Kaggle Elliptic Data EDA ver2 .ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import libraries\n",
    "import networkx as nx\n",
    "import pandas as pd \n",
    "import numpy as np \n",
    "import matplotlib.pyplot as plt \n",
    "import seaborn as sns\n",
    "import plotly.graph_objs as go \n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import precision_recall_fscore_support\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.neural_network import MLPClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Data loading \n",
    "df_classes = pd.read_csv('elliptic_txs_classes.csv')\n",
    "df_features = pd.read_csv('elliptic_txs_features.csv', header=None)\n",
    "df_edgelist = pd.read_csv('elliptic_txs_edgelist.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>txId</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>230425980</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5530458</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>232022460</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>232438397</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>230460314</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        txId    class\n",
       "0  230425980  unknown\n",
       "1    5530458  unknown\n",
       "2  232022460  unknown\n",
       "3  232438397        2\n",
       "4  230460314  unknown"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# exploring the basics of our classes \n",
    "df_classes.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<bound method DataFrame.info of              txId    class\n",
       "0       230425980  unknown\n",
       "1         5530458  unknown\n",
       "2       232022460  unknown\n",
       "3       232438397        2\n",
       "4       230460314  unknown\n",
       "...           ...      ...\n",
       "203764  173077460  unknown\n",
       "203765  158577750  unknown\n",
       "203766  158375402        1\n",
       "203767  158654197  unknown\n",
       "203768  157597225  unknown\n",
       "\n",
       "[203769 rows x 2 columns]>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_classes.info"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>ID</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>230425980</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5530458</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>232022460</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>232438397</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>230460314</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          ID    Class\n",
       "0  230425980  unknown\n",
       "1    5530458  unknown\n",
       "2  232022460  unknown\n",
       "3  232438397        2\n",
       "4  230460314  unknown"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# renaming my columns \n",
    "df_classes.columns = ['ID'] + ['Class']\n",
    "# check \n",
    "df_classes.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_classes['Class'].value_counts()\n",
    "# 1 means the transaction is illicit / illegal \n",
    "# 2 means the transaction is licit / legal \n",
    "\n",
    "# we will change the 'unknown class' to 3 so that is well suited with the form of the other classes \n",
    "df_classes.loc[df_classes['Class'] == 'unknown', 'Class'] = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "           ID\n",
      "Class        \n",
      "3      157205\n",
      "1        4545\n",
      "2       42019\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<BarContainer object of 3 artists>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualize the total number of each class  \n",
    "group_class = df_classes.groupby('Class').count()\n",
    "print(group_class)\n",
    "plt.bar(['Unknown','Ilicit', 'Licit'], group_class['ID'].values, color=['b', 'r', 'g'] )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>157</th>\n",
       "      <th>158</th>\n",
       "      <th>159</th>\n",
       "      <th>160</th>\n",
       "      <th>161</th>\n",
       "      <th>162</th>\n",
       "      <th>163</th>\n",
       "      <th>164</th>\n",
       "      <th>165</th>\n",
       "      <th>166</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>230425980</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.171469</td>\n",
       "      <td>-0.184668</td>\n",
       "      <td>-1.201369</td>\n",
       "      <td>-0.121970</td>\n",
       "      <td>-0.043875</td>\n",
       "      <td>-0.113002</td>\n",
       "      <td>-0.061584</td>\n",
       "      <td>-0.162097</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.562153</td>\n",
       "      <td>-0.600999</td>\n",
       "      <td>1.461330</td>\n",
       "      <td>1.461369</td>\n",
       "      <td>0.018279</td>\n",
       "      <td>-0.087490</td>\n",
       "      <td>-0.131155</td>\n",
       "      <td>-0.097524</td>\n",
       "      <td>-0.120613</td>\n",
       "      <td>-0.119792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5530458</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.171484</td>\n",
       "      <td>-0.184668</td>\n",
       "      <td>-1.201369</td>\n",
       "      <td>-0.121970</td>\n",
       "      <td>-0.043875</td>\n",
       "      <td>-0.113002</td>\n",
       "      <td>-0.061584</td>\n",
       "      <td>-0.162112</td>\n",
       "      <td>...</td>\n",
       "      <td>0.947382</td>\n",
       "      <td>0.673103</td>\n",
       "      <td>-0.979074</td>\n",
       "      <td>-0.978556</td>\n",
       "      <td>0.018279</td>\n",
       "      <td>-0.087490</td>\n",
       "      <td>-0.131155</td>\n",
       "      <td>-0.097524</td>\n",
       "      <td>-0.120613</td>\n",
       "      <td>-0.119792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>232022460</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.172107</td>\n",
       "      <td>-0.184668</td>\n",
       "      <td>-1.201369</td>\n",
       "      <td>-0.121970</td>\n",
       "      <td>-0.043875</td>\n",
       "      <td>-0.113002</td>\n",
       "      <td>-0.061584</td>\n",
       "      <td>-0.162749</td>\n",
       "      <td>...</td>\n",
       "      <td>0.670883</td>\n",
       "      <td>0.439728</td>\n",
       "      <td>-0.979074</td>\n",
       "      <td>-0.978556</td>\n",
       "      <td>-0.098889</td>\n",
       "      <td>-0.106715</td>\n",
       "      <td>-0.131155</td>\n",
       "      <td>-0.183671</td>\n",
       "      <td>-0.120613</td>\n",
       "      <td>-0.119792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>232438397</td>\n",
       "      <td>1</td>\n",
       "      <td>0.163054</td>\n",
       "      <td>1.963790</td>\n",
       "      <td>-0.646376</td>\n",
       "      <td>12.409294</td>\n",
       "      <td>-0.063725</td>\n",
       "      <td>9.782742</td>\n",
       "      <td>12.414558</td>\n",
       "      <td>-0.163645</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.577099</td>\n",
       "      <td>-0.613614</td>\n",
       "      <td>0.241128</td>\n",
       "      <td>0.241406</td>\n",
       "      <td>1.072793</td>\n",
       "      <td>0.085530</td>\n",
       "      <td>-0.131155</td>\n",
       "      <td>0.677799</td>\n",
       "      <td>-0.120613</td>\n",
       "      <td>-0.119792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>230460314</td>\n",
       "      <td>1</td>\n",
       "      <td>1.011523</td>\n",
       "      <td>-0.081127</td>\n",
       "      <td>-1.201369</td>\n",
       "      <td>1.153668</td>\n",
       "      <td>0.333276</td>\n",
       "      <td>1.312656</td>\n",
       "      <td>-0.061584</td>\n",
       "      <td>-0.163523</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.511871</td>\n",
       "      <td>-0.400422</td>\n",
       "      <td>0.517257</td>\n",
       "      <td>0.579382</td>\n",
       "      <td>0.018279</td>\n",
       "      <td>0.277775</td>\n",
       "      <td>0.326394</td>\n",
       "      <td>1.293750</td>\n",
       "      <td>0.178136</td>\n",
       "      <td>0.179117</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 167 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         0    1         2         3         4          5         6    \\\n",
       "0  230425980    1 -0.171469 -0.184668 -1.201369  -0.121970 -0.043875   \n",
       "1    5530458    1 -0.171484 -0.184668 -1.201369  -0.121970 -0.043875   \n",
       "2  232022460    1 -0.172107 -0.184668 -1.201369  -0.121970 -0.043875   \n",
       "3  232438397    1  0.163054  1.963790 -0.646376  12.409294 -0.063725   \n",
       "4  230460314    1  1.011523 -0.081127 -1.201369   1.153668  0.333276   \n",
       "\n",
       "        7          8         9    ...       157       158       159       160  \\\n",
       "0 -0.113002  -0.061584 -0.162097  ... -0.562153 -0.600999  1.461330  1.461369   \n",
       "1 -0.113002  -0.061584 -0.162112  ...  0.947382  0.673103 -0.979074 -0.978556   \n",
       "2 -0.113002  -0.061584 -0.162749  ...  0.670883  0.439728 -0.979074 -0.978556   \n",
       "3  9.782742  12.414558 -0.163645  ... -0.577099 -0.613614  0.241128  0.241406   \n",
       "4  1.312656  -0.061584 -0.163523  ... -0.511871 -0.400422  0.517257  0.579382   \n",
       "\n",
       "        161       162       163       164       165       166  \n",
       "0  0.018279 -0.087490 -0.131155 -0.097524 -0.120613 -0.119792  \n",
       "1  0.018279 -0.087490 -0.131155 -0.097524 -0.120613 -0.119792  \n",
       "2 -0.098889 -0.106715 -0.131155 -0.183671 -0.120613 -0.119792  \n",
       "3  1.072793  0.085530 -0.131155  0.677799 -0.120613 -0.119792  \n",
       "4  0.018279  0.277775  0.326394  1.293750  0.178136  0.179117  \n",
       "\n",
       "[5 rows x 167 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#  exploring the basics of our features \n",
    "df_features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 203769 entries, 0 to 203768\n",
      "Columns: 167 entries, 0 to 166\n",
      "dtypes: float64(165), int64(2)\n",
      "memory usage: 259.6 MB\n"
     ]
    }
   ],
   "source": [
    "df_features.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>157</th>\n",
       "      <th>158</th>\n",
       "      <th>159</th>\n",
       "      <th>160</th>\n",
       "      <th>161</th>\n",
       "      <th>162</th>\n",
       "      <th>163</th>\n",
       "      <th>164</th>\n",
       "      <th>165</th>\n",
       "      <th>166</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>203769.000000</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>...</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "      <td>2.037690e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.711310e+08</td>\n",
       "      <td>23.843961</td>\n",
       "      <td>5.132771e-17</td>\n",
       "      <td>2.178534e-14</td>\n",
       "      <td>1.609496e-13</td>\n",
       "      <td>-1.966206e-15</td>\n",
       "      <td>-4.104424e-15</td>\n",
       "      <td>6.953564e-15</td>\n",
       "      <td>4.018656e-14</td>\n",
       "      <td>-1.027389e-15</td>\n",
       "      <td>...</td>\n",
       "      <td>-8.620828e-15</td>\n",
       "      <td>1.915670e-15</td>\n",
       "      <td>5.923961e-15</td>\n",
       "      <td>2.052502e-14</td>\n",
       "      <td>6.975481e-15</td>\n",
       "      <td>2.943190e-14</td>\n",
       "      <td>-1.748264e-14</td>\n",
       "      <td>8.625937e-14</td>\n",
       "      <td>-6.533539e-14</td>\n",
       "      <td>-1.622942e-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.104655e+08</td>\n",
       "      <td>15.172170</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "      <td>1.000002e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.076000e+03</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.729826e-01</td>\n",
       "      <td>-2.105526e-01</td>\n",
       "      <td>-1.756361e+00</td>\n",
       "      <td>-1.219696e-01</td>\n",
       "      <td>-6.372457e-02</td>\n",
       "      <td>-1.130020e-01</td>\n",
       "      <td>-6.158379e-02</td>\n",
       "      <td>-1.636459e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.770994e-01</td>\n",
       "      <td>-6.262286e-01</td>\n",
       "      <td>-9.790738e-01</td>\n",
       "      <td>-9.785560e-01</td>\n",
       "      <td>-2.160569e-01</td>\n",
       "      <td>-1.259391e-01</td>\n",
       "      <td>-1.311553e-01</td>\n",
       "      <td>-2.698175e-01</td>\n",
       "      <td>-1.760926e+00</td>\n",
       "      <td>-1.760984e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>8.433452e+07</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>-1.725317e-01</td>\n",
       "      <td>-1.803266e-01</td>\n",
       "      <td>-1.201369e+00</td>\n",
       "      <td>-1.219696e-01</td>\n",
       "      <td>-4.387455e-02</td>\n",
       "      <td>-1.130020e-01</td>\n",
       "      <td>-6.158379e-02</td>\n",
       "      <td>-1.635168e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.696264e-01</td>\n",
       "      <td>-5.946915e-01</td>\n",
       "      <td>-9.790738e-01</td>\n",
       "      <td>-9.785560e-01</td>\n",
       "      <td>-9.888874e-02</td>\n",
       "      <td>-8.749016e-02</td>\n",
       "      <td>-1.311553e-01</td>\n",
       "      <td>-1.405971e-01</td>\n",
       "      <td>-1.206134e-01</td>\n",
       "      <td>-1.197925e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.624375e+08</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>-1.692045e-01</td>\n",
       "      <td>-1.328975e-01</td>\n",
       "      <td>4.636092e-01</td>\n",
       "      <td>-1.219696e-01</td>\n",
       "      <td>-4.387455e-02</td>\n",
       "      <td>-1.130020e-01</td>\n",
       "      <td>-6.158379e-02</td>\n",
       "      <td>-1.620440e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>-4.799511e-01</td>\n",
       "      <td>-4.559278e-01</td>\n",
       "      <td>2.411283e-01</td>\n",
       "      <td>2.414064e-01</td>\n",
       "      <td>1.827940e-02</td>\n",
       "      <td>-8.749016e-02</td>\n",
       "      <td>-1.311553e-01</td>\n",
       "      <td>-9.752359e-02</td>\n",
       "      <td>-1.206134e-01</td>\n",
       "      <td>-1.197925e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.454798e+08</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>-1.318553e-01</td>\n",
       "      <td>-5.524241e-02</td>\n",
       "      <td>1.018602e+00</td>\n",
       "      <td>-1.219696e-01</td>\n",
       "      <td>-4.387455e-02</td>\n",
       "      <td>-1.130020e-01</td>\n",
       "      <td>-6.158379e-02</td>\n",
       "      <td>-1.355932e-01</td>\n",
       "      <td>...</td>\n",
       "      <td>1.552495e-01</td>\n",
       "      <td>1.212026e-01</td>\n",
       "      <td>1.305594e+00</td>\n",
       "      <td>1.398764e+00</td>\n",
       "      <td>1.827940e-02</td>\n",
       "      <td>-8.749016e-02</td>\n",
       "      <td>-8.467423e-02</td>\n",
       "      <td>-9.752359e-02</td>\n",
       "      <td>1.520067e-01</td>\n",
       "      <td>1.199710e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>4.032446e+08</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>7.168197e+01</td>\n",
       "      <td>7.359505e+01</td>\n",
       "      <td>2.683580e+00</td>\n",
       "      <td>4.902760e+01</td>\n",
       "      <td>2.600907e+02</td>\n",
       "      <td>5.456518e+01</td>\n",
       "      <td>1.134409e+02</td>\n",
       "      <td>7.335457e+01</td>\n",
       "      <td>...</td>\n",
       "      <td>7.862953e+00</td>\n",
       "      <td>7.914041e+00</td>\n",
       "      <td>1.461330e+00</td>\n",
       "      <td>1.461369e+00</td>\n",
       "      <td>1.170692e+02</td>\n",
       "      <td>2.518490e+02</td>\n",
       "      <td>2.387835e+02</td>\n",
       "      <td>1.057340e+02</td>\n",
       "      <td>1.519700e+00</td>\n",
       "      <td>1.521399e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 167 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                0              1             2             3             4    \\\n",
       "count  2.037690e+05  203769.000000  2.037690e+05  2.037690e+05  2.037690e+05   \n",
       "mean   1.711310e+08      23.843961  5.132771e-17  2.178534e-14  1.609496e-13   \n",
       "std    1.104655e+08      15.172170  1.000002e+00  1.000002e+00  1.000002e+00   \n",
       "min    1.076000e+03       1.000000 -1.729826e-01 -2.105526e-01 -1.756361e+00   \n",
       "25%    8.433452e+07       9.000000 -1.725317e-01 -1.803266e-01 -1.201369e+00   \n",
       "50%    1.624375e+08      23.000000 -1.692045e-01 -1.328975e-01  4.636092e-01   \n",
       "75%    2.454798e+08      38.000000 -1.318553e-01 -5.524241e-02  1.018602e+00   \n",
       "max    4.032446e+08      49.000000  7.168197e+01  7.359505e+01  2.683580e+00   \n",
       "\n",
       "                5             6             7             8             9    \\\n",
       "count  2.037690e+05  2.037690e+05  2.037690e+05  2.037690e+05  2.037690e+05   \n",
       "mean  -1.966206e-15 -4.104424e-15  6.953564e-15  4.018656e-14 -1.027389e-15   \n",
       "std    1.000002e+00  1.000002e+00  1.000002e+00  1.000002e+00  1.000002e+00   \n",
       "min   -1.219696e-01 -6.372457e-02 -1.130020e-01 -6.158379e-02 -1.636459e-01   \n",
       "25%   -1.219696e-01 -4.387455e-02 -1.130020e-01 -6.158379e-02 -1.635168e-01   \n",
       "50%   -1.219696e-01 -4.387455e-02 -1.130020e-01 -6.158379e-02 -1.620440e-01   \n",
       "75%   -1.219696e-01 -4.387455e-02 -1.130020e-01 -6.158379e-02 -1.355932e-01   \n",
       "max    4.902760e+01  2.600907e+02  5.456518e+01  1.134409e+02  7.335457e+01   \n",
       "\n",
       "       ...           157           158           159           160  \\\n",
       "count  ...  2.037690e+05  2.037690e+05  2.037690e+05  2.037690e+05   \n",
       "mean   ... -8.620828e-15  1.915670e-15  5.923961e-15  2.052502e-14   \n",
       "std    ...  1.000002e+00  1.000002e+00  1.000002e+00  1.000002e+00   \n",
       "min    ... -5.770994e-01 -6.262286e-01 -9.790738e-01 -9.785560e-01   \n",
       "25%    ... -5.696264e-01 -5.946915e-01 -9.790738e-01 -9.785560e-01   \n",
       "50%    ... -4.799511e-01 -4.559278e-01  2.411283e-01  2.414064e-01   \n",
       "75%    ...  1.552495e-01  1.212026e-01  1.305594e+00  1.398764e+00   \n",
       "max    ...  7.862953e+00  7.914041e+00  1.461330e+00  1.461369e+00   \n",
       "\n",
       "                161           162           163           164           165  \\\n",
       "count  2.037690e+05  2.037690e+05  2.037690e+05  2.037690e+05  2.037690e+05   \n",
       "mean   6.975481e-15  2.943190e-14 -1.748264e-14  8.625937e-14 -6.533539e-14   \n",
       "std    1.000002e+00  1.000002e+00  1.000002e+00  1.000002e+00  1.000002e+00   \n",
       "min   -2.160569e-01 -1.259391e-01 -1.311553e-01 -2.698175e-01 -1.760926e+00   \n",
       "25%   -9.888874e-02 -8.749016e-02 -1.311553e-01 -1.405971e-01 -1.206134e-01   \n",
       "50%    1.827940e-02 -8.749016e-02 -1.311553e-01 -9.752359e-02 -1.206134e-01   \n",
       "75%    1.827940e-02 -8.749016e-02 -8.467423e-02 -9.752359e-02  1.520067e-01   \n",
       "max    1.170692e+02  2.518490e+02  2.387835e+02  1.057340e+02  1.519700e+00   \n",
       "\n",
       "                166  \n",
       "count  2.037690e+05  \n",
       "mean  -1.622942e-14  \n",
       "std    1.000002e+00  \n",
       "min   -1.760984e+00  \n",
       "25%   -1.197925e-01  \n",
       "50%   -1.197925e-01  \n",
       "75%    1.199710e-01  \n",
       "max    1.521399e+00  \n",
       "\n",
       "[8 rows x 167 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_features.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The file with features is anonimyzed, it doesn't even have column names.\n",
    "#Let's see what can be derived from the data description:\n",
    "\n",
    "#1) The first column (column 0) is the transaction id\n",
    "\n",
    "#2) Column 1 represents timesteps for each node. These timestamps have an interval of ~2 weeks. \n",
    "#   Each timestamp contains connected component of transactions,\n",
    "#   which appeared on the blockchain within less than three hours between each other\n",
    "\n",
    "#3) Next 93 features represent local information about the transaction — including the time step described above, \n",
    "#   number of inputs/outputs, transaction fee, output volume and aggregated figures \n",
    "#   such as average BTC received (spent) by the inputs/outputs and average number \n",
    "#   of incoming (outgoing) transactions associated with the inputs/outputs.\n",
    "\n",
    "#4) The  72 features represent nonlocal (graph) information in the form of aggregated features,\n",
    "#   obtained using transaction information one-hop backward/forward from the center node — giving the maximum\n",
    "#   ,minimum, standard deviation and correlation coefficients of the neighbour transactions \n",
    "#   for the same information data (number of inputs/outputs, transaction fee, etc.)\n",
    "\n",
    "# so we will rename our features acoordingly \n",
    "df_features.columns = ['ID', 'Time step'] + [f'Trans_feature_{i}' for i in range(93)] + [f'Graph_feature_{i}' for i in range(72)] \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>ID</th>\n",
       "      <th>Time step</th>\n",
       "      <th>Trans_feature_0</th>\n",
       "      <th>Trans_feature_1</th>\n",
       "      <th>Trans_feature_2</th>\n",
       "      <th>Trans_feature_3</th>\n",
       "      <th>Trans_feature_4</th>\n",
       "      <th>Trans_feature_5</th>\n",
       "      <th>Trans_feature_6</th>\n",
       "      <th>Trans_feature_7</th>\n",
       "      <th>...</th>\n",
       "      <th>Graph_feature_62</th>\n",
       "      <th>Graph_feature_63</th>\n",
       "      <th>Graph_feature_64</th>\n",
       "      <th>Graph_feature_65</th>\n",
       "      <th>Graph_feature_66</th>\n",
       "      <th>Graph_feature_67</th>\n",
       "      <th>Graph_feature_68</th>\n",
       "      <th>Graph_feature_69</th>\n",
       "      <th>Graph_feature_70</th>\n",
       "      <th>Graph_feature_71</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>230425980</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.171469</td>\n",
       "      <td>-0.184668</td>\n",
       "      <td>-1.201369</td>\n",
       "      <td>-0.121970</td>\n",
       "      <td>-0.043875</td>\n",
       "      <td>-0.113002</td>\n",
       "      <td>-0.061584</td>\n",
       "      <td>-0.162097</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.562153</td>\n",
       "      <td>-0.600999</td>\n",
       "      <td>1.461330</td>\n",
       "      <td>1.461369</td>\n",
       "      <td>0.018279</td>\n",
       "      <td>-0.087490</td>\n",
       "      <td>-0.131155</td>\n",
       "      <td>-0.097524</td>\n",
       "      <td>-0.120613</td>\n",
       "      <td>-0.119792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5530458</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.171484</td>\n",
       "      <td>-0.184668</td>\n",
       "      <td>-1.201369</td>\n",
       "      <td>-0.121970</td>\n",
       "      <td>-0.043875</td>\n",
       "      <td>-0.113002</td>\n",
       "      <td>-0.061584</td>\n",
       "      <td>-0.162112</td>\n",
       "      <td>...</td>\n",
       "      <td>0.947382</td>\n",
       "      <td>0.673103</td>\n",
       "      <td>-0.979074</td>\n",
       "      <td>-0.978556</td>\n",
       "      <td>0.018279</td>\n",
       "      <td>-0.087490</td>\n",
       "      <td>-0.131155</td>\n",
       "      <td>-0.097524</td>\n",
       "      <td>-0.120613</td>\n",
       "      <td>-0.119792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>232022460</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.172107</td>\n",
       "      <td>-0.184668</td>\n",
       "      <td>-1.201369</td>\n",
       "      <td>-0.121970</td>\n",
       "      <td>-0.043875</td>\n",
       "      <td>-0.113002</td>\n",
       "      <td>-0.061584</td>\n",
       "      <td>-0.162749</td>\n",
       "      <td>...</td>\n",
       "      <td>0.670883</td>\n",
       "      <td>0.439728</td>\n",
       "      <td>-0.979074</td>\n",
       "      <td>-0.978556</td>\n",
       "      <td>-0.098889</td>\n",
       "      <td>-0.106715</td>\n",
       "      <td>-0.131155</td>\n",
       "      <td>-0.183671</td>\n",
       "      <td>-0.120613</td>\n",
       "      <td>-0.119792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>232438397</td>\n",
       "      <td>1</td>\n",
       "      <td>0.163054</td>\n",
       "      <td>1.963790</td>\n",
       "      <td>-0.646376</td>\n",
       "      <td>12.409294</td>\n",
       "      <td>-0.063725</td>\n",
       "      <td>9.782742</td>\n",
       "      <td>12.414558</td>\n",
       "      <td>-0.163645</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.577099</td>\n",
       "      <td>-0.613614</td>\n",
       "      <td>0.241128</td>\n",
       "      <td>0.241406</td>\n",
       "      <td>1.072793</td>\n",
       "      <td>0.085530</td>\n",
       "      <td>-0.131155</td>\n",
       "      <td>0.677799</td>\n",
       "      <td>-0.120613</td>\n",
       "      <td>-0.119792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>230460314</td>\n",
       "      <td>1</td>\n",
       "      <td>1.011523</td>\n",
       "      <td>-0.081127</td>\n",
       "      <td>-1.201369</td>\n",
       "      <td>1.153668</td>\n",
       "      <td>0.333276</td>\n",
       "      <td>1.312656</td>\n",
       "      <td>-0.061584</td>\n",
       "      <td>-0.163523</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.511871</td>\n",
       "      <td>-0.400422</td>\n",
       "      <td>0.517257</td>\n",
       "      <td>0.579382</td>\n",
       "      <td>0.018279</td>\n",
       "      <td>0.277775</td>\n",
       "      <td>0.326394</td>\n",
       "      <td>1.293750</td>\n",
       "      <td>0.178136</td>\n",
       "      <td>0.179117</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 167 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          ID  Time step  Trans_feature_0  Trans_feature_1  Trans_feature_2  \\\n",
       "0  230425980          1        -0.171469        -0.184668        -1.201369   \n",
       "1    5530458          1        -0.171484        -0.184668        -1.201369   \n",
       "2  232022460          1        -0.172107        -0.184668        -1.201369   \n",
       "3  232438397          1         0.163054         1.963790        -0.646376   \n",
       "4  230460314          1         1.011523        -0.081127        -1.201369   \n",
       "\n",
       "   Trans_feature_3  Trans_feature_4  Trans_feature_5  Trans_feature_6  \\\n",
       "0        -0.121970        -0.043875        -0.113002        -0.061584   \n",
       "1        -0.121970        -0.043875        -0.113002        -0.061584   \n",
       "2        -0.121970        -0.043875        -0.113002        -0.061584   \n",
       "3        12.409294        -0.063725         9.782742        12.414558   \n",
       "4         1.153668         0.333276         1.312656        -0.061584   \n",
       "\n",
       "   Trans_feature_7  ...  Graph_feature_62  Graph_feature_63  Graph_feature_64  \\\n",
       "0        -0.162097  ...         -0.562153         -0.600999          1.461330   \n",
       "1        -0.162112  ...          0.947382          0.673103         -0.979074   \n",
       "2        -0.162749  ...          0.670883          0.439728         -0.979074   \n",
       "3        -0.163645  ...         -0.577099         -0.613614          0.241128   \n",
       "4        -0.163523  ...         -0.511871         -0.400422          0.517257   \n",
       "\n",
       "   Graph_feature_65  Graph_feature_66  Graph_feature_67  Graph_feature_68  \\\n",
       "0          1.461369          0.018279         -0.087490         -0.131155   \n",
       "1         -0.978556          0.018279         -0.087490         -0.131155   \n",
       "2         -0.978556         -0.098889         -0.106715         -0.131155   \n",
       "3          0.241406          1.072793          0.085530         -0.131155   \n",
       "4          0.579382          0.018279          0.277775          0.326394   \n",
       "\n",
       "   Graph_feature_69  Graph_feature_70  Graph_feature_71  \n",
       "0         -0.097524         -0.120613         -0.119792  \n",
       "1         -0.097524         -0.120613         -0.119792  \n",
       "2         -0.183671         -0.120613         -0.119792  \n",
       "3          0.677799         -0.120613         -0.119792  \n",
       "4          1.293750          0.178136          0.179117  \n",
       "\n",
       "[5 rows x 167 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Now let's check our features again \n",
    "df_features.head()\n",
    "# Everything is more clear now "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>txId1</th>\n",
       "      <th>txId2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>230425980</td>\n",
       "      <td>5530458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>232022460</td>\n",
       "      <td>232438397</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>230460314</td>\n",
       "      <td>230459870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>230333930</td>\n",
       "      <td>230595899</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>232013274</td>\n",
       "      <td>232029206</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       txId1      txId2\n",
       "0  230425980    5530458\n",
       "1  232022460  232438397\n",
       "2  230460314  230459870\n",
       "3  230333930  230595899\n",
       "4  232013274  232029206"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Exploring the basics of edgelist \n",
    "df_edgelist.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 234355 entries, 0 to 234354\n",
      "Data columns (total 2 columns):\n",
      " #   Column  Non-Null Count   Dtype\n",
      "---  ------  --------------   -----\n",
      " 0   txId1   234355 non-null  int64\n",
      " 1   txId2   234355 non-null  int64\n",
      "dtypes: int64(2)\n",
      "memory usage: 3.6 MB\n"
     ]
    }
   ],
   "source": [
    "df_edgelist.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>ID1</th>\n",
       "      <th>ID2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>230425980</td>\n",
       "      <td>5530458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>232022460</td>\n",
       "      <td>232438397</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>230460314</td>\n",
       "      <td>230459870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>230333930</td>\n",
       "      <td>230595899</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>232013274</td>\n",
       "      <td>232029206</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         ID1        ID2\n",
       "0  230425980    5530458\n",
       "1  232022460  232438397\n",
       "2  230460314  230459870\n",
       "3  230333930  230595899\n",
       "4  232013274  232029206"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# renaming the columns \n",
    "df_edgelist.columns = ['ID1'] + ['ID2']\n",
    "# Check \n",
    "df_edgelist.head()\n",
    "# we can clearly see the 2 parts (ID1 and ID2) that carry out the transactions "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # create a graph from my edgelist \n",
    "# G = nx.read_edgelist('elliptic_txs_edgelist.csv')\n",
    "# # type of G \n",
    "# type(G)\n",
    "# # created graph info \n",
    "# nx.info(G)\n",
    "# # Draw graph \n",
    "# nx.draw(G)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Total Number of Transactions in each time step')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEICAYAAABYoZ8gAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAe4UlEQVR4nO3de7gcVZ3u8e9rAnI3BBKEJBjQHBV8BmS2EMVHkTAQEAkzwpx4vEQmnugMODA6I8HhGLk56HFAnVE8GYgGjhIiyhAZRsjDxdtzuARBFJDJ5pptQrIhF1AEDPzOH2s1VHa6d3cne/fe2ev9PE8/XbVqVdVa1dW/Wr2qukoRgZmZleNVQ10AMzPrLAd+M7PCOPCbmRXGgd/MrDAO/GZmhXHgNzMrjAN/h0jaQVJImjjUZQGQdKGkS4dw/f9b0lOSHh2qMnSSpCskfaqD63uzpCc7tb5WSHpS0tStmL+j23AkKzrwS/pd5fWSpD9Uxj/YZN7pkroHsCy3Sfq9pNdW0o6X9JuBWsdwIWkK8NfAlIiY3Gfa7Mpn8If8udTGh1Uga0TSaZJ+VE2LiA9HxEWdKkNEPBARe3ZqfQNtOGzDerb24DVcFB34I2KX2gt4HHhfJe07Q1Ck54DPDsF6t4qk0W3O8jrgiYhY23dCRFxW+Uz+HHi48plsFsi2YN1mxSs68DcjaUdJX5e0SlJP7p7YTtIewDXA/pXW6B6SDpd0u6QNklZKurjNwPQV4BRJ+9Ypy2ZdRZIWSTo7D0+X1C3p7Nwq+a2k4yTNkPRQ7lb5dJ/F7izp+5KekXSnpAMry54k6dq8rIclfaIy7UJJ35V0laRngJl1yjs25+mV9Iikzyg5HvhhZdt9s43tU1v2k5I+Jel+YG1OO0fSo7kuv5J0bCX/aZJuzJ/lhryd3lOZ/glJj+V5H5L0Fzn9AEk/kbRW0hpJ35K0S2W+/SX9MJenV9KXJL0N+GfgqFy/npz3aklzK/P+bd6uT+Zp43P6Lvlz/lievlbSlyvzHSjp57kevZK+1WAbvUXSc5XxZZL+l6Q7JD0t6TpJr+lnG78/b8f1kn4s6U2Vaf1ta0n6pKQH8/R7JR1QWfRhku7P5b9c0nZ11t10Gyr/GpY0L+/bPZKOkfQXlf399MoyR0v6fN4Xn1TqNtqtQd33kXRDrvtTkm7M6dcAewA353L9TU5/d96u6yXdJentfbb7OZLuznX+XqP1dlRE+JVuW/EocFSftC8BPwX2BPYC7gT+MU+bDnT3yX8o8DZgFPB6oBv4RJ62AxDAxAbrvw34EPAN4NKcdjzwm0bzA4uAsyvl+SNwJjAa+CTwBHAFsDPwVtIvigk5/4XAC8AJwHbA2cCDueyjgF/lZW0P/DfSL6J3V+Z9HjiO1HjYsU59FgPfA3YB3gA8Anyw0bZrsE3q5gOezNtr79q6SQef1+aynwI8DYzN007L2+Z/5Ol/X1suMI508Ng/j08A3pSHDwCOyNtnb+AO4Pw8bXvgv4DzgZ3y6x2V9f2oT5mvBubm4ROAVcBbgB2By4D/zNN2yZ/z94Bd8370NPDOPP2HwBmA8ryHN9h2bwGeq4wvAx4A9svruL2279SZ953ASuCQvL3+Js87qoVtfQrwMHBQLuObeGWfexL4Sd7m4/M+8aEGZWi2DY8HNubPcjTwd3mbLiTt712k/f21Of/ZwK253DuSvhf/1mDd/0I68IzOn/O7+ux7UyvjrweeAo4kfRdOANYAr6ls90dI36FdgeuBbw55vBvqAgyXF/UD/2+BIyvjM3glEDcNXsBc4Mo83Grg3wd4BphC+4F/A/CqPD4u5z+okv8+YHoevhC4tTJtdN6B3wa8G1jep3znAJdU5r2xn3q/GniRHExz2um1L3Ir266/fPnL95dN5u0GpuXh04B7KtPG522zC6kFtw54H/DqJsv8EPDTPPxnwGO17d0nX7OgdRXwucq0PXJ59uSVwH9wZfr1wGl5+AfAV8kBrZ+y1gv8Z1TGPwNc3WDeK4Az63wX/rSFbf1zYHaDfE8CJ1bGvwF8uUHeVgL/U4Dy+N55ux1Yyf8g+TsNrAAOq0ybAjzTYN0Xkb5b+zWoQzXwn1f7XlTSfg68v7Ldz65MOxTY0GzfH+yXu3oakCRS6+CxSvJjpBZho3kOkPSfklZLehr4HOnL3LKIWAnMBz7fdqGhNyJeysN/yO+rK9P/QAosNSsq691IauXtQ+qDn5x/uq6XtB74FGl7bDZvHa8ltX4er6T1u+22wCbrlzSn0jWxHpjMptv+icrws/l9l4h4CphFajGuVureen1e5sTcvbAyf57frCxzEvBIZXu3Yx8q+1Uuw+/ZdPv0LW/tczsd2A24R9IvJX2gjfU2WmZfrwM+1+fz371WvibbehLw0ACUoRW9kaMp/ezvkkblst9YKfOdwHaSxtRZ7nlAL/BjSf8l6Yx+yvA64KN9ttXBpM+4prqvPgbsJmnnVis5GBz4G8g71BOkD7ZmX1LLB1Lroq9/A34BvD4idgPOJf3cbdeFpBbNWyppL5C6K3aqpFUD8ZaYVBvIX459SMF/BemXxpjKa9eI+PPKvP3d1vUJ4CXS9qqpbruB8PL6cx/yRcBsUpfDGNIvuJa2fUQsiYgjScFhJfCvedI/k35FHZA/z09UlrkC2C83EBqWrYGVVPYrSWNJ3RNNt09ErIiIU0gt3E8Bl0vap8ls7VoBfLbP579TRCxpYVuvIHV/bK0Bu21wRLxI6gZ6V5867RAR6+vkXxcRn4yIfYH/Dnxe0mENyrWC1HVTXe7OEfEvlTyTKsP7Ak9HxO8Hqn5bwoG/f1cC85RO3I4H/hH4v3naamC8Kif7SH14GyLid0onSv/nlqw0InqBrwH/UEl7idTv/kFJoyS9D3h7g0W06h35JNl2pJ/+T5EOXD8DkHSG0knl0ZL+RNIhLZb/edLJ7y9I2jm3oE/nlW030HYhHWh6gVdJOpXUCm1K6ST2cZJ2JPUJ/57UTQXp83wGeFrSZNKvgpofk85znKN0EcBOkt6Rp60G9lXjE/tXAh9XOlG7I/BF4IaIaHq5qqSZkvbODZP1pED0YpPZ2vV/gDMkHaJkV0knStqB5tv6UuCzeX+RpDdJ2pJfes22Ybu+CXyxVhZJeyldaLAZpQsiagf1DaT61rbxamD/SvZvAx+Q9B5Jr8r7wlGS9qrk+StJUyTtSvolf9UA1WmLOfD373PA/aS+8XtIfXdfytN+CSwBHss/8caSAsPHJP0O+Dpb9wF/mc1brKeRWiDrSJc6XrcVywf4PvBXeXnvJ/VLvhgRfySduH0H6adpL3AJ7f0s/3h+fwy4mRQQBuUS2Yi4g3SC9G5Sa3oC6fNqxWjSAX01qf/2YNJBCtIJwfeQTl5eTTrhWlvnC8CxpJOIvyW1ek/Ik6/Pab2Sql2FtXn/nfT5Xgf0kPr4Z7VY3sOBX+R97EpSf/rqJvO0JSJ+QtqXLyUdXB4k7XfRwrb+Nunk6PdJ2+0qUtdUu/rdhlvgn0gnln+cu+1+RrrgoZ4DSQf2Z4BbgC9ExLI87XzgS5LWSfrriFgOnJzTnyLtB59k0+/uFaSLHXpIjYt/YIjVToyYmdkAk7SMdAJ70VCXpcotfjOzwjjwm5kVxl09ZmaFcYvfzKwww/oGV3vuuWdMnjx5qIthZrZNueuuu56MiHGNpg/rwD958mSWLVvWPKOZmb2s2SWw7uoxMyuMA7+ZWWFaCvyS/k7SfZJ+LenK/Df+/ZTuPb9c6b7s2+e8r87j3Xn65MpyzsrpD0o6ZnCqZGZm/Wka+PO9Lf4W6IqIt5DuwT2TdH+RiyNiCukv/7PzLLOBdRHxBuDinK92I62ZpL9DTwe+kW8MZmZmHdRqV89oYMd8w6SdSHe6O5J0/xJIDz84MQ/PyOPk6dPyzY5mAIsi4vmIeIR0D+9Dt74KZmbWjqaBPyJ+S7qh1OOkgL8BuAtYn+/hDunmQ7U78E0g3386T99AugnVy+l15jEzsw5ppatnd1JrfT/S/dp3Jt2VsK/aX4Ab3Z+8pfuW54c8LJO0rLe3t1nxzMysTa109RxFetJQb75d7w9It+sdU7lX9kTSLVohteQnQXrAMfAa0jNNX06vM8/LImJ+RHRFRNe4cQ3/f2BmZluolcD/ODA1P2hCwDTSPepvAU7KeWYB1+bhJbxyb/GTgJvzQyOWADPzVT/7kZ55ecfAVMPMzFrV9J+7EXG7pKtJT2baSHoAw3zgP4BFks7PaZflWS4DrpDUTWrpz8zLuU/SYtJBYyNwan4k2qCZPPc/Nkt79ML3DuYqzcyGvZZu2RAR84B5fZIfps5VORHxHOmJNPWWcwFwQZtlNDOzAeR/7pqZFcaB38ysMA78ZmaFceA3MyuMA7+ZWWEc+M3MCuPAb2ZWGAd+M7PCOPCbmRXGgd/MrDAO/GZmhXHgNzMrjAO/mVlhHPjNzArjwG9mVhgHfjOzwjjwm5kVpqUncJmZbUv82NX+NW3xS3qjpHsqr6clnSFprKSlkpbn991zfkn6mqRuSfdKOqSyrFk5/3JJsxqv1czMBkvTwB8RD0bEwRFxMPCnwLPANcBc4KaImALclMcBjgWm5Ncc4BIASWNJz+09jPSs3nm1g4WZmXVOu33804CHIuIxYAawMKcvBE7MwzOAyyO5DRgjaW/gGGBpRKyNiHXAUmD6VtfAzMza0m7gnwlcmYf3iohVAPl9fE6fAKyozNOT0xqlb0LSHEnLJC3r7e1ts3hmZtZMy4Ff0vbACcD3mmWtkxb9pG+aEDE/IroiomvcuHGtFs/MzFrUTov/WOAXEbE6j6/OXTjk9zU5vQeYVJlvIrCyn3QzM+ugdgL/B3ilmwdgCVC7MmcWcG0l/SP56p6pwIbcFXQDcLSk3fNJ3aNzmpmZdVBL1/FL2gn4M+DjleQLgcWSZgOPAyfn9OuB44Bu0hVApwBExFpJ5wF35nznRsTara6BmZm1paXAHxHPAnv0SXuKdJVP37wBnNpgOQuABe0Xc3jyn0TMbFvkWzaYmRXGt2zYhvkXh5ltCbf4zcwK48BvZlYYd/VUuOvEzErgFr+ZWWEc+M3MCuPAb2ZWGAd+M7PCOPCbmRXGV/WY2Vbx1XDbHrf4zcwK48BvZlYYB34zs8I48JuZFcaB38ysMA78ZmaFaSnwSxoj6WpJv5H0gKS3Sxoraamk5fl995xXkr4mqVvSvZIOqSxnVs6/XNKsxms0M7PB0mqL/6vAjyLiTcBBwAPAXOCmiJgC3JTHAY4FpuTXHOASAEljgXnAYcChwLzawcLMzDqn6R+4JO0GvAv4KEBEvAC8IGkGcETOthC4FTgTmAFcnp+9e1v+tbB3zru09oB1SUuB6cCVA1ed4c9/djGzodZKi39/oBf4lqS7JV0qaWdgr4hYBZDfx+f8E4AVlfl7clqj9E1ImiNpmaRlvb29bVfIzMz610rgHw0cAlwSEW8Ffs8r3Tr1qE5a9JO+aULE/IjoioiucePGtVA8MzNrRyuBvwfoiYjb8/jVpAPB6tyFQ35fU8k/qTL/RGBlP+lmZtZBTfv4I+IJSSskvTEiHgSmAffn1yzgwvx+bZ5lCXCapEWkE7kbImKVpBuAL1RO6B4NnDWw1TGrz+dWzF7R6t05Pwl8R9L2wMPAKaRfC4slzQYeB07Oea8HjgO6gWdzXiJiraTzgDtzvnNrJ3rNzKxzWgr8EXEP0FVn0rQ6eQM4tcFyFgAL2imgmZkNLP9z18ysMA78ZmaF8RO4WuATg2Y2krjFb2ZWGAd+M7PCuKvHzKwfI7Gr1y1+M7PCuMVvZi0ZiS3fUrnFb2ZWGAd+M7PCOPCbmRXGffxmts3yeYct4xa/mVlhHPjNzArjwG9mVhgHfjOzwjjwm5kVpqWreiQ9CjwDvAhsjIguSWOBq4DJwKPAX0bEOkkCvkp6/OKzwEcj4hd5ObOAs/Niz4+IhQNXFTOzobctXGnUTov/PRFxcETUHsE4F7gpIqYAN+VxgGOBKfk1B7gEIB8o5pEewH4oMK/y4HUzM+uQrbmOfwZwRB5eCNwKnJnTL8/P3r1N0hhJe+e8S2sPWJe0FJgOXLkVZTAza9m20BrvhFZb/AHcKOkuSXNy2l4RsQogv4/P6ROAFZV5e3Jao/RNSJojaZmkZb29va3XxMzMWtJqi//wiFgpaTywVNJv+smrOmnRT/qmCRHzgfkAXV1dm00vjVsoZp1R0netpRZ/RKzM72uAa0h99KtzFw75fU3O3gNMqsw+EVjZT7qZmXVQ08AvaWdJu9aGgaOBXwNLgFk52yzg2jy8BPiIkqnAhtwVdANwtKTd80ndo3OamZl1UCtdPXsB16SrNBkNfDcifiTpTmCxpNnA48DJOf/1pEs5u0mXc54CEBFrJZ0H3JnznVs70WtmZp3TNPBHxMPAQXXSnwKm1UkP4NQGy1oALGi/mGZmNlD8z10zs8I48JuZFcaB38ysMA78ZmaFceA3MyuMn7lrVkdJ/+K08hQZ+P2lNrOSFRn4bVM+EI5c/mytHgd+Mxv2fAAbWA78w4R3bDPrFF/VY2ZWGAd+M7PCuKtnkLjrxsyGKwf+EWikH3RcP7Ot464eM7PCOPCbmRXGXT1WNHerWInc4jczK0zLgV/SKEl3S7ouj+8n6XZJyyVdJWn7nP7qPN6dp0+uLOOsnP6gpGMGujJmZtZcOy3+04EHKuNfBC6OiCnAOmB2Tp8NrIuINwAX53xIOgCYCRwITAe+IWnU1hXfzMza1VLglzQReC9waR4XcCRwdc6yEDgxD8/I4+Tp03L+GcCiiHg+Ih4BuoFDB6ISZmbWulZb/F8BPgO8lMf3ANZHxMY83gNMyMMTgBUAefqGnP/l9DrzvEzSHEnLJC3r7e1toypmZtaKpoFf0vHAmoi4q5pcJ2s0mdbfPK8kRMyPiK6I6Bo3blyz4pmZWZtauZzzcOAESccBOwC7kX4BjJE0OrfqJwIrc/4eYBLQI2k08BpgbSW9pjqPmZl1SNMWf0ScFRETI2Iy6eTszRHxQeAW4KScbRZwbR5eksfJ02+OiMjpM/NVP/sBU4A7BqwmZmbWkq35A9eZwCJJ5wN3A5fl9MuAKyR1k1r6MwEi4j5Ji4H7gY3AqRHx4las38zMtkBbgT8ibgVuzcMPU+eqnIh4Dji5wfwXABe0W0gzMxs4/ueumVlhHPjNzArjwG9mVhjfndO2iO9qabbtcovfzKwwDvxmZoVxV4+NGO5+sk7alvc3B34btrblL5bZcOauHjOzwjjwm5kVxl091hHutjEbPtziNzMrjFv8ZjYo/Ctv+HKL38ysMA78ZmaFceA3MyuM+/jNCuT+97I1bfFL2kHSHZJ+Kek+Sefk9P0k3S5puaSrJG2f01+dx7vz9MmVZZ2V0x+UdMxgVcrMzBprpavneeDIiDgIOBiYLmkq8EXg4oiYAqwDZuf8s4F1EfEG4OKcD0kHkJ6/eyAwHfiGpFEDWRkzM2uuaeCP5Hd5dLv8CuBI4OqcvhA4MQ/PyOPk6dMkKacviojnI+IRoJs6z+w1M7PB1dLJXUmjJN0DrAGWAg8B6yNiY87SA0zIwxOAFQB5+gZgj2p6nXmq65ojaZmkZb29ve3XyMzM+tXSyd2IeBE4WNIY4BrgzfWy5Xc1mNYove+65gPzAbq6ujabbsObTxqaNTZcvh9tXc4ZEeuBW4GpwBhJtQPHRGBlHu4BJgHk6a8B1lbT68xjZmYd0rTFL2kc8MeIWC9pR+Ao0gnbW4CTgEXALODaPMuSPP7/8vSbIyIkLQG+K+kiYB9gCnDHANfHzIa54dLqLVkrXT17AwvzFTivAhZHxHWS7gcWSTofuBu4LOe/DLhCUjeppT8TICLuk7QYuB/YCJyau5DMzKyDmgb+iLgXeGud9Iepc1VORDwHnNxgWRcAF7RfTDPrj1vR1g7/c9fMhg0fwDrD9+oxMyuMA7+ZWWEc+M3MCuPAb2ZWGAd+M7PC+KoeswHgq1FsW+LAb2Yv8wGsDA78ZtsIB+WRq9Ofrfv4zcwK48BvZlYYB34zs8K4j9+sDe5nt5HALX4zs8I48JuZFcaB38ysMA78ZmaFaRr4JU2SdIukByTdJ+n0nD5W0lJJy/P77jldkr4mqVvSvZIOqSxrVs6/XNKswauWmZk10kqLfyPw6Yh4MzAVOFXSAcBc4KaImALclMcBjiU9SH0KMAe4BNKBApgHHEZ6ZOO82sHCzMw6p5Vn7q4CVuXhZyQ9AEwAZgBH5GwLgVuBM3P65RERwG2SxkjaO+ddGhFrASQtBaYDVw5gfawfvhTRzKDN6/glTSY9eP12YK98UCAiVkkan7NNAFZUZuvJaY3S+65jDumXAvvuu287xbNB4IOF2cjTcuCXtAvwfeCMiHhaUsOsddKin/RNEyLmA/MBurq6NptuI48PLmad1dJVPZK2IwX970TED3Ly6tyFQ35fk9N7gEmV2ScCK/tJNzOzDmrlqh4BlwEPRMRFlUlLgNqVObOAayvpH8lX90wFNuQuoRuAoyXtnk/qHp3TzMysg1rp6jkc+DDwK0n35LTPAhcCiyXNBh4HTs7TrgeOA7qBZ4FTACJiraTzgDtzvnNrJ3rNzKxzWrmq52fU758HmFYnfwCnNljWAmBBOwU0M7OB5X/umpkVxoHfzKwwDvxmZoXxg1jMhoj/v2BDxS1+M7PCOPCbmRXGgd/MrDAO/GZmhXHgNzMrjAO/mVlhHPjNzArjwG9mVhgHfjOzwjjwm5kVxoHfzKwwDvxmZoVx4DczK0wrz9xdIGmNpF9X0sZKWippeX7fPadL0tckdUu6V9IhlXlm5fzLJc2qty4zMxt8rdyW+dvAvwKXV9LmAjdFxIWS5ubxM4FjgSn5dRhwCXCYpLHAPKALCOAuSUsiYt1AVcRsuPLtl224adrij4ifAH0fij4DWJiHFwInVtIvj+Q2YIykvYFjgKURsTYH+6XA9IGogJmZtWdL+/j3iohVAPl9fE6fAKyo5OvJaY3Szcyswwb65K7qpEU/6ZsvQJojaZmkZb29vQNaODMz2/LAvzp34ZDf1+T0HmBSJd9EYGU/6ZuJiPkR0RURXePGjdvC4pmZWSNbGviXALUrc2YB11bSP5Kv7pkKbMhdQTcAR0vaPV8BdHROMzOzDmt6VY+kK4EjgD0l9ZCuzrkQWCxpNvA4cHLOfj1wHNANPAucAhARayWdB9yZ850bEX1PGJuZWQc0DfwR8YEGk6bVyRvAqQ2WswBY0FbpzMxswPmfu2ZmhXHgNzMrjAO/mVlhHPjNzArjwG9mVhgHfjOzwjjwm5kVxoHfzKwwDvxmZoVx4DczK4wDv5lZYRz4zcwK48BvZlYYB34zs8I48JuZFcaB38ysMA78ZmaFceA3MytMxwO/pOmSHpTULWlup9dvZla6jgZ+SaOArwPHAgcAH5B0QCfLYGZWuk63+A8FuiPi4Yh4AVgEzOhwGczMiqaI6NzKpJOA6RHxsTz+YeCwiDitkmcOMCePvhF4cABWvSfw5AAsZ1tTar2h3LqXWm8ot+716v26iBjXaIbRg1uezahO2iZHnoiYD8wf0JVKyyKiayCXuS0otd5Qbt1LrTeUW/ctqXenu3p6gEmV8YnAyg6XwcysaJ0O/HcCUyTtJ2l7YCawpMNlMDMrWke7eiJio6TTgBuAUcCCiLivA6se0K6jbUip9YZy615qvaHcurdd746e3DUzs6Hnf+6amRXGgd/MrDAjPvCXcosISQskrZH060raWElLJS3P77sPZRkHg6RJkm6R9ICk+ySdntNLqPsOku6Q9Mtc93Ny+n6Sbs91vypfSDHiSBol6W5J1+XxEV9vSY9K+pWkeyQty2lt7+sjOvAXdouIbwPT+6TNBW6KiCnATXl8pNkIfDoi3gxMBU7Nn3EJdX8eODIiDgIOBqZLmgp8Ebg4130dMHsIyziYTgceqIyXUu/3RMTBlWv3297XR3Tgp6BbRETET4C1fZJnAAvz8ELgxI4WqgMiYlVE/CIPP0MKBBMoo+4REb/Lo9vlVwBHAlfn9BFZd0kTgfcCl+ZxUUC9G2h7Xx/pgX8CsKIy3pPTSrFXRKyCFCCB8UNcnkElaTLwVuB2Cql77u64B1gDLAUeAtZHxMacZaTu818BPgO8lMf3oIx6B3CjpLvy7W1gC/b1Tt+yodOa3iLCRgZJuwDfB86IiKdTA3Dki4gXgYMljQGuAd5cL1tnSzW4JB0PrImIuyQdUUuuk3VE1Ts7PCJWShoPLJX0my1ZyEhv8Zd+i4jVkvYGyO9rhrg8g0LSdqSg/52I+EFOLqLuNRGxHriVdJ5jjKRao24k7vOHAydIepTUfXsk6RfASK83EbEyv68hHegPZQv29ZEe+Eu/RcQSYFYengVcO4RlGRS5b/cy4IGIuKgyqYS6j8stfSTtCBxFOsdxC3BSzjbi6h4RZ0XExIiYTPpO3xwRH2SE11vSzpJ2rQ0DRwO/Zgv29RH/z11Jx5FaA7VbRFwwxEUaFJKuBI4g3aJ1NTAP+HdgMbAv8DhwckT0PQG8TZP0TuCnwK94pb/3s6R+/pFe9z8hncwbRWrELY6IcyXtT2oJjwXuBj4UEc8PXUkHT+7q+fuIOH6k1zvX75o8Ohr4bkRcIGkP2tzXR3zgNzOzTY30rh4zM+vDgd/MrDAO/GZmhXHgNzMrjAO/mVlhHPjNzArjwG9mVpj/D3J87uDNspNUAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# But how are these transactions distributed through our various time steps ? \n",
    "# Let's find out \n",
    "# We will create a barplot to represent the development of the transactions through our defined time steps \n",
    "group1 = df_features.groupby('Time step').count()['ID']\n",
    "timesteps = np.arange(0,49)\n",
    "bar1 = plt.bar(timesteps,group1)\n",
    "plt.title('Total Number of Transactions in each time step')\n",
    "\n",
    "# Considering the fact that each time step is 2 weeks we an overall time period of 98 weeks i.e. 24.5 months i.e. \n",
    "# near 2 years in total we can observe some significant bursts and drops of activity especially around weeks 10,12 and 25 \n",
    "\n",
    "# but we can not understand which kind of activities (legal, illegal, unknown) causes these sudden bursts and drops"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 203769 entries, 0 to 203768\n",
      "Columns: 168 entries, ID to Graph_feature_71\n",
      "dtypes: float64(165), int64(2), object(1)\n",
      "memory usage: 262.7+ MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "#In order to explore further  which kind of activities (legal, illegal, unknown) causes \n",
    "#these sudden bursts and drops this fact we will split our transactions to the 3 provided classes and visualize again \n",
    "\n",
    "# but first we have to merge our features and class dataframes into one \n",
    "df_class_and_features = pd.merge(df_classes, df_features)\n",
    "df_class_and_features.head()\n",
    "print(df_class_and_features.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                   ID  Trans_feature_0  Trans_feature_1  Trans_feature_2  \\\n",
      "Time step Class                                                            \n",
      "1         3      5733             5733             5733             5733   \n",
      "          1        17               17               17               17   \n",
      "          2      2130             2130             2130             2130   \n",
      "2         3      3427             3427             3427             3427   \n",
      "          1        18               18               18               18   \n",
      "...               ...              ...              ...              ...   \n",
      "48        1        36               36               36               36   \n",
      "          2       435              435              435              435   \n",
      "49        3      1978             1978             1978             1978   \n",
      "          1        56               56               56               56   \n",
      "          2       420              420              420              420   \n",
      "\n",
      "                 Trans_feature_3  Trans_feature_4  Trans_feature_5  \\\n",
      "Time step Class                                                      \n",
      "1         3                 5733             5733             5733   \n",
      "          1                   17               17               17   \n",
      "          2                 2130             2130             2130   \n",
      "2         3                 3427             3427             3427   \n",
      "          1                   18               18               18   \n",
      "...                          ...              ...              ...   \n",
      "48        1                   36               36               36   \n",
      "          2                  435              435              435   \n",
      "49        3                 1978             1978             1978   \n",
      "          1                   56               56               56   \n",
      "          2                  420              420              420   \n",
      "\n",
      "                 Trans_feature_6  Trans_feature_7  Trans_feature_8  ...  \\\n",
      "Time step Class                                                     ...   \n",
      "1         3                 5733             5733             5733  ...   \n",
      "          1                   17               17               17  ...   \n",
      "          2                 2130             2130             2130  ...   \n",
      "2         3                 3427             3427             3427  ...   \n",
      "          1                   18               18               18  ...   \n",
      "...                          ...              ...              ...  ...   \n",
      "48        1                   36               36               36  ...   \n",
      "          2                  435              435              435  ...   \n",
      "49        3                 1978             1978             1978  ...   \n",
      "          1                   56               56               56  ...   \n",
      "          2                  420              420              420  ...   \n",
      "\n",
      "                 Graph_feature_62  Graph_feature_63  Graph_feature_64  \\\n",
      "Time step Class                                                         \n",
      "1         3                  5733              5733              5733   \n",
      "          1                    17                17                17   \n",
      "          2                  2130              2130              2130   \n",
      "2         3                  3427              3427              3427   \n",
      "          1                    18                18                18   \n",
      "...                           ...               ...               ...   \n",
      "48        1                    36                36                36   \n",
      "          2                   435               435               435   \n",
      "49        3                  1978              1978              1978   \n",
      "          1                    56                56                56   \n",
      "          2                   420               420               420   \n",
      "\n",
      "                 Graph_feature_65  Graph_feature_66  Graph_feature_67  \\\n",
      "Time step Class                                                         \n",
      "1         3                  5733              5733              5733   \n",
      "          1                    17                17                17   \n",
      "          2                  2130              2130              2130   \n",
      "2         3                  3427              3427              3427   \n",
      "          1                    18                18                18   \n",
      "...                           ...               ...               ...   \n",
      "48        1                    36                36                36   \n",
      "          2                   435               435               435   \n",
      "49        3                  1978              1978              1978   \n",
      "          1                    56                56                56   \n",
      "          2                   420               420               420   \n",
      "\n",
      "                 Graph_feature_68  Graph_feature_69  Graph_feature_70  \\\n",
      "Time step Class                                                         \n",
      "1         3                  5733              5733              5733   \n",
      "          1                    17                17                17   \n",
      "          2                  2130              2130              2130   \n",
      "2         3                  3427              3427              3427   \n",
      "          1                    18                18                18   \n",
      "...                           ...               ...               ...   \n",
      "48        1                    36                36                36   \n",
      "          2                   435               435               435   \n",
      "49        3                  1978              1978              1978   \n",
      "          1                    56                56                56   \n",
      "          2                   420               420               420   \n",
      "\n",
      "                 Graph_feature_71  \n",
      "Time step Class                    \n",
      "1         3                  5733  \n",
      "          1                    17  \n",
      "          2                  2130  \n",
      "2         3                  3427  \n",
      "          1                    18  \n",
      "...                           ...  \n",
      "48        1                    36  \n",
      "          2                   435  \n",
      "49        3                  1978  \n",
      "          1                    56  \n",
      "          2                   420  \n",
      "\n",
      "[147 rows x 166 columns]\n"
     ]
    }
   ],
   "source": [
    "# then we will group by the time step and class features \n",
    "group_class_and_feature = df_class_and_features.groupby(['Time step', 'Class']).count()\n",
    "print(group_class_and_feature)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Time step Class  Count\n",
      "0            1     3   5733\n",
      "1            1     1     17\n",
      "2            1     2   2130\n",
      "3            2     3   3427\n",
      "4            2     1     18\n",
      "..         ...   ...    ...\n",
      "142         48     1     36\n",
      "143         48     2    435\n",
      "144         49     3   1978\n",
      "145         49     1     56\n",
      "146         49     2    420\n",
      "\n",
      "[147 rows x 3 columns]\n"
     ]
    }
   ],
   "source": [
    "# we collect our counts in one column \n",
    "group_class_and_feature = group_class_and_feature['ID'].reset_index().rename(columns={'ID': 'Count'})\n",
    "print(group_class_and_feature)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Time step Class  Count\n",
      "1            1     1     17\n",
      "4            2     1     18\n",
      "7            3     1     11\n",
      "10           4     1     30\n",
      "13           5     1      8\n",
      "16           6     1      5\n",
      "19           7     1    102\n",
      "22           8     1     67\n",
      "25           9     1    248\n",
      "28          10     1     18\n",
      "31          11     1    131\n",
      "34          12     1     16\n",
      "37          13     1    291\n",
      "40          14     1     43\n",
      "43          15     1    147\n",
      "46          16     1    128\n",
      "49          17     1     99\n",
      "52          18     1     52\n",
      "55          19     1     80\n",
      "58          20     1    260\n",
      "61          21     1    100\n",
      "64          22     1    158\n",
      "67          23     1     53\n",
      "70          24     1    137\n",
      "73          25     1    118\n",
      "76          26     1     96\n",
      "79          27     1     24\n",
      "82          28     1     85\n",
      "85          29     1    329\n",
      "88          30     1     83\n",
      "91          31     1    106\n",
      "94          32     1    342\n",
      "97          33     1     23\n",
      "100         34     1     37\n",
      "103         35     1    182\n",
      "106         36     1     33\n",
      "109         37     1     40\n",
      "112         38     1    111\n",
      "115         39     1     81\n",
      "118         40     1    112\n",
      "121         41     1    116\n",
      "124         42     1    239\n",
      "127         43     1     24\n",
      "130         44     1     24\n",
      "133         45     1      5\n",
      "136         46     1      2\n",
      "139         47     1     22\n",
      "142         48     1     36\n",
      "145         49     1     56\n",
      "\n",
      "\n",
      "     Time step Class  Count\n",
      "2            1     2   2130\n",
      "5            2     2   1099\n",
      "8            3     2   1268\n",
      "11           4     2   1410\n",
      "14           5     2   1874\n",
      "17           6     2    480\n",
      "20           7     2   1101\n",
      "23           8     2   1098\n",
      "26           9     2    530\n",
      "29          10     2    954\n",
      "32          11     2    565\n",
      "35          12     2    490\n",
      "38          13     2    518\n",
      "41          14     2    374\n",
      "44          15     2    471\n",
      "47          16     2    402\n",
      "50          17     2    712\n",
      "53          18     2    337\n",
      "56          19     2    665\n",
      "59          20     2    640\n",
      "62          21     2    541\n",
      "65          22     2   1605\n",
      "68          23     2   1134\n",
      "71          24     2    989\n",
      "74          25     2    476\n",
      "77          26     2    421\n",
      "80          27     2    182\n",
      "83          28     2    199\n",
      "86          29     2    845\n",
      "89          30     2    441\n",
      "92          31     2    604\n",
      "95          32     2    981\n",
      "98          33     2    418\n",
      "101         34     2    478\n",
      "104         35     2   1159\n",
      "107         36     2   1675\n",
      "110         37     2    458\n",
      "113         38     2    645\n",
      "116         39     2   1102\n",
      "119         40     2   1099\n",
      "122         41     2   1016\n",
      "125         42     2   1915\n",
      "128         43     2   1346\n",
      "131         44     2   1567\n",
      "134         45     2   1216\n",
      "137         46     2    710\n",
      "140         47     2    824\n",
      "143         48     2    435\n",
      "146         49     2    420\n",
      "\n",
      "\n",
      "     Time step Class  Count\n",
      "0            1     3   5733\n",
      "3            2     3   3427\n",
      "6            3     3   5342\n",
      "9            4     3   4253\n",
      "12           5     3   4921\n",
      "15           6     3   3843\n",
      "18           7     3   4845\n",
      "21           8     3   3292\n",
      "24           9     3   4218\n",
      "27          10     3   5755\n",
      "30          11     3   3600\n",
      "33          12     3   1541\n",
      "36          13     3   3719\n",
      "39          14     3   1605\n",
      "42          15     3   3021\n",
      "45          16     3   2445\n",
      "48          17     3   2574\n",
      "51          18     3   1587\n",
      "54          19     3   2761\n",
      "57          20     3   3391\n",
      "60          21     3   2896\n",
      "63          22     3   4131\n",
      "66          23     3   2978\n",
      "69          24     3   3466\n",
      "72          25     3   1720\n",
      "75          26     3   2006\n",
      "78          27     3    883\n",
      "81          28     3   1369\n",
      "84          29     3   3101\n",
      "87          30     3   1959\n",
      "90          31     3   2106\n",
      "93          32     3   3202\n",
      "96          33     3   2710\n",
      "99          34     3   1971\n",
      "102         35     3   4166\n",
      "105         36     3   4685\n",
      "108         37     3   2808\n",
      "111         38     3   2135\n",
      "114         39     3   1577\n",
      "117         40     3   3270\n",
      "120         41     3   4210\n",
      "123         42     3   4986\n",
      "126         43     3   3693\n",
      "129         44     3   3384\n",
      "132         45     3   4377\n",
      "135         46     3   2807\n",
      "138         47     3   4275\n",
      "141         48     3   2483\n",
      "144         49     3   1978\n"
     ]
    }
   ],
   "source": [
    "# we then separate each class \n",
    "class1 = group_class_and_feature[group_class_and_feature['Class'] == '1']\n",
    "class2 = group_class_and_feature[group_class_and_feature['Class'] == '2']\n",
    "class3 = group_class_and_feature[group_class_and_feature['Class'] == 3 ]\n",
    "print(class1)\n",
    "print('\\n')\n",
    "print(class2)\n",
    "print('\\n')\n",
    "print(class3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1e1404d7ac8>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# and finally we proceed to the creation of our barplots \n",
    "\n",
    "bar1 = plt.bar(class3['Time step'], class3['Count'], color = 'b', label = 'Unknown')\n",
    "bar2 = plt.bar(class2['Time step'], class2['Count'], color='g',\n",
    "             bottom=class3['Count'], label = 'Licit')\n",
    "bar3 = plt.bar(class1['Time step'], class1['Count'], color='r',\n",
    "             bottom=np.array(class3['Count'])+np.array(class2['Count']), label = 'Illicit')\n",
    "plt.legend()\n",
    "# Now it's pretty clear how the total number of transactions is formed and which kind of activity \n",
    "# (illicit, licit or unknown) i have in each time step \n",
    "\n",
    "# We remember that the time steps are evenly spaced with an interval of about two weeks.\n",
    "# It is important to mention that each one of them contains a single connected component of transactions \n",
    "# that appeared on the blockchain within less than three hours between each other. \n",
    "# Finally there are no edges connecting the different time steps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             ID Class  Time step  Trans_feature_0  Trans_feature_1  \\\n",
      "51818  43523344     1         10        -0.172905        -0.158783   \n",
      "52165  43327192     1         10        -0.172424        -0.158783   \n",
      "52574  42627141     1         10        -0.170057        -0.158783   \n",
      "52898  43003119     1         10        -0.169909        -0.158783   \n",
      "53572  43438203     1         10        -0.167401        -0.167462   \n",
      "54102  43523018     1         10        -0.171491        -0.158783   \n",
      "54514  43525013     1         10        -0.172237        -0.158783   \n",
      "55431  43327208     1         10        -0.171963        -0.158783   \n",
      "55707  42627147     1         10        -0.165382        -0.158783   \n",
      "55848  43004044     1         10        -0.172369        -0.158783   \n",
      "55976  43007143     1         10        -0.170712        -0.158783   \n",
      "56139  43525228     1         10        -0.172748        -0.158783   \n",
      "57323  43007170     1         10        -0.172369        -0.158783   \n",
      "57324  43327204     1         10        -0.172828        -0.158783   \n",
      "57433  43328043     1         10        -0.172051        -0.158783   \n",
      "57713  43522396     1         10        -0.171350        -0.158783   \n",
      "57802  43527419     1         10         0.007805        -0.158783   \n",
      "57986  43003669     1         10        -0.171754        -0.158783   \n",
      "\n",
      "       Trans_feature_2  Trans_feature_3  Trans_feature_4  Trans_feature_5  \\\n",
      "51818        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "52165        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "52574        -1.201369        -0.121970        -0.043875        -0.113002   \n",
      "52898        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "53572         1.018602         0.028105        -0.043875         0.054722   \n",
      "54102        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "54514        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "55431        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "55707        -1.201369        -0.121970        -0.043875        -0.113002   \n",
      "55848        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "55976        -1.201369        -0.121970        -0.043875        -0.113002   \n",
      "56139        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "57323        -1.201369        -0.121970        -0.043875        -0.113002   \n",
      "57324        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "57433        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "57713        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "57802        -1.201369        -0.121970        -0.043875        -0.113002   \n",
      "57986        -1.201369        -0.121970        -0.063725        -0.113002   \n",
      "\n",
      "       Trans_feature_6  ...  Graph_feature_62  Graph_feature_63  \\\n",
      "51818        -0.061584  ...         -0.577099         -0.588384   \n",
      "52165        -0.061584  ...         -0.577099         -0.563154   \n",
      "52574        -0.061584  ...         -0.494897         -0.544232   \n",
      "52898        -0.061584  ...         -0.577099          0.332502   \n",
      "53572        -0.061584  ...         -0.562153         -0.600999   \n",
      "54102        -0.061584  ...         -0.577099         -0.563154   \n",
      "54514        -0.061584  ...         -0.577099          0.092819   \n",
      "55431        -0.061584  ...         -0.577099         -0.512695   \n",
      "55707        -0.061584  ...         -0.524789         -0.569462   \n",
      "55848        -0.061584  ...         -0.577099         -0.613614   \n",
      "55976        -0.061584  ...         -0.577099         -0.613614   \n",
      "56139        -0.061584  ...         -0.577099         -0.563154   \n",
      "57323        -0.061584  ...         -0.502370         -0.550539   \n",
      "57324        -0.061584  ...         -0.577099         -0.588384   \n",
      "57433        -0.061584  ...         -0.577099         -0.550539   \n",
      "57713        -0.061584  ...         -0.577099         -0.525310   \n",
      "57802        -0.061584  ...         -0.390276         -0.455928   \n",
      "57986        -0.061584  ...         -0.577099          0.332502   \n",
      "\n",
      "       Graph_feature_64  Graph_feature_65  Graph_feature_66  Graph_feature_67  \\\n",
      "51818          0.241128          0.241406          0.018279         -0.087490   \n",
      "52165          0.241128          0.241406         -0.098889         -0.106715   \n",
      "52574         -0.979074         -0.978556          0.018279         -0.087490   \n",
      "52898          0.241128          0.241406         -0.098889         -0.106715   \n",
      "53572         -0.979074         -0.978556         -0.098889         -0.087490   \n",
      "54102          0.241128          0.241406         -0.098889         -0.106715   \n",
      "54514          0.241128          0.241406          0.018279         -0.087490   \n",
      "55431          0.241128          0.241406         -0.098889         -0.106715   \n",
      "55707         -0.979074         -0.978556         -0.098889         -0.087490   \n",
      "55848          0.241128          0.241406          0.018279         -0.087490   \n",
      "55976          0.241128          0.241406         -0.098889         -0.087490   \n",
      "56139          0.241128          0.241406         -0.098889         -0.106715   \n",
      "57323         -0.979074         -0.978556         -0.098889         -0.106715   \n",
      "57324          0.241128          0.241406          0.018279         -0.087490   \n",
      "57433          0.241128          0.241406         -0.098889         -0.106715   \n",
      "57713          0.241128          0.241406         -0.098889         -0.106715   \n",
      "57802         -0.979074         -0.978556          0.018279         -0.087490   \n",
      "57986          0.241128          0.241406         -0.098889         -0.106715   \n",
      "\n",
      "       Graph_feature_68  Graph_feature_69  Graph_feature_70  Graph_feature_71  \n",
      "51818         -0.131155         -0.097524         -0.120613         -0.119792  \n",
      "52165         -0.131155         -0.183671         -0.120613         -0.119792  \n",
      "52574         -0.131155         -0.097524         -0.120613         -0.119792  \n",
      "52898         -0.131155         -0.183671         -0.120613         -0.119792  \n",
      "53572         -0.084674         -0.140597         -1.760926         -1.760984  \n",
      "54102         -0.131155         -0.183671         -0.120613         -0.119792  \n",
      "54514         -0.131155         -0.097524         -0.120613         -0.119792  \n",
      "55431         -0.131155         -0.183671         -0.120613         -0.119792  \n",
      "55707         -0.084674         -0.140597          1.519700          1.521399  \n",
      "55848         -0.131155         -0.097524         -0.120613         -0.119792  \n",
      "55976         -0.084674         -0.140597         -1.760926         -1.760984  \n",
      "56139         -0.131155         -0.183671         -0.120613         -0.119792  \n",
      "57323         -0.131155         -0.183671         -0.120613         -0.119792  \n",
      "57324         -0.131155         -0.097524         -0.120613         -0.119792  \n",
      "57433         -0.131155         -0.183671         -0.120613         -0.119792  \n",
      "57713         -0.131155         -0.183671         -0.120613         -0.119792  \n",
      "57802         -0.131155         -0.097524         -0.120613         -0.119792  \n",
      "57986         -0.131155         -0.183671         -0.120613         -0.119792  \n",
      "\n",
      "[18 rows x 168 columns]\n"
     ]
    }
   ],
   "source": [
    "# so we should explore a little bit better the form of this connected components in some time steps and to \n",
    "# visualize the graphs that are created\n",
    "\n",
    "# lets start with timestamp 10 and illicit transactions (class 1 )\n",
    "# we will create first the directed graph that is formed in this case\n",
    "# first we find the illicit in timestep 10 \n",
    "illicit = df_class_and_features.loc[(df_class_and_features['Time step'] == 10) & (df_class_and_features['Class'] == '1')]\n",
    "print(illicit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "51818    43523344\n",
      "52165    43327192\n",
      "52574    42627141\n",
      "52898    43003119\n",
      "53572    43438203\n",
      "54102    43523018\n",
      "54514    43525013\n",
      "55431    43327208\n",
      "55707    42627147\n",
      "55848    43004044\n",
      "55976    43007143\n",
      "56139    43525228\n",
      "57323    43007170\n",
      "57324    43327204\n",
      "57433    43328043\n",
      "57713    43522396\n",
      "57802    43527419\n",
      "57986    43003669\n",
      "Name: ID, dtype: int64\n",
      "\n",
      "\n",
      " There are 18 illicit IDs in timestamp 10\n"
     ]
    }
   ],
   "source": [
    "# we keep their ID\n",
    "illicit_ID = illicit['ID']\n",
    "print(illicit_ID)\n",
    "print('\\n')\n",
    "print(f' There are {len(illicit_ID)} illicit IDs in timestamp 10')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            ID1       ID2\n",
      "64058  43327208  43521039\n",
      "64507  43007170  43327204\n",
      "66822  42627141  43007143\n",
      "67534  43327192  43517733\n",
      "68430  43527419  68629305\n",
      "68559  42627141  37669635\n",
      "69072  42627147  42627141\n",
      "We have 7 illicit transactions in timestamp 10\n"
     ]
    }
   ],
   "source": [
    "# now we will check in the edgelist dataframe to find which illicit IDs are included there\n",
    "# so we will track the IDs that they traded with (ID2) \n",
    "edges = df_edgelist.loc[df_edgelist['ID1'].isin(illicit_ID)]\n",
    "print(edges)\n",
    "print(f'We have {len(edges)} illicit transactions in timestamp 10')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# we will first visualize them with a directed graph \n",
    "G = nx.from_pandas_edgelist(edges, source = 'ID1', target = 'ID2', \n",
    "                                 create_using = nx.DiGraph())\n",
    "\n",
    "#Position nodes using Fruchterman-Reingold force-directed algorithm.\n",
    "#The algorithm simulates a force-directed representation of the network treating edges \n",
    "#as springs holding nodes close, while treating nodes as repelling objects, \n",
    "#sometimes called an anti-gravity force. Simulation continues until the positions are close to an equilibrium\n",
    "pos = nx.spring_layout(G)\n",
    "# print(pos)\n",
    "#without labels \n",
    "# nx.draw(G, pos = pos)\n",
    "#with labels\n",
    "nx.draw(G, pos = pos, with_labels = True)\n",
    "# nx.draw(graph, cmap = plt.get_cmap('terrain'), with_labels=True, pos=pos)\n",
    "\n",
    "# As we can see from the graph not even one user had illicit transactions with more than one receiver "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# now we will visualize the corresponding undirected graph\n",
    "G = nx.from_pandas_edgelist(edges, source = 'ID1', target = 'ID2', \n",
    "                                 create_using = nx.Graph())\n",
    "pos = nx.spring_layout(G)\n",
    "# print(pos)\n",
    "nx.draw(G, pos = pos, with_labels = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             ID Class  Time step  Trans_feature_0  Trans_feature_1  \\\n",
      "51371  43388675     2         10        -0.125817         4.448752   \n",
      "51379  13690412     2         10        -0.014147        -0.132897   \n",
      "51384  43473398     2         10        -0.030952         0.048298   \n",
      "51386  68806101     2         10        -0.075933         0.103500   \n",
      "51392  68706499     2         10        -0.069634         2.390893   \n",
      "...         ...   ...        ...              ...              ...   \n",
      "58070  43388427     2         10        -0.171928        -0.158783   \n",
      "58075  43028760     2         10        -0.138367        -0.181302   \n",
      "58087  43352022     2         10        -0.172207        -0.181561   \n",
      "58091  43445815     2         10        -0.172198        -0.158783   \n",
      "58092  43410473     2         10        -0.147256        -0.169302   \n",
      "\n",
      "       Trans_feature_2  Trans_feature_3  Trans_feature_4  Trans_feature_5  \\\n",
      "51371        -0.646376        26.816496        -0.063725        25.548843   \n",
      "51379        -1.201369        -0.046932         0.114926        -0.029140   \n",
      "51384        -1.201369         1.678931        -0.063725         1.815829   \n",
      "51386         1.573595         1.003593        -0.043875         1.061069   \n",
      "51392         0.463609        14.885532        -0.063725        12.969507   \n",
      "...                ...              ...              ...              ...   \n",
      "58070        -1.201369         0.028105        -0.063725         0.054722   \n",
      "58075        -0.091383        -0.121970        -0.043875        -0.113002   \n",
      "58087        -0.091383        -0.121970        -0.043875        -0.113002   \n",
      "58091        -1.201369         0.028105        -0.063725         0.054722   \n",
      "58092         1.018602        -0.121970        -0.043875        -0.113002   \n",
      "\n",
      "       Trans_feature_6  ...  Graph_feature_62  Graph_feature_63  \\\n",
      "51371         0.242712  ...         -0.577099         -0.613614   \n",
      "51379        -0.061584  ...          0.816851          0.467060   \n",
      "51384        -0.061584  ...         -0.577099         -0.613614   \n",
      "51386         0.242712  ...         -0.397749         -0.462235   \n",
      "51392         2.068489  ...         -0.577099         -0.613614   \n",
      "...                ...  ...               ...               ...   \n",
      "58070        -0.061584  ...         -0.577099         -0.600999   \n",
      "58075        -0.061584  ...         -0.562153         -0.600999   \n",
      "58087        -0.061584  ...         -0.293127         -0.235168   \n",
      "58091        -0.061584  ...         -0.577099         -0.588384   \n",
      "58092        -0.061584  ...         -0.547208         -0.588384   \n",
      "\n",
      "       Graph_feature_64  Graph_feature_65  Graph_feature_66  Graph_feature_67  \\\n",
      "51371          0.241128          0.241406          1.189961          0.104754   \n",
      "51379          0.527415          0.169944         -0.098889          9.486290   \n",
      "51384          0.241128          0.241406          0.018279         -0.087490   \n",
      "51386         -0.979074         -0.978556          0.018279          0.104754   \n",
      "51392          0.241128          0.241406          0.838456          0.047081   \n",
      "...                 ...               ...               ...               ...   \n",
      "58070          0.241128          0.241406          0.721288          0.027857   \n",
      "58075         -0.979074         -0.978556          0.018279         -0.087490   \n",
      "58087          1.461330          1.461369          0.018279         -0.087490   \n",
      "58091          0.241128          0.241406          0.135448         -0.068266   \n",
      "58092         -0.979074         -0.978556          0.018279         -0.029817   \n",
      "\n",
      "       Graph_feature_68  Graph_feature_69  Graph_feature_70  Graph_feature_71  \n",
      "51371         -0.131155          0.763946         -0.120613         -0.119792  \n",
      "51379         10.775522          4.688419          0.103651         -0.105442  \n",
      "51384         -0.131155         -0.097524         -0.120613         -0.119792  \n",
      "51386          0.333655          0.333211          1.519700          1.521399  \n",
      "51392         -0.131155          0.505505         -0.120613         -0.119792  \n",
      "...                 ...               ...               ...               ...  \n",
      "58070         -0.131155          0.419358         -0.120613         -0.119792  \n",
      "58075         -0.131155         -0.097524         -0.120613         -0.119792  \n",
      "58087         -0.131155         -0.097524         -0.120613         -0.119792  \n",
      "58091         -0.131155         -0.011377         -0.120613         -0.119792  \n",
      "58092          0.008288          0.031697         -1.760926         -1.760984  \n",
      "\n",
      "[954 rows x 168 columns]\n"
     ]
    }
   ],
   "source": [
    "# ok now we figured out what happens visualy when an illicit transaction occurs \n",
    "# but what about the licit transactions in timestamp 10?\n",
    "# lets see, we will follow the same steps as in the illicit case  \n",
    "\n",
    "# first we find the licit in timestep 10 \n",
    "licit = df_class_and_features.loc[(df_class_and_features['Time step'] == 10) & (df_class_and_features['Class'] == '2')]\n",
    "print(licit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "51371    43388675\n",
      "51379    13690412\n",
      "51384    43473398\n",
      "51386    68806101\n",
      "51392    68706499\n",
      "           ...   \n",
      "58070    43388427\n",
      "58075    43028760\n",
      "58087    43352022\n",
      "58091    43445815\n",
      "58092    43410473\n",
      "Name: ID, Length: 954, dtype: int64\n",
      "\n",
      "\n",
      " There are 954 licit IDs in timestamp 10\n"
     ]
    }
   ],
   "source": [
    "# we keep their ID\n",
    "licit_ID = licit['ID']\n",
    "print(licit_ID)\n",
    "print('\\n')\n",
    "print(f' There are {len(licit_ID)} licit IDs in timestamp 10')\n",
    "# we can see again that as expected the licit IDs are more than the illicit ones "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            ID1       ID2\n",
      "63155  43019093  43023869\n",
      "63163  43470002  35808292\n",
      "63166  49172826  68532148\n",
      "63182  40420842  42638123\n",
      "63184  13091685  68589263\n",
      "...         ...       ...\n",
      "71683   4907336  68452703\n",
      "71694  43468144  43468337\n",
      "71707  43015861  43015850\n",
      "71709  68802682  36892268\n",
      "71723  68443063  42960585\n",
      "\n",
      "[1065 rows x 2 columns]\n",
      "We have 1065 licit transactions in timestamp 10\n"
     ]
    }
   ],
   "source": [
    "# now let's check in the edgelist dataframe to find which licit IDs are included there\n",
    "# so we will track the IDs that they traded with (ID2) \n",
    "edges = df_edgelist.loc[df_edgelist['ID1'].isin(licit_ID)]\n",
    "print(edges)\n",
    "print(f'We have {len(edges)} licit transactions in timestamp 10')\n",
    "# Some users made more than one trasactions (1065 transactions >954 users)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# lets start the visualization again with directed graph \n",
    "G = nx.from_pandas_edgelist(edges, source = 'ID1', target = 'ID2', \n",
    "                                 create_using = nx.DiGraph())\n",
    "pos = nx.spring_layout(G)\n",
    "# print(pos)\n",
    "# we remove the labels now (IDs) because the plot is more dense than before and the visualization \n",
    "#will be worse with them \n",
    "nx.draw(G, pos = pos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# now we will visualize the corresponding undirected graph\n",
    "G = nx.from_pandas_edgelist(edges, source = 'ID1', target = 'ID2', \n",
    "                                 create_using = nx.DiGraph())\n",
    "pos = nx.spring_layout(G)\n",
    "# print(pos)\n",
    "nx.draw(G, pos = pos)\n",
    "# we can see the expected growth in the graph in this licit transactions case study \n",
    "#in comparsion with the illicit transactions that were less \n",
    "\n",
    "# we can also explore some further metrics in this diagrams with the help of gephi \n",
    "#like betweenness, closseness centrality, community detection \n",
    "#etc "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we can see that networkx visualization is not the best\n",
    "# we can not even distinguish the undirected from the directed graph they see the same \n",
    "# maybe the use of gephi would help in this case "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Regarding the features\n",
    "# maybe some features will help us to separate effectively illicit from licit transactions than others \n",
    "# so the correlation of the features to find the right class classification may prove to be important as well \n",
    "# lets explore for example a random feature (Trans_feature_10)\n",
    "\n",
    "# we remind that  \n",
    "# 1 means the transaction is illicit / illegal \n",
    "# 2 means the transaction is licit / legal \n",
    "\n",
    "grouped = df_class_and_features.groupby(['Time step', 'Class'])['Trans_feature_10'].mean().reset_index()\n",
    "# print(grouped)\n",
    "sns.lineplot(x='Time step', y='Trans_feature_10',hue = 'Class', data=grouped, legend = 'full');\n",
    "plt.title('Average trans_feat_10 in each time step by type');\n",
    "\n",
    "# Trans_feature_10 helps us to distinguish illicit from licit transactions "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# lets now fit our data in some models to see how well we can predict the classes \n",
    "# lets start with Logistic regression \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3         2\n",
      "9         2\n",
      "10        2\n",
      "11        2\n",
      "16        2\n",
      "         ..\n",
      "203752    2\n",
      "203754    2\n",
      "203759    1\n",
      "203763    2\n",
      "203766    1\n",
      "Name: Class, Length: 46564, dtype: object\n"
     ]
    }
   ],
   "source": [
    "# first we split our data into features and target \n",
    "\n",
    "#we remind our merged dataset\n",
    "# print(df_class_and_features.head())\n",
    "\n",
    "# we select classes 1 and 2 and we discard the unknown class \n",
    "data = df_class_and_features[(df_class_and_features['Class']=='1') | (df_class_and_features['Class']=='2')]\n",
    "\n",
    "# features \n",
    "features =  data.drop(['ID','Class','Time step'], axis =1 )\n",
    "#check \n",
    "# print(features)\n",
    "\n",
    "# target \n",
    "target = data['Class']\n",
    "#check \n",
    "print(target)\n",
    "\n",
    "# we remind that  \n",
    "# 1 means the transaction is illicit / illegal \n",
    "# 2 means the transaction is licit / legal \n",
    "\n",
    "# we rename the licit transactions from 0 to 2 \n",
    "target = target.apply(lambda x: 0 if x == '2' else 1 )\n",
    "#check \n",
    "# print(target)\n",
    "\n",
    "# now we will have \n",
    "# 0 means the transaction is licit / legal \n",
    "# 1 means the transaction is illicit / illegal \n",
    "\n",
    "# so we split our data \n",
    "x_train, x_test, y_train, y_test = train_test_split(features,target,test_size=0.2,random_state=0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression\n",
      "Precision:0.853 \n",
      "Recall:0.770 \n",
      "F1 Score:0.809\n",
      "Macro-Average F1 Score: 0.8955563996816439\n",
      "Micro-Average F1 Score: 0.9667131966068936\n"
     ]
    }
   ],
   "source": [
    "# time to fit our log regr model\n",
    "reg = LogisticRegression(max_iter=500)\n",
    "reg.fit(x_train,y_train)\n",
    "\n",
    "# predictions \n",
    "pred = reg.predict(x_test)\n",
    "\n",
    "# metrics for log regression \n",
    "prec,rec,f1,num = precision_recall_fscore_support(y_test,pred, average=None)\n",
    "print(\"Logistic Regression\")\n",
    "print(\"Precision:%.3f \\nRecall:%.3f \\nF1 Score:%.3f\"%(prec[1],rec[1],f1[1]))\n",
    "\n",
    "# lets try macro-f1 also \n",
    "macro_f1 = f1_score(y_test,pred,average='macro')\n",
    "print(\"Macro-Average F1 Score:\",macro_f1)\n",
    "\n",
    "# lets try micro-f1 also \n",
    "micro_f1 = f1_score(y_test,pred,average='micro')\n",
    "print(\"Micro-Average F1 Score:\",micro_f1)\n",
    "\n",
    "# looks like our logistic regression does a descent job "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Classifier\n",
      "Precision:0.999 \n",
      "Recall:0.885 \n",
      "F1 Score:0.939\n",
      "Macro-Average F1 Score: 0.9664023015258125\n",
      "Micro-Average F1 Score: 0.9893696982712338\n"
     ]
    }
   ],
   "source": [
    "# now lets try a random forest classifier \n",
    "\n",
    "rf = RandomForestClassifier(n_estimators=50, max_depth=100,random_state=0)\n",
    "rf.fit(x_train,y_train)\n",
    "pred = rf.predict(x_test)\n",
    "prec,rec,f1,num = precision_recall_fscore_support(y_test,pred,average=None)\n",
    "print(\"Random Forest Classifier\")\n",
    "print(\"Precision:%.3f \\nRecall:%.3f \\nF1 Score:%.3f\"%(prec[1],rec[1],f1[1]))\n",
    "macro_f1 = f1_score(y_test,pred,average='macro')\n",
    "print(\"Macro-Average F1 Score:\",macro_f1)\n",
    "micro_f1 = f1_score(y_test,pred,average='micro')\n",
    "print(\"Micro-Average F1 Score:\",micro_f1)\n",
    "\n",
    "#The random forest classifier improves further our scores "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MLP Classifier\n",
      "Precision:0.890 \n",
      "Recall:0.889 \n",
      "F1 Score:0.889\n",
      "Macro-Average F1 Score: 0.9391184087331279\n",
      "Micro-Average F1 Score: 0.979705787608719\n"
     ]
    }
   ],
   "source": [
    "# we will finally try an MLP classifier \n",
    "\n",
    "MLP = MLPClassifier(random_state=0, max_iter=300, solver = 'adam', activation = 'relu')\n",
    "MLP.fit(x_train,y_train)\n",
    "pred = MLP.predict(x_test)\n",
    "prec,rec,f1,num = precision_recall_fscore_support(y_test,pred,average=None)\n",
    "print(\"MLP Classifier\")\n",
    "print(\"Precision:%.3f \\nRecall:%.3f \\nF1 Score:%.3f\"%(prec[1],rec[1],f1[1]))\n",
    "macro_f1 = f1_score(y_test,pred,average='macro')\n",
    "print(\"Macro-Average F1 Score:\",macro_f1)\n",
    "micro_f1 = f1_score(y_test,pred,average='micro')\n",
    "print(\"Micro-Average F1 Score:\",micro_f1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# now we will try to explore what happens when we try to predict the unknown class \n",
    "# The unknown class has much more observations that the licit and illicit classes \n",
    "# lets use our best model in terms of scores the random forest classifier to see what is happening \n",
    "\n",
    "# we select the class 3 (unknown class) in this case from our initial dataframes \n",
    "Class3_ID = df_class_and_features.loc[(df_class_and_features['Class'] == 3), 'ID']\n",
    "#CHECK \n",
    "# print(Class3_ID)\n",
    "#check again how many unknown IDs i have \n",
    "# print(len(Class3_ID))\n",
    "\n",
    "# picking the according values from all my dataframes \n",
    "df_edgelist_Class3_ID = df_edgelist.loc[df_edgelist['ID1'].isin(Class3_ID)]\n",
    "df_classes_Class3_ID = df_classes.loc[df_classes['ID'].isin(Class3_ID)]\n",
    "df_features_Class3_ID = df_features.loc[df_features['ID'].isin(Class3_ID)]\n",
    "\n",
    "# Check \n",
    "# print(df_edgelist_Class3_ID.head())\n",
    "# print(df_classes_Class3_ID.head())\n",
    "# print(df_features_Class3_ID.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 1]\n",
      "               ID  Class\n",
      "0       230425980      0\n",
      "1         5530458      0\n",
      "2       232022460      0\n",
      "4       230460314      0\n",
      "5       230459870      0\n",
      "...           ...    ...\n",
      "203762  157581340      0\n",
      "203764  173077460      0\n",
      "203765  158577750      0\n",
      "203767  158654197      0\n",
      "203768  157597225      0\n",
      "\n",
      "[157205 rows x 2 columns]\n"
     ]
    }
   ],
   "source": [
    "# we create our predictions \n",
    "x_Class3_ID = df_features_Class3_ID.drop(columns=['ID', 'Time step'])\n",
    "# we use our trained rf model \n",
    "pred_Class3_ID = rf.predict(x_Class3_ID.values)\n",
    "print(np.unique(pred_Class3_ID))\n",
    "\n",
    "# we replace the 'unknown' class with the predicted one \n",
    "df_classes_Class3_ID = df_classes_Class3_ID.copy()\n",
    "df_classes_Class3_ID.loc[:, 'Class'] = pred_Class3_ID\n",
    "# check \n",
    "print(df_classes_Class3_ID)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          ID  Class  Time step  Trans_feature_0  Trans_feature_1  \\\n",
      "0  230425980      0          1        -0.171469        -0.184668   \n",
      "1    5530458      0          1        -0.171484        -0.184668   \n",
      "2  232022460      0          1        -0.172107        -0.184668   \n",
      "3  230460314      0          1         1.011523        -0.081127   \n",
      "4  230459870      0          1         0.961040        -0.081127   \n",
      "\n",
      "   Trans_feature_2  Trans_feature_3  Trans_feature_4  Trans_feature_5  \\\n",
      "0        -1.201369        -0.121970        -0.043875        -0.113002   \n",
      "1        -1.201369        -0.121970        -0.043875        -0.113002   \n",
      "2        -1.201369        -0.121970        -0.043875        -0.113002   \n",
      "3        -1.201369         1.153668         0.333276         1.312656   \n",
      "4        -1.201369         1.303743         0.333276         1.480381   \n",
      "\n",
      "   Trans_feature_6  ...  Graph_feature_62  Graph_feature_63  Graph_feature_64  \\\n",
      "0        -0.061584  ...         -0.562153         -0.600999          1.461330   \n",
      "1        -0.061584  ...          0.947382          0.673103         -0.979074   \n",
      "2        -0.061584  ...          0.670883          0.439728         -0.979074   \n",
      "3        -0.061584  ...         -0.511871         -0.400422          0.517257   \n",
      "4        -0.061584  ...         -0.504702         -0.422589         -0.226790   \n",
      "\n",
      "   Graph_feature_65  Graph_feature_66  Graph_feature_67  Graph_feature_68  \\\n",
      "0          1.461369          0.018279         -0.087490         -0.131155   \n",
      "1         -0.978556          0.018279         -0.087490         -0.131155   \n",
      "2         -0.978556         -0.098889         -0.106715         -0.131155   \n",
      "3          0.579382          0.018279          0.277775          0.326394   \n",
      "4         -0.117629          0.018279          0.277775          0.413931   \n",
      "\n",
      "   Graph_feature_69  Graph_feature_70  Graph_feature_71  \n",
      "0         -0.097524         -0.120613         -0.119792  \n",
      "1         -0.097524         -0.120613         -0.119792  \n",
      "2         -0.183671         -0.120613         -0.119792  \n",
      "3          1.293750          0.178136          0.179117  \n",
      "4          1.149556         -0.696053         -0.695540  \n",
      "\n",
      "[5 rows x 168 columns]\n"
     ]
    }
   ],
   "source": [
    "# we merge class and features \n",
    "df_class_feature_Class3_ID = pd.merge(df_classes_Class3_ID, df_features_Class3_ID)\n",
    "print(df_class_feature_Class3_ID.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb4AAAEuCAYAAADx63eqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOydeXhTVfrHPzdJ23ShlKWU0rIUyioUKFsRZFURUJA1oAgVZ2AUEZxxQ3CZcVxnEB2XYfSnlAGUFBzREZF9lzKACAIt+9ICLS1QSqFbkvv7o6Z2b5J7k9zb3s/z+KhtenOS3JzvOe953+8riKIooqGhoaGhUUfQeXsAGhoaGhoankQTPg0NDQ2NOoUmfBoaGhoadQpN+DQ0NDQ06hSa8GloaGho1Ck04dPQ0NDQqFNowqehoaGhUafQhE9DQ0NDo06hCZ+GhoaGRp1CEz4NDQ0NjTqFJnwaGhoaGnUKTfg0NDQ0NOoUmvBpaGhoaNQpNOHT0NDQ0KhTaMKnoaGhoVGn0IRPQ0NDQ6NOoQmfhoaGhkadQhM+DQ0NDY06hSZ8GhoaGhp1Ck34NDQ0NDTqFJrwaWhoaGjUKTTh09DQ0NCoUxi8PQANjdpAVm4Bqw+kkZKeQ06+hWCjgQ5Ng5nQI5JGQX7eHp6GhkYpBFEURW8PQkNDrRxKzeajbafYfiITgAKLreR3RoMOERjUPpQnBkbTtXlItdfSxFNDwzNowqeh4SLLk87x+vcp5FusVPctEgQwGvTMH9GBKXGtKvxeTvEETUA1NGpCEz4NDRcoFr1k8opsNT/4V/x9dMwf0bGM+MklniC/gGpo1FY04dPQcJJDqdlM+jSJvCKr03/r76PHPCOOmMgQ2cQT5BVQDY3ajpbcouE11BqS+2jbKfItzoseQL7FysfbTvH4wGhe/z7FKdEDyCuy8fr3KcREhhATWbxrc0ZARRHyiqy8/n0ygCzip9bPUaPuou34NDyOmkNyJzJuMvz9HVglfGv8DDr6tm7E9pOZ1e7OqkIQYFinMBZP6SnL7rNZiL9LwqXmz1GjbqMJn4ZHUWtIzj7Jb065gtUm7SvjZ9BhsdokiadegHvvaMq+c9fIyi106RoCEFrPjxt5RYBzwqXWz1FDAzTh0/Agcp5peRJHJ3ln0AkgUT89QmXC5cnPUQujargDTfg0PIKUkJxBJ7BkWi/uahfqhpFVjyuTfG3ELlxdIkJkSeypCS2MquFONOHTqBS5V9ozlu1nY3KGyzsmnQD3dArz6ES3/fgVHvv3fixq2Jp5AH8fPTGR9dl79ppLf1/6bLI6tDCqhrvRhE+jDO5YaWflFtDv7S1lruUq/j7un+js78Gm5Ay3hCPVEuYsjwBIHbafQcePzw+pcvGk1nC4hrrQhE+jBKkr7ap2iWczc1n1U5psk707J7qS96DIKnmSrwq9TkAvQKGU7BaVYjToePqedswc0KbC7+Sqj9TQqAmtjk8DkFYL1iUipMpdokG4hEXm+b2yWjY58MR5niDAgLaN+fH0VaTvn9RHvsVGyuWblf5OjvrImsKoGhqgCZ8GxSttV4upX/3vMfQ6gUKrrdJdotyiZ0fuic7V98BZjAY9T9/dDh/9KUlnnmomJ7+ows+ycgvYfsK1ukYoXoxtPZ7J1dwCLdtTo0Y04dOQtNK22ESvJH/IPdFJeQ8cxc+gY/6IDsREhjBrUDQ7T2a5FNZTO8FGnwo/W30gTfJ1BWD1T2mVhlErQyuVqLtowlfHkbrS9ibOTnRV4an3oE1oYMm5ZNfmIcwf0aHOlUoYDTo6hNer8POU9BzJyU/VhVFLU30CVzqLNp3QSiVqOZrw1VIcXc3KsdL2Fo5OdDXhqffgdOatMjtUuwjKXRyvZIqsNn5OzWb60n1l7smcfIss168sjFqamhK48n8VwQ3HMthxIksrlailaMJXy3B2NSvHStub1DTROYKn3oPKdqhT4loRExnCx9tOsfV4JgK/Tb61ERFYdyS95P/t96SfQZDl+pWFUe0s3n6Kv2844VBo3h1m3hrKQRO+WoQrq1m5VtreorqJzlE89R5UtUONiQxh8ZSeXM0toJdpNqGtY4jp2YdfLt7gys0CWZ7bXoNZ399Htmu6QnnNsd+TBTJ8BFWFUQ+lZvPGumSXCu/dlUGs4V004asluFqO0Das4kShFqqa6Jwl2Oi5r0F1O9SDSTs5/8NnXPH357vXMhiwcIfk52sU6MuAtqF0CK/H+NhI0q7nuVwrp3REYHxsZJmf2ReDUl6vVipR+9B5ewAa0pFSjnDsUg4+ennCTJ6msonOFTo0DcbP4JmvQlU71KKiIn73u99hs9mw2Wy8tmKTLM/3QNdwFpm6MXNAGxoF+ZUk1fj71L6v/uD2oWXOr39bDEoT+dIZxBq1A23Hp2AcTVCRkopvtYkIKtQ9Qag40bnK+B6RLNp0QoZRVU91O9QPP/yQjIwMAPLy8ti47ygF4V0lP+eN2xVjiLUxqcagE3hiUHTJ/8tdlylXBrGGMtCET4E4k6AS0cBfWuEvxV9qtWE06MtMdFJoHOTHwHahbi8or26HeujQIQICAigoKN5VXL+VT4AMz1lVaNWeVPPB5hNsPZGJTVSnf6idzs2Cy5zBfbTtFPkyhnPlyiDWUAaa8CkMZxNU7mrbSPJzGnQCNhHVdCEw6ATmDI2WNdnA3QXlNe1QExISAAgLC2P//v28tf0ya49mSn7e6pJ/7givx+GPZ3PpQibh0z9EzVu/0u9rVm4BW1KuyG4IV9UiQiuEVx+a8CkIVxJUNidfkdTJG4rNkrtF1ud4xk11FFOLIos2neRgarZsRcb2s6+/fp9MvhveA0d2qKIocv36dZo0aUKX5oVsOn5VUpmFr16oMrQqiiJTp07lp59+IrDnaAw6sKrgo6+K0gL/4teH3bKIK7+I0Arh1UvtO+FWKa6eSchl8N8oyI/5Izri76OX54JuxCIWTzIbjmUw6dMkliedk+W6U+JasWBER3Qyx36Lu0l0qHGHevv2bQwGA35+fozvIT1pRxCESkOroigya9Ys1qxZg8ViwRgeTYFMpqq+XkiUKn12eig1m03JV9z6HFC8SJ30aRIbkzMosNgqLFDyf/2Z3PeohjxowqcQPOEVWR1ZuQVMiWuFeUYc4fWNXhuHM5Quy5BT/N4ZFyOL+AmCvX+gYy2Url+/TkhIsTjazx1dTTyqLrR64cIFPvnkE0o6kvn4u/Yk5WgabKRVo0BZruUM9rPTQ6nZ/H7ZfrecVZY+ny2dLVpTdNgd96iGdDThUwBK8Ms8eimHw2nZxESG0CWivtuex1BOUYwGHb56HbqcdIoupbiUaGMvMj6cli3LGMf3aM5fRt2B0UU3EYOu2JB6WKcwzDPiHHb9yM7OpkGDBiX/P2tQNEaDaztwAbinY1ilv2vZsiVZWVkMHz6c6OhodP7yfN6xLUI4f+22LNdyFLvArztyGdMne9xSnF96ESGldEjOe1RDGprwKQAl+GVabSIfbzvF8qRz7DghPamiPOH1jQQk/R+T7whkTLcIhnZowphuETx9Tzv2vDCEQXm7Kbx5DZvNtYMme5GxXEyJa8WCkZ2cD/3+unp5+u62LJ7S06kEnOvXr5cRPik1d6IIC745WuUuIyQkhDNnzvDwK//EP6Kt09cvj9Gg47YXiuKNBj1RjQJ5/ftkt1m9lT6flaNnoIb30YRPASjBL1MENqdc4a9r3TOBdAwPJnnbGt6adjc+B77gL/e3p33TeiRfzuHZrw5zuulgAqJ7I+hcuyXdUWQ8Ja4V8Xe2dG4XKghYbPD+5pNOh7bKC599DK6cvYpUH2K7cOECF/MMrDxRJEtSi0jxWaYn72N/Hx3T+rYkYc95tyVllT6flbNnoIZ30bI6FYBS/DKtNpEiN8Vb6xt9CAoKojAonH+fNWJ+awu+vj78Nl81BJ2055a7yHh50jk+333OpbR4Vzwes7OzS874SjMlrhX+Pnqe/eqw0+dXeUU2/ro2mZNXcrmRV1SSbn/97FEi7ptBtgxCZQ8FypUg48jzGQ165o/owI6TWW47Gy8+n/2tO4O7ewZqZRGeQxM+BeBJr8jqcFcZnz0jrkGvURA7FkHvi6DTUX6RLki0kJGzyNh+liNlB+Osx2NlOz47649luFyXlm+xsSzpfNnP1xYE/iHIUexmDwV+vvus9ItR/H0osNgqdKqwG20Pbh/KE4OiaRbiz2trk91yNt63dUMeH9iGY5dvMtd8kJx8CycybrqlZ6BWFuF5lDHj1nGKvSLTJX2pdIJynTdEAFFE13M86H3d+lxytCkCeZw/nO0SX5XwyZH8VOHe0EnvagHFyUr2UKAc97FBgFmDoxkfG8nqn9JIuXyT7zdtpWvHdgzt2Y7xsb/tfhZvPy3LayhPjxYh1DP68PtlBwBkD9+Wvke1/oDeQRM+L1A+pOFn0GFRc/VwNQgCdG8RwnubT7ld9ECeNkUlQiPDeJwJv16/fp2oqKgKP1dC8lNV3NEsuGQilsPzVK/XlYjbzAFtOHv2LO9N+iO5zZrxxfnzGAy/TVnuOBsXKM5wLrDa3JZlbb9HXe2oAlp/QKlowudBqgtpSKkbEwSIDg3i/LXbXk+SKY/RoEcU8UiNolxtiuQUGmfCr1Wd8Skh+akqGpfayUr1PK2s9vCTTz5Bp9ORlZXF+++/z5/+9KeS37njbFwQ3NsI2O/Xe1RqWYTWH1AaWlanh6jJ6UFKmNJo0LNgZEeJI5Qffx8dc4ZG83NqtkdqFG2iKEubIrmFxtHwa1WhTqUkP1VG+R22lNrD8rZuVquVTz75BJvNRmFhIS+//DJpab8tStxxNu7u4wKLrfge1coivIsmfDKRlVvA4u2nmWs+yPSl+5hrPsji7ae5mlvglNODs/j76LDsT+TR4X0Js11VRKeF0o4lnuz9YLGJvLjmFw6lSisSlltoqgu/lr5vTjQdgvmCseS++e3vlRmYqWqH3aqx830lKrN127ZtG9nZ2RiNRnx8fMjLy+OLL74o+b2cfRR1gmfuVF+9wBzzz2yS0AlEK4uQjjK/USqipoyshRuPY7GKsjvFl07pPnirAe+uO8OlhNdoMvlNBB/vpT776nUM6VCcdRcTGcJc80GPhelsojxJAHIKTVXiUOl906gtP16y8NOVE2Uy+eRIGnEH5dsslU7UcJTS93H5z6t79+6sXLmSY8eOsXv3blatWkVwcHDJ7+Xqo+inF7CKYPFAWCKvyMauU1mSr6P1B5SGtuOTgCNGtUUSRa/82Z/RoKtghzV//nyMRiP5l05wbfP/IRZ5ZyXYPiyIPS8MKeNY4ukwnRzeiHLuJCrrweeswbEsNQcyU/48ztWoRsem9aq0dWvYsCETJkxgyJAh3Lp1i/r165cpeZHqZwoQVs+PSb1boJfbmdzNaP0BpaHt+MrhaBGpMxlZUhAEgRF3hFFgsRFs9KFDeL0yKd3w2wSxYsUK8o9sxCcsDH2vCRRaRY/5f/r76Pnb+K4V0va9FaaTkgQgZ0f2+v4+pF3Pq0QgHM/ke3/zSVqHBpKSflMxLfNKn8dJ6XZ+Nqtmb8+mTZuWdKcvj5Q+ikaDjk+n9uTz3WcVt5t2BLlKd+oimvD9ijNFpIDLX3Rn8dEJdG0eUmNI47nnnmPFihUkJiZy8uRJ/rH8dfr9/s/8fKWoQiGw3PjqhSrb7ngzTOdsAbmdkuxECUXjdq7cLGDSp0nMH9GBLhEhLmfynblyC1+9Zy3BqqL8eZwciRrVfUZhYWGkp6dX+ju7n6mzi9Di19CRmMgQRScPVYccpTt1FU34cL6ItFXjAI+1EHI0pNG5c2fS09MJDQ0FoG3btsyYMZm/f7CY/GbdWPfLZQ5fvCF/1pooUlBYxA8//MD9HSZVSMeXc/fkwtCcKiAvzaxB0ZISEEpjD722bBTo8n1TaLPRsWk9zmbd8lqz4MrO4+T0r6zqM6pXrx42m43c3FyCgoIq/N4+luq+w9W9BqUmD1WHXKU7dZU6f8bnSm+t5MueDTk5GtKwix7AmDFj2LBhAwuencvOLZtIybjpnlRtQUDQ+7AjN5TW905j1qxZ5OTklPxajnMYScOjOAnAWSIa+Mua5ZdXZJMUqhRFOJ15izlD2+Lvo3fr+1n+fLOyc2U7cvpXVvl7Qag23AmU9JIc1ikMP4MOoxOvQc4zXU9R2dmxhuOob6kjI1LOJjyJqyGN7t27s+SbLfxu5RHQu/c16nyMBA2Yyr9XvcLjj1+gc+fOJb+Tcg4jFVeTAFYfSMOg12FVQGjRjkCxCJhnxPHxtlNsPZ5ZIYwtxbpOEGBwu1D6tG5EyuWb5OQXVXmubEeOmkdHPiN7uLNNm6pD/jGRISye0pOruQUldmeOvAZvRiVcobomwxqOUaeFz9tdzx1Bakgj8egNBL2vR/ICBb0v98z9WxnRA9fPYeTyH3UlCUCJbil2gZg5oE2VE3ywvwHzvlSXznSNBj1z727nVDKQXOdjNX1GNe34SmO3O3MUqY4znqZ8ob+G89RZ4VNC13NHkBLSkNNz0hEEnY6fLhdwMesGEY3LdvV25RymbVgQh9NuSB6XKztmpSY8lBaIqib4tk2CXEz2qDxBqTrkOh+r6TOqLsFFDrwZlXAGVz8njbKoK7AtI0o2/i2NlJCGN16jIAg8/NIHiJUom7PnMCM6h+Ojl3aY5asXXNoxKzXhwRERL928tqazwNIuO64U/MtxPuZIVKNp06ZuFT57VMLoQrd7TyD1c9IoizK/3R5AiaGs8hh0gqSQhjdeo6gzcOm2wMKFC3nmmWcq/N6Zc5hmIf68sz5F0njs3ojOokS3FGfC3lPiWhETGVLlWWD53nau7iDkOB9zJKrRtGlTDh48KOl5asIuKAu+OerW53GFtqFBvHR/J+5qG1rzgzVqpM4Kn1JDWaXpH92Iz3efdbkbs7deY8++d/HuXybRpUsXhg0bVuljnD2H8TRKTHhwNuztarKHM7ijI0NluDvUaWdKXCv+c/AiP12Q5vcqN+ev3eZ3/96vNaSViTorfEoNZdkRgD1nrknqxuyt1xjWIJjExETGjRvHrl27aNu2bcnvHHXGgV8zK3UChVbXTykNOsElT0P7hL7hmGMJFe5GSiafuxcZ0txTHEvUcCa5RSohAe7vG+ksBeVqibWGtNJQ9uzvRpQYyiqNSMXOz852Y/bGa7SH4/r378Zf/vIXRo8eTVJSEmdv2Bx2xrELekp6jiTRAyi0ii57GtqL2JXQ2V7JmXzS3FMcS9Tw1I4PlL0o1hrSyoMyT3I9wPgeMhR/eikl1H7zv/TNUeLe3MQjn+0t0wbJjiyv0dmx8Vs4bubMmQwYMIDhs9/A9Okeh02Z7ebSnkqVr4quzUPoHFG/5gc6iCBAp/B6+DuZQKGGTD53J9TYha+ypCm5kSNhx6ATcGdNvN2L9nCaskKyakG5Sxs3I0vtjoz2GaIolnGed+hvgPScAtJzCn4dziXeWZ9Cm9AgXhrZkQHtmni0PqmycFzclGdZ/+0viE6YMttXs55Kla+OEZ3DOXrxBhI3nkDxru2tsTEcTst22V5LybgzoSYwMBBfX19ycnKoX1++xUhlyHG+q9cJ/PHudryz/rgs905lVOVz6sxxQl2lzgofqKd2x1FEsVgMT17JZeqSfQQbDUSE+KMDPPEKy4fjDqVm8/aGk4h654THvpqd2DMSP4M0Y2apBgD2SVCqg4vR8JuRt/0fd2dcegN3JtTYd33uFj65EnZmDozm0o18liWdd0u4vLzPqTNG+3U9OaZOC5+rZxPuwNndniPk5FvISfdMz67KwnFSXfsvXKu5ZU1NSPU0lCMyoBNgwchOZXZtnsi49CbuSKixJ7i0b99e1uuW5urVqxw4cICg88kYDe0lJ+zMHtKWlftS3XrO/qfVh7h2q5Aj1ZjQO5sfUNup08IHZR1FasvOTwr28NrdHZuwKfmKy+E4OVz7fzx9lb5tGrl8Hbk8DaVEBnQC/G18DONim1f6e6WXdSgJdyW4HDhwgLlz53L06FFyc3OxWCw0bNiQ977bz1+/TybfiUWxj2AjxnqGD//yH9LS0jh48CD3/mUVO87ecMtxQ4HFxrbjmQ4/XkuOKabWC58j8e4pca1o0TCA+IR9isjg8waVhdcOp2W7HI6Ty7W/ZcMAjAa9W1Pla8LVyIDRR8eCER2rFD0N53CXe0tAQAB79+6lqKg4CcrX15e/fLiUHSezKHJwpyYIYEAkfd0/Sfx5XZnfPdqnGf9LzVXUwlpKo+baQK0VPmfj3ccu38RHIY0+PYVeEIhtEUJkg4BKw2tSwnFyufbn5FncnirvCFJ7vmlIx9FaPmeTO6Kioujbty87d+5EFEVaDH2YfxzTUWBxLLxt0Anc3bEJTwyKpt8Hk8v87q233mJQTGvm39Yp4kilNK42aq4N1Erhc7ax7PwRHVRhYSY3NkQaBfmyyNSt2se5Eo6TsxRBKaLjKRswjcoJCwsjKSmpyt+7ktxx4sQJJk6cSLt27Rg1ahSbLxRC93FOhTcNOujUUM+jo4eSm5tb8vPmzZvz/PPPA84tnDyFlEbNaqfWCd9vjWWdS59v0TDAA6NTFu688eUuRVCK6NT2pBQlU92Oz5XFruFcErNnz+a1115j5syZJJ28zJElB3B2U5ZvEfn7ljP4GUI4evQov/vd79izZw9bt24t87jy9zCiSIG7ah0cxN4EuK6dM9cq4XO1sWxekY1TV3JrfmAtxF03vhyuMeVLEZQkOlpSiuep6ozPlcXuy2sOwcF1bNiwge7duwOwZO8lrIIOV/p46Xz8GDr7bTp16sS2bdtISUkp0zT3mWeeYfPmzdy6dYsbN25wI99Kr4mzsbTpz+Ub+c4/oUy42qhZ7dQq4ZOSPm8V5Wt8qibcdeO707VfE526SWVZna4udm2CAWPvSehDowA5+nMKZaInMTExZX5bUFDAkSNHsFiKjwD0ej3/nDMOa/1IJn2a5NXEF1edjdRMrbEsk6OxbF0TPTtZuQUs3n6aueaDTF+6r1L7M2ex17+5Wp4oVymCRu0hLCyMK1euYLP9JnJSFrsFVhsfbzsFyJeFvPqnitexWCy0atUKq7V4nAEBAbz//vt07ty5JGPYWRs7OZHibKRWas2OT44bV+qOTwBsLliPeZvdp7P43zlpnSAqwxOu/Rp1Bz8/PwIDA7l+/TqNGjWSpVbUvkuTKwu5fPRk3bp1PPPMM4SFhfGnP/2JRYsW0a1bN5544omSx3gz8cXVRs1qp9bs+OS4cW1icYq/K/j76Pn39N70DvdFtCq/119pbGLlnSAqM452BldXs2owZdbwDqXP+eTcpcltiH7kyBHuu+8+5syZw5tvvsnmzZt54403MJlMfPnllxUWx1PiWmGeEcewTmH4GXQYyzlc69y0lna1UbPaqTXCJ9eNG90k0OWJ+q62oayacy+T2hmwFeXj0im5wiid+eqK+LnbtV+jblE6s1POXZpcWci+WPjDH/7AkCFDGDFiBEeOHGHUqFEIgoCPjw8rVqygRYsWlf6tPXnruyf7E9emEZEN/GkU6EtkA39aNQrAi9HQWketCXXKdeN2Cq/PI3EtJdWMvf27kfTZtp8fUkW2n8iqkHqvRqQ4PSilFEFD/ZROcJFzl9arVUPJWch6bKxZ+jGTu4WSkpJCw4YNnfr7KusQb4GfQed0mYUjCIJrjZrVTq0QvkOp2Ry9lCP5OkaDjnB/G+31mTzRvoD/ni7kTL4/Br3e6Yl67KCejIUyqfe/XMzmTNYt1SbRSHF6UFIpgoZ6Kb3jk7NWdHyPSBZuPC7pOjabjW/fe5FeMR2d/tua6hDdZa5htYms++WyJnxqo+SGkSEduKCwkHmTBuNjK8DX15e8vDx63zWEqa9+7PJEXTr1fq75IKcyb0kep7eQo+BdK0XQkELpHZ8ctaIGAVo38mPRm38h77QvhqgeFJ/8OYcADOsSIUH0vGdndvjiDQ6nZdepSIuqo8a/3TBWyadpggBDOzYlunlTioqKuHnzJqIo0rNLB6b1jmCRqRufTevFIlM3Zg5o49LEL1doxpuIoo1VMiQVaGi4QunklvE9pCdlWESRvy9fy66r/nwwZyL+Pq7tBYw+rmUhu1qHKCc2kZKyjrqCand8ct8wRoOep+5uz99GJ9GzZ0/Onz+PwWDg5+RTtBr+e1rH9ie8RWuaN23scjdjuUIz3qTQCn/7dAUZ20RMJhMdOnTw9pA06hClQ532WtENx2o2rq4aASG8IxkGHS98f47WoYGcuZLrlJWYlCxkKXWIclLXPDtVOxPLecOU7o4NsGvXLmLvGUPkvY+RHhhJsChyySpyKb2QA+mX0JPGuxuPM7hDE6dq3OQIzSiBrj3jyD6/jqFDh9K4cWMmTpyIyWQiOlqru9NwL+XdWx7u3UKi8BVjP8NPSb+JQSdg0AlYRdGthuhymG7IRV3z7FRlqFPuG8Zig52nsjiUmg3A5vMF1B/zCpl+zSiw2Cgst/qzoqPQKrL+yGXGL97F0t2nHXoeOUIzSqBleBMWLVpEamoqH374IZcvX6Z///706NGDd955h3Pnznl7iBq1lPJG1cm/CpVciCIUWUX0OoGOTethEABLYZnHGA06/Aw6hnUKwzwjzuXSG7lMN+Sgrnl2CqKohPWGcyzefppFm07IunMq3Xl8w9F0p0IdYlEB3YWz/HXaMDp37lztY2cs28/GZMf6fCkRo0HH0/e0q7AytFqtbN++HbPZzH/+8x/atGmDyWRiwoQJREbWDsHX8D5FRUUEBASQn5+PXq9nrvkga36+5JbnEqxF6HZ/yv1T/oCuYeSvvf3ky0J259hdoVGgL3e1df0oR02oUviUdsMACIhYkrfQ5MoBHptiYvLkyTRo0KDC4w6lZrtsSiuINgwGPUVebGXiZ9Dx4/NDqv1SFBUVsWXLFhITE1mzZg0dO3YsEcGmTZt6cLQatZHQ0FCOHDlCWFgY05fuY0vKFTc9k8i9HcP4ZGovt1xdrrHX89Nzs0C+c0J7uZYUu0Klo8pQpxKzI0UEfDoN5cag5/g0GaJ7D2Xy5Mls2LChxJwWXLfx8jPosF74mWb+ogvJ1vLgqHG0j37FqZcAACAASURBVI8Pw4YN47PPPuPy5cvMmzePffv20bFjRwYPHszixYvJzMz00Kg1ahvuqOWrHIHtJ7MkmbVXZwAv19gjG/ijl3FSkMOuUOmoMrlFqdmRNhFsQHa9loROep2goCu8+OKLZGRkMG3aNOLj44mOjnauozjFgmOx2vCLiuW8F9sGumIc7evry8iRIxk5ciT5+fn88MMPmM1mXnjhBXr37o3JZGLMmDFOu1xo1F3sCS4xMTFuTxhzNenDkW7wLRsG4KsXKuQQOIOtqIDL+3/E0Ko/Vpnfg9J2hUCtshFU5Y7vdqH303+rQxSLV027bjVh7oerWbt2Lbdu3eLOO+9kwIABLFmyhAc7N67WlNZo0GHQCQi/doywirjFsshR5DCONhqNPPjgg3z55ZdcunSJ3//+96xbt46oqChGjhzJv//9b27cuCHjqDVqI6V3fO5OGHMl6WN50jkmfZrExuQMCn7dPZW/ZoHFxokruZJED0Cn07F20XMMbBcq6TrVYbcrPJyW7bbn8DT6V1999VVvD8IZliedY1nSeSwq8P2y2ER+PH0VXWAI+U270GKQCUOLrmzdf5QFc2aQnXaSqYM68/y4ftQP8CU0yI+IEH+6RYYQEeJP6rVbTomdXoB2YUH0iWrImcxcWSyy3WUc7ePjwx133MHEiRN58skn8fPzY9WqVcyZM4e9e/dis9mIiorC19dXtufUqB38+OOPAPTr148AXwO/XLzBmSz3OSJFhPgzuluEQ4/1qAuLzcbdnZoy5c42HL10g/3nr7vtqSw2kU3JGRy7nMO5q7eJahxIgK8yI2+OoKrkFimJId6kfJ8/o0GHTRSJ1N3g0uYErFfOEB8fz7Rp04iMjJT0Ov199JhnxPHh1lOSskd1AvjodR43jr5x4wbffPMNZrOZXbt2cc8992AymRg5ciQBAQEeGYOGsvnb3/5Geno6CxcuBNw/L4zpVuzcVBOenp+MBh2JM/tyOC3bo5ZntSH5RVWhzo+2nZLFk9PTlN+c5v9aG3jWEgxD5zL1L5+SmppKTEwMw4YN48UV211+nXYj6VmDojEa9C5dQy/A1LiW/Pj8EBZP6elRD7/69eszdepU1q5dy9mzZxk+fDiffvopzZo1Y/LkyaxZs4b8/HyPjUdDeZSv5XNnF3OjQedwo1ZPurD4++hYMLIjoojHLc9qQ/KLaoQvK7eArcev1IIOd78hipBfZOOL5Dz6x7/AxYsXGfdwPMeuu97Jz24kHdnA3+UmsH8edQevjurs9Tqehg0b8thjj7FhwwZOnjzJwIEDef/99wkPDy8Rx8LCwpovpFGrKO3XaceZvo/OIIJDjVplMdUQRWr65pc/evCm5ZnUXp3eRDXCt/pAGlYVnOu5gv3w+OTVAmwte0s+17JnotWmJrChoaH84Q9/YOvWrRw7doxevXrx5ptvEh4ezmOPPcb69espKiry9jA1PEB52zI7NXUxdxZHy3dAHheW4i9p5V/U4qMHoYxbjFIsz9SY/KIa4TuUlq3aPnaOYA9RytlVGmqeDOSyX/Ik4eHhzJ49m127dnHo0CE6d+7MK6+8QrNmzZg5cyZbtmwpUzupUbsoH+osjb3v44/PD+Hpe9oxqF2oy7ZezpTvyPG9rQ6bCHpBoH9045KjB1nEVibs85daUE1yy9B3t3Faxb3sHMHPoKNXq4bsOpUl+VpDOzThs2llHSdqexPYc+fOkZiYiNls5uLFi4wfPx6TyUS/fv3Q6VSzxtOoAavVitFo5Pbt2/j4+NT4eFcyLYvLdxyPfrjXQeY3So+ry4y/c7OR8/3/3IUjrk5KQTX5qGrvaOAIAnD9lusuEaUJNlacEGp7E9hWrVrx3HPP8dxzz3Hy5EkSExOZNWsW165dY8KECZhMJvr06YMg5yGQhsfR6/U0btyYzMxMmjVrVuPjnTKMcLHjgp/EsKqj2MOKXSLqk3Eth4BGHnlah1BThwfVLIM9dWN5k3yLDQRB8mt1JhOtttK2bVvmz5/P4cOH2bhxIyEhITz66KNERUXx3HPPsX//flQS7NCohMoSXKrDXSH/Q6nZzFi2nw1HHR+LVPKKrPxp2S5sBcqKgKmpw4NqdnztmgS5NdQpCGDQCVisolczRxsGSC/YttpEhzLR6godO3bklVde4eWXX+bIkSOYzWYmT56MKIolvQRjYmK0naCKqCrBpTrs539yhfyLQ6g17yLdwckcAZ1fELaiAnQ+ygkt5uSrI8FMNcLXtXkD1h/LkJzgoheK7b/s2Isx7YXaP57O4q0fjkt7EgnYu0q7XnwuYrv4CwH6wXIPTfUIgkCXLl3o0qULr732GgcPHsRsNjN69GiMRiMmkwmTyUSnTp28PVSNGqguwaUm5Aj5e9ShpRIEQSCgbU9Eq7KOgCo7YlEiqhG+8T0iWbjxODYJ3nY6Aab3iyIrt7DKlV5MZAjbjmeSdPaaXEN3GHuIMi6qETtPZrno3GIgWkzl6aefZvHixW4YZe1AEARiY2OJjY3lrbfe4n//+x9ms5l7772XBg0alIhg27ZtvT1UjUpwNtQpJ4dSsz1eNF45OgShuFuLEoL2ajpiUc3BWeMgPwa3byL5Ov9OOs/tIgtPDWnLIlM3Zg5oUyG8MW94ce2bp7EXy7rqRGE3kl7xwZts2rSJVatWuWegtQxBEOjTpw/vvvsuFy5c4J///CcZGRkMGDCA2NhY3n77bc6ePevtYWqUIiwszOUdn1Q+2nZKObaJOnkL9qXgaLG/ElCN8AHMGhQtSZBsIg5Z7bjTAqkqRJuNwBvn2L97G1arVVLxeXBwMCtXrmTWrFnahO0kOp2O/v3788EHH5CWlsa7777LuXPn6NOnD71792bhwoWkpqZ6e5h1Hm/t+LafuMKmZO8IblVYrTZEm3eF2JlifyWgKuGzC5KPxK6LjljtuGKBVN9owFfv2lvq72tgZGsfXnrpJVq2bMmLL75IrwYFLmei9ezZk3nz5jFp0iTN1stF9Ho9gwYN4p///CeXLl3i9ddfJzk5me7du9OvXz/+8Y9/cPnyZW8Ps07iSnKLVJYnneOxpfsVZ6Qh6HQIOj2iKHotU9mVXp3eRDUF7HYOpWYzbvGPsrUlsnczqMqI+XBaNi/85zDHnEjTNegEp8ZXvlj26NGjJCQksHz5cqKiooiPj+eeB8ay4eQNpzLRRFFk1KhRdOzYkXfeecfh8WhUT2FhIZs2bcJsNvPf//6XmJgYTCYT48aNo0kT6eF4jZo5duwY48aNIzk52SPP5+1kFqmU7xAj97Vn3NWaF4Yrp5i+JlQnfDOW7ZfUbqc8ggDDOoWxeErPSn/v6g2vF4q7sUsplrVYLKxfv56EhAQ2btzIiBEjePTRRxkyZAh6vWMh36ysLLp3784nn3zC8OHDnXoNGjWTn5/P+vXrMZvNfP/99/Tq1QuTycTYsWO1rvJu5Nq1a7Rp04br193Xg86OWtuh6QWIbhJEm9Ag1h9NR2LP22opPmpxrujfm6hK+LJyC+j39hbZXVyqstqResNHNQrg0o18BH4tTv+V8iUUjrT9uXr1KitXriQhIYH09HSmTp3KtGnTaNeuXY1/u337diZNmsSBAwcccrrQcI3bt2+zbt06zGYz69evp1+/fphMJh588EHq16/v7eHVKkRRxGg0kpOTg5+fe8+V5F5sexJ/Hz0Te0aycl+q292vnLV58yaqEr7F20+zaNMJ2T9Ao0HH0/e0K1PbI4oiM5cfYOOxDJdThXUCLH20N8cu58jqj/nLL7+wdOlSli9fTps2bYiPj2fixInVTq5//vOf2b59Oxs3bnR4t6jhOrm5uXz33XeYzWa2bNnCwIEDMZlMjBo1inr11JHyrXSaN2/O7t27adGihduew12LbU9Sz0/PzQJP9Qms/uhIKagqucVdDuh2qx1RFDlw4AAmk4mABmFsSU6XVB9jE2HJ7rPMHNCGRaZufDatV5UlFM7QpUsX/v73v5Oamsq8efNYv349LVu25OGHH2bTpk2VdiZYsGABoijyxhtvSHhFGo4SFBTEpEmT+Prrr0lNTWXChAl8+eWXREZGMm7cOBITE7l1S1mWU2rDE5mdSuqA4CqeEj1QT5cGVQlfTr7FbdfeuXc/QUFB9OrVi8TERHw73CVLhtT2E5lczZXHeLo8Pj4+3H///axevZrTp0/Tt29fXnjhBaKioliwYAEnT54seaxer2fFihV89NFH7Ny50y3j0aic4OBgHnnkEb777jvOnTvHyJEj+eyzz4iIiCgRR62rvPN4opbP3e2Gahv2RtjumvPkQlXCF2x0n9HMhVMp3L59u0TsfEJbYZXh7REpdix3N40aNeLJJ59k//79fPfdd+Tl5dG/f3/69+/P//3f/5GTk0OzZs34/PPPefjhh7l69arbx6RRkQYNGjB9+nTWr1/PyZMnGTx4MB988AHh4eEl4uiO8pOs3AIWbz/NXPNBpi/dx1zzQRZvP634Cao6PLHjc+diu7Zi79KgZFQlfB2aBrulS4PRoGP+rEe5++67S4yKdX6BslzbJuJxx/KYmBgWLlxIWloazz//POvWraNFixY88sgj+Pn5MWHCBOLj47XuBF4mNDS0pHFucnIyffr04e233yY8PJzp06fzww8/SO4qb+8e0O/tLSzadII1P19iS8oV1vx8ifc2neDOt7cwc/l+DqWqp3u2HSl+nY7izsV2bUUNXRpUJXzje7jHDkcEpvSLZuPGjaxevbo4S6wwT7bre8ux3MfHhwceeICvvvqKU6dO0atXL5599llWr17NgQMHeOmll7wyLo2KNG3alCeffJKdO3dy+PBhYmJi+POf/0yzZs2YMWMGmzdvdrqr/PKkc0z6NImNyRkUWGwVQnb5v/6sJicjpeKJInZ3LbahuNxAoheHYlF6lwZVCZ+9c4Hc3nQtGwbw2tpjTF+6jx1FUbz+n71MHnUvOpmeRwmO5Y0bN+app57ip59+4ttvv2X48OG88cYbdO/enc8++4ybN5W9QqtLREREMHfuXPbs2cP+/ftp27Ytzz//PBEREcyaNYsdO3Zgs1V/7vRb/WnNLXMccTJSIp4IdbprsQ3w5OBo/jzqDvxcdHtSMkqY86pDde/4rEHRGA3ypuOfu3qrTAjon7sv8t8zhbI4HfjpBcU5lnft2pXPPvuM5cuXk56ezpo1a2jevDlTp05ly5YtNU6qGp6jZcuWPPvss+zfv59du3YRERHB7Nmzad68eYk4lg9Zu9o9wN7d+3CaOsKenkhucddiWyfA1L6tmBLXilV/6Mt9d4ThqxdqxQ5QDV0aVCd8XZuHcHdHeW2hCstZGlhsonz2PoKgWMfyhx56qKSu7MSJE/To0YM//elPREVF8fLLL3P69GlvD1GjFNHR0bz44oscOnSIzZs306BBAx577DFatWrFs88+y759+xBFkY+2nSLf4loKu1rS0cFzRtXuWGz3iWpYphXa4ik92fPCUJ67rwNjukUwtEMTxnSLYN7wDrRtEiTrc7sbNXRpUFUBO6jPPui+O6q2Q1MCt2/fpnfv3vzxj39k+vTpAPz8888sXbqUFStW0LFjR+Lj4xk/frxWeK1ARFHk6NGjmM1mzGYzFoM/jP4rNglr2qqcjJRGTk4OERERHgnTy+nVadAJ/OfxOx0q8l6edI7XvjtGgZv8xgREijv6yYfS5zxQofCpyT5ILS4Gx44dY+DAgezYsYOOHX8zmi0sLOT7778nISGBbdu2MXr0aOLj4xk4cCA6neqCBbUeURR5+cudfHE4G6vg+g6lMicjJSKKIgEBAWRlZREYKE8WdnUUi18K+Zaaz02rwqCDVx+4wyFbL08s8u8ID6ZdWD1+uZjNmaxbkiNdzoi6N1HN7JWVW8DCjcfZpBrRK24Kq/QbAKBTp0689dZbTJw4kby837JZfX19efDBB1mzZg0nTpygW7duzJ07l9atW/PKK69w5swZL45aozyCIHBTX0+S6IE60tGh+PV6oqTBzpS4VtW2CasJP73OYdEDJIWsHRuPwKhuzVhk6kbniPqyHO+E1vMjIsRf+oXcjOJ3fIdSs/lo2ym2n8jEahNla0fkTtTmVA7Fq+eHHnqI+vXrs3jx4mof+/PPP5OQkMAXX3xBp06dSkKhQUHqOouojUxfuo8tKVckX2dohyZ8Nq2XDCNyL3379uXdd9+lb9++Hn3eq7kFrP4pjZTLN0m7fpuL2Xmk38hHECjTBUEvgF4nMKRDE4cN6cEzHqGipZCwpA+YPPYBfgq+k52nsiRfUyeAj17HoPahPDEwmq7NlbnwV3R1phyhBU+hF4oPdRsF+tKpWX1yC6xczS1Q/DmJHUEQ+Ne//kVsbCyrVq1iwoQJVT62W7duvPfee7zzzjusXbuWhIQEnn766ZJQ6IABA7RQqJeQq+Ba6enodrzRkBagUZBfhVBwaTGUakjvbo9Q0Wbj9ul9HDqVxoUfr+PfVrroQbFhh702dMeJLMVuABQrfGpq/ChQLBxWm0hmbiHbT2Sy98xVFm06ofiVT2mCg4NZuXIlI0aMoGfPnkRFRVX7eF9fX8aMGcOYMWPIyMjgiy++4KmnniI3N5dp06YxderUGq+hIS/FBdfpknYKSizBqQpPZXY6QmVi6Cru9ggVrYWINzKInPZ3bIJe9o1F6dpQQHHip8hluat1SN5ChAohWLW6YvTs2ZN58+YxadIkpzwjw8LCePrppzl06BBfffUV165do3fv3gwePJilS5eSm5vrxlFr2JGj4Dq/oIClrzzOhx9+qBhRqQpP1PJ5A3d6hArWInSXjtCw71isgl5SB5qaUGptqP7VV1991duDKM/L3x7hxBXlH647isUmsufMVUL8fVSR7BIXF8e3337LL7/8wj333OPU3wqCQHh4OMOHD2fOnDk0aNCAL7/8krlz53L8+HEaNGhAixYtSjxRNeQlwNfALxdvcCbLtZZHggD3dGrK74d2Zv369cyZM4cNGzaQl5dHy5YtPZI96ShZuQUkHkznf9f8+DFdZOvxK5y7epuoxoEE+Co2mOUQW49fISVd3jlQEIrzDx6Ja05yfjAFFs+cH1lFkeu3Crk/RjlNsBWX3FIbGj9WhVrKGwCysrLo3r07n3zyCcOHD5d8vYyMDFasWMGSJUu4fft2SSi0VatW0gerUQYpafDl79H8/Hx++OEHzGYz69ato3fv3kycOJGxY8fSsGFDuYfuEKUT3mxWK0Xib4soo0GHCKo6YqgMOZtu29+Twe1DeWJQNB9uPeXxkjCl1YYqTvjc1WVdCQgCDOuk/OJOOzt27MBkMnHgwAGaNZNntSaKIgcPHmTJkiWsXLmSLl26EB8fz7hx4xS1m1A7rpyRF5fgdKzyPOb27dusXbuWxMRENmzYQL9+/TCZTIwePZqQEM8IjKMJb4IARoP6sqvtyLEB0Alw3x1N6do8pCTBxlsbC6XVhirujK82N35US5NGOwMGDODxxx9nypQpTncGqApBEIiNjeWDDz4gLS2NWbNmsWrVKiIjI5k+fTo7duzQ2iXJwJS4Vswf0RF/H32NPpP2EFh1ogcQEBDAhAkTWLVqFRcvXuSRRx7h66+/pmXLlowaNYoVK1aQk5Mj7wspRV0w3rYj1SNUEODeTmF8/HAPZg5oU7LT8lZHeaXVhipO+Gp740c1NGkszfz58xFFkTfeeEP2a/v5+TFu3Dj++9//kpyczB133METTzxBdHQ0r732GufPn5f9OesSNRVcGw06/Aw6hnUKwzwjzqmdUVBQEJMnT2bNmjVcuHCBCRMm8OWXX9K8eXPGjh2L2Wzm1i3Xzhkro64Yb5dGikeo0aDniUHRFX7uzY2FkloVKS7UOdd8kDU/X/L2MNzKmG4RLDJ18/YwHObSpUv06NGDxMRE7rrrLrc+lyiK/PTTTyWh0K5duxIfH8/YsWO1UKgE5Kwxq47r16+zZs0azGYze/bs4b777mPixImMGDECf3/XHT2kWBWq7YihNHKHrOUyOHAFJc17ihO+2nzGZ0ctrhil+f777/nDH/7AwYMHadSokUees6CggP/+978kJCSwe/duxo4dy6OPPkq/fv20rFAVkJWVxddff43ZbGb//v2MHDmSiRMnct999xU3e3b0OjKcSyktucIZ5DzX9NbGQmlnfIoTvtqc1WlHSSsfZ3jmmWc4fvw43377rceF5/LlyyVZoQUFBcTHxzN16lRatGjh0XFouEZGRgZfffUViYmJHD58mAceeACTycTdd9+Nr69vtX8rx2JYaROvsxxOy+bjbafYejwTgeIzMzvlszZLZ41n5Raw+kAaKek55ORbSL+Rz/H0HNzU7KFKlLbwUJzwgbo6MDiLmr+AhYWF3HXXXTz00EPMmTPHK2MQRZH9+/eTkJCA2WymW7duJaHQgIAAr4xJwzkuXbrE6tWrMZvNpKSk8OCDD2IymRg8eDA+PhWt0uTapah1wVkaR0PWpUs+AK9uJJQYalak8Kmt554zKG3l4yxnzpwhLi6OdevW0aNHD6+OJT8/vyQUumfPHsaNG0d8fDx33nmnFgpVCampqaxatQqz2cyZM2cYO3YsJpOJgQMHotcXJ3bUNeNtqSjN41iJ9cuKy+qE4i7r80d0wN9HkcNzGUEoDkeoVfQAWrduzYcffojJZHJr6rojGI1GJkyYwNq1azly5Ajt2rXj97//Pe3ateP111/nwoULXh2fRs00b96cP/7xj+zdu5f//e9/tGnThmeffZaIiAhmzZrFjh07qOdXt4y3peBMyYcnUGp7NkXu+OwobeUiFSWufFxl5syZ3Lx5kxUrVihqdyWKIvv27SsJhcbGxhIfH8+YMWO0UKiKOHnyZMlO8FpYLD6xY7DVgea6UlBSpEzpBgKKFj747VB3w7EMWRoleouaXDHURl5eHr179+bpp59m+vTp3h5OpeTn5/Ptt9+SkJBAUlIS48ePJz4+nr59+ypKrDWqJ+ngER5OPItVQoBK7UcMjqCU3AgfvcBQJ/sPehrFC5+d0R/t4lDaDW8Pw2mUvvKRwrFjxxg4cCDbt2+nU6dO3h5OtVy6dIlly5aRkJCA1WolPj6eRx55hObNm3t7aBoOUFfr+BxFSdnwQ9qH8nl8b28Po1pUcYi2POkcRy4qW/R05TYQUlwx1EKnTp146623MJlM5OXleXs41dKsWTOef/55jh07xrJly0hNTaVbt27ce++9fPHFF4off13HHS4mtYWs3AKeWXWIIqv3RQ9g9+mrirdlVPyOTy0NaRsH+XJXdKhbXTGUiCiKPPzwwwQHB7N48WJvD8cp8vLySkKhe/fuZcKECcTHxxMXFycpFFq+dirYaKBD02Am9Kj994M7cYfxtpopXbJQZLUp5ihIDeepihY+JR3W1oSvXmDPC0Pr5MSWk5NDbGwsb775JhMmTPD2cFzi4sWLJaFQURRLQqGRkY43dq2udqq2tMvxNg4nvIk2RGsRxuR1TO3biokTJ9K2bVuPjdPdKD3xT+k1k4oOdX607RT5FuWLHoBOEFRlPi0nwcHBrFy5klmzZnH27FlvD8clIiIieOGFF0hOTmbp0qWcP3+erl27ct9997Fy5coaQ6HLk84x6dMkNiZnUGCxVThryf/1ZxuOZTDp0yRVdgxQAo4ab9/XOZxvZg/m/56fyuXLl7nrrruIjY3l7bffVu09akdpJQuVoSRD6spQ7I5PSYe1jqL0VY67ee+99/jyyy/ZuXNnjTZUaiAvL481a9aQkJDA/v37S0Khffr0KRMK1UJw3qG0i8mGbbu4o10UQ3t0qPSIwWq1smPHDsxmM1999RVRUVGYTCYmTJigKtu7Q6nZmD7dQ77Cj34GtQsl4VHlJrgoVvjUaFZdV5whqkIURUaNGkXHjh155513vD0cWUlLSysJhQqCUBIKzbIFytbtXMN14uPjGThwII8++miNj7VYLGzZsgWz2cyaNWto3759iQjK1XDZGaxWK9euXSMzM5MrV66QmZlZ5X/fiDGhb9EdQafoYB0GncCrD3RS7MJOscKnxvZEdX3HB8WO/LGxsfzrX/9i+PDh3h6O7IiiSFJSEgkJCaxatYpmple5FdIaEeeTYepCmr2nmDdvHvXq1ePFF1906u8KCwvZtGkTZrOZb7/9lpiYGCZOnMj48eMJCwtzaSx2IatJxOz/XLt2jZCQEEJDQwkNDaVJkyYl/136/33qNeSxby5R6GmHaRdRclRDscLnzb5RrqCGTCZPsWPHDkwmEwcOHPDKCtpTpGXdYNC7O7GIrmeA1oXCak/w/vvvc/r0af7xj3+4fI2CggLWr1+P2Wxm7dq19OjRA5PJxOjRoxEEwSERu3LlCtnZ2dSvX79aESv9340aNcJgqNmWTY1RMKVGNeQxwXMDwUbFDq1SRGB8rOMZgLWZAQMG8PjjjzNlyhQ2btxYYjZc2/juaBZ6vR6LhImoyGpj4id76BJRXyt5kEB4eDi7du1y6LFWq5WrV69WKWKFhYV07tyZ48ePM3v2bGbOnImPjw+NGzemVatWNGvWrES0OnXqVEHQHBUyZ/Fm93RXybdY+XjbKcVFNRSrLh2aBuNnSFfFB10bzKflZv78+Wzbto033niDl156ydvDcQtyTEQ2EU5n3uJ05i2MhnQWbTqhlTw4idVqxc/PjzNnzrB169Yad2XXr1+nQYMGle7COnXqVGFX5ufnxw8//EBiYiKbNm2iQYMGxMXFMXr0aOrXr+/QGOWo7czJt0h5m7yCKMLW45lczS1Q1Pyo2FCnmrI6lbqd9zaXLl2iR48emM1mBgwYANSu4m53heNrs82dI1gslpIdmSOhxRs3blCvXj1yc3O58847HQotuhqFuHnzJt9++y1ms5lt27YxePBgTCYTDzzwAPXq1avweDlrO+XLe7BP+Z7xq1XiMZBihQ+UY7paHUo+wFUC69atY+bMt1lBfQAAIABJREFUmSz/fifLD2bWquJudydg1ZZ7q7SQ1SRimZmZ3LhxoyTZozLhqiy0ePv2bcLDw8nNzfXY68rOzuabb77BbDaze/du7rnnHkwmEyNGjCAwMNDhInNHFzpynfEF+urJL7J6tAu70hL/FC18Sndu8fepu6tyZxjz3CJ+FqJA7yPLBKAUPJFsoMRogl3IHBGxzMxMsrOzadCggUMi1qRJExo2bOj0jkwURYKCgsjIyCAoKMhNr7xqrl27xtdff43ZbGbv3r3ETnyKtCZ9KLI5vquqaaHj6SiYQSfQIMCHzNxCyddSWqmXooUPlOvVGRrky2fTeilqQlIiy5PO8dfvk50quFXLTscTE5EnSh4sFgtZWVkOiZg9tOhuIXOF6Oho1q1b53Vrsm2/nOV3Xx7BIjpfa1fTQseTUTCjQaBd02AOy9AVR2k7PsUmt9ixT36vfHvUo1vzmugfHaqJXg0cSs0uDvU4uWjJK7Lx+vcpxESGKPo9bhzkx8B2oW6diFxJDigtZDWJmD202LBhw0qFq3Pnzl4TMmdp2rQp6enpXhe+L36+6nLvwJqyIGcNimbnySyPRMHyLSK/pN1A4LdTQVcwGnR0CK94/ulNFC98UCx+3/x8iX3nr3t7KIAyP0glIsVrValp0OXxxERktVr5ZNMvjGztW6OI1SRkMTExFX6uVCFzlqZNm3L58mWvjiErt4DtJzJdXgiJImw4lsG7G48zrW+rCoudrs1DmD+iA39de4x8i/t3AnI8gxJLvVQhfAARDfwVI3xK/CCVhhwTgBLToMtjn4jcGY63iAILP/uSv+1eQlRUFK1bty4RrdosZM4SHh5Oenq6V8ew+oB0o3qbCB9tPcVHW0/RJNiPiJAAIhv4l2Q/T4lrxY6TmWw4pnyDD6WWeqlG+Or7+3h7CIByP0ilIccEIACrf0pTVBp0ZUzqGUnOzZu8vz2VQqsoyyq5PPeOHE3nvo1ISEjgyJEjxMfHM2XKFCIiItzwbOpECTs+uYrM7cc6l28UcPlGAfvPXy9T56mGMi9QbhNg1Qjf+Wu3vT0EQLkfpNKQYwLIt9hIuXxTphE5TnVei5X9Ozs7m5CQEBq1i8XY+T7yG7RBEMAmyPf1CmsQzIt/eJF58+axZ88eEhIS6NKlC3369CE+Pp7Ro0djNBplez41Eh4ezs6dO706BncWmef/+n3acCxDnhikmylOUuugyHN6VQhfVm4Be05f9fYwFP1BKg25JgA5+nrZbLZq3e/L//vatWvUr1+/0kzF8hZV9mSP0hZV9nY5S3afIz0nX/L4dQIlZ8qCIHDnnXdy55138t5777FmzRo+++wznnjiiSrbJtUV7Mkt3sQTVovKzsNXR1mSKoRPjrCZVHQCqkixVwpyTQDBxoohbpvNRnZ2dgXBqkrMrl27Rr169SoIVmhoKO3ataN///4VnD18fFwPrTcK8mPmgDYkX86RrcC9sjPlgIAAHnroIR566CFSU1NZtmwZU6dORafTlbRNqkuh0PDwcK+HOtVktSg3diOKwe1DeWJQtKI3CKoQPiWYsw5oG6qJnhPIMQHosXH8f1uYvObtMmKWlZVFUFBQhd1YkyZNiI6OLmNb1aRJExo3bixJyFxFrkmwbZOgGs+UmzdvzosvVgyF9u7duyQU6u/vL2kcSkcJO77xPSJZtOmEV8fgaXQCdGwazKhuzSptAqxEVCF83jZnNRp09G3TyKtjUBOiKHJv22AWbpQWkxGB/hE+tIy9v4y4NW7cWBUd3uWaBBeM7OTwYysLhX7++efMmjWr1odCQ0NDuXr1KhaLxS3dERzBE7WdSsMmQruweopPQiuNKoTP2y2K6nr5giiK3Lx5s9pzsfL/7efnR8PRL0BEFxCcL+YVBLi3UzgvTnnADa/IM8gxCXYKr8ddbUNd+tu6Fgo1GAw0atSIzMxMwsPDvTYOTxaZKwU5zuI9iSqEz5tx89pYviCKIrdu3ao2U7G8mPn4+FQaWoyMjCQ2NrZCobTRaJTktVpbsmelTII+eoG3xsbIMo66Egq1lzR4U/g8UdupNCo7i1cyqhA+b8bN1TIB37p1q8a0+9L/1ul0lWYthoeHExMTUyEJxJVJ0dUJoDZlz7r6HvgZdLw0sqPs70HpUOj777/PmjVrWLJkSa0JhSqhiB1+s1p0pDuDpxBF0W2f6+1CdfUKVIXweStu7s0J+Pbt2w7VkNkfY7PZaNKkSQUxa9KkCZ07d66wIwsMDPTI63BmAhAAm6WAJwe3rVWJRM6+B0YPdf3w9/dn8uTJTJ48mdTUVJYvX860adMQBEG1oVAlFLHbmRLXipjIED7edorNKVco8qLZsE6Adk3rcSojlyKrDUHnmpdoVWw/kcnhtGzVLFYV353BjidbFLmjDiU/P9/hgujMzEwsFkuFXVdl6fj2fwcGBip6lX44LZuPt51i6/FMBH4rxoWyadC6lE2c+t8WfvjhB3Qyfzm9jaPvgTdTwUVRJCkpiYSEBFatWqW6UOiLL75IYGAg8+fP9/ZQSliedI6/rk0u83l7GtvFo7S9uJ6YQfez/GAm/m16gSii85HnCMcTXUTkRDXCB+5vUSTabOh1cO8d4TVOPgUFBQ4XRF+5coXCwsJKBasqMQsKClK0kLmKvbg75fJNcvKLCDb60CG8XkkatMViYdCgQTzwwAM8//zz3h6uW6jpPVAKeXl5rFmzhoSEBPbv36+KUOg//vEPTp48yQcffODtoQDKaasmnt3LlW/+Rn5+saGCzj+YwC534xvaCkNIGAEt7sAmUQn8DDp+fH6Iou7hqlCV8AG8tS6ZxTvOuOXaBZdPckfWNp576okaxSwvL6+CcNVvEkFWcFtu+4Qg+hgJCfSjY3h9Hr6zDa2aNlLsZKE0Lly4QK9evVizZg19+/b19nA0gLS0NJYtW0ZCQoKiQ6GJiYkkJiayevVqbw9FMY20DTqB9mFBhNX3x18vsvzDt8k9vAlbXk7JYyZ9uJk9aXmSnsdo0PH0Pe1UUdagKuErXj2luOVGshUVkL1zBdYjP9C3b98ad2b169cvEbJDqdl8tO0U209kApTJPrWHsAa1D+WJgdF0ba6OGLi3WbNmDU8//TQHDx4kJER7z5RC+VBor169iI+P58EHH1REKHTnzp3MmzePXbt2eXsoHm0a6wy2ogIEQSDo5gVendCXCUN7M9d8UBaXIaU1nK0K1Qif28OclkLSPoonyAcGDRpETk4OrVq1YsmSJQ6My4GkBSfODbNyC1h9II2U9Bxy8i0EGw0lLUnUEEaQi9mzZ3P58mVWrVql7ZYVSOlQ6L59+0pCoXFxcV77vE6ePMl9993H6dOnvfL8drJyC+j39havO05Vh31OmjM0mlUH0jideUvyNYd2aMJn03rJMDr3ogrh80TIIMpwg21/fbjMz0aMGMHatWur/BtXxLg4U7Ryz09t51iW/Px84uLiePzxx5k5c6a3h6NRDaVDoUBJKDQy0rPGD7m5uYSFhZGbm+vVxdLi7adZtOmEooWvNFK7rNtRy45PFWlzUjp5O8pfHrmHRYsWlWntEhgYWOXK8VBq9q9hV+du7LwiG69/n8LhtOwyP1+edI5JnyaxMTmDAoutwhcm/9efbTiWwaRPk1iedM6p51UjRqMRs9nMggUL+OWXX7w9HI1qiIyMZN68eaSkpLB06VLOnz9PTEwMw4YN48svvyQvT9r5kaPYk8Ju3vR8O6vSKMFf2BnkED2jQVfSRUTpKF74pHbydgSDAMfSc5gzZw4mkwmDwUDbtm1p06YNcXFxjBkzhp07d1J6cyxFjPMtVj7edqrk/3/bOdZc6CqKkFdk5fXvk+uE+LVv356///3vmEwmbt2SHorRcC+CIBAXF8fixYu5ePEijz76KEuXLiUiIoKZM2eyZ88e3B1kUoJZtVz+wmH1/BjaoQljukUwrFMYRoMOpUb91WTtqHjh80RLIosIKZdvIggCn376KUOGDGHhwoW8+eabnDt3jnvuuYfHHnuMXr16sWLFCi5fy5UkxqIIW49ncjW3QPadY21k6tSp9OjRg7lz53p7KBpO4O/vz6RJk/jhhx84fPgwUVFRPProo3To0IE333yTtDT3fLeV0J5ILn/hc/9bz96/PUrGN+8QdXETc7uIDIiqj0GnLPVTm7Wj4oXPUyEDu8mqj48P69ev54EHis2RAwMDeeKJJ0hJSeHVV1/l888/p/ekp7BYpK3oBGD1T2my7hxrK4Ig8PHHH7N9+3ZWrlzp7eFouEBkZCQvvPACycnJbg+FKmHHV+wvLG16NRp0vDz7dyxbtoxhw4aRmZnJ6k/eZcOyf1BUIL3BsZyoxdrRjuKFz1MtiWoyWdXpdNx///1s3ryZIWOnYJX41uVbbBxKzZZt51jbqVevHitXrmT27NmcOeOeOk4N91NdKHTGjBn8+OOPkkOhSvDrHN9DeshPBCb1aUX37t2ZNm0aCxcuZOGS1YQMmo4gk+OKHKjRW1fxwueJlkTOHsrq/eU5wD2RIf0A3r5zrAvExsayYMECJk2aRGFhobeHoyGR8qHQ1q1bM336dNq3b88bb7xBamqqS9dVgl+n3V/Y1fO4qkKHnkj0cxRBAH8ffZVZ6kpG8cInR8igJpw9lA30kWc8+ZVkb7pyjZTL3s1g8yRPPfUUYWFhivJi1JBO6VDov//9by5cuEDXrl259957+eKLL7h9+7bD11LCGR8Ut6QyGvQu/W1loUNPJPo5gk4oticb1ikM84w41YkeqKA7gydaEpVeWX3zzTcsWLCAgwcPlnRxtlgs/PTTT2zZsoXNmzfzc0FjAuNMoHe9B5XRoJNN0NXWBFIKgiCwZMkSunfvzpAhQxg+fLi3h6QhI/ZQaFxcHIsWLeKbb74hISGBJ598kvHjxxMfH0/fvn2rrdErf8YntyGEo9eTuy2XJxL9HKF140DMM/qqJpGlMhQvfJ5oSXTlZj7Tl+7j0tlT7Nv0DXkXr5CYmEhmZiabN29mx44dtGjRgiFDhvDkk0/SpWdfhv1zv6Q2IxabDd3NDCBI8vjV1gRSKo0bN2b58uVMmjSJAwcO0KxZM28PScMN2EOhkyZNIi0tjeXLlzN9+nRsNltJgXzz5s0r/J1d+Ko3hEhn0aYTThlCuHI9p1pS1eDupJTawC4RIaoWPdCcWypgKyoAQcCaepi7Gt5mzMBYBg0aRFhYWJnH9X5jE1duuphUIorYbmcTbU3lQv0uWETXU5PVZAwrN3/+85/ZsWMHGzZsQK93LaSkoS5EUWTv3r0kJCSQmJhIz549S7xCAwICAEhPT6f7xKcIGTRdNitBqdaEcrSkmr50H1tSrlT95B6gtsw3qhA+8EJ7D9v/t3fmcU2d6R7/nXMSdgGRVa3FDQUVNzqDS9EWqa1iO/V61SpTtQte6mixvdPW20Vvp07tdKpdZtRpx9at0zLttWoXp5aidKpYERFUwAVEQdn3NSQ55/4RgwGSkOQsOUne7+czn88Uw8kb4JznfZ73eX4/Fp7uSqM3RG2rCtO2/Mgr46PB4VBKLBZ9lM1rF+dIViBCo9VqER8fj4SEBHLm54J0dHR0l0JPnz7dXQq9woXi1a9yQSs9+r/IbcxJCfKRJnxwfFiPsqi7gkaHWgsvJQOVhrXKkkooIWk+OMvzxmECH2D5rktIjN0QW4+cx18yr4GleGQZHAdtQznCAnxRAz+bJIMczfxRDMrLyxETE4Mvv/wSM2fOtPdyCHbi5s2b2L9/P3Z9lY6ue58BFG5WX8NTySAtObZHxsWn2kRTAENToCmqx+ZWP3s+MsgHr86PRFxEsEXXs7f+pzM9bxwq8AHmSwZiYXhD3LhxA/e+8HdQw38tyLVpTgMtC1CM9cet+nUN9vd0aTeHb775BmvWrEFubi4CAgLsvRyCHUnedwY/FFQJtpGUwlooMmwAtjwa3e85oxCVJj4Y2xg4Kg4X+PToXaw/OVGKymZxVQz0N8S6KV649957wdy/Bl6jhAl8gF4Znbv9/yzDU0ljxbRwlNS2ETcHAOvXr0dpaSkOHDhALIxcFCGsgAxLeVJaCyloCpsWRPU7GsCrt4AH5krBjojs5/hMMcjHHavjRiJ2hPg7fI4DfiyswsRfzUB9fT3YTmHFknU7D8se1vqh0TmRIdiTdZ24Odxmy5YtuHHjBnbs2GHvpRDshBDt/oaCEFKOD2hYDn/4psDsfVrbqkJDu7TCDY48pG4O2Y8z9IduwL1S9F0ZTdNIXPcGlFczcbKhDqxaBdoOskGRoQMwa3QQdmeVWnTYbujmAMCp/ngNcXd3x+eff47p06djxowZmDhxor2XRJAYIdr9DQUhpB4fUGk5bP6uCNFD/Y2WE7/MKQdNCeWcZx5LO00dFYcPfFIMuAO6MuLFylY0ZWYicdFj+MndHVI1mBpypaoFxdVtUGltc3MwdVM5A6NHj8a2bduwZMkS5OTkwNvb295LIkiIULq+TbcFIaTSCTakQ60TnjfWQCJmIFbQFMaEDECon4dVnaaOisOWOvXw1cSzBnXoeDz89jdY99Im3Dc2xC6+WF1aDiri5mCSpKQkxMbGYu3atfZeCkFihNL1Pfb9t9i8eTMYrX3E300Jz4sZiDUsh4iQAdi14h5sWzIJq+NGOm3QA5wg8AH8NPGsgQOQXlSNpR+dwohAb0neszcURcHWiOsqbg5/+ctfcOLECXz66af2XgpBQoSyAlo2bxbKy8vx9b6/Aaz0WZ8p4XmxBftdSfrQKQKfXhPPQyF+CqY/M9uTVYo5kcHwFEiwWipcwc3Bx8cHaWlpSE1NxdWrzp3hEu4ghBUQy3F47jfTsWPHDrz8yisALf3m1pTwvNiC/a4kfehYT20zJMWGIy4iSLL361CzSC+sxsrp4XBnKN4eYlLhKm4OkyZNwsaNG7FkyRKoVM6d4RJ08D/24NB29TTeev1V/C2jEH85XgJrRoyEpLlTjbq6uh5fEyKwm8JaazZHx2kCHwB4uUnbq9Op0aKwvB5lu5+Duua6wwQ/VylprFmzBnfddRc2bNhg76UQjFDbqsLOzGKkpuXiiT3ZSE3Lxc7MYl6leD7HHp5KBfa+mIQrrQr88V/2U0gBgLbGWgQGBuKZZ57p/pqY/QzWWrM5Og7f1WmI1F1YHAdkXW/GuieT0NFZgUPcUICS/4/UVUoaFEXh448/7rYwSkxMtPeSCLDN5cBS+FgBrZh2N9IuNCLPfwYoO+5hWbUKxw58CQDYsWMHfHx88Kc//QmALrBnFFVDI+Am25TprTMj/6e0FUjh1t4bCsDw+MewOm4kOiSQN+KLG0O5VEkjICAAn376KRYtWoScnBwMGTLE3ktyafrT29VLEB4tqMJPl2v7dU0whv71rx44BzBKs1NvejeFOZHB2JN1HZ1qrQRTcuahaRq12d92//fbb78NhmHw5ptvArC5t80kxkxvnR2nKnVK4dbeG8MzszWzR8GdkfePVKVS4ev3XkZmZqbDlGb5MnPmTPzud7/D8uXLodWKb21FMM4dl4P+ReYNhRdsUR1Kig1HZMX3iPJVw11Bw6PXc0FvBD03KgQrp9+N9MIq3bqsfidhoQBEDeTAdjT3+PqWLVtQX1+Pvx6/Cg0r3CpNmd46O/J+SluJmIe/5iioaAIAjAnygPLCYbBqcbVDbYWigITxg5GYMBurV6/G1KlTsW/fPnR1SSuDZA82bNgAmqaxefNmey/FJckra8Tm74qsthXTCy/klzda/Z6jBiqR4FmKky/ej/UJEXh00hDEjw3Go5OGYH1CBE6+eD9SZo3C7pPXpbM76wcOQEOrCu5hEfD09ER4eDgSEhJ0Brxu3si8XCNIRclZpcgsxalKnVK4tRvjanUbcq/X4bfzZ6GgoADeE69jYPyToBg3ULSwewt3BQ1wsFq5BdCVNNbFj0H00F8jOTkZR44cwbZt2/DSSy9hzZo1WL16NQYNGmT0e2tbVQ7tAMEwDPbv34+pU6di9uzZiIuLs/eSXIq/Hr+KTp7CC9ba4YSFhaGioqJb19cYG746b/O6xKKSG4DhT73Xp8y7M7NYkOszFJAQFeKUUmSW4lSBD9CVG/99pVYSt3Y9Wo7DO0cuoLKyEj4+Pmg/fxRc/XV4Tn0EA6NmgqaoPo7Lai0LDoA1VQu9QjoAG40x75Q0aJrG/PnzMX/+fOTn5+Pdd9/FqFGjsGTJEqSmpmLs2LEArGtEGDJQ3vZIgwcPxq5du7B8+XKcO3fOZJAnCEttq4pXpmIovGDN31FoaCiKiopEW5dYmNLXFUqybO64UGxfPpX3dRwZpwt8tnZ18eWX8jZcKr2JQT7uYFkWNTU1qKmpgW/QYKz5817cPWkGOjS6jsrBXhya845i/08FwMTfAAol+psXoilgdMgAtKq0+M/bJV1LTHn1h/cvzxuLB8eHYWdmsZHANAYff/wx3nzzTWzfvh2zZs3C1KlTMfk/1+H/rnFQaVjzjQgXq/BDQRUoAAqGFrRLT2jmzZuHJUuWYNWqVTh06BCxMJIAIV0TTGVuxggLC0NlZaWo6xKT3vq6QnWtt9hBg1RuMJs2bdpk70UITfRQf/h7KpFVUg+tRNs5lgOu17VhwcTBoCgKPj4+CAgIwKJHH8YPn32IVXMmY8ogFum738E/3n8Ddw8Nwx+fX43fzpmKxnY1yho6oKQpkwfXHICqZhXOlNZj14lroGkKT88cDoqC0e/1UNBgaApzIoPx9MzhyLhUg9cOX8SpkjpcvNWMa7VtKKps6b7ehVtNiLwrGEsfeQhr167FqTolDlwDNBYeA3O3/6fttX4Ny0HLciipbcPBc7fg76mwe3ll9uzZeO+999DZ2Ylf/1o4X0WCcfb/ch0XbzX3/0IzaFgOQT7ueHB8qMXf09rain379iElJUW0dYmNluPQ0NaFxOjBOHapGkWV/MUnyhvaUVDRjGEDvRDq5yHAKh0PhzWitQRzbu16243Jw/xxuqQeQuSGDAX878PjkBQbDo7jsGzZMhw8eBCdnZ1gGAbTp0/Hk08+iUWLFvVxDtAb6x45X4H8m039lkANM7mHxofhy7PlKKpoQXOnuoe6+pELFVZnhhOG+GPpR6dEKRfLxdCyuLgYsbGx+P777zFlyhS7rsXZeWJPNjKKqnlfJ35sMHatuMfi11dWViI6OhrV1cbfW6h1iY2C4nAsdQa+LazHtnRhBusN73d734v2wOlKnYZED/XHzqSY7qBiLDAM8nHH3Hczcamqlff7aTngjW8LET3UH6/9biUOHTrU/W8KhQI7d+5EVFSU0e8d5OMObzcGl6paLDr3630OYKwEdKd93DrfvrsHeYt24C8Xe6SRI0figw8+wNKlS5GTk4MBA1xntlFqhJqvtVZ4ISgoCA0NDVCr1VAq+36vPeZ+bUGr1SBm8e8wd7QfNIEPQYhmfFfx6TSFY/zmeWKuqwsAIsN8BQl8gC6rfGrvGbhx3hg7dixUKhXq6urQ0tKC9PR0k4GPb7t370DC53pClFPMYWuXntAsXboU6enpeOaZZ7Bv3z67rsXRMdf1K4RZtC1akgzDIDAwENXV1UaFC6QyseYLRyvx6Mo1GFlzAj9kn4Hb8BjBusXlshGVGqea47MVoQffq1tUqI1OwsufHEFJSQmamprQ1dVl1iNOiHZvoa4nNnKyR3rvvfeQk5ODvXv32nspDkleWSOS953BjLcysC39Mg6eu4WMomocPHcL76ZfxvS3MnDqWh1YnicqtmpJmmtwWTR1qOw6Ok2hppR46qmn0HgyDZxW2LlbV/Dp7A0JfBBn8L1Tw/ZQnVAoFCY7CIVs9xbielIgF3skb29vpKWl4fnnn8elS5fsvRyHYv+pUiz96BR+KKyCSsP2yZw6b38t83JNn6Yna+CjJRkaGoqKioo+X88ra8T/fHUeGlbe2Z4eXw8lKisrMTrADUFl/wbNCdeZKaeNqFS4RKmzP8QafO9dRjBVDmrv4v9HbNjuLfc2bUBe9kgTJkzAH/7wByxduhRZWVnw8HDNTjdrsPb8mM9txUdLMjQ0tE/G159eqNzQl3kjI0ciPz8fgPCfwZZxEUeGBL7biKF6DujKCH88UghfD6XJIfAuLWvVILvx97kTSIQadBUbOdkjrV69Gunp6XjhhRfw/vvv23s5ssbW82NboFg1XnhgjM3nT3r1Fj3WBGzzcLqITolfNDMs8xpunscN9sXNxg5UNPGXSJTTRlQKSOC7zfmbjeL4XHHAqZJ6k//eKWCA0gcSqe2ZbEVO9kgUReGjjz7C5MmTER8fj0ceecTeS5ItUpwfU5ROni+s4iw+e307lhw8CC8vL6uvY6jeIkTA1o8BrJx+N3LLGnH6Wj3vTWt/73ffmCCUN3Rgw1fnjW6eKfDLqPXIaSMqNuSMD3duCLXWAeoeZtAHEkdp0/bzktc6Bw4ciM8++wzJyckoKyuz93Jkidjnx4auCf9MnoYfdryGkJAQJCYmoq2tzerrGTa38A3YNAXMjQpBWnIshvh7Iq+s/3lbvngoGAwf5G32LFWoJchpIyo2JPBB3h2Q1qA/KxSmS1XcO5rjOHxz/BezWor2YNq0aVi/fj2WLVsGjcYxMmcpEeP8mKEojAnx6eGasDMpBtFD/aFQKLB7926Eh4fjoYceQkuLdeU4fXOLEAFbQVPY/JsJyC9v7LZXEhNPJY05kcHYnVVqkZUTH2wZF3FkXD7wOUIHpKVkXq5Bfnnj7TZtnu3jIv88KIpCg1sw4h6Yj8TERBw7dkxwf8DaVhV2ZhYjNS0XT+zJRmpaLnZmFvfbvfbCCy/Aw8MDr7/+uqDrcQbEOD/Wchxu1Hdg1YxwrI4b2ad7k2EY/P3vf0dkZCQefPBBNDfrZMYyMzPB9tOVqc/4hAjYNEXh/Ywrop9vciwLBcVixbRwpBdWS3KWauu4iKMir1qTHXCEDkhLUWlZbD9+FXPjMCR0AAAXAUlEQVQ8StFRUgDm7inoT/zaKCyLiGBvlDepRL3plAoFXvvkW7gV/4SUlBR4eXnh+eefx+LFi40qbViKNY4SxkSzaZrGvn37MGXKFNx333247777bF6LsyHW+bF+luyN30wwOQi/Y8cOrF27FgkJCYiLi8Of//xnZGRkmP39hISEoKKiQpCA3alhcbSgSrzqEKsFBQ5eDcVYMmEgSmrDJKlE8RkXcVScWqvTElLTcnHw3C17L0MwKE4L5bebsPHt9/HGyVabyjEeShr/TJ52u6Qjbtv3o5OGYNuSSWBZFt999x22bt2KK1euYO3atUhOToa/v3XdfJa2eVuiVXj06FE88cQTyM3NRVBQkFXrEAq5+SCKeb/QlK6cSFFUr82KTld39pggpMwaiZTF83D69GmwLIukpKR+VXcGDBiAR7d9j5+KGwRZoxjnegHeblgdN6JbRrG2VYUZb2VI0p3tqWSQlhzrUsotLp/xOUoHpKVQANZt+weWzxkLzt/61m29iLR+9jB6qL9JoW8h0HeS0TSNxMREJCYm4uzZs9i6dStGjBiBxx9/HM8++yyGDx/e77Vs1SYFjGsVPvDAA0hKSsKKFSvwzTffgBbYVNgcfLNWsRBT5ovlgC6t3ufjDt32VwVV+LGgAjVdAd0lzgMHDmDXrl1wc3Prfn3vzULggv9GeYP1jTFS4uXG4Jdr9SisaBZsttcSevt0ugpOaUtkDUJZfcgFjqIR4ueFB8eH6poDaAqnSvpvuabAwVOp6OOcEOLrgdgRg6BkKDR2qMFQFLzdGTC9duW2Mmmofx+rmbCwMCxcuBDLly9HTk4OUlJSkJ2djWHDhmHoUOPnEHlljXg27ZzVpVkNyyGrpB5xowMR4tt3cH3WrFnYvn07WlpaMG3aNKuubSv7T5Xi2bRzuFzd0m3r1HvN9rJ6Cg/0xscnrvFSYuEDCwqe4RNBdbVDW3MNKpUKY8aMQXR0NPLKGvHa4Qt97LfgPxjNnVre7VpiZXuAbgNuaBWWVVJn0qJMCChKl+nJwSnFHrh84Cuta0d2ab3dbmQxGOLvifBB3njt8AV8nq1ryzf18TiNGm4KBglRofjTf0QjIepOEDJ8kORcb0BFUydaVRq0qbTQshzvh4CHgsaCiYMRc3eA0X/39fVFQkICUlJSUFdXh9///vf44osvMHDgQERERPTIwF47fAGXq23bwBh6nvWGYRjEx8dj5cqVmD17tlGx497UtqqwN+s69v9yHf/MKcexS9UorWvH8EBveLmZL7JYO2CtC9x18PdUShL8vNwUOH+zCSW1dsygaAb+Eb9C6tK5KC3Mg7u7OxoGjTO7WRDi7qYosXuddWhYTrD3oXut2dCns/f97kq4/BmflLV0qZg41A+Xq1r7P5vjODA0hRfmjsHqWT0loaSQdXJX0Dj54v0Wn1VpNBocPHgQ77zzDqqrq5GamopVq1ahE0rev0Nza6ltVeG1T77DdyfzEDdnLgJ8PIyes5kvT945pzJVnswra7TZB1GKcxp9CTGrpBb/vlIr+gybOajbM3U7k2IEVGMxDceyCPXzQFWLsALRYjMuzBcRIQOM2rG5Mi4f+AAged8ZQXU6hVJSsAUFBYAy7eRujN7msFI8SAwfXLZw8uRJbN26FcePH8eMpzehkBlx+3zINjwUNNYnRPTQKjQMZBzH9bg+Tel+z3ERQUiNj8D5m5Y1AplrquH7dxjk44YZowIFb34xF9DtibuCxke/jcHq/Tmiz9Rx6k5M8VehQDVQNp+/P4z9TRN0kMAHfjttQ/S7+vvGBKGhXY3TpfUOMx+ozxg4DqK5rxt7P74ZSnFxMR7fmYGbir5lSmvRd5gCdzJeS34OFHSB0Jq423uzIWTlwZLs0lLkLOjsoaBxV4AXrta0iro2tqsTDRm74NN4Fb5J7zlM4LO2ouJKuHxXJwBMvMsfL88ba3WWo6ApjBvsi0Af9z5lhMzL1cjZ0yC46LVY6OeoWA6izw4J2Uk2cuRIjBlfj5tF1byvpe8wtTbj5WBd0AP6OncIOU9q2AX50+VasyMb5pAi8+dDp4ZFsYhBjwKgoDgMbT6HsyUnUNncjNiBQG6NOO8nJK44m2cNJPDdRv9g4Fuuyitr7BaTdaRkmuOAjKJqAJR4DxILZudsQShtUl8PpaTOA4ZO9GIoolgysmEKKX8O9obpla0bVm6emT0K0UMTwX34Oi5evIivb9DIrblmt7VaCh8rJ1eABD4DkmLDzc6t9b0hemYsci4LWYKW5QRxqOj/QSJsA4YQs2V6rcK/Hr8qeplXj6EBqJjzpL2zS0twFP1aIRpsIsN8MTrYfAMIRVHQ+g3F3lOn+L+hyLjqbJ41kMDXi+ih/tiZFIO6VhW+PFuOoooWizqi5F4WsgTd7DD/J4klDxIhWTR1KLalX+Z1DQ7A/WOCsfUHftexFr0BqNiOGobZZX8IpV8rdpOXUHN1Ib4e3We75pD7ZkCsioozQgKfCQb5uFvcDeVKZSFL6FRrwUE3i8SBEz37DfRxx6yIIJs7IvXnIT8WVUtentYbgIqpiAL0zC7724AIcd7oxlAID/TG9br2PpUTdwUFlUoNMAxs0pIVmNp+RMv1r5GrmL3YFRVnhAQ+AZD7TlBqSmrbcLXmzoCzFBJba2aPwr+v1NpUplSAQ/LMcOz9pYzXSIStNHeqBcla+0OfXfa3oRPivLFLy2FcmB8+eyrWaOXkXznFOFvVxau0PirYB9fr2nmv9eKtZuw/VWo2S5KTmL2CpjAmZABC/TzIbJ6NkMDHEznvBK2FoXRnGXylknp/u1BdhuawtTOXAuBVfhqP3JuCESvfBiC9J5mvh5J31moJ+uyyP8obOgR5v+ZONfTV8061BqV1bVBpWOSWNeB6nRoUzwPldfePxvNf5PFep4blus9AB/t7GhUFzytvkM0Yg4blEBEywKLyLME4JPDxRE47Qb6wHMDbv9YMfLoMLcGwM7ejSw1Qln2YrvDpSHknAV/9nAd4Sxv4DA1A+WStlnKypBZP7Mk26fKw/1Qpcm/wdzEAgIKKZkzb8qMg8na9YSjgZmOHYJuFTrUWT+09g6YO3UiLYZCjqVt2Vakxhn70hmAbLm9Eyxcx2tDtBYfbnZ0iv4++yzC/vFHwayfFhmPBmAEWn9Vx0AXjj/NaMG7yryQ/cTI0ANVnrZ5K8W7LqmYVMoqqcfDcLbybfhnT38rA6v1nkFfW2N2gJVS1t6KpE2qt8EEP0DVi7T5xDQ9EhsBDwfC+HgegukUFlYbtcz/LLegBuioBwXZI4OOJs9kasZw0cmv6LkOhyStrxP+drwVFW/cw1LAcjl+ukVRqztiQcVJsOF6eFwlPJSPIaIk5Om8/5I8WVGHxh1nY9HWBQzVoVTR14pXDFzAnMljUSoXcMKwSEGzDhf5cxEHsNnRnxbDLUEje/aEQGs4x/qxNDRknxYYjLTkWc6NC4K6g4SHyU53jdKU9MW1wRIGi0KlmkV5YZdRSylkxrBIQbMMxnhAyRteG7pw/RpqCqFmHWsvi+S/zBAt+ta0q/HSlFpSEhrG20t+QsX6e9OSL92N9QgQenTQEQT5usJ/8uXzpULOoaOq09zIkgUiRCQMRqeaJEOLC9nRzMIeSoTBzVCBOFtf1mcUSaniYpgAlQwsy6rDj2BW89d0FQOHW/4vtiM4A1PrO1ryyRizccUKwMzjngoMcZgLFRgr7KVeA1Ol4IkQbOkUJIpgiOAxFIXbEIPx50cQ+s1j5NxtRXMPfjJS9XWYTYtQhPadQ1kGPpoAHokJsHjI+f7NRl4HL8G/F/ty+icQ+GLUBBa0bE1Lz3LEQKTLhIIFPAHgNT9MUaAp2GZzuD/3clzEVm9S0XEECnx4hRh0KrlwDAoQW5hUmk1DQFD5ZcQ/ujQiy6fv16kBO0kAsCnLL+QwlxADLBPD7uw6RIhMGEvgEwNbhaU8ljYiQAcgrbxJxdfwwNS8klsSWtYLK2dnZOHfuHMaPH4+2xlrQggc+Ct5uNDrVLDjYVt51Zyi8mhhlc9ADiDqQJVAUBY5jQVk4vykWpiTEzAngW3MdAn9I4BMIW22NMi7J29zL1LyQmBJb1ggqHz16FBs3bgRFUYhcuA6dDCV49tzWxcJDQWOUtxo3r5cgeMxkXK1u7fesjWNZQNuFxp8+w7j/eBlAuE3v7xDqQCwLyKGpiOPslvYN8nZD3OggkxJi5gTwhwV4AgBu1HdIJu7uypDAJyC22BqduS6MSoYYmJsXElNiyxpBZV9fXygUCqhUKlz61x6EDpsJSoRzvk4Ni/N1GihL87F4WAeeX/goDhS14vvzNwGOA8fc2SDof8/1F0/g81efQOr3mzB9+nTMmTMHzz33HGbPnm2VXJcjqANRNAVOqwYY+w5WjxyoQEmdyi5nvXGjgyySEbNGAJ8gDiTwCYy1tkZilAw9FDRYDtCwLK/OS5bj0KbSIDUtt4duYfzYYPxYVA2VhhWtI1WtZfH4x79gwcQhfWS1DPH19QXL6n52Ck0HJgQyuNgkTrMQpXBH17j52PT+S/h66lTsTHoAsx7YiNwmD7gFhYP28IaXksL6lMfxyIQQ3BX8MOIP/BEnTpzA5MmTERYWhpSUFHh7e+O5557D4sWLoVT2Hyjkrg5EUcCs0cE4XnjLruvwUNBYMj0C3m4MNn1dIOlcIhkqdyzIOIOdEWIcgqGAueNCodKwPQLshq/O88rI9KMG9nzo6rMnU+MOX3zxBRYvXgx/f39kZGSADhyOpR+dEk3vkmNZDGy/gXMfrAEAHD58GMuWLUNbWxu8vLxw5MgRxMXFobW1FSEhIWhr0zUAnTlzBvPmzUNWVhYKCwuxdetWXLlyBevWrcPTTz8Nf/87n+vcuXOYOHFid1b4xJ5sZBRVi/J5hEDfYv/S3mMoaFZYrJEqNO4KGidfvB9HLlTgD98WSvp3q39vUpZ0DEjGZ2eE8JJLiArB9uVT+/wbX9Fj/aiBPel2drhYhR8LqxE12BeBPu7d2WeAmzdCQ0Nx9uxZhIWFAYBNjUaWQtE02v2Gd5dh582bB/r22daCBQsQFxcHAOjq6oK7+52HYExMDDZs2IBly5bh559/RmJiIs6ePYutW7dixIgRePzxx/Hss8+iq6sLkydPxqZNm7Bx40YA8lYHMmyxf2NZHBb+9Se7jZTcHeCFi7easPm7Ikn/bslQueMhg9NowprZo2wW2jUlewVII3osFRx0epr55U09RJZfPUPh4be/RrXGs/u1er1LNwaiGMvSFIUvz+rO3RQKBRISEsAwDD744AOo1WpERUUhOjoazc3NiIqKwpNPPgkASE1NRVBQEF555RUAwJQpU7B//37k5+fDzc0NMTExmDt3LmiaxltvvYVDhw4BkF4dyI2h+h2Hoyj9IH5kd2PXlPBBmBvcBmi7bHpfhqZ49aWU1rVh5e5sUd0tjGHuHiTIE1LqlAl6ZXxrxyEMHzzmr23bDJEjYGrO6fvsIqzY9hXch03g7f3Wm0cnDeluZCgrK0N2djYWLlwIQBfQcnNzAegC46pVq/Dhhx8CAGpqajB58mTs2rULc+fO7XHN0tJSjB49GhqNBrSXH/wmzcWC3/4XWKUXfr5aK9nceuyIAFAAzt5o7NOg5UYDXWo17osajGEBXmjqUHef/7q11+LwuxtwkwnGwPufBM24WdXpyVBwOFUaS+9BgrwggU9GWBqgbBlozS9vNNtt2qXl1wgjB3o/hDo7O+Hp6YnEF95DoWKkoMPfg7zdkBgd1uPBr/e3O/3vDCxatAjt7e3w9PTEtWvXEBIS0v29x48fx2OPPdajPFvbqsK6bf9AxvkyuIWOAOXpB3AsKEb6Mqf+7+vZ+FGgKKpHg5avpwLX69uRVVwHoGcpnOG00Gi1aC8+g47LpxA0JR6KYZNM+vE5sggNGSp3bEjgkxn9BSi+A63Guk3vCvDEzswSdGnl2zloKQwFTBk2EEMGemJsqC9iAtSImTAW+0+V4n8PX0AXC8GzPz3638+siCD86531KMv7GS+++CK2bNnS57UbN27EiRMn8PbH/8S7RwuRVdoELcvJKuPpvZGwdGPGsSw4bReY/MM4c+BDfHm2HHlljbhc1QKVhoW7goa7ksGlymZZKtGMCPRGp1qLymad8LVh0CZD5c4BCXwyxdJxCCHYmVmMbemX7d7IIjS9O0IvXSrCc//MBRNwl2jBD9BlAwzHouboTtzI+Af8/Pz6vEaj0eCex55D04h4aEHL1lFC37GZX95odSmeU6uwedGUPhlRXlmjqJ23fIkfG4xdK+6R9B4kSAsJfASkpuXi4Dn7zmCJib4s9fuEkfiv+bEIW/13UYbce0NTwPghfpg3PqzPLOLun4vx+tcXwNLy7dgEdD+7Xw8PQF5Zk02BykNBY8k9d/UoCV+81Yyr1a2yLXMant8SnBMS+AiynxMTCk8ljeqjf4NiyHh4RUyT7MGr65KkujNPAFi4/SdoYVsnr9TwtaASysJKCjwUNNYnRBBlFSdH3ttNgiTIeU5MSDrULHxnrcSCsDb8UM9IVmrTaYdy3dZL4YFeYB0k6AH8g5ajBD2AuJu7CvI8WCBIijO7yPeGo5Vo8R9tl/lGvfVSYUWLbMt8rgwZRHcdXONpRzDLoqmutcM9dqkGPxZVY8W0cHgqHSfzIogLGUR3HUjgI3TLpsnQvFo0jl2qwZ6s61g5/W4E+cjXtZ0gDcTd3LUggY8AgJ9smqPSodZi98lSDPb37P/FBKfEmPQawfkhXZ2EbmyRTXMGHFEqi2A5NAW4MbTgYhAEx4UEPkIPutU51FrSgEFweDyVDD787VQUVDSTQXRCNyTwEfqgl01LL6yW1MyTIB6umNUSAWmCKUjgI5ikrlWFDzKu4GhBlVHdQoJj4KGgMSsiCD8WVclSG9NSlAwFjgO0HCe4iDvBtSCBj2ARxnQLfT0VSMsu63F2QpAfenfwJ/ZkI6+8yd7LsRrDQBY91F9UEXeCa+Aakh0E3gzycTcq4zQ62Mfqhhjdzp2zKvvwUNKYNToIXm4KlDe04+yNRmjJnq1fDIeyHe08y1Qg25kUQwSkCbwggY/AC30pyVofQVu+x7Bs5egdqBQFRIYOwOWqVlHPUQ2HsoWSphvs54G6tq4+GRdfTc4xIT6ICvOzKJCZ2ogRCJZAAh+BN0mx4TaVoPiUrQwDrlztbczhoWCwZWE0GtvVWLn7tChnp72HsnXSdJW87Kc8FDRWTA/HoilDBS19eyoZvL1oIilPEiSBnPERBMWWEhSfslV+eSO2pV/GsUs1YnwcUejdbZi87wyOFlQJdn1TWXJtqwoz3srgFfj054Wmfi+2ZOKk+5IgNSTwEZyClZ+cxvHL0gc/fcnyWm27zWVboYxZLWnuSN53Bj8UVpldpykoCpgbFYKdSTFmX2epUzvpviTYCxL4CE4Bn+DhztAABZsyIb1DOQBe3Ya2ZEoKGhg/2A+DfNwtzpL5/Jz0n9WScqR+FpR0XxLkCAl8BKeBT5kNgCAlOj5lW6kyJSnLkaT7kiBHSOAjOBV8goccSnRSZUpy+KwEgr0ggY/gdPAJHnIp0UmRKcnlsxIIUkMCH8Fp4RM8XKlE50qflUAASOAjEAgEgotBjGgJBAKB4FKQwEcgEAgEl4IEPgKBQCC4FCTwEQgEAsGlIIGPQCAQCC4FCXwEAoFAcClI4CMQCASCS0ECH4FAIBBcChL4CAQCgeBSkMBHIBAIBJeCBD4CgUAguBQk8BEIBALBpSCBj0AgEAguBQl8BAKBQHApSOAjEAgEgktBAh+BQCAQXAoS+AgEAoHgUpDARyAQCASXggQ+AoFAILgUJPARCAQCwaUggY9AIBAILsX/AxMo5iC9e14sAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# finally \n",
    "# we will try to visualize some predicted graphs from our best performing model i.e. random forest \n",
    "# for the same timestamp as before (10) and both for the illicit and licit transactions \n",
    "\n",
    "# we remind that  \n",
    "# 1 means the transaction is illicit / illegal \n",
    "# 0 means the transaction is licit / legal \n",
    "\n",
    "# we start with illicit directed graph \n",
    "\n",
    "illicit_ids = df_class_feature_Class3_ID.loc[(df_class_feature_Class3_ID['Time step'] == 10) & (df_class_feature_Class3_ID['Class'] == 1), 'ID']\n",
    "illicit_edges = df_edgelist_Class3_ID.loc[df_edgelist_Class3_ID['ID1'].isin(illicit_ids)]\n",
    "\n",
    "G = nx.from_pandas_edgelist(illicit_edges, source = 'ID1', target = 'ID2', \n",
    "                                 create_using = nx.DiGraph())\n",
    "pos = nx.spring_layout(G)\n",
    "nx.draw(G, with_labels=False, pos=pos)\n",
    "\n",
    "\n",
    "# we show the growth of the graph in comparison with the previous visualized illicit directed graph because now \n",
    "# my unknown observations that have been classified as illicit are more than the original illicit ones \n",
    "# we are always refering to the timestamp 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Illicit case directed graph\n",
      "Number of nodes: 283\n",
      "Number of edges: 177\n"
     ]
    }
   ],
   "source": [
    "print('Illicit case directed graph')\n",
    "\n",
    "# number of the edges\n",
    "print(f'Number of nodes: {len(list(G.nodes))}')\n",
    "#number of edges \n",
    "print(f'Number of edges: {len(G.edges)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# we continue with the predicted illicit undirected graph \n",
    "G2 = G.to_undirected()\n",
    "pos = nx.spring_layout(G2)\n",
    "nx.draw(G2, with_labels=False, pos=pos)\n",
    "\n",
    "# we show the growth of the graph in comparison with the previous visualized illicit directed graph because now \n",
    "# my unknown observations that have been classified as illicit are more than the original illicit ones \n",
    "# we are always refering to the timestamp 10 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Illicit case undirected graph\n",
      "Number of nodes: 283\n",
      "Number of edges: 177\n"
     ]
    }
   ],
   "source": [
    "print('Illicit case undirected graph')\n",
    "\n",
    "# number of the edges\n",
    "print(f'Number of nodes: {len(list(G2.nodes))}')\n",
    "#number of edges \n",
    "print(f'Number of edges: {len(G2.edges)}')\n",
    "\n",
    "# as we expected the same as the directed "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb4AAAEuCAYAAADx63eqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAYzElEQVR4nO3dfWydZ3nH8d/znGMfO3EcN35L2hhC6ya2S5OpnVAKK22TEkbKkAqZQFqkMWmDAoJtGojStBovbVcGQptQUQAN2Ob9wcgaDUg70qZNqEZcRiqcNLbz0tatkyZ2nNSxT2If+5zz7A/H6XHql+P4+Nz3/Tzfzz/9o4502XH88/0813XdXhAEgQAAiAjfdAEAABQTwQcAiBSCDwAQKQQfACBSCD4AQKQQfACASCH4AACRQvABACKF4AMARArBBwCIFIIPABApBB8AIFIIPgBApBB8AIBIIfgAAJFC8AEAIoXgAwBECsEHAIgUgg8AECkEHwAgUgg+AECkEHwAgEiJmy4AcF1/MqUdB06o6/SgBkfSqiyLq2l5pf701pWqrkjM++MBFJYXBEFgugjAhHwDaLqPa1lRqdYXXtO+o2ckSal09vKf8T3J8zy9c9kiNSwrVyYrnRy4qP6hlJKpjDxPyub8y/M9KZC0vDKh2xtrtXxpmV4/d1H9yVG9eSGlQJ6qF5equqJ02pAkUIH8EHyInPaeAT2+9/iUgVUW95UNAq2qXqyKRFyvv3lR5y6MypOUyfmXEvc9pbNm/umUxjxJnt5ZvUjXVpXpQiqjkwPD6htKKe57U3w+uvyxMd8nEBF5BB+cNNPpJpCm/X9PvXRKjzzZpZF0RlH+zo/7nrJBoGWLS3XTtZW67fqaSUHI6RFhRvDBKTOd1kpjb53CYr6nsZwjWiLmaTQTiG/26fmetGxRqcpLY+odHJHnvf30GEi6c02tPntHo9Y1VJkrFpgHgg9O6E+m9MDOQ9rT2TvpkSOKz/OksnhM2zY3aev6VdOeDjc21WlPVx+nRliH4IN1cn+QnnhzWCcHhnX6/AinNcv4nlS9uFTnh8fedjr0LzXv+Fc08XBqhA0IPlhjpseYCJeJU+Nfb2yU5HEqRFERfFgwUz0Ca7hmkTxPev3cxUk/6KRA/7zneOSbTqIo5k3umOVUiIVG8GHOZuv4m+vJLe5Jab4LMYWYJ737uqWqrkhwGkTBEHzI22zzb4Gk62sX65W+CxrNZjm5oeA4DaIQCD68TX8ypX/9Tbee7epV39CohsfSSmcCjfDODRaJ+56+tGm1Pn1Ho+lS4BiCD5e19wzo0ac69X/d52RoKQkwZ80rluixe9dy+kPeCD5IklrbuvXVX3QYW8MFzIfvSTfUVqixrkKpdJb3gZgRwQe1tnXr739+mMFwhEri0iafusqErqtapJXXlBOGkETwRcpU3ZiVZSX6j7bXlDFdHFAEs+0oRTQQfBHAYDgwNU/jw/TVi0vVcu1S3XZ9NUEYAQRfyLW2dXMbAZCnuO/J9zyucQo5gi/ExkOvU8NjnPCA+WB+MFwIvhC5crnzi6+9KSIPKCxP0vLKhG6oW6KailJOgw4i+EKAd3iAOaUxT57ncRp0CMHnmCs7M/uTKXW8MahMlktWAZOuvKcQ9oqbLgD54VQH2C0IpOGxjB55slOSCD+LceKz2MTp7qmXTunQyfOsEQMcUV4S008/tV5rV/LY00YEn2FTDZUvLS/Ra+cuav/LZ5XJBqwRAxzjSfrgTfXavvUPTZeCKRB8hvDoEgi/915frbLSGLOAliH4DGCoHIgeZgHtQfAVGUPlADxJ9ZUJ3Vi3RNXMAhYdwVdE7T0D+sQP2zQ8xkpoAG/xvPEwvKG2Qg/d06z3r64zXVKoEXxF9PEf7NcLr54zXQYsEASBPM8zXQYsxeW6C4vgWwBTdWoOXBzV3qP9pksD4Ii47+mrf9LCPOACYIC9gOjUBFAo6Wygr/2yQxLD8IXGia9A6NQEsBBKYp7+6773MgxfQL7pAsLgrU5NQg9AYY1lAv35j17Q2WTKdCmhwYlvnujUBFAMMU+6u6WeGcACIPjm6ZM/+a32HjljugwAERHzpI3NdfqHe9cy93eVCL6r1N4zoO88c/RyIwsAFFPc97SxuY4T4FUg+K7CRCMLjzcBmORJKivhDsC5YpxhjrbvO65v7z4iJhUAmBaIOwCvBie+PLX3DOjRpzrZvALAStwBmD/GGfLQ2tatT/ywjdADYK2RdEbf23vcdBlOIPhmkTujBwC2CgJpT1cf8355IPhm0N4zoId3cYUQADeMZQI9sPOQ6TKsR/BNo71nQH/177/TCF0sAByyu6NX2/fxyHMmNLdMobWtWw/v6iT0ADhr/fXL9JU/bmbGbwoE3xW4IR1AWJQz4zclHnXmaO8ZuDSYTugBcN/EjF9rW7fpUqxC8OV4fO9xjaTp3gQQHsNjWT3yZJcOnhgwXYo1CL5L+pMpPXekj2uFAITO8FhG//TMUdNlWIPgu+SBnYc0liH1AITTs0fO6C9+8lu193Dyo7lF0mNPdWr7r18xXQYALDgaXjjxEXoAIoWGl4gHX2tbt37wPKEHIFqi3vAS2eBr7xnQ135+WNnIP+gFEEVRXmodyeBr7xnQn/1Lm8YIPQARFQTSc0fORHKpdeSCr7WtWx/b/hslU8zrAYg2T9KOF0+YLqPoIhV8rW3d+sauTqV5vgkAGkln1XVqyHQZRReZ4GvvGdA3dnUqxeJpALhscGTMdAlFF5ngu3/nQUIPAK6QiEcmBi6LxGe8fd9xdUbwOA8Asxkei16/Q+iDr71nQN/8nyOmywAAK6UieBtN6IPvy08cFK0sADC1/a+cjdwWl1AH34//9xV1neYRJwBMJ5D0jV92RCr8QrukurWtWw/+92HTZQCAE+K+pyc+816tXVllupQFF8oTX2tbt77+C0IPAPKVzga6/4mDpssoitAFX3vPgB55skuj0XtfCwDz0nFqSD/5zaumy1hwoQu+x/cej2R7LgAUwtd+Ef73faEKvv5kSs8d6TNdBgA4K5D08K5wh1+ogu+BnQc1lgllrw4AFM1IOgj1fX2hCb7Wtm7t6eS0BwCFEOb7+kIRfBMNLRz2AKAwgkDa09UXyvv6QhF8j+89rpE0DS0AUEhjmUAP7DxkuoyCcz74+pMp7Tt6RuEcwwcAs57p7Atdo4vzwbfjQPRuDwaAYskE4Wt0cT74uk4Pcs8eACyg4bFwNbo4H3yDI2nTJQBA6D3d0RuaRhfng68/JH8RAGCzTCB997ljpssoCKeDr71nQIdPnjddBgBEwu7DvaZLKAing+/xvceZ3QOAIjl1fiQUjzudDb7+ZErPdoXjtw8AcEEghWKuz9nge2DnQdHMCQDF9Uxnr/NzfU4GH3s5AcCMTCDn5/qcCz72cgKAWa7P9TkXfOzlBADzdjs81+dU8LGXEwDskA3cbXRxKvjYywkA9tjd0evkuz6ngo+9nABgj0DS/U8cNF3GnDkVfB2nBk2XAADI0XFqSM8fO2O6jDlxJvha27r1cl/SdBkAgCv83c9+b7qEOXEi+BhhAAB79Q2N6vv73BlvcCL4GGEAEDWBY+3r39p91JlGF+uDjxEGAFHkeZ7pEuYknQ2cGWq3Pvh2HDihTJbUAwDbuXJZrfXBt/+Vs0oTfABgPVcuq7U++A6/wUWzAOAKFy6rtTr4+pMpnbswaroMAECeTg/af1mt1cG348AJufV6FwCw40W710taHXxdpweZ3QMAh2QDqevUkOkyZmR18A2OpE2XAACYo8GRMdMlzMjq4Kssi5suAQAwR5VlJaZLmJHVwdewbJHpEgAAc3Rx1O6ndVYHX+/giOkSAABztO/oGavXl1kdfM8f6zddAgBgjlLprNXry6wNvv5kihMfADgokPTckTPWzvNZG3w7Dtg9BwIAmF4mG1g7z2dt8HWdHhQrOgHATelsoP0vnzVdxpSsDT5m+ADAbbbuWrY2+JjhAwC3nUmOWvmez9rga7iGGT4AcJ2N1xRZG3xspwYA99l4TZG1wddz7qLpEgAA82TjNUXWBh/NLQAQDraNNVgbfDS3AID7bLymyNrga1peqUTc2vIAAHmy7Zoia5Nly60rlQ2YYAcA19l2TZG1wVdTkdA1i0pNlwEAmAffk5pWLDFdxiTWBl9/MqU3L46aLgMAME9bbllpuoRJrA2+HQdOiCedAOC2G+sqVF2RMF3GJNYG31MvnVKaLdUA4LQH72kxXcLbWBl87T0DOnTSzuWmAID8eJJaVlSaLuNtrAy+x/ce50oiAHCc73vWDa9LFgZffzKlfUfPmC4DADBPGUvv5LMu+Lh5HQDCw8Y7+awLvq7Tg0qls6bLAAAUwNkL9t3JZ13wsZwaAMLDE0uqZ8VyagAIjwxLqmfHcmoACBeWVM9iy612rbYBAMwPS6pnUVOR0B2ra02XAQAoAJZU5+lzdzYq7numywAAzFM2kDauqTNdxiRWBt+6hiq9q2aR6TIAAAXQ+sJrpkuYxMrgk6Q3BkZMlwAAKIDdHb2mS5jEyuDbd7RPF0YzpssAABTA6cERq4bYrQy+h3d1mi4BAFBANg2xWxd8/cmUXj6TNF0GAKBAspYNsVsXfCypBoDwsWmI3brg6zo9yF18ABAyNg2xWxd8LKkGgHCxbYjduuBjSTUAhM+WW+xZR2ld8DUtrxRLWwAgPJZXlqm6ImG6jMusC74NTXW84wOAENnUUm+6hEmsC75nu/o48QFASHiSPr/hRtNlTGJd8NHVCQDhUVeZsOoxp2Rh8NHVCQDhURKzLmbsCz66OgEgPGx8c2Vd8DUtr1RpzMYvFQBgrhJx62LGvuDbcutKZXjJBwChsLrensH1CdYFX01FQnWVdr0IBQDMne+NXyxuG+uCT5Kuq+L2dQBwXcz3rNrYMsHK4Ft5TbnpEgAA87Sxqc66UQbJ0uBrWl5p5QtRAEB+Yp702TsbTZcxJSvTZcut9h2NAQD5u7u5XmtX2vd+T7I0+GoqErpjda08phoAwEmP3nuz6RKmZWXwSdLn7mxUWTxmugwAwByVl/hWvtubYG3wrWuo0rbNTbzrAwDH2HTb+lSsTpWt61fpXTWLTZcBAJiDJZavnrQ6+PqTKb3af8F0GQCAObBxW0suq4Nvx4ETpksAAMyBrdtaclkdfF2nB5VKZ02XAQDIk63bWnJZHXzczQcAbrF1W0suq4OPu/kAwB1lcd/abS25rA4+VpcBgDsevKfZ2m0tuaxOFVaXAYAbKhIxbV2/ynQZebE6+CZWlwEA7FZr+Xu9XFYHnzS+uizGzk4AsJpLr6Wsr3RdQ5U2NteZLgMAMBOHDijWB58kPXrvWsV9h76qABAxqTF3Zq6dCL6aioQ2NHHqAwBbpbOB6RLy5kTwSbzrAwCb8Y5vAaxrqNIf3UiHJwDYyPbF1LmcCT5JumaR3Xc8AUAUubCYOpdTwcfuTgCwjwuLqXM5FXzs7gQA+7iwmDqXU8HH7k4AsEt5ScyJxdS5nEoRdncCgD3ivrRtc5MTi6lzORV8E7s7PcYaAMC4d1+71JnF1LmcCj5pfJ6vLB4zXQYARJ5L7/VyORd86xqqtG1zk8pLnCsdAEKlsszNETMn2yQnjtaPPNml4bGM2WIAIILK4r6aVrgztJ7L2WPT1vWr9NNPrVdtRanpUgAgcgLJqdm9XM4GnyStXVml9zXWmC4DACLnrjW1vOMzhdk+ACiuuO85N7uXy/nEYLYPAIrri5tWOze7l8v54JuY7QMALLyWFUt03x3unvakEASfxF19AFAMiZivxz661nQZ8xaK4FvXUKWNzfWmywCA0Cov8fXQh5udfsQ5IRTBJ0mP3nszpz4AWADlJTFt29zs5HqyqYQm+GoqEpz6AKDA7lpTq59+an1oQk8KUfBJ4+/6ykvY4wkAhbBhTa1+/Mn3hOLxZq5QBd/EHk/G+gBgfspLYvqbu1ebLmNBhC4itq5fpS9uWmO6DABwlqv37OUrdMEnSffd0aj171pmugwAcJKr9+zlK5TBJ0lf+VCz4j5tngAwV67u4MxXaINvXUOVvrgpnM+nAWAhuXrPXr5CG3zS+CPPpuVu3hcFACa4fM9evkIdfJL0zY+uVSIW+k8TAArC5Xv28hX6RFjXUKWHPtys8pLQf6oAMC+e5/Y9e/mKRBpsXb9K2zY3i14XAJheWTzm9D17+YpE8Enj4Xf7jdzWDgBTKS/xQz27lysywSdJt11fw23tAJDD88K3hHo2kUoBbmsHgLfEPOmDLfWhW0I9m7jpAopp4rb23R29pksBAKNivqdffeF2NdaHe3RhKpE68UnjNziw0QVAlHme9IHmukiGnhTB4GOjC4Coi0r35nQiF3zS+EaX5pBvJgCAqUSpe3M6kQw+SXrs3rV0eAKIlKh1b07HC4IgMF2EKa1t3XrkyU4Nj2VNlwIAC+quNbX627tXR/qkNyFSXZ1Xmvit55EnuzSSzii6vwIACLMNa2r1o0++x3QZ1oj0iW/CwRMD+t7e43ruyBl5kkbSnAABhENZ3Nd/fvo2Tno5CL4cZ5Mp7XjxhLpODenV/qR+f+K86ZIA4Kr5nvT1j9wU+Xd6VyL4ZnDbY3t06vyI6TIA4Kp85v3X68sfajZdhnVoa5zBppZ60yUAwFVZXVdB6E2D4JvB5zfcqBhLXgA4Ju57+hi7iadF8M2gpiKhjc2c+gC4JeZ7ob9FfT4Ivll87s5GlZfETJcBAHmJyi3q80HwzWJdQ5W2bW5SeQlfKgD2i/oeznzw0zwPW9ev0rbNzSpjxRkAi7GHMz+MM8zBwRMD+st/+536hlKmSwGAyzxv/KS3bXMTM3t5IPjmqL1nQB//wX62uwAwrjTmyfM83bWmVp+9s5GTXp4IvqvAcmsApt20Yok+8gfXacstK2lkmaNIL6m+Wiy3BmBSy4ol2vWF95suw1mc+OZhYrn1nq4+jWX4MgJYeImYr5/dx9Lp+SD4CuBsMqUHdh7SM529Iv8ALJTxrk0ukp0vgq+Axt/98fgTQGHRtVlYBF+BcbcfgEIpi/sKJLo2C4zgWyC5d/sNjoypsqxEL71xXsf6kqZLA2ChmopSfXjtCp2/mL78M6NpxRK6NhcAwVdE7T0D+sQP2zQ8ljFdCgCLcEt6cbGDq4jY+wngSuUlvh68p5nQKyLm+IosdwaQkx8QXZ6kshIaVkzgUachuU0wCgKlcuYg/EuX31YvKtXZi6PK8jcEhErc93R3cx0NK4YQfIZN1QST+0L7bDKl7z53TDtfPKnzI2nT5QKYh5gn3d1cr0fvvZmGFYMIPods33dc//irI5wAAcdwwrMLweeYgycGdP8TB9Vxash0KQBmwQnPTgSfo54/dkYP7+rQ0d6k+AsE7BH3pFjMZ+jcYgSf4yb2hP6qo9d0KUCk+Z5083VLtfnmFQydW47gC4nHnurU9l+/YroMIJJaVizRYx9dy+nOEczxhcT9H2qWJ+n7z79C8wtQJHHf05c2rdan72g0XQrmgBNfyLS2devhXZ0sxwYWCIuj3UfwhVDucHwmGyjNERC4anHf003XVqqmIsHi6JAg+EJsYjh+/8tndfiN8+pPjtIBCuSJ2bvwIvgihJvigelNrApcXlmmTS31+vyGGznVhRTBF0ETj0Kf7iAAEX4xz9PN11UqnQ0UBNKyxaVaUhbX8FhG5SUxpdJZHmFGDMEXYRN7QHcf7tXpwREFgXgUilApL/G1bXMztx9gEoIPkiYvy+5PpnTuwqg8T1qciCs5kpbnSSNjGb3af4FTIqzneVJZnCt/MDWCD3MyqWM0k1Wa7x5YhHVhyAfBh6tyZcfo2Quj8j1v0uiEJx6dojhYF4a5IPhQENPdKxgEgb7z9DGNZhiox+w8T0rE/LwWMDBIjqtF8GHBcZUSZuNJ2tTy1vU9V/4ilYj7dGGiYAg+FM339x3Xt3YfZZMMJmHBM4qN4ENR5TbHeNKkR1oxT3SMRsDEoHhjXYUe2tyi21fXmi0IkUPwwYjp3gluuWWlTg4MTxuOU4l540001YtLddO1S3XbDdXacstK/exAj77NCXNB+Z5mvA3Ev/Tf5UvLdF1VuVZes4hHlDCO4IO1pgrHdywrlyS9fm74bYE51Q/S3BNmEAQa5Uh51RJxT3eurtOi0vikr/3GNXXac6RP7T0DOto7pFQ6q0Tc1+r6JVrXUEXIwToEHyLhygH9NwaG1TeUUjKVZuRiFgyDI2wIPkTa2WRK3332mHZ3jK9tk2Z+dBcGuWMAH2ip19MdvVM+VmZcAGFF8AGX5NtCP/Fob//LZ3Xo5HmduzAqqTjD+mVxX9lAWlWzSNcuLVfM95TJBjp1fliv9l+Q73mTwisR95XOBqpfkpjxHdtM71x5TImwIfiAebryMeqbF0cVBFJJzFPfUEq9QynFfU+pnEAqjb215Sbue5PePU6ctN53Q7XesWyRzg+n8wojwgvID8EHLLCZAkkSYQUUGcEHAIgUf/YPAQAgPAg+AECkEHwAgEgh+AAAkULwAQAiheADAEQKwQcAiBSCDwAQKQQfACBSCD4AQKQQfACASCH4AACRQvABACKF4AMARArBBwCIFIIPABApBB8AIFIIPgBApBB8AIBIIfgAAJFC8AEAIuX/AW567AeM3DDcAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Now we will visualize the predicted licit directed graph \n",
    "\n",
    "licit_ids = df_class_feature_Class3_ID.loc[(df_class_feature_Class3_ID['Time step'] == 10) & (df_class_feature_Class3_ID['Class'] == 0), 'ID']\n",
    "licit_edges = df_edgelist_Class3_ID.loc[df_edgelist_Class3_ID['ID1'].isin(licit_ids)]\n",
    "\n",
    "G = nx.from_pandas_edgelist(licit_edges, source = 'ID1', target = 'ID2', \n",
    "                                 create_using = nx.DiGraph())\n",
    "pos = nx.spring_layout(G)\n",
    "nx.draw(G, with_labels=False, pos=pos)\n",
    "\n",
    "# as we can see the graph is so dence that we can not observe the edges "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Licit case directed graph\n",
      "Number of nodes: 5849\n",
      "Number of edges: 7339\n"
     ]
    }
   ],
   "source": [
    "print('Licit case directed graph')\n",
    "\n",
    "# number of the edges\n",
    "print(f'Number of nodes: {len(list(G.nodes))}')\n",
    "#number of edges \n",
    "print(f'Number of edges: {len(G.edges)}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# we continue with the predicted licit undirected graph \n",
    "G2 = G.to_undirected()\n",
    "pos = nx.spring_layout(G2)\n",
    "nx.draw(G2, with_labels=False, pos=pos)\n",
    "\n",
    "# as we can see the graph is so dence that we can not observe the edges "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Licit case undirected graph\n",
      "Number of nodes: 5849\n",
      "Number of edges: 7339\n"
     ]
    }
   ],
   "source": [
    "print('Licit case undirected graph')\n",
    "\n",
    "# number of the edges\n",
    "print(f'Number of nodes: {len(list(G2.nodes))}')\n",
    "#number of edges \n",
    "print(f'Number of edges: {len(G2.edges)}')\n",
    "\n",
    "# as we expected the same as the directed "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Conclusions \n",
    "\n",
    "# 1) Regarding the metric scores of our models it seems that the random forest clf is suited better to this \n",
    "# graph-based dataset , then comes the MLP classifier and finally the Logistic Regression , but overall all our \n",
    "# models do a pretty good job \n",
    "# 2) The micro-average f1-score and macro-average f1-score outscore the 'classic' f1-score in all cases "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "# next steps \n",
    "\n",
    "#1) gephi visualization \n",
    "#2) calculate valuable metrics for all the computed graphs (nodes, edges, betweeness, closenness etc.)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}