0.034506  0.021717  0.011369  0.494881   \n",
      "feat_4   0.070077  0.019618  0.069462  0.039103  0.015802  0.068900  0.354402   \n",
      "feat_5   0.142221  0.001560  0.037162  0.001546  0.038893  0.008796  0.001559   \n",
      "feat_6   0.062765  0.004014  0.036655  0.110743  0.001285  0.022289  0.002227   \n",
      "feat_7   0.343752  0.003145  0.073969  0.106203  0.009624  0.012336  0.006284   \n",
      "feat_8   0.063415  0.002285  0.011968  0.025674  0.118223  0.020328  0.045203   \n",
      "feat_9   0.042138  0.022639  0.049335  0.028290  0.021451  0.066333  0.044343   \n",
      "feat_10  0.213003  0.009786  0.017525  0.020933  0.051189  0.000619  0.004397   \n",
      "feat_11  0.022723  0.033221  0.007302  0.179361  0.039646  0.025990  0.543858   \n",
      "feat_12  0.081504  0.007665  0.042518  0.082574  0.026684  0.027027  0.050080   \n",
      "feat_13  0.223364  0.002066  0.075397  0.132569  0.174173  0.046147  0.009951   \n",
      "feat_14  0.055732  0.004436  0.101382  0.059363  0.026693  0.118618  0.093809   \n",
      "feat_15  0.040297  0.022519  0.069910  0.044920  0.017773  0.079152  0.056169   \n",
      "feat_16  0.131218  0.021978  0.049373  0.018927  0.046633  0.043111  0.150952   \n",
      "feat_17  0.163377  0.012085  0.173169  0.040370  0.058797  0.090963  0.030037   \n",
      "feat_18  0.183218  0.014345  0.284435  0.048980  0.099073  0.234885  0.084745   \n",
      "feat_19  0.018046  0.008824  0.026194  0.234503  0.008378  0.002039  0.014633   \n",
      "feat_20  0.254142  0.012953  0.136076  0.310481  0.020746  0.036649  0.027135   \n",
      "feat_21  0.003326  0.026285  0.022876  0.004761  0.024209  0.032751  0.298307   \n",
      "feat_22  0.339893  0.003457  0.185503  0.057172  0.043127  0.166049  0.136120   \n",
      "feat_23  0.087014  0.013547  0.105362  0.110441  0.004395  0.045739  0.005186   \n",
      "feat_24  0.149115  0.016847  0.208961  0.370922  0.041787  0.108238  0.018301   \n",
      "feat_25  0.010211  0.014930  0.110428  0.091991  0.089081  0.131640  0.132584   \n",
      "feat_26  0.066164  0.007013  0.264678  0.004414  0.024493  0.344495  0.165298   \n",
      "feat_27  0.004804  0.022714  0.015006  0.046311  0.028375  0.026060  0.500029   \n",
      "feat_28  0.061357  0.009810  0.062168  0.015941  0.009859  0.070262  0.393108   \n",
      "feat_29  0.273112  0.001742  0.042796  0.003237  0.024546  0.039505  0.013307   \n",
      "feat_30  0.068682  0.000284  0.021861  0.010901  0.004690  0.006984  0.002622   \n",
      "feat_31  0.042039  0.045396  0.116651  0.045329  0.014443  0.104792  0.120430   \n",
      "feat_32  0.083579  0.016834  0.012744  0.121410  0.007351  0.018205  0.028499   \n",
      "feat_33  0.014856  0.015381  0.036934  0.024569  0.042400  0.072889  0.071736   \n",
      "feat_34  0.034814  0.030780  0.017135  0.030054  0.021977  0.043871  0.036210   \n",
      "feat_35  0.069446  0.045052  0.133482  0.066752  0.007550  0.099420  0.143230   \n",
      "feat_36  0.067892  0.018979  0.021359  0.017621  0.085643  0.007885  0.032847   \n",
      "feat_37  0.087968  0.065605  0.027198  0.110513  0.047061  0.016753  0.020982   \n",
      "feat_38  0.194682  0.035271  0.086578  0.279557  0.028466  0.019733  0.023578   \n",
      "feat_39  0.045411  0.011425  0.007114  0.025101  0.007327  0.002745  0.018494   \n",
      "feat_40  0.046788  0.013689  0.088847  0.057899  0.028204  0.099869  0.079754   \n",
      "feat_41  0.115234  0.009312  0.029363  0.139469  0.069765  0.026557  0.026785   \n",
      "feat_42  0.110436  0.003905  0.271951  0.022257  0.038613  0.332167  0.001546   \n",
      "feat_43  0.016999  0.176251  0.034859  0.014288  0.020581  0.039480  0.016680   \n",
      "feat_44  0.284885  0.010393  0.111068  0.044159  0.109311  0.081152  0.128809   \n",
      "feat_45  0.034931  0.008955  0.003581  0.028467  0.004403  0.000943  0.009329   \n",
      "feat_46  0.020171  0.014555  0.005103  0.031832  0.028600  0.003954  0.502794   \n",
      "feat_47  0.263623  0.013103  0.053890  0.013860  0.010021  0.016341  0.012562   \n",
      "feat_48  0.045665  0.030145  0.000509  0.042172  0.086275  0.023404  0.086675   \n",
      "feat_49  0.134802  0.019829  0.125088  0.057051  0.001289  0.073149  0.189569   \n",
      "feat_50  0.105915  0.011528  0.018667  0.051836  0.024369  0.005157  0.024572   \n",
      "feat_51  0.077797  0.005882  0.197049  0.014615  0.003985  0.083940  0.054041   \n",
      "feat_52  0.014103  0.045495  0.028884  0.063267  0.018623  0.010198  0.253560   \n",
      "feat_53  0.172746  0.008639  0.020143  0.010057  0.113257  0.013433  0.028828   \n",
      "feat_54  0.189662  0.024859  0.108515  0.055111  0.009751  0.097616  0.418198   \n",
      "feat_55  1.000000  0.006551  0.148330  0.055781  0.054114  0.087929  0.019997   \n",
      "feat_56  0.006551  1.000000  0.014550  0.005532  0.010537  0.022879  0.022271   \n",
      "feat_57  0.148330  0.014550  1.000000  0.038391  0.009761  0.471676  0.000989   \n",
      "feat_58  0.055781  0.005532  0.038391  1.000000  0.015981  0.019522  0.031040   \n",
      "feat_59  0.054114  0.010537  0.009761  0.015981  1.000000  0.008088  0.026870   \n",
      "feat_60  0.087929  0.022879  0.471676  0.019522  0.008088  1.000000  0.010940   \n",
      "feat_61  0.019997  0.022271  0.000989  0.031040  0.026870  0.010940  1.000000   \n",
      "feat_62  0.139948  0.003510  0.003693  0.005153  0.145860  0.006826  0.012790   \n",
      "feat_63  0.105285  0.005410  0.017826  0.010079  0.181146  0.012982  0.021335   \n",
      "feat_64  0.004749  0.018979  0.008348  0.001655  0.006894  0.055960  0.056514   \n",
      "feat_65  0.217883  0.008946  0.045171  0.019621  0.060893  0.025368  0.040793   \n",
      "feat_66  0.196955  0.010837  0.114737  0.059694  0.053516  0.073185  0.044301   \n",
      "feat_67  0.122626  0.030794  0.078598  0.134576  0.398837  0.037814  0.091471   \n",
      "feat_68  0.056319  0.002327  0.039628  0.022746  0.020244  0.021800  0.020848   \n",
      "feat_69  0.062175  0.002325  0.073636  0.392839  0.003780  0.014202  0.030713   \n",
      "feat_70  0.195079  0.089528  0.028737  0.038319  0.229503  0.008525  0.039559   \n",
      "feat_71  0.185377  0.020002  0.034841  0.020289  0.015826  0.020767  0.046490   \n",
      "feat_72  0.004484  0.019436  0.055874  0.024774  0.014535  0.062609  0.042920   \n",
      "feat_73  0.001851  0.000268  0.006254  0.057573  0.015477  0.019131  0.006029   \n",
      "feat_74  0.016692  0.032422  0.010981  0.058768  0.002264  0.006299  0.104516   \n",
      "feat_75  0.039229  0.020258  0.056871  0.053835  0.000676  0.006871  0.036555   \n",
      "feat_76  0.136822  0.004393  0.066284  0.365075  0.013573  0.011742  0.020551   \n",
      "feat_77  0.068623  0.005466  0.006597  0.001763  0.048736  0.024297  0.001620   \n",
      "feat_78  0.246042  0.004568  0.052778  0.044968  0.007320  0.002122  0.020607   \n",
      "feat_79  0.106119  0.013829  0.040919  0.023352  0.189173  0.029188  0.019744   \n",
      "feat_80  0.051735  0.043107  0.049279  0.078906  0.024168  0.026903  0.575101   \n",
      "feat_81  0.182156  0.002544  0.084484  0.001446  0.055354  0.055102  0.007611   \n",
      "feat_82  0.053607  0.023436  0.036242  0.031475  0.017358  0.007344  0.474954   \n",
      "feat_83  0.113962  0.005914  0.144449  0.235803  0.011922  0.068019  0.021269   \n",
      "feat_84  0.102831  0.003549  0.043523  0.005284  0.002065  0.020174  0.002943   \n",
      "feat_85  0.025289  0.016538  0.034281  0.015734  0.037042  0.043094  0.047659   \n",
      "feat_86  0.021798  0.072723  0.055232  0.026106  0.148417  0.078149  0.083301   \n",
      "feat_87  0.127128  0.005218  0.102572  0.005737  0.149499  0.099639  0.027363   \n",
      "feat_88  0.033113  0.012290  0.068615  0.045055  0.002890  0.091812  0.070815   \n",
      "feat_89  0.157031  0.053164  0.051941  0.022821  0.013638  0.031379  0.160405   \n",
      "feat_90  0.069305  0.008288  0.087563  0.407341  0.011079  0.014203  0.018112   \n",
      "feat_91  0.055702  0.003705  0.010013  0.056724  0.003928  0.011737  0.017045   \n",
      "feat_92  0.049364  0.028787  0.100979  0.014156  0.010828  0.128371  0.042357   \n",
      "feat_93  0.055275  0.007368  0.045129  0.034472  0.047987  0.055350  0.001258   \n",
      "\n",
      "          feat_62   feat_63   feat_64   feat_65   feat_66   feat_67   feat_68  \\\n",
      "feat_1   0.091636  0.069799  0.010499  0.110041  0.053010  0.154301  0.014674   \n",
      "feat_2   0.029040  0.037020  0.005354  0.078801  0.175620  0.068667  0.012802   \n",
      "feat_3   0.010733  0.031241  0.065105  0.065492  0.088017  0.110081  0.030992   \n",
      "feat_4   0.025228  0.059903  0.047288  0.062285  0.129654  0.080457  0.020092   \n",
      "feat_5   0.073215  0.082383  0.021017  0.228349  0.048364  0.061964  0.107405   \n",
      "feat_6   0.022874  0.017050  0.002764  0.066867  0.033285  0.038289  0.021619   \n",
      "feat_7   0.025485  0.039585  0.011165  0.202346  0.122660  0.148598  0.040309   \n",
      "feat_8   0.214717  0.148673  0.003194  0.025544  0.115175  0.320949  0.075384   \n",
      "feat_9   0.021138  0.013012  0.702951  0.038163  0.001778  0.176921  0.012192   \n",
      "feat_10  0.054733  0.090421  0.022536  0.182756  0.100722  0.043117  0.001693   \n",
      "feat_11  0.051554  0.013436  0.082868  0.041014  0.050764  0.114944  0.040588   \n",
      "feat_12  0.126326  0.038777  0.012603  0.121550  0.084591  0.079377  0.014873   \n",
      "feat_13  0.164920  0.070109  0.003286  0.108567  0.122272  0.211335  0.026881   \n",
      "feat_14  0.026822  0.020837  0.066800  0.011517  0.081709  0.086404  0.076979   \n",
      "feat_15  0.097920  0.018947  0.003848  0.034954  0.018487  0.052486  0.049076   \n",
      "feat_16  0.337177  0.104440  0.043076  0.182735  0.289082  0.027728  0.015351   \n",
      "feat_17  0.105504  0.032391  0.021551  0.072733  0.165143  0.306513  0.038816   \n",
      "feat_18  0.197339  0.119569  0.001922  0.114910  0.285667  0.108521  0.078124   \n",
      "feat_19  0.014714  0.012208  0.016986  0.015797  0.030787  0.035876  0.005129   \n",
      "feat_20  0.082477  0.047287  0.058862  0.143265  0.215576  0.199391  0.011770   \n",
      "feat_21  0.037588  0.015895  0.035863  0.044702  0.022228  0.075966  0.016256   \n",
      "feat_22  0.319551  0.105657  0.017547  0.180556  0.382385  0.061315  0.036537   \n",
      "feat_23  0.005936  0.018242  0.029366  0.066825  0.114657  0.115559  0.008228   \n",
      "feat_24  0.009847  0.035863  0.099968  0.074939  0.116615  0.435017  0.096420   \n",
      "feat_25  0.077595  0.023574  0.009117  0.027847  0.035200  0.001504  0.032516   \n",
      "feat_26  0.005080  0.025983  0.060043  0.047372  0.119656  0.059968  0.016605   \n",
      "feat_27  0.029421  0.010165  0.056230  0.022504  0.025894  0.097900  0.020866   \n",
      "feat_28  0.030228  0.037190  0.056711  0.067139  0.117781  0.080253  0.000733   \n",
      "feat_29  0.252822  0.043699  0.017184  0.158624  0.201489  0.065628  0.000229   \n",
      "feat_30  0.002875  0.013605  0.016485  0.116697  0.019762  0.082248  0.022408   \n",
      "feat_31  0.025281  0.042732  0.018383  0.046102  0.087548  0.018839  0.008982   \n",
      "feat_32  0.006367  0.034355  0.457911  0.057767  0.083371  0.378091  0.022240   \n",
      "feat_33  0.238590  0.027324  0.030341  0.065155  0.056500  0.040761  0.031524   \n",
      "feat_34  0.002139  0.022741  0.052562  0.029832  0.052581  0.046257  0.015014   \n",
      "feat_35  0.046613  0.055321  0.025843  0.058910  0.158498  0.038446  0.005791   \n",
      "feat_36  0.161397  0.108435  0.028963  0.031601  0.119379  0.429543  0.057658   \n",
      "feat_37  0.028921  0.063563  0.012852  0.075276  0.077615  0.124339  0.002614   \n",
      "feat_38  0.140945  0.101778  0.019928  0.160865  0.244178  0.128729  0.030065   \n",
      "feat_39  0.000659  0.011183  0.012954  0.025084  0.038623  0.108151  0.005995   \n",
      "feat_40  0.084960  0.007991  0.073709  0.034486  0.019922  0.110168  0.053138   \n",
      "feat_41  0.241020  0.059253  0.005579  0.071459  0.116906  0.166986  0.066465   \n",
      "feat_42  0.081785  0.035641  0.033987  0.062684  0.147385  0.025730  0.014355   \n",
      "feat_43  0.016113  0.011440  0.023057  0.064798  0.000381  0.080169  0.037935   \n",
      "feat_44  0.116861  0.110515  0.021038  0.295100  0.245185  0.052369  0.037033   \n",
      "feat_45  0.001492  0.005855  0.009361  0.017275  0.028869  0.081525  0.010629   \n",
      "feat_46  0.015610  0.015763  0.059347  0.035316  0.046867  0.107314  0.030122   \n",
      "feat_47  0.046582  0.023698  0.003333  0.079686  0.120077  0.053932  0.244780   \n",
      "feat_48  0.049943  0.036264  0.102866  0.019454  0.015508  0.407350  0.004573   \n",
      "feat_49  0.050396  0.076030  0.024424  0.150720  0.218145  0.021548  0.001992   \n",
      "feat_50  0.008057  0.064145  0.081781  0.105297  0.061823  0.082293  0.001275   \n",
      "feat_51  0.001125  0.034871  0.010992  0.037801  0.039897  0.023045  0.030563   \n",
      "feat_52  0.002664  0.025127  0.015069  0.049985  0.042590  0.027666  0.004855   \n",
      "feat_53  0.254328  0.059414  0.030776  0.098366  0.062345  0.036349  0.028684   \n",
      "feat_54  0.144185  0.077873  0.058250  0.159901  0.252159  0.059267  0.006572   \n",
      "feat_55  0.139948  0.105285  0.004749  0.217883  0.196955  0.122626  0.056319   \n",
      "feat_56  0.003510  0.005410  0.018979  0.008946  0.010837  0.030794  0.002327   \n",
      "feat_57  0.003693  0.017826  0.008348  0.045171  0.114737  0.078598  0.039628   \n",
      "feat_58  0.005153  0.010079  0.001655  0.019621  0.059694  0.134576  0.022746   \n",
      "feat_59  0.145860  0.181146  0.006894  0.060893  0.053516  0.398837  0.020244   \n",
      "feat_60  0.006826  0.012982  0.055960  0.025368  0.073185  0.037814  0.021800   \n",
      "feat_61  0.012790  0.021335  0.056514  0.040793  0.044301  0.091471  0.020848   \n",
      "feat_62  1.000000  0.140069  0.017422  0.104989  0.342523  0.131518  0.092759   \n",
      "feat_63  0.140069  1.000000  0.038846  0.069119  0.142239  0.117070  0.019061   \n",
      "feat_64  0.017422  0.038846  1.000000  0.015070  0.106675  0.226275  0.019150   \n",
      "feat_65  0.104989  0.069119  0.015070  1.000000  0.120274  0.041425  0.021648   \n",
      "feat_66  0.342523  0.142239  0.106675  0.120274  1.000000  0.142036  0.044337   \n",
      "feat_67  0.131518  0.117070  0.226275  0.041425  0.142036  1.000000  0.078876   \n",
      "feat_68  0.092759  0.019061  0.019150  0.021648  0.044337  0.078876  1.000000   \n",
      "feat_69  0.011587  0.014252  0.004913  0.015448  0.079321  0.117455  0.017943   \n",
      "feat_70  0.159988  0.141850  0.029886  0.237573  0.212931  0.127927  0.008038   \n",
      "feat_71  0.089270  0.049447  0.058239  0.045051  0.072601  0.131528  0.196309   \n",
      "feat_72  0.162735  0.030482  0.039423  0.067476  0.037809  0.035209  0.035915   \n",
      "feat_73  0.020436  0.000880  0.018469  0.006393  0.006615  0.098574  0.004141   \n",
      "feat_74  0.001716  0.013859  0.007032  0.050069  0.021841  0.027789  0.008749   \n",
      "feat_75  0.044457  0.004057  0.011456  0.014663  0.005674  0.121735  0.008292   \n",
      "feat_76  0.001444  0.026610  0.000467  0.072170  0.081866  0.323678  0.013405   \n",
      "feat_77  0.214828  0.035163  0.019363  0.115692  0.147127  0.052970  0.001655   \n",
      "feat_78  0.001336  0.006096  0.029796  0.099054  0.058784  0.188128  0.084630   \n",
      "feat_79  0.169192  0.194618  0.020209  0.066714  0.085545  0.152964  0.062394   \n",
      "feat_80  0.027840  0.066939  0.053117  0.062834  0.107806  0.076209  0.004003   \n",
      "feat_81  0.090282  0.088525  0.010230  0.127931  0.110691  0.075488  0.084468   \n",
      "feat_82  0.026090  0.052736  0.047284  0.088551  0.093782  0.070049  0.000787   \n",
      "feat_83  0.022311  0.005221  0.001432  0.057058  0.088982  0.127444  0.000817   \n",
      "feat_84  0.012320  0.013394  0.013347  0.120772  0.033794  0.039005  0.036198   \n",
      "feat_85  0.031092  0.026954  0.009820  0.066922  0.031291  0.020048  0.024620   \n",
      "feat_86  0.006503  0.080492  0.064883  0.030362  0.088686  0.081445  0.038904   \n",
      "feat_87  0.081320  0.056860  0.075964  0.050138  0.206406  0.295803  0.001672   \n",
      "feat_88  0.048727  0.026337  0.117596  0.011600  0.000679  0.058706  0.059843   \n",
      "feat_89  0.246666  0.163009  0.015218  0.125884  0.205289  0.005220  0.125150   \n",
      "feat_90  0.000511  0.018277  0.001930  0.029076  0.094925  0.089262  0.023839   \n",
      "feat_91  0.279845  0.009487  0.020846  0.001188  0.098063  0.112052  0.022515   \n",
      "feat_92  0.327176  0.031464  0.071886  0.044286  0.123694  0.011247  0.095970   \n",
      "feat_93  0.012774  0.023298  0.023064  0.015500  0.067957  0.129018  0.004602   \n",
      "\n",
      "          feat_69   feat_70   feat_71   feat_72   feat_73   feat_74   feat_75  \\\n",
      "feat_1   0.007544  0.165442  0.013712  0.029983  0.140815  0.051365  0.011596   \n",
      "feat_2   0.307406  0.112968  0.002336  0.023267  0.039192  0.070724  0.093689   \n",
      "feat_3   0.032748  0.018774  0.053020  0.045339  0.013972  0.041559  0.044724   \n",
      "feat_4   0.014461  0.020798  0.042413  0.029796  0.011285  0.049097  0.031454   \n",
      "feat_5   0.003294  0.118510  0.056428  0.005177  0.001609  0.017265  0.015279   \n",
      "feat_6   0.074836  0.052401  0.011901  0.011090  0.025023  0.043160  0.006951   \n",
      "feat_7   0.131430  0.237907  0.115813  0.014921  0.022819  0.053059  0.039865   \n",
      "feat_8   0.046258  0.023089  0.081664  0.029868  0.028999  0.000431  0.031466   \n",
      "feat_9   0.029335  0.056205  0.043286  0.058147  0.022679  0.007594  0.027313   \n",
      "feat_10  0.077354  0.322857  0.104834  0.004225  0.000240  0.008912  0.003828   \n",
      "feat_11  0.010697  0.050540  0.071717  0.086496  0.006893  0.186561  0.034513   \n",
      "feat_12  0.057074  0.071630  0.028629  0.031888  0.181833  0.087012  0.052536   \n",
      "feat_13  0.086611  0.145294  0.052337  0.008543  0.030927  0.021072  0.022496   \n",
      "feat_14  0.058352  0.054226  0.076935  0.022358  0.040175  0.033498  0.079099   \n",
      "feat_15  0.052924  0.118795  0.033807  0.764664  0.019194  0.013632  0.048894   \n",
      "feat_16  0.021699  0.255740  0.001594  0.394998  0.026285  0.034125  0.059511   \n",
      "feat_17  0.100656  0.108789  0.150286  0.023103  0.002826  0.001770  0.025506   \n",
      "feat_18  0.071926  0.169238  0.074981  0.002190  0.008668  0.038278  0.011790   \n",
      "feat_19  0.397621  0.031171  0.020871  0.022584  0.056837  0.022614  0.036357   \n",
      "feat_20  0.348327  0.316436  0.098633  0.012246  0.029960  0.033726  0.053898   \n",
      "feat_21  0.044378  0.007589  0.040361  0.072693  0.009870  0.123215  0.047657   \n",
      "feat_22  0.045054  0.231848  0.072454  0.113100  0.012546  0.040243  0.020839   \n",
      "feat_23  0.143339  0.137402  0.056965  0.012941  0.098657  0.085407  0.026762   \n",
      "feat_24  0.419143  0.098407  0.123829  0.050412  0.048055  0.113057  0.156134   \n",
      "feat_25  0.091005  0.126664  0.012005  0.015754  0.041218  0.040263  0.059513   \n",
      "feat_26  0.042414  0.012608  0.031770  0.046341  0.018681  0.036063  0.022976   \n",
      "feat_27  0.037023  0.048459  0.050763  0.043237  0.005898  0.134597  0.040042   \n",
      "feat_28  0.037170  0.001360  0.038653  0.023556  0.013072  0.043981  0.039447   \n",
      "feat_29  0.002000  0.079299  0.050342  0.010273  0.001577  0.005190  0.016475   \n",
      "feat_30  0.009822  0.053508  0.026446  0.006538  0.073560  0.044089  0.121927   \n",
      "feat_31  0.039048  0.076045  0.005662  0.016424  0.003859  0.128573  0.008119   \n",
      "feat_32  0.112176  0.076867  0.073278  0.060626  0.049874  0.069656  0.100877   \n",
      "feat_33  0.038998  0.119803  0.002155  0.303326  0.031310  0.024346  0.047442   \n",
      "feat_34  0.026018  0.065941  0.003020  0.056553  0.011981  0.020616  0.041175   \n",
      "feat_35  0.032932  0.090731  0.024095  0.014162  0.001981  0.073560  0.009315   \n",
      "feat_36  0.017015  0.021627  0.060492  0.035706  0.018678  0.021245  0.051641   \n",
      "feat_37  0.122514  0.184344  0.042454  0.022233  0.169925  0.086901  0.145731   \n",
      "feat_38  0.301431  0.409175  0.060743  0.060632  0.023959  0.048949  0.046876   \n",
      "feat_39  0.030387  0.105113  0.058650  0.000943  0.012927  0.022546  0.008305   \n",
      "feat_40  0.065651  0.090232  0.068913  0.102698  0.034949  0.002428  0.082554   \n",
      "feat_41  0.084362  0.126442  0.053532  0.003314  0.004248  0.007097  0.071319   \n",
      "feat_42  0.035006  0.072523  0.004829  0.024687  0.022221  0.007210  0.032157   \n",
      "feat_43  0.025677  0.321611  0.032762  0.026895  0.012752  0.022302  0.034776   \n",
      "feat_44  0.033513  0.384838  0.073183  0.098485  0.005774  0.036903  0.014398   \n",
      "feat_45  0.014862  0.066091  0.022953  0.008630  0.018758  0.022145  0.015586   \n",
      "feat_46  0.036424  0.040795  0.049396  0.042077  0.010842  0.047543  0.043645   \n",
      "feat_47  0.007611  0.054501  0.392560  0.031692  0.013389  0.002744  0.004454   \n",
      "feat_48  0.041509  0.049905  0.030908  0.036547  0.084806  0.016403  0.050548   \n",
      "feat_49  0.031680  0.139793  0.007692  0.009323  0.001989  0.034674  0.006816   \n",
      "feat_50  0.072500  0.347168  0.104623  0.016511  0.009222  0.016972  0.015501   \n",
      "feat_51  0.012447  0.016843  0.027463  0.019303  0.005856  0.022998  0.013396   \n",
      "feat_52  0.010622  0.001866  0.018361  0.024320  0.007225  0.554737  0.012264   \n",
      "feat_53  0.012123  0.104217  0.013776  0.017736  0.008826  0.000711  0.011755   \n",
      "feat_54  0.007706  0.127314  0.019349  0.064084  0.016801  0.066009  0.050655   \n",
      "feat_55  0.062175  0.195079  0.185377  0.004484  0.001851  0.016692  0.039229   \n",
      "feat_56  0.002325  0.089528  0.020002  0.019436  0.000268  0.032422  0.020258   \n",
      "feat_57  0.073636  0.028737  0.034841  0.055874  0.006254  0.010981  0.056871   \n",
      "feat_58  0.392839  0.038319  0.020289  0.024774  0.057573  0.058768  0.053835   \n",
      "feat_59  0.003780  0.229503  0.015826  0.014535  0.015477  0.002264  0.000676   \n",
      "feat_60  0.014202  0.008525  0.020767  0.062609  0.019131  0.006299  0.006871   \n",
      "feat_61  0.030713  0.039559  0.046490  0.042920  0.006029  0.104516  0.036555   \n",
      "feat_62  0.011587  0.159988  0.089270  0.162735  0.020436  0.001716  0.044457   \n",
      "feat_63  0.014252  0.141850  0.049447  0.030482  0.000880  0.013859  0.004057   \n",
      "feat_64  0.004913  0.029886  0.058239  0.039423  0.018469  0.007032  0.011456   \n",
      "feat_65  0.015448  0.237573  0.045051  0.067476  0.006393  0.050069  0.014663   \n",
      "feat_66  0.079321  0.212931  0.072601  0.037809  0.006615  0.021841  0.005674   \n",
      "feat_67  0.117455  0.127927  0.131528  0.035209  0.098574  0.027789  0.121735   \n",
      "feat_68  0.017943  0.008038  0.196309  0.035915  0.004141  0.008749  0.008292   \n",
      "feat_69  1.000000  0.077678  0.000955  0.032479  0.045667  0.034456  0.065643   \n",
      "feat_70  0.077678  1.000000  0.088064  0.192594  0.002807  0.010436  0.016054   \n",
      "feat_71  0.000955  0.088064  1.000000  0.015731  0.006121  0.012457  0.025972   \n",
      "feat_72  0.032479  0.192594  0.015731  1.000000  0.020547  0.024074  0.043639   \n",
      "feat_73  0.045667  0.002807  0.006121  0.020547  1.000000  0.035633  0.190639   \n",
      "feat_74  0.034456  0.010436  0.012457  0.024074  0.035633  1.000000  0.002297   \n",
      "feat_75  0.065643  0.016054  0.025972  0.043639  0.190639  0.002297  1.000000   \n",
      "feat_76  0.409691  0.122972  0.036850  0.040593  0.220832  0.058736  0.172375   \n",
      "feat_77  0.002209  0.067065  0.002787  0.016045  0.001634  0.009406  0.010904   \n",
      "feat_78  0.044834  0.105585  0.099693  0.022904  0.018161  0.069628  0.013876   \n",
      "feat_79  0.036435  0.144247  0.063309  0.017522  0.015294  0.016904  0.006885   \n",
      "feat_80  0.014590  0.004856  0.029219  0.038547  0.007285  0.174775  0.037439   \n",
      "feat_81  0.019568  0.192565  0.132951  0.002439  0.002868  0.008959  0.011281   \n",
      "feat_82  0.017440  0.009776  0.025602  0.028420  0.005143  0.055336  0.022992   \n",
      "feat_83  0.204569  0.069588  0.026032  0.034569  0.027131  0.024667  0.063463   \n",
      "feat_84  0.007221  0.045636  0.011858  0.011992  0.012885  0.028021  0.028478   \n",
      "feat_85  0.006954  0.361941  0.013894  0.294384  0.010675  0.000453  0.026329   \n",
      "feat_86  0.025538  0.225792  0.015410  0.008897  0.000841  0.015945  0.031401   \n",
      "feat_87  0.027690  0.212133  0.060004  0.013536  0.004759  0.003992  0.001201   \n",
      "feat_88  0.022918  0.140850  0.048676  0.004066  0.026363  0.025207  0.058630   \n",
      "feat_89  0.011806  0.163631  0.076348  0.057040  0.006704  0.042104  0.014925   \n",
      "feat_90  0.549489  0.074178  0.019694  0.030673  0.070001  0.055372  0.160418   \n",
      "feat_91  0.041206  0.030560  0.050622  0.008936  0.007193  0.016941  0.002625   \n",
      "feat_92  0.037961  0.007310  0.000368  0.005300  0.024017  0.004497  0.037710   \n",
      "feat_93  0.032052  0.093488  0.002001  0.008233  0.000163  0.021967  0.006208   \n",
      "\n",
      "          feat_76   feat_77   feat_78   feat_79   feat_80   feat_81   feat_82  \\\n",
      "feat_1   0.153808  0.123752  0.279202  0.228912  0.013303  0.032427  0.026085   \n",
      "feat_2   0.259360  0.014911  0.094256  0.033668  0.155768  0.052101  0.119109   \n",
      "feat_3   0.028670  0.001584  0.021979  0.020566  0.442036  0.013089  0.438458   \n",
      "feat_4   0.013792  0.015318  0.014499  0.010835  0.405772  0.028284  0.436541   \n",
      "feat_5   0.035570  0.030462  0.070709  0.055115  0.026223  0.129333  0.057400   \n",
      "feat_6   0.073867  0.006501  0.061250  0.009942  0.017648  0.044136  0.014907   \n",
      "feat_7   0.375114  0.005769  0.567084  0.066753  0.028860  0.144308  0.022059   \n",
      "feat_8   0.081682  0.027486  0.079623  0.083714  0.038382  0.035102  0.034409   \n",
      "feat_9   0.027424  0.020185  0.015922  0.036116  0.046721  0.005847  0.039806   \n",
      "feat_10  0.106752  0.019069  0.091760  0.113659  0.019042  0.135928  0.029741   \n",
      "feat_11  0.021357  0.008064  0.009866  0.024491  0.561843  0.004795  0.420361   \n",
      "feat_12  0.113921  0.021418  0.026525  0.036666  0.064162  0.052954  0.054571   \n",
      "feat_13  0.189751  0.178263  0.224472  0.233604  0.010431  0.065423  0.010074   \n",
      "feat_14  0.058841  0.007714  0.041530  0.045473  0.104507  0.037389  0.084089   \n",
      "feat_15  0.061486  0.016673  0.027200  0.039033  0.057948  0.019557  0.050702   \n",
      "feat_16  0.015473  0.129884  0.003965  0.045703  0.205393  0.076330  0.168008   \n",
      "feat_17  0.134841  0.099665  0.069539  0.114670  0.015193  0.123831  0.012734   \n",
      "feat_18  0.059279  0.102684  0.042958  0.080390  0.182025  0.135914  0.156176   \n",
      "feat_19  0.179548  0.007951  0.006204  0.012499  0.002391  0.000812  0.006879   \n",
      "feat_20  0.359596  0.017077  0.196400  0.074068  0.025698  0.144703  0.023469   \n",
      "feat_21  0.041637  0.000132  0.027742  0.016071  0.279618  0.004052  0.221444   \n",
      "feat_22  0.060522  0.132590  0.084217  0.093992  0.189565  0.155515  0.177255   \n",
      "feat_23  0.208441  0.010125  0.091219  0.039133  0.037434  0.116575  0.028166   \n",
      "feat_24  0.490795  0.033078  0.223999  0.083032  0.024981  0.084692  0.016468   \n",
      "feat_25  0.080761  0.093776  0.038806  0.092507  0.138356  0.021283  0.114821   \n",
      "feat_26  0.039677  0.001871  0.019367  0.012442  0.206267  0.028376  0.154009   \n",
      "feat_27  0.028521  0.007386  0.020715  0.027651  0.430435  0.004720  0.321354   \n",
      "feat_28  0.029717  0.002422  0.020470  0.013202  0.375019  0.048691  0.404849   \n",
      "feat_29  0.012754  0.612847  0.026352  0.050629  0.008607  0.030520  0.003870   \n",
      "feat_30  0.073491  0.015577  0.025439  0.003858  0.004286  0.095881  0.019710   \n",
      "feat_31  0.027571  0.016217  0.002431  0.006050  0.241754  0.045104  0.191724   \n",
      "feat_32  0.227310  0.053867  0.153970  0.056263  0.014739  0.079886  0.021769   \n",
      "feat_33  0.031347  0.063717  0.027899  0.035588  0.068258  0.019694  0.048967   \n",
      "feat_34  0.033004  0.010891  0.020402  0.012131  0.030723  0.012386  0.031212   \n",
      "feat_35  0.025874  0.014190  0.000374  0.002152  0.274259  0.052736  0.253594   \n",
      "feat_36  0.076780  0.032473  0.152192  0.073767  0.023800  0.021000  0.023039   \n",
      "feat_37  0.201531  0.045286  0.136333  0.045819  0.014978  0.048283  0.007138   \n",
      "feat_38  0.344171  0.025131  0.135069  0.067772  0.109521  0.179604  0.096153   \n",
      "feat_39  0.077277  0.058445  0.018723  0.035878  0.001440  0.081091  0.006898   \n",
      "feat_40  0.070334  0.017066  0.022609  0.043747  0.065813  0.031681  0.068693   \n",
      "feat_41  0.121279  0.072372  0.011902  0.127535  0.003674  0.070796  0.005659   \n",
      "feat_42  0.029785  0.033537  0.016144  0.104605  0.028310  0.044863  0.003465   \n",
      "feat_43  0.022225  0.005309  0.015939  0.024274  0.003795  0.002574  0.009240   \n",
      "feat_44  0.074762  0.072331  0.093156  0.110654  0.179266  0.202801  0.182162   \n",
      "feat_45  0.077474  0.063837  0.025537  0.031569  0.000147  0.026733  0.002928   \n",
      "feat_46  0.033789  0.008934  0.022669  0.031722  0.403677  0.000241  0.363145   \n",
      "feat_47  0.066191  0.008686  0.193238  0.023517  0.010398  0.151613  0.007708   \n",
      "feat_48  0.125008  0.090005  0.074560  0.057616  0.079256  0.006612  0.073208   \n",
      "feat_49  0.047059  0.023213  0.023846  0.041352  0.265545  0.135250  0.343092   \n",
      "feat_50  0.088080  0.005459  0.051061  0.073377  0.001354  0.173281  0.001742   \n",
      "feat_51  0.026127  0.015799  0.028184  0.026622  0.093025  0.107963  0.118212   \n",
      "feat_52  0.012416  0.001000  0.006928  0.006210  0.366544  0.019771  0.189014   \n",
      "feat_53  0.011055  0.111723  0.051046  0.349737  0.020713  0.029944  0.018108   \n",
      "feat_54  0.008459  0.058522  0.019843  0.025234  0.444238  0.082381  0.394613   \n",
      "feat_55  0.136822  0.068623  0.246042  0.106119  0.051735  0.182156  0.053607   \n",
      "feat_56  0.004393  0.005466  0.004568  0.013829  0.043107  0.002544  0.023436   \n",
      "feat_57  0.066284  0.006597  0.052778  0.040919  0.049279  0.084484  0.036242   \n",
      "feat_58  0.365075  0.001763  0.044968  0.023352  0.078906  0.001446  0.031475   \n",
      "feat_59  0.013573  0.048736  0.007320  0.189173  0.024168  0.055354  0.017358   \n",
      "feat_60  0.011742  0.024297  0.002122  0.029188  0.026903  0.055102  0.007344   \n",
      "feat_61  0.020551  0.001620  0.020607  0.019744  0.575101  0.007611  0.474954   \n",
      "feat_62  0.001444  0.214828  0.001336  0.169192  0.027840  0.090282  0.026090   \n",
      "feat_63  0.026610  0.035163  0.006096  0.194618  0.066939  0.088525  0.052736   \n",
      "feat_64  0.000467  0.019363  0.029796  0.020209  0.053117  0.010230  0.047284   \n",
      "feat_65  0.072170  0.115692  0.099054  0.066714  0.062834  0.127931  0.088551   \n",
      "feat_66  0.081866  0.147127  0.058784  0.085545  0.107806  0.110691  0.093782   \n",
      "feat_67  0.323678  0.052970  0.188128  0.152964  0.076209  0.075488  0.070049   \n",
      "feat_68  0.013405  0.001655  0.084630  0.062394  0.004003  0.084468  0.000787   \n",
      "feat_69  0.409691  0.002209  0.044834  0.036435  0.014590  0.019568  0.017440   \n",
      "feat_70  0.122972  0.067065  0.105585  0.144247  0.004856  0.192565  0.009776   \n",
      "feat_71  0.036850  0.002787  0.099693  0.063309  0.029219  0.132951  0.025602   \n",
      "feat_72  0.040593  0.016045  0.022904  0.017522  0.038547  0.002439  0.028420   \n",
      "feat_73  0.220832  0.001634  0.018161  0.015294  0.007285  0.002868  0.005143   \n",
      "feat_74  0.058736  0.009406  0.069628  0.016904  0.174775  0.008959  0.055336   \n",
      "feat_75  0.172375  0.010904  0.013876  0.006885  0.037439  0.011281  0.022992   \n",
      "feat_76  1.000000  0.003596  0.329988  0.073039  0.004267  0.056915  0.002648   \n",
      "feat_77  0.003596  1.000000  0.005717  0.068083  0.009147  0.022957  0.003631   \n",
      "feat_78  0.329988  0.005717  1.000000  0.036415  0.005224  0.047949  0.012932   \n",
      "feat_79  0.073039  0.068083  0.036415  1.000000  0.005482  0.080384  0.003830   \n",
      "feat_80  0.004267  0.009147  0.005224  0.005482  1.000000  0.054154  0.522616   \n",
      "feat_81  0.056915  0.022957  0.047949  0.080384  0.054154  1.000000  0.067345   \n",
      "feat_82  0.002648  0.003631  0.012932  0.003830  0.522616  0.067345  1.000000   \n",
      "feat_83  0.291364  0.004793  0.166186  0.028770  0.018600  0.049550  0.003285   \n",
      "feat_84  0.031277  0.001243  0.035861  0.002540  0.029430  0.096658  0.054646   \n",
      "feat_85  0.000682  0.005602  0.004071  0.004663  0.035876  0.054972  0.034368   \n",
      "feat_86  0.010324  0.020294  0.018797  0.095254  0.081888  0.013808  0.065189   \n",
      "feat_87  0.063411  0.019275  0.063539  0.099579  0.004588  0.084096  0.012153   \n",
      "feat_88  0.050417  0.007396  0.030010  0.018615  0.076250  0.017469  0.059553   \n",
      "feat_89  0.023242  0.021591  0.014639  0.073207  0.350787  0.166234  0.266249   \n",
      "feat_90  0.291884  0.004988  0.043339  0.031099  0.012623  0.009379  0.001795   \n",
      "feat_91  0.175163  0.026376  0.068450  0.021616  0.017815  0.017243  0.014641   \n",
      "feat_92  0.050887  0.076551  0.028596  0.162033  0.063401  0.018565  0.049661   \n",
      "feat_93  0.029426  0.001715  0.016047  0.029082  0.012651  0.019378  0.005497   \n",
      "\n",
      "          feat_83   feat_84   feat_85   feat_86   feat_87   feat_88   feat_89  \\\n",
      "feat_1   0.059165  0.049634  0.008739  0.107947  0.089374  0.020830  0.096851   \n",
      "feat_2   0.371691  0.009845  0.006764  0.039090  0.047451  0.047035  0.105527   \n",
      "feat_3   0.019914  0.011159  0.048626  0.096093  0.009838  0.082336  0.174781   \n",
      "feat_4   0.001052  0.005684  0.033153  0.071029  0.005055  0.067484  0.183715   \n",
      "feat_5   0.008006  0.467329  0.034062  0.013879  0.013999  0.019201  0.119951   \n",
      "feat_6   0.035145  0.177777  0.004290  0.010455  0.015256  0.015437  0.035042   \n",
      "feat_7   0.282069  0.062634  0.037874  0.009169  0.089574  0.033646  0.063511   \n",
      "feat_8   0.033479  0.005064  0.003416  0.029395  0.059929  0.050931  0.007974   \n",
      "feat_9   0.032875  0.013569  0.031462  0.019144  0.016925  0.001160  0.019147   \n",
      "feat_10  0.052025  0.017939  0.086758  0.159447  0.077421  0.054635  0.061498   \n",
      "feat_11  0.000190  0.017724  0.074293  0.123339  0.032969  0.114491  0.137374   \n",
      "feat_12  0.036529  0.009807  0.019283  0.007214  0.016089  0.024324  0.082220   \n",
      "feat_13  0.115747  0.023221  0.002594  0.004850  0.093870  0.036259  0.062990   \n",
      "feat_14  0.070643  0.027058  0.021455  0.145787  0.020229  0.323089  0.038881   \n",
      "feat_15  0.052090  0.009311  0.246847  0.002529  0.023191  0.010840  0.029547   \n",
      "feat_16  0.017558  0.035211  0.110850  0.003610  0.077770  0.007257  0.248364   \n",
      "feat_17  0.113987  0.000334  0.015559  0.049102  0.214221  0.034139  0.035390   \n",
      "feat_18  0.085116  0.028752  0.001555  0.029295  0.126886  0.035981  0.247462   \n",
      "feat_19  0.056721  0.002847  0.008292  0.014560  0.000412  0.018485  0.011116   \n",
      "feat_20  0.364803  0.001723  0.084570  0.016850  0.220475  0.004081  0.111231   \n",
      "feat_21  0.031940  0.004702  0.006180  0.045562  0.016862  0.030401  0.105392   \n",
      "feat_22  0.083639  0.041519  0.044396  0.018347  0.219974  0.045439  0.244779   \n",
      "feat_23  0.181739  0.011832  0.056994  0.121170  0.111837  0.014039  0.059743   \n",
      "feat_24  0.300629  0.091092  0.018990  0.015444  0.123298  0.043479  0.023581   \n",
      "feat_25  0.092796  0.018320  0.021119  0.263924  0.011294  0.207974  0.012866   \n",
      "feat_26  0.001433  0.048907  0.048889  0.072464  0.015937  0.078470  0.094521   \n",
      "feat_27  0.024884  0.002502  0.046053  0.082510  0.028097  0.070194  0.099536   \n",
      "feat_28  0.020382  0.029148  0.039784  0.080806  0.002941  0.074442  0.169794   \n",
      "feat_29  0.015374  0.007335  0.013104  0.011960  0.038800  0.032585  0.055398   \n",
      "feat_30  0.015350  0.716862  0.011743  0.003346  0.024016  0.024413  0.028606   \n",
      "feat_31  0.049387  0.004130  0.012365  0.030215  0.028640  0.030582  0.147243   \n",
      "feat_32  0.112246  0.286689  0.000020  0.055348  0.172447  0.044493  0.014314   \n",
      "feat_33  0.034326  0.005488  0.082534  0.038480  0.030021  0.098167  0.055363   \n",
      "feat_34  0.012072  0.007639  0.047765  0.080038  0.019620  0.084706  0.029953   \n",
      "feat_35  0.035014  0.003819  0.005873  0.041072  0.023173  0.034481  0.192518   \n",
      "feat_36  0.036176  0.011182  0.027171  0.029148  0.109157  0.037274  0.010805   \n",
      "feat_37  0.115052  0.013236  0.160492  0.005594  0.095706  0.015467  0.031177   \n",
      "feat_38  0.215912  0.033801  0.181822  0.047868  0.166591  0.010355  0.263962   \n",
      "feat_39  0.050781  0.003668  0.042419  0.001503  0.070500  0.024618  0.017853   \n",
      "feat_40  0.063668  0.022284  0.002764  0.050871  0.030951  0.162950  0.056494   \n",
      "feat_41  0.056999  0.002849  0.004431  0.000603  0.045929  0.031026  0.100392   \n",
      "feat_42  0.032126  0.011315  0.001051  0.029038  0.145685  0.045588  0.066354   \n",
      "feat_43  0.008312  0.002255  0.415152  0.109338  0.007965  0.158878  0.049449   \n",
      "feat_44  0.084283  0.118820  0.104923  0.049617  0.165350  0.010699  0.205417   \n",
      "feat_45  0.045234  0.002054  0.026840  0.007039  0.051954  0.018612  0.007262   \n",
      "feat_46  0.027402  0.001093  0.049848  0.092302  0.019250  0.077723  0.114977   \n",
      "feat_47  0.037423  0.105327  0.024970  0.033694  0.030387  0.042558  0.139668   \n",
      "feat_48  0.049315  0.006009  0.040097  0.109048  0.099603  0.074914  0.009900   \n",
      "feat_49  0.079155  0.036612  0.006069  0.043704  0.059635  0.045283  0.217931   \n",
      "feat_50  0.065188  0.003742  0.184132  0.132234  0.125120  0.042334  0.076471   \n",
      "feat_51  0.043710  0.079998  0.014491  0.025348  0.081376  0.026219  0.063843   \n",
      "feat_52  0.005840  0.008737  0.010815  0.040474  0.003709  0.041361  0.115156   \n",
      "feat_53  0.006772  0.005698  0.010533  0.076241  0.004038  0.015673  0.081352   \n",
      "feat_54  0.026663  0.026109  0.004352  0.081509  0.057848  0.083348  0.256305   \n",
      "feat_55  0.113962  0.102831  0.025289  0.021798  0.127128  0.033113  0.157031   \n",
      "feat_56  0.005914  0.003549  0.016538  0.072723  0.005218  0.012290  0.053164   \n",
      "feat_57  0.144449  0.043523  0.034281  0.055232  0.102572  0.068615  0.051941   \n",
      "feat_58  0.235803  0.005284  0.015734  0.026106  0.005737  0.045055  0.022821   \n",
      "feat_59  0.011922  0.002065  0.037042  0.148417  0.149499  0.002890  0.013638   \n",
      "feat_60  0.068019  0.020174  0.043094  0.078149  0.099639  0.091812  0.031379   \n",
      "feat_61  0.021269  0.002943  0.047659  0.083301  0.027363  0.070815  0.160405   \n",
      "feat_62  0.022311  0.012320  0.031092  0.006503  0.081320  0.048727  0.246666   \n",
      "feat_63  0.005221  0.013394  0.026954  0.080492  0.056860  0.026337  0.163009   \n",
      "feat_64  0.001432  0.013347  0.009820  0.064883  0.075964  0.117596  0.015218   \n",
      "feat_65  0.057058  0.120772  0.066922  0.030362  0.050138  0.011600  0.125884   \n",
      "feat_66  0.088982  0.033794  0.031291  0.088686  0.206406  0.000679  0.205289   \n",
      "feat_67  0.127444  0.039005  0.020048  0.081445  0.295803  0.058706  0.005220   \n",
      "feat_68  0.000817  0.036198  0.024620  0.038904  0.001672  0.059843  0.125150   \n",
      "feat_69  0.204569  0.007221  0.006954  0.025538  0.027690  0.022918  0.011806   \n",
      "feat_70  0.069588  0.045636  0.361941  0.225792  0.212133  0.140850  0.163631   \n",
      "feat_71  0.026032  0.011858  0.013894  0.015410  0.060004  0.048676  0.076348   \n",
      "feat_72  0.034569  0.011992  0.294384  0.008897  0.013536  0.004066  0.057040   \n",
      "feat_73  0.027131  0.012885  0.010675  0.000841  0.004759  0.026363  0.006704   \n",
      "feat_74  0.024667  0.028021  0.000453  0.015945  0.003992  0.025207  0.042104   \n",
      "feat_75  0.063463  0.028478  0.026329  0.031401  0.001201  0.058630  0.014925   \n",
      "feat_76  0.291364  0.031277  0.000682  0.010324  0.063411  0.050417  0.023242   \n",
      "feat_77  0.004793  0.001243  0.005602  0.020294  0.019275  0.007396  0.021591   \n",
      "feat_78  0.166186  0.035861  0.004071  0.018797  0.063539  0.030010  0.014639   \n",
      "feat_79  0.028770  0.002540  0.004663  0.095254  0.099579  0.018615  0.073207   \n",
      "feat_80  0.018600  0.029430  0.035876  0.081888  0.004588  0.076250  0.350787   \n",
      "feat_81  0.049550  0.096658  0.054972  0.013808  0.084096  0.017469  0.166234   \n",
      "feat_82  0.003285  0.054646  0.034368  0.065189  0.012153  0.059553  0.266249   \n",
      "feat_83  1.000000  0.001047  0.009157  0.029711  0.072006  0.052930  0.035181   \n",
      "feat_84  0.001047  1.000000  0.010210  0.003459  0.013631  0.017903  0.103643   \n",
      "feat_85  0.009157  0.010210  1.000000  0.109643  0.049250  0.027886  0.053582   \n",
      "feat_86  0.029711  0.003459  0.109643  1.000000  0.073685  0.426972  0.011822   \n",
      "feat_87  0.072006  0.013631  0.049250  0.073685  1.000000  0.023053  0.066008   \n",
      "feat_88  0.052930  0.017903  0.027886  0.426972  0.023053  1.000000  0.022552   \n",
      "feat_89  0.035181  0.103643  0.053582  0.011822  0.066008  0.022552  1.000000   \n",
      "feat_90  0.243942  0.006013  0.003931  0.019803  0.014696  0.031679  0.027764   \n",
      "feat_91  0.095801  0.003444  0.023091  0.024005  0.028850  0.033653  0.015917   \n",
      "feat_92  0.018325  0.048431  0.043484  0.049393  0.001424  0.070120  0.129622   \n",
      "feat_93  0.054188  0.003723  0.023390  0.029035  0.499990  0.008631  0.030650   \n",
      "\n",
      "          feat_90   feat_91   feat_92   feat_93  \n",
      "feat_1   0.010310  0.037264  0.054777  0.081783  \n",
      "feat_2   0.515022  0.026383  0.008219  0.054593  \n",
      "feat_3   0.015068  0.012417  0.066921  0.006814  \n",
      "feat_4   0.009454  0.010312  0.087631  0.015746  \n",
      "feat_5   0.004842  0.012012  0.065331  0.002038  \n",
      "feat_6   0.054034  0.012465  0.015479  0.008521  \n",
      "feat_7   0.129578  0.068506  0.032261  0.034912  \n",
      "feat_8   0.026807  0.095990  0.013608  0.005131  \n",
      "feat_9   0.020698  0.014742  0.069707  0.006038  \n",
      "feat_10  0.049908  0.024025  0.006869  0.041316  \n",
      "feat_11  0.045074  0.029511  0.013179  0.003326  \n",
      "feat_12  0.062721  0.063965  0.063922  0.012722  \n",
      "feat_13  0.107722  0.044338  0.071953  0.038989  \n",
      "feat_14  0.060240  0.038444  0.040133  0.018127  \n",
      "feat_15  0.046616  0.034402  0.018206  0.020369  \n",
      "feat_16  0.016863  0.048494  0.210499  0.031467  \n",
      "feat_17  0.045218  0.088508  0.006538  0.056695  \n",
      "feat_18  0.094336  0.037275  0.126640  0.058100  \n",
      "feat_19  0.450925  0.004085  0.027662  0.014243  \n",
      "feat_20  0.370282  0.079181  0.018715  0.110054  \n",
      "feat_21  0.033193  0.019779  0.058008  0.007677  \n",
      "feat_22  0.098595  0.104921  0.200593  0.113276  \n",
      "feat_23  0.141869  0.010438  0.031837  0.084945  \n",
      "feat_24  0.357270  0.090833  0.024375  0.089200  \n",
      "feat_25  0.088187  0.045759  0.030135  0.015708  \n",
      "feat_26  0.021565  0.018447  0.199974  0.016709  \n",
      "feat_27  0.025263  0.018778  0.023790  0.000318  \n",
      "feat_28  0.021330  0.015242  0.122653  0.005275  \n",
      "feat_29  0.000185  0.040526  0.084445  0.008301  \n",
      "feat_30  0.016571  0.016378  0.002712  0.006260  \n",
      "feat_31  0.068371  0.008808  0.022346  0.030656  \n",
      "feat_32  0.122931  0.008934  0.066435  0.130842  \n",
      "feat_33  0.048377  0.003487  0.099836  0.012546  \n",
      "feat_34  0.028500  0.006878  0.076567  0.015377  \n",
      "feat_35  0.099522  0.009463  0.064610  0.024180  \n",
      "feat_36  0.008579  0.066842  0.000525  0.032804  \n",
      "feat_37  0.174533  0.031573  0.011842  0.073010  \n",
      "feat_38  0.356354  0.027772  0.005320  0.083854  \n",
      "feat_39  0.019496  0.033654  0.028631  0.016089  \n",
      "feat_40  0.057816  0.036275  0.031490  0.023675  \n",
      "feat_41  0.064128  0.071538  0.159939  0.007970  \n",
      "feat_42  0.011508  0.020277  0.197315  0.095720  \n",
      "feat_43  0.014281  0.019450  0.040629  0.000640  \n",
      "feat_44  0.080785  0.001806  0.073727  0.105498  \n",
      "feat_45  0.017729  0.033762  0.019553  0.014994  \n",
      "feat_46  0.026419  0.022626  0.049284  0.004403  \n",
      "feat_47  0.013074  0.008806  0.016421  0.000629  \n",
      "feat_48  0.070611  0.047508  0.023148  0.053749  \n",
      "feat_49  0.075120  0.006273  0.037485  0.029427  \n",
      "feat_50  0.080146  0.008383  0.040728  0.052934  \n",
      "feat_51  0.017031  0.000449  0.013779  0.017484  \n",
      "feat_52  0.043089  0.012082  0.026573  0.014812  \n",
      "feat_53  0.016199  0.004092  0.135595  0.006926  \n",
      "feat_54  0.034802  0.018972  0.153724  0.029959  \n",
      "feat_55  0.069305  0.055702  0.049364  0.055275  \n",
      "feat_56  0.008288  0.003705  0.028787  0.007368  \n",
      "feat_57  0.087563  0.010013  0.100979  0.045129  \n",
      "feat_58  0.407341  0.056724  0.014156  0.034472  \n",
      "feat_59  0.011079  0.003928  0.010828  0.047987  \n",
      "feat_60  0.014203  0.011737  0.128371  0.055350  \n",
      "feat_61  0.018112  0.017045  0.042357  0.001258  \n",
      "feat_62  0.000511  0.279845  0.327176  0.012774  \n",
      "feat_63  0.018277  0.009487  0.031464  0.023298  \n",
      "feat_64  0.001930  0.020846  0.071886  0.023064  \n",
      "feat_65  0.029076  0.001188  0.044286  0.015500  \n",
      "feat_66  0.094925  0.098063  0.123694  0.067957  \n",
      "feat_67  0.089262  0.112052  0.011247  0.129018  \n",
      "feat_68  0.023839  0.022515  0.095970  0.004602  \n",
      "feat_69  0.549489  0.041206  0.037961  0.032052  \n",
      "feat_70  0.074178  0.030560  0.007310  0.093488  \n",
      "feat_71  0.019694  0.050622  0.000368  0.002001  \n",
      "feat_72  0.030673  0.008936  0.005300  0.008233  \n",
      "feat_73  0.070001  0.007193  0.024017  0.000163  \n",
      "feat_74  0.055372  0.016941  0.004497  0.021967  \n",
      "feat_75  0.160418  0.002625  0.037710  0.006208  \n",
      "feat_76  0.291884  0.175163  0.050887  0.029426  \n",
      "feat_77  0.004988  0.026376  0.076551  0.001715  \n",
      "feat_78  0.043339  0.068450  0.028596  0.016047  \n",
      "feat_79  0.031099  0.021616  0.162033  0.029082  \n",
      "feat_80  0.012623  0.017815  0.063401  0.012651  \n",
      "feat_81  0.009379  0.017243  0.018565  0.019378  \n",
      "feat_82  0.001795  0.014641  0.049661  0.005497  \n",
      "feat_83  0.243942  0.095801  0.018325  0.054188  \n",
      "feat_84  0.006013  0.003444  0.048431  0.003723  \n",
      "feat_85  0.003931  0.023091  0.043484  0.023390  \n",
      "feat_86  0.019803  0.024005  0.049393  0.029035  \n",
      "feat_87  0.014696  0.028850  0.001424  0.499990  \n",
      "feat_88  0.031679  0.033653  0.070120  0.008631  \n",
      "feat_89  0.027764  0.015917  0.129622  0.030650  \n",
      "feat_90  1.000000  0.014812  0.035311  0.039864  \n",
      "feat_91  0.014812  1.000000  0.104226  0.000045  \n",
      "feat_92  0.035311  0.104226  1.000000  0.003653  \n",
      "feat_93  0.039864  0.000045  0.003653  1.000000  \n"
     ]
    }
   ],
   "source": [
    "# Create correlation matrix\n",
    "corr_matrix = X.corr().abs()\n",
    "print(); print(corr_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "         feat_1    feat_2    feat_3    feat_4    feat_5    feat_6    feat_7  \\\n",
      "feat_1      NaN  0.031332  0.027807  0.027529  0.042973  0.043603  0.298952   \n",
      "feat_2      NaN       NaN  0.082573  0.134987  0.020926  0.041343  0.222386   \n",
      "feat_3      NaN       NaN       NaN  0.583523  0.010880  0.004288  0.001294   \n",
      "feat_4      NaN       NaN       NaN       NaN  0.017290  0.014059  0.014490   \n",
      "feat_5      NaN       NaN       NaN       NaN       NaN  0.145355  0.075047   \n",
      "feat_6      NaN       NaN       NaN       NaN       NaN       NaN  0.088014   \n",
      "feat_7      NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_8      NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_9      NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_10     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_11     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_12     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_13     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_14     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_15     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_16     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_17     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_18     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_19     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_20     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_21     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_22     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_23     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_24     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_25     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_26     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_27     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_28     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_29     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_30     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_31     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_32     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_33     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_34     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_35     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_36     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_37     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_38     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_39     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_40     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_41     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_42     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_43     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_44     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_45     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_46     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_47     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_48     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_49     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_50     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_51     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_52     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_53     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_54     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_55     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_56     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_57     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_58     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_59     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_60     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_61     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_62     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_63     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_64     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_65     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_66     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_67     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_68     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_69     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_70     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_71     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_72     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_73     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_74     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_75     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_76     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93     NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "           feat_8    feat_9   feat_10   feat_11   feat_12   feat_13   feat_14  \\\n",
      "feat_1   0.056321  0.032285  0.097776  0.042928  0.056934  0.139254  0.063517   \n",
      "feat_2   0.019815  0.025630  0.051925  0.118534  0.090153  0.157467  0.070057   \n",
      "feat_3   0.053462  0.063551  0.036944  0.596243  0.050037  0.013870  0.111105   \n",
      "feat_4   0.046184  0.046250  0.059514  0.389409  0.057434  0.028973  0.099215   \n",
      "feat_5   0.035861  0.024708  0.091324  0.004882  0.036668  0.059081  0.037607   \n",
      "feat_6   0.012867  0.009373  0.041940  0.014504  0.028588  0.036293  0.027350   \n",
      "feat_7   0.038121  0.027146  0.194258  0.012418  0.056230  0.199142  0.044671   \n",
      "feat_8        NaN  0.039281  0.000023  0.065923  0.091424  0.095365  0.061799   \n",
      "feat_9        NaN       NaN  0.024323  0.075820  0.021885  0.040164  0.110188   \n",
      "feat_10       NaN       NaN       NaN  0.006010  0.048969  0.086682  0.029598   \n",
      "feat_11       NaN       NaN       NaN       NaN  0.053391  0.005500  0.151635   \n",
      "feat_12       NaN       NaN       NaN       NaN       NaN  0.115288  0.021695   \n",
      "feat_13       NaN       NaN       NaN       NaN       NaN       NaN  0.021555   \n",
      "feat_14       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_15       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_16       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_17       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_18       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_19       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_20       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_21       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_22       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_23       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_24       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_25       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_26       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_27       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_28       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_29       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_30       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_31       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_32       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_33       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_34       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_35       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_36       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_37       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_38       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_39       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_40       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_41       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_42       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_43       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_44       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_45       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_46       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_47       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_48       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_49       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_50       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_51       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_52       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_53       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_54       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_55       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_56       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_57       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_58       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_59       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_60       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_61       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_62       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_63       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_64       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_65       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_66       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_67       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_68       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_69       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_70       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_71       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_72       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_73       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_74       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_75       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_76       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_15   feat_16   feat_17   feat_18       feat_19   feat_20  \\\n",
      "feat_1   0.045738  0.027086  0.053004  0.084856  2.302499e-03  0.070511   \n",
      "feat_2   0.048798  0.108046  0.074902  0.242716  1.766549e-01  0.449160   \n",
      "feat_3   0.065285  0.221426  0.023093  0.115655  1.222845e-02  0.011069   \n",
      "feat_4   0.051222  0.211078  0.007554  0.214895  3.519107e-07  0.044657   \n",
      "feat_5   0.007000  0.062877  0.062197  0.052186  8.555966e-03  0.046200   \n",
      "feat_6   0.018328  0.021934  0.015488  0.048710  3.849262e-02  0.057813   \n",
      "feat_7   0.035721  0.043957  0.127245  0.098972  5.807104e-02  0.364972   \n",
      "feat_8   0.056960  0.004659  0.173912  0.087777  1.938742e-02  0.062595   \n",
      "feat_9   0.009858  0.082664  0.028709  0.043642  1.671680e-04  0.023397   \n",
      "feat_10  0.021700  0.063997  0.092959  0.071635  9.015307e-03  0.176373   \n",
      "feat_11  0.100255  0.134924  0.031807  0.079620  3.367902e-02  0.005641   \n",
      "feat_12  0.028407  0.113804  0.031434  0.100066  3.474800e-02  0.071224   \n",
      "feat_13  0.032725  0.087992  0.319352  0.177439  4.295357e-02  0.194816   \n",
      "feat_14  0.039565  0.018640  0.040148  0.076035  4.274776e-02  0.060775   \n",
      "feat_15       NaN  0.344157  0.045991  0.044733  3.202887e-02  0.034214   \n",
      "feat_16       NaN       NaN  0.014567  0.244849  1.011014e-03  0.088638   \n",
      "feat_17       NaN       NaN       NaN  0.154259  3.731379e-02  0.261954   \n",
      "feat_18       NaN       NaN       NaN       NaN  3.652979e-02  0.204010   \n",
      "feat_19       NaN       NaN       NaN       NaN           NaN  0.105428   \n",
      "feat_20       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_21       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_22       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_23       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_24       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_25       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_26       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_27       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_28       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_29       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_30       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_31       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_32       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_33       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_34       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_35       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_36       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_37       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_38       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_39       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_40       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_41       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_42       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_43       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_44       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_45       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_46       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_47       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_48       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_49       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_50       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_51       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_52       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_53       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_54       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_55       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_56       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_57       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_58       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_59       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_60       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_61       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_62       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_63       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_64       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_65       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_66       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_67       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_68       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_69       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_70       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_71       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_72       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_73       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_74       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_75       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_76       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN           NaN       NaN   \n",
      "\n",
      "          feat_21   feat_22   feat_23   feat_24   feat_25   feat_26   feat_27  \\\n",
      "feat_1   0.027026  0.063283  0.048686  0.067255  0.187237  0.022813  0.038826   \n",
      "feat_2   0.014113  0.215106  0.162065  0.253684  0.096366  0.064856  0.037841   \n",
      "feat_3   0.354925  0.251082  0.002427  0.031596  0.157459  0.268112  0.508370   \n",
      "feat_4   0.232923  0.247738  0.030622  0.003728  0.134231  0.365757  0.308629   \n",
      "feat_5   0.003288  0.075161  0.017281  0.075222  0.003610  0.025116  0.002098   \n",
      "feat_6   0.008046  0.038939  0.043651  0.082124  0.023319  0.004680  0.001943   \n",
      "feat_7   0.022908  0.162620  0.186462  0.244813  0.048820  0.008782  0.015429   \n",
      "feat_8   0.041095  0.029032  0.012774  0.161848  0.036939  0.041599  0.050272   \n",
      "feat_9   0.028409  0.062348  0.006940  0.073618  0.025279  0.066414  0.042531   \n",
      "feat_10  0.005134  0.141405  0.096666  0.081684  0.009792  0.003721  0.001551   \n",
      "feat_11  0.324080  0.154861  0.026500  0.064213  0.206207  0.191095  0.599484   \n",
      "feat_12  0.048215  0.119511  0.070561  0.085558  0.012281  0.042363  0.041377   \n",
      "feat_13  0.005092  0.209427  0.056463  0.192249  0.006019  0.016335  0.012657   \n",
      "feat_14  0.005645  0.063087  0.046284  0.118920  0.557929  0.101571  0.093146   \n",
      "feat_15  0.086686  0.056194  0.029100  0.062806  0.008228  0.056035  0.045859   \n",
      "feat_16  0.199338  0.433154  0.037233  0.016491  0.039487  0.187627  0.118580   \n",
      "feat_17  0.040070  0.157098  0.133157  0.257106  0.034996  0.004430  0.036484   \n",
      "feat_18  0.043272  0.364125  0.098466  0.134752  0.063139  0.234748  0.043358   \n",
      "feat_19  0.024122  0.019088  0.100028  0.290475  0.065112  0.021169  0.018954   \n",
      "feat_20  0.037885  0.244537  0.225089  0.344270  0.070439  0.004728  0.038583   \n",
      "feat_21       NaN  0.118657  0.006174  0.041367  0.046066  0.113374  0.376264   \n",
      "feat_22       NaN       NaN  0.084268  0.088580  0.047433  0.234156  0.112099   \n",
      "feat_23       NaN       NaN       NaN  0.211602  0.033920  0.006505  0.002618   \n",
      "feat_24       NaN       NaN       NaN       NaN  0.127708  0.037106  0.017972   \n",
      "feat_25       NaN       NaN       NaN       NaN       NaN  0.124594  0.130455   \n",
      "feat_26       NaN       NaN       NaN       NaN       NaN       NaN  0.141360   \n",
      "feat_27       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_28       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_29       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_30       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_31       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_32       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_33       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_34       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_35       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_36       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_37       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_38       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_39       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_40       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_41       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_42       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_43       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_44       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_45       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_46       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_47       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_48       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_49       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_50       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_51       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_52       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_53       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_54       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_55       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_56       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_57       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_58       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_59       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_60       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_61       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_62       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_63       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_64       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_65       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_66       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_67       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_68       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_69       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_70       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_71       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_72       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_73       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_74       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_75       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_76       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_28   feat_29   feat_30   feat_31   feat_32   feat_33   feat_34  \\\n",
      "feat_1   0.030257  0.069266  0.033108  0.011210  0.061361  0.049454  0.042756   \n",
      "feat_2   0.072494  0.025689  0.026896  0.193216  0.087699  0.033927  0.026717   \n",
      "feat_3   0.551398  0.004141  0.007667  0.138548  0.053753  0.078520  0.039765   \n",
      "feat_4   0.486417  0.014271  0.000873  0.351263  0.036730  0.054819  0.029087   \n",
      "feat_5   0.047688  0.065957  0.318117  0.013188  0.122957  0.006046  0.001032   \n",
      "feat_6   0.017132  0.002389  0.196493  0.017112  0.103699  0.019617  0.010199   \n",
      "feat_7   0.000998  0.046231  0.050535  0.032291  0.149537  0.008628  0.022144   \n",
      "feat_8   0.036668  0.104985  0.009574  0.018342  0.017910  0.045822  0.015314   \n",
      "feat_9   0.055545  0.021328  0.015830  0.016227  0.568417  0.045937  0.050045   \n",
      "feat_10  0.022349  0.068243  0.012623  0.038503  0.028812  0.003292  0.030941   \n",
      "feat_11  0.422271  0.021024  0.008832  0.109761  0.002610  0.130168  0.068741   \n",
      "feat_12  0.053406  0.063105  0.013299  0.056052  0.017584  0.048145  0.018258   \n",
      "feat_13  0.015961  0.177028  0.042503  0.024556  0.072300  0.048620  0.004043   \n",
      "feat_14  0.097257  0.027883  0.028994  0.052896  0.141330  0.258664  0.094934   \n",
      "feat_15  0.044554  0.022266  0.007801  0.029179  0.029320  0.273124  0.049157   \n",
      "feat_16  0.235443  0.142485  0.025419  0.125445  0.025200  0.239344  0.047032   \n",
      "feat_17  0.020020  0.135787  0.023838  0.028863  0.100666  0.049150  0.009384   \n",
      "feat_18  0.163886  0.118086  0.018357  0.234090  0.018512  0.033070  0.020361   \n",
      "feat_19  0.018858  0.010786  0.006245  0.032551  0.115506  0.040924  0.022311   \n",
      "feat_20  0.003058  0.059671  0.015289  0.123438  0.162073  0.043688  0.038096   \n",
      "feat_21  0.255093  0.014585  0.004026  0.090587  0.041779  0.027701  0.026739   \n",
      "feat_22  0.253812  0.235942  0.015912  0.148829  0.023853  0.083295  0.043086   \n",
      "feat_23  0.003651  0.006970  0.042710  0.123996  0.133619  0.020804  0.017839   \n",
      "feat_24  0.021632  0.039872  0.133829  0.060356  0.406661  0.042170  0.017694   \n",
      "feat_25  0.126680  0.033676  0.021807  0.067072  0.062620  0.228878  0.116002   \n",
      "feat_26  0.348072  0.028692  0.025203  0.163779  0.042307  0.061111  0.034864   \n",
      "feat_27  0.335252  0.018366  0.005345  0.094907  0.034932  0.076713  0.040586   \n",
      "feat_28       NaN  0.004385  0.006272  0.158226  0.037948  0.051453  0.031705   \n",
      "feat_29       NaN       NaN  0.023337  0.003476  0.052170  0.041168  0.011341   \n",
      "feat_30       NaN       NaN       NaN  0.007509  0.357488  0.019171  0.011542   \n",
      "feat_31       NaN       NaN       NaN       NaN  0.006080  0.024233  0.017199   \n",
      "feat_32       NaN       NaN       NaN       NaN       NaN  0.069877  0.057669   \n",
      "feat_33       NaN       NaN       NaN       NaN       NaN       NaN  0.011244   \n",
      "feat_34       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_35       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_36       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_37       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_38       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_39       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_40       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_41       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_42       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_43       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_44       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_45       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_46       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_47       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_48       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_49       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_50       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_51       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_52       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_53       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_54       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_55       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_56       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_57       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_58       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_59       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_60       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_61       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_62       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_63       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_64       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_65       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_66       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_67       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_68       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_69       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_70       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_71       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_72       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_73       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_74       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_75       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_76       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_35   feat_36   feat_37   feat_38   feat_39   feat_40   feat_41  \\\n",
      "feat_1   0.010667  0.095475  0.082306  0.104666  0.006898  0.053603  0.080960   \n",
      "feat_2   0.293374  0.026988  0.124475  0.373022  0.020626  0.057890  0.072120   \n",
      "feat_3   0.233923  0.034305  0.029425  0.046097  0.016279  0.094511  0.028920   \n",
      "feat_4   0.554474  0.029660  0.010646  0.092488  0.002700  0.076805  0.009977   \n",
      "feat_5   0.027013  0.043504  0.035320  0.070100  0.000710  0.025652  0.038764   \n",
      "feat_6   0.015739  0.014673  0.043213  0.078088  0.033470  0.018090  0.028930   \n",
      "feat_7   0.043001  0.079760  0.141960  0.315177  0.053230  0.039712  0.097971   \n",
      "feat_8   0.024043  0.606707  0.119417  0.034287  0.071908  0.044847  0.075854   \n",
      "feat_9   0.026879  0.035874  0.024868  0.028766  0.013354  0.085491  0.023446   \n",
      "feat_10  0.051793  0.007687  0.187069  0.177524  0.042223  0.038610  0.078533   \n",
      "feat_11  0.108392  0.047595  0.011404  0.055483  0.013556  0.133424  0.038465   \n",
      "feat_12  0.077834  0.063250  0.080605  0.097602  0.008417  0.016455  0.064854   \n",
      "feat_13  0.050881  0.128211  0.077847  0.191271  0.015840  0.036971  0.123191   \n",
      "feat_14  0.063796  0.042550  0.038171  0.061308  0.001582  0.487804  0.029650   \n",
      "feat_15  0.033329  0.058118  0.040783  0.001730  0.003620  0.069358  0.026326   \n",
      "feat_16  0.208736  0.012829  0.040773  0.193696  0.043227  0.287290  0.079613   \n",
      "feat_17  0.026671  0.199284  0.079853  0.136263  0.061983  0.071674  0.153653   \n",
      "feat_18  0.380595  0.076773  0.062521  0.248949  0.058047  0.024125  0.104176   \n",
      "feat_19  0.040617  0.010993  0.067629  0.246811  0.006468  0.042588  0.025658   \n",
      "feat_20  0.153165  0.079299  0.196200  0.489655  0.085462  0.062804  0.190383   \n",
      "feat_21  0.093071  0.036644  0.006567  0.014056  0.015944  0.041815  0.018627   \n",
      "feat_22  0.255825  0.044453  0.065120  0.253535  0.052726  0.018002  0.128977   \n",
      "feat_23  0.075955  0.018374  0.117174  0.229581  0.097809  0.032760  0.021231   \n",
      "feat_24  0.049722  0.193335  0.186706  0.298310  0.181969  0.130764  0.110562   \n",
      "feat_25  0.083226  0.034228  0.030234  0.054273  0.024830  0.417171  0.023498   \n",
      "feat_26  0.270053  0.030216  0.036926  0.021670  0.015289  0.067956  0.010296   \n",
      "feat_27  0.082270  0.035222  0.008916  0.000735  0.018293  0.072274  0.032546   \n",
      "feat_28  0.270286  0.029080  0.017669  0.043109  0.009859  0.068621  0.010145   \n",
      "feat_29  0.012444  0.082159  0.026360  0.041284  0.002196  0.022948  0.154367   \n",
      "feat_30  0.005045  0.033858  0.047492  0.036073  0.006347  0.029433  0.020869   \n",
      "feat_31  0.556203  0.006242  0.033776  0.192466  0.037898  0.019174  0.009391   \n",
      "feat_32  0.020074  0.055867  0.096995  0.141960  0.117487  0.132667  0.039299   \n",
      "feat_33  0.018519  0.020444  0.006702  0.044232  0.020577  0.283048  0.138983   \n",
      "feat_34  0.022267  0.036960  0.030082  0.041901  0.020456  0.063950  0.017078   \n",
      "feat_35       NaN  0.006591  0.022481  0.213707  0.014495  0.032130  0.013647   \n",
      "feat_36       NaN       NaN  0.048980  0.033062  0.000932  0.047362  0.080966   \n",
      "feat_37       NaN       NaN       NaN  0.208764  0.221596  0.015906  0.075154   \n",
      "feat_38       NaN       NaN       NaN       NaN  0.158922  0.021407  0.145597   \n",
      "feat_39       NaN       NaN       NaN       NaN       NaN  0.021638  0.012608   \n",
      "feat_40       NaN       NaN       NaN       NaN       NaN       NaN  0.018424   \n",
      "feat_41       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_42       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_43       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_44       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_45       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_46       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_47       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_48       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_49       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_50       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_51       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_52       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_53       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_54       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_55       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_56       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_57       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_58       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_59       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_60       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_61       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_62       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_63       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_64       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_65       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_66       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_67       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_68       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_69       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_70       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_71       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_72       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_73       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_74       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_75       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_76       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_42   feat_43   feat_44   feat_45   feat_46   feat_47   feat_48  \\\n",
      "feat_1   0.073611  0.008884  0.090439  0.008601  0.040647  0.026463  0.130564   \n",
      "feat_2   0.043863  0.017600  0.208450  0.018259  0.042491  0.014574  0.023310   \n",
      "feat_3   0.036605  0.014068  0.242901  0.007984  0.777517  0.009560  0.111987   \n",
      "feat_4   0.071397  0.000893  0.238665  0.002809  0.436274  0.012914  0.085788   \n",
      "feat_5   0.023272  0.048320  0.177137  0.004808  0.003583  0.111428  0.013228   \n",
      "feat_6   0.020709  0.015515  0.065082  0.025133  0.001892  0.018736  0.013193   \n",
      "feat_7   0.001220  0.010259  0.213066  0.036262  0.012643  0.177082  0.056163   \n",
      "feat_8   0.006578  0.004214  0.001733  0.076982  0.050246  0.015016  0.104416   \n",
      "feat_9   0.055850  0.046033  0.062023  0.013405  0.053175  0.027254  0.077385   \n",
      "feat_10  0.024055  0.111706  0.222353  0.009960  0.013546  0.046108  0.008297   \n",
      "feat_11  0.002356  0.027301  0.172523  0.006148  0.579272  0.004880  0.125103   \n",
      "feat_12  0.028966  0.007790  0.091486  0.004838  0.042124  0.018930  0.035591   \n",
      "feat_13  0.036274  0.026543  0.162659  0.011546  0.008579  0.083185  0.089556   \n",
      "feat_14  0.054232  0.073237  0.061226  0.015501  0.099296  0.059736  0.252006   \n",
      "feat_15  0.045392  0.010929  0.043123  0.018245  0.054633  0.043214  0.051239   \n",
      "feat_16  0.142924  0.046106  0.353285  0.041837  0.179348  0.038439  0.030415   \n",
      "feat_17  0.045206  0.018660  0.085796  0.028681  0.033466  0.017601  0.102710   \n",
      "feat_18  0.229794  0.007073  0.265708  0.038576  0.059749  0.072757  0.019632   \n",
      "feat_19  0.024769  0.014469  0.042021  0.006416  0.018800  0.012064  0.038394   \n",
      "feat_20  0.039543  0.043772  0.239186  0.046258  0.032060  0.103850  0.049880   \n",
      "feat_21  0.009244  0.001126  0.094507  0.011902  0.381266  0.019792  0.050541   \n",
      "feat_22  0.238647  0.009844  0.416059  0.034295  0.171096  0.135443  0.002619   \n",
      "feat_23  0.001777  0.039410  0.115325  0.079246  0.011156  0.059817  0.061294   \n",
      "feat_24  0.015382  0.047241  0.086296  0.150948  0.034435  0.079616  0.192535   \n",
      "feat_25  0.012705  0.084195  0.027719  0.014604  0.142185  0.029371  0.255795   \n",
      "feat_26  0.378864  0.023106  0.177793  0.007335  0.219206  0.002630  0.078122   \n",
      "feat_27  0.011435  0.017084  0.113467  0.008254  0.572678  0.012920  0.095155   \n",
      "feat_28  0.103324  0.019070  0.229184  0.002959  0.467904  0.008369  0.087888   \n",
      "feat_29  0.065757  0.017998  0.103415  0.001680  0.014711  0.021081  0.073798   \n",
      "feat_30  0.006329  0.001860  0.110357  0.007040  0.008163  0.028166  0.022090   \n",
      "feat_31  0.037239  0.055828  0.147659  0.022507  0.088787  0.001308  0.025759   \n",
      "feat_32  0.030480  0.033969  0.076621  0.083419  0.050622  0.049702  0.158067   \n",
      "feat_33  0.012912  0.011619  0.025225  0.015520  0.076527  0.026461  0.083435   \n",
      "feat_34  0.040423  0.036566  0.051081  0.019361  0.037194  0.016706  0.019349   \n",
      "feat_35  0.108742  0.057303  0.240699  0.006098  0.125971  0.003474  0.044776   \n",
      "feat_36  0.004834  0.038372  0.014352  0.001001  0.037652  0.014814  0.140863   \n",
      "feat_37  0.001908  0.210933  0.094774  0.226044  0.029782  0.033234  0.035257   \n",
      "feat_38  0.043037  0.116811  0.287322  0.121745  0.005029  0.073738  0.037572   \n",
      "feat_39  0.003142  0.016880  0.068093  0.824146  0.017705  0.009785  0.069612   \n",
      "feat_40  0.049079  0.068687  0.010021  0.012797  0.079144  0.042989  0.115672   \n",
      "feat_41  0.051125  0.024384  0.064104  0.004404  0.033962  0.022343  0.031844   \n",
      "feat_42       NaN  0.005319  0.211003  0.008854  0.017923  0.017745  0.012537   \n",
      "feat_43       NaN       NaN  0.080740  0.017203  0.014968  0.029769  0.051727   \n",
      "feat_44       NaN       NaN       NaN  0.042672  0.168826  0.154684  0.012963   \n",
      "feat_45       NaN       NaN       NaN       NaN  0.007301  0.001732  0.071787   \n",
      "feat_46       NaN       NaN       NaN       NaN       NaN  0.017981  0.109526   \n",
      "feat_47       NaN       NaN       NaN       NaN       NaN       NaN  0.015034   \n",
      "feat_48       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_49       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_50       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_51       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_52       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_53       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_54       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_55       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_56       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_57       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_58       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_59       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_60       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_61       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_62       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_63       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_64       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_65       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_66       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_67       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_68       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_69       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_70       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_71       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_72       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_73       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_74       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_75       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_76       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_49   feat_50   feat_51   feat_52   feat_53   feat_54   feat_55  \\\n",
      "feat_1   0.011179  0.040114  0.000902  0.013082  0.156400  0.001399  0.165044   \n",
      "feat_2   0.311278  0.071664  0.085345  0.105528  0.008015  0.179173  0.134617   \n",
      "feat_3   0.242692  0.016765  0.031979  0.169111  0.024222  0.694048  0.055445   \n",
      "feat_4   0.300011  0.005343  0.041647  0.209517  0.016308  0.525456  0.070077   \n",
      "feat_5   0.054311  0.024087  0.060318  0.003930  0.040202  0.055299  0.142221   \n",
      "feat_6   0.022864  0.027150  0.006902  0.030929  0.005914  0.031379  0.062765   \n",
      "feat_7   0.124320  0.140433  0.036493  0.006189  0.048544  0.079607  0.343752   \n",
      "feat_8   0.019114  0.001231  0.009513  0.028689  0.018674  0.025421  0.063415   \n",
      "feat_9   0.042280  0.009925  0.013181  0.009792  0.030075  0.078499  0.042138   \n",
      "feat_10  0.079449  0.252350  0.014012  0.007879  0.082810  0.074637  0.213003   \n",
      "feat_11  0.250991  0.019520  0.040465  0.272738  0.052811  0.494180  0.022723   \n",
      "feat_12  0.073520  0.041978  0.016439  0.068520  0.018344  0.102486  0.081504   \n",
      "feat_13  0.075534  0.043774  0.021930  0.009553  0.161234  0.119941  0.223364   \n",
      "feat_14  0.070134  0.014922  0.041262  0.060335  0.006808  0.112305  0.055732   \n",
      "feat_15  0.040979  0.020619  0.024504  0.022071  0.006163  0.034175  0.040297   \n",
      "feat_16  0.227799  0.058954  0.024675  0.093180  0.086820  0.468766  0.131218   \n",
      "feat_17  0.049887  0.081685  0.076780  0.004599  0.018729  0.033721  0.163377   \n",
      "feat_18  0.346609  0.074758  0.114754  0.107743  0.042412  0.272898  0.183218   \n",
      "feat_19  0.019607  0.056299  0.004868  0.017126  0.013756  0.000260  0.018046   \n",
      "feat_20  0.194854  0.240263  0.073081  0.033363  0.028131  0.096978  0.254142   \n",
      "feat_21  0.093235  0.020847  0.020342  0.230007  0.002909  0.318332  0.003326   \n",
      "feat_22  0.307542  0.122040  0.056819  0.081117  0.089161  0.502168  0.339893   \n",
      "feat_23  0.090079  0.152904  0.085160  0.071561  0.003534  0.036249  0.087014   \n",
      "feat_24  0.045497  0.098027  0.053253  0.064581  0.021068  0.016601  0.149115   \n",
      "feat_25  0.088119  0.001294  0.043517  0.073940  0.149136  0.136167  0.010211   \n",
      "feat_26  0.186126  0.020735  0.040346  0.110586  0.015321  0.375083  0.066164   \n",
      "feat_27  0.130413  0.027077  0.037532  0.254923  0.033497  0.399163  0.004804   \n",
      "feat_28  0.262104  0.012332  0.052675  0.152144  0.016824  0.540238  0.061357   \n",
      "feat_29  0.024018  0.011039  0.012631  0.007411  0.147231  0.084520  0.273112   \n",
      "feat_30  0.039086  0.002592  0.038955  0.017820  0.003433  0.007768  0.068682   \n",
      "feat_31  0.234752  0.052848  0.058908  0.243705  0.010286  0.201311  0.042039   \n",
      "feat_32  0.011588  0.125057  0.024916  0.032964  0.058391  0.025977  0.083579   \n",
      "feat_33  0.026104  0.008562  0.025569  0.040093  0.077197  0.002876  0.014856   \n",
      "feat_34  0.033517  0.024418  0.009592  0.029361  0.028083  0.057204  0.034814   \n",
      "feat_35  0.400371  0.037632  0.054189  0.212755  0.010598  0.338959  0.069446   \n",
      "feat_36  0.001935  0.002601  0.012007  0.014003  0.012986  0.004917  0.067892   \n",
      "feat_37  0.028267  0.099837  0.028112  0.052793  0.003284  0.007759  0.087968   \n",
      "feat_38  0.232639  0.290446  0.038608  0.082314  0.024398  0.177940  0.194682   \n",
      "feat_39  0.040913  0.220654  0.005479  0.012768  0.007054  0.018700  0.045411   \n",
      "feat_40  0.050584  0.013975  0.033740  0.020696  0.018717  0.027177  0.046788   \n",
      "feat_41  0.033257  0.044417  0.009971  0.005032  0.130334  0.037902  0.115234   \n",
      "feat_42  0.117745  0.003906  0.013406  0.017937  0.037325  0.162022  0.110436   \n",
      "feat_43  0.026975  0.056477  0.007465  0.026134  0.022839  0.001902  0.016999   \n",
      "feat_44  0.388665  0.188333  0.047933  0.085201  0.071061  0.436212  0.284885   \n",
      "feat_45  0.031328  0.075324  0.001455  0.009299  0.007686  0.018264  0.034931   \n",
      "feat_46  0.146562  0.028279  0.053718  0.166742  0.029967  0.569484  0.020171   \n",
      "feat_47  0.032821  0.033893  0.039878  0.004416  0.009543  0.044086  0.263623   \n",
      "feat_48  0.028316  0.067493  0.009996  0.034030  0.088261  0.085181  0.045665   \n",
      "feat_49       NaN  0.072909  0.095188  0.105964  0.002183  0.377186  0.134802   \n",
      "feat_50       NaN       NaN  0.009406  0.005567  0.023457  0.038062  0.105915   \n",
      "feat_51       NaN       NaN       NaN  0.052590  0.008589  0.041812  0.077797   \n",
      "feat_52       NaN       NaN       NaN       NaN  0.013411  0.186552  0.014103   \n",
      "feat_53       NaN       NaN       NaN       NaN       NaN  0.051162  0.172746   \n",
      "feat_54       NaN       NaN       NaN       NaN       NaN       NaN  0.189662   \n",
      "feat_55       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_56       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_57       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_58       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_59       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_60       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_61       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_62       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_63       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_64       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_65       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_66       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_67       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_68       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_69       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_70       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_71       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_72       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_73       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_74       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_75       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_76       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_56   feat_57   feat_58   feat_59   feat_60   feat_61   feat_62  \\\n",
      "feat_1   0.015738  0.014598  0.028696  0.139364  0.020267  0.038124  0.091636   \n",
      "feat_2   0.018908  0.104263  0.371146  0.013283  0.018219  0.063873  0.029040   \n",
      "feat_3   0.016288  0.018887  0.034506  0.021717  0.011369  0.494881  0.010733   \n",
      "feat_4   0.019618  0.069462  0.039103  0.015802  0.068900  0.354402  0.025228   \n",
      "feat_5   0.001560  0.037162  0.001546  0.038893  0.008796  0.001559  0.073215   \n",
      "feat_6   0.004014  0.036655  0.110743  0.001285  0.022289  0.002227  0.022874   \n",
      "feat_7   0.003145  0.073969  0.106203  0.009624  0.012336  0.006284  0.025485   \n",
      "feat_8   0.002285  0.011968  0.025674  0.118223  0.020328  0.045203  0.214717   \n",
      "feat_9   0.022639  0.049335  0.028290  0.021451  0.066333  0.044343  0.021138   \n",
      "feat_10  0.009786  0.017525  0.020933  0.051189  0.000619  0.004397  0.054733   \n",
      "feat_11  0.033221  0.007302  0.179361  0.039646  0.025990  0.543858  0.051554   \n",
      "feat_12  0.007665  0.042518  0.082574  0.026684  0.027027  0.050080  0.126326   \n",
      "feat_13  0.002066  0.075397  0.132569  0.174173  0.046147  0.009951  0.164920   \n",
      "feat_14  0.004436  0.101382  0.059363  0.026693  0.118618  0.093809  0.026822   \n",
      "feat_15  0.022519  0.069910  0.044920  0.017773  0.079152  0.056169  0.097920   \n",
      "feat_16  0.021978  0.049373  0.018927  0.046633  0.043111  0.150952  0.337177   \n",
      "feat_17  0.012085  0.173169  0.040370  0.058797  0.090963  0.030037  0.105504   \n",
      "feat_18  0.014345  0.284435  0.048980  0.099073  0.234885  0.084745  0.197339   \n",
      "feat_19  0.008824  0.026194  0.234503  0.008378  0.002039  0.014633  0.014714   \n",
      "feat_20  0.012953  0.136076  0.310481  0.020746  0.036649  0.027135  0.082477   \n",
      "feat_21  0.026285  0.022876  0.004761  0.024209  0.032751  0.298307  0.037588   \n",
      "feat_22  0.003457  0.185503  0.057172  0.043127  0.166049  0.136120  0.319551   \n",
      "feat_23  0.013547  0.105362  0.110441  0.004395  0.045739  0.005186  0.005936   \n",
      "feat_24  0.016847  0.208961  0.370922  0.041787  0.108238  0.018301  0.009847   \n",
      "feat_25  0.014930  0.110428  0.091991  0.089081  0.131640  0.132584  0.077595   \n",
      "feat_26  0.007013  0.264678  0.004414  0.024493  0.344495  0.165298  0.005080   \n",
      "feat_27  0.022714  0.015006  0.046311  0.028375  0.026060  0.500029  0.029421   \n",
      "feat_28  0.009810  0.062168  0.015941  0.009859  0.070262  0.393108  0.030228   \n",
      "feat_29  0.001742  0.042796  0.003237  0.024546  0.039505  0.013307  0.252822   \n",
      "feat_30  0.000284  0.021861  0.010901  0.004690  0.006984  0.002622  0.002875   \n",
      "feat_31  0.045396  0.116651  0.045329  0.014443  0.104792  0.120430  0.025281   \n",
      "feat_32  0.016834  0.012744  0.121410  0.007351  0.018205  0.028499  0.006367   \n",
      "feat_33  0.015381  0.036934  0.024569  0.042400  0.072889  0.071736  0.238590   \n",
      "feat_34  0.030780  0.017135  0.030054  0.021977  0.043871  0.036210  0.002139   \n",
      "feat_35  0.045052  0.133482  0.066752  0.007550  0.099420  0.143230  0.046613   \n",
      "feat_36  0.018979  0.021359  0.017621  0.085643  0.007885  0.032847  0.161397   \n",
      "feat_37  0.065605  0.027198  0.110513  0.047061  0.016753  0.020982  0.028921   \n",
      "feat_38  0.035271  0.086578  0.279557  0.028466  0.019733  0.023578  0.140945   \n",
      "feat_39  0.011425  0.007114  0.025101  0.007327  0.002745  0.018494  0.000659   \n",
      "feat_40  0.013689  0.088847  0.057899  0.028204  0.099869  0.079754  0.084960   \n",
      "feat_41  0.009312  0.029363  0.139469  0.069765  0.026557  0.026785  0.241020   \n",
      "feat_42  0.003905  0.271951  0.022257  0.038613  0.332167  0.001546  0.081785   \n",
      "feat_43  0.176251  0.034859  0.014288  0.020581  0.039480  0.016680  0.016113   \n",
      "feat_44  0.010393  0.111068  0.044159  0.109311  0.081152  0.128809  0.116861   \n",
      "feat_45  0.008955  0.003581  0.028467  0.004403  0.000943  0.009329  0.001492   \n",
      "feat_46  0.014555  0.005103  0.031832  0.028600  0.003954  0.502794  0.015610   \n",
      "feat_47  0.013103  0.053890  0.013860  0.010021  0.016341  0.012562  0.046582   \n",
      "feat_48  0.030145  0.000509  0.042172  0.086275  0.023404  0.086675  0.049943   \n",
      "feat_49  0.019829  0.125088  0.057051  0.001289  0.073149  0.189569  0.050396   \n",
      "feat_50  0.011528  0.018667  0.051836  0.024369  0.005157  0.024572  0.008057   \n",
      "feat_51  0.005882  0.197049  0.014615  0.003985  0.083940  0.054041  0.001125   \n",
      "feat_52  0.045495  0.028884  0.063267  0.018623  0.010198  0.253560  0.002664   \n",
      "feat_53  0.008639  0.020143  0.010057  0.113257  0.013433  0.028828  0.254328   \n",
      "feat_54  0.024859  0.108515  0.055111  0.009751  0.097616  0.418198  0.144185   \n",
      "feat_55  0.006551  0.148330  0.055781  0.054114  0.087929  0.019997  0.139948   \n",
      "feat_56       NaN  0.014550  0.005532  0.010537  0.022879  0.022271  0.003510   \n",
      "feat_57       NaN       NaN  0.038391  0.009761  0.471676  0.000989  0.003693   \n",
      "feat_58       NaN       NaN       NaN  0.015981  0.019522  0.031040  0.005153   \n",
      "feat_59       NaN       NaN       NaN       NaN  0.008088  0.026870  0.145860   \n",
      "feat_60       NaN       NaN       NaN       NaN       NaN  0.010940  0.006826   \n",
      "feat_61       NaN       NaN       NaN       NaN       NaN       NaN  0.012790   \n",
      "feat_62       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_63       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_64       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_65       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_66       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_67       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_68       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_69       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_70       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_71       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_72       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_73       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_74       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_75       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_76       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_63   feat_64   feat_65   feat_66   feat_67   feat_68   feat_69  \\\n",
      "feat_1   0.069799  0.010499  0.110041  0.053010  0.154301  0.014674  0.007544   \n",
      "feat_2   0.037020  0.005354  0.078801  0.175620  0.068667  0.012802  0.307406   \n",
      "feat_3   0.031241  0.065105  0.065492  0.088017  0.110081  0.030992  0.032748   \n",
      "feat_4   0.059903  0.047288  0.062285  0.129654  0.080457  0.020092  0.014461   \n",
      "feat_5   0.082383  0.021017  0.228349  0.048364  0.061964  0.107405  0.003294   \n",
      "feat_6   0.017050  0.002764  0.066867  0.033285  0.038289  0.021619  0.074836   \n",
      "feat_7   0.039585  0.011165  0.202346  0.122660  0.148598  0.040309  0.131430   \n",
      "feat_8   0.148673  0.003194  0.025544  0.115175  0.320949  0.075384  0.046258   \n",
      "feat_9   0.013012  0.702951  0.038163  0.001778  0.176921  0.012192  0.029335   \n",
      "feat_10  0.090421  0.022536  0.182756  0.100722  0.043117  0.001693  0.077354   \n",
      "feat_11  0.013436  0.082868  0.041014  0.050764  0.114944  0.040588  0.010697   \n",
      "feat_12  0.038777  0.012603  0.121550  0.084591  0.079377  0.014873  0.057074   \n",
      "feat_13  0.070109  0.003286  0.108567  0.122272  0.211335  0.026881  0.086611   \n",
      "feat_14  0.020837  0.066800  0.011517  0.081709  0.086404  0.076979  0.058352   \n",
      "feat_15  0.018947  0.003848  0.034954  0.018487  0.052486  0.049076  0.052924   \n",
      "feat_16  0.104440  0.043076  0.182735  0.289082  0.027728  0.015351  0.021699   \n",
      "feat_17  0.032391  0.021551  0.072733  0.165143  0.306513  0.038816  0.100656   \n",
      "feat_18  0.119569  0.001922  0.114910  0.285667  0.108521  0.078124  0.071926   \n",
      "feat_19  0.012208  0.016986  0.015797  0.030787  0.035876  0.005129  0.397621   \n",
      "feat_20  0.047287  0.058862  0.143265  0.215576  0.199391  0.011770  0.348327   \n",
      "feat_21  0.015895  0.035863  0.044702  0.022228  0.075966  0.016256  0.044378   \n",
      "feat_22  0.105657  0.017547  0.180556  0.382385  0.061315  0.036537  0.045054   \n",
      "feat_23  0.018242  0.029366  0.066825  0.114657  0.115559  0.008228  0.143339   \n",
      "feat_24  0.035863  0.099968  0.074939  0.116615  0.435017  0.096420  0.419143   \n",
      "feat_25  0.023574  0.009117  0.027847  0.035200  0.001504  0.032516  0.091005   \n",
      "feat_26  0.025983  0.060043  0.047372  0.119656  0.059968  0.016605  0.042414   \n",
      "feat_27  0.010165  0.056230  0.022504  0.025894  0.097900  0.020866  0.037023   \n",
      "feat_28  0.037190  0.056711  0.067139  0.117781  0.080253  0.000733  0.037170   \n",
      "feat_29  0.043699  0.017184  0.158624  0.201489  0.065628  0.000229  0.002000   \n",
      "feat_30  0.013605  0.016485  0.116697  0.019762  0.082248  0.022408  0.009822   \n",
      "feat_31  0.042732  0.018383  0.046102  0.087548  0.018839  0.008982  0.039048   \n",
      "feat_32  0.034355  0.457911  0.057767  0.083371  0.378091  0.022240  0.112176   \n",
      "feat_33  0.027324  0.030341  0.065155  0.056500  0.040761  0.031524  0.038998   \n",
      "feat_34  0.022741  0.052562  0.029832  0.052581  0.046257  0.015014  0.026018   \n",
      "feat_35  0.055321  0.025843  0.058910  0.158498  0.038446  0.005791  0.032932   \n",
      "feat_36  0.108435  0.028963  0.031601  0.119379  0.429543  0.057658  0.017015   \n",
      "feat_37  0.063563  0.012852  0.075276  0.077615  0.124339  0.002614  0.122514   \n",
      "feat_38  0.101778  0.019928  0.160865  0.244178  0.128729  0.030065  0.301431   \n",
      "feat_39  0.011183  0.012954  0.025084  0.038623  0.108151  0.005995  0.030387   \n",
      "feat_40  0.007991  0.073709  0.034486  0.019922  0.110168  0.053138  0.065651   \n",
      "feat_41  0.059253  0.005579  0.071459  0.116906  0.166986  0.066465  0.084362   \n",
      "feat_42  0.035641  0.033987  0.062684  0.147385  0.025730  0.014355  0.035006   \n",
      "feat_43  0.011440  0.023057  0.064798  0.000381  0.080169  0.037935  0.025677   \n",
      "feat_44  0.110515  0.021038  0.295100  0.245185  0.052369  0.037033  0.033513   \n",
      "feat_45  0.005855  0.009361  0.017275  0.028869  0.081525  0.010629  0.014862   \n",
      "feat_46  0.015763  0.059347  0.035316  0.046867  0.107314  0.030122  0.036424   \n",
      "feat_47  0.023698  0.003333  0.079686  0.120077  0.053932  0.244780  0.007611   \n",
      "feat_48  0.036264  0.102866  0.019454  0.015508  0.407350  0.004573  0.041509   \n",
      "feat_49  0.076030  0.024424  0.150720  0.218145  0.021548  0.001992  0.031680   \n",
      "feat_50  0.064145  0.081781  0.105297  0.061823  0.082293  0.001275  0.072500   \n",
      "feat_51  0.034871  0.010992  0.037801  0.039897  0.023045  0.030563  0.012447   \n",
      "feat_52  0.025127  0.015069  0.049985  0.042590  0.027666  0.004855  0.010622   \n",
      "feat_53  0.059414  0.030776  0.098366  0.062345  0.036349  0.028684  0.012123   \n",
      "feat_54  0.077873  0.058250  0.159901  0.252159  0.059267  0.006572  0.007706   \n",
      "feat_55  0.105285  0.004749  0.217883  0.196955  0.122626  0.056319  0.062175   \n",
      "feat_56  0.005410  0.018979  0.008946  0.010837  0.030794  0.002327  0.002325   \n",
      "feat_57  0.017826  0.008348  0.045171  0.114737  0.078598  0.039628  0.073636   \n",
      "feat_58  0.010079  0.001655  0.019621  0.059694  0.134576  0.022746  0.392839   \n",
      "feat_59  0.181146  0.006894  0.060893  0.053516  0.398837  0.020244  0.003780   \n",
      "feat_60  0.012982  0.055960  0.025368  0.073185  0.037814  0.021800  0.014202   \n",
      "feat_61  0.021335  0.056514  0.040793  0.044301  0.091471  0.020848  0.030713   \n",
      "feat_62  0.140069  0.017422  0.104989  0.342523  0.131518  0.092759  0.011587   \n",
      "feat_63       NaN  0.038846  0.069119  0.142239  0.117070  0.019061  0.014252   \n",
      "feat_64       NaN       NaN  0.015070  0.106675  0.226275  0.019150  0.004913   \n",
      "feat_65       NaN       NaN       NaN  0.120274  0.041425  0.021648  0.015448   \n",
      "feat_66       NaN       NaN       NaN       NaN  0.142036  0.044337  0.079321   \n",
      "feat_67       NaN       NaN       NaN       NaN       NaN  0.078876  0.117455   \n",
      "feat_68       NaN       NaN       NaN       NaN       NaN       NaN  0.017943   \n",
      "feat_69       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_70       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_71       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_72       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_73       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_74       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_75       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_76       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_70   feat_71   feat_72   feat_73   feat_74   feat_75   feat_76  \\\n",
      "feat_1   0.165442  0.013712  0.029983  0.140815  0.051365  0.011596  0.153808   \n",
      "feat_2   0.112968  0.002336  0.023267  0.039192  0.070724  0.093689  0.259360   \n",
      "feat_3   0.018774  0.053020  0.045339  0.013972  0.041559  0.044724  0.028670   \n",
      "feat_4   0.020798  0.042413  0.029796  0.011285  0.049097  0.031454  0.013792   \n",
      "feat_5   0.118510  0.056428  0.005177  0.001609  0.017265  0.015279  0.035570   \n",
      "feat_6   0.052401  0.011901  0.011090  0.025023  0.043160  0.006951  0.073867   \n",
      "feat_7   0.237907  0.115813  0.014921  0.022819  0.053059  0.039865  0.375114   \n",
      "feat_8   0.023089  0.081664  0.029868  0.028999  0.000431  0.031466  0.081682   \n",
      "feat_9   0.056205  0.043286  0.058147  0.022679  0.007594  0.027313  0.027424   \n",
      "feat_10  0.322857  0.104834  0.004225  0.000240  0.008912  0.003828  0.106752   \n",
      "feat_11  0.050540  0.071717  0.086496  0.006893  0.186561  0.034513  0.021357   \n",
      "feat_12  0.071630  0.028629  0.031888  0.181833  0.087012  0.052536  0.113921   \n",
      "feat_13  0.145294  0.052337  0.008543  0.030927  0.021072  0.022496  0.189751   \n",
      "feat_14  0.054226  0.076935  0.022358  0.040175  0.033498  0.079099  0.058841   \n",
      "feat_15  0.118795  0.033807  0.764664  0.019194  0.013632  0.048894  0.061486   \n",
      "feat_16  0.255740  0.001594  0.394998  0.026285  0.034125  0.059511  0.015473   \n",
      "feat_17  0.108789  0.150286  0.023103  0.002826  0.001770  0.025506  0.134841   \n",
      "feat_18  0.169238  0.074981  0.002190  0.008668  0.038278  0.011790  0.059279   \n",
      "feat_19  0.031171  0.020871  0.022584  0.056837  0.022614  0.036357  0.179548   \n",
      "feat_20  0.316436  0.098633  0.012246  0.029960  0.033726  0.053898  0.359596   \n",
      "feat_21  0.007589  0.040361  0.072693  0.009870  0.123215  0.047657  0.041637   \n",
      "feat_22  0.231848  0.072454  0.113100  0.012546  0.040243  0.020839  0.060522   \n",
      "feat_23  0.137402  0.056965  0.012941  0.098657  0.085407  0.026762  0.208441   \n",
      "feat_24  0.098407  0.123829  0.050412  0.048055  0.113057  0.156134  0.490795   \n",
      "feat_25  0.126664  0.012005  0.015754  0.041218  0.040263  0.059513  0.080761   \n",
      "feat_26  0.012608  0.031770  0.046341  0.018681  0.036063  0.022976  0.039677   \n",
      "feat_27  0.048459  0.050763  0.043237  0.005898  0.134597  0.040042  0.028521   \n",
      "feat_28  0.001360  0.038653  0.023556  0.013072  0.043981  0.039447  0.029717   \n",
      "feat_29  0.079299  0.050342  0.010273  0.001577  0.005190  0.016475  0.012754   \n",
      "feat_30  0.053508  0.026446  0.006538  0.073560  0.044089  0.121927  0.073491   \n",
      "feat_31  0.076045  0.005662  0.016424  0.003859  0.128573  0.008119  0.027571   \n",
      "feat_32  0.076867  0.073278  0.060626  0.049874  0.069656  0.100877  0.227310   \n",
      "feat_33  0.119803  0.002155  0.303326  0.031310  0.024346  0.047442  0.031347   \n",
      "feat_34  0.065941  0.003020  0.056553  0.011981  0.020616  0.041175  0.033004   \n",
      "feat_35  0.090731  0.024095  0.014162  0.001981  0.073560  0.009315  0.025874   \n",
      "feat_36  0.021627  0.060492  0.035706  0.018678  0.021245  0.051641  0.076780   \n",
      "feat_37  0.184344  0.042454  0.022233  0.169925  0.086901  0.145731  0.201531   \n",
      "feat_38  0.409175  0.060743  0.060632  0.023959  0.048949  0.046876  0.344171   \n",
      "feat_39  0.105113  0.058650  0.000943  0.012927  0.022546  0.008305  0.077277   \n",
      "feat_40  0.090232  0.068913  0.102698  0.034949  0.002428  0.082554  0.070334   \n",
      "feat_41  0.126442  0.053532  0.003314  0.004248  0.007097  0.071319  0.121279   \n",
      "feat_42  0.072523  0.004829  0.024687  0.022221  0.007210  0.032157  0.029785   \n",
      "feat_43  0.321611  0.032762  0.026895  0.012752  0.022302  0.034776  0.022225   \n",
      "feat_44  0.384838  0.073183  0.098485  0.005774  0.036903  0.014398  0.074762   \n",
      "feat_45  0.066091  0.022953  0.008630  0.018758  0.022145  0.015586  0.077474   \n",
      "feat_46  0.040795  0.049396  0.042077  0.010842  0.047543  0.043645  0.033789   \n",
      "feat_47  0.054501  0.392560  0.031692  0.013389  0.002744  0.004454  0.066191   \n",
      "feat_48  0.049905  0.030908  0.036547  0.084806  0.016403  0.050548  0.125008   \n",
      "feat_49  0.139793  0.007692  0.009323  0.001989  0.034674  0.006816  0.047059   \n",
      "feat_50  0.347168  0.104623  0.016511  0.009222  0.016972  0.015501  0.088080   \n",
      "feat_51  0.016843  0.027463  0.019303  0.005856  0.022998  0.013396  0.026127   \n",
      "feat_52  0.001866  0.018361  0.024320  0.007225  0.554737  0.012264  0.012416   \n",
      "feat_53  0.104217  0.013776  0.017736  0.008826  0.000711  0.011755  0.011055   \n",
      "feat_54  0.127314  0.019349  0.064084  0.016801  0.066009  0.050655  0.008459   \n",
      "feat_55  0.195079  0.185377  0.004484  0.001851  0.016692  0.039229  0.136822   \n",
      "feat_56  0.089528  0.020002  0.019436  0.000268  0.032422  0.020258  0.004393   \n",
      "feat_57  0.028737  0.034841  0.055874  0.006254  0.010981  0.056871  0.066284   \n",
      "feat_58  0.038319  0.020289  0.024774  0.057573  0.058768  0.053835  0.365075   \n",
      "feat_59  0.229503  0.015826  0.014535  0.015477  0.002264  0.000676  0.013573   \n",
      "feat_60  0.008525  0.020767  0.062609  0.019131  0.006299  0.006871  0.011742   \n",
      "feat_61  0.039559  0.046490  0.042920  0.006029  0.104516  0.036555  0.020551   \n",
      "feat_62  0.159988  0.089270  0.162735  0.020436  0.001716  0.044457  0.001444   \n",
      "feat_63  0.141850  0.049447  0.030482  0.000880  0.013859  0.004057  0.026610   \n",
      "feat_64  0.029886  0.058239  0.039423  0.018469  0.007032  0.011456  0.000467   \n",
      "feat_65  0.237573  0.045051  0.067476  0.006393  0.050069  0.014663  0.072170   \n",
      "feat_66  0.212931  0.072601  0.037809  0.006615  0.021841  0.005674  0.081866   \n",
      "feat_67  0.127927  0.131528  0.035209  0.098574  0.027789  0.121735  0.323678   \n",
      "feat_68  0.008038  0.196309  0.035915  0.004141  0.008749  0.008292  0.013405   \n",
      "feat_69  0.077678  0.000955  0.032479  0.045667  0.034456  0.065643  0.409691   \n",
      "feat_70       NaN  0.088064  0.192594  0.002807  0.010436  0.016054  0.122972   \n",
      "feat_71       NaN       NaN  0.015731  0.006121  0.012457  0.025972  0.036850   \n",
      "feat_72       NaN       NaN       NaN  0.020547  0.024074  0.043639  0.040593   \n",
      "feat_73       NaN       NaN       NaN       NaN  0.035633  0.190639  0.220832   \n",
      "feat_74       NaN       NaN       NaN       NaN       NaN  0.002297  0.058736   \n",
      "feat_75       NaN       NaN       NaN       NaN       NaN       NaN  0.172375   \n",
      "feat_76       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_77       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_78       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_79       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_80       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_77   feat_78   feat_79   feat_80   feat_81   feat_82   feat_83  \\\n",
      "feat_1   0.123752  0.279202  0.228912  0.013303  0.032427  0.026085  0.059165   \n",
      "feat_2   0.014911  0.094256  0.033668  0.155768  0.052101  0.119109  0.371691   \n",
      "feat_3   0.001584  0.021979  0.020566  0.442036  0.013089  0.438458  0.019914   \n",
      "feat_4   0.015318  0.014499  0.010835  0.405772  0.028284  0.436541  0.001052   \n",
      "feat_5   0.030462  0.070709  0.055115  0.026223  0.129333  0.057400  0.008006   \n",
      "feat_6   0.006501  0.061250  0.009942  0.017648  0.044136  0.014907  0.035145   \n",
      "feat_7   0.005769  0.567084  0.066753  0.028860  0.144308  0.022059  0.282069   \n",
      "feat_8   0.027486  0.079623  0.083714  0.038382  0.035102  0.034409  0.033479   \n",
      "feat_9   0.020185  0.015922  0.036116  0.046721  0.005847  0.039806  0.032875   \n",
      "feat_10  0.019069  0.091760  0.113659  0.019042  0.135928  0.029741  0.052025   \n",
      "feat_11  0.008064  0.009866  0.024491  0.561843  0.004795  0.420361  0.000190   \n",
      "feat_12  0.021418  0.026525  0.036666  0.064162  0.052954  0.054571  0.036529   \n",
      "feat_13  0.178263  0.224472  0.233604  0.010431  0.065423  0.010074  0.115747   \n",
      "feat_14  0.007714  0.041530  0.045473  0.104507  0.037389  0.084089  0.070643   \n",
      "feat_15  0.016673  0.027200  0.039033  0.057948  0.019557  0.050702  0.052090   \n",
      "feat_16  0.129884  0.003965  0.045703  0.205393  0.076330  0.168008  0.017558   \n",
      "feat_17  0.099665  0.069539  0.114670  0.015193  0.123831  0.012734  0.113987   \n",
      "feat_18  0.102684  0.042958  0.080390  0.182025  0.135914  0.156176  0.085116   \n",
      "feat_19  0.007951  0.006204  0.012499  0.002391  0.000812  0.006879  0.056721   \n",
      "feat_20  0.017077  0.196400  0.074068  0.025698  0.144703  0.023469  0.364803   \n",
      "feat_21  0.000132  0.027742  0.016071  0.279618  0.004052  0.221444  0.031940   \n",
      "feat_22  0.132590  0.084217  0.093992  0.189565  0.155515  0.177255  0.083639   \n",
      "feat_23  0.010125  0.091219  0.039133  0.037434  0.116575  0.028166  0.181739   \n",
      "feat_24  0.033078  0.223999  0.083032  0.024981  0.084692  0.016468  0.300629   \n",
      "feat_25  0.093776  0.038806  0.092507  0.138356  0.021283  0.114821  0.092796   \n",
      "feat_26  0.001871  0.019367  0.012442  0.206267  0.028376  0.154009  0.001433   \n",
      "feat_27  0.007386  0.020715  0.027651  0.430435  0.004720  0.321354  0.024884   \n",
      "feat_28  0.002422  0.020470  0.013202  0.375019  0.048691  0.404849  0.020382   \n",
      "feat_29  0.612847  0.026352  0.050629  0.008607  0.030520  0.003870  0.015374   \n",
      "feat_30  0.015577  0.025439  0.003858  0.004286  0.095881  0.019710  0.015350   \n",
      "feat_31  0.016217  0.002431  0.006050  0.241754  0.045104  0.191724  0.049387   \n",
      "feat_32  0.053867  0.153970  0.056263  0.014739  0.079886  0.021769  0.112246   \n",
      "feat_33  0.063717  0.027899  0.035588  0.068258  0.019694  0.048967  0.034326   \n",
      "feat_34  0.010891  0.020402  0.012131  0.030723  0.012386  0.031212  0.012072   \n",
      "feat_35  0.014190  0.000374  0.002152  0.274259  0.052736  0.253594  0.035014   \n",
      "feat_36  0.032473  0.152192  0.073767  0.023800  0.021000  0.023039  0.036176   \n",
      "feat_37  0.045286  0.136333  0.045819  0.014978  0.048283  0.007138  0.115052   \n",
      "feat_38  0.025131  0.135069  0.067772  0.109521  0.179604  0.096153  0.215912   \n",
      "feat_39  0.058445  0.018723  0.035878  0.001440  0.081091  0.006898  0.050781   \n",
      "feat_40  0.017066  0.022609  0.043747  0.065813  0.031681  0.068693  0.063668   \n",
      "feat_41  0.072372  0.011902  0.127535  0.003674  0.070796  0.005659  0.056999   \n",
      "feat_42  0.033537  0.016144  0.104605  0.028310  0.044863  0.003465  0.032126   \n",
      "feat_43  0.005309  0.015939  0.024274  0.003795  0.002574  0.009240  0.008312   \n",
      "feat_44  0.072331  0.093156  0.110654  0.179266  0.202801  0.182162  0.084283   \n",
      "feat_45  0.063837  0.025537  0.031569  0.000147  0.026733  0.002928  0.045234   \n",
      "feat_46  0.008934  0.022669  0.031722  0.403677  0.000241  0.363145  0.027402   \n",
      "feat_47  0.008686  0.193238  0.023517  0.010398  0.151613  0.007708  0.037423   \n",
      "feat_48  0.090005  0.074560  0.057616  0.079256  0.006612  0.073208  0.049315   \n",
      "feat_49  0.023213  0.023846  0.041352  0.265545  0.135250  0.343092  0.079155   \n",
      "feat_50  0.005459  0.051061  0.073377  0.001354  0.173281  0.001742  0.065188   \n",
      "feat_51  0.015799  0.028184  0.026622  0.093025  0.107963  0.118212  0.043710   \n",
      "feat_52  0.001000  0.006928  0.006210  0.366544  0.019771  0.189014  0.005840   \n",
      "feat_53  0.111723  0.051046  0.349737  0.020713  0.029944  0.018108  0.006772   \n",
      "feat_54  0.058522  0.019843  0.025234  0.444238  0.082381  0.394613  0.026663   \n",
      "feat_55  0.068623  0.246042  0.106119  0.051735  0.182156  0.053607  0.113962   \n",
      "feat_56  0.005466  0.004568  0.013829  0.043107  0.002544  0.023436  0.005914   \n",
      "feat_57  0.006597  0.052778  0.040919  0.049279  0.084484  0.036242  0.144449   \n",
      "feat_58  0.001763  0.044968  0.023352  0.078906  0.001446  0.031475  0.235803   \n",
      "feat_59  0.048736  0.007320  0.189173  0.024168  0.055354  0.017358  0.011922   \n",
      "feat_60  0.024297  0.002122  0.029188  0.026903  0.055102  0.007344  0.068019   \n",
      "feat_61  0.001620  0.020607  0.019744  0.575101  0.007611  0.474954  0.021269   \n",
      "feat_62  0.214828  0.001336  0.169192  0.027840  0.090282  0.026090  0.022311   \n",
      "feat_63  0.035163  0.006096  0.194618  0.066939  0.088525  0.052736  0.005221   \n",
      "feat_64  0.019363  0.029796  0.020209  0.053117  0.010230  0.047284  0.001432   \n",
      "feat_65  0.115692  0.099054  0.066714  0.062834  0.127931  0.088551  0.057058   \n",
      "feat_66  0.147127  0.058784  0.085545  0.107806  0.110691  0.093782  0.088982   \n",
      "feat_67  0.052970  0.188128  0.152964  0.076209  0.075488  0.070049  0.127444   \n",
      "feat_68  0.001655  0.084630  0.062394  0.004003  0.084468  0.000787  0.000817   \n",
      "feat_69  0.002209  0.044834  0.036435  0.014590  0.019568  0.017440  0.204569   \n",
      "feat_70  0.067065  0.105585  0.144247  0.004856  0.192565  0.009776  0.069588   \n",
      "feat_71  0.002787  0.099693  0.063309  0.029219  0.132951  0.025602  0.026032   \n",
      "feat_72  0.016045  0.022904  0.017522  0.038547  0.002439  0.028420  0.034569   \n",
      "feat_73  0.001634  0.018161  0.015294  0.007285  0.002868  0.005143  0.027131   \n",
      "feat_74  0.009406  0.069628  0.016904  0.174775  0.008959  0.055336  0.024667   \n",
      "feat_75  0.010904  0.013876  0.006885  0.037439  0.011281  0.022992  0.063463   \n",
      "feat_76  0.003596  0.329988  0.073039  0.004267  0.056915  0.002648  0.291364   \n",
      "feat_77       NaN  0.005717  0.068083  0.009147  0.022957  0.003631  0.004793   \n",
      "feat_78       NaN       NaN  0.036415  0.005224  0.047949  0.012932  0.166186   \n",
      "feat_79       NaN       NaN       NaN  0.005482  0.080384  0.003830  0.028770   \n",
      "feat_80       NaN       NaN       NaN       NaN  0.054154  0.522616  0.018600   \n",
      "feat_81       NaN       NaN       NaN       NaN       NaN  0.067345  0.049550   \n",
      "feat_82       NaN       NaN       NaN       NaN       NaN       NaN  0.003285   \n",
      "feat_83       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_84       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_85       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_86       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_87       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_84   feat_85   feat_86   feat_87   feat_88   feat_89   feat_90  \\\n",
      "feat_1   0.049634  0.008739  0.107947  0.089374  0.020830  0.096851  0.010310   \n",
      "feat_2   0.009845  0.006764  0.039090  0.047451  0.047035  0.105527  0.515022   \n",
      "feat_3   0.011159  0.048626  0.096093  0.009838  0.082336  0.174781  0.015068   \n",
      "feat_4   0.005684  0.033153  0.071029  0.005055  0.067484  0.183715  0.009454   \n",
      "feat_5   0.467329  0.034062  0.013879  0.013999  0.019201  0.119951  0.004842   \n",
      "feat_6   0.177777  0.004290  0.010455  0.015256  0.015437  0.035042  0.054034   \n",
      "feat_7   0.062634  0.037874  0.009169  0.089574  0.033646  0.063511  0.129578   \n",
      "feat_8   0.005064  0.003416  0.029395  0.059929  0.050931  0.007974  0.026807   \n",
      "feat_9   0.013569  0.031462  0.019144  0.016925  0.001160  0.019147  0.020698   \n",
      "feat_10  0.017939  0.086758  0.159447  0.077421  0.054635  0.061498  0.049908   \n",
      "feat_11  0.017724  0.074293  0.123339  0.032969  0.114491  0.137374  0.045074   \n",
      "feat_12  0.009807  0.019283  0.007214  0.016089  0.024324  0.082220  0.062721   \n",
      "feat_13  0.023221  0.002594  0.004850  0.093870  0.036259  0.062990  0.107722   \n",
      "feat_14  0.027058  0.021455  0.145787  0.020229  0.323089  0.038881  0.060240   \n",
      "feat_15  0.009311  0.246847  0.002529  0.023191  0.010840  0.029547  0.046616   \n",
      "feat_16  0.035211  0.110850  0.003610  0.077770  0.007257  0.248364  0.016863   \n",
      "feat_17  0.000334  0.015559  0.049102  0.214221  0.034139  0.035390  0.045218   \n",
      "feat_18  0.028752  0.001555  0.029295  0.126886  0.035981  0.247462  0.094336   \n",
      "feat_19  0.002847  0.008292  0.014560  0.000412  0.018485  0.011116  0.450925   \n",
      "feat_20  0.001723  0.084570  0.016850  0.220475  0.004081  0.111231  0.370282   \n",
      "feat_21  0.004702  0.006180  0.045562  0.016862  0.030401  0.105392  0.033193   \n",
      "feat_22  0.041519  0.044396  0.018347  0.219974  0.045439  0.244779  0.098595   \n",
      "feat_23  0.011832  0.056994  0.121170  0.111837  0.014039  0.059743  0.141869   \n",
      "feat_24  0.091092  0.018990  0.015444  0.123298  0.043479  0.023581  0.357270   \n",
      "feat_25  0.018320  0.021119  0.263924  0.011294  0.207974  0.012866  0.088187   \n",
      "feat_26  0.048907  0.048889  0.072464  0.015937  0.078470  0.094521  0.021565   \n",
      "feat_27  0.002502  0.046053  0.082510  0.028097  0.070194  0.099536  0.025263   \n",
      "feat_28  0.029148  0.039784  0.080806  0.002941  0.074442  0.169794  0.021330   \n",
      "feat_29  0.007335  0.013104  0.011960  0.038800  0.032585  0.055398  0.000185   \n",
      "feat_30  0.716862  0.011743  0.003346  0.024016  0.024413  0.028606  0.016571   \n",
      "feat_31  0.004130  0.012365  0.030215  0.028640  0.030582  0.147243  0.068371   \n",
      "feat_32  0.286689  0.000020  0.055348  0.172447  0.044493  0.014314  0.122931   \n",
      "feat_33  0.005488  0.082534  0.038480  0.030021  0.098167  0.055363  0.048377   \n",
      "feat_34  0.007639  0.047765  0.080038  0.019620  0.084706  0.029953  0.028500   \n",
      "feat_35  0.003819  0.005873  0.041072  0.023173  0.034481  0.192518  0.099522   \n",
      "feat_36  0.011182  0.027171  0.029148  0.109157  0.037274  0.010805  0.008579   \n",
      "feat_37  0.013236  0.160492  0.005594  0.095706  0.015467  0.031177  0.174533   \n",
      "feat_38  0.033801  0.181822  0.047868  0.166591  0.010355  0.263962  0.356354   \n",
      "feat_39  0.003668  0.042419  0.001503  0.070500  0.024618  0.017853  0.019496   \n",
      "feat_40  0.022284  0.002764  0.050871  0.030951  0.162950  0.056494  0.057816   \n",
      "feat_41  0.002849  0.004431  0.000603  0.045929  0.031026  0.100392  0.064128   \n",
      "feat_42  0.011315  0.001051  0.029038  0.145685  0.045588  0.066354  0.011508   \n",
      "feat_43  0.002255  0.415152  0.109338  0.007965  0.158878  0.049449  0.014281   \n",
      "feat_44  0.118820  0.104923  0.049617  0.165350  0.010699  0.205417  0.080785   \n",
      "feat_45  0.002054  0.026840  0.007039  0.051954  0.018612  0.007262  0.017729   \n",
      "feat_46  0.001093  0.049848  0.092302  0.019250  0.077723  0.114977  0.026419   \n",
      "feat_47  0.105327  0.024970  0.033694  0.030387  0.042558  0.139668  0.013074   \n",
      "feat_48  0.006009  0.040097  0.109048  0.099603  0.074914  0.009900  0.070611   \n",
      "feat_49  0.036612  0.006069  0.043704  0.059635  0.045283  0.217931  0.075120   \n",
      "feat_50  0.003742  0.184132  0.132234  0.125120  0.042334  0.076471  0.080146   \n",
      "feat_51  0.079998  0.014491  0.025348  0.081376  0.026219  0.063843  0.017031   \n",
      "feat_52  0.008737  0.010815  0.040474  0.003709  0.041361  0.115156  0.043089   \n",
      "feat_53  0.005698  0.010533  0.076241  0.004038  0.015673  0.081352  0.016199   \n",
      "feat_54  0.026109  0.004352  0.081509  0.057848  0.083348  0.256305  0.034802   \n",
      "feat_55  0.102831  0.025289  0.021798  0.127128  0.033113  0.157031  0.069305   \n",
      "feat_56  0.003549  0.016538  0.072723  0.005218  0.012290  0.053164  0.008288   \n",
      "feat_57  0.043523  0.034281  0.055232  0.102572  0.068615  0.051941  0.087563   \n",
      "feat_58  0.005284  0.015734  0.026106  0.005737  0.045055  0.022821  0.407341   \n",
      "feat_59  0.002065  0.037042  0.148417  0.149499  0.002890  0.013638  0.011079   \n",
      "feat_60  0.020174  0.043094  0.078149  0.099639  0.091812  0.031379  0.014203   \n",
      "feat_61  0.002943  0.047659  0.083301  0.027363  0.070815  0.160405  0.018112   \n",
      "feat_62  0.012320  0.031092  0.006503  0.081320  0.048727  0.246666  0.000511   \n",
      "feat_63  0.013394  0.026954  0.080492  0.056860  0.026337  0.163009  0.018277   \n",
      "feat_64  0.013347  0.009820  0.064883  0.075964  0.117596  0.015218  0.001930   \n",
      "feat_65  0.120772  0.066922  0.030362  0.050138  0.011600  0.125884  0.029076   \n",
      "feat_66  0.033794  0.031291  0.088686  0.206406  0.000679  0.205289  0.094925   \n",
      "feat_67  0.039005  0.020048  0.081445  0.295803  0.058706  0.005220  0.089262   \n",
      "feat_68  0.036198  0.024620  0.038904  0.001672  0.059843  0.125150  0.023839   \n",
      "feat_69  0.007221  0.006954  0.025538  0.027690  0.022918  0.011806  0.549489   \n",
      "feat_70  0.045636  0.361941  0.225792  0.212133  0.140850  0.163631  0.074178   \n",
      "feat_71  0.011858  0.013894  0.015410  0.060004  0.048676  0.076348  0.019694   \n",
      "feat_72  0.011992  0.294384  0.008897  0.013536  0.004066  0.057040  0.030673   \n",
      "feat_73  0.012885  0.010675  0.000841  0.004759  0.026363  0.006704  0.070001   \n",
      "feat_74  0.028021  0.000453  0.015945  0.003992  0.025207  0.042104  0.055372   \n",
      "feat_75  0.028478  0.026329  0.031401  0.001201  0.058630  0.014925  0.160418   \n",
      "feat_76  0.031277  0.000682  0.010324  0.063411  0.050417  0.023242  0.291884   \n",
      "feat_77  0.001243  0.005602  0.020294  0.019275  0.007396  0.021591  0.004988   \n",
      "feat_78  0.035861  0.004071  0.018797  0.063539  0.030010  0.014639  0.043339   \n",
      "feat_79  0.002540  0.004663  0.095254  0.099579  0.018615  0.073207  0.031099   \n",
      "feat_80  0.029430  0.035876  0.081888  0.004588  0.076250  0.350787  0.012623   \n",
      "feat_81  0.096658  0.054972  0.013808  0.084096  0.017469  0.166234  0.009379   \n",
      "feat_82  0.054646  0.034368  0.065189  0.012153  0.059553  0.266249  0.001795   \n",
      "feat_83  0.001047  0.009157  0.029711  0.072006  0.052930  0.035181  0.243942   \n",
      "feat_84       NaN  0.010210  0.003459  0.013631  0.017903  0.103643  0.006013   \n",
      "feat_85       NaN       NaN  0.109643  0.049250  0.027886  0.053582  0.003931   \n",
      "feat_86       NaN       NaN       NaN  0.073685  0.426972  0.011822  0.019803   \n",
      "feat_87       NaN       NaN       NaN       NaN  0.023053  0.066008  0.014696   \n",
      "feat_88       NaN       NaN       NaN       NaN       NaN  0.022552  0.031679   \n",
      "feat_89       NaN       NaN       NaN       NaN       NaN       NaN  0.027764   \n",
      "feat_90       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_91       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_92       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "feat_93       NaN       NaN       NaN       NaN       NaN       NaN       NaN   \n",
      "\n",
      "          feat_91   feat_92   feat_93  \n",
      "feat_1   0.037264  0.054777  0.081783  \n",
      "feat_2   0.026383  0.008219  0.054593  \n",
      "feat_3   0.012417  0.066921  0.006814  \n",
      "feat_4   0.010312  0.087631  0.015746  \n",
      "feat_5   0.012012  0.065331  0.002038  \n",
      "feat_6   0.012465  0.015479  0.008521  \n",
      "feat_7   0.068506  0.032261  0.034912  \n",
      "feat_8   0.095990  0.013608  0.005131  \n",
      "feat_9   0.014742  0.069707  0.006038  \n",
      "feat_10  0.024025  0.006869  0.041316  \n",
      "feat_11  0.029511  0.013179  0.003326  \n",
      "feat_12  0.063965  0.063922  0.012722  \n",
      "feat_13  0.044338  0.071953  0.038989  \n",
      "feat_14  0.038444  0.040133  0.018127  \n",
      "feat_15  0.034402  0.018206  0.020369  \n",
      "feat_16  0.048494  0.210499  0.031467  \n",
      "feat_17  0.088508  0.006538  0.056695  \n",
      "feat_18  0.037275  0.126640  0.058100  \n",
      "feat_19  0.004085  0.027662  0.014243  \n",
      "feat_20  0.079181  0.018715  0.110054  \n",
      "feat_21  0.019779  0.058008  0.007677  \n",
      "feat_22  0.104921  0.200593  0.113276  \n",
      "feat_23  0.010438  0.031837  0.084945  \n",
      "feat_24  0.090833  0.024375  0.089200  \n",
      "feat_25  0.045759  0.030135  0.015708  \n",
      "feat_26  0.018447  0.199974  0.016709  \n",
      "feat_27  0.018778  0.023790  0.000318  \n",
      "feat_28  0.015242  0.122653  0.005275  \n",
      "feat_29  0.040526  0.084445  0.008301  \n",
      "feat_30  0.016378  0.002712  0.006260  \n",
      "feat_31  0.008808  0.022346  0.030656  \n",
      "feat_32  0.008934  0.066435  0.130842  \n",
      "feat_33  0.003487  0.099836  0.012546  \n",
      "feat_34  0.006878  0.076567  0.015377  \n",
      "feat_35  0.009463  0.064610  0.024180  \n",
      "feat_36  0.066842  0.000525  0.032804  \n",
      "feat_37  0.031573  0.011842  0.073010  \n",
      "feat_38  0.027772  0.005320  0.083854  \n",
      "feat_39  0.033654  0.028631  0.016089  \n",
      "feat_40  0.036275  0.031490  0.023675  \n",
      "feat_41  0.071538  0.159939  0.007970  \n",
      "feat_42  0.020277  0.197315  0.095720  \n",
      "feat_43  0.019450  0.040629  0.000640  \n",
      "feat_44  0.001806  0.073727  0.105498  \n",
      "feat_45  0.033762  0.019553  0.014994  \n",
      "feat_46  0.022626  0.049284  0.004403  \n",
      "feat_47  0.008806  0.016421  0.000629  \n",
      "feat_48  0.047508  0.023148  0.053749  \n",
      "feat_49  0.006273  0.037485  0.029427  \n",
      "feat_50  0.008383  0.040728  0.052934  \n",
      "feat_51  0.000449  0.013779  0.017484  \n",
      "feat_52  0.012082  0.026573  0.014812  \n",
      "feat_53  0.004092  0.135595  0.006926  \n",
      "feat_54  0.018972  0.153724  0.029959  \n",
      "feat_55  0.055702  0.049364  0.055275  \n",
      "feat_56  0.003705  0.028787  0.007368  \n",
      "feat_57  0.010013  0.100979  0.045129  \n",
      "feat_58  0.056724  0.014156  0.034472  \n",
      "feat_59  0.003928  0.010828  0.047987  \n",
      "feat_60  0.011737  0.128371  0.055350  \n",
      "feat_61  0.017045  0.042357  0.001258  \n",
      "feat_62  0.279845  0.327176  0.012774  \n",
      "feat_63  0.009487  0.031464  0.023298  \n",
      "feat_64  0.020846  0.071886  0.023064  \n",
      "feat_65  0.001188  0.044286  0.015500  \n",
      "feat_66  0.098063  0.123694  0.067957  \n",
      "feat_67  0.112052  0.011247  0.129018  \n",
      "feat_68  0.022515  0.095970  0.004602  \n",
      "feat_69  0.041206  0.037961  0.032052  \n",
      "feat_70  0.030560  0.007310  0.093488  \n",
      "feat_71  0.050622  0.000368  0.002001  \n",
      "feat_72  0.008936  0.005300  0.008233  \n",
      "feat_73  0.007193  0.024017  0.000163  \n",
      "feat_74  0.016941  0.004497  0.021967  \n",
      "feat_75  0.002625  0.037710  0.006208  \n",
      "feat_76  0.175163  0.050887  0.029426  \n",
      "feat_77  0.026376  0.076551  0.001715  \n",
      "feat_78  0.068450  0.028596  0.016047  \n",
      "feat_79  0.021616  0.162033  0.029082  \n",
      "feat_80  0.017815  0.063401  0.012651  \n",
      "feat_81  0.017243  0.018565  0.019378  \n",
      "feat_82  0.014641  0.049661  0.005497  \n",
      "feat_83  0.095801  0.018325  0.054188  \n",
      "feat_84  0.003444  0.048431  0.003723  \n",
      "feat_85  0.023091  0.043484  0.023390  \n",
      "feat_86  0.024005  0.049393  0.029035  \n",
      "feat_87  0.028850  0.001424  0.499990  \n",
      "feat_88  0.033653  0.070120  0.008631  \n",
      "feat_89  0.015917  0.129622  0.030650  \n",
      "feat_90  0.014812  0.035311  0.039864  \n",
      "feat_91       NaN  0.104226  0.000045  \n",
      "feat_92       NaN       NaN  0.003653  \n",
      "feat_93       NaN       NaN       NaN  \n"
     ]
    }
   ],
   "source": [
    "# Select upper triangle of correlation matrix\n",
    "upper_tri = corr_matrix.where(np.triu(np.ones(corr_matrix.shape), k=1).astype(np.bool))\n",
    "print(); print(upper_tri)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Find index of feature columns with correlation greater than 0.95\n",
    "to_drop = [column for column in upper_tri.columns if any(upper_tri[column] > 0.80)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['feat_45']\n"
     ]
    }
   ],
   "source": [
    "print(to_drop)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(to_drop)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "#### Drop Marked Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   feat_1  feat_2  feat_3  feat_4  feat_5  feat_6  feat_7  feat_8  feat_9  \\\n",
      "0       1       0       0       0       0       0       0       0       0   \n",
      "1       0       0       0       0       0       0       0       1       0   \n",
      "2       0       0       0       0       0       0       0       1       0   \n",
      "3       1       0       0       1       6       1       5       0       0   \n",
      "4       0       0       0       0       0       0       0       0       0   \n",
      "\n",
      "   feat_10  feat_11  feat_12  feat_13  feat_14  feat_15  feat_16  feat_17  \\\n",
      "0        0        1        0        0        0        0        0        2   \n",
      "1        0        0        0        0        0        0        0        0   \n",
      "2        0        0        0        0        0        0        0        1   \n",
      "3        1        1        0        1        0        0        1        1   \n",
      "4        0        0        0        0        0        0        0        4   \n",
      "\n",
      "   feat_18  feat_19  feat_20  feat_21  feat_22  feat_23  feat_24  feat_25  \\\n",
      "0        0        0        0        0        1        0        4        1   \n",
      "1        2        0        0        0        0        0        0        0   \n",
      "2        0        0        0        0        0        0        0        0   \n",
      "3        0        0        0        0        0        0        7        2   \n",
      "4        0        0        0        0        0        0        1        0   \n",
      "\n",
      "   feat_26  feat_27  feat_28  feat_29  feat_30  feat_31  feat_32  feat_33  \\\n",
      "0        1        0        0        2        0        0        0        0   \n",
      "1        0        0        0        0        0        0        0        0   \n",
      "2        0        0        0        0        0        0        0        1   \n",
      "3        2        0        0        0       58        0       10        0   \n",
      "4        0        0        0        0        0        0        0        0   \n",
      "\n",
      "   feat_34  feat_35  feat_36  feat_37  feat_38  feat_39  feat_40  feat_41  \\\n",
      "0        0        1        0        0        0        0        1        0   \n",
      "1        0        0        0        1        0        0        0        0   \n",
      "2        0        0        0        0        0        0        0        0   \n",
      "3        0        0        0        0        3        0        0        0   \n",
      "4        0        0        0        0        0        0        0        1   \n",
      "\n",
      "   feat_42  feat_43  feat_44  feat_46  feat_47  feat_48  feat_49  feat_50  \\\n",
      "0        5        0        0        0        0        2        0        0   \n",
      "1        0        0        0        0        0        0        0        0   \n",
      "2        0        0        0        0        0        1        0        0   \n",
      "3        0        0        2        2        0        1        2        1   \n",
      "4        0        0        0        0        0        0        0        2   \n",
      "\n",
      "   feat_51  feat_52  feat_53  feat_54  feat_55  feat_56  feat_57  feat_58  \\\n",
      "0        0        0        0        1        0        0        2        0   \n",
      "1        0        0        0        0        0        0        0        1   \n",
      "2        0        0        0        0        1        0        0        0   \n",
      "3        3        0        0        3        1        0        0        0   \n",
      "4        0        0        0        0        0        0        0        0   \n",
      "\n",
      "   feat_59  feat_60  feat_61  feat_62  feat_63  feat_64  feat_65  feat_66  \\\n",
      "0        0       11        0        1        1        0        1        0   \n",
      "1        0        0        0        0        0        1        0        0   \n",
      "2        0        0        0        0        0        0        0        0   \n",
      "3        0        0        0        0        0        0        2        1   \n",
      "4        0        0        0        1        0        0        0        0   \n",
      "\n",
      "   feat_67  feat_68  feat_69  feat_70  feat_71  feat_72  feat_73  feat_74  \\\n",
      "0        7        0        0        0        1        0        0        0   \n",
      "1        1        0        0        0        0        0        2        1   \n",
      "2        6        0        0        2        0        0        0        0   \n",
      "3        5        0        0        4        0        0        2        1   \n",
      "4        0        1        0        0        3        0        0        0   \n",
      "\n",
      "   feat_75  feat_76  feat_77  feat_78  feat_79  feat_80  feat_81  feat_82  \\\n",
      "0        0        0        0        0        2        1        0        0   \n",
      "1        0        1        0        1        0        0        0        0   \n",
      "2        0        0        0        0        0        0        1        0   \n",
      "3        0        1        0        0        1        1        2        2   \n",
      "4        0        0        0        0        4        0        1        0   \n",
      "\n",
      "   feat_83  feat_84  feat_85  feat_86  feat_87  feat_88  feat_89  feat_90  \\\n",
      "0        0        0        1        0        0        0        0        0   \n",
      "1        0        0        0        0        0        0        0        0   \n",
      "2        0        0        0        0        0        0        0        0   \n",
      "3        0       22        0        1        2        0        0        0   \n",
      "4        0        0        1        0        0        0        0        1   \n",
      "\n",
      "   feat_91  feat_92  feat_93   target  \n",
      "0        0        0        0  Class_1  \n",
      "1        0        0        0  Class_1  \n",
      "2        0        0        0  Class_1  \n",
      "3        0        0        0  Class_1  \n",
      "4        0        0        0  Class_1  \n"
     ]
    }
   ],
   "source": [
    "# Drop features \n",
    "if len(to_drop) > 0:\n",
    "    df_train = df_train.drop(df_train[to_drop], axis=1)\n",
    "    print (df_train.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### PyCaret Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "#T_7e907896_25c5_11eb_b3a6_44032c87eac3row29_col1,#T_7e907896_25c5_11eb_b3a6_44032c87eac3row31_col1,#T_7e907896_25c5_11eb_b3a6_44032c87eac3row33_col1,#T_7e907896_25c5_11eb_b3a6_44032c87eac3row40_col1,#T_7e907896_25c5_11eb_b3a6_44032c87eac3row56_col1{\n",
       "            background-color:  lightgreen;\n",
       "        }</style><table id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Description</th>        <th class=\"col_heading level0 col1\" >Value</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row0_col0\" class=\"data row0 col0\" >session_id</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row0_col1\" class=\"data row0 col1\" >123</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row1_col0\" class=\"data row1 col0\" >Target</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row1_col1\" class=\"data row1 col1\" >target</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row2_col0\" class=\"data row2 col0\" >Target Type</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row2_col1\" class=\"data row2 col1\" >Multiclass</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row3_col0\" class=\"data row3 col0\" >Label Encoded</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row3_col1\" class=\"data row3 col1\" >Class_1: 0, Class_2: 1, Class_3: 2, Class_4: 3, Class_5: 4, Class_6: 5, Class_7: 6, Class_8: 7, Class_9: 8</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row4_col0\" class=\"data row4 col0\" >Original Data</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row4_col1\" class=\"data row4 col1\" >(61878, 93)</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row5_col0\" class=\"data row5 col0\" >Missing Values</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row5_col1\" class=\"data row5 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row6_col0\" class=\"data row6 col0\" >Numeric Features</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row6_col1\" class=\"data row6 col1\" >82</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row7_col0\" class=\"data row7 col0\" >Categorical Features</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row7_col1\" class=\"data row7 col1\" >10</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row8_col0\" class=\"data row8 col0\" >Ordinal Features</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row8_col1\" class=\"data row8 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row9_col0\" class=\"data row9 col0\" >High Cardinality Features</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row9_col1\" class=\"data row9 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row10_col0\" class=\"data row10 col0\" >High Cardinality Method</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row10_col1\" class=\"data row10 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row11_col0\" class=\"data row11 col0\" >Transformed Train Set</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row11_col1\" class=\"data row11 col1\" >(41148, 46)</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row12_col0\" class=\"data row12 col0\" >Transformed Test Set</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row12_col1\" class=\"data row12 col1\" >(18564, 46)</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row13_col0\" class=\"data row13 col0\" >Shuffle Train-Test</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row13_col1\" class=\"data row13 col1\" >True</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row14_col0\" class=\"data row14 col0\" >Stratify Train-Test</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row14_col1\" class=\"data row14 col1\" >True</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row15_col0\" class=\"data row15 col0\" >Fold Generator</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row15_col1\" class=\"data row15 col1\" >StratifiedKFold</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row16_col0\" class=\"data row16 col0\" >Fold Number</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row16_col1\" class=\"data row16 col1\" >10</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row17_col0\" class=\"data row17 col0\" >CPU Jobs</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row17_col1\" class=\"data row17 col1\" >-1</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row18\" class=\"row_heading level0 row18\" >18</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row18_col0\" class=\"data row18 col0\" >Use GPU</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row18_col1\" class=\"data row18 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row19\" class=\"row_heading level0 row19\" >19</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row19_col0\" class=\"data row19 col0\" >Log Experiment</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row19_col1\" class=\"data row19 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row20\" class=\"row_heading level0 row20\" >20</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row20_col0\" class=\"data row20 col0\" >Experiment Name</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row20_col1\" class=\"data row20 col1\" >clf-default-name</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row21\" class=\"row_heading level0 row21\" >21</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row21_col0\" class=\"data row21 col0\" >USI</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row21_col1\" class=\"data row21 col1\" >28e7</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row22\" class=\"row_heading level0 row22\" >22</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row22_col0\" class=\"data row22 col0\" >Imputation Type</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row22_col1\" class=\"data row22 col1\" >simple</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row23\" class=\"row_heading level0 row23\" >23</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row23_col0\" class=\"data row23 col0\" >Iterative Imputation Iteration</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row23_col1\" class=\"data row23 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row24\" class=\"row_heading level0 row24\" >24</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row24_col0\" class=\"data row24 col0\" >Numeric Imputer</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row24_col1\" class=\"data row24 col1\" >mean</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row25\" class=\"row_heading level0 row25\" >25</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row25_col0\" class=\"data row25 col0\" >Iterative Imputation Numeric Model</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row25_col1\" class=\"data row25 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row26\" class=\"row_heading level0 row26\" >26</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row26_col0\" class=\"data row26 col0\" >Categorical Imputer</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row26_col1\" class=\"data row26 col1\" >constant</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row27\" class=\"row_heading level0 row27\" >27</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row27_col0\" class=\"data row27 col0\" >Iterative Imputation Categorical Model</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row27_col1\" class=\"data row27 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row28\" class=\"row_heading level0 row28\" >28</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row28_col0\" class=\"data row28 col0\" >Unknown Categoricals Handling</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row28_col1\" class=\"data row28 col1\" >least_frequent</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row29\" class=\"row_heading level0 row29\" >29</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row29_col0\" class=\"data row29 col0\" >Normalize</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row29_col1\" class=\"data row29 col1\" >True</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row30\" class=\"row_heading level0 row30\" >30</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row30_col0\" class=\"data row30 col0\" >Normalize Method</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row30_col1\" class=\"data row30 col1\" >minmax</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row31\" class=\"row_heading level0 row31\" >31</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row31_col0\" class=\"data row31 col0\" >Transformation</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row31_col1\" class=\"data row31 col1\" >True</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row32\" class=\"row_heading level0 row32\" >32</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row32_col0\" class=\"data row32 col0\" >Transformation Method</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row32_col1\" class=\"data row32 col1\" >quantile</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row33\" class=\"row_heading level0 row33\" >33</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row33_col0\" class=\"data row33 col0\" >PCA</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row33_col1\" class=\"data row33 col1\" >True</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row34\" class=\"row_heading level0 row34\" >34</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row34_col0\" class=\"data row34 col0\" >PCA Method</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row34_col1\" class=\"data row34 col1\" >incremental</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row35\" class=\"row_heading level0 row35\" >35</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row35_col0\" class=\"data row35 col0\" >PCA Components</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row35_col1\" class=\"data row35 col1\" >46</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row36\" class=\"row_heading level0 row36\" >36</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row36_col0\" class=\"data row36 col0\" >Ignore Low Variance</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row36_col1\" class=\"data row36 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row37\" class=\"row_heading level0 row37\" >37</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row37_col0\" class=\"data row37 col0\" >Combine Rare Levels</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row37_col1\" class=\"data row37 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row38\" class=\"row_heading level0 row38\" >38</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row38_col0\" class=\"data row38 col0\" >Rare Level Threshold</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row38_col1\" class=\"data row38 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row39\" class=\"row_heading level0 row39\" >39</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row39_col0\" class=\"data row39 col0\" >Numeric Binning</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row39_col1\" class=\"data row39 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row40\" class=\"row_heading level0 row40\" >40</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row40_col0\" class=\"data row40 col0\" >Remove Outliers</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row40_col1\" class=\"data row40 col1\" >True</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row41\" class=\"row_heading level0 row41\" >41</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row41_col0\" class=\"data row41 col0\" >Outliers Threshold</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row41_col1\" class=\"data row41 col1\" >0.050000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row42\" class=\"row_heading level0 row42\" >42</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row42_col0\" class=\"data row42 col0\" >Remove Multicollinearity</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row42_col1\" class=\"data row42 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row43\" class=\"row_heading level0 row43\" >43</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row43_col0\" class=\"data row43 col0\" >Multicollinearity Threshold</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row43_col1\" class=\"data row43 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row44\" class=\"row_heading level0 row44\" >44</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row44_col0\" class=\"data row44 col0\" >Clustering</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row44_col1\" class=\"data row44 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row45\" class=\"row_heading level0 row45\" >45</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row45_col0\" class=\"data row45 col0\" >Clustering Iteration</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row45_col1\" class=\"data row45 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row46\" class=\"row_heading level0 row46\" >46</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row46_col0\" class=\"data row46 col0\" >Polynomial Features</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row46_col1\" class=\"data row46 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row47\" class=\"row_heading level0 row47\" >47</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row47_col0\" class=\"data row47 col0\" >Polynomial Degree</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row47_col1\" class=\"data row47 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row48\" class=\"row_heading level0 row48\" >48</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row48_col0\" class=\"data row48 col0\" >Trignometry Features</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row48_col1\" class=\"data row48 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row49\" class=\"row_heading level0 row49\" >49</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row49_col0\" class=\"data row49 col0\" >Polynomial Threshold</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row49_col1\" class=\"data row49 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row50\" class=\"row_heading level0 row50\" >50</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row50_col0\" class=\"data row50 col0\" >Group Features</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row50_col1\" class=\"data row50 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row51\" class=\"row_heading level0 row51\" >51</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row51_col0\" class=\"data row51 col0\" >Feature Selection</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row51_col1\" class=\"data row51 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row52\" class=\"row_heading level0 row52\" >52</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row52_col0\" class=\"data row52 col0\" >Features Selection Threshold</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row52_col1\" class=\"data row52 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row53\" class=\"row_heading level0 row53\" >53</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row53_col0\" class=\"data row53 col0\" >Feature Interaction</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row53_col1\" class=\"data row53 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row54\" class=\"row_heading level0 row54\" >54</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row54_col0\" class=\"data row54 col0\" >Feature Ratio</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row54_col1\" class=\"data row54 col1\" >False</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row55\" class=\"row_heading level0 row55\" >55</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row55_col0\" class=\"data row55 col0\" >Interaction Threshold</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row55_col1\" class=\"data row55 col1\" >None</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row56\" class=\"row_heading level0 row56\" >56</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row56_col0\" class=\"data row56 col0\" >Fix Imbalance</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row56_col1\" class=\"data row56 col1\" >True</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3level0_row57\" class=\"row_heading level0 row57\" >57</th>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row57_col0\" class=\"data row57 col0\" >Fix Imbalance Method</td>\n",
       "                        <td id=\"T_7e907896_25c5_11eb_b3a6_44032c87eac3row57_col1\" class=\"data row57 col1\" >SMOTE</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f7553010128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Execution time :  0:00:24.594763\n",
      "Job Ended at: 2020-11-13 21:32:52.807448\n"
     ]
    }
   ],
   "source": [
    "init_time = datetime.now()\n",
    "print (f\"Job Started at: {init_time}\")\n",
    "#fold_strategy = 'kfold' Default = 'stratifiedkfold'\n",
    "clf = setup(data = df_train, target = 'target', session_id=123,\n",
    "            data_split_stratify = True,\n",
    "            normalize = True, \n",
    "            normalize_method = 'minmax',\n",
    "            transformation = True, \n",
    "            transformation_method = 'quantile',\n",
    "            pca = True,\n",
    "            pca_method = 'incremental',\n",
    "            #feature_selection = True,\n",
    "            #feature_selection_threshold = 0.5,\n",
    "            #feature_selection_method = 'boruta',\n",
    "            remove_outliers = True,\n",
    "            outliers_threshold = 0.05,\n",
    "            fix_imbalance = True\n",
    "           )\n",
    "fin_time = datetime.now()\n",
    "print(\"Execution time : \", (fin_time-init_time))\n",
    "print (f\"Job Ended at: {fin_time}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Finding optimal model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Job Started at: 2020-11-13 21:33:16.794172\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "be5cb8bc9d794e3a9652b5630cded861",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "IntProgress(value=0, description='Processing: ', max=79)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Initiated</th>\n",
       "      <td>. . . . . . . . . . . . . . . . . .</td>\n",
       "      <td>21:33:16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Status</th>\n",
       "      <td>. . . . . . . . . . . . . . . . . .</td>\n",
       "      <td>Fitting 10 Folds</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Estimator</th>\n",
       "      <td>. . . . . . . . . . . . . . . . . .</td>\n",
       "      <td>Gradient Boosting Classifier</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                            \n",
       "                                                                            \n",
       "Initiated  . . . . . . . . . . . . . . . . . .                      21:33:16\n",
       "Status     . . . . . . . . . . . . . . . . . .              Fitting 10 Folds\n",
       "Estimator  . . . . . . . . . . . . . . . . . .  Gradient Boosting Classifier"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_07cb80d0_25c9_11eb_b3a6_44032c87eac3 th {\n",
       "          text-align: left;\n",
       "    }#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col0,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col1,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col2,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col3,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col4,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col5,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col6,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col7,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col8,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col0,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col1,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col2,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col3,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col4,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col5,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col6,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col7,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col8,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col0,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col1,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col2,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col3,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col4,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col5,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col6,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col7,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col8,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col0,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col1,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col2,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col3,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col4,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col5,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col6,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col7,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col8,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col0,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col1,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col2,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col3,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col4,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col5,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col6,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col7,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col8,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col0,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col1,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col2,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col3,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col4,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col5,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col6,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col7,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col8,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col0,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col1,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col2,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col3,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col4,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col5,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col6,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col7,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col8,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col0,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col1,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col2,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col3,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col4,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col5,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col6,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col7,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col8,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col0,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col1,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col2,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col3,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col4,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col5,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col6,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col7,#T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col8{\n",
       "            text-align:  left;\n",
       "        }</style><table id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Model</th>        <th class=\"col_heading level0 col1\" >Accuracy</th>        <th class=\"col_heading level0 col2\" >AUC</th>        <th class=\"col_heading level0 col3\" >Recall</th>        <th class=\"col_heading level0 col4\" >Prec.</th>        <th class=\"col_heading level0 col5\" >F1</th>        <th class=\"col_heading level0 col6\" >Kappa</th>        <th class=\"col_heading level0 col7\" >MCC</th>        <th class=\"col_heading level0 col8\" >TT (Sec)</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3level0_row0\" class=\"row_heading level0 row0\" >rf</th>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col0\" class=\"data row0 col0\" >Random Forest Classifier</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col1\" class=\"data row0 col1\" >0.7421</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col2\" class=\"data row0 col2\" >0.9499</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col3\" class=\"data row0 col3\" >0.6904</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col4\" class=\"data row0 col4\" >0.7412</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col5\" class=\"data row0 col5\" >0.7410</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col6\" class=\"data row0 col6\" >0.6895</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col7\" class=\"data row0 col7\" >0.6896</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row0_col8\" class=\"data row0 col8\" >36.6140</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3level0_row1\" class=\"row_heading level0 row1\" >knn</th>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col0\" class=\"data row1 col0\" >K Neighbors Classifier</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col1\" class=\"data row1 col1\" >0.6882</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col2\" class=\"data row1 col2\" >0.8972</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col3\" class=\"data row1 col3\" >0.6967</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col4\" class=\"data row1 col4\" >0.7407</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col5\" class=\"data row1 col5\" >0.7029</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col6\" class=\"data row1 col6\" >0.6339</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col7\" class=\"data row1 col7\" >0.6382</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row1_col8\" class=\"data row1 col8\" >28.2520</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3level0_row2\" class=\"row_heading level0 row2\" >lr</th>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col0\" class=\"data row2 col0\" >Logistic Regression</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col1\" class=\"data row2 col1\" >0.6424</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col2\" class=\"data row2 col2\" >0.9331</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col3\" class=\"data row2 col3\" >0.6509</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col4\" class=\"data row2 col4\" >0.6992</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col5\" class=\"data row2 col5\" >0.6582</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col6\" class=\"data row2 col6\" >0.5815</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col7\" class=\"data row2 col7\" >0.5860</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row2_col8\" class=\"data row2 col8\" >18.2460</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3level0_row3\" class=\"row_heading level0 row3\" >qda</th>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col0\" class=\"data row3 col0\" >Quadratic Discriminant Analysis</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col1\" class=\"data row3 col1\" >0.6355</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col2\" class=\"data row3 col2\" >0.9327</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col3\" class=\"data row3 col3\" >0.6288</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col4\" class=\"data row3 col4\" >0.6983</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col5\" class=\"data row3 col5\" >0.6526</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col6\" class=\"data row3 col6\" >0.5732</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col7\" class=\"data row3 col7\" >0.5797</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row3_col8\" class=\"data row3 col8\" >8.7550</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3level0_row4\" class=\"row_heading level0 row4\" >ridge</th>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col0\" class=\"data row4 col0\" >Ridge Classifier</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col1\" class=\"data row4 col1\" >0.5986</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col2\" class=\"data row4 col2\" >0.0000</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col3\" class=\"data row4 col3\" >0.6046</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col4\" class=\"data row4 col4\" >0.6620</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col5\" class=\"data row4 col5\" >0.6067</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col6\" class=\"data row4 col6\" >0.5344</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col7\" class=\"data row4 col7\" >0.5423</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row4_col8\" class=\"data row4 col8\" >6.7650</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3level0_row5\" class=\"row_heading level0 row5\" >svm</th>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col0\" class=\"data row5 col0\" >SVM - Linear Kernel</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col1\" class=\"data row5 col1\" >0.5901</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col2\" class=\"data row5 col2\" >0.0000</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col3\" class=\"data row5 col3\" >0.6017</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col4\" class=\"data row5 col4\" >0.6648</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col5\" class=\"data row5 col5\" >0.6046</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col6\" class=\"data row5 col6\" >0.5238</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col7\" class=\"data row5 col7\" >0.5320</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row5_col8\" class=\"data row5 col8\" >9.7930</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3level0_row6\" class=\"row_heading level0 row6\" >nb</th>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col0\" class=\"data row6 col0\" >Naive Bayes</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col1\" class=\"data row6 col1\" >0.5775</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col2\" class=\"data row6 col2\" >0.9005</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col3\" class=\"data row6 col3\" >0.5746</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col4\" class=\"data row6 col4\" >0.6443</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col5\" class=\"data row6 col5\" >0.5954</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col6\" class=\"data row6 col6\" >0.5081</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col7\" class=\"data row6 col7\" >0.5138</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row6_col8\" class=\"data row6 col8\" >7.7730</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3level0_row7\" class=\"row_heading level0 row7\" >dt</th>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col0\" class=\"data row7 col0\" >Decision Tree Classifier</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col1\" class=\"data row7 col1\" >0.5718</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col2\" class=\"data row7 col2\" >0.7528</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col3\" class=\"data row7 col3\" >0.5279</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col4\" class=\"data row7 col4\" >0.6085</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col5\" class=\"data row7 col5\" >0.5858</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col6\" class=\"data row7 col6\" >0.4936</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col7\" class=\"data row7 col7\" >0.4951</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row7_col8\" class=\"data row7 col8\" >12.0510</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3level0_row8\" class=\"row_heading level0 row8\" >ada</th>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col0\" class=\"data row8 col0\" >Ada Boost Classifier</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col1\" class=\"data row8 col1\" >0.5438</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col2\" class=\"data row8 col2\" >0.8788</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col3\" class=\"data row8 col3\" >0.5522</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col4\" class=\"data row8 col4\" >0.6268</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col5\" class=\"data row8 col5\" >0.5646</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col6\" class=\"data row8 col6\" >0.4725</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col7\" class=\"data row8 col7\" >0.4803</td>\n",
       "                        <td id=\"T_07cb80d0_25c9_11eb_b3a6_44032c87eac3row8_col8\" class=\"data row8 col8\" >20.9490</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7f75530184e0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "init_time = datetime.now()\n",
    "print (f\"Job Started at: {init_time}\")\n",
    "#top3_models = compare_models(n_select = 3)\n",
    "best = compare_models()\n",
    "fin_time = datetime.now()\n",
    "print(\"Execution time : \", (fin_time-init_time))\n",
    "print (f\"Job Ended at: {fin_time}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Job Started at: 2020-11-12 20:35:10.135452\n",
      "Execution time :  0:05:17.129745\n",
      "Job Ended at: 2020-11-12 20:40:27.265197\n"
     ]
    }
   ],
   "source": [
    "init_time = datetime.now()\n",
    "print (f\"Job Started at: {init_time}\")\n",
    "best_model = finalize_model(best)\n",
    "fin_time = datetime.now()\n",
    "print(\"Execution time : \", (fin_time-init_time))\n",
    "print (f\"Job Ended at: {fin_time}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight=None,\n",
      "                       criterion='gini', max_depth=None, max_features='auto',\n",
      "                       max_leaf_nodes=None, max_samples=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
      "                       n_jobs=-1, oob_score=False, random_state=123, verbose=0,\n",
      "                       warm_start=False)\n"
     ]
    }
   ],
   "source": [
    "print (best_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "#best = compare_models()\n",
    "#lgb = create_model('lightgbm')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Tune Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "#T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col0,#T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col1,#T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col2,#T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col3,#T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col4,#T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col5,#T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col6{\n",
       "            background:  yellow;\n",
       "        }</style><table id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Accuracy</th>        <th class=\"col_heading level0 col1\" >AUC</th>        <th class=\"col_heading level0 col2\" >Recall</th>        <th class=\"col_heading level0 col3\" >Prec.</th>        <th class=\"col_heading level0 col4\" >F1</th>        <th class=\"col_heading level0 col5\" >Kappa</th>        <th class=\"col_heading level0 col6\" >MCC</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row0_col0\" class=\"data row0 col0\" >0.6519</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row0_col1\" class=\"data row0 col1\" >0.9251</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row0_col2\" class=\"data row0 col2\" >0.6325</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row0_col3\" class=\"data row0 col3\" >0.6725</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row0_col4\" class=\"data row0 col4\" >0.6534</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row0_col5\" class=\"data row0 col5\" >0.5874</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row0_col6\" class=\"data row0 col6\" >0.5899</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row1_col0\" class=\"data row1 col0\" >0.6734</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row1_col1\" class=\"data row1 col1\" >0.9285</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row1_col2\" class=\"data row1 col2\" >0.6452</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row1_col3\" class=\"data row1 col3\" >0.6966</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row1_col4\" class=\"data row1 col4\" >0.6771</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row1_col5\" class=\"data row1 col5\" >0.6123</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row1_col6\" class=\"data row1 col6\" >0.6145</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row2_col0\" class=\"data row2 col0\" >0.6717</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row2_col1\" class=\"data row2 col1\" >0.9280</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row2_col2\" class=\"data row2 col2\" >0.6338</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row2_col3\" class=\"data row2 col3\" >0.6921</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row2_col4\" class=\"data row2 col4\" >0.6752</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row2_col5\" class=\"data row2 col5\" >0.6100</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row2_col6\" class=\"data row2 col6\" >0.6117</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row3_col0\" class=\"data row3 col0\" >0.6674</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row3_col1\" class=\"data row3 col1\" >0.9256</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row3_col2\" class=\"data row3 col2\" >0.6421</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row3_col3\" class=\"data row3 col3\" >0.6842</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row3_col4\" class=\"data row3 col4\" >0.6692</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row3_col5\" class=\"data row3 col5\" >0.6046</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row3_col6\" class=\"data row3 col6\" >0.6062</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row4_col0\" class=\"data row4 col0\" >0.6571</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row4_col1\" class=\"data row4 col1\" >0.9254</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row4_col2\" class=\"data row4 col2\" >0.6392</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row4_col3\" class=\"data row4 col3\" >0.6842</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row4_col4\" class=\"data row4 col4\" >0.6611</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row4_col5\" class=\"data row4 col5\" >0.5945</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row4_col6\" class=\"data row4 col6\" >0.5974</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row5_col0\" class=\"data row5 col0\" >0.6634</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row5_col1\" class=\"data row5 col1\" >0.9249</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row5_col2\" class=\"data row5 col2\" >0.6250</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row5_col3\" class=\"data row5 col3\" >0.6840</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row5_col4\" class=\"data row5 col4\" >0.6662</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row5_col5\" class=\"data row5 col5\" >0.5997</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row5_col6\" class=\"data row5 col6\" >0.6016</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row6_col0\" class=\"data row6 col0\" >0.6638</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row6_col1\" class=\"data row6 col1\" >0.9248</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row6_col2\" class=\"data row6 col2\" >0.6440</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row6_col3\" class=\"data row6 col3\" >0.6817</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row6_col4\" class=\"data row6 col4\" >0.6652</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row6_col5\" class=\"data row6 col5\" >0.6004</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row6_col6\" class=\"data row6 col6\" >0.6025</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row7_col0\" class=\"data row7 col0\" >0.6631</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row7_col1\" class=\"data row7 col1\" >0.9249</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row7_col2\" class=\"data row7 col2\" >0.6275</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row7_col3\" class=\"data row7 col3\" >0.6827</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row7_col4\" class=\"data row7 col4\" >0.6654</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row7_col5\" class=\"data row7 col5\" >0.5995</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row7_col6\" class=\"data row7 col6\" >0.6014</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row8_col0\" class=\"data row8 col0\" >0.6684</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row8_col1\" class=\"data row8 col1\" >0.9258</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row8_col2\" class=\"data row8 col2\" >0.6315</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row8_col3\" class=\"data row8 col3\" >0.6868</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row8_col4\" class=\"data row8 col4\" >0.6711</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row8_col5\" class=\"data row8 col5\" >0.6058</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row8_col6\" class=\"data row8 col6\" >0.6076</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row9_col0\" class=\"data row9 col0\" >0.6666</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row9_col1\" class=\"data row9 col1\" >0.9263</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row9_col2\" class=\"data row9 col2\" >0.6386</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row9_col3\" class=\"data row9 col3\" >0.6859</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row9_col4\" class=\"data row9 col4\" >0.6704</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row9_col5\" class=\"data row9 col5\" >0.6038</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row9_col6\" class=\"data row9 col6\" >0.6054</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row10\" class=\"row_heading level0 row10\" >Mean</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col0\" class=\"data row10 col0\" >0.6647</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col1\" class=\"data row10 col1\" >0.9259</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col2\" class=\"data row10 col2\" >0.6359</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col3\" class=\"data row10 col3\" >0.6851</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col4\" class=\"data row10 col4\" >0.6674</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col5\" class=\"data row10 col5\" >0.6018</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row10_col6\" class=\"data row10 col6\" >0.6038</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3level0_row11\" class=\"row_heading level0 row11\" >SD</th>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row11_col0\" class=\"data row11 col0\" >0.0061</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row11_col1\" class=\"data row11 col1\" >0.0012</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row11_col2\" class=\"data row11 col2\" >0.0066</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row11_col3\" class=\"data row11 col3\" >0.0061</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row11_col4\" class=\"data row11 col4\" >0.0065</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row11_col5\" class=\"data row11 col5\" >0.0069</td>\n",
       "                        <td id=\"T_0b2ef98c_2501_11eb_8373_44032c87eac3row11_col6\" class=\"data row11 col6\" >0.0067</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7ff9befa2cc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForestClassifier(bootstrap=False, ccp_alpha=0.0,\n",
      "                       class_weight='balanced_subsample', criterion='gini',\n",
      "                       max_depth=6, max_features='log2', max_leaf_nodes=None,\n",
      "                       max_samples=None, min_impurity_decrease=0.001,\n",
      "                       min_impurity_split=None, min_samples_leaf=6,\n",
      "                       min_samples_split=9, min_weight_fraction_leaf=0.0,\n",
      "                       n_estimators=190, n_jobs=-1, oob_score=False,\n",
      "                       random_state=123, verbose=0, warm_start=False)\n",
      "Execution time :  0:53:02.500396\n",
      "Job Ended at: 2020-11-12 22:06:37.855853\n"
     ]
    }
   ],
   "source": [
    "init_time = datetime.now()\n",
    "print (f\"Job Started at: {init_time}\")\n",
    "#tuned_model = tune_model(best_model, optimize = 'AUC') # 'Recall'\n",
    "tuned_model = tune_model(best_model, optimize = 'F1')\n",
    "print(tuned_model)\n",
    "fin_time = datetime.now()\n",
    "print(\"Execution time : \", (fin_time-init_time))\n",
    "print (f\"Job Ended at: {fin_time}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "hidden": true
   },
   "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>Parameters</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>boosting_type</th>\n",
       "      <td>gbdt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>class_weight</th>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>colsample_bytree</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>importance_type</th>\n",
       "      <td>split</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>learning_rate</th>\n",
       "      <td>0.064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max_depth</th>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min_child_samples</th>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min_child_weight</th>\n",
       "      <td>0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min_split_gain</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_estimators</th>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_jobs</th>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>num_leaves</th>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>objective</th>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>random_state</th>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reg_alpha</th>\n",
       "      <td>0.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reg_lambda</th>\n",
       "      <td>0.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>silent</th>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subsample</th>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subsample_for_bin</th>\n",
       "      <td>200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subsample_freq</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_fraction</th>\n",
       "      <td>0.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bagging_freq</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bagging_fraction</th>\n",
       "      <td>0.4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Parameters\n",
       "boosting_type           gbdt\n",
       "class_weight            None\n",
       "colsample_bytree         1.0\n",
       "importance_type        split\n",
       "learning_rate          0.064\n",
       "max_depth                 -1\n",
       "min_child_samples        100\n",
       "min_child_weight       0.001\n",
       "min_split_gain             0\n",
       "n_estimators             170\n",
       "n_jobs                    -1\n",
       "num_leaves                80\n",
       "objective               None\n",
       "random_state             123\n",
       "reg_alpha                0.7\n",
       "reg_lambda               0.4\n",
       "silent                  True\n",
       "subsample                1.0\n",
       "subsample_for_bin     200000\n",
       "subsample_freq             0\n",
       "feature_fraction         0.9\n",
       "bagging_freq               4\n",
       "bagging_fraction         0.4"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_model(tuned_model, plot = 'parameter')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "init_time = datetime.now()\n",
    "print (f\"Job Started at: {init_time}\")\n",
    "custom_grid = {\n",
    "                'n_estimators': [100, 200, 300, 400, 500, 600, 700, 800, 900, 1000],\n",
    "                'max_depth': [3, 4, 5, 6, 7, 8, 9, 10],\n",
    "                'learning_rate': [0.0001, 0.001, 0.01, 0.02, 0.1, 0.2, 0.3], \n",
    "                'subsample' : [0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 1.0],\n",
    "                'colsample_bytree' : [0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 1.0],\n",
    "                'colsample_bylevel' : [0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 1.0],\n",
    "                'gamma': [0],\n",
    "                'reg_lambda': [0, 1.0, 3.0, 5.0, 7.0, 10.0, 12.0],\n",
    "                'scale_pos_weight': [1, 2, 3, 4, 5, 6]\n",
    "              }\n",
    "#tuned_lgb = tune_model(lgb, custom_grid = custom_grid)\n",
    "fin_time = datetime.now()\n",
    "print(\"Execution time : \", (fin_time-init_time))\n",
    "print (f\"Job Ended at: {fin_time}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Feature Importance Plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_model(tuned_model, plot='feature')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Model visualization and interpretation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_model(tuned_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjQAAAGACAYAAAC6OPj9AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAACT80lEQVR4nOzdd1gU19vG8S+CgCBdqh3sIqJiVzQaSyzBGmPvJXaNJfbejRq7sWvsvbdEY+9iwQYiggqCAqLSWXj/8BfebFCwsAwLz+e69ko4Z3bnPp7Z3WdnZnd0kpKSkhBCCCGE0GI5lA4ghBBCCPG1pKARQgghhNaTgkYIIYQQWk8KGiGEEEJoPSlohBBCCKH1pKARQgghhNaTgkaIr1S8eHFevHjxwb6XL18yatQoGjRoQMOGDWnQoAHTp08nKipK7f716tVL7u/RowdPnz5V6x84cGCKxx4zZgzFixdP/jspKYm1a9fSpEkTGjRowLfffsvEiRN5+/YtAL/88gtLly5Nr2EDcPv2bbp37w7A06dPqVevHh4eHmrtX+Pt27dMnjyZ+vXr06BBAxo1asSaNWv4ml+bePv2LR4eHtSvX5/w8PDPvv8ff/zBggULvnj9/1WnTh1atGiRon3p0qUUL16cZ8+epfkY27dv/2hf586duXv37ldlFEIbSEEjhIZERUXRoUMH8uXLx+HDhzl69Chbt27Fz8+PcePGqS27ceNGjh49yrFjxyhZsiTTpk1T63/48CHv3r1L/jsuLo47d+6oLTN37lwOHz7M6tWrOXbsGPv37yc+Pp7evXt/VQGQGhcXF1avXg3A9evXsba2Zt++fWrtXyoxMZGePXsSGxvLgQMHOHbsGGvWrOHQoUNfVVA8fPiQ169fc/z4cSwsLD77/h06dGDw4MFfvP4PCQsLw8/PT63tzz//xNLSMs37vnz5klWrVn20f/369ZQuXfqrMwqR2UlBI4SG7Nmzhzx58tCvXz90dXUBsLCwYPHixcyaNeuj96tSpYraHhqAypUrc+LEieS/z507R5kyZZL/fv36NRs3bmTmzJnY2toCYGRkxPjx4+nRo0eKgsbT05MWLVrQsGFDGjVqxIULFwBISEhgzJgxNGjQgHr16tG/f3/evXv30fbLly9Tr149PD09mTt3Lvfu3eP7779Pbof3xdfUqVNp0KABderUYfny5ck56tSpw+LFi2nQoAGBgYFqGc+cOUNwcDATJ07EwMAAADs7O+bPn0/dunUBCAwMpHv37jRo0IAmTZqwd+9eAJ49e0aNGjXYsGEDTZs2pWbNmhw+fJjAwECGDRtGaGgoDRs25M6dO5QqVSp5nc+ePUv+Ozg4mM6dO9OoUSO+/fZb5s+fD8CiRYsYM2bMF63/Y9zd3Tl06FDy3w8fPsTU1JRcuXIlt/311180bdqUBg0a0KJFC+7fvw/Ajz/+SGBgIA0bNiQuLi7Fv2mdOnW4du0aa9asoU+fPsmP161bNzZt2vTRTEJoGylohNCQK1euUKtWrRTtBgYG6OnpffA+cXFx7N+/nzp16qi1f/fddxw8eDD570OHDtGwYcPkv2/duoWdnR1OTk4p1lWnTh1y5FB/qo8fP57u3btz9OhRevXqxYQJE4D3hdKzZ884evQox48fp0iRInh6en60/R/lypVj6NChuLq6sn//frV1rVy5kkePHnHgwAEOHjzIsWPHOHXqVHJ/cHAwx44dw8HBIcW/X/Xq1cmZM6dae4ECBXBxcQFg3LhxVKpUiWPHjrFixQqmTp2afIgmPDycHDlycODAAUaPHs2CBQtwcHBg1qxZ2Nvbc/To0VT30Kxbt46KFSty+PBhDhw4wNOnTwkJCVFb5nPX/zENGzZMdX4TEhL45ZdfmDJlCseOHaNOnTrJRfH06dOTx6Ovr//Rf9POnTsTEhLCuXPn+PPPP4mMjKRt27YfzSSEtpGCRggNefPmjdob5q5du2jYsCENGzakSpUqBAUFJfd17NiRhg0bUr16de7cuZPinIpKlSrh4+NDaGgo0dHReHp6UrVq1eT+169fY2Vl9cnZ9u7dy3fffQdAhQoVkvcIWVpa4uvry4kTJ4iOjmbw4MHUrFnzo+2f4tSpU7Rr1w59fX2MjIzw8PDg+PHjyf21a9f+4P0iIiJSHVN8fDwXLlygXbt2AOTNm5fKlStz6dIl4H0R8M+/Y+nSpVPsAUqLlZUV586d49q1a+jr6zNv3jxsbGw0sv4CBQpgbGyMl5cXAMeOHaN+/frJ/Xp6ely4cAFXV1cA3NzcUuzF+7cP/Zvq6uoyZcoUZs2axa+//sqUKVNSFLpCaLMPf0wUQnw1S0tLtU/0LVu2pGXLlgCUKlUKlUqV3Ldx40bs7OwAuHr1Kh07dmT37t3Jb6C6urrUr1+fI0eOYGlpSY0aNdT28lhYWBAcHPzJ2Q4cOMCGDRuIjIwkMTEx+ZCUi4sLY8eOZePGjYwcOZI6deowYcKEj7Z/irdv3zJjxgzmzZsHvN8L9c8eFgAzM7MP3s/CwiLFHpF/e/36NUlJSZiYmCS3mZqaEhYWBrz/NzMyMgIgR44cJCYmflLef3Tp0oXExEQmTZpESEgI7du3Z8CAARpbf5MmTTh06BCJiYnky5cvxfkzGzduZM+ePcTFxREXF4eOjs5HH+tj/6alS5fG2NgYXV1dihUrlvo/gBBaRspzITSkatWqnDhx4rNPyK1YsSIODg5cv35drb1Ro0YcO3aMo0eP0qhRI7U+V1dXQkNDU3ybJT4+nvnz5xMdHZ3cFhwczNixY5k2bRrHjh1j5cqVavdp2LAhGzdu5NSpU0RHRyef3Pux9rTY2Ngwfvx4jh49ytGjRzl58uQnndRbuXJlzpw5Q0xMjFp7QEAAa9euxcLCghw5chAREZHc97l7qnR1ddUKujdv3iT36enp0atXLw4cOMDWrVvZv39/8rlGQLqs/98aNWrE8ePHOXLkSIr5vXHjBitXrmTZsmUcO3aMqVOnftE6/v77b/T09IiLi+P06dNf9BhCZFZS0AihId9//z0A06ZNIy4uDnh/XsXYsWMxNzdX+2T/b35+fvj5+eHo6KjWXq5cOUJCQvDx8aFSpUpqfaampvTo0YORI0fi7+8PQHR0NOPHj+fevXtqJ5eGhYVhZGSEo6MjCQkJbNu2DYDIyEh27drFkiVLADA3N0/O8LH2T1G3bl127NiBSqUiKSmJpUuXcubMmTTvV6NGDRwdHRkxYkTyN7xevHjB4MGDSUhIQE9Pjxo1aiTnDwgI4Nq1a1SrVu2Ts1lYWKCrq8vDhw8Bkk/qhffnGZ0/fx54f0goT548antF0mP9/2Zra4u9vT1HjhxJPqH6H2FhYVhZWeHg4EB0dDR79uwhKiqKpKQk9PT0iIqKIiEhIdXHj4qKYtq0aYwbN45x48YxadIktZ8PEELbySEnIdJBx44dk7/JBDB16lTc3NxYt24d8+bNo2nTpgCoVCpq1qzJnj171A4L/Pv++vr6TJo0Se03ZgB0dHSoV68e0dHRHzz3YcCAAZiZmfHTTz+hUqnIkSMHdevWZeLEiWrLlShRAnd3dxo0aICVlRW//PILN27coGPHjqxZs4bRo0dTv359dHV1KViwIDNnzgT4YPs/hUBq2rVrx7Nnz2jcuDFJSUk4OzvTuXPnNO+no6PD8uXLmT9/Ps2aNUNPT49cuXLRvn17WrVqBcCkSZMYO3Ysu3fvJmfOnEydOhV7e/tP+u0WAENDQwYMGECPHj2wsbGhY8eOyX0//vgj48ePZ8qUKSQlJVGnTh2qVq2qtufsa9f/X40bN+b06dOYmpqqtdesWZPNmzfz7bffYmtry+jRo7l16xYDBw5kxowZmJmZUb16dfbs2fPRx160aBG1a9dO3q6qVq3KggULGD169BdlFSKz0UnS1A9UCCGEEEJkEDnkJIQQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC62W6bzklJiYSGRlJzpw5U/3hKCGEEEJkH0lJScTHx2NsbPzBb3pmuoImMjISb29vpWMIIYQQIhMqVqzYB3/HK9MVNP9ciK6QUwFy6udMY2ntZaCbi1hVdNoLarnsNM7sIqvPZ3baZmWcWUd2GGd8XDxPfANSXLD2H5muoPnnMFNO/ZzoG+grnEZzDHQNSPrXtXyyquw0zuwiq89ndtpmZZxZR3YZJ/DR01HkpGAhhBBCaD0paIQQQgih9aSgEUIIIYTWk4JGCCGEEFpPChohhBBCaD0paIQQQgih9aSgEUIIIYTWk4JGCCGEEFpPChohhBBCaL0sXdDcu3uPlh6tyGuTH3NjS6pWrMaB/QeT+9+9e8fAfoMwypmbjes3prj/jeueNKrfmLw2+Smc34lO7Tvz6tWrjBzCF3ni94T6dRqSS88Y/yf+an3eD735vpEHecxssM+Tl47ttGNMnyIkJIRe3XtTOL8TNhZ2uFerzam/TikdK92pVCq8vb25dOkSZ8+e5caNG4SFhSkdK91l9fm8cP4ixvomTJ00Lblt29btVK1YDWtzW5xLuDBh7ERUWvjrr2nN3Yplv1OuTAXymNlQokgppkycSmJiooKJv4z3Q29aN/+BAvYFsbNywL1abQ4fPAJAdHQ0E8ZOpHTxMlia5MHNtSJHDh1VOPGnSe095FO20V/nzKO4U0kscltRrkwFtmzemiG5s2xBExUVRf06DXF0cuSB7z2Cw4Jo1rwZbVu34/69+3g/9KZS+SrA+yt4/ldYWBgejZtRvkJ57vl4cenaBSJeR9C+TceMHspn2bd3P7Wq16ZAwfwp+sLDw2lQtyGu5VzxDfDh2q2rxMTEsGThUgWSpr/WzdsQFBjExavnCQh6Qq3a7rRu0YbAwCClo6UrHx8fIiIicHFxoVq1atjZ2XHnzh2ioqKUjpausvJ8RkdH07tHH7UL7J09fZaeXXsxbOQwngUHsHXHZrZs3srMabMUTPplUpu7Vb+vZsLYiSxcvIDgsCDWrF/Not8Ws3TxMqVjf5bExESaNvLAMFcubt71xD/Qj5atW/Bj67Z4P/Tml+Gj2LxpC5u2buRFaCBjJ4ylQ9uOeN3xUjp6qlJ7D/mUbXTOrLmsXLGKTVv/IOjVc8ZNHMus6bN59uy5xrNrtKCJjo5m4sSJ1KlThwoVKtCmTRvOnz+vyVUmi4yMZMr0yUyaOhETExMMDAzo0683KpWKu3fvEfwimHkL5jLvt18/eP9tW7aTlJTExCkTMDMzw9bWlinTJ3Pm9Blu37qdIWP4EuFh4fz59wnatW+bom/1yjVYWFoyedokzMzMyJvXgR27tzFh8ngFkqaviIgISpYqwZx5s7Gzs8PQ0JCfRwwlMjKSq1euKh0v3cTHxxMcHEyhQoUwMjJCV1cXBwcHjI2NCQwMVDpeusnq8zl+zASKFy+GS1mX5LalS5bT4LsGtGzVAgMDA5zLODNw8ACWLVmuVXsv0pq72NhYps2cSs1aNdHV1aVa9arU+qYWp0+dVjr6Z3n58iUB/gF06NgOS0tLDAwM6P1TL+Lj47l16zZ7du2hV+8euJZzRV9fn2bNPWjyfWNWrlildPRUpfYektY2Ghsby7w585k6YwpuFStgaGhIi5bNuel1g3z58mo8u0YLmsmTJ+Pp6cnq1au5cOECzZs3p0+fPjx+/FiTqwXA2tqart27YGRkBEBoaCgzps4kb768fFOnNjVr1aRho4Yfvf+Vy1dwLVcWPb3/v35nGZcyGBoacuVy5n1B7dKtM0WLFf1g3+lTp3EtV5ZB/QeT364gTgWLMqj/YN69e5fBKdOfmZkZy1cuo0TJEsltfo/9AMiXL59SsdLd27dvSUpKwtTUVK3dxMSEN2/eKJQq/WXl+Tx/7gKb/9jCwqUL1dqvXL5CxYpuam1uFd0IDQ3lkc+jjIz4VdKau34D+tK9Z7fkvqSkJAL8/cmbX7vm1dbWlmrVq7F2zXpevnxJfHw8K1esIk+ePNSq7Y6Ojg6Jiep7/63y5OHatesKJf40qb2HpLWNet64yevXr4mPi6NqxWrYWNhRvUpN/jrxV0ZE11xBExERwYEDBxgwYACFCxfGwMCAH3/8EScnJ7ZuzZjjaf8wM7Ign20Bzp09z6GjB7CyskrzPq9evsLC0lKtTUdHBwtLC16GvNRUVI169uw5+/bsx7WcKz5PHrJ91zYOHzrCsCHDlY6W7t68eUPvHn1o8n0TKriVVzpOuomPjwdQK7QBcubMSVxcnBKRMkRWmc+oqCh69+jDzDnTcXCwV+t7/5pjodaWJ8/716oQLX3NgbTnbvqUGQT4P2XI0EEKpPs6W3ZsIsDfnwL2hTAzsmDOzDls2roRGxsbmrVoxu8rVnLt6nXi4+P5++Tf7Nuzj9BXoUrH/mJpbaPPnj0DYP3aDWzetolH/t7Url2LFh6t8H3kq/F8Gito7t69S3x8PGXKlFFrd3Fx4datW5pa7QdFRIUTEPSEht81oG6tevh4+3zV433s0uWZXVJSEq7lXOnavQuGhoZUcCvPz8OHsmXTVhISEpSOl278/QOo414Xa2tr1m1co3Qc8ZWy0nyOHzuBokWL0LHz552Lp62vOanNnUqlYtjQ4SxdvIw9B3ZTsFBBhVJ+mbi4ODwaN6NY8WL4PfMlOCyIMePH0LJZa+7fu8/0WdP4vtn3tGn1I4XyOrJu7Xq6de9Kzpw5lY6uETo6Osnno44cNYLCjoUxNTVl8rRJWFiYs33rDo1n0FhB88+3LszNzdXaLSwsCA3N+ArV2tqasRPG4JDX/pOOYdrY2hD2n5xJSUmEh4Vja2erqZgaZW9vh6WV+l4nR8fCxMXF8fKl9n4C/LdrV6/jXtWd6jWqs/fgHoyNjZWOlK709fUBUhSg8fHxyX1ZSVaaz/PnLrB54xYWL1/8wf73rznq31Z79b9P83Za+JqT2txFR0fTqllr/jpxktPnT1GlamUFk36Zv0/+zU3PW8z+dRZ2dnaYmJjQ+6deFCpckA3rNmJsbMz8337F19+H5yFPWbdxLe/evaNAgZQn22qLtLZRe/v3ex3//T6jq6tLgYIFef5cy08K/piM+LRx8MAhijuVJCYmRq09NjYuxe76D6lStQo3PW8l7+IHuH7tBjExMVSrXjXd82YE5zLO3Lp5S+0rdr6+j8mVKxd2dnYKJksfd73u4tG4GcNGDuO3xQuy5CchExMTdHR0UpwvExERkeLDg7bLavO5fu16IiMjqVy+CvlsC5DPtgAXL1xk3tz5VK1YjSpVq3Dl8hW1+1w4fwE7ezscnRwVSv1lUps7lUrFj63aEhkZxd/nTlKkaBEFk345lSrxf/9V/8pyQoKKxMRELpy/yN8n/1brO3b0OLVq18qoiOkurW20ZKkS6Onpcf3ajeR+lUpFgL9/huyB01hB8895Kq9fv1ZrDw8PJ0+ePJpabbLKVSoRHRXFkIFDCQsLIyYmhsULl+D7yBeP5h5p3v+HH1ujr5+TcaPH8+bNG549e86oEaP5rlFDipcorvH8mvBTvz68evmKsaPG8e7dO7zueDFv7nx69u6htbu0/6FSqejZrRddu3dhwKD+SsfRGD09Pezt7fHz8yMqKur9i0VAADExMTg4OCgdL91kxfmcNXcmd729uHT9YvKtfIXy9OjVnT0HdtN/YD9OHP+THdt3Ehsby/VrN1g4fyEDBw/QqudnWnO3ZNFSHj3yZff+nZiZmSmQMH1UrV4FW1sbxo4aR2hoKDExMaxZtRbvh960aNWCixcu0rlDF+7fu09cXByTxk/m1cuXdO/VLe0Hz6TS2katrKzo1KUj06ZMx/OGJ9HR0UwaP5nIyCg6dGqv8XwaK2icnZ3R19fn5s2bau03btzAzc3tw3dKR9bW1hz98wiBzwMp7liSgg6F2bZlO9t2bqVylUr07d0Pc2NL8pjZANC3d3/MjS1xKeUKvD9T/+DRA9z0vEmhvI5UKleZwo6FWPfHWo1n/xoupVwxN7akWdMWan/37d0PRydH9h/ex9kz58hvV5DGDZrQrn1bpkyfrHDqr3fp4mU8b9xk0W+LMTe2VLv17d1P6XjpqkiRIlhYWODp6cn58+d59eoVZcuWxdDQUOlo6SYrzqeFhQX58uVVuxkYGGBqaoqdnR2Vq1Riw+b1zJw2ExsLO35s3Za+/fsyWMtOlk1r7lYs+x3/J/7ksy2Qol+bmJubc+DIfsJCwyhfpgJ2Vg6sWLaCbbvev8cMGjKQlj+0on6dhtjnycuF8xc5cuLwJ30pRUmpvYd8yjY677dfadbCg2ZNmmOfJy+n/z7D0T+PJB+O0iSdpA/9qlw6mThxIteuXWPRokXY2dmxefNmFi9ezMGDB8mb98PfSY+NjcXLy4uiJZ3QN8h65wT8w1DXiBhV1vohtA/JTuPMLrL6fGanbVbGmXVkh3HGxcbhc98XZ2dnDAwMUvSnfTLJVxg9ejSzZ8+mXbt2REZGUrJkSVatWvXRYkYIIYQQ4ktotKDR19dn7NixjB07VpOrEUIIIUQ2l2Wv5SSEEEKI7EMKGiGEEEJoPSlohBBCCKH1pKARQgghhNaTgkYIIYQQWk8KGiGEEEJoPSlohBBCCKH1pKARQgghhNaTgkYIIYQQWk8KGiGEEEJoPY1e+uBrVK5QleDgYKVjaExYWFi2uaBhdhlndpEd5jM7jBFknFlNVh+njq5uqv2ZtqC5fP1ilr7atqWlJeEVsvbGB5B04hmBUf5Kx9A4B6OCWf5Kt5A9ruibHcYIMs6sJjuMM04Vl2q/HHISQgghhNaTgkYIIYQQWk8KGiGEEEJoPSlohBBCCKH1pKARQgghhNaTgkYIIYQQWk8KGiGEEEJoPSlohBBCCKH1pKARQgghhNaTgkYIIYQQWk8KGiGEEEJovWxd0ISEhNCre28K53fCxsIO92q1OfXXKaVjfbJqpd1IOOrPhI5Dk9t+/MaD60uP8GbfA7zXnWVq1xHkyKE+zcN/+Am/jReJOviIu6tO0q5Oc7V+Y0Mjlg6cjupYAJ3rt86QsXyKbRt34F6uDo6Wxahauia/L1qV3Ld3+z4aVGtMMdvSVHepzcyJc1CpVMn9/n4B9GzXB5eCFSiV14UfGrXjjqeXEsP4IhvXb6Rs6XKYG1tSsmhpFi5YBIBLKVfMjS3Vbqa5zMmlZ6xw4q9T3KkkJoZmKcbm4+2jdLR09fTpM7p26oZjgSJYmVrTuEGTLDHG1F5b4+PjmThuEqWKOWNlak2pYs6MGz2euLjUr9OTGT3xe0L9Og3JpWeM/5P/v2ZdYmIiC+b9RrkyFbAytca5hAsb1m1QMGn6OXfmXIrnpbmxJcb6JvTq3lvRbNm6oGndvA1BgUFcvHqegKAn1KrtTusWbQgMDFI6WpoM9Q1ZO2web6PfJbe5u1Rh/fD5zNi6mDytXGgxsScd6rZgbLtBycuMbNOPn5p0pPWUPlg0L8349b8ytv1A8uaxB6BYPkdurTgOkKIQUtK+HfuZMmY6U+dO4n7QHX5dNptNa7dw2/MOF89eYnCvYfQf1pc7ATdYtXk5u7fu5bdZ79/0Y2Ji+LFJe4yMjTh76ySX75/HPq8dnVt1IyYmRuGRpW37th2MGjGGeQvmEhwWxIqVy1izai03rnty+95NXkeGqd2atfCgU5eOSsf+aktXLE4xtqLFiiodK92oVCpafN+CkOAQzl06i3+gHxUrVaRpIw+t2C5Tk9pr6/QpM1i3Zh1btm8iJPwFW7ZvYuP6jcyaPlvp2J9l39791KpemwIF86fom//rAqZPmcFvi+bzIjSQJcsWMWrEGI4dOaZA0vRVw71GiuflA997WFhY0LFTB0WzafQd6+nTp3Ts2JHixYvz7NkzTa7qs0VERFCyVAnmzJuNnZ0dhoaG/DxiKJGRkVy9clXpeGma3m0kD54+4qbvveS2AR5dOXzlJDvPHCIuPg6vJw+Yt2slA5p1RUdHB/2c+oxs8xMjV03nmvctYuNj2XX2EKW6f8PzV++LODtLGwYsGc+AxeOUGtoHzZ+5kL6De+NetyYGBgZUc6/K6Rt/4VKuDGuWr6dOg9o0bdEYAwMDSjqXoNeA7qxZvp7ExERCXrykcvVKTJg5DjNzM0xMTejZvzvBL0J49OCR0kNL04ypMxkybDB169XFwMAA99ru3PS6QfkK5VIsu3/fAc6dPc+suTMVSCo+h/dDb7zu3GXcxHE4ONiTO3duxk8aR3x8PAcPHFI63hdL67X1xvUb1HCvSVnXsujq6lLWtSw1a7lz7eo1paN/lvCwcP78+wTt2rdN0bdn115++LE17rXdyZkzJ7W+qUX3nt1YtmS5Akk1b0DfQbRs3YKatWoqmkNjBc2JEydo06YNDg4OmlrFVzEzM2P5ymWUKFkiuc3vsR8A+fLlUyrWJ6leuiKd6rWiz2+j1NqrlCzPlYc31dquPPAkj5klRfMWpnwRZyxMzNHPqc/1pUeI2HufK4sP8m35/98Iz9y+xJErJzNiGJ8sOCgEnwePMMptTLNvW1Hczpm6FRuwZ9s+AG5c8cTVzVXtPq5uZQkPDefxIz8KFMrPgt9/xdLKIrk/4MlTdHV1sbW3zcihfLagoCAe3H9A7ty5qeP+LTYWdri5VmTrlm0plo2OjmbIwKHMmD0dc3PzjA+bznbt2E25MhWwtbSnWqXqHNh/UOlI6UpHRwd4f3jiHzly5MDS0oLrV68rFeurpfXa2rxlc878fZprV6+jUqm4c/sO586cpUWr5h97yEypS7fOH91jqKOjozavAFZ5rLh+7UZGRMtQhw4c5tKFS0yZPlnpKJoraF6/fs2mTZvw8PDQ1CrS1Zs3b+jdow9Nvm9CBbfySsf5qFwGhqwdPo+fV0wmKDRYrc/azJKwt6/V2l69CQfAxjwP+a3fF5fdGrSh1eTe5GtXkZM3L3BwyjqcHAplRPwvEvg8EIA/Vm9ixoKpePpeoV3XH+nfbRCXz18h7FUY5hZmavextLIEIPRlaIrHCwp8wbhhE+nSuxPWttaaH8BXePb0OQCrV65m4ZIF+D3zpWv3rnTt2I1zZ8+rLbtk0VKsrCz5oU3mOe/pS5Up40yx4sU4fvIoPk8e4tHMgzYtf+TypStKR0s3RYsVpbRzKSZPmMzTp8+Ijo5m2ZLlPPb1IzQ05Xarrf772tqlW2e6du+Ke7VamBiaUblCVdq0bUPnrp2VjppumrXwYMe2nZz88yRxcXHcuO7J+rUbCA0NJSkpSel46SYxMZHxY8fz84ihmJqaKh1HcwVN69atKVy4sKYePl35+wdQx70u1tbWrNu4Ruk4qZre7Re8nz1m/fEdn3W/pKSk5E+E07YsxO9FAG+j3jF6zUzC30XQ9pvMW3j+8wLQvW9XSjqXwMjYiO59u1K2vAvb/9iZ6n3/GfM/vG7dpWnt5lSvVY0JM8dqLHN6+Wfsffv3xbmMM8bGxvQb0JfybuX5Y8MfycvFxsYyf+4Cho0clmLM2mjn3h3M/nUW1tbWmJqaMnL0CMq6urB29Vqlo6UbXV1ddu7ZQe7cualSoSoupVwJfhFMvQb1yJkzp9Lx0sWHXlvn/7qALZu3cursScLfhXLmwmn27d3P9CkzFE6bfgYNGcjAwQPo99MACtgXYtrkafTs3YOcOXNmiefnP/bu2Ufg8yD69FX2ZOB/ZJ6zPhVy7ep13Ku6U71GdfYe3IOxceb9dkj10hXp9G0res0f+cH+4NevsDK1UGvL87+/X4S/JPB/e3RC/7fXBt5X2E+Cn5HvfycFZ0b/HBaysDRXay/oWICg50HksclDeNhrtb6w0DAAtT0wfx09RcsGbejQvR0LV81DV1dXo7nTg73D+3n5Z4/TPxwdHXn+7Hny38eOHic6OprGTRplaL6M5OjklLy3LqsoVLgQO/fu4HnIU3z8HjJxygSePn1KgQIFlI721T722vrbvN/o/VMvKlephIGBAW4VK9Cnb+8sdX6Jnp4e4yaO5b7PXV6EBrJr3050dHTIXyDlCcTabMumrTRu2ghDQ0OlowDZvKC563UXj8bNGDZyGL8tXpDpPxV1/+5HjA1zcWvFcV7uvM3LnbepXtqNET/8xPWlR7hw9xpVSqgfLqvhXInA0Bf4Bj7hnr838QnxVCzmmtyfI0cOCtnmw+/F0wwezaezs7fFwsqCm9dvq7U/8fUnf6H8uFUpz40rnmp9Vy9cw9bOhkKOBQE49/cFfurcn1+XzWbwyAEZlv1rOTjYY2VlleKcise+vhQqXCj57107dlH32zqZuiD/VE/8njB4wBBev36t1v7wwQMcizgpE0pDdu/aw8MHD5P/DgoK4pbnLdxruyuY6uul9tqqUiWS+K+fVABISEhIcc6JNrt963aKE7uPHj5KLS2f1397+/Ytx48ep6lHU6WjJMu2BY1KpaJnt1507d6FAYP6Kx3nkwxdPhmnzjVw7dMg+XbN+zbLD26k0ZhOLNizmgZutfihVlP0c+pToZgLP7fqxbydKwEIe/uatce2M6HjEMoVccZQ35ApXYZjbGjE+hOfdwgrI+nq6tJrQA/WrdjAmZPniI2NZd3vG/G6dZeO3dvTo183Tv95hn07DxAbG8utG7dZsXAlvQb0QEdHh8h3kQzu9TNjp42iSXPt2oOhq6vLwMEDWL50BSf/PElsbCwrlv3OTc9b9OjVPXm5K5evUta1rIJJ04+NrQ0H9h9kUP/BhIaGEhkZyfQpM/DxfkTffn2UjpeuNqzdwIC+gwgNDSU0NJQ+PX6ipnsNqlaronS0L5bWa2uzFh6s/H01njc8k08KXr1yDa3btFIgrWZ4e/vQqV1nzp+7gEqlYuWKVZw5fZZBQwYqHS3d3PK8RVxcHK6Z6HVHT+kASrl08TKeN25y1+seixcuUetr16EtS1cs+cg9lfP6XQSv30WotcXGx/Em6h3B4S8JDn/Jj9P7MbnTz2wYsYDg169YuHcNv+5ckbz8gCXjiI6L4cj0jZgZm+D56C7fDP+BF2EhAPw+ZDYdv22RvPzKIbNZPmgG/sHPKdGtVsYM9AMGDOuLKiGBoX2GE/oyFKeijvyxZx3OZUsDsHT9IuZOmcfgnj+TxyYP3fp2pfegngAcPXCcoOdBTBwxmYkj1M/EHzhyQKbfYzP8l2EkJCTQq0cfXoa8pFjxouw7uEetgAkKDMLaJnOf4PypjIyMOHT0AGNHjcO1dHkiIyNxLefK8ZPHKFa8mNLx0tWylUvp17s/JYuURldXl0ZNvmPu/DlKx/oqab22zpk3G1NTUzq07UTg80DMzc1o16EdY8aPVijxl3Ep5UqAf0DyniWXUq7o6Ogkv3888n5Eh7YdeR3+mtLOpdl/aG+W2n4Dg97/1Edmet3RSdLwKdcXLlyga9eu/PXXX5/0dejY2Fi8vLwoWtIJfQN9TUZTlIN1PsIrGCkdQ+OSTjwjMMo/7QW1nINRQWJUUUrH0DhDXaMsP87sMEaQcWY12WGccbFx+Nz3xdnZGQMDgxT9GttD06BBAwIDA5O/pdGwYUN0dHTw8PBg6tSpmlqtEEIIIbIhjRU0x45p/088CyGEEEI7ZNuTgoUQQgiRdUhBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitJwWNEEIIIbSexq7l9LUMdHNhoJvyappZicX1rH1l1H84GBVUOkKGMNTN+ldPh+wxzuwwRpBxZjVZfZw6urqp9mfagiZWFU2SSqV0DI0JCwvjUshppWNonKWlJYcf7FE6hsZVsalFjCrrF6hZ/QXzH9llLmWcWUd2GGecKi7VfjnkJIQQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HrZuqAJCQmhV/feFM7vhI2FHe7VanPqr1NKx0pToH8QfZsPoqptbYICgtT6ju/+i87f9qSu43e0rtKe5dNXofrXNbGePwlkVLfxNCrVjPrFmtC/5VAe3vZO7o98G8ncUQtoVv4H6hRuSJvqHfljydYMG9u/pTbOXWv30rZmZ+oUbkgLtx9ZOXstiYmJyf1nj56ne8OfqOv4HXUdv2Nou5H43n/8yY+fWdy7e4+WHq3Ia5Mfc2NLqlasxoH9BwHo2a0XxvommBtbqt3Wrl6nbOh0oFKp8Pb25tKlS5w9e5YbN24QFhamdKx0l9r8ZhUXzl/EWN+EqZOmATB10jSMcuZOsd1OGj9Z4aSf74nfE+rXaUguPWP8n/gnt8fFxTHml7EUcyyBubElhfM7MXrkGGJjYxVMqxn/nV8labSgCQ0NZdSoUdSoUYPy5cvzww8/cPHiRU2u8rO0bt6GoMAgLl49T0DQE2rVdqd1izYEBmbONzeAvw+fpUejvtjls0vRd+PCTaYMnEGnge05cn8fM9ZM4diuE6ybvxGA2JhYBrb+mVzGudh2cSN7rm/DxsGaYR1GERvz/ok295cF3Dh/k0U753Hi0SFGzBrKqjlrObjlSKYZ554N+1k+fSXDZw3hxKNDTFg6hq0rdrBj1W4A7ly9y+geE/ihRwuOeR9g97Wt5NTPyZC2I5KLntQeP7OIioqifp2GODo58sD3HsFhQTRr3oy2rdtx/959ANp1aMvryDC1W9fuXZQNng58fHyIiIjAxcWFatWqYWdnx507d4iKyjoX3/uU+dV20dHR9O7RBxMTE7X2GjVrpNhuJ0wer1DKL7Nv735qVa9NgYL5U/QNGzKco0eOcvDIfsLevmLH7u1sWLeRaZOnK5BUcz42v0rRaEHTt29fQkJC2LNnDxcvXqRy5cr07duX4OBgTa72k0RERFCyVAnmzJuNnZ0dhoaG/DxiKJGRkVy9clXpeB/1JvwNy/cvpGGrein6dq7eTbW6lan7fW30DfQpUsqRH3u3Zsfq3SQmJhIaHIZrVRcGTeqLiZkJxibG/Ni7Fa+CQ3niEwDA/VsPqV6vKvkd86Grq0uFGuUoUtKRe54Z+wKb2jjjY+PpN64P5au5oqurS9lKZXCrUY7r5zwByGWciwmLR9OgVT309PQwszSjcZuGvAx6xevQiDQfP7OIjIxkyvTJTJo6ERMTEwwMDOjTrzcqlYq7d+8pHU9j4uPjCQ4OplChQhgZGaGrq4uDgwPGxsYEBgYqHS/dZIf5HT9mAsWLF8OlrIvSUdJdeFg4f/59gnbt26boq/NtHVavW0Wx4sXIkSMHbhUrUK1GNW7fuq1AUs3JbPOrsYLm7du3ODk5MXr0aKytrTEwMKBnz55ERUVx+7byk2pmZsbylcsoUbJEcpvfYz8A8uXLp1SsNH3fvjEFnFJ+IgDwun6PkuVKqrWVKleSiLA3PH38DIeC9oxbOAozS7Pk/kD/IHR1c2BtawXAN01qceboOZ74+JOYmIjnxVv4PXxC7cbumhvUB6Q2zh96tqRZp6bJfyclJRH09AU2ea0BKFLKkW+b1Unue/4kkB2rd1O1bmUsrS3SfPzMwtramq7du2BkZAS83+M5Y+pM8ubLyzd1agPgdceLb2rWxc7KgTIlyzJn5hy1Q4za6O3btyQlJWFqaqrWbmJiwps3bxRKlf4+ZX612flzF9j8xxYWLl2You/58+c0btCEvDb5KVGkFL8MH0V0dLQCKb9cl26dKVqs6Af7mjX3wLWcK/C+QD925BjnzpyjXcd2GZhQs1KbX6XoaeqBTUxMmD5dfffa06dPAbCzy3y7+d+8eUPvHn1o8n0TKriVVzrOF3kdGoGphfquP3Or98VL+KvXFCxSQK0vJOgl88YspGW35ljaWALQa2Q3XjwLpm2Nzujo6KCrp0u/cb2pVMstYwbxBdb8up4Xz4KZtW6qWrvnxVsMaDUUVYKKes3qMGrecIUSfj0zIwvi4uKo4FaBQ0cPYGVlRaFChXj79h2TpkygUOFCnDj+J106dEUnRw6GjfhZ6chfLD4+HgA9PfWXp5w5cxIXF6dEJI370Pxqs6ioKHr36MPMOdNxcLBX67N3sMfRsTATp0zEpWwZLl+8TMd2nYiKimLhkt8USqwZfXv3Y92a9ZibmzN91jR+aNNa6UjpIrX5VVKGnRT87t07Ro0aRd26dSlTpkxGrfaT+PsHUMe9LtbW1qzbuEbpOBqho6P+t7eXDz0b9aVCjfIMnNQ3uf3X0b/he8+XP/5ew6knR5m3eRZr52/k0LajGZw4bSqVivljF7F91W5+3TQL+wLqT6xyVcty9tmfbL/0B7GxcfzUbBDxcfEKpf06EVHhBAQ9oeF3Dahbqx4+3j6MGT+arTs2U7xEcQwMDGjStDHdenRlzaq1SscVn+lD86vNxo+dQNGiRejYuWOKvu49u3HgyH4quJUnZ86c1HCvwbCRw9iwbiMJCQkKpNWcpSuW8DoyjPV/rGXS+EnM/3WB0pHSRWrzq6QMKWieP39O27ZtsbKyYu7cuRmxyk927ep13Ku6U71GdfYe3IOxsbHSkb6YpbUFb8LUd8n/c87IP3tgAC78eYmfPAbRrGNTJiweja6uLgAxUTHsWbefToM64FTSEQNDAyq6V6BBi2/ZtWZPxg3kE8RExzKi0xiunL7GqsNLKVOx9AeX09HRIX/hfIxZMJKHt725dPJKBidNP9bW1oydMAaHvPasXLHqg8s4FXEi8Ll2n2eir68PkOLNLT4+PrkvK/qU+dUG589dYPPGLSxevviT7+Pk5EhsbCyvXr3SYDJl6OvrU69BPYYMG8KcmZnr/e9LfMn8ZhSNFzS3b9+mdevWVKhQgd9//z35eHFmcNfrLh6NmzFs5DB+W7yAnDlzKh3pq5RxK43XdfWTCW9duUMeWyvyFcoLwLWzNxjXaxJjFoyg69BOassmJiWSlJREoipRrV2lUpGYmKTZ8J9BpVIxqts4oqOiWXloCfkd1c95WjZtJSM6jVFri//foQpdPd0My/m1Dh44RHGnksTExKi1x8bGoaury+iRY7h8Sb1Ae/DgIU5FHDMyZrozMTFBR0cnxfkyERERmJubKxNKA1Kb3/8ebtMm69euJzIyksrlq5DPtgD5bAtw8cJF5s2dT9WK1Zg1fTZHDqnv8X3w4CG5c+fG1tZWodTpJyEhAVfn8mzZrP5zF7Gxsehp0evPx6Q1v0rSaEHj7e1Nz5496dWrFxMnTsxUBYNKpaJnt1507d6FAYP6Kx0nXbTp1YrLf1/lz70niYuN4/7NB2xZto0f+7RGR0eHqMgopgycQb8JfajTtHaK+xsZG1Gpthublm4lwPcpCQkJ3Lp8mxN7TyafZJsZbF+5m6ePnzH3jxnkNs2dor9SLTfOn7jIng37iY2JJSIsgkUTl5HHLg8ulZwVSPxlKlepRHRUFEMGDiUsLIyYmBgWL1yC7yNfmrVohq/vY/r/1B/vh97Ex8dzYP9B1q9Zz6AhA5WO/lX09PSwt7fHz8+PqKgoVCoVAQEBxMTE4ODgoHS8dJPa/Ho091A63hebNXcmd729uHT9YvKtfIXy9OjVnT0HdhMaGkr/vgO4fu0GCQkJnDtzjvlz5zNw8AB0/ntsXAvp6elRsZIbUydN49bNW6hUKm5c92TFst9p0aqF0vG+WlrzqySdpKQkjXz0VqlUtG7dmmrVqjFs2LBPvl9sbCxeXl4ULemEvoHmdi+fP3eBb2vXQ19fP8WTqF2HtixdsURj6wYw1DXiUsjpz75fm2odefHsBYmJSSTEJ5BTPyc6OtCwVX1GzRvO34fOsHLWGp76PcfS2oJmnb6n86D26OjocHTncSb1m46+QcrCssvgjnQd2onXoa9ZPmM1l09dIexlGJbWlrTo4kG7vm2SD019jkYlmnP4wecfrkptnNfPe/Li6YsP7m05HXACgLPHLrBy9hr8HwVgZJyL0uVL0XdsLxxLFE7z8b/k5OEqNrWIUaX/b6Tcu3uPUSNGc+H8RXLkyEGx4sX4ZfRIGjdtREREBOPHTODQwcO8evmKAgULMOKXYXTo1CHdc/zDUDdj9rAmJibi6+tLSEgIKpWK3Llz4+TkhJmZWdp3TgeamMsPSW1+Nc1Q1yjDxlm/TkPca9Vk7IQxxMXFMW3ydLZv20FQYBC2drb81K8PAwb1/6LXmLRoapwupVwJ8A8gMTEx+XCojo4O7Tq0Ze78OUyfMoNNf2wmPCwcewd7WrVuyehxo8iVK1e6Z4GMnc//+vf8alJcbBw+931xdnbGwMAgRb/GCppr167Rvn17cubMmaJg8PDwYOrUqR+8X0YVNEr70oJG23xpQaNtNFXQZDYZVdAoLbvMpYwz68gO40yroNHYgVo3NzcePnyoqYcXQgghhEiWra/lJIQQQoisQQoaIYQQQmg9KWiEEEIIofWkoBFCCCGE1pOCRgghhBBaTwoaIYQQQmg9KWiEEEIIofWkoBFCCCGE1pOCRgghhBBaTwoaIYQQQmg9KWiEEEIIofU0di2nr2WgmwsD3ZQXn8pKqtjUUjpChmhUornSETQuLCws21y4MTvILnMp48xasvo4ddK4GnumLWhiVdEkqVRKx9CY7HBlVHj/Rp8dxmlpaUngy2dKx9C47LDdZocxQvYaZ3aR1eczThWXar8cchJCCCGE1pOCRgghhBBaTwoaIYQQQmg9KWiEEEIIofWkoBFCCCGE1pOCRgghhBBaTwoaIYQQQmg9KWiEEEIIofWkoBFCCCGE1pOCRgghhBBaTwoaIYQQQmi9bF/QPPF7Qv06DcmlZ4z/E3+l43yRkJAQenXvTeH8TthY2OFerTan/joFwNRJ0zDKmRtzY0u126Txk9Ue4++Tf1PcqSTFnUoqMYRPktpcbdu6naoVq2FtbotzCRcmjJ2I6l/XArt39x6tmrWmgH1BbCzs+LF1O168eJHRQ0jTvbv3aOnRirw2+TE3tqRqxWoc2H8QgDdv3jB44FCKFi5OHjMbypYux7y589Xuv3DBIsqWLoeVqTVFCxdnYL9BvH79WoGRfJ3iTiUxMTRLsd36ePsoHe2rfWw7jo+PZ+K4SZQq5oyVqTWlijkzbvR44uJSv35NZvSxMcbFxTHml7EUcyyBubElhfM7MXrkGGJjYxVMm35UKhXe3t5cunSJs2fPcuPGDcLCwpSOle5Se89RkkYLGh8fH/r06UPlypUpU6YMzZs3588//9TkKj/Lvr37qVW9NgUK5lc6yldp3bwNQYFBXLx6noCgJ9Sq7U7rFm0IDAwCoEbNGryODFO7TZg8Pvn+40aP56fe/ShZsoRSQ0hTanN19vRZenbtxbCRw3gWHMDWHZvZsnkrM6fNAiAiIoLGDZtiYmrCrXs3efj4PqamJrRp2Tajh5GqqKgo6tdpiKOTIw987xEcFkSz5s1o27od9+/dZ/CAIZz5+wxHjh8iOCyIhYsXMHXSNDas2wDAujXrmThuEgsXLyAk/AVHjh/i3NnzDBsyXOGRfZmlKxan2G6LFiuqdKyvktp2PH3KDNatWceW7ZsICX/Blu2b2Lh+I7Omz1Yg6ZdLbYzDhgzn6JGjHDyyn7C3r9ixezsb1m1k2uTpCiRNfz4+PkRERODi4kK1atWws7Pjzp07REVlrYtGpvWeoxSNFTTR0dF06NCBAgUK8Ndff3H9+nXq16/PwIEDefTokaZW+1nCw8L58+8TtGufud7YPkdERAQlS5VgzrzZ2NnZYWhoyM8jhhIZGcnVK1c/6TFy587N5esXcS3nqtmwXyG1uVq6ZDkNvmtAy1YtMDAwwLmMMwMHD2DZkuUkJiZy8fxFXgS9YPqsaVhYWGBhYcH8hfPwvOHJ1SvXFBjNh0VGRjJl+mQmTZ2IiYkJBgYG9OnXG5VKxd2797hx3ZPvGjWkSNEi6OrqUuubWjiXceba1esA3Lh+g9LOpan1TS10dXUpUrQIjRp/l9wvlJfadnzj+g1quNekrGtZdHV1Ketalpq13Ll2NfNso58itTHW+bYOq9etoljxYuTIkQO3ihWoVqMat2/dViBp+oqPjyc4OJhChQphZGSErq4uDg4OGBsbExgYqHS8dJMe7zmaotGCZtiwYQwZMoTcuXOjr69Phw4dknfJZQZdunXW+k98ZmZmLF+5jBL/2rvi99gPgHz58gHw/PlzGjdoQl6b/JQoUopfho8iOjo6efmRo0dgamqascE/U2pzdeXyFSpWdFNrc6voRmhoKI98HqGjowNAYmJScr+RkRG5cuXi+rXM82ZvbW1N1+5dMDIyAiA0NJQZU2eSN19evqlTm+YtmnHwwCEePnhIYmIi586c4/69+3g0+x6A7z2acu/uPf468Rfx8fE88XvCkcNHaNmqhZLD+mK7duymXJkK2FraU61S9eRDb9oste24ecvmnPn7NNeuXkelUnHn9h3OnTlLi1bNMzjl10ltjM2aeyR/cIqPj+fYkWOcO3OOdh3bZWBCzXj79i1JSUkpXktNTEx48+aNQqnS36e85yhFT1MPbGlpSevWrZP/Dg8P5/fff8fOzo6qVatqarXZ3ps3b+jdow9Nvm9CBbfy3PS8iaNjYSZOmYhL2TJcvniZju06ERUVxcIlvykdN128evkKC0sLtbY8eawACAl5SdXqVbGzs2XUiFH8umAuhoaGzJk5l/j4eEJfhSoROU1mRhbExcVRwa0Ch44ewMrKivGTxhEQEICrc3l0dHTQ09Nj2syp1K1XF4Bv63/LjNnTaf59SxISEkhKSqLVD60YM360wqP5fGXKOOPo5MjKtb9jYGDAssXLadPyR06dPUnlKpWUjqcRXbp1xu+xH+7VaiW3DRw8gM5dOyuYSjP69u7HujXrMTc3Z/qsafzQpnXad8rk4uPjAdDTU39bzZkzp1aeB/Wp/vueo6QMOSnY2dmZKlWqcPXqVdasWYOFhUXadxKfzd8/gDrudbG2tmbdxjUAdO/ZjQNH9lPBrTw5c+akhnsNho0cxoZ1G0lISFA4sebp6OhgamrK3oN7CAkOoUyJslStWB0bWxtKlS5Fzpw5lY74QRFR4QQEPaHhdw2oW6sePt4+DBn0M153vLh28wphb1+x7+AeZk6bxR8b/gBgx/adTBw3iV17dxD29hU3bl/jsa8vP/Xqq/BoPt/OvTuY/essrK2tMTU1ZeToEZR1dWHt6rVKR9OY+b8uYMvmrZw6e5Lwd6GcuXCafXv3M33KDKWjpbulK5bwOjKM9X+sZdL4Scz/dYHSkcQX+NB7jpIypKDx8vLi4sWL1KpVi3bt2uHn55cRq81Wrl29jntVd6rXqM7eg3swNjb+6LJOTo7Exsby6tWrDEyoOTa2NoSFqn+T4NX/9rzY2dkCUNa1LEf/PMKL0EC8Htym/8B++D/xJ38mPiHc2tqasRPG4JDXnpUrVrFy+UqG/zKc0s6lMTQ05Ju639CmXRuWLV0BwKIFi2ndphX1GtTD0NCQkqVKMvyX4Wxc/wdv375VeDRfz9HJicDnWedchP/6bd5v9P6pF5WrVMLAwAC3ihXo07c3y5YsVzqaRujr61OvQT2GDBvCnJlzlY7z1fT19QFSfFCMj49P7stKPuc9J6Nk2Ne2LS0tGTBgALa2tmzdujWjVpst3PW6i0fjZgwbOYzfFi9Q2+swa/psjhw6qrb8gwcPyZ07N7a2thkdVSOqVK3ClctX1NounL+Anb0djv8r3rZs3qp2Bv7VK9cIDQ2lpnvNjI77UQcPHKK4U0liYmLU2mNj3++uTkpKUvsqOoAqIYGkxMT3/5+oStH/z4trUlIS2uKJ3xMGDxiS4uvmDx88wLGIkzKhMoBKlUjiB+Yv8X/zq+0SEhJwdS7Pls3qr/+xsbHo6ekqlCr9mJiYoKOjk+J8mYiICMzNzZUJpSGpvecoSWMFzV9//UWdOnVS/L5AXFwcurrav/FmFiqVip7detG1excGDOqfoj80NJT+fQdw/doNEhISOHfmHPPnzmfg4AHJJ8tqu/4D+3Hi+J/s2L6T2NhYrl+7wcL5C5PHaGBgwJyZcxk57BciIyN5+vQZg/oPomPnDuTLl1fp+MkqV6lEdFQUQwYOJSwsjJiYGBYvXILvI19atm7Jt/XqsuDXBfh4+5CQkMD5cxfYsW0nrX5oBbw/4XLn9l2cPnWahIQE/B77seDX32jQsH6mP+n732xsbTiw/yCD+g8mNDSUyMhIpk+ZgY/3I/r266N0PI1p1sKDlb+vxvOGZ/JJwatXrqF1m1ZKR0sXenp6VKzkxtRJ07h18xYqlYob1z1Zsex3Wmjpiev/pqenh729PX5+fkRFRaFSqQgICCAmJgYHBwel46WbtN5zlKSxk4LLlStHdHQ0kydPZvjw4eTKlYutW7cSEBBA/fr1NbXaz+JSypUA/4DkT0AupVzR0dGhXYe2LF2xROF0n+bSxct43rjJXa97LF6onrldh7YsWDSfXLly0aFtR4ICg7C1s2XIsCHJG6K/fwBlS7kC7z9BqVQqzI0tgfe/A9KuQ+b49kFac7Vh83qmTJxCjy49sbG1oW//vgweOij5/pu3/cGAvoMo6FAYI6NctP6xNdNnTlNqOB9kbW3N0T+PMGrEaIo7liRHjhwUK16MbTu3UrlKJdZsWM3EcZNo3LApIcEh2NjaMHT4EAYNGQjAkJ8HAzBowBAC/AMwMjLCo/n3TJ42ScFRfT4jIyMOHT3A2FHjcC1dnsjISFzLuXL85DGKFS+mdLyvktp2PGfebExNTenQthOBzwMxNzejXYd2WndSd2pj/G3xAqZPmUGzpi0IDwvH3sGetu1+ZPS4UQqnTh9FihTB19cXT8/3RWnu3LkpW7YshoaGSkdLN2m95yj53qmTpMF90T4+PsyaNYvr16+TI0cOHB0d+emnn6hTp85H7xMbG4uXlxdFSzqhb5D1jjv+w1DXiBhV1vqxpQ/JLuN0sM5H4MtnSsfQuOwwn9lhjJC9xpldZPX5jIuNw+e+L87OzhgYGKTo19geGoCiRYuyatUqTa5CCCGEEEKu5SSEEEII7ScFjRBCCCG0nhQ0QgghhNB6UtAIIYQQQuulWdDEx8fz4sULAB48eMDevXvVLmwohBBCCKG0NAuaX375hZs3bxIcHMyAAQPw9vbml19+yYhsQgghhBCfJM2CJjg4mIYNG3L48GHatWvHiBEjiIiIyIhsQgghhBCfJM2CJi4ujqSkJE6cOEHt2rUBiIrK2j/eI4QQQgjtkmZBU6lSJSpUqIC1tTWFCxdm3bp1FC5cOCOyCSGEEEJ8kjR/KXjYsGH06tUr+eJ23377Le3bt9d4MCGEEEKIT5XmHprTp09z6tQpAH7++We6deuW/LcQQgghRGaQ5h6apUuXsmzZMk6fPk1iYiJ79uyhT58+Gr9itoFuLgx0U158KivJLhdNyy7jdLDOp3QEjQsLC8sW85kdxgjZZ5zZRVafTx1d3VT70yxoDA0NsbS05PTp03h4eGBsbEyOHPJ7fOkhq18ZFbLPFX3DwsKUjpAhLC0t8XrmqXQMjXIwKqh0hAyTHZ6b2eU1KDuMM04Vl2p/mpVJbGwsq1at4uzZs1StWpUnT57w9u3bdAsohBBCCPG10ixopkyZQnBwMDNmzMDAwIBz584xfPjwjMgmhBBCCPFJ0ixoihYtypgxY3BzcwPghx9+YMuWLRoPJoQQQgjxqdI8h2bv3r3MnDkz+deBc+TIQZUqVTQeTAghhBDiU6VZ0GzcuJEDBw4wdOhQVqxYwYEDBzAxMcmIbEIIIYQQnyTNQ04mJiZYW1ujUqkwMjKiTZs27Nq1KyOyCSGEEEJ8kjT30Ojq6nLq1Cns7e1ZtGgRRYoU4fnz5xmRTQghhBDik6S5h2b27NnY2dkxevRoQkJC2L9/P+PGjcuIbEIIIYQQn+Sje2gSExMBsLCwwMLCAoBJkyZlTCohhBBCiM/w0YKmVKlS6OjopGhPSkpCR0eH+/fvazSYEEIIIcSn+mhB8+DBg4zMIYQQQgjxxT56Dk1SUhJLly5FpVIlt/n6+rJs2bIMCZYRVCoV3t7eXLp0ibNnz3Ljxo0scU2eJ35PqF+nIbn0jPF/4q/Wt23rdqpWrIa1uS3OJVyYMHai2hxrs3t379HSoxV5bfJjbmxJ1YrVOLD/oNKx0l1W2G4fefvS9YceuBSsQEmHMjSp3YwTR/4CICEhgfkzfqN6mVoUsS5JTdc6rF2+Pvm+F85cJK9xIRwti6ndWjb4QanhfLGsMJef68L5ixjrmzB10jSlo6S71F57s5KQkBB6de9N4fxO2FjY4V6tNqf+OqV0rI8XNIsXL+bu3bvExf3/xaBsbW158OABGzZs+OwVXb9+nZIlS7Jo0aIvS6oBPj4+RERE4OLiQrVq1bCzs+POnTtERWnvBb727d1Preq1KVAwf4q+s6fP0rNrL4aNHMaz4AC27tjMls1bmTltlgJJ01dUVBT16zTE0cmRB773CA4LolnzZrRt3Y7797LW4VFt324TExNp79EJw1yGnPb8i5t+12jaogk92/bhkbcvcyb/ypb12/l90zIevvBizJSRTBgxmeOHTqg9zuMwb7XbrmPbFRrRl9P2ufxc0dHR9O7RJ0v+lllqr71ZTevmbQgKDOLi1fMEBD2hVm13WrdoQ2BgkKK5PlrQnDp1ivnz55MrV67ktty5czNr1iwOHz78WSuJiYlh9OjRGBsbf3nSdBYfH09wcDCFChXCyMgIXV1dHBwcMDY2JjAwUOl4Xyw8LJw//z5Bu/ZtU/QtXbKcBt81oGWrFhgYGOBcxpmBgwewbMny5JPAtVVkZCRTpk9m0tSJmJiYYGBgQJ9+vVGpVNy9e0/peOkmK2y3oS9DeRbwnNbtWmJhaY6BgQGde3UgPj6eu7fvoZczJxNnjaO0Syl0dXVp2LQBxUsV49yp80pHT1dZYS4/1/gxEyhevBguZV2UjpLuUnvtzUoiIiIoWaoEc+a9/wa0oaEhP48YSmRkJFevXFU020cLGkNDQ/T19T/YniNHmt/2VjNv3jwKFy5MyZIlPz+hhrx9+5akpCRMTU3V2k1MTHjz5o1Cqb5el26dKVqs6Af7rly+QsWKbmptbhXdCA0N5ZHPo4yIpzHW1tZ07d4FIyMjAEJDQ5kxdSZ58+Xlmzq1lQ2XjrLCdmtta02lahXZsn4boS9DiY+PZ+OqTVjmsaS6e1WGjxtKI4+GycvHxsYSHBSCQz4HtccZ1HMoFYpUpmwhN3q2/4nnz7SrCMgKc/k5zp+7wOY/trBw6UKlo2hEaq+9WYmZmRnLVy6jRMkSyW1+j/0AyJcvn1KxgFQKmqioqA/u9oyIiCAyMvKTV3Dt2jX27duX6b7yHR8fD4Cenvp50Tlz5lQ7zJaVvHr5CgtLC7W2PHmsAAgJealEJI0wM7Ign20Bzp09z6GjB7CyslI6UrrJKtvtyk3LeOr/DJdCFShsUYxFc5ayYuMS8tjkUVsuKSmJUYPGYpjLkPbd3n/yzW2Sm3JuZanfuB4X751l7187eRn8ks4tupGQkKDEcL5IVpnLTxEVFUXvHn2YOWc6Dg72SscR6ejNmzf07tGHJt83oYJbeUWzfLSg8fDwoH///jx58iS57cGDB/Tp04euXbt+0oNHR0czevRoRo4cia2t7VeHFZrzoa/oa6uIqHACgp7Q8LsG1K1VDx9vH6UjiX+Ji4ujfbPOFCnmhKfvFR4E3WHomMF0ad0D7/v/P1fR0TH06dSf86cvsOXARkxM35934VKuDAdP76Nxs+/Q19ensFMhps+fwv27D7h++YZSwxKpGD92AkWLFqFj545KRxHpyN8/gDrudbG2tmbdxjVKx/l4QdO1a1fq1q1L586dqVChAuXKlaN///60aNGCZs2afdKDz5s3j0KFCtGiRYv0yptu/jmc9t9PdPHx8R881JYV2NjaEBaq/g2KV69CAbCzy1oFp7W1NWMnjMEhrz0rV6xSOk66yQrb7fm/L+B16y4TZ43Dxs6G3Ca56dKrIwUK5mfbxh0AhL0Ko1WDNgQHBXPw7704FXVM9TELOxUC4EXgC03HTzdZYS4/xflzF9i8cQuLly9WOopIR9euXse9qjvVa1Rn78E9meIc2VSv5dS+fXvat2/Pu3fv0NHR+azA/xxqOnDgwFeH1AQTExN0dHR48+YN1tbWye0RERHkyZMnlXtqrypVq3Dl8hW1tgvnL2Bnb4ejU+pvGJndwQOH+HnwMG7d9cTQ0DC5PTY2LsUufW2WFbZb1f9OQFep1E9ET1AlkJiYyLu372j7fUccixbmt5XzUry57991kBeBL+g1oEdym8/D9+eAFfpfYaMNssJcfor1a9cTGRlJ5fJVktsiIiK4dvUahw4e4uLVCwqmE1/irtddPBo345cxIxkwqL/ScZJ90tm9uXPn/uzqa9euXURFRfH9999TuXJlKleuzI0bN1i1ahXNmzf/orDpSU9PD3t7e/z8/IiKikKlUhEQEEBMTAwODg5pP4AW6j+wHyeO/8mO7TuJjY3l+rUbLJy/kIGDB2j9IafKVSoRHRXFkIFDCQsLIyYmhsULl+D7yBeP5h5Kx0s3WWG7rVilAtY2eZg2biZhoeHExMSwae0WfL0f06RFI2ZPmksuo1wsWr3gg3sqDAz0mTZ2Jvt27Cc+Pp4nj/0ZM3Q8VWtWpmx57fn2TFaYy08xa+5M7np7cen6xeRb+Qrl6dGrO3sO7FY6nvhMKpWKnt160bV7l0xVzADoJCUlJWnigSMiIoiOjlZrGzRoEK6urvTo0UPtE8m/xcbG4uXlhbOzMwYGBpqIliwxMRFfX19CQkJQqVTkzp0bJycnzMzMNLref8So0v+3JlxKuRLgH0BiYmLyrmsdHR3adWjL0hVL2LtnH1MmTuGRjy82tjb07NWD4b8M01hBY6hrpJFxfsi9u/cYNWI0F85fJEeOHBQrXoxfRo+kcdNGGl+3oa6RxtfxDyW3W0tLS7yeeX7149y7c58ZE2Zz+8Yd3r55g1MxJ4aPG0r9xvUoYOqEjo4Ourrqn7fyFsjH2ZsnAdixaRcrFq7kyWN/DHMZ0rjZd4yePBIz86//N3AwKvjVj/GpsuJr0KeoX6ch7rVqMnbCGI2vK6Neg9J67dW0jBrn+XMX+LZ2veTx/ZumxxoXG4fPfd+P1gcaK2g+pGPHjlSqVIkBAwZ8dJmMLGiUptSLSUbKyIJGSRlZ0CgpvQqazCwjCxqlZZfnpowza0iroEnz5ILnz58za9YswsPD2bhxI9u3b6dSpUoUKlTos8Ns3Ljxs+8jhBBCCJGWNM+hGTduHB4eHvyzI6dw4cKMGzdO48GEEEIIIT5VmgVNfHw8devWTT5WVrFiRY2HEkIIIYT4HJ/0Lac3b94kFzQ+Pj7ExsZqNJQQQgghxOdI8xyafv368cMPP/Dy5UuaNm1KeHg4c+bMyYhsQgghhBCfJM2CpkqVKuzduxdvb+/3PzNeuHCW//aREEIIIbRLmgXNb7/99sH2QYMGpXsYIYQQQogvkeY5NLq6usm3xMRELl++zNu3bzMimxBCCCHEJ0lzD03//uo/baxSqVL9YTwhhBBCiIz2Sd9y+reEhAQCAgI0kUUIIYQQ4oukuYemVq1aatdriIiIyBQXlxRCCCGE+EeaBc3mzZuT/19HR4fcuXNjamqq0VBCCCGEEJ8jzYJmzpw5LFiwIAOiZD/Z5YKG2WWc2YVzvnJKR9CosLAwpSNkmOzy3JRxZg06urqp9qdZ0OTLl4+dO3dSrlw59PX1k9vz58//9elSEauKJkml0ug6lJQdrowKMs6sJiwsLMuP09LSkvAKWfuNASDpxLMsP5eQfZ6b2WGccaq4VPvTLGgOHz6cok1HR4e//vrry1MJIYQQQqSjjxY0+/fv5/vvv+fkyZMZmUcIIYQQ4rN99GvbO3fuzMgcQgghhBBf7LN/h0YIIYQQIrP56CEnT09PateunaI9KSkJHR0d/v77bw3GEkIIIYT4dB8taEqVKsW8efMyMosQQgghxBf5aEGjr69P3rx5MzKLEEIIIcQX+eg5NC4uLhmZQwghhBDii320oBk+fHhG5hBCCCGE+GLyLSchhBBCaD0paIQQQgih9bJVQRMSEkKv7r0pnN8JGws73KvV5tRfp5L7Vyz7nXJlKpDHzIYSRUoxZeJUEhMTFUz8Ze7dvUdLj1bktcmPubElVStW48D+gymWCwsLo3A+R+rXaahAyq/3xO8J9es0JJeeMf5P/NX6tm3dTtWK1bA2t8W5hAsTxk5EpaXXBkttnPHx8UyZOBUTQzOmTpqmUELNSOv5mplVK+1GwlF/JnQcmtz24zceXF96hDf7HuC97ixTu44gR47/fwmuXroiJ+dsJ3zPXSIP+HBm3i6ql674wcd3tC/Iu/3erB2eOb+JWtypJCaGZpgbW6rdfLx9gKy13W5cv5GypcthbmxJyaKlWbhgUYpl4uLiqFiuEsWdSiqQMH2kNs7M8t6ZrQqa1s3bEBQYxMWr5wkIekKt2u60btGGwMAgVv2+mgljJ7Jw8QKCw4JYs341i35bzNLFy5SO/VmioqKoX6chjk6OPPC9R3BYEM2aN6Nt63bcv3dfbdmfBw8jOjpGoaRfZ9/e/dSqXpsCBVNeJPXs6bP07NqLYSOH8Sw4gK07NrNl81ZmTpulQNKvk9o4Q0JCqFnVnQcPHmJubp7x4TQstedrZmaob8jaYfN4G/0uuc3dpQrrh89nxtbF5GnlQouJPelQtwVj2w0CoIBNXk7M2sKhy39h90N5bFqX5bbffY5M34iliXmKdawd9ivxqviMGtIXWbpiMa8jw9RuRYsVzVLb7fZtOxg1YgzzFswlOCyIFSuXsWbVWm5c91RbbvqUGTwNeKZQyq+X2jgz03unRguaOnXqULp0acqUKaN28/Pz0+RqPygiIoKSpUowZ95s7OzsMDQ05OcRQ4mMjOTqlavExsYybeZUataqia6uLtWqV6XWN7U4fep0hmf9GpGRkUyZPplJUydiYmKCgYEBffr1RqVScffuveTlDuw/yJnTZ+nctZOCab9ceFg4f/59gnbt26boW7pkOQ2+a0DLVi0wMDDAuYwzAwcPYNmS5Vq3xy21cYaFhtFvQF82bd2IvoG+Auk0J63na2Y2vdtIHjx9xE3f/3++DfDoyuErJ9l55hBx8XF4PXnAvF0rGdCsKzo6OuTIkYN+i8bw684VxMbHEhkTxe+HNmFilBsnh0Jqjz+weXdy5zLmwKU/M3hk6SMrbbczps5kyLDB1K1XFwMDA9xru3PT6wblK5RLXubGdU9WLPudgYMHKJj066Q2zsz03qnxPTRTpkzhzp07arfChQtrerUpmJmZsXzlMkqULJHc5vf4fWGVL18++g3oS/ee3ZL7kpKSCPD3J2/+fBme9WtYW1vTtXsXjIyMAAgNDWXG1JnkzZeXb+rUBt4fahrYdyBLli3CxMREwbRfrku3zhQtVvSDfVcuX6FiRTe1NreKboSGhvLI51FGxEs3qY2zRMkSdOzcMYMTZYy0nq+ZVfXSFelUrxV9fhul1l6lZHmuPLyp1nblgSd5zCwpmrcwT148Ze2xbcl99la2jPihL56PvLjpeze53cmhEJM6DaXr3KGoEjP3IdRdO3ZTrkwFbC3tqVapevJh76yy3QYFBfHg/gNy585NHfdvsbGww821Ilu3/P88xsXF0at7LyZOmUD+Apl3u01NWuPMTO+d2eqQ07+9efOG3j360OT7JlRwK5+if/qUGQT4P2XI0EEKpEsfZkYW5LMtwLmz5zl09ABWVlYADB30M9/W/5aGjbTz3Jm0vHr5CgtLC7W2PHnejz0k5KUSkcRXSuv5mhnkMjBk7fB5/LxiMkGhwWp91maWhL19rdb26k04ADbmeZLbCtjkJfbwYwK3XiePmQXfje5IfML7Q0s6OjqsHfYrC3av5vZj9cPHmU2ZMs4UK16M4yeP4vPkIR7NPGjT8kcuX7qidLR08+zpcwBWr1zNwiUL8HvmS9fuXenasRvnzp4HYNrk6eTJk4defXoqGfWrfMo4/03J906NFzRHjhyhUaNGVKhQgRYtWvDnn8rvJvX3D6COe12sra1Zt3GNWp9KpWLY0OEsXbyMPQd2U7BQQYVSfr2IqHACgp7Q8LsG1K1VDx9vH/bvO8CZ02eZM2+20vEUoaOjo3QE8ZlSe75mJtO7/YL3s8esP77js+6XlJSU/P8BIc8xaORIvrYV8XvxlEsL9yefQzOwWTdMjHIzbfPC9IytETv37mD2r7OwtrbG1NSUkaNHUNbVhbWr1yodLd38M299+/fFuYwzxsbG9BvQl/Ju5fljwx9cv3aDFct+Z/nKZVr9upPWOP+RGd47NVrQFCtWDEdHR/744w9Onz5NvXr16N+/Pzdv3tTkalN17ep13Ku6U71GdfYe3IOxsXFyX3R0NK2ateavEyc5ff4UVapWVixnerG2tmbshDE45LVn7uxfGdRvEIuXLtL6k/FSY2NrQ1homFrbq1ehANjZ2SoRSXyh1J6vmUn10hXp9G0res0f+cH+4NevsDL9z17D//39IjzlXsPnr4Lou3A0ZsYmtK/bAieHQkzsNJSuc4aSoEpI/wFkAEcnJwKfByodI93YO9gDYGllqdbu6OiI3+Mn9OreiwmTx1OocCEF0qWf1Mb5/Nn7vTeZ5b3zo9dySg/Lly9X+/unn37i+PHjbN++HVdXV02u+oPuet3Fo3EzfhkzkgGD+qv1qVQqfmzVlujoGP4+dxIzM7MMz5ceDh44xM+Dh3HrrieGhobJ7bGxcZz++wzBwe+/CvuPqKgo4uPjyWdbgIvXLpBfy84Z+pAqVatw5bL6ru0L5y9gZ2+Ho5OjQqnE50rt+ZrZdP/uR4wNc3FrxfHkNjNjEyoVd+X7qvW4cPcaVUqoHyqr4VyJwNAX+AY+4aemnejWsA0V+zVWW0ZfT58EVQLt6zTH2NCIE7O2JPflzvX+PLkmlb/FulXmuVTNE78nLJj3GxOnTFD74PTwwQNquNdULlg6c3Cwx8rKiutXr/O9R9Pk9se+vuTObcK9u/eZNnk60yZPByA2NpaoqCjy2RZg++5tVKteVanonyW1cZavUD5TvXdqtKD5kAIFChAcHJz2gulMpVLRs1svunbv8sEXxyWLlvLokS+Xr18kd+7cGZ4vvVSuUonoqCiGDBzKtJlTMTIyYtXvq/F95MuJU8fJX0D9678L5y/k8qUrbNr2B/b2dgqlTl/9B/aj3jf12bF9J997NMXrzl0Wzl/IoKGDtHrXb3aS1vM1sxm6fDLj1s1Va9sxbjkX711n9vZlFLLLz5lfd/JDrabsvXCMMoVL8HOrXszdsQKAkzfP82vv8UzqPIzZ25YCMKHjUBKTEjly9RShb8JZfXSr2uPP6zM+ed2ZiY2tDQf2HyQ8PJx5v/2KoaEhv81biI/3IzZv26R0vHSjq6vLwMED+HXOPGrVdqd6zeqsW7Oem563OHX2JA55HdSW371zNwsXLOLvc6ewts7zkUfNfFIb59IVSzLVe6fGCpqnT5+yZs0ahgwZgqmpaXL748ePqVjxwz8WpUmXLl7G88ZN7nrdY/HCJWp97Tq05fTfZ/B/4k8+2wIp7vs6MixFW2ZlbW3N0T+PMGrEaIo7liRHjhwUK16MbTu3UrValRTLm5qaYmBgQL582nVldZdSrgT4ByR/DdullCs6Ojq069CWpSuWsGHzeqZMnEKPLj2xsbWhb/++DNbCE7xTG2eBAgWYOf39b+vExsYyc/os5s7+FdCubfZD0nq+Ll2x5CP3VMbrdxG8fheh1hYbH8ebqHcEh78kOPwlP07vx+ROP7NhxAKCX79i4d41/LrzfUHz8Kkv3478kVk9RjOsVW9i4mK59fgeDUd34MmLpwC8jXqn9vhRsdHA+8NTmYmRkRGHjh5g7KhxuJYuT2RkJK7lXDl+8hjFihdj5rRZWWa7Hf7LMBISEujVow8vQ15SrHhR9h3cQ6XKKd/jLCzM0dXV1brXWvj4OMu6lqVdmw6Z5r1TJ+nfZ6Slo+joaBo0aEDFihUZO3YsBgYGrFmzhuXLl7N//34cHT+86z82NhYvLy+KlnTS+t8oSI2hrhExqiilY2icjDNryQ7jdLDOR3gFI6VjaFzSiWdZfi4he2yzkD3GGRcbh899X5ydnTEwMEjRr7GTgnPlysXatWuJjIzku+++o2rVqpw/f54//vjjo8WMEEIIIcSX0Og5NE5OTilODBZCCCGESG/Z9of1hBBCCJF1SEEjhBBCCK0nBY0QQgghtJ4UNEIIIYTQelLQCCGEEELrSUEjhBBCCK0nBY0QQgghtJ4UNEIIIYTQelLQCCGEEELrSUEjhBBCCK2n0UsffA0D3VwY6Ka8+FRWYqib9S+ABzLOrCY7jNPieta+yN8/ssNcgowzq9DR1U21P9MWNLGqaJJUKqVjaEx2uDIqyDizmuwwzrCwsCw/RgBLS0u8nnkqHUPjHIwKZov5zA7PzThVXKr9cshJCCGEEFpPChohhBBCaD0paIQQQgih9aSgEUIIIYTWk4JGCCGEEFpPChohhBBCaD0paIQQQgih9aSgEUIIIYTWk4JGCCGEEFpPChohhBBCaD0paIQQQgih9bJVQfPE7wn16zQkl54x/k/8k9vj4+OZOG4SpYo5Y2VqTalizowbPZ64uNSvG5EZnTtzDnNjyxQ3Y30TenXvzdRJ0zDKmTtF/6Txk5WO/tmKO5XExNAsxVh8vH0A2LZ1O1UrVsPa3BbnEi5MGDsRlRZeH+xj2y2A90Nvvm/kQR4zG+zz5KVju868evVKoaTpx/uhN62b/0AB+4LYWTngXq02hw8eUTrWV0ttm9Xm52blktUpaFYER8tiajdfn8cAPPL2pb1HJ4ralKJUXhd+6tyfsFdhH3ysRXOWkNe4EBfOXMzIIXyRtF6D4uPjmTJxKiaGZkydNE3htOnn6dNndO3UDccCRbAytaZxgybJY1ZStilo9u3dT63qtSlQMH+KvulTZrBuzTq2bN9ESPgLtmzfxMb1G5k1fbYCSb9ODfcavI4MU7s98L2HhYUFHTt1eL9MzZTLTJg8XuHkX2bpisUpxlK0WFHOnj5Lz669GDZyGM+CA9i6YzNbNm9l5rRZSkf+LKltt+Hh4TSo2xDXcq74Bvhw7dZVYmJiWLJwqQJJ009iYiJNG3lgmCsXN+964h/oR8vWLfixdVu8H3orHe+rfWybBe1+bs5ePIPHYd5qN6eijrwOj6B1wx9xdnXmus8l/rp6nNiYWFYtXZviMR7e82bVkpTtmdnH5jMkJISaVd158OAh5ubmSsdMNyqVihbftyAkOIRzl87iH+hHxUoVadrIg5iYGEWzabyg2b17Nw0bNqRMmTLUrVuXdevWaXqVHxQeFs6ff5+gXfu2KfpuXL9BDfealHUti66uLmVdy1KzljvXrl5TIGn6G9B3EC1bt6BmrZpKR8kwS5csp8F3DWjZqgUGBgY4l3Fm4OABLFuynMTERKXjfbLUttvVK9dgYWnJ5GmTMDMzI29eB3bs3qY1b4Af8/LlSwL8A+jQsR2WlpYYGBjQ+6dexMfHc+vWbaXjic+0ac1mzC3NGTVpBKZmptg72LFm20pGjP9ZbTmVSsWQ3sPoO7SPQknTV1hoGP0G9GXT1o3oG+grHSfdeD/0xuvOXcZNHIeDgz25c+dm/KRxxMfHc/DAIUWzabSgOXToELNmzWLcuHFcv36d6dOns23bNry8vDS52g/q0q1z8qeg/2resjln/j7NtavXUalU3Ll9h3NnztKiVfMMTpn+Dh04zKULl5gy/f93Wz9//pzGDZqQ1yY/JYqU4pfho4iOjlYw5ZfbtWM35cpUwNbSnmqVqnNg/0EArly+QsWKbmrLulV0IzQ0lEc+j5SI+kVS225PnzqNa7myDOo/mPx2BXEqWJRB/Qfz7t27DE6ZvmxtbalWvRpr16zn5cuXxMfHs3LFKvLkyUOt2u5Kx/tqH9tmQbufmwd3H6J2hW8pYV+GhtWbcOzgcQDOn76Ac1lnRg0eS5mC5alQtAqjBo8l8l2k2v2XzFuOrp4uPft3UyL+F/vYfJYoWYKOnTsqnC796ejoAKh9MMyRIweWlhZcv3pdqVjvc2jywZcsWUKPHj2oXr06+vr6VK5cmSNHjuDs7KzJ1X62Lt0607V7V9yr1cLE0IzKFarSpm0bOnftrHS0r5KYmMj4seP5ecRQTE1NAbB3sMfRsTCTp03myfPHrFrzO9u2bGPksF8UTvv5ypRxpljxYhw/eRSfJw/xaOZBm5Y/cvnSFV69fIWFpYXa8nnyWAEQEvJSibjp7tmz5+zbsx/Xcq74PHnI9l3bOHzoCMOGDFc62lfbsmMTAf7+FLAvhJmRBXNmzmHT1o3Y2NgoHe2rpLbNavNzs6RzSZyKObHr6DauPrzAdx4N6f5jb65fuUHg8yCO7D9KGVdnrj68wJptv/PnkZOMHz4p+f4P7j5k6bzlzF8+hxw5tOdMiNTmM6sqWqwopZ1LMXnCZJ4+fUZ0dDTLliznsa8foaGhimbT2JYTEhKCr68vRkZGtG3blvLly9O0aVMOHDigqVV+sfm/LmDL5q2cOnuS8HehnLlwmn179zN9ygylo32VvXv2Efg8iD59eye3de/ZjQNH9lPBrTw5c+akhnsNho0cxoZ1G0lISFAw7efbuXcHs3+dhbW1NaampowcPYKyri6sXZ36Mfh/PmFou6SkJFzLudK1excMDQ2p4Faen4cPZcumrVo3l/8WFxeHR+NmFCteDL9nvgSHBTFm/BhaNmvN/Xv3lY73VVLbZrX5ubluxyomzhqHlbUVJqYmDBrRn9Iupdi8ditJSUmUcXWmXZcfMTQ0pGx5F/oN7cPurXtJSEggISGBIb2HMXB4P4oUL6L0UD7Ll74GaTNdXV127tlB7ty5qVKhKi6lXAl+EUy9BvXImTOnotk0VtC8ePECgG3btjFx4kTOnTtH69atGTZsGNeuZa5zU36b9xu9f+pF5SqVMDAwwK1iBfr07c2yJcuVjvZVtmzaSuOmjTA0NEx1OScnR2JjY7PEt2McnZwIfB6Ija0NYaHq36J49er9pwc7O1sloqU7e3s7LK0s1docHQsTFxfHy5fauxfq75N/c9PzFrN/nYWdnR0mJib0/qkXhQoXZMO6jUrHS3f/bLMfos3PzUJOhXgR+AJbO1ssLM3V+goWLkBcXByhL8OSDzX1HtRTmaDpLLX5zCoKFS7Ezr07eB7yFB+/h0ycMoGnT59SoEABRXNprKBJSkoCoGPHjhQvXhwjIyM6deqEs7Mzu3fv1tRqv4hKlUjif77Om5CQoFUnj/7X27dvOX70OE09mqq1z5o+myOHjqq1PXjwkNy5c2Nrqz1v9E/8njB4wBBev36t1v7wwQMcizhRpWoVrlxW3+174fwF7OztcHRyzMCkmuNcxplbN2+pfRXd1/cxuXLlws7OTsFkX0elSvzff//7nFRp9XMyrW1WW5+bAU+eMnrIOCJeR6i1P3rwiEJOhSjpXAKvW/fU5vPJY38McxliY2fNlnVbeXjPm7KF3XAuUA7nAuUA6NamF2N/npChY/kcac1nVrZ71x4ePniY/HdQUBC3PG/hrvA5bhoraP451m1hoX4eQ4ECBQgODtbUar9IsxYerPx9NZ43PJNPCl69cg2t27RSOtoXu+V5i7i4OFxdy6q1h4aG0r/vAK5fu0FCQgLnzpxj/tz5DBw8QKsOxdjY2nBg/0EG9R9MaGgokZGRTJ8yAx/vR/Tt14f+A/tx4vif7Ni+k9jYWK5fu8HC+Qu1bpyp+alfH169fMXYUeN49+4dXne8mDd3Pj1799DqMVatXgVbWxvGjhpHaGgoMTExrFm1Fu+H3rRo1ULpeF8srW1WW5+b1jZ5OHbwBKMHjyMsNJyoyCjmz/iNx4/86NqnM137dCb0VSjTx80i8l0k970esHT+Cjr16ICOjg77T+7mtOdfHL94OPkGMGfJTIaNHarw6D4urfnMyjas3cCAvoMIDQ0lNDSUPj1+oqZ7DapWq6JoLj1NPbCNjQ3m5ubcuXOHb7/9Nrnd399fkZOCXUq5EuAfkPwJz6WUKzo6OrTr0JY582ZjampKh7adCHweiLm5Ge06tGPM+NEZnjO9BAYFAWBtY63WPnXGFHLlykWHth0JCgzC1s6WIcOGMGBQfyVifjEjIyMOHT3A2FHjcC1dnsjISFzLuXL85DGKFS8GwIbN65kycQo9uvTExtaGvv37MnjoIIWTf57UttulK5aw//A+fhk+ivx2BTE1NaFLty5avd0CmJubc+DIfsaPmUD5MhWIiHhD8RLF2LZrK5WrVFI63hdLa5vV1udmLqNcbD3wB9PGzaRW+bpERUZRxtWZnce2UqTY+z0Vm/dtYPKoabgULE9uExPadmnD0NHvn4s2dh8+0dsqjyXmFmYZNo7PldZ8zpw2i5nT3//uVWxsLDOnz2Lu7F8BeB354R8V1BbLVi6lX+/+lCxSGl1dXRo1+Y658+coHQudpH+ODWnA8uXLWblyJYsWLcLNzY0dO3YwdepUdu/eTcmSJT94n9jYWLy8vCha0ilLfXf/vwx1jYhRRSkdQ+NknFlLdhhndhgjgIN1PryeeSodQ+McjApmi/nMDtttXGwcPvd9cXZ2xsDAIEW/xvbQAPTu3ZuEhARGjRpFaGgohQsXZuXKlR8tZoQQQgghvoRGCxodHR369+9P//6Ze5epEEIIIbSb9vyCkRBCCCHER0hBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitp9FLH3wNA91cGOimvPhUVmKoa6R0hAwh48xassM4s8MYAZzzlVM6gsaFhYVlm/nM6uPU0dVNtT/TFjSxqmiSVCqlY2hMdrgyKmSvcWYXWX0+s8s2GxYWli3GaWlpievvTZSOoXEnW23I8vMZp4pLtV8OOQkhhBBC60lBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HrZqqC5d/ceLT1akdcmP+bGllStWI0D+w8m9y9euITyLm7kMbOhXJkKLPptsYJpv1xISAi9uvemcH4nbCzscK9Wm1N/nUruj4+PZ8rEqZgYmjF10jQFk6Yv74fetG7+AwXsC2Jn5YB7tdocPnhE6VjpTqVS4e3tzaVLlzh79iw3btwgLCxM6VjpLq3tWBsVdyqJiaEZ5saWajcfbx/g/Tb8fSMP8pjZYJ8nLx3bdebVq1cKp/58T/yeUL9OQ3LpGeP/xD+53aWUa4qxm+YyJ5eesYJpPy4uIpr7Sy5wvtdOznTayvXRRwi7EwRAYrwK3z9ucOGn3Zxut4nzvXbyaON1EuP//6LK4XeCuD72KGc6beV8jx3cW3SeuIiY5P6YV5HcW3iO8712crrDZm5OPkFU4JsMH2daUnsuZqY51VhBc/XqVcqUKZPiVrJkSUaNGqWp1X5UVFQU9es0xNHJkQe+9wgOC6JZ82a0bd2O+/fus2njJiZPmML8334l8OUzFi9bxNRJ0/hjwx8ZnvVrtW7ehqDAIC5ePU9A0BNq1XandYs2BAYGERISQs2q7jx48BBzc3Olo6abxMREmjbywDBXLm7e9cQ/0I+WrVvwY+u2eD/0VjpeuvLx8SEiIgIXFxeqVauGnZ0dd+7cISoqa11pN7XtWJstXbGY15FhareixYoSHh5Og7oNcS3nim+AD9duXSUmJoYlC5cqHfmz7Nu7n1rVa1OgYP4Ufbfv3Uwx9mYtPOjUpaMCSdN2Z9bfxIZH4TarMdVXtca8tC1es/4mNiwKn7XXCL3xDNexdXH/ox1lRtTmxSlf/HbcBuCdfzi3Z5zEtnohaqz5gQozviMyIJyHKy4BkKRK5PaMk8S9jsZtZiOqr2yNadE83Jr6J6o4VWqxMlxqz8XMNKcaK2gqVqzInTt31G4nT57E1NSU5s2ba2q1HxUZGcmU6ZOZNHUiJiYmGBgY0Kdfb1QqFXfv3mPp4mV06tqRWt/UQl9fn+o1qtGpS0cWL1yS4Vm/RkREBCVLlWDOvNnY2dlhaGjIzyOGEhkZydUrVwkLDaPfgL5s2roRfQN9peOmm5cvXxLgH0CHju2wtLTEwMCA3j/1Ij4+nlu3bisdL93Ex8cTHBxMoUKFMDIyQldXFwcHB4yNjQkMDFQ6XrpJazvOilavXIOFpSWTp03CzMyMvHkd2LF7GxMmj1c62mcJDwvnz79P0K592zSX3b/vAOfOnmfW3JkZkOzzJETGYZzPjKJdKmJgkQtdfV0KNHNGFZvAG59XWJa1p+SAGhjlNUMnhw6mRfJgVsKGd0/e7y2NDY8mb4Pi5PuuBDn0cmBonRu72k6Ee70AICrwDZEBryncxhUDSyP0cuWkcBtXklRJhF57quTQ1Xzuc1HJOc3QQ04TJkzgu+++o1KlShm5WgCsra3p2r0LRkZGAISGhjJj6kzy5stLTfca3Lp5G7eKbmr3cavkxp3bXlr1ydfMzIzlK5dRomSJ5Da/x34A5MuXjxIlS9Cxc+b8NPQ1bG1tqVa9GmvXrOfly5fEx8ezcsUq8uTJQ63a7krHSzdv374lKSkJU1NTtXYTExPevMl8u6q/VFrbsTbbtWM35cpUwNbSnmqVqicf9j596jSu5coyqP9g8tsVxKlgUQb1H8y7d+8UTvx5unTrTNFiRdNcLjo6miEDhzJj9vRMubdYz1ifEn2rYZzPLLktJvgtAAZ5jLGuXACTwpYAJCYkEur5nNf3grGr5QiAlasDRTqrv6dEB7/FMM/79yB0dABISkpK7tfJoYNebn3ePArV2Lg+1+c8F5WeU72MWtHJkye5ceMGf/75Z0at8qPMjCyIi4ujglsFDh09QGJiIiqVCksLC7Xl8lhZkZiYSGhoWHIhpG3evHlD7x59aPJ9Eyq4lVc6jkZt2bGJZk2aU8C+EDo6OuTJY8WmrRuxsbFROlq6iY+PB0BPT/2pmzNnTuLi4pSIlCGyynZcpowzjk6OrFz7OwYGBixbvJw2LX/k1NmTPHv2nAvnLzJ3/hxmzZ3JXa97/Ni6LbGxsSxfuUzp6OluyaKlWFlZ8kOb1kpH+SQJUXHcX3qRPBXzYepkldz+YPlFgk4+Qs9IH6eO5bGtXviD9w+7E0TgcR9KD6kJgJG9Ccb5zfHbepOS/auT08SAoJOPiA5+S/zb2AwZ05dI7bmo9JxmyB6axMRE5s2bR69evcidO3dGrDJVEVHhBAQ9oeF3Dahbqx6PvB+lurzO/yppbePvH0Ad97pYW1uzbuMapeNoVFxcHB6Nm1GseDH8nvkSHBbEmPFjaNmsNffv3Vc6nvgKWWk73rl3B7N/nYW1tTWmpqaMHD2Csq4urF29lqSkJFzLudK1excMDQ2p4Faen4cPZcumrSQkJCgdPV3FxsYyf+4Cho0cphWvrzEv33Fj7DH0TQ0oNbCGWl+JPlWptakdpQbXxG/rLQL2301x/+CzftyZdYqiXd2wrlwAAB3dHJQZWRtdw5xcHX6Qy4P2Efc6GivXvOjoZs5/k9Sei5lhTjOkoDl+/DjBwcG0b98+I1b3SaytrRk7YQwOee05sP8genp6hP7nmyKvQkPR09MjTx6rjzxK5nXt6nXcq7pTvUZ19h7cg7Fx5vwWQXr5++Tf3PS8xexfZ2FnZ4eJiQm9f+pFocIF2bBuo9Lx0o2+/vvznv77BhcfH5/cl5Vkh+3Y0cmJwOeB2NvbYWllqd7nWJi4uDhevnypUDrNOHb0ONHR0TRu0kjpKGl68+gV10YdwaykDS6j66JrmDPFMjly6mLl6kABj1L47/ZS6/PbcQvv1VdwHlqLvA2Kq/XlsjXB5ZdvqLm2DdWWt8SxbTliXkViaK38B///Suu5mBnmNEMKmv3791OnTh0MDAwyYnUfdPDAIYo7lSQmJkatPTY2Dj09PcqVL8eVy+onOF08f5HyFcpjaGiYkVG/2l2vu3g0bsawkcP4bfECcuZM+QTMalSqxP/9V/3bAQkJKhITE5WIpBEmJibo6OikOF8mIiIiU56H8DWy2nb8xO8JgwcM4fXr12rtDx88wLGIE85lnLl185baNuzr+5hcuXJhZ2eXwWk1a9eOXdT9tk6mL1DfBYRza9pfFGzuTPGelcmh9/4tM1GVyOXB+3hx9rHa8onxiWp7V57sukPgCR/KT2mAVfm8KR4/5KI/kc8jkv+ODY/inV8YFs62GhrRl/mU52JmmFONFzTv3r3jzJkzfPvtt5peVaoqV6lEdFQUQwYOJSwsjJiYGBYvXILvI188mnswcMgANq7byKm/ThEXF8dfJ/5i4/o/GDR0oKK5P5dKpaJnt1507d6FAYP6Kx0nw1StXgVbWxvGjhpHaGgoMTExrFm1Fu+H3rRo1ULpeOlGT08Pe3t7/Pz8iIqKQqVSERAQQExMDA4ODkrHSzdZcTu2sbXhwP6DDOo/mNDQUCIjI5k+ZQY+3o/o268PP/Xrw6uXrxg7ahzv3r3D644X8+bOp2fvHlpxWOZzXLl8lbKuZZWOkaokVSL3F1/AoW5R8jcuqdaXQzcHpkXy8GTbbd76hZGkSuStbyjPjz3EpmpBAN76huK/6w5lx9TFOL/5B9cRdOoR3r9fJv5tLPFvY3mw9CJmpWwxK555zvv71OdiZphTjZ8UfP/+feLj4ylZsmTaC2uQtbU1R/88wqgRoynuWJIcOXJQrHgxtu3cSuUqlahcpRKvw1/T76cBPH/2nPwF8jN3/hxatMz4r5h/jUsXL+N54yZ3ve6l+Mp5uw5tKVCgADOnzwLeH/OcOX0Wc2f/CsDrSO39cTZzc3MOHNnP+DETKF+mAhERbyheohjbdr2f36ykSJEi+Pr64unpiUqlInfu3JQtW1br9iSmJq3teOkK7fo5BQAjIyMOHT3A2FHjcC1dnsjISFzLuXL85DGKFS8GwP7D+/hl+Cjy2xXE1NSELt26MGb8aIWTfx6XUq4E+Ack7xl1KeWKjo6O2rwFBQZhbWOtZMw0RXi/4p1fGJFPX/PssPp5eLbujhTrWZknO25ze8ZJEt7Fom9hhG3NwhRq5QLA8+PeJCaouD7qcIrHLjv2W8xL2VLip6o8XH6Ji/32oJNDB6sK+Sja1S3F8kr61OdiZphTnaR/f2dMAw4dOsTQoUO5efMmuXLlSnP52NhYvLy8KFrSKUv9Tsp/GeoaEaPSnq+Df6nsNM7sIqvPZ3baZrPDOB2s8+H6exOlY2jcyVYbsvx8xsXG4XPfF2dn5w+ewqLxQ06NGzfm4cOHn1TMCCGEEEJ8iWx1LSchhBBCZE1S0AghhBBC60lBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitJwWNEEIIIbSeFDRCCCGE0HpS0AghhBBC60lBI4QQQgitp/GrbX8pA91cGOimvPhUVpJdLmiYXcaZXWSH+cwOY4TsM86bvQ4qHUHzWmX9+dTR1U21P9MWNLGqaJJUKqVjaEx2udJtdhpndpHV5zM7bbPZYZxhYWHZYpyWlpYcfrBH6RialaBDTow/2i2HnIQQQgih9aSgEUIIIYTWk4JGCCGEEFpPChohhBBCaD0paIQQQgih9aSgEUIIIYTWk4JGCCGEEFpPChohhBBCaD0paIQQQgih9aSgEUIIIYTWk4JGCCGEEFov2xY0F85fxFjfhKmTpgHgUsoVc2NLtZtpLnNy6X38uhGZVUhICL2696ZwfidsLOxwr1abU3+dArLWOP/rid8T6tdpSC49Y/yf+CsdR2NUKhXe3t5cunSJs2fPcuPGDcLCwpSOle5S24612cb1GylbuhzmxpaULFqahQsWJfd5P/Tm+0Ye5DGzwT5PXjq268yrV68UTPv1zp05l+I1x9zYEmN9E3p17610vK+W2nxu27qdqhWrYW1ui3MJFyaMnYgqE16jMNA/iL7NB1HVtjZBAUHJ7fFx8SyZsoLmFdpQq0A9mrq0ZPHk5cTFxn3wcWYO+zXFY/g/CmBEpzE0KtWMekUb0+O7nzh3/IJGxqHRgubx48f89NNPVK1aFTc3N3744QdOnVL+BSk6OprePfpgYmKS3Hb73k1eR4ap3Zq18KBTl44KJv0yrZu3ISgwiItXzxMQ9IRatd1p3aINgYFBWWqc/7Zv735qVa9NgYL5lY6icT4+PkRERODi4kK1atWws7Pjzp07REVlrQvwpbYda6vt23YwasQY5i2YS3BYECtWLmPNqrXcuO5JeHg4Deo2xLWcK74BPly7dZWYmBiWLFyqdOyvUsO9RorXnAe+97CwsKBjpw5Kx/sqqc3n2dNn6dm1F8NGDuNZcABbd2xmy+atzJw2S+nYav4+fJYejfpil88uRd/8sYu4cOIiC7bN4dSTY8xaP42DWw6zeu76FMteOX2Nv/adVGtLTExk8I/DMTDUZ8v5DRzy2kNdj28Y1W08/o8C0n0sGitoEhMT6dGjB4aGhhw5coQLFy7w3XffMWDAAB4/fqyp1X6S8WMmULx4MVzKunx0mf37DnDu7HlmzZ2Zgcm+XkREBCVLlWDOvNnY2dlhaGjIzyOGEhkZydUrV1Msr63j/K/wsHD+/PsE7dq3VTqKRsXHxxMcHEyhQoUwMjJCV1cXBwcHjI2NCQwMVDpeuvnc7VhbzJg6kyHDBlO3Xl0MDAxwr+3OTa8blK9QjtUr12BhacnkaZMwMzMjb14HduzexoTJ45WOne4G9B1Ey9YtqFmrptJRvkpq87l0yXIafNeAlq1aYGBggHMZZwYOHsCyJctJTExUOnqyN+FvWL5/IQ1b1UvRV6mWG+OXjKFgkQLkyJGDUuVKULayCz53H6ktF/kuihlD59Dt585q7a9fvebF02AatWmImYUp+gb6tOjiQUJ8Aj5e6o+RHjRW0ISFhfH8+XOaNWuGubk5+vr6tGvXjvj4eB48eKCp1abp/LkLbP5jCwuXLvzoMtHR0QwZOJQZs6djbm6eceHSgZmZGctXLqNEyRLJbX6P/QDIly+f2rLaPM7/6tKtM0WLFVU6hsa9ffuWpKQkTE1N1dpNTEx48+aNQqnS3+dsx9oiKCiIB/cfkDt3buq4f4uNhR1urhXZumUbAKdPnca1XFkG9R9MfruCOBUsyqD+g3n37p3CydPXoQOHuXThElOmT1Y6yldJaz6vXL5CxYpuavdxq+hGaGgoj3zS/838S33fvjEFnD68Z7t2Y3eKl3n/upoQn8DFvy7jefEW37Wur7bcoglLKVW+JLUbqReoljaWlK1chv2bDhH+6jUJ8QnsWb8fcyszylcvl+5j0Uv3R/yfPHnyUKFCBXbu3EmZMmUwMTFhy5YtWFhYULlyZU2tNlVRUVH07tGHmXOm4+Bg/9HllixaipWVJT+0aZ2B6TTjzZs39O7RhybfN6GCW3m1vqw0zuwiPj4eAD099aduzpw5iYv78HHtrCC17VhbPHv6HIDVK1ezZv1qCjsWZt2a9XTt2I18+fLx7NlzLpy/yNz5c5g1dyZ3ve7xY+u2xMbGsnzlMoXTp4/ExETGjx3PzyOGpijKtU1a8/nq5SssLC3U7pMnjxUAISEvKVa8WIZn/lIzhs7hwObD5DbLTf/xfajXvG5y3+W/r3LmyDk2nVlHTFR0ivtOXz2Zoe1G0qh0M3R0dDC3NGPayolYWlukWPZrafQcmkWLFvH8+XOqVq1KmTJlWLFiBb/99htWVlaaXO1HjR87gaJFi9Cx88fPF4mNjWX+3AUMGzkMHR2dDEyX/vz9A6jjXhdra2vWbVyj1peVximyttS2Y22SlJQEQN/+fXEu44yxsTH9BvSlvFt5/tjwB0lJSbiWc6Vr9y4YGhpSwa08Pw8fypZNW0lISFA4ffrYu2cfgc+D6NNX+08GTms+U6Ntr7mj5g3ndMBxJi8fx+8zV7Np6VYAIt9GMmPoHIbOGIRFHvMU94uPi2dI2xEULFKAg3d2ceLRIboP78LwjqPxe/gk3XNqrKCJi4ujR48eFC5cmHPnznHt2jX69+9Pnz59ePQo43e3nT93gc0bt7B4+eJUlzt29DjR0dE0btIog5JpxrWr13Gv6k71GtXZe3APxsbq32LKKuPMbvT19QFSvMHFx8cn92UlaW3H2sT+f3uFLa0s1dodHR15/uw59vZ2H+grTFxcHC9fvsywnJq0ZdNWGjdthKGhodJRvlpa82lja0NYqPq3D1+9CgXAzs42Y0Kmo5z6OanyTSXa92vL+t82AbBwwlJKlSvBtx7ffPA+187dwPuOD4Mm98PKxgrj3Ea07NoM+wL2HNx6JN0zaqyguXTpEvfu3WP06NFYW1uTO3du2rdvT758+di1a5emVvtR69euJzIyksrlq5DPtgD5bAtw8cJF5s2dT9WK1ZKX27VjF3W/raPVL5x3ve7i0bgZw0YO47fFC8iZM2eKZbLCOLMjExMTdHR0UpwvExERofXnQf3Xp2zH2sTBwR4rKyuuX72u1v7Y15dChQvhXMaZWzdvqX2t19f3Mbly5cLOLuU3ULTN27dvOX70OE09miodJV2kNZ9VqlbhyuUran0Xzl/Azt4ORyfHjIz6RRISEvixRieO7Tyh1h4fF4eeri4A+zcd4srp6zQs+T0NS35P53q9AOhcrxd/LN5Cour9yc//PQlapVKRlJiU7pk1+i0nIMV37lUqVfKuuow0a+5M7np7cen6xeRb+Qrl6dGrO3sO7E5e7srlq5R1LZvh+dKLSqWiZ7dedO3ehQGD+n90OW0fZ3alp6eHvb09fn5+REVFoVKpCAgIICYmBgcHB6XjpZtP3Y61ia6uLgMHD2D50hWc/PMksbGxrFj2Ozc9b9GjV3d+6teHVy9fMXbUON69e4fXHS/mzZ1Pz949tO4QxYfc8rxFXFwcrlnkdSet+ew/sB8njv/Jju07iY2N5fq1Gyycv5CBgwdoxXzq6elRunxJVs5Zi7eXDyqVige3HrJr7V6++b42APs8t7P5zFo2/LWKDX+tYt6m99+WnbdpJs07f49LJWcsrS1YMmUFEWERxMbEsm/jQQIePaVO01rpnzndH/F/ypcvT548eZg7dy6jRo3CyMiIffv24efnx/Tp0zW12o+ysLDAwkL9JCQDAwNMTU3VPv0EBQZhbWOd0fHSzaWLl/G8cZO7XvdYvHCJWl+7Dm1ZuuJ9m7aP879cSrkS4B+QXEi7lHJFR0dHbcxZRZEiRfD19cXT0xOVSkXu3LkpW7ZsltiN/49P3Y61zfBfhpGQkECvHn14GfKSYsWLsu/gnuQPF/sP7+OX4aPIb1cQU1MTunTrwpjxoxVOnT4Cg97/flBWet1Jaz43bF7PlIlT6NGlJza2NvTt35fBQwcpnFpdm2odefHsBYn/22PSpnondHSgYav6DJ/5f+3de1BV5b/H8fdWQPBGoAc0CB01FbkIAiFq/RI9HmHQMW9ckkY0VBQ1CZOg0SQ65Sg2QoiaiRe8ktpoKl6oZnLSGhS5eBkNhQESI7G4CGzYrPMHh33kiP0yNm43fF8z/MF6NovP2mux9pdnrfU8K9gRv4t3g6Ko+KOCvtZ9+a8Z/6l9PNvqRasW69I0NHVg9LGypEevpt7/TQc3kPzfXxD46lyqK6sYMMSOT1M+wtHdQefbolLasbvkxo0bbNy4kby8PCorKxk0aBDLli1jwoQJT/yZuro68vLyeNl+MCbdOt49Ac1Mu3anVtOxBkJrTWfazs6io+/PznTMynZ2HC/+hy0nbxzVd4z21aDC+G4PHB0d6dat22PN7dZDAzB8+HC2bdvWnr9CCCGEEKLzzuUkhBBCiI5DChohhBBCGDwpaIQQQghh8KSgEUIIIYTBk4JGCCGEEAZPChohhBBCGDwpaIQQQghh8KSgEUIIIYTBk4JGCCGEEAZPChohhBBCGLx2nfrgn2ieWqpeXa/nJO1L1bUrao1a3zHaXWfazs6io+/PznTMynZ2HNbW1tDw/M/i3Sb/u31PmoKyXSen/CcqKyu5efOmvmMIIYQQ4jk0dOhQevXq9djy566gaWxspLq6GmNjY1SqDl5tCiGEEOJvURSF+vp6evToQZcuj98x89wVNEIIIYQQT0tuChZCCCGEwZOCRgghhBAGTwoaIYQQQhg8KWiEEEIIYfCkoBFCCCGEwZOCRgghhBAG77kbKbi9FRQUcODAAa5cuUJ5eTkqlYq+ffvi7u5OQEAA/fv313dE8RTKysrIzs6mf//+ODg4PNa+evVqYmNj9ZBMt/Lz87l16xZeXl6Ym5tTWFjIgQMH6Nq1K5MmTcLZ2VnfEdvN9OnT+eKLL+jTp4++o7RZYWEhAwYM0H6fk5PDoUOHuHfvHgMGDGDOnDkMHDhQfwF16Nq1a4wYMQKAS5cucfToUUpLS7G1tSUwMJBhw4bpOWHbRUdHM336dNzd3fUdpd0VFBRw/fp1PD09sbS0JD8/nyNHjtDY2IiPj89zcQ7qVOPQ/PjjjyxatIjBgwfj5OSEhYUFAA8ePCArK4uSkhJ27NiBi4uLfoM+I4b+YX/hwgUWL15MTU0NKpWKiRMnsn79ekxNTbWvGTlyJNnZ2XpM2XYZGRksW7YMjUbDiy++yJdffklgYCDW1tao1WqKiopISkriX//6l76j/mOff/75E9u2b99OYGAgPXr0IDw8/Bmm0r1Hj8eMjAzCw8Oxt7fHzs6O/Px8CgsLSU1NfS4+HNpi586dpKWlceLECb755hsiIyNxdHTEzs6OwsJCbty4wfbt2/Hy8tJ31DYZMWIE5ubm+Pr6EhERQY8ePfQdqV18//33hIeH09DQgKWlJZs2bWLhwoXY2tqi0WgoKCggKSmJ119/Xb9BlU7E399fOXDgwBPbt2/frvj7+z/DRPrl7Oys7whtMnv2bCU+Pl6pqKhQ8vLylKlTpyrz5s1TGhoatK9xcnLSY0LdeOONN5Tk5GSlsrJS+fTTT5VZs2YpiYmJ2vadO3cqs2fP1mPCtrO3t1c8PDyU4OBgZc6cOS2+HBwcFH9/fyU4OFjfMdvs0eOxeb8+KjExUQkKCnrWsXRu4sSJys8//6woiqL4+fkpu3fvbtGempqqTJ06VR/RdMrZ2VkpKytTwsLCFC8vLyUlJUWpq6vTdyyd8/f3125bSkqKMmbMGGXv3r3a9oMHDyrTpk3TY8ImnaqgcXd3/8uDraamRhk5cuSzC9SOSkpK/vKruLjY4D/s3dzcWuzPyspKxc/PT4mNjdUuM/SiTVGatkGtViuKoijl5eXK8OHDlfLycm17XV2d4u7urq94OpGZmalMmjRJCQsLU3777bcWbZ6enkppaamekunWo8ejm5ubUlVV1aK9urpacXFxedaxdM7JyUl7zI4aNUqpqalp0V5XV2fw5x9Fabk/MzIyFD8/P+WVV15RoqKilLNnzyolJSVKbW2tHhPqhoeHh3Z/1tfXK8OHD1eqq6u17XV1dc/FubZT3UPTu3dvSktLsbOza7W9tLSU7t27P+NU7cPb2/sv58JSFMXg58oyMzOjqqoKS0tLAHr27MmWLVuYPXs2AwcOJDg4+ImzshoSU1NT6uvrMTY2xsLCAiMjI+3lUoCamhoaGxv1mLDt3NzcOHbsGPHx8UyZMoWoqCimTZum71jtqvmS4aOXKdRqdYtLpobKxsaGzMxMvLy8GDZsGMXFxQwZMkTbfunSJe3fbUfh7e3N+PHjOX36NMePH2fFihXU19ejUqm4fv26vuO1iZGREbW1tRgbG1NRUYGiKNTW1mo/Lx8+fEjPnj31nLKT3RT86quvsnz5cpYuXYqTkxPm5uYA/PHHH2RnZ5OQkICfn5+eU+qGh4cHtra2TJ06tdV2RVFYuHDhM06lW+PGjeO9994jOjqaQYMGAU0n0uTkZEJDQykrK9NzQt1wdXUlMTGRd955h27dupGbm6ttU6vVbNiwATc3Nz0m1I1u3boRHR3NpEmTiImJIT09ndjYWIMvvB+l0Wj4+uuvgaZjddu2baxatQqAqqoq4uLicHV11WNC3QgNDSUiIoLQ0FD8/PyIjIzkrbfewtzcnNzcXPbu3UtISIi+Y+qcSqVi8uTJTJ48merqaq5du9YhzkMuLi58+OGHTJs2jbS0NEaMGMEnn3xCdHQ0Go2GDRs24OTkpO+Yneum4NraWtauXcvx48fRaDQt2oyNjZkxYwbR0dEYGxvrKaHuFBUVERAQwL59+1o8VfEoQ79htry8nMWLFzNkyBDi4uJatN28eZOoqCiuX79u8P8d3bp1i+DgYCIjI5k5c2aLNm9vb+rq6ti9ezeDBw/WU0Ldq62tJT4+nmPHjlFTU8PZs2extrbWd6w28/b2bvH98OHD2bx5MwCxsbFkZGSQkpKiLdAN2YkTJ9i8eTP5+fktlr/wwguEhIQY/D9UAM7OzuTk5Og7Rru7c+cO8+fP59dff8Xd3Z2EhATmz5/PjRs3AOjTpw87duxg6NChes3ZqQqaZhUVFVy9epXy8nKgaWc4Ojo+F11munTu3Dn+/PNPZsyY0Wr75MmTSU9Pf8apdK+qqqrVfdfY2EhWVlaH6L2ora3l4cOHj3XTf/vtt7i4uHS47vtmmZmZHD58mOjoaHr16qXvOO2quLiYvn37dohLTo8qKyvj7t27AFhZWWFlZUWXLh1jCLTMzMxO8cg2NPXql5eXa4dPqK+v5+LFizQ2NjJq1Kjn4u+zUxY0QgghhOhYOkaZLIQQQohOTQoaIYQQQhg8KWiEEBQXF+Po6EhwcDDBwcEEBATw7rvvUlFR8Y/XmZaWRlRUFAArVqzg3r17T3zt5cuXKSoq+tvrbmhoeOLQ+Tk5OcydO5fp06cza9YswsLCtOuOiooiLS3tKbZCCGEopKARQgBgaWnJnj172LNnDwcOHMDKyork5GSdrPuzzz77y6eUjhw58lQFzZOUlZURHh7O8uXLOXLkCGlpafj6+vL222/T0NDQ5vULIZ5fnWocGiHE3+fh4cHBgweBpseNfXx8KCoqIiEhgZMnT5KamoqiKFhaWhIXF4eFhQV79+5l//799OvXDysrK+26vL29SUlJ4aWXXiIuLo68vDwAQkJCMDIyIj09nZycHN5//30GDBjA2rVrqamp4eHDh0RERDBmzBhu377NypUrMTMzw9PTs9XMqampTJ06tcVYLlOmTOG1117DyKjl6W7Tpk1cuHABgH79+rF+/XpUKhUffPABd+7cQaVSYW9vz5o1a7h48SLx8fGYmpqiVquJiYkx+PmWhOhopKARQjxGo9Fw9uzZFo+8Dxw4kJUrV3L37l22bNnCV199hYmJCbt27WLr1q0sWbKEhIQE0tPTsbCwICwsTDt4ZbNjx47x+++/c+jQISoqKoiMjCQ5ORl7e3vCwsLw8vJiwYIFzJs3j9GjR1NWVoa/vz9nzpwhKSmJGTNmEBQUxJkzZ1rN/csvv7Q6mOT/z9HQ0ICZmRn79u2jS5cuzJ8/n/Pnz2NtbU12djanTp0C4NChQ1RWVrJr1y5CQkLw9fXl9u3b3Llzp61vsRBCx6SgEUIATQMVBgcHA01j+Li7uzN37lxte3OvR1ZWFmVlZcyfPx9oGq3Y1taWwsJCbGxstNMyeHp6agfeapaTk6PtXenduzfbtm17LMdPP/1EdXU1SUlJQNOw6/fv3+fmzZssWLAAgNGjR7e6DV27dn1s0MzWGBkZ0aVLF4KCgjAyMuL27ds8ePCAMWPGYGFhQWhoKOPHj8fHx4devXoxZcoUNm7cSE5ODhMmTGDChAn/9ncIIZ4tKWiEEMD/3UPzJM0jaJuYmODs7MzWrVtbtOfm5raYpqC1+aVUKtW/nXfKxMSExMTExwYLVBRFOyDbk4qWoUOHcvnyZXx9fVssz87ObnGJ6NKlSxw+fJjDhw/TvXt3li1bBjRNv7Bv3z6uXr3Kd999x8yZM9m/fz++vr6MGzeO8+fPk5SUhLOzMxEREX+5HUKIZ0tuChZCPBUnJydycnK0c9ScOnWKc+fOYWdnR3FxsXbyuub7Ux7l6urKDz/8ADSN8Dxr1izUajUqlYr6+nqgaaLK5ks+5eXlfPzxxwAMHjyYK1euALS6boCgoCDS09O5ePGidtnJkyeJiYnRrh/g/v372NjY0L17d0pKSrhy5QpqtZrc3FyOHj2Kg4MD4eHhODg4UFBQQEJCAhqNBl9fX2JiYsjKymrjuyiE0DXpoRFCPBVra2tiYmJYuHAhZmZmmJqasm7dOszNzVm0aBFvvvkmNjY22NjYUFtb2+JnfXx8uHz5MgEBAWg0GkJCQjAxMWHs2LGsWbOG6OhoYmJiWL16NSdOnECtVhMWFgbAkiVLWLVqFenp6bi6uj52ky809TKlpqby0UcfsW7dOkxNTbGxsWHnzp2YmJhoXzd27Fh27NhBYGAgL7/8MkuXLiUpKYlNmzZx+vRpDh48iImJCXZ2dowaNYq7d+8yb948evfuTWNjI0uXLm3fN1kI8dRk6gMhhBBCGDy55CSEEEIIgycFjRBCCCEMnhQ0QgghhDB4UtAIIYQQwuBJQSOEEEIIgycFjRBCCCEMnhQ0QgghhDB4UtAIIYQQwuD9D/nzBzAqNj3FAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_model(tuned_model, 'confusion_matrix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_model(tuned_model, 'class_report')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Model interpretation with SHAP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x684 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "interpret_model(tuned_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "</style><table id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Model</th>        <th class=\"col_heading level0 col1\" >Accuracy</th>        <th class=\"col_heading level0 col2\" >AUC</th>        <th class=\"col_heading level0 col3\" >Recall</th>        <th class=\"col_heading level0 col4\" >Prec.</th>        <th class=\"col_heading level0 col5\" >F1</th>        <th class=\"col_heading level0 col6\" >Kappa</th>        <th class=\"col_heading level0 col7\" >MCC</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3row0_col0\" class=\"data row0 col0\" >Random Forest Classifier</td>\n",
       "                        <td id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3row0_col1\" class=\"data row0 col1\" >0.6586</td>\n",
       "                        <td id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3row0_col2\" class=\"data row0 col2\" >0.9241</td>\n",
       "                        <td id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3row0_col3\" class=\"data row0 col3\" >0.6272</td>\n",
       "                        <td id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3row0_col4\" class=\"data row0 col4\" >0.6793</td>\n",
       "                        <td id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3row0_col5\" class=\"data row0 col5\" >0.6614</td>\n",
       "                        <td id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3row0_col6\" class=\"data row0 col6\" >0.5947</td>\n",
       "                        <td id=\"T_0cf6837c_2509_11eb_8373_44032c87eac3row0_col7\" class=\"data row0 col7\" >0.5968</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x7ff9d13117f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictions = predict_model(tuned_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Model saving and loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LGBMClassifier(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
       "               importance_type='split', learning_rate=0.1, max_depth=-1,\n",
       "               min_child_samples=20, min_child_weight=0.001, min_split_gain=0.0,\n",
       "               n_estimators=100, n_jobs=-1, num_leaves=31, objective=None,\n",
       "               random_state=42, reg_alpha=0.0, reg_lambda=0.0, silent=True,\n",
       "               subsample=1.0, subsample_for_bin=200000, subsample_freq=0)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_model = finalize_model(tuned_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transformation Pipeline and Model Succesfully Saved\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(Pipeline(memory=None,\n",
       "          steps=[('dtypes',\n",
       "                  DataTypes_Auto_infer(categorical_features=[],\n",
       "                                       display_types=True, features_todrop=[],\n",
       "                                       id_columns=[],\n",
       "                                       ml_usecase='classification',\n",
       "                                       numerical_features=[], target='target',\n",
       "                                       time_features=[])),\n",
       "                 ('imputer',\n",
       "                  Simple_Imputer(categorical_strategy='not_available',\n",
       "                                 fill_value_categorical=None,\n",
       "                                 fill_value_numerical=None,\n",
       "                                 numeric_strat...\n",
       "                  LGBMClassifier(boosting_type='gbdt', class_weight=None,\n",
       "                                 colsample_bytree=1.0, importance_type='split',\n",
       "                                 learning_rate=0.1, max_depth=-1,\n",
       "                                 min_child_samples=20, min_child_weight=0.001,\n",
       "                                 min_split_gain=0.0, n_estimators=100, n_jobs=-1,\n",
       "                                 num_leaves=31, objective=None, random_state=42,\n",
       "                                 reg_alpha=0.0, reg_lambda=0.0, silent=True,\n",
       "                                 subsample=1.0, subsample_for_bin=200000,\n",
       "                                 subsample_freq=0)]],\n",
       "          verbose=False),\n",
       " 'otto_Group_LGB.pkl')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "save_model(final_model, 'otto_Group_LGB')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transformation Pipeline and Model Successfully Loaded\n"
     ]
    }
   ],
   "source": [
    "model = load_model('otto_Group_LGB')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "PyCaret",
   "language": "python",
   "name": "py_caret"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}