RNN-LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.preprocessing import sequence\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Embedding\n",
    "from tensorflow.keras.layers import LSTM\n",
    "from tensorflow.keras.layers import Bidirectional\n",
    "from tensorflow import keras\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>Date</th>\n",
       "      <th>SP500</th>\n",
       "      <th>SP500_%Change</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1995-01-03</td>\n",
       "      <td>459.11</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1995-01-04</td>\n",
       "      <td>460.71</td>\n",
       "      <td>0.348500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1995-01-05</td>\n",
       "      <td>460.34</td>\n",
       "      <td>-0.080311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1995-01-06</td>\n",
       "      <td>460.68</td>\n",
       "      <td>0.073858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1995-01-09</td>\n",
       "      <td>460.83</td>\n",
       "      <td>0.032561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6400</th>\n",
       "      <td>2020-06-04</td>\n",
       "      <td>3,112.35</td>\n",
       "      <td>-0.336870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6401</th>\n",
       "      <td>2020-06-05</td>\n",
       "      <td>3,193.93</td>\n",
       "      <td>2.621170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6402</th>\n",
       "      <td>2020-06-08</td>\n",
       "      <td>3,232.39</td>\n",
       "      <td>1.204159</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6403</th>\n",
       "      <td>2020-06-09</td>\n",
       "      <td>3,207.18</td>\n",
       "      <td>-0.779918</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6404</th>\n",
       "      <td>2020-06-10</td>\n",
       "      <td>3,190.14</td>\n",
       "      <td>-0.531308</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6405 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           Date     SP500  SP500_%Change\n",
       "0    1995-01-03    459.11            NaN\n",
       "1    1995-01-04    460.71       0.348500\n",
       "2    1995-01-05    460.34      -0.080311\n",
       "3    1995-01-06    460.68       0.073858\n",
       "4    1995-01-09    460.83       0.032561\n",
       "...         ...       ...            ...\n",
       "6400 2020-06-04  3,112.35      -0.336870\n",
       "6401 2020-06-05  3,193.93       2.621170\n",
       "6402 2020-06-08  3,232.39       1.204159\n",
       "6403 2020-06-09  3,207.18      -0.779918\n",
       "6404 2020-06-10  3,190.14      -0.531308\n",
       "\n",
       "[6405 rows x 3 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Load the data for the use of demo\n",
    "df = pd.read_csv('SP500_for_practice_csv.csv', sep = ',', parse_dates = ['Date'])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          Date    SP500  SP500_%Change\n",
      "0   1995-01-03   459.11            NaN\n",
      "1   1995-01-04   460.71       0.348500\n",
      "2   1995-01-05   460.34      -0.080311\n",
      "3   1995-01-06   460.68       0.073858\n",
      "4   1995-01-09   460.83       0.032561\n",
      "..         ...      ...            ...\n",
      "995 1998-12-09  1183.49       0.178605\n",
      "996 1998-12-10  1165.02      -1.560638\n",
      "997 1998-12-11  1166.46       0.123603\n",
      "998 1998-12-14   1141.2      -2.165526\n",
      "999 1998-12-15  1162.83       1.895373\n",
      "\n",
      "[1000 rows x 3 columns]\n"
     ]
    },
    {
     "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>SP500_%Change</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>999.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.097732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.964696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-6.865681</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.360472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.100876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.590136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.115224</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       SP500_%Change\n",
       "count     999.000000\n",
       "mean        0.097732\n",
       "std         0.964696\n",
       "min        -6.865681\n",
       "25%        -0.360472\n",
       "50%         0.100876\n",
       "75%         0.590136\n",
       "max         5.115224"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Chop out one portion of dataframe for demo\n",
    "df_new = df.head(1000)\n",
    "print(df_new)\n",
    "df_new.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Basic preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              SP500  SP500_%Change\n",
      "Date                              \n",
      "1995-01-04   460.71       0.348500\n",
      "1995-01-05   460.34      -0.080311\n",
      "1995-01-06   460.68       0.073858\n",
      "1995-01-09   460.83       0.032561\n",
      "1995-01-10   461.68       0.184450\n",
      "...             ...            ...\n",
      "1998-12-09  1183.49       0.178605\n",
      "1998-12-10  1165.02      -1.560638\n",
      "1998-12-11  1166.46       0.123603\n",
      "1998-12-14   1141.2      -2.165526\n",
      "1998-12-15  1162.83       1.895373\n",
      "\n",
      "[999 rows x 2 columns]\n"
     ]
    }
   ],
   "source": [
    "#drop the missing value\n",
    "df_new = df_new.dropna()\n",
    "#print(df_new)\n",
    "\n",
    "#Set date time index to easily master the data\n",
    "df_new.set_index('Date', inplace = True)\n",
    "print(df_new)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Choose one single dimension --- SP500_%Change ---as the target for forecasting\n",
    "#This data is a demo for univariate LSTM modelling\n",
    "df_new = df_new.drop(['SP500'], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            SP500_%Change\n",
      "Date                     \n",
      "1995-01-04       0.348500\n",
      "1995-01-05      -0.080311\n",
      "1995-01-06       0.073858\n",
      "1995-01-09       0.032561\n",
      "1995-01-10       0.184450\n",
      "...                   ...\n",
      "1998-12-09       0.178605\n",
      "1998-12-10      -1.560638\n",
      "1998-12-11       0.123603\n",
      "1998-12-14      -2.165526\n",
      "1998-12-15       1.895373\n",
      "\n",
      "[999 rows x 1 columns]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualisation\n",
    "%matplotlib inline\n",
    "df_new.head(500).plot()\n",
    "print(df_new)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# More time-based visualisation using seaborn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://seaborn.pydata.org/generated/seaborn.lineplot.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ASUS\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:8: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \n",
      "C:\\Users\\ASUS\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:9: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  if __name__ == '__main__':\n",
      "C:\\Users\\ASUS\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:10: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  # Remove the CWD from sys.path while we load stuff.\n",
      "C:\\Users\\ASUS\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:11: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  # This is added back by InteractiveShellApp.init_path()\n"
     ]
    },
    {
     "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>SP500</th>\n",
       "      <th>SP500_%Change</th>\n",
       "      <th>day</th>\n",
       "      <th>month</th>\n",
       "      <th>year</th>\n",
       "      <th>day_of_week</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995-01-04</th>\n",
       "      <td>460.71</td>\n",
       "      <td>0.348500</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1995</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-01-05</th>\n",
       "      <td>460.34</td>\n",
       "      <td>-0.080311</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1995</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-01-06</th>\n",
       "      <td>460.68</td>\n",
       "      <td>0.073858</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>1995</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-01-09</th>\n",
       "      <td>460.83</td>\n",
       "      <td>0.032561</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>1995</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-01-10</th>\n",
       "      <td>461.68</td>\n",
       "      <td>0.184450</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>1995</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-12-10</th>\n",
       "      <td>1165.02</td>\n",
       "      <td>-1.560638</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>1998</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-12-11</th>\n",
       "      <td>1166.46</td>\n",
       "      <td>0.123603</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "      <td>1998</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-12-14</th>\n",
       "      <td>1141.2</td>\n",
       "      <td>-2.165526</td>\n",
       "      <td>14</td>\n",
       "      <td>12</td>\n",
       "      <td>1998</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-12-15</th>\n",
       "      <td>1162.83</td>\n",
       "      <td>1.895373</td>\n",
       "      <td>15</td>\n",
       "      <td>12</td>\n",
       "      <td>1998</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-12-16</th>\n",
       "      <td>1161.94</td>\n",
       "      <td>-0.076537</td>\n",
       "      <td>16</td>\n",
       "      <td>12</td>\n",
       "      <td>1998</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              SP500  SP500_%Change  day  month  year  day_of_week\n",
       "Date                                                             \n",
       "1995-01-04   460.71       0.348500    4      1  1995            2\n",
       "1995-01-05   460.34      -0.080311    5      1  1995            3\n",
       "1995-01-06   460.68       0.073858    6      1  1995            4\n",
       "1995-01-09   460.83       0.032561    9      1  1995            0\n",
       "1995-01-10   461.68       0.184450   10      1  1995            1\n",
       "...             ...            ...  ...    ...   ...          ...\n",
       "1998-12-10  1165.02      -1.560638   10     12  1998            3\n",
       "1998-12-11  1166.46       0.123603   11     12  1998            4\n",
       "1998-12-14   1141.2      -2.165526   14     12  1998            0\n",
       "1998-12-15  1162.83       1.895373   15     12  1998            1\n",
       "1998-12-16  1161.94      -0.076537   16     12  1998            2\n",
       "\n",
       "[1000 rows x 6 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Use original data to show more datetime-based line plots\n",
    "#Data processing for visualisation\n",
    "df2 = df.copy()\n",
    "df2_new = df2.dropna() #NAs is suggested to be dropped before datetime index is set.\n",
    "df2_new.set_index('Date', inplace = True)\n",
    "#print(df2)\n",
    "\n",
    "#Create new columns recording correspinding time stamp\n",
    "df2_new['day'] = df2_new.index.day\n",
    "df2_new['month'] = df2_new.index.month\n",
    "df2_new['year'] = df2_new.index.year\n",
    "df2_new['day_of_week'] = df2_new.index.dayofweek\n",
    "\n",
    "df2_new = df2_new.head(1000)\n",
    "df2_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f2cef1a288>"
      ]
     },
     "execution_count": 44,
     "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": [
    "import seaborn as sns\n",
    "sns.lineplot(data = df2_new, x = df2_new.index, y = df2_new['SP500_%Change'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f2cef56588>"
      ]
     },
     "execution_count": 46,
     "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": [
    "#Show the change of return patterns by day of week (0 for Mon., and etc.)\n",
    "#The interval is 95% confidence interval, and this applies to the sae things as below\n",
    "sns.lineplot(data = df2_new, x = df2_new.day_of_week, y = df2_new['SP500_%Change'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f2d0237e88>"
      ]
     },
     "execution_count": 48,
     "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": [
    "#Monthly pattern for return changes\n",
    "sns.lineplot(data = df2_new, x = df2_new.month, y = df2_new['SP500_%Change'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f2d02a5b48>"
      ]
     },
     "execution_count": 49,
     "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": [
    "#Yearly patterns\n",
    "sns.lineplot(data = df2_new, x = df2_new.year, y = df2_new['SP500_%Change'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Slice out training and test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(758, 1)\n",
      "(242, 1)\n"
     ]
    }
   ],
   "source": [
    "#Split training and testing set (using univariate df_new)\n",
    "df_train = df_new[\"1995/1/4\":\"1997/12/31\"]\n",
    "df_test = df_new[\"1997/12/31\":] # [\"start_date\": ] start date to end\n",
    "print(df_train.shape)\n",
    "print(df_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sequential Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "<class 'pandas.core.frame.DataFrame'>\n"
     ]
    }
   ],
   "source": [
    "print(type(df_train)); print(type(df_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Define the sequential data transformation\n",
    "#Would work better on the pandas.DataFrame type of data\n",
    "#Resources: #Resources: https://github.com/jeffheaton/t81_558_deep_learning/blob/master/t81_558_class_10_2_lstm.ipynb \n",
    "\n",
    "def to_sequential(data, seq_size):\n",
    "    x = []\n",
    "    y = []\n",
    "    for i in range(len(data) - seq_size):\n",
    "        first_window = data.iloc[i:(i + seq_size), 0]\n",
    "        further_window = data.iloc[i + seq_size, 0]\n",
    "        #print(first_window)\n",
    "        first_window = [[value] for value in first_window]\n",
    "        x.append(first_window)\n",
    "        y.append(further_window)\n",
    "    return np.array(x), np.array(y).reshape(len(y), 1)\n",
    "    #Series input into array: some data format issue occurs, so .reshape() is called to fix it into proper results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Transform the training dataset into sequential-typed data\n",
    "seq_train_x, seq_train_y = to_sequential(df_train, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(757, 1, 1)\n",
      "(757, 1)\n"
     ]
    }
   ],
   "source": [
    "print(seq_train_x.shape); print(seq_train_y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[ 3.48500359e-01]]\n",
      "\n",
      " [[-8.03108250e-02]]\n",
      "\n",
      " [[ 7.38584520e-02]]\n",
      "\n",
      " [[ 3.25605630e-02]]\n",
      "\n",
      " [[ 1.84449797e-01]]\n",
      "\n",
      " [[-4.33200500e-03]]\n",
      "\n",
      " [[-4.33219300e-03]]\n",
      "\n",
      " [[ 9.37960315e-01]]\n",
      "\n",
      " [[ 7.31806769e-01]]\n",
      "\n",
      " [[ 1.42741489e-01]]\n",
      "\n",
      " [[-7.02052970e-02]]\n",
      "\n",
      " [[-5.89713021e-01]]\n",
      "\n",
      " [[-4.64717850e-01]]\n",
      "\n",
      " [[ 2.21610224e-01]]\n",
      "\n",
      " [[ 1.07339900e-02]]\n",
      "\n",
      " [[ 3.39157687e-01]]\n",
      "\n",
      " [[ 1.88259456e-01]]\n",
      "\n",
      " [[ 4.42005466e-01]]\n",
      "\n",
      " [[-3.99668360e-01]]\n",
      "\n",
      " [[ 4.07675396e-01]]\n",
      "\n",
      " [[-4.25152000e-03]]\n",
      "\n",
      " [[ 5.05952381e-01]]\n",
      "\n",
      " [[ 1.23947714e+00]]\n",
      "\n",
      " [[ 5.22313221e-01]]\n",
      "\n",
      " [[-6.85871060e-02]]\n",
      "\n",
      " [[ 7.90332980e-02]]\n",
      "\n",
      " [[-2.07818118e-01]]\n",
      "\n",
      " [[ 2.64478644e-01]]\n",
      "\n",
      " [[ 3.94632990e-02]]\n",
      "\n",
      " [[ 1.86857677e-01]]\n",
      "\n",
      " [[ 4.12392498e-01]]\n",
      "\n",
      " [[ 1.40339291e-01]]\n",
      "\n",
      " [[-6.69799266e-01]]\n",
      "\n",
      " [[ 1.59760981e-01]]\n",
      "\n",
      " [[ 4.82661474e-01]]\n",
      "\n",
      " [[ 3.79326695e-01]]\n",
      "\n",
      " [[ 2.46452116e-01]]\n",
      "\n",
      " [[-8.80948966e-01]]\n",
      "\n",
      " [[ 7.39959902e-01]]\n",
      "\n",
      " [[-3.57003632e-01]]\n",
      "\n",
      " [[-1.07072995e-01]]\n",
      "\n",
      " [[ 5.97777920e-02]]\n",
      "\n",
      " [[ 4.32615060e-02]]\n",
      "\n",
      " [[-7.22772481e-01]]\n",
      "\n",
      " [[ 2.11565585e-01]]\n",
      "\n",
      " [[ 4.13958700e-03]]\n",
      "\n",
      " [[ 1.32668267e+00]]\n",
      "\n",
      " [[ 9.80452230e-02]]\n",
      "\n",
      " [[ 5.79532701e-01]]\n",
      "\n",
      " [[-2.04913875e-01]]\n",
      "\n",
      " [[ 7.17654713e-01]]\n",
      "\n",
      " [[ 2.22038310e-02]]\n",
      "\n",
      " [[ 1.27139167e-01]]\n",
      "\n",
      " [[-2.17676106e-01]]\n",
      "\n",
      " [[ 1.21194983e-01]]\n",
      "\n",
      " [[ 5.64891960e-02]]\n",
      "\n",
      " [[ 1.01219881e+00]]\n",
      "\n",
      " [[ 4.45136435e-01]]\n",
      "\n",
      " [[ 1.39109698e-01]]\n",
      "\n",
      " [[-1.54792618e-01]]\n",
      "\n",
      " [[-1.78883765e-01]]\n",
      "\n",
      " [[-3.00665047e-01]]\n",
      "\n",
      " [[ 2.27676699e-01]]\n",
      "\n",
      " [[ 6.75500648e-01]]\n",
      "\n",
      " [[ 6.53154940e-02]]\n",
      "\n",
      " [[ 1.00876239e-01]]\n",
      "\n",
      " [[ 6.71830540e-02]]\n",
      "\n",
      " [[ 1.16504088e-01]]\n",
      "\n",
      " [[-2.91907457e-01]]\n",
      "\n",
      " [[ 3.24412003e-01]]\n",
      "\n",
      " [[ 4.06175444e-01]]\n",
      "\n",
      " [[-6.08762249e-01]]\n",
      "\n",
      " [[-1.50159050e-01]]\n",
      "\n",
      " [[-8.90436710e-02]]\n",
      "\n",
      " [[ 7.32789350e-02]]\n",
      "\n",
      " [[ 6.33299689e-01]]\n",
      "\n",
      " [[ 8.65307086e-01]]\n",
      "\n",
      " [[-1.54029129e-01]]\n",
      "\n",
      " [[ 1.09353642e-01]]\n",
      "\n",
      " [[ 1.73604338e-01]]\n",
      "\n",
      " [[ 2.25878688e-01]]\n",
      "\n",
      " [[-8.74278720e-02]]\n",
      "\n",
      " [[ 1.16672500e-01]]\n",
      "\n",
      " [[ 1.09155887e+00]]\n",
      "\n",
      " [[ 1.15278200e-02]]\n",
      "\n",
      " [[-8.06854420e-02]]\n",
      "\n",
      " [[ 7.38291164e-01]]\n",
      "\n",
      " [[-7.63417050e-02]]\n",
      "\n",
      " [[ 1.52800061e-01]]\n",
      "\n",
      " [[ 1.90708700e-03]]\n",
      "\n",
      " [[ 2.25031943e-01]]\n",
      "\n",
      " [[ 4.16706308e-01]]\n",
      "\n",
      " [[ 8.52692610e-02]]\n",
      "\n",
      " [[-2.12044908e-01]]\n",
      "\n",
      " [[-1.42106362e+00]]\n",
      "\n",
      " [[-7.50606260e-02]]\n",
      "\n",
      " [[ 8.59030413e-01]]\n",
      "\n",
      " [[ 9.43378211e-01]]\n",
      "\n",
      " [[ 3.78365100e-03]]\n",
      "\n",
      " [[-3.78350800e-03]]\n",
      "\n",
      " [[-9.34561759e-01]]\n",
      "\n",
      " [[-1.33677070e-02]]\n",
      "\n",
      " [[ 1.87554910e+00]]\n",
      "\n",
      " [[ 1.68728910e-02]]\n",
      "\n",
      " [[-1.83696039e-01]]\n",
      "\n",
      " [[ 5.80270793e-01]]\n",
      "\n",
      " [[-9.33532500e-03]]\n",
      "\n",
      " [[-4.51871907e-01]]\n",
      "\n",
      " [[-1.46305779e-01]]\n",
      "\n",
      " [[-8.28402367e-01]]\n",
      "\n",
      " [[ 5.56881464e-01]]\n",
      "\n",
      " [[ 9.73854732e-01]]\n",
      "\n",
      " [[ 7.83509000e-02]]\n",
      "\n",
      " [[ 1.21162414e-01]]\n",
      "\n",
      " [[ 5.04542746e-01]]\n",
      "\n",
      " [[ 9.98462479e-01]]\n",
      "\n",
      " [[-4.40189280e-02]]\n",
      "\n",
      " [[-1.83492972e-01]]\n",
      "\n",
      " [[ 1.30335674e+00]]\n",
      "\n",
      " [[-2.46792603e-01]]\n",
      "\n",
      " [[-1.01508068e+00]]\n",
      "\n",
      " [[-3.12425340e-01]]\n",
      "\n",
      " [[ 4.24017846e-01]]\n",
      "\n",
      " [[-1.57876379e-01]]\n",
      "\n",
      " [[ 1.61803372e-01]]\n",
      "\n",
      " [[ 4.29554842e-01]]\n",
      "\n",
      " [[ 3.10734980e-02]]\n",
      "\n",
      " [[ 1.22976282e+00]]\n",
      "\n",
      " [[ 4.29610643e-01]]\n",
      "\n",
      " [[ 1.47383935e-01]]\n",
      "\n",
      " [[-4.32527504e-01]]\n",
      "\n",
      " [[ 1.10133747e+00]]\n",
      "\n",
      " [[ 1.96116890e-02]]\n",
      "\n",
      " [[-1.97860963e-01]]\n",
      "\n",
      " [[ 5.05456429e-01]]\n",
      "\n",
      " [[-7.57037248e-01]]\n",
      "\n",
      " [[-1.33939763e+00]]\n",
      "\n",
      " [[ 4.64626665e-01]]\n",
      "\n",
      " [[ 1.44524330e-02]]\n",
      "\n",
      " [[ 5.43694231e-01]]\n",
      "\n",
      " [[ 8.03046907e-01]]\n",
      "\n",
      " [[ 9.08928890e-02]]\n",
      "\n",
      " [[ 6.42794822e-01]]\n",
      "\n",
      " [[-4.05151976e-01]]\n",
      "\n",
      " [[-1.54548523e-01]]\n",
      "\n",
      " [[-4.30559015e-01]]\n",
      "\n",
      " [[-1.50096491e-01]]\n",
      "\n",
      " [[-8.94774500e-03]]\n",
      "\n",
      " [[ 3.40044740e-02]]\n",
      "\n",
      " [[ 1.95011987e-01]]\n",
      "\n",
      " [[ 6.42822710e-02]]\n",
      "\n",
      " [[-1.21344064e-01]]\n",
      "\n",
      " [[-4.03780529e-01]]\n",
      "\n",
      " [[-4.19768589e-01]]\n",
      "\n",
      " [[ 8.34068923e-01]]\n",
      "\n",
      " [[-2.09025619e-01]]\n",
      "\n",
      " [[ 2.50640027e-01]]\n",
      "\n",
      " [[-1.66080326e-01]]\n",
      "\n",
      " [[ 3.04092730e-02]]\n",
      "\n",
      " [[-1.96706067e-01]]\n",
      "\n",
      " [[ 2.52638369e-01]]\n",
      "\n",
      " [[-4.25364598e-01]]\n",
      "\n",
      " [[ 5.74361920e-02]]\n",
      "\n",
      " [[ 4.73576579e-01]]\n",
      "\n",
      " [[-1.87466524e-01]]\n",
      "\n",
      " [[ 1.69931133e-01]]\n",
      "\n",
      " [[ 1.64285714e-01]]\n",
      "\n",
      " [[ 1.71147401e-01]]\n",
      "\n",
      " [[ 3.48828931e-01]]\n",
      "\n",
      " [[ 9.45303632e-01]]\n",
      "\n",
      " [[ 1.75694432e-01]]\n",
      "\n",
      " [[ 2.10463550e-02]]\n",
      "\n",
      " [[ 4.19085027e-01]]\n",
      "\n",
      " [[ 2.14779633e-01]]\n",
      "\n",
      " [[ 4.53032705e-01]]\n",
      "\n",
      " [[ 3.92014015e-01]]\n",
      "\n",
      " [[ 8.36256198e-01]]\n",
      "\n",
      " [[-4.45502990e-02]]\n",
      "\n",
      " [[-9.94257310e-02]]\n",
      "\n",
      " [[ 2.45379824e-01]]\n",
      "\n",
      " [[ 4.39917836e-01]]\n",
      "\n",
      " [[-6.42500469e-01]]\n",
      "\n",
      " [[-2.17838765e-01]]\n",
      "\n",
      " [[ 1.37520840e-02]]\n",
      "\n",
      " [[-6.87509670e-02]]\n",
      "\n",
      " [[-6.36383960e-02]]\n",
      "\n",
      " [[ 8.31268071e-01]]\n",
      "\n",
      " [[-2.49202041e-01]]\n",
      "\n",
      " [[-4.60293287e-01]]\n",
      "\n",
      " [[ 1.06580485e-01]]\n",
      "\n",
      " [[-1.49397259e-01]]\n",
      "\n",
      " [[ 1.99494385e-01]]\n",
      "\n",
      " [[-2.40289720e-02]]\n",
      "\n",
      " [[-7.07308280e-01]]\n",
      "\n",
      " [[-1.46964746e-01]]\n",
      "\n",
      " [[ 3.35919102e-01]]\n",
      "\n",
      " [[ 6.28171056e-01]]\n",
      "\n",
      " [[ 2.40096038e-01]]\n",
      "\n",
      " [[-2.51497006e-01]]\n",
      "\n",
      " [[ 6.43191602e-01]]\n",
      "\n",
      " [[ 1.12478271e-01]]\n",
      "\n",
      " [[ 5.46438785e-01]]\n",
      "\n",
      " [[-5.40082959e-01]]\n",
      "\n",
      " [[-4.08538454e-01]]\n",
      "\n",
      " [[ 2.52965508e-01]]\n",
      "\n",
      " [[-6.93899819e-01]]\n",
      "\n",
      " [[-9.87175305e-01]]\n",
      "\n",
      " [[ 5.16715217e-01]]\n",
      "\n",
      " [[ 6.12385717e-01]]\n",
      "\n",
      " [[-3.00042863e-01]]\n",
      "\n",
      " [[ 4.67755804e-01]]\n",
      "\n",
      " [[ 9.41426175e-01]]\n",
      "\n",
      " [[ 1.44136200e-01]]\n",
      "\n",
      " [[-3.57281948e-01]]\n",
      "\n",
      " [[-3.63661081e-01]]\n",
      "\n",
      " [[ 9.19293219e-01]]\n",
      "\n",
      " [[ 2.61952646e-01]]\n",
      "\n",
      " [[-9.10224860e-02]]\n",
      "\n",
      " [[-7.08597650e-02]]\n",
      "\n",
      " [[-5.08188418e-01]]\n",
      "\n",
      " [[ 7.92479085e-01]]\n",
      "\n",
      " [[ 5.69061890e-01]]\n",
      "\n",
      " [[ 4.57026149e-01]]\n",
      "\n",
      " [[-5.36604063e-01]]\n",
      "\n",
      " [[ 5.67981905e-01]]\n",
      "\n",
      " [[-3.06544049e-01]]\n",
      "\n",
      " [[ 2.62366310e-01]]\n",
      "\n",
      " [[ 2.25011251e-01]]\n",
      "\n",
      " [[ 8.53123129e-01]]\n",
      "\n",
      " [[ 1.96223926e-01]]\n",
      "\n",
      " [[-3.73576460e-01]]\n",
      "\n",
      " [[ 2.65953054e-01]]\n",
      "\n",
      " [[ 1.10382550e+00]]\n",
      "\n",
      " [[ 6.51805501e-01]]\n",
      "\n",
      " [[ 4.04740319e-01]]\n",
      "\n",
      " [[-6.46586475e-01]]\n",
      "\n",
      " [[ 2.12603665e-01]]\n",
      "\n",
      " [[ 3.30375073e-01]]\n",
      "\n",
      " [[-1.19447314e-01]]\n",
      "\n",
      " [[ 4.70280229e-01]]\n",
      "\n",
      " [[-7.67263427e-01]]\n",
      "\n",
      " [[-9.40154310e-02]]\n",
      "\n",
      " [[-1.54622449e+00]]\n",
      "\n",
      " [[ 8.43756695e-01]]\n",
      "\n",
      " [[-9.78870132e-01]]\n",
      "\n",
      " [[ 7.50899429e-01]]\n",
      "\n",
      " [[ 2.40790185e-01]]\n",
      "\n",
      " [[ 3.82377933e-01]]\n",
      "\n",
      " [[ 3.74409900e-02]]\n",
      "\n",
      " [[-6.67176540e-02]]\n",
      "\n",
      " [[ 2.94730672e-01]]\n",
      "\n",
      " [[ 7.79309337e-01]]\n",
      "\n",
      " [[ 9.50493770e-02]]\n",
      "\n",
      " [[-5.82630529e-01]]\n",
      "\n",
      " [[-1.60271977e-01]]\n",
      "\n",
      " [[ 2.83763844e-01]]\n",
      "\n",
      " [[-1.45684442e+00]]\n",
      "\n",
      " [[-1.79998359e+00]]\n",
      "\n",
      " [[ 7.03448737e-01]]\n",
      "\n",
      " [[-1.46012046e-01]]\n",
      "\n",
      " [[-3.30669148e-01]]\n",
      "\n",
      " [[ 1.43709780e+00]]\n",
      "\n",
      " [[-3.40214319e-01]]\n",
      "\n",
      " [[ 3.08392566e-01]]\n",
      "\n",
      " [[ 5.90227542e-01]]\n",
      "\n",
      " [[ 2.56607227e-01]]\n",
      "\n",
      " [[-9.94457120e-02]]\n",
      "\n",
      " [[ 1.17005826e+00]]\n",
      "\n",
      " [[-4.72611136e-01]]\n",
      "\n",
      " [[ 7.43886035e-01]]\n",
      "\n",
      " [[ 4.18261961e-01]]\n",
      "\n",
      " [[ 9.49985582e-01]]\n",
      "\n",
      " [[ 9.31524240e-01]]\n",
      "\n",
      " [[ 3.83635735e-01]]\n",
      "\n",
      " [[-4.10362435e-01]]\n",
      "\n",
      " [[ 8.79151988e-01]]\n",
      "\n",
      " [[ 7.63918120e-01]]\n",
      "\n",
      " [[ 5.56991011e-01]]\n",
      "\n",
      " [[ 9.44717123e-01]]\n",
      "\n",
      " [[ 4.57268280e-02]]\n",
      "\n",
      " [[ 7.73953715e-01]]\n",
      "\n",
      " [[-1.42112027e-01]]\n",
      "\n",
      " [[-7.46392939e-01]]\n",
      "\n",
      " [[-6.49806278e-01]]\n",
      "\n",
      " [[-5.12804766e-01]]\n",
      "\n",
      " [[-1.13120775e+00]]\n",
      "\n",
      " [[ 1.16288145e+00]]\n",
      "\n",
      " [[ 1.66023762e+00]]\n",
      "\n",
      " [[ 3.33910090e-02]]\n",
      "\n",
      " [[-1.30788372e+00]]\n",
      "\n",
      " [[-4.95034283e-01]]\n",
      "\n",
      " [[-3.84710463e-01]]\n",
      "\n",
      " [[-6.70027142e-01]]\n",
      "\n",
      " [[ 6.15211655e-01]]\n",
      "\n",
      " [[ 9.99425796e-01]]\n",
      "\n",
      " [[ 7.65200289e-01]]\n",
      "\n",
      " [[-5.77928910e-01]]\n",
      "\n",
      " [[ 2.53067485e-01]]\n",
      "\n",
      " [[-3.08268951e+00]]\n",
      "\n",
      " [[ 1.02920284e+00]]\n",
      "\n",
      " [[-4.57798194e-01]]\n",
      "\n",
      " [[ 2.29166994e-01]]\n",
      "\n",
      " [[ 3.63323154e-01]]\n",
      "\n",
      " [[ 8.73812160e-02]]\n",
      "\n",
      " [[ 1.74921659e+00]]\n",
      "\n",
      " [[-1.47092622e-01]]\n",
      "\n",
      " [[-2.62394697e-01]]\n",
      "\n",
      " [[-1.21542201e-01]]\n",
      "\n",
      " [[ 2.20274496e-01]]\n",
      "\n",
      " [[-8.91457380e-02]]\n",
      "\n",
      " [[ 4.50741493e-01]]\n",
      "\n",
      " [[-6.21774354e-01]]\n",
      "\n",
      " [[ 4.62313700e-03]]\n",
      "\n",
      " [[-5.30095232e-01]]\n",
      "\n",
      " [[ 1.27498063e+00]]\n",
      "\n",
      " [[ 2.34041577e-01]]\n",
      "\n",
      " [[ 9.46189300e-02]]\n",
      "\n",
      " [[-3.04933800e-03]]\n",
      "\n",
      " [[-1.77171957e+00]]\n",
      "\n",
      " [[-3.18204396e-01]]\n",
      "\n",
      " [[-1.35318208e+00]]\n",
      "\n",
      " [[-3.66219416e-01]]\n",
      "\n",
      " [[ 8.76136760e-01]]\n",
      "\n",
      " [[ 9.07791616e-01]]\n",
      "\n",
      " [[ 3.90667559e-01]]\n",
      "\n",
      " [[-5.25581395e-01]]\n",
      "\n",
      " [[ 3.11715840e-01]]\n",
      "\n",
      " [[ 2.26845450e-01]]\n",
      "\n",
      " [[ 4.37161858e-01]]\n",
      "\n",
      " [[ 5.69541126e-01]]\n",
      "\n",
      " [[-2.16397066e-01]]\n",
      "\n",
      " [[ 4.15276005e-01]]\n",
      "\n",
      " [[ 9.03702120e-02]]\n",
      "\n",
      " [[ 1.07122089e-01]]\n",
      "\n",
      " [[ 1.52867800e-03]]\n",
      "\n",
      " [[ 6.26748400e-02]]\n",
      "\n",
      " [[-1.71102081e+00]]\n",
      "\n",
      " [[-2.72000995e-01]]\n",
      "\n",
      " [[-1.27799511e-01]]\n",
      "\n",
      " [[-3.97933865e-01]]\n",
      "\n",
      " [[ 1.01996052e+00]]\n",
      "\n",
      " [[ 1.03913023e-01]]\n",
      "\n",
      " [[ 1.03030491e+00]]\n",
      "\n",
      " [[ 1.44458587e+00]]\n",
      "\n",
      " [[ 6.18282414e-01]]\n",
      "\n",
      " [[-2.70432690e-02]]\n",
      "\n",
      " [[-8.56601850e-02]]\n",
      "\n",
      " [[ 6.10664060e-01]]\n",
      "\n",
      " [[ 6.33867037e-01]]\n",
      "\n",
      " [[-5.79365670e-02]]\n",
      "\n",
      " [[ 8.41310423e-01]]\n",
      "\n",
      " [[-3.56711182e-01]]\n",
      "\n",
      " [[ 3.71301775e-01]]\n",
      "\n",
      " [[-9.25557472e-01]]\n",
      "\n",
      " [[-6.39662020e-01]]\n",
      "\n",
      " [[ 5.64430404e-01]]\n",
      "\n",
      " [[-3.84100045e-01]]\n",
      "\n",
      " [[-2.15208034e-01]]\n",
      "\n",
      " [[ 7.30889049e-01]]\n",
      "\n",
      " [[ 8.74271440e-01]]\n",
      "\n",
      " [[-7.97417605e-01]]\n",
      "\n",
      " [[ 4.16029000e-02]]\n",
      "\n",
      " [[-1.70797998e-01]]\n",
      "\n",
      " [[-1.77041181e-01]]\n",
      "\n",
      " [[-2.87643263e-01]]\n",
      "\n",
      " [[-1.67404042e-01]]\n",
      "\n",
      " [[-3.09917355e-01]]\n",
      "\n",
      " [[-1.03626943e-01]]\n",
      "\n",
      " [[-4.66053280e-01]]\n",
      "\n",
      " [[-1.51043710e-02]]\n",
      "\n",
      " [[ 2.11493140e-02]]\n",
      "\n",
      " [[ 7.15903942e-01]]\n",
      "\n",
      " [[ 3.01421630e-01]]\n",
      "\n",
      " [[-5.53188310e-02]]\n",
      "\n",
      " [[-6.11835807e-01]]\n",
      "\n",
      " [[ 6.26138262e-01]]\n",
      "\n",
      " [[ 3.11121083e-01]]\n",
      "\n",
      " [[ 7.82845981e-01]]\n",
      "\n",
      " [[-3.35858436e-01]]\n",
      "\n",
      " [[-1.79629162e-01]]\n",
      "\n",
      " [[-2.22486615e+00]]\n",
      "\n",
      " [[-7.45315162e-01]]\n",
      "\n",
      " [[ 3.38676556e-01]]\n",
      "\n",
      " [[ 2.00076365e-01]]\n",
      "\n",
      " [[-1.58369661e+00]]\n",
      "\n",
      " [[ 8.05364970e-02]]\n",
      "\n",
      " [[-2.53640570e+00]]\n",
      "\n",
      " [[-2.27056208e-01]]\n",
      "\n",
      " [[ 9.07108869e-01]]\n",
      "\n",
      " [[ 1.49668018e+00]]\n",
      "\n",
      " [[-7.50512773e-01]]\n",
      "\n",
      " [[-7.76540948e-01]]\n",
      "\n",
      " [[-1.08872304e+00]]\n",
      "\n",
      " [[-3.50949960e-02]]\n",
      "\n",
      " [[ 7.21295779e-01]]\n",
      "\n",
      " [[ 7.49401904e-01]]\n",
      "\n",
      " [[-7.84714578e-01]]\n",
      "\n",
      " [[ 6.89480275e-01]]\n",
      "\n",
      " [[ 7.38280389e-01]]\n",
      "\n",
      " [[ 1.57356043e+00]]\n",
      "\n",
      " [[ 1.91840251e+00]]\n",
      "\n",
      " [[-3.41137225e-01]]\n",
      "\n",
      " [[ 3.25644094e-01]]\n",
      "\n",
      " [[ 2.68727921e-01]]\n",
      "\n",
      " [[-2.36388822e-01]]\n",
      "\n",
      " [[-7.39522180e-02]]\n",
      "\n",
      " [[ 5.54296934e-01]]\n",
      "\n",
      " [[-8.36625261e-01]]\n",
      "\n",
      " [[ 2.80218116e-01]]\n",
      "\n",
      " [[ 3.47405790e-02]]\n",
      "\n",
      " [[ 4.42411065e-01]]\n",
      "\n",
      " [[ 2.05950001e-01]]\n",
      "\n",
      " [[-1.33517357e-01]]\n",
      "\n",
      " [[-9.31364450e-02]]\n",
      "\n",
      " [[ 8.43520231e-01]]\n",
      "\n",
      " [[-5.44223773e-01]]\n",
      "\n",
      " [[-4.72242628e-01]]\n",
      "\n",
      " [[ 3.79586672e-01]]\n",
      "\n",
      " [[-2.38595438e-01]]\n",
      "\n",
      " [[-1.11460417e+00]]\n",
      "\n",
      " [[-8.22938850e-01]]\n",
      "\n",
      " [[ 4.18718078e-01]]\n",
      "\n",
      " [[ 1.35935973e-01]]\n",
      "\n",
      " [[-9.41108281e-01]]\n",
      "\n",
      " [[ 9.60827790e-01]]\n",
      "\n",
      " [[ 1.23230844e+00]]\n",
      "\n",
      " [[ 7.53284300e-03]]\n",
      "\n",
      " [[ 5.22739941e-01]]\n",
      "\n",
      " [[ 5.79966431e-01]]\n",
      "\n",
      " [[ 1.39909111e+00]]\n",
      "\n",
      " [[ 5.05480942e-01]]\n",
      "\n",
      " [[-1.52051230e-01]]\n",
      "\n",
      " [[-2.15245849e-01]]\n",
      "\n",
      " [[ 2.24514652e-01]]\n",
      "\n",
      " [[ 5.90043924e-01]]\n",
      "\n",
      " [[-8.00547280e-02]]\n",
      "\n",
      " [[-1.26733481e-01]]\n",
      "\n",
      " [[ 3.20882130e-02]]\n",
      "\n",
      " [[ 4.37426200e-03]]\n",
      "\n",
      " [[ 4.81147760e-02]]\n",
      "\n",
      " [[ 1.63220099e-01]]\n",
      "\n",
      " [[ 2.57525716e-01]]\n",
      "\n",
      " [[ 7.15446683e-01]]\n",
      "\n",
      " [[-1.77230876e-01]]\n",
      "\n",
      " [[ 1.25292301e+00]]\n",
      "\n",
      " [[ 2.68012431e-01]]\n",
      "\n",
      " [[-3.83882617e-01]]\n",
      "\n",
      " [[-5.56633935e-01]]\n",
      "\n",
      " [[-3.05709447e-01]]\n",
      "\n",
      " [[ 8.70992355e-01]]\n",
      "\n",
      " [[ 4.11041018e-01]]\n",
      "\n",
      " [[-1.37874179e-01]]\n",
      "\n",
      " [[ 2.61895612e-01]]\n",
      "\n",
      " [[ 3.66263966e-01]]\n",
      "\n",
      " [[ 5.41733264e-01]]\n",
      "\n",
      " [[-1.36462114e-01]]\n",
      "\n",
      " [[-4.62069451e-01]]\n",
      "\n",
      " [[ 9.90701560e-02]]\n",
      "\n",
      " [[-7.04115826e-01]]\n",
      "\n",
      " [[-1.95076108e-01]]\n",
      "\n",
      " [[-5.22170861e-01]]\n",
      "\n",
      " [[ 6.08094541e-01]]\n",
      "\n",
      " [[-8.55310050e-02]]\n",
      "\n",
      " [[ 6.23484092e-01]]\n",
      "\n",
      " [[-2.12684504e-01]]\n",
      "\n",
      " [[ 4.20591955e-01]]\n",
      "\n",
      " [[ 1.04849094e+00]]\n",
      "\n",
      " [[ 1.46329851e+00]]\n",
      "\n",
      " [[ 4.22307788e-01]]\n",
      "\n",
      " [[ 4.35649007e-01]]\n",
      "\n",
      " [[ 1.43674229e-01]]\n",
      "\n",
      " [[-3.15629825e-01]]\n",
      "\n",
      " [[ 2.15198202e-01]]\n",
      "\n",
      " [[ 6.49679264e-01]]\n",
      "\n",
      " [[ 2.36451595e-01]]\n",
      "\n",
      " [[-8.13426970e-02]]\n",
      "\n",
      " [[ 6.97403056e-01]]\n",
      "\n",
      " [[ 2.41187884e-01]]\n",
      "\n",
      " [[-1.61301163e-01]]\n",
      "\n",
      " [[ 8.05116123e-01]]\n",
      "\n",
      " [[ 1.10854380e+00]]\n",
      "\n",
      " [[-1.41341823e-01]]\n",
      "\n",
      " [[-1.26990846e-01]]\n",
      "\n",
      " [[ 2.67549669e-01]]\n",
      "\n",
      " [[-6.07645770e-02]]\n",
      "\n",
      " [[-1.09442741e+00]]\n",
      "\n",
      " [[-4.24974608e-01]]\n",
      "\n",
      " [[-9.66313250e-02]]\n",
      "\n",
      " [[-6.42145141e-01]]\n",
      "\n",
      " [[ 1.38047593e+00]]\n",
      "\n",
      " [[-3.02743362e-01]]\n",
      "\n",
      " [[-9.10988041e-01]]\n",
      "\n",
      " [[-1.53902232e+00]]\n",
      "\n",
      " [[-9.46073790e-02]]\n",
      "\n",
      " [[-1.05127361e+00]]\n",
      "\n",
      " [[ 7.01822519e-01]]\n",
      "\n",
      " [[ 7.57534020e-01]]\n",
      "\n",
      " [[ 1.94384449e+00]]\n",
      "\n",
      " [[ 4.17024244e-01]]\n",
      "\n",
      " [[-2.60392324e-01]]\n",
      "\n",
      " [[ 5.50259733e-01]]\n",
      "\n",
      " [[ 6.37790767e-01]]\n",
      "\n",
      " [[ 1.28337435e-01]]\n",
      "\n",
      " [[-3.88482934e-01]]\n",
      "\n",
      " [[-1.73907276e+00]]\n",
      "\n",
      " [[-5.03550504e-01]]\n",
      "\n",
      " [[ 1.49523073e+00]]\n",
      "\n",
      " [[-5.08001020e-02]]\n",
      "\n",
      " [[ 7.46338527e-01]]\n",
      "\n",
      " [[-6.39910784e-01]]\n",
      "\n",
      " [[ 8.60490907e-01]]\n",
      "\n",
      " [[ 6.16016427e-01]]\n",
      "\n",
      " [[ 1.31665600e-03]]\n",
      "\n",
      " [[ 1.23105687e+00]]\n",
      "\n",
      " [[-2.15904066e-01]]\n",
      "\n",
      " [[ 3.32377477e-01]]\n",
      "\n",
      " [[ 8.34037025e-01]]\n",
      "\n",
      " [[ 6.82840100e-02]]\n",
      "\n",
      " [[ 7.75074031e-01]]\n",
      "\n",
      " [[ 4.48436222e-01]]\n",
      "\n",
      " [[-1.10273075e+00]]\n",
      "\n",
      " [[-9.05396368e-01]]\n",
      "\n",
      " [[-7.13803665e-01]]\n",
      "\n",
      " [[ 0.00000000e+00]]\n",
      "\n",
      " [[ 9.77752216e-01]]\n",
      "\n",
      " [[ 1.51067961e+00]]\n",
      "\n",
      " [[ 2.53771504e-01]]\n",
      "\n",
      " [[ 7.25043250e-02]]\n",
      "\n",
      " [[ 3.21584279e-01]]\n",
      "\n",
      " [[-1.39117654e+00]]\n",
      "\n",
      " [[ 2.40273423e-01]]\n",
      "\n",
      " [[ 1.20617830e+00]]\n",
      "\n",
      " [[-5.23076144e-01]]\n",
      "\n",
      " [[ 5.29646181e-01]]\n",
      "\n",
      " [[ 1.66922073e+00]]\n",
      "\n",
      " [[ 1.12734656e+00]]\n",
      "\n",
      " [[-4.11421251e-01]]\n",
      "\n",
      " [[ 9.66010291e-01]]\n",
      "\n",
      " [[-4.65520832e-01]]\n",
      "\n",
      " [[-1.19263006e+00]]\n",
      "\n",
      " [[-1.28300947e-01]]\n",
      "\n",
      " [[ 1.06140165e+00]]\n",
      "\n",
      " [[ 2.24613714e-01]]\n",
      "\n",
      " [[-7.90543037e-01]]\n",
      "\n",
      " [[-1.31690001e+00]]\n",
      "\n",
      " [[-5.34544128e-01]]\n",
      "\n",
      " [[ 5.67765105e-01]]\n",
      "\n",
      " [[-5.48213904e-01]]\n",
      "\n",
      " [[ 1.39578987e+00]]\n",
      "\n",
      " [[-4.27686131e-01]]\n",
      "\n",
      " [[ 8.02694851e-01]]\n",
      "\n",
      " [[ 1.07830106e+00]]\n",
      "\n",
      " [[-2.83905856e-01]]\n",
      "\n",
      " [[-8.72630463e-01]]\n",
      "\n",
      " [[-1.82776714e+00]]\n",
      "\n",
      " [[ 4.57216678e-01]]\n",
      "\n",
      " [[ 3.20233998e-01]]\n",
      "\n",
      " [[-7.60327255e-01]]\n",
      "\n",
      " [[-4.92617076e-01]]\n",
      "\n",
      " [[-3.97062754e-01]]\n",
      "\n",
      " [[ 1.85268000e-01]]\n",
      "\n",
      " [[ 8.65960974e-01]]\n",
      "\n",
      " [[-2.30120497e-01]]\n",
      "\n",
      " [[ 1.81226000e-01]]\n",
      "\n",
      " [[-2.10246679e+00]]\n",
      "\n",
      " [[-2.16571045e+00]]\n",
      "\n",
      " [[ 3.32840237e-01]]\n",
      "\n",
      " [[-1.25454163e+00]]\n",
      "\n",
      " [[ 2.79958940e-02]]\n",
      "\n",
      " [[ 1.01023563e+00]]\n",
      "\n",
      " [[ 5.58121124e-01]]\n",
      "\n",
      " [[ 5.23532731e-01]]\n",
      "\n",
      " [[-7.20513758e-01]]\n",
      "\n",
      " [[-2.97133842e-01]]\n",
      "\n",
      " [[-2.72832766e+00]]\n",
      "\n",
      " [[ 8.24239138e-01]]\n",
      "\n",
      " [[ 1.47768680e+00]]\n",
      "\n",
      " [[ 1.16732033e+00]]\n",
      "\n",
      " [[-2.30508297e-01]]\n",
      "\n",
      " [[ 5.99918611e-01]]\n",
      "\n",
      " [[-7.79027586e-01]]\n",
      "\n",
      " [[ 1.87277247e+00]]\n",
      "\n",
      " [[-1.25224306e-01]]\n",
      "\n",
      " [[-3.17977354e-01]]\n",
      "\n",
      " [[-7.53390907e-01]]\n",
      "\n",
      " [[ 9.91677228e-01]]\n",
      "\n",
      " [[ 2.72847237e+00]]\n",
      "\n",
      " [[ 9.18078207e-01]]\n",
      "\n",
      " [[-3.50662640e-01]]\n",
      "\n",
      " [[ 1.80832279e+00]]\n",
      "\n",
      " [[ 2.12430963e+00]]\n",
      "\n",
      " [[-2.98708807e-01]]\n",
      "\n",
      " [[-1.46660868e+00]]\n",
      "\n",
      " [[ 5.68892376e-01]]\n",
      "\n",
      " [[ 5.51044791e-01]]\n",
      "\n",
      " [[ 1.56162855e+00]]\n",
      "\n",
      " [[-5.40792207e-01]]\n",
      "\n",
      " [[ 3.49285226e-01]]\n",
      "\n",
      " [[ 6.98531171e-01]]\n",
      "\n",
      " [[-1.44082292e+00]]\n",
      "\n",
      " [[ 4.24224164e-01]]\n",
      "\n",
      " [[ 1.00687652e+00]]\n",
      "\n",
      " [[-2.74457619e-01]]\n",
      "\n",
      " [[-4.39625901e-01]]\n",
      "\n",
      " [[ 1.36060120e+00]]\n",
      "\n",
      " [[ 3.16399655e-01]]\n",
      "\n",
      " [[-2.94218027e-01]]\n",
      "\n",
      " [[-3.69447953e-01]]\n",
      "\n",
      " [[ 4.97583167e-01]]\n",
      "\n",
      " [[-2.26340359e-01]]\n",
      "\n",
      " [[-1.03974668e-01]]\n",
      "\n",
      " [[-6.35142168e-01]]\n",
      "\n",
      " [[ 3.95186345e-01]]\n",
      "\n",
      " [[ 1.72865561e+00]]\n",
      "\n",
      " [[ 5.71088915e-01]]\n",
      "\n",
      " [[ 2.73493180e-01]]\n",
      "\n",
      " [[ 4.96954708e-01]]\n",
      "\n",
      " [[ 1.59964120e+00]]\n",
      "\n",
      " [[ 1.10811790e+00]]\n",
      "\n",
      " [[ 7.05273880e-02]]\n",
      "\n",
      " [[ 5.81720550e-02]]\n",
      "\n",
      " [[-5.99271036e-01]]\n",
      "\n",
      " [[ 1.00443165e+00]]\n",
      "\n",
      " [[ 7.90654680e-02]]\n",
      "\n",
      " [[-2.23433849e+00]]\n",
      "\n",
      " [[ 2.01679907e+00]]\n",
      "\n",
      " [[-8.20001339e-01]]\n",
      "\n",
      " [[-5.97307056e-01]]\n",
      "\n",
      " [[ 4.09650552e-01]]\n",
      "\n",
      " [[-2.43435140e-01]]\n",
      "\n",
      " [[ 6.65431457e-01]]\n",
      "\n",
      " [[ 1.45898567e+00]]\n",
      "\n",
      " [[ 1.42583764e+00]]\n",
      "\n",
      " [[-5.14766828e-01]]\n",
      "\n",
      " [[ 7.18044289e-01]]\n",
      "\n",
      " [[-1.22013605e+00]]\n",
      "\n",
      " [[ 6.87573000e-01]]\n",
      "\n",
      " [[ 3.17363041e-01]]\n",
      "\n",
      " [[ 1.85451848e-01]]\n",
      "\n",
      " [[ 8.03588928e-01]]\n",
      "\n",
      " [[ 1.16984964e+00]]\n",
      "\n",
      " [[-5.31716119e-01]]\n",
      "\n",
      " [[-1.75073260e+00]]\n",
      "\n",
      " [[-2.57838960e-01]]\n",
      "\n",
      " [[ 2.30464214e+00]]\n",
      "\n",
      " [[ 2.76237179e-01]]\n",
      "\n",
      " [[ 3.99333732e-01]]\n",
      "\n",
      " [[-1.60587047e-01]]\n",
      "\n",
      " [[-2.49257022e-01]]\n",
      "\n",
      " [[ 6.23631801e-01]]\n",
      "\n",
      " [[ 1.06124441e+00]]\n",
      "\n",
      " [[ 2.10020057e-01]]\n",
      "\n",
      " [[-7.49248132e-01]]\n",
      "\n",
      " [[ 3.33635999e-01]]\n",
      "\n",
      " [[ 2.17825950e-01]]\n",
      "\n",
      " [[ 8.34759600e-01]]\n",
      "\n",
      " [[-9.50724758e-01]]\n",
      "\n",
      " [[-1.85557039e+00]]\n",
      "\n",
      " [[ 3.70632217e-01]]\n",
      "\n",
      " [[-1.11739594e+00]]\n",
      "\n",
      " [[-4.86762436e-01]]\n",
      "\n",
      " [[ 2.98258172e-01]]\n",
      "\n",
      " [[-2.59091450e+00]]\n",
      "\n",
      " [[ 1.29661083e+00]]\n",
      "\n",
      " [[ 1.48166007e+00]]\n",
      "\n",
      " [[ 1.44058920e+00]]\n",
      "\n",
      " [[-1.52232927e+00]]\n",
      "\n",
      " [[-1.62153397e-01]]\n",
      "\n",
      " [[-3.67061881e-01]]\n",
      "\n",
      " [[-7.75952008e-01]]\n",
      "\n",
      " [[ 7.44781060e-02]]\n",
      "\n",
      " [[-1.09773449e+00]]\n",
      "\n",
      " [[-4.64771432e-01]]\n",
      "\n",
      " [[ 3.12517371e+00]]\n",
      "\n",
      " [[ 3.01860760e-02]]\n",
      "\n",
      " [[ 3.24402388e-01]]\n",
      "\n",
      " [[-1.95516023e-01]]\n",
      "\n",
      " [[ 2.31419192e-01]]\n",
      "\n",
      " [[ 2.59879725e-01]]\n",
      "\n",
      " [[-1.56273430e+00]]\n",
      "\n",
      " [[-7.00738822e-01]]\n",
      "\n",
      " [[ 1.24042560e+00]]\n",
      "\n",
      " [[-4.48095594e-01]]\n",
      "\n",
      " [[ 2.81265969e+00]]\n",
      "\n",
      " [[-2.79176008e-01]]\n",
      "\n",
      " [[ 4.54931071e-01]]\n",
      "\n",
      " [[ 3.39917026e-01]]\n",
      "\n",
      " [[ 5.17616858e-01]]\n",
      "\n",
      " [[-3.66327203e-01]]\n",
      "\n",
      " [[-7.82620571e-01]]\n",
      "\n",
      " [[-6.95620871e-01]]\n",
      "\n",
      " [[ 7.79392479e-01]]\n",
      "\n",
      " [[ 8.59059267e-01]]\n",
      "\n",
      " [[-6.35659890e-01]]\n",
      "\n",
      " [[ 8.58246770e-01]]\n",
      "\n",
      " [[ 5.28568887e-01]]\n",
      "\n",
      " [[ 4.75813673e-01]]\n",
      "\n",
      " [[ 7.93757707e-01]]\n",
      "\n",
      " [[ 1.07228408e+00]]\n",
      "\n",
      " [[-9.43933599e-01]]\n",
      "\n",
      " [[-3.30649799e-01]]\n",
      "\n",
      " [[-3.75018030e-01]]\n",
      "\n",
      " [[ 1.15824526e-01]]\n",
      "\n",
      " [[ 2.25183349e-01]]\n",
      "\n",
      " [[-4.69967432e-01]]\n",
      "\n",
      " [[-1.08623618e+00]]\n",
      "\n",
      " [[-1.15888320e+00]]\n",
      "\n",
      " [[ 1.21271818e+00]]\n",
      "\n",
      " [[ 1.74443549e+00]]\n",
      "\n",
      " [[-3.89805406e-01]]\n",
      "\n",
      " [[-1.83791263e+00]]\n",
      "\n",
      " [[-9.51940170e-01]]\n",
      "\n",
      " [[-6.86568115e+00]]\n",
      "\n",
      " [[ 5.11522366e+00]]\n",
      "\n",
      " [[-2.91804524e-01]]\n",
      "\n",
      " [[-1.68414639e+00]]\n",
      "\n",
      " [[ 1.21060552e+00]]\n",
      "\n",
      " [[ 2.66449454e+00]]\n",
      "\n",
      " [[ 1.88500410e-01]]\n",
      "\n",
      " [[ 2.12594073e-01]]\n",
      "\n",
      " [[-5.01718359e-01]]\n",
      "\n",
      " [[-1.12149931e+00]]\n",
      "\n",
      " [[-6.87863204e-01]]\n",
      "\n",
      " [[ 2.87690120e-01]]\n",
      "\n",
      " [[-1.92903072e+00]]\n",
      "\n",
      " [[ 1.18106760e+00]]\n",
      "\n",
      " [[ 1.27528200e+00]]\n",
      "\n",
      " [[ 1.92276620e+00]]\n",
      "\n",
      " [[-8.42316635e-01]]\n",
      "\n",
      " [[ 6.77872164e-01]]\n",
      "\n",
      " [[ 1.52341227e+00]]\n",
      "\n",
      " [[ 4.28580367e-01]]\n",
      "\n",
      " [[-1.70492893e+00]]\n",
      "\n",
      " [[ 4.38378738e-01]]\n",
      "\n",
      " [[ 8.62413500e-02]]\n",
      "\n",
      " [[ 3.95107394e-01]]\n",
      "\n",
      " [[ 2.02846975e+00]]\n",
      "\n",
      " [[-3.18020476e-01]]\n",
      "\n",
      " [[ 5.23835007e-01]]\n",
      "\n",
      " [[-3.75728165e-01]]\n",
      "\n",
      " [[ 1.09855102e+00]]\n",
      "\n",
      " [[-1.44339747e-01]]\n",
      "\n",
      " [[-6.70826674e-01]]\n",
      "\n",
      " [[-6.13867880e-01]]\n",
      "\n",
      " [[-1.53125935e+00]]\n",
      "\n",
      " [[-1.62313863e-01]]\n",
      "\n",
      " [[ 1.04888870e+00]]\n",
      "\n",
      " [[ 4.82670570e-01]]\n",
      "\n",
      " [[-2.58253791e-01]]\n",
      "\n",
      " [[-1.06054643e+00]]\n",
      "\n",
      " [[-8.91866429e-01]]\n",
      "\n",
      " [[ 7.30898414e-01]]\n",
      "\n",
      " [[-1.52773409e+00]]\n",
      "\n",
      " [[-6.84676243e-01]]\n",
      "\n",
      " [[ 4.03130696e-01]]\n",
      "\n",
      " [[ 1.80360079e+00]]\n",
      "\n",
      " [[ 1.83458331e+00]]]\n"
     ]
    }
   ],
   "source": [
    "print(seq_train_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-8.03108250e-02]\n",
      " [ 7.38584520e-02]\n",
      " [ 3.25605630e-02]\n",
      " [ 1.84449797e-01]\n",
      " [-4.33200500e-03]\n",
      " [-4.33219300e-03]\n",
      " [ 9.37960315e-01]\n",
      " [ 7.31806769e-01]\n",
      " [ 1.42741489e-01]\n",
      " [-7.02052970e-02]\n",
      " [-5.89713021e-01]\n",
      " [-4.64717850e-01]\n",
      " [ 2.21610224e-01]\n",
      " [ 1.07339900e-02]\n",
      " [ 3.39157687e-01]\n",
      " [ 1.88259456e-01]\n",
      " [ 4.42005466e-01]\n",
      " [-3.99668360e-01]\n",
      " [ 4.07675396e-01]\n",
      " [-4.25152000e-03]\n",
      " [ 5.05952381e-01]\n",
      " [ 1.23947714e+00]\n",
      " [ 5.22313221e-01]\n",
      " [-6.85871060e-02]\n",
      " [ 7.90332980e-02]\n",
      " [-2.07818118e-01]\n",
      " [ 2.64478644e-01]\n",
      " [ 3.94632990e-02]\n",
      " [ 1.86857677e-01]\n",
      " [ 4.12392498e-01]\n",
      " [ 1.40339291e-01]\n",
      " [-6.69799266e-01]\n",
      " [ 1.59760981e-01]\n",
      " [ 4.82661474e-01]\n",
      " [ 3.79326695e-01]\n",
      " [ 2.46452116e-01]\n",
      " [-8.80948966e-01]\n",
      " [ 7.39959902e-01]\n",
      " [-3.57003632e-01]\n",
      " [-1.07072995e-01]\n",
      " [ 5.97777920e-02]\n",
      " [ 4.32615060e-02]\n",
      " [-7.22772481e-01]\n",
      " [ 2.11565585e-01]\n",
      " [ 4.13958700e-03]\n",
      " [ 1.32668267e+00]\n",
      " [ 9.80452230e-02]\n",
      " [ 5.79532701e-01]\n",
      " [-2.04913875e-01]\n",
      " [ 7.17654713e-01]\n",
      " [ 2.22038310e-02]\n",
      " [ 1.27139167e-01]\n",
      " [-2.17676106e-01]\n",
      " [ 1.21194983e-01]\n",
      " [ 5.64891960e-02]\n",
      " [ 1.01219881e+00]\n",
      " [ 4.45136435e-01]\n",
      " [ 1.39109698e-01]\n",
      " [-1.54792618e-01]\n",
      " [-1.78883765e-01]\n",
      " [-3.00665047e-01]\n",
      " [ 2.27676699e-01]\n",
      " [ 6.75500648e-01]\n",
      " [ 6.53154940e-02]\n",
      " [ 1.00876239e-01]\n",
      " [ 6.71830540e-02]\n",
      " [ 1.16504088e-01]\n",
      " [-2.91907457e-01]\n",
      " [ 3.24412003e-01]\n",
      " [ 4.06175444e-01]\n",
      " [-6.08762249e-01]\n",
      " [-1.50159050e-01]\n",
      " [-8.90436710e-02]\n",
      " [ 7.32789350e-02]\n",
      " [ 6.33299689e-01]\n",
      " [ 8.65307086e-01]\n",
      " [-1.54029129e-01]\n",
      " [ 1.09353642e-01]\n",
      " [ 1.73604338e-01]\n",
      " [ 2.25878688e-01]\n",
      " [-8.74278720e-02]\n",
      " [ 1.16672500e-01]\n",
      " [ 1.09155887e+00]\n",
      " [ 1.15278200e-02]\n",
      " [-8.06854420e-02]\n",
      " [ 7.38291164e-01]\n",
      " [-7.63417050e-02]\n",
      " [ 1.52800061e-01]\n",
      " [ 1.90708700e-03]\n",
      " [ 2.25031943e-01]\n",
      " [ 4.16706308e-01]\n",
      " [ 8.52692610e-02]\n",
      " [-2.12044908e-01]\n",
      " [-1.42106362e+00]\n",
      " [-7.50606260e-02]\n",
      " [ 8.59030413e-01]\n",
      " [ 9.43378211e-01]\n",
      " [ 3.78365100e-03]\n",
      " [-3.78350800e-03]\n",
      " [-9.34561759e-01]\n",
      " [-1.33677070e-02]\n",
      " [ 1.87554910e+00]\n",
      " [ 1.68728910e-02]\n",
      " [-1.83696039e-01]\n",
      " [ 5.80270793e-01]\n",
      " [-9.33532500e-03]\n",
      " [-4.51871907e-01]\n",
      " [-1.46305779e-01]\n",
      " [-8.28402367e-01]\n",
      " [ 5.56881464e-01]\n",
      " [ 9.73854732e-01]\n",
      " [ 7.83509000e-02]\n",
      " [ 1.21162414e-01]\n",
      " [ 5.04542746e-01]\n",
      " [ 9.98462479e-01]\n",
      " [-4.40189280e-02]\n",
      " [-1.83492972e-01]\n",
      " [ 1.30335674e+00]\n",
      " [-2.46792603e-01]\n",
      " [-1.01508068e+00]\n",
      " [-3.12425340e-01]\n",
      " [ 4.24017846e-01]\n",
      " [-1.57876379e-01]\n",
      " [ 1.61803372e-01]\n",
      " [ 4.29554842e-01]\n",
      " [ 3.10734980e-02]\n",
      " [ 1.22976282e+00]\n",
      " [ 4.29610643e-01]\n",
      " [ 1.47383935e-01]\n",
      " [-4.32527504e-01]\n",
      " [ 1.10133747e+00]\n",
      " [ 1.96116890e-02]\n",
      " [-1.97860963e-01]\n",
      " [ 5.05456429e-01]\n",
      " [-7.57037248e-01]\n",
      " [-1.33939763e+00]\n",
      " [ 4.64626665e-01]\n",
      " [ 1.44524330e-02]\n",
      " [ 5.43694231e-01]\n",
      " [ 8.03046907e-01]\n",
      " [ 9.08928890e-02]\n",
      " [ 6.42794822e-01]\n",
      " [-4.05151976e-01]\n",
      " [-1.54548523e-01]\n",
      " [-4.30559015e-01]\n",
      " [-1.50096491e-01]\n",
      " [-8.94774500e-03]\n",
      " [ 3.40044740e-02]\n",
      " [ 1.95011987e-01]\n",
      " [ 6.42822710e-02]\n",
      " [-1.21344064e-01]\n",
      " [-4.03780529e-01]\n",
      " [-4.19768589e-01]\n",
      " [ 8.34068923e-01]\n",
      " [-2.09025619e-01]\n",
      " [ 2.50640027e-01]\n",
      " [-1.66080326e-01]\n",
      " [ 3.04092730e-02]\n",
      " [-1.96706067e-01]\n",
      " [ 2.52638369e-01]\n",
      " [-4.25364598e-01]\n",
      " [ 5.74361920e-02]\n",
      " [ 4.73576579e-01]\n",
      " [-1.87466524e-01]\n",
      " [ 1.69931133e-01]\n",
      " [ 1.64285714e-01]\n",
      " [ 1.71147401e-01]\n",
      " [ 3.48828931e-01]\n",
      " [ 9.45303632e-01]\n",
      " [ 1.75694432e-01]\n",
      " [ 2.10463550e-02]\n",
      " [ 4.19085027e-01]\n",
      " [ 2.14779633e-01]\n",
      " [ 4.53032705e-01]\n",
      " [ 3.92014015e-01]\n",
      " [ 8.36256198e-01]\n",
      " [-4.45502990e-02]\n",
      " [-9.94257310e-02]\n",
      " [ 2.45379824e-01]\n",
      " [ 4.39917836e-01]\n",
      " [-6.42500469e-01]\n",
      " [-2.17838765e-01]\n",
      " [ 1.37520840e-02]\n",
      " [-6.87509670e-02]\n",
      " [-6.36383960e-02]\n",
      " [ 8.31268071e-01]\n",
      " [-2.49202041e-01]\n",
      " [-4.60293287e-01]\n",
      " [ 1.06580485e-01]\n",
      " [-1.49397259e-01]\n",
      " [ 1.99494385e-01]\n",
      " [-2.40289720e-02]\n",
      " [-7.07308280e-01]\n",
      " [-1.46964746e-01]\n",
      " [ 3.35919102e-01]\n",
      " [ 6.28171056e-01]\n",
      " [ 2.40096038e-01]\n",
      " [-2.51497006e-01]\n",
      " [ 6.43191602e-01]\n",
      " [ 1.12478271e-01]\n",
      " [ 5.46438785e-01]\n",
      " [-5.40082959e-01]\n",
      " [-4.08538454e-01]\n",
      " [ 2.52965508e-01]\n",
      " [-6.93899819e-01]\n",
      " [-9.87175305e-01]\n",
      " [ 5.16715217e-01]\n",
      " [ 6.12385717e-01]\n",
      " [-3.00042863e-01]\n",
      " [ 4.67755804e-01]\n",
      " [ 9.41426175e-01]\n",
      " [ 1.44136200e-01]\n",
      " [-3.57281948e-01]\n",
      " [-3.63661081e-01]\n",
      " [ 9.19293219e-01]\n",
      " [ 2.61952646e-01]\n",
      " [-9.10224860e-02]\n",
      " [-7.08597650e-02]\n",
      " [-5.08188418e-01]\n",
      " [ 7.92479085e-01]\n",
      " [ 5.69061890e-01]\n",
      " [ 4.57026149e-01]\n",
      " [-5.36604063e-01]\n",
      " [ 5.67981905e-01]\n",
      " [-3.06544049e-01]\n",
      " [ 2.62366310e-01]\n",
      " [ 2.25011251e-01]\n",
      " [ 8.53123129e-01]\n",
      " [ 1.96223926e-01]\n",
      " [-3.73576460e-01]\n",
      " [ 2.65953054e-01]\n",
      " [ 1.10382550e+00]\n",
      " [ 6.51805501e-01]\n",
      " [ 4.04740319e-01]\n",
      " [-6.46586475e-01]\n",
      " [ 2.12603665e-01]\n",
      " [ 3.30375073e-01]\n",
      " [-1.19447314e-01]\n",
      " [ 4.70280229e-01]\n",
      " [-7.67263427e-01]\n",
      " [-9.40154310e-02]\n",
      " [-1.54622449e+00]\n",
      " [ 8.43756695e-01]\n",
      " [-9.78870132e-01]\n",
      " [ 7.50899429e-01]\n",
      " [ 2.40790185e-01]\n",
      " [ 3.82377933e-01]\n",
      " [ 3.74409900e-02]\n",
      " [-6.67176540e-02]\n",
      " [ 2.94730672e-01]\n",
      " [ 7.79309337e-01]\n",
      " [ 9.50493770e-02]\n",
      " [-5.82630529e-01]\n",
      " [-1.60271977e-01]\n",
      " [ 2.83763844e-01]\n",
      " [-1.45684442e+00]\n",
      " [-1.79998359e+00]\n",
      " [ 7.03448737e-01]\n",
      " [-1.46012046e-01]\n",
      " [-3.30669148e-01]\n",
      " [ 1.43709780e+00]\n",
      " [-3.40214319e-01]\n",
      " [ 3.08392566e-01]\n",
      " [ 5.90227542e-01]\n",
      " [ 2.56607227e-01]\n",
      " [-9.94457120e-02]\n",
      " [ 1.17005826e+00]\n",
      " [-4.72611136e-01]\n",
      " [ 7.43886035e-01]\n",
      " [ 4.18261961e-01]\n",
      " [ 9.49985582e-01]\n",
      " [ 9.31524240e-01]\n",
      " [ 3.83635735e-01]\n",
      " [-4.10362435e-01]\n",
      " [ 8.79151988e-01]\n",
      " [ 7.63918120e-01]\n",
      " [ 5.56991011e-01]\n",
      " [ 9.44717123e-01]\n",
      " [ 4.57268280e-02]\n",
      " [ 7.73953715e-01]\n",
      " [-1.42112027e-01]\n",
      " [-7.46392939e-01]\n",
      " [-6.49806278e-01]\n",
      " [-5.12804766e-01]\n",
      " [-1.13120775e+00]\n",
      " [ 1.16288145e+00]\n",
      " [ 1.66023762e+00]\n",
      " [ 3.33910090e-02]\n",
      " [-1.30788372e+00]\n",
      " [-4.95034283e-01]\n",
      " [-3.84710463e-01]\n",
      " [-6.70027142e-01]\n",
      " [ 6.15211655e-01]\n",
      " [ 9.99425796e-01]\n",
      " [ 7.65200289e-01]\n",
      " [-5.77928910e-01]\n",
      " [ 2.53067485e-01]\n",
      " [-3.08268951e+00]\n",
      " [ 1.02920284e+00]\n",
      " [-4.57798194e-01]\n",
      " [ 2.29166994e-01]\n",
      " [ 3.63323154e-01]\n",
      " [ 8.73812160e-02]\n",
      " [ 1.74921659e+00]\n",
      " [-1.47092622e-01]\n",
      " [-2.62394697e-01]\n",
      " [-1.21542201e-01]\n",
      " [ 2.20274496e-01]\n",
      " [-8.91457380e-02]\n",
      " [ 4.50741493e-01]\n",
      " [-6.21774354e-01]\n",
      " [ 4.62313700e-03]\n",
      " [-5.30095232e-01]\n",
      " [ 1.27498063e+00]\n",
      " [ 2.34041577e-01]\n",
      " [ 9.46189300e-02]\n",
      " [-3.04933800e-03]\n",
      " [-1.77171957e+00]\n",
      " [-3.18204396e-01]\n",
      " [-1.35318208e+00]\n",
      " [-3.66219416e-01]\n",
      " [ 8.76136760e-01]\n",
      " [ 9.07791616e-01]\n",
      " [ 3.90667559e-01]\n",
      " [-5.25581395e-01]\n",
      " [ 3.11715840e-01]\n",
      " [ 2.26845450e-01]\n",
      " [ 4.37161858e-01]\n",
      " [ 5.69541126e-01]\n",
      " [-2.16397066e-01]\n",
      " [ 4.15276005e-01]\n",
      " [ 9.03702120e-02]\n",
      " [ 1.07122089e-01]\n",
      " [ 1.52867800e-03]\n",
      " [ 6.26748400e-02]\n",
      " [-1.71102081e+00]\n",
      " [-2.72000995e-01]\n",
      " [-1.27799511e-01]\n",
      " [-3.97933865e-01]\n",
      " [ 1.01996052e+00]\n",
      " [ 1.03913023e-01]\n",
      " [ 1.03030491e+00]\n",
      " [ 1.44458587e+00]\n",
      " [ 6.18282414e-01]\n",
      " [-2.70432690e-02]\n",
      " [-8.56601850e-02]\n",
      " [ 6.10664060e-01]\n",
      " [ 6.33867037e-01]\n",
      " [-5.79365670e-02]\n",
      " [ 8.41310423e-01]\n",
      " [-3.56711182e-01]\n",
      " [ 3.71301775e-01]\n",
      " [-9.25557472e-01]\n",
      " [-6.39662020e-01]\n",
      " [ 5.64430404e-01]\n",
      " [-3.84100045e-01]\n",
      " [-2.15208034e-01]\n",
      " [ 7.30889049e-01]\n",
      " [ 8.74271440e-01]\n",
      " [-7.97417605e-01]\n",
      " [ 4.16029000e-02]\n",
      " [-1.70797998e-01]\n",
      " [-1.77041181e-01]\n",
      " [-2.87643263e-01]\n",
      " [-1.67404042e-01]\n",
      " [-3.09917355e-01]\n",
      " [-1.03626943e-01]\n",
      " [-4.66053280e-01]\n",
      " [-1.51043710e-02]\n",
      " [ 2.11493140e-02]\n",
      " [ 7.15903942e-01]\n",
      " [ 3.01421630e-01]\n",
      " [-5.53188310e-02]\n",
      " [-6.11835807e-01]\n",
      " [ 6.26138262e-01]\n",
      " [ 3.11121083e-01]\n",
      " [ 7.82845981e-01]\n",
      " [-3.35858436e-01]\n",
      " [-1.79629162e-01]\n",
      " [-2.22486615e+00]\n",
      " [-7.45315162e-01]\n",
      " [ 3.38676556e-01]\n",
      " [ 2.00076365e-01]\n",
      " [-1.58369661e+00]\n",
      " [ 8.05364970e-02]\n",
      " [-2.53640570e+00]\n",
      " [-2.27056208e-01]\n",
      " [ 9.07108869e-01]\n",
      " [ 1.49668018e+00]\n",
      " [-7.50512773e-01]\n",
      " [-7.76540948e-01]\n",
      " [-1.08872304e+00]\n",
      " [-3.50949960e-02]\n",
      " [ 7.21295779e-01]\n",
      " [ 7.49401904e-01]\n",
      " [-7.84714578e-01]\n",
      " [ 6.89480275e-01]\n",
      " [ 7.38280389e-01]\n",
      " [ 1.57356043e+00]\n",
      " [ 1.91840251e+00]\n",
      " [-3.41137225e-01]\n",
      " [ 3.25644094e-01]\n",
      " [ 2.68727921e-01]\n",
      " [-2.36388822e-01]\n",
      " [-7.39522180e-02]\n",
      " [ 5.54296934e-01]\n",
      " [-8.36625261e-01]\n",
      " [ 2.80218116e-01]\n",
      " [ 3.47405790e-02]\n",
      " [ 4.42411065e-01]\n",
      " [ 2.05950001e-01]\n",
      " [-1.33517357e-01]\n",
      " [-9.31364450e-02]\n",
      " [ 8.43520231e-01]\n",
      " [-5.44223773e-01]\n",
      " [-4.72242628e-01]\n",
      " [ 3.79586672e-01]\n",
      " [-2.38595438e-01]\n",
      " [-1.11460417e+00]\n",
      " [-8.22938850e-01]\n",
      " [ 4.18718078e-01]\n",
      " [ 1.35935973e-01]\n",
      " [-9.41108281e-01]\n",
      " [ 9.60827790e-01]\n",
      " [ 1.23230844e+00]\n",
      " [ 7.53284300e-03]\n",
      " [ 5.22739941e-01]\n",
      " [ 5.79966431e-01]\n",
      " [ 1.39909111e+00]\n",
      " [ 5.05480942e-01]\n",
      " [-1.52051230e-01]\n",
      " [-2.15245849e-01]\n",
      " [ 2.24514652e-01]\n",
      " [ 5.90043924e-01]\n",
      " [-8.00547280e-02]\n",
      " [-1.26733481e-01]\n",
      " [ 3.20882130e-02]\n",
      " [ 4.37426200e-03]\n",
      " [ 4.81147760e-02]\n",
      " [ 1.63220099e-01]\n",
      " [ 2.57525716e-01]\n",
      " [ 7.15446683e-01]\n",
      " [-1.77230876e-01]\n",
      " [ 1.25292301e+00]\n",
      " [ 2.68012431e-01]\n",
      " [-3.83882617e-01]\n",
      " [-5.56633935e-01]\n",
      " [-3.05709447e-01]\n",
      " [ 8.70992355e-01]\n",
      " [ 4.11041018e-01]\n",
      " [-1.37874179e-01]\n",
      " [ 2.61895612e-01]\n",
      " [ 3.66263966e-01]\n",
      " [ 5.41733264e-01]\n",
      " [-1.36462114e-01]\n",
      " [-4.62069451e-01]\n",
      " [ 9.90701560e-02]\n",
      " [-7.04115826e-01]\n",
      " [-1.95076108e-01]\n",
      " [-5.22170861e-01]\n",
      " [ 6.08094541e-01]\n",
      " [-8.55310050e-02]\n",
      " [ 6.23484092e-01]\n",
      " [-2.12684504e-01]\n",
      " [ 4.20591955e-01]\n",
      " [ 1.04849094e+00]\n",
      " [ 1.46329851e+00]\n",
      " [ 4.22307788e-01]\n",
      " [ 4.35649007e-01]\n",
      " [ 1.43674229e-01]\n",
      " [-3.15629825e-01]\n",
      " [ 2.15198202e-01]\n",
      " [ 6.49679264e-01]\n",
      " [ 2.36451595e-01]\n",
      " [-8.13426970e-02]\n",
      " [ 6.97403056e-01]\n",
      " [ 2.41187884e-01]\n",
      " [-1.61301163e-01]\n",
      " [ 8.05116123e-01]\n",
      " [ 1.10854380e+00]\n",
      " [-1.41341823e-01]\n",
      " [-1.26990846e-01]\n",
      " [ 2.67549669e-01]\n",
      " [-6.07645770e-02]\n",
      " [-1.09442741e+00]\n",
      " [-4.24974608e-01]\n",
      " [-9.66313250e-02]\n",
      " [-6.42145141e-01]\n",
      " [ 1.38047593e+00]\n",
      " [-3.02743362e-01]\n",
      " [-9.10988041e-01]\n",
      " [-1.53902232e+00]\n",
      " [-9.46073790e-02]\n",
      " [-1.05127361e+00]\n",
      " [ 7.01822519e-01]\n",
      " [ 7.57534020e-01]\n",
      " [ 1.94384449e+00]\n",
      " [ 4.17024244e-01]\n",
      " [-2.60392324e-01]\n",
      " [ 5.50259733e-01]\n",
      " [ 6.37790767e-01]\n",
      " [ 1.28337435e-01]\n",
      " [-3.88482934e-01]\n",
      " [-1.73907276e+00]\n",
      " [-5.03550504e-01]\n",
      " [ 1.49523073e+00]\n",
      " [-5.08001020e-02]\n",
      " [ 7.46338527e-01]\n",
      " [-6.39910784e-01]\n",
      " [ 8.60490907e-01]\n",
      " [ 6.16016427e-01]\n",
      " [ 1.31665600e-03]\n",
      " [ 1.23105687e+00]\n",
      " [-2.15904066e-01]\n",
      " [ 3.32377477e-01]\n",
      " [ 8.34037025e-01]\n",
      " [ 6.82840100e-02]\n",
      " [ 7.75074031e-01]\n",
      " [ 4.48436222e-01]\n",
      " [-1.10273075e+00]\n",
      " [-9.05396368e-01]\n",
      " [-7.13803665e-01]\n",
      " [ 0.00000000e+00]\n",
      " [ 9.77752216e-01]\n",
      " [ 1.51067961e+00]\n",
      " [ 2.53771504e-01]\n",
      " [ 7.25043250e-02]\n",
      " [ 3.21584279e-01]\n",
      " [-1.39117654e+00]\n",
      " [ 2.40273423e-01]\n",
      " [ 1.20617830e+00]\n",
      " [-5.23076144e-01]\n",
      " [ 5.29646181e-01]\n",
      " [ 1.66922073e+00]\n",
      " [ 1.12734656e+00]\n",
      " [-4.11421251e-01]\n",
      " [ 9.66010291e-01]\n",
      " [-4.65520832e-01]\n",
      " [-1.19263006e+00]\n",
      " [-1.28300947e-01]\n",
      " [ 1.06140165e+00]\n",
      " [ 2.24613714e-01]\n",
      " [-7.90543037e-01]\n",
      " [-1.31690001e+00]\n",
      " [-5.34544128e-01]\n",
      " [ 5.67765105e-01]\n",
      " [-5.48213904e-01]\n",
      " [ 1.39578987e+00]\n",
      " [-4.27686131e-01]\n",
      " [ 8.02694851e-01]\n",
      " [ 1.07830106e+00]\n",
      " [-2.83905856e-01]\n",
      " [-8.72630463e-01]\n",
      " [-1.82776714e+00]\n",
      " [ 4.57216678e-01]\n",
      " [ 3.20233998e-01]\n",
      " [-7.60327255e-01]\n",
      " [-4.92617076e-01]\n",
      " [-3.97062754e-01]\n",
      " [ 1.85268000e-01]\n",
      " [ 8.65960974e-01]\n",
      " [-2.30120497e-01]\n",
      " [ 1.81226000e-01]\n",
      " [-2.10246679e+00]\n",
      " [-2.16571045e+00]\n",
      " [ 3.32840237e-01]\n",
      " [-1.25454163e+00]\n",
      " [ 2.79958940e-02]\n",
      " [ 1.01023563e+00]\n",
      " [ 5.58121124e-01]\n",
      " [ 5.23532731e-01]\n",
      " [-7.20513758e-01]\n",
      " [-2.97133842e-01]\n",
      " [-2.72832766e+00]\n",
      " [ 8.24239138e-01]\n",
      " [ 1.47768680e+00]\n",
      " [ 1.16732033e+00]\n",
      " [-2.30508297e-01]\n",
      " [ 5.99918611e-01]\n",
      " [-7.79027586e-01]\n",
      " [ 1.87277247e+00]\n",
      " [-1.25224306e-01]\n",
      " [-3.17977354e-01]\n",
      " [-7.53390907e-01]\n",
      " [ 9.91677228e-01]\n",
      " [ 2.72847237e+00]\n",
      " [ 9.18078207e-01]\n",
      " [-3.50662640e-01]\n",
      " [ 1.80832279e+00]\n",
      " [ 2.12430963e+00]\n",
      " [-2.98708807e-01]\n",
      " [-1.46660868e+00]\n",
      " [ 5.68892376e-01]\n",
      " [ 5.51044791e-01]\n",
      " [ 1.56162855e+00]\n",
      " [-5.40792207e-01]\n",
      " [ 3.49285226e-01]\n",
      " [ 6.98531171e-01]\n",
      " [-1.44082292e+00]\n",
      " [ 4.24224164e-01]\n",
      " [ 1.00687652e+00]\n",
      " [-2.74457619e-01]\n",
      " [-4.39625901e-01]\n",
      " [ 1.36060120e+00]\n",
      " [ 3.16399655e-01]\n",
      " [-2.94218027e-01]\n",
      " [-3.69447953e-01]\n",
      " [ 4.97583167e-01]\n",
      " [-2.26340359e-01]\n",
      " [-1.03974668e-01]\n",
      " [-6.35142168e-01]\n",
      " [ 3.95186345e-01]\n",
      " [ 1.72865561e+00]\n",
      " [ 5.71088915e-01]\n",
      " [ 2.73493180e-01]\n",
      " [ 4.96954708e-01]\n",
      " [ 1.59964120e+00]\n",
      " [ 1.10811790e+00]\n",
      " [ 7.05273880e-02]\n",
      " [ 5.81720550e-02]\n",
      " [-5.99271036e-01]\n",
      " [ 1.00443165e+00]\n",
      " [ 7.90654680e-02]\n",
      " [-2.23433849e+00]\n",
      " [ 2.01679907e+00]\n",
      " [-8.20001339e-01]\n",
      " [-5.97307056e-01]\n",
      " [ 4.09650552e-01]\n",
      " [-2.43435140e-01]\n",
      " [ 6.65431457e-01]\n",
      " [ 1.45898567e+00]\n",
      " [ 1.42583764e+00]\n",
      " [-5.14766828e-01]\n",
      " [ 7.18044289e-01]\n",
      " [-1.22013605e+00]\n",
      " [ 6.87573000e-01]\n",
      " [ 3.17363041e-01]\n",
      " [ 1.85451848e-01]\n",
      " [ 8.03588928e-01]\n",
      " [ 1.16984964e+00]\n",
      " [-5.31716119e-01]\n",
      " [-1.75073260e+00]\n",
      " [-2.57838960e-01]\n",
      " [ 2.30464214e+00]\n",
      " [ 2.76237179e-01]\n",
      " [ 3.99333732e-01]\n",
      " [-1.60587047e-01]\n",
      " [-2.49257022e-01]\n",
      " [ 6.23631801e-01]\n",
      " [ 1.06124441e+00]\n",
      " [ 2.10020057e-01]\n",
      " [-7.49248132e-01]\n",
      " [ 3.33635999e-01]\n",
      " [ 2.17825950e-01]\n",
      " [ 8.34759600e-01]\n",
      " [-9.50724758e-01]\n",
      " [-1.85557039e+00]\n",
      " [ 3.70632217e-01]\n",
      " [-1.11739594e+00]\n",
      " [-4.86762436e-01]\n",
      " [ 2.98258172e-01]\n",
      " [-2.59091450e+00]\n",
      " [ 1.29661083e+00]\n",
      " [ 1.48166007e+00]\n",
      " [ 1.44058920e+00]\n",
      " [-1.52232927e+00]\n",
      " [-1.62153397e-01]\n",
      " [-3.67061881e-01]\n",
      " [-7.75952008e-01]\n",
      " [ 7.44781060e-02]\n",
      " [-1.09773449e+00]\n",
      " [-4.64771432e-01]\n",
      " [ 3.12517371e+00]\n",
      " [ 3.01860760e-02]\n",
      " [ 3.24402388e-01]\n",
      " [-1.95516023e-01]\n",
      " [ 2.31419192e-01]\n",
      " [ 2.59879725e-01]\n",
      " [-1.56273430e+00]\n",
      " [-7.00738822e-01]\n",
      " [ 1.24042560e+00]\n",
      " [-4.48095594e-01]\n",
      " [ 2.81265969e+00]\n",
      " [-2.79176008e-01]\n",
      " [ 4.54931071e-01]\n",
      " [ 3.39917026e-01]\n",
      " [ 5.17616858e-01]\n",
      " [-3.66327203e-01]\n",
      " [-7.82620571e-01]\n",
      " [-6.95620871e-01]\n",
      " [ 7.79392479e-01]\n",
      " [ 8.59059267e-01]\n",
      " [-6.35659890e-01]\n",
      " [ 8.58246770e-01]\n",
      " [ 5.28568887e-01]\n",
      " [ 4.75813673e-01]\n",
      " [ 7.93757707e-01]\n",
      " [ 1.07228408e+00]\n",
      " [-9.43933599e-01]\n",
      " [-3.30649799e-01]\n",
      " [-3.75018030e-01]\n",
      " [ 1.15824526e-01]\n",
      " [ 2.25183349e-01]\n",
      " [-4.69967432e-01]\n",
      " [-1.08623618e+00]\n",
      " [-1.15888320e+00]\n",
      " [ 1.21271818e+00]\n",
      " [ 1.74443549e+00]\n",
      " [-3.89805406e-01]\n",
      " [-1.83791263e+00]\n",
      " [-9.51940170e-01]\n",
      " [-6.86568115e+00]\n",
      " [ 5.11522366e+00]\n",
      " [-2.91804524e-01]\n",
      " [-1.68414639e+00]\n",
      " [ 1.21060552e+00]\n",
      " [ 2.66449454e+00]\n",
      " [ 1.88500410e-01]\n",
      " [ 2.12594073e-01]\n",
      " [-5.01718359e-01]\n",
      " [-1.12149931e+00]\n",
      " [-6.87863204e-01]\n",
      " [ 2.87690120e-01]\n",
      " [-1.92903072e+00]\n",
      " [ 1.18106760e+00]\n",
      " [ 1.27528200e+00]\n",
      " [ 1.92276620e+00]\n",
      " [-8.42316635e-01]\n",
      " [ 6.77872164e-01]\n",
      " [ 1.52341227e+00]\n",
      " [ 4.28580367e-01]\n",
      " [-1.70492893e+00]\n",
      " [ 4.38378738e-01]\n",
      " [ 8.62413500e-02]\n",
      " [ 3.95107394e-01]\n",
      " [ 2.02846975e+00]\n",
      " [-3.18020476e-01]\n",
      " [ 5.23835007e-01]\n",
      " [-3.75728165e-01]\n",
      " [ 1.09855102e+00]\n",
      " [-1.44339747e-01]\n",
      " [-6.70826674e-01]\n",
      " [-6.13867880e-01]\n",
      " [-1.53125935e+00]\n",
      " [-1.62313863e-01]\n",
      " [ 1.04888870e+00]\n",
      " [ 4.82670570e-01]\n",
      " [-2.58253791e-01]\n",
      " [-1.06054643e+00]\n",
      " [-8.91866429e-01]\n",
      " [ 7.30898414e-01]\n",
      " [-1.52773409e+00]\n",
      " [-6.84676243e-01]\n",
      " [ 4.03130696e-01]\n",
      " [ 1.80360079e+00]\n",
      " [ 1.83458331e+00]\n",
      " [-4.22314700e-02]]\n"
     ]
    }
   ],
   "source": [
    "print(seq_train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Transform the test dataset into sequential-typed data\n",
    "seq_test_x, seq_test_y = to_sequential(df_test, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(241, 1, 1)\n",
      "(241, 1)\n"
     ]
    }
   ],
   "source": [
    "print(seq_test_x.shape); print(seq_test_y.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Build the LSTM model (with bidirectional learning layer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Define the model (the most simple form: 1 LSTM layer and then output layer)\n",
    "model = Sequential()\n",
    "model.add(keras.layers.Bidirectional(keras.layers.LSTM(\n",
    "        units = 128, activation = 'tanh', \n",
    "        dropout = 0.0, \n",
    "        recurrent_dropout = 0.0, \n",
    "        input_shape = (seq_train_x.shape[1], seq_train_x.shape[2])\n",
    "        )\n",
    "    )\n",
    ")\n",
    "\n",
    "model.add(Dense(1))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer = 'adam', \n",
    "             loss = 'mean_squared_error')\n",
    "             #metrics = ['mean_absolute_percentage_error'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n",
      "137/137 - 0s - loss: 0.5472 - val_loss: 1.9577\n",
      "Epoch 2/15\n",
      "137/137 - 0s - loss: 0.5472 - val_loss: 1.9559\n",
      "Epoch 3/15\n",
      "137/137 - 0s - loss: 0.5472 - val_loss: 1.9542\n",
      "Epoch 4/15\n",
      "137/137 - 0s - loss: 0.5471 - val_loss: 1.9527\n",
      "Epoch 5/15\n",
      "137/137 - 0s - loss: 0.5471 - val_loss: 1.9513\n",
      "Epoch 6/15\n",
      "137/137 - 0s - loss: 0.5470 - val_loss: 1.9501\n",
      "Epoch 7/15\n",
      "137/137 - 0s - loss: 0.5470 - val_loss: 1.9490\n",
      "Epoch 8/15\n",
      "137/137 - 0s - loss: 0.5470 - val_loss: 1.9480\n",
      "Epoch 9/15\n",
      "137/137 - 0s - loss: 0.5469 - val_loss: 1.9470\n",
      "Epoch 10/15\n",
      "137/137 - 0s - loss: 0.5469 - val_loss: 1.9462\n",
      "Epoch 11/15\n",
      "137/137 - 0s - loss: 0.5469 - val_loss: 1.9453\n",
      "Epoch 12/15\n",
      "137/137 - 0s - loss: 0.5469 - val_loss: 1.9446\n",
      "Epoch 13/15\n",
      "137/137 - 0s - loss: 0.5468 - val_loss: 1.9439\n",
      "Epoch 14/15\n",
      "137/137 - 0s - loss: 0.5468 - val_loss: 1.9432\n",
      "Epoch 15/15\n",
      "137/137 - 0s - loss: 0.5468 - val_loss: 1.9426\n",
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "bidirectional (Bidirectional (None, 256)               133120    \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1)                 257       \n",
      "=================================================================\n",
      "Total params: 133,377\n",
      "Trainable params: 133,377\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#Fit/Train the model\n",
    "model_fit = model.fit(seq_train_x, seq_train_y, \n",
    "              batch_size = 5, \n",
    "              #validation_data = (seq_test_x, seq_test_y),\n",
    "              epochs = 15,\n",
    "              validation_split = 0.1,\n",
    "              verbose = 2,\n",
    "              shuffle = False)\n",
    "\n",
    "#Summarise the model\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot some results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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": [
    "#Plot the outcomes from the training process (recorded in object via .fit())\n",
    "#plot loss and validation loss against epochs\n",
    "plt.plot(model_fit.history['loss'], label = \"Training Loss\")\n",
    "plt.plot(model_fit.history['val_loss'], label = \"Validation Loss\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Obviously, this is a failed training. It is purely for demo."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Prediction: in-sample prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Predict on training set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.11591307]\n",
      " [0.07768746]\n",
      " [0.08994175]\n",
      " [0.08649451]\n",
      " [0.09976438]\n",
      " [0.08351678]\n",
      " [0.08351678]\n",
      " [0.18918256]\n",
      " [0.16088796]\n",
      " [0.09595893]\n",
      " [0.07843922]\n",
      " [0.04918706]\n",
      " [0.05447334]\n",
      " [0.10325765]\n",
      " [0.08472119]\n",
      " [0.11494303]\n",
      " [0.10011806]\n",
      " [0.1259548 ]\n",
      " [0.05766349]\n",
      " [0.12219774]\n",
      " [0.08352318]\n",
      " [0.13316976]\n",
      " [0.2355689 ]\n",
      " [0.13506085]\n",
      " [0.07856027]\n",
      " [0.09038219]\n",
      " [0.06882396]\n",
      " [0.10740753]\n",
      " [0.08706232]\n",
      " [0.0999878 ]\n",
      " [0.12270915]\n",
      " [0.09574346]\n",
      " [0.04638261]\n",
      " [0.09749703]\n",
      " [0.13050935]\n",
      " [0.11915668]\n",
      " [0.10564683]\n",
      " [0.04116147]\n",
      " [0.16195314]\n",
      " [0.0599192 ]\n",
      " [0.07573151]\n",
      " [0.08875289]\n",
      " [0.08737619]\n",
      " [0.04477709]\n",
      " [0.10230387]\n",
      " [0.08419205]\n",
      " [0.2500633 ]\n",
      " [0.09201651]\n",
      " [0.14181879]\n",
      " [0.069013  ]\n",
      " [0.15904965]\n",
      " [0.08564888]\n",
      " [0.09456676]\n",
      " [0.06818675]\n",
      " [0.09404088]\n",
      " [0.08847724]\n",
      " [0.20005877]\n",
      " [0.12630151]\n",
      " [0.09563334]\n",
      " [0.07236986]\n",
      " [0.07073411]\n",
      " [0.06309611]\n",
      " [0.10383711]\n",
      " [0.1536539 ]\n",
      " [0.08921878]\n",
      " [0.09226205]\n",
      " [0.08937639]\n",
      " [0.09362764]\n",
      " [0.06361021]\n",
      " [0.11342435]\n",
      " [0.12203544]\n",
      " [0.0484788 ]\n",
      " [0.0726892 ]\n",
      " [0.07704362]\n",
      " [0.08989255]\n",
      " [0.1483725 ]\n",
      " [0.17888895]\n",
      " [0.07242237]\n",
      " [0.09300068]\n",
      " [0.0987631 ]\n",
      " [0.10366511]\n",
      " [0.07716235]\n",
      " [0.09364244]\n",
      " [0.21208161]\n",
      " [0.0847851 ]\n",
      " [0.07765972]\n",
      " [0.16173476]\n",
      " [0.07798186]\n",
      " [0.0968655 ]\n",
      " [0.08401361]\n",
      " [0.10358417]\n",
      " [0.12317818]\n",
      " [0.09091545]\n",
      " [0.06854989]\n",
      " [0.04194857]\n",
      " [0.07807712]\n",
      " [0.17801604]\n",
      " [0.18996412]\n",
      " [0.08416358]\n",
      " [0.08356036]\n",
      " [0.04033514]\n",
      " [0.08280215]\n",
      " [0.351838  ]\n",
      " [0.08521655]\n",
      " [0.07041229]\n",
      " [0.14190742]\n",
      " [0.08312036]\n",
      " [0.0550795 ]\n",
      " [0.07295591]\n",
      " [0.04216715]\n",
      " [0.1391168 ]\n",
      " [0.19439626]\n",
      " [0.090324  ]\n",
      " [0.094038  ]\n",
      " [0.13300769]\n",
      " [0.19801924]\n",
      " [0.08042092]\n",
      " [0.07042584]\n",
      " [0.24613976]\n",
      " [0.06634498]\n",
      " [0.03947198]\n",
      " [0.06241432]\n",
      " [0.1239761 ]\n",
      " [0.07215818]\n",
      " [0.09768298]\n",
      " [0.12458282]\n",
      " [0.08637262]\n",
      " [0.23398378]\n",
      " [0.12458894]\n",
      " [0.09637645]\n",
      " [0.05601443]\n",
      " [0.21359117]\n",
      " [0.08543842]\n",
      " [0.06947459]\n",
      " [0.13311271]\n",
      " [0.04384422]\n",
      " [0.04054412]\n",
      " [0.12847492]\n",
      " [0.08502092]\n",
      " [0.13755988]\n",
      " [0.17034587]\n",
      " [0.09139869]\n",
      " [0.14955027]\n",
      " [0.05738296]\n",
      " [0.07238664]\n",
      " [0.05611105]\n",
      " [0.07269352]\n",
      " [0.08315101]\n",
      " [0.08661301]\n",
      " [0.10074744]\n",
      " [0.08913169]\n",
      " [0.07470925]\n",
      " [0.05745292]\n",
      " [0.05664563]\n",
      " [0.17457037]\n",
      " [0.06874553]\n",
      " [0.10605385]\n",
      " [0.07159831]\n",
      " [0.08631822]\n",
      " [0.0695505 ]\n",
      " [0.10624849]\n",
      " [0.05636736]\n",
      " [0.08855654]\n",
      " [0.12948172]\n",
      " [0.0701613 ]\n",
      " [0.09842585]\n",
      " [0.09790938]\n",
      " [0.09853742]\n",
      " [0.1159473 ]\n",
      " [0.19024232]\n",
      " [0.09895543]\n",
      " [0.08555484]\n",
      " [0.12343737]\n",
      " [0.10260829]\n",
      " [0.12717889]\n",
      " [0.12051082]\n",
      " [0.17487064]\n",
      " [0.08038022]\n",
      " [0.07628524]\n",
      " [0.10554281]\n",
      " [0.12572402]\n",
      " [0.04728752]\n",
      " [0.06817628]\n",
      " [0.0849644 ]\n",
      " [0.07854801]\n",
      " [0.07893162]\n",
      " [0.17418633]\n",
      " [0.06619526]\n",
      " [0.05468079]\n",
      " [0.0927585 ]\n",
      " [0.07274184]\n",
      " [0.101167  ]\n",
      " [0.08196636]\n",
      " [0.04522527]\n",
      " [0.07291023]\n",
      " [0.11460818]\n",
      " [0.14773887]\n",
      " [0.10503142]\n",
      " [0.06605305]\n",
      " [0.14959964]\n",
      " [0.0932742 ]\n",
      " [0.13788292]\n",
      " [0.05115324]\n",
      " [0.05721078]\n",
      " [0.10628039]\n",
      " [0.04562756]\n",
      " [0.03971986]\n",
      " [0.13441174]\n",
      " [0.1457999 ]\n",
      " [0.06313246]\n",
      " [0.1288263 ]\n",
      " [0.1896823 ]\n",
      " [0.0960842 ]\n",
      " [0.05990406]\n",
      " [0.0595587 ]\n",
      " [0.18650457]\n",
      " [0.10715944]\n",
      " [0.07689849]\n",
      " [0.07839032]\n",
      " [0.05250907]\n",
      " [0.16892144]\n",
      " [0.1405654 ]\n",
      " [0.12762426]\n",
      " [0.05129761]\n",
      " [0.14043656]\n",
      " [0.06275405]\n",
      " [0.10720003]\n",
      " [0.1035822 ]\n",
      " [0.17719686]\n",
      " [0.10086074]\n",
      " [0.05902761]\n",
      " [0.10755255]\n",
      " [0.21397623]\n",
      " [0.15067364]\n",
      " [0.1218803 ]\n",
      " [0.04714871]\n",
      " [0.10240211]\n",
      " [0.11403668]\n",
      " [0.07484429]\n",
      " [0.12911026]\n",
      " [0.04358187]\n",
      " [0.0766795 ]\n",
      " [0.04497253]\n",
      " [0.17590275]\n",
      " [0.03980412]\n",
      " [0.16338935]\n",
      " [0.10509849]\n",
      " [0.11948133]\n",
      " [0.08689562]\n",
      " [0.07870036]\n",
      " [0.11041331]\n",
      " [0.1671567 ]\n",
      " [0.09175729]\n",
      " [0.04945696]\n",
      " [0.0719942 ]\n",
      " [0.10931629]\n",
      " [0.04270582]\n",
      " [0.05428628]\n",
      " [0.15721789]\n",
      " [0.07297629]\n",
      " [0.0613761 ]\n",
      " [0.26909187]\n",
      " [0.06084234]\n",
      " [0.11179163]\n",
      " [0.14310671]\n",
      " [0.10663591]\n",
      " [0.07628379]\n",
      " [0.22437274]\n",
      " [0.05410669]\n",
      " [0.16246767]\n",
      " [0.12334765]\n",
      " [0.19091985]\n",
      " [0.18825665]\n",
      " [0.11961535]\n",
      " [0.05711838]\n",
      " [0.18082365]\n",
      " [0.16510902]\n",
      " [0.13912977]\n",
      " [0.19015756]\n",
      " [0.08758046]\n",
      " [0.1664424 ]\n",
      " [0.0732474 ]\n",
      " [0.04412513]\n",
      " [0.04704016]\n",
      " [0.05230836]\n",
      " [0.03902209]\n",
      " [0.22323269]\n",
      " [0.30979943]\n",
      " [0.08656264]\n",
      " [0.04012319]\n",
      " [0.05308929]\n",
      " [0.0584396 ]\n",
      " [0.04637527]\n",
      " [0.14614579]\n",
      " [0.19816186]\n",
      " [0.16527902]\n",
      " [0.04963809]\n",
      " [0.10629033]\n",
      " [0.16134079]\n",
      " [0.20260051]\n",
      " [0.05479839]\n",
      " [0.10397986]\n",
      " [0.11746453]\n",
      " [0.09109666]\n",
      " [0.32685053]\n",
      " [0.07290136]\n",
      " [0.06538285]\n",
      " [0.07469516]\n",
      " [0.10313042]\n",
      " [0.07703613]\n",
      " [0.12692386]\n",
      " [0.04800979]\n",
      " [0.08423074]\n",
      " [0.05157005]\n",
      " [0.24141249]\n",
      " [0.10444785]\n",
      " [0.0917201 ]\n",
      " [0.0836187 ]\n",
      " [0.05304091]\n",
      " [0.06208289]\n",
      " [0.04074948]\n",
      " [0.059421  ]\n",
      " [0.1804012 ]\n",
      " [0.18486597]\n",
      " [0.12036658]\n",
      " [0.05176074]\n",
      " [0.11212887]\n",
      " [0.10375756]\n",
      " [0.1254198 ]\n",
      " [0.1406226 ]\n",
      " [0.06826903]\n",
      " [0.12302253]\n",
      " [0.09135368]\n",
      " [0.09280576]\n",
      " [0.0839834 ]\n",
      " [0.08899635]\n",
      " [0.05054139]\n",
      " [0.06479907]\n",
      " [0.07425158]\n",
      " [0.05775268]\n",
      " [0.20121664]\n",
      " [0.09252606]\n",
      " [0.20276588]\n",
      " [0.2704094 ]\n",
      " [0.14652224]\n",
      " [0.08173151]\n",
      " [0.07729243]\n",
      " [0.14558946]\n",
      " [0.14844272]\n",
      " [0.07936163]\n",
      " [0.17556572]\n",
      " [0.05993511]\n",
      " [0.11830594]\n",
      " [0.04045985]\n",
      " [0.04738463]\n",
      " [0.14001338]\n",
      " [0.05847161]\n",
      " [0.06834563]\n",
      " [0.16076833]\n",
      " [0.18014017]\n",
      " [0.04285119]\n",
      " [0.08723901]\n",
      " [0.07127853]\n",
      " [0.07085776]\n",
      " [0.0638625 ]\n",
      " [0.07150842]\n",
      " [0.06255889]\n",
      " [0.07598052]\n",
      " [0.05441099]\n",
      " [0.08266545]\n",
      " [0.08556321]\n",
      " [0.15882316]\n",
      " [0.11108672]\n",
      " [0.07955983]\n",
      " [0.04836693]\n",
      " [0.14748825]\n",
      " [0.11206845]\n",
      " [0.16762948]\n",
      " [0.06108511]\n",
      " [0.07068415]\n",
      " [0.07909302]\n",
      " [0.04415403]\n",
      " [0.11489324]\n",
      " [0.10122157]\n",
      " [0.0460813 ]\n",
      " [0.09051048]\n",
      " [0.10423304]\n",
      " [0.06758683]\n",
      " [0.18476897]\n",
      " [0.27966958]\n",
      " [0.04401546]\n",
      " [0.04335023]\n",
      " [0.03907812]\n",
      " [0.08110735]\n",
      " [0.15952131]\n",
      " [0.16319229]\n",
      " [0.04315118]\n",
      " [0.15543002]\n",
      " [0.16173336]\n",
      " [0.29363427]\n",
      " [0.36051774]\n",
      " [0.06079108]\n",
      " [0.11355067]\n",
      " [0.10782587]\n",
      " [0.06699616]\n",
      " [0.07815964]\n",
      " [0.13881071]\n",
      " [0.04199696]\n",
      " [0.10896341]\n",
      " [0.08667348]\n",
      " [0.12599969]\n",
      " [0.10177372]\n",
      " [0.07384866]\n",
      " [0.07674374]\n",
      " [0.17587014]\n",
      " [0.05098251]\n",
      " [0.05412371]\n",
      " [0.11918433]\n",
      " [0.0668574 ]\n",
      " [0.03902911]\n",
      " [0.04228286]\n",
      " [0.12339736]\n",
      " [0.09534957]\n",
      " [0.04024804]\n",
      " [0.19249432]\n",
      " [0.23439857]\n",
      " [0.08446395]\n",
      " [0.13511042]\n",
      " [0.14187086]\n",
      " [0.26245737]\n",
      " [0.13311553]\n",
      " [0.0725586 ]\n",
      " [0.06834319]\n",
      " [0.10353477]\n",
      " [0.14308453]\n",
      " [0.07770641]\n",
      " [0.07432695]\n",
      " [0.08645577]\n",
      " [0.08421083]\n",
      " [0.08777874]\n",
      " [0.09781213]\n",
      " [0.10672573]\n",
      " [0.15876403]\n",
      " [0.07084502]\n",
      " [0.23777267]\n",
      " [0.10775535]\n",
      " [0.05848301]\n",
      " [0.05047812]\n",
      " [0.06280246]\n",
      " [0.17968188]\n",
      " [0.12256248]\n",
      " [0.07354321]\n",
      " [0.10715386]\n",
      " [0.11777415]\n",
      " [0.13732935]\n",
      " [0.07364207]\n",
      " [0.05459735]\n",
      " [0.09210534]\n",
      " [0.0453199 ]\n",
      " [0.06965782]\n",
      " [0.05190578]\n",
      " [0.14527571]\n",
      " [0.07730195]\n",
      " [0.1471614 ]\n",
      " [0.06850854]\n",
      " [0.12360177]\n",
      " [0.20550637]\n",
      " [0.2737168 ]\n",
      " [0.12378915]\n",
      " [0.12525302]\n",
      " [0.09604269]\n",
      " [0.06223024]\n",
      " [0.10264799]\n",
      " [0.15040806]\n",
      " [0.10467983]\n",
      " [0.0776111 ]\n",
      " [0.15644246]\n",
      " [0.10513694]\n",
      " [0.07192388]\n",
      " [0.17062566]\n",
      " [0.21470758]\n",
      " [0.07330107]\n",
      " [0.07430875]\n",
      " [0.10770976]\n",
      " [0.07914807]\n",
      " [0.03906333]\n",
      " [0.05638668]\n",
      " [0.07648861]\n",
      " [0.04729964]\n",
      " [0.25924015]\n",
      " [0.06297491]\n",
      " [0.04067366]\n",
      " [0.04477013]\n",
      " [0.07663626]\n",
      " [0.03923129]\n",
      " [0.15700905]\n",
      " [0.16426432]\n",
      " [0.36571893]\n",
      " [0.12321281]\n",
      " [0.06550536]\n",
      " [0.13833351]\n",
      " [0.14892882]\n",
      " [0.09467307]\n",
      " [0.05824236]\n",
      " [0.05166679]\n",
      " [0.05271224]\n",
      " [0.27940974]\n",
      " [0.07990309]\n",
      " [0.1627896 ]\n",
      " [0.0473761 ]\n",
      " [0.1782189 ]\n",
      " [0.14624439]\n",
      " [0.08396647]\n",
      " [0.2341946 ]\n",
      " [0.06830078]\n",
      " [0.11424285]\n",
      " [0.174566  ]\n",
      " [0.08946941]\n",
      " [0.16659164]\n",
      " [0.12666763]\n",
      " [0.03904583]\n",
      " [0.04075967]\n",
      " [0.04503496]\n",
      " [0.08386146]\n",
      " [0.19496743]\n",
      " [0.28218612]\n",
      " [0.106359  ]\n",
      " [0.08982681]\n",
      " [0.11313485]\n",
      " [0.04138222]\n",
      " [0.10504856]\n",
      " [0.23016027]\n",
      " [0.05186721]\n",
      " [0.1359144 ]\n",
      " [0.3115    ]\n",
      " [0.2176362 ]\n",
      " [0.05706485]\n",
      " [0.19324964]\n",
      " [0.05443583]\n",
      " [0.03916164]\n",
      " [0.07421615]\n",
      " [0.2074649 ]\n",
      " [0.10354423]\n",
      " [0.04301214]\n",
      " [0.04023672]\n",
      " [0.05138351]\n",
      " [0.1404107 ]\n",
      " [0.05081914]\n",
      " [0.26188523]\n",
      " [0.05625257]\n",
      " [0.17029831]\n",
      " [0.21004477]\n",
      " [0.0640847 ]\n",
      " [0.04130775]\n",
      " [0.05556066]\n",
      " [0.12764552]\n",
      " [0.1129968 ]\n",
      " [0.04375901]\n",
      " [0.05319725]\n",
      " [0.05779754]\n",
      " [0.09984026]\n",
      " [0.17898004]\n",
      " [0.0673922 ]\n",
      " [0.09946589]\n",
      " [0.07079645]\n",
      " [0.0749694 ]\n",
      " [0.11429054]\n",
      " [0.03956506]\n",
      " [0.08612087]\n",
      " [0.19976653]\n",
      " [0.13926375]\n",
      " [0.13520256]\n",
      " [0.0448415 ]\n",
      " [0.06330276]\n",
      " [0.12247794]\n",
      " [0.17322487]\n",
      " [0.27627435]\n",
      " [0.22393744]\n",
      " [0.06736761]\n",
      " [0.1442805 ]\n",
      " [0.04328918]\n",
      " [0.35127914]\n",
      " [0.0744338 ]\n",
      " [0.06209586]\n",
      " [0.04393955]\n",
      " [0.1970163 ]\n",
      " [0.54251504]\n",
      " [0.18633106]\n",
      " [0.0602655 ]\n",
      " [0.3384287 ]\n",
      " [0.40362096]\n",
      " [0.06321047]\n",
      " [0.04292741]\n",
      " [0.14054517]\n",
      " [0.13842621]\n",
      " [0.29144377]\n",
      " [0.05112391]\n",
      " [0.11599484]\n",
      " [0.15658696]\n",
      " [0.0423561 ]\n",
      " [0.12399867]\n",
      " [0.19926706]\n",
      " [0.06465083]\n",
      " [0.05566826]\n",
      " [0.2558288 ]\n",
      " [0.11260549]\n",
      " [0.06347404]\n",
      " [0.0592479 ]\n",
      " [0.1322095 ]\n",
      " [0.06763239]\n",
      " [0.07595536]\n",
      " [0.04754046]\n",
      " [0.12085115]\n",
      " [0.32286972]\n",
      " [0.14080746]\n",
      " [0.10829652]\n",
      " [0.13213758]\n",
      " [0.29845166]\n",
      " [0.21464154]\n",
      " [0.08965924]\n",
      " [0.0886182 ]\n",
      " [0.04882848]\n",
      " [0.19890401]\n",
      " [0.09038493]\n",
      " [0.07977298]\n",
      " [0.38082987]\n",
      " [0.04234594]\n",
      " [0.04890163]\n",
      " [0.12241169]\n",
      " [0.06655428]\n",
      " [0.1523828 ]\n",
      " [0.27295265]\n",
      " [0.26711705]\n",
      " [0.05222351]\n",
      " [0.15910006]\n",
      " [0.03930838]\n",
      " [0.15518692]\n",
      " [0.11270371]\n",
      " [0.09985731]\n",
      " [0.17041916]\n",
      " [0.22433954]\n",
      " [0.05150192]\n",
      " [0.05214962]\n",
      " [0.065662  ]\n",
      " [0.44320637]\n",
      " [0.10856825]\n",
      " [0.12129713]\n",
      " [0.07197266]\n",
      " [0.06619185]\n",
      " [0.14717956]\n",
      " [0.20744097]\n",
      " [0.10215774]\n",
      " [0.044049  ]\n",
      " [0.11437257]\n",
      " [0.1028975 ]\n",
      " [0.17466515]\n",
      " [0.04012553]\n",
      " [0.05688559]\n",
      " [0.11823516]\n",
      " [0.0390266 ]\n",
      " [0.05346045]\n",
      " [0.11076794]\n",
      " [0.10920667]\n",
      " [0.24501139]\n",
      " [0.2769828 ]\n",
      " [0.26970574]\n",
      " [0.04431426]\n",
      " [0.07186571]\n",
      " [0.05937576]\n",
      " [0.04336476]\n",
      " [0.0899944 ]\n",
      " [0.03905579]\n",
      " [0.05447084]\n",
      " [0.64265496]\n",
      " [0.08629996]\n",
      " [0.11342337]\n",
      " [0.06962883]\n",
      " [0.10419587]\n",
      " [0.10695618]\n",
      " [0.04544956]\n",
      " [0.04542078]\n",
      " [0.23572397]\n",
      " [0.05525992]\n",
      " [0.56321853]\n",
      " [0.06436732]\n",
      " [0.12739047]\n",
      " [0.11502165]\n",
      " [0.13451613]\n",
      " [0.05941521]\n",
      " [0.04320174]\n",
      " [0.04557521]\n",
      " [0.16716781]\n",
      " [0.17802003]\n",
      " [0.04752253]\n",
      " [0.17790721]\n",
      " [0.13578877]\n",
      " [0.12973425]\n",
      " [0.1690934 ]\n",
      " [0.20912412]\n",
      " [0.04021138]\n",
      " [0.06137719]\n",
      " [0.05895098]\n",
      " [0.0935679 ]\n",
      " [0.10359865]\n",
      " [0.054229  ]\n",
      " [0.03908527]\n",
      " [0.03905275]\n",
      " [0.23121697]\n",
      " [0.3259227 ]\n",
      " [0.05817346]\n",
      " [0.05603838]\n",
      " [0.04011051]\n",
      " [0.83375764]\n",
      " [1.2164607 ]\n",
      " [0.06361629]\n",
      " [0.04951138]\n",
      " [0.23087534]\n",
      " [0.52698976]\n",
      " [0.10014047]\n",
      " [0.1024012 ]\n",
      " [0.05279292]\n",
      " [0.03902388]\n",
      " [0.04581282]\n",
      " [0.10970807]\n",
      " [0.06062356]\n",
      " [0.226128  ]\n",
      " [0.24146244]\n",
      " [0.36140725]\n",
      " [0.04188195]\n",
      " [0.1539543 ]\n",
      " [0.28448513]\n",
      " [0.12447589]\n",
      " [0.05030377]\n",
      " [0.12555405]\n",
      " [0.09099882]\n",
      " [0.12084267]\n",
      " [0.38327402]\n",
      " [0.0620934 ]\n",
      " [0.1352377 ]\n",
      " [0.05891329]\n",
      " [0.2131604 ]\n",
      " [0.07309241]\n",
      " [0.04634958]\n",
      " [0.04829334]\n",
      " [0.04455582]\n",
      " [0.07185476]\n",
      " [0.20556656]\n",
      " [0.13051037]\n",
      " [0.06563652]\n",
      " [0.03918429]\n",
      " [0.04097685]\n",
      " [0.16076955]\n",
      " [0.04445983]\n",
      " [0.04591167]\n",
      " [0.12170647]\n",
      " [0.33749643]\n",
      " [0.34363645]]\n"
     ]
    }
   ],
   "source": [
    "# Try predict on training set\n",
    "y_predict = model.predict(seq_train_x)\n",
    "print(y_predict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.11591307, 0.07768746, 0.08994175, 0.08649451, 0.09976438,\n",
       "       0.08351678, 0.08351678, 0.18918256, 0.16088796, 0.09595893,\n",
       "       0.07843922, 0.04918706, 0.05447334, 0.10325765, 0.08472119,\n",
       "       0.11494303, 0.10011806, 0.1259548 , 0.05766349, 0.12219774,\n",
       "       0.08352318, 0.13316976, 0.2355689 , 0.13506085, 0.07856027,\n",
       "       0.09038219, 0.06882396, 0.10740753, 0.08706232, 0.0999878 ,\n",
       "       0.12270915, 0.09574346, 0.04638261, 0.09749703, 0.13050935,\n",
       "       0.11915668, 0.10564683, 0.04116147, 0.16195314, 0.0599192 ,\n",
       "       0.07573151, 0.08875289, 0.08737619, 0.04477709, 0.10230387,\n",
       "       0.08419205, 0.2500633 , 0.09201651, 0.14181879, 0.069013  ,\n",
       "       0.15904965, 0.08564888, 0.09456676, 0.06818675, 0.09404088,\n",
       "       0.08847724, 0.20005877, 0.12630151, 0.09563334, 0.07236986,\n",
       "       0.07073411, 0.06309611, 0.10383711, 0.1536539 , 0.08921878,\n",
       "       0.09226205, 0.08937639, 0.09362764, 0.06361021, 0.11342435,\n",
       "       0.12203544, 0.0484788 , 0.0726892 , 0.07704362, 0.08989255,\n",
       "       0.1483725 , 0.17888895, 0.07242237, 0.09300068, 0.0987631 ,\n",
       "       0.10366511, 0.07716235, 0.09364244, 0.21208161, 0.0847851 ,\n",
       "       0.07765972, 0.16173476, 0.07798186, 0.0968655 , 0.08401361,\n",
       "       0.10358417, 0.12317818, 0.09091545, 0.06854989, 0.04194857,\n",
       "       0.07807712, 0.17801604, 0.18996412, 0.08416358, 0.08356036,\n",
       "       0.04033514, 0.08280215, 0.351838  , 0.08521655, 0.07041229,\n",
       "       0.14190742, 0.08312036, 0.0550795 , 0.07295591, 0.04216715,\n",
       "       0.1391168 , 0.19439626, 0.090324  , 0.094038  , 0.13300769,\n",
       "       0.19801924, 0.08042092, 0.07042584, 0.24613976, 0.06634498,\n",
       "       0.03947198, 0.06241432, 0.1239761 , 0.07215818, 0.09768298,\n",
       "       0.12458282, 0.08637262, 0.23398378, 0.12458894, 0.09637645,\n",
       "       0.05601443, 0.21359117, 0.08543842, 0.06947459, 0.13311271,\n",
       "       0.04384422, 0.04054412, 0.12847492, 0.08502092, 0.13755988,\n",
       "       0.17034587, 0.09139869, 0.14955027, 0.05738296, 0.07238664,\n",
       "       0.05611105, 0.07269352, 0.08315101, 0.08661301, 0.10074744,\n",
       "       0.08913169, 0.07470925, 0.05745292, 0.05664563, 0.17457037,\n",
       "       0.06874553, 0.10605385, 0.07159831, 0.08631822, 0.0695505 ,\n",
       "       0.10624849, 0.05636736, 0.08855654, 0.12948172, 0.0701613 ,\n",
       "       0.09842585, 0.09790938, 0.09853742, 0.1159473 , 0.19024232,\n",
       "       0.09895543, 0.08555484, 0.12343737, 0.10260829, 0.12717889,\n",
       "       0.12051082, 0.17487064, 0.08038022, 0.07628524, 0.10554281,\n",
       "       0.12572402, 0.04728752, 0.06817628, 0.0849644 , 0.07854801,\n",
       "       0.07893162, 0.17418633, 0.06619526, 0.05468079, 0.0927585 ,\n",
       "       0.07274184, 0.101167  , 0.08196636, 0.04522527, 0.07291023,\n",
       "       0.11460818, 0.14773887, 0.10503142, 0.06605305, 0.14959964,\n",
       "       0.0932742 , 0.13788292, 0.05115324, 0.05721078, 0.10628039,\n",
       "       0.04562756, 0.03971986, 0.13441174, 0.1457999 , 0.06313246,\n",
       "       0.1288263 , 0.1896823 , 0.0960842 , 0.05990406, 0.0595587 ,\n",
       "       0.18650457, 0.10715944, 0.07689849, 0.07839032, 0.05250907,\n",
       "       0.16892144, 0.1405654 , 0.12762426, 0.05129761, 0.14043656,\n",
       "       0.06275405, 0.10720003, 0.1035822 , 0.17719686, 0.10086074,\n",
       "       0.05902761, 0.10755255, 0.21397623, 0.15067364, 0.1218803 ,\n",
       "       0.04714871, 0.10240211, 0.11403668, 0.07484429, 0.12911026,\n",
       "       0.04358187, 0.0766795 , 0.04497253, 0.17590275, 0.03980412,\n",
       "       0.16338935, 0.10509849, 0.11948133, 0.08689562, 0.07870036,\n",
       "       0.11041331, 0.1671567 , 0.09175729, 0.04945696, 0.0719942 ,\n",
       "       0.10931629, 0.04270582, 0.05428628, 0.15721789, 0.07297629,\n",
       "       0.0613761 , 0.26909187, 0.06084234, 0.11179163, 0.14310671,\n",
       "       0.10663591, 0.07628379, 0.22437274, 0.05410669, 0.16246767,\n",
       "       0.12334765, 0.19091985, 0.18825665, 0.11961535, 0.05711838,\n",
       "       0.18082365, 0.16510902, 0.13912977, 0.19015756, 0.08758046,\n",
       "       0.1664424 , 0.0732474 , 0.04412513, 0.04704016, 0.05230836,\n",
       "       0.03902209, 0.22323269, 0.30979943, 0.08656264, 0.04012319,\n",
       "       0.05308929, 0.0584396 , 0.04637527, 0.14614579, 0.19816186,\n",
       "       0.16527902, 0.04963809, 0.10629033, 0.16134079, 0.20260051,\n",
       "       0.05479839, 0.10397986, 0.11746453, 0.09109666, 0.32685053,\n",
       "       0.07290136, 0.06538285, 0.07469516, 0.10313042, 0.07703613,\n",
       "       0.12692386, 0.04800979, 0.08423074, 0.05157005, 0.24141249,\n",
       "       0.10444785, 0.0917201 , 0.0836187 , 0.05304091, 0.06208289,\n",
       "       0.04074948, 0.059421  , 0.1804012 , 0.18486597, 0.12036658,\n",
       "       0.05176074, 0.11212887, 0.10375756, 0.1254198 , 0.1406226 ,\n",
       "       0.06826903, 0.12302253, 0.09135368, 0.09280576, 0.0839834 ,\n",
       "       0.08899635, 0.05054139, 0.06479907, 0.07425158, 0.05775268,\n",
       "       0.20121664, 0.09252606, 0.20276588, 0.2704094 , 0.14652224,\n",
       "       0.08173151, 0.07729243, 0.14558946, 0.14844272, 0.07936163,\n",
       "       0.17556572, 0.05993511, 0.11830594, 0.04045985, 0.04738463,\n",
       "       0.14001338, 0.05847161, 0.06834563, 0.16076833, 0.18014017,\n",
       "       0.04285119, 0.08723901, 0.07127853, 0.07085776, 0.0638625 ,\n",
       "       0.07150842, 0.06255889, 0.07598052, 0.05441099, 0.08266545,\n",
       "       0.08556321, 0.15882316, 0.11108672, 0.07955983, 0.04836693,\n",
       "       0.14748825, 0.11206845, 0.16762948, 0.06108511, 0.07068415,\n",
       "       0.07909302, 0.04415403, 0.11489324, 0.10122157, 0.0460813 ,\n",
       "       0.09051048, 0.10423304, 0.06758683, 0.18476897, 0.27966958,\n",
       "       0.04401546, 0.04335023, 0.03907812, 0.08110735, 0.15952131,\n",
       "       0.16319229, 0.04315118, 0.15543002, 0.16173336, 0.29363427,\n",
       "       0.36051774, 0.06079108, 0.11355067, 0.10782587, 0.06699616,\n",
       "       0.07815964, 0.13881071, 0.04199696, 0.10896341, 0.08667348,\n",
       "       0.12599969, 0.10177372, 0.07384866, 0.07674374, 0.17587014,\n",
       "       0.05098251, 0.05412371, 0.11918433, 0.0668574 , 0.03902911,\n",
       "       0.04228286, 0.12339736, 0.09534957, 0.04024804, 0.19249432,\n",
       "       0.23439857, 0.08446395, 0.13511042, 0.14187086, 0.26245737,\n",
       "       0.13311553, 0.0725586 , 0.06834319, 0.10353477, 0.14308453,\n",
       "       0.07770641, 0.07432695, 0.08645577, 0.08421083, 0.08777874,\n",
       "       0.09781213, 0.10672573, 0.15876403, 0.07084502, 0.23777267,\n",
       "       0.10775535, 0.05848301, 0.05047812, 0.06280246, 0.17968188,\n",
       "       0.12256248, 0.07354321, 0.10715386, 0.11777415, 0.13732935,\n",
       "       0.07364207, 0.05459735, 0.09210534, 0.0453199 , 0.06965782,\n",
       "       0.05190578, 0.14527571, 0.07730195, 0.1471614 , 0.06850854,\n",
       "       0.12360177, 0.20550637, 0.2737168 , 0.12378915, 0.12525302,\n",
       "       0.09604269, 0.06223024, 0.10264799, 0.15040806, 0.10467983,\n",
       "       0.0776111 , 0.15644246, 0.10513694, 0.07192388, 0.17062566,\n",
       "       0.21470758, 0.07330107, 0.07430875, 0.10770976, 0.07914807,\n",
       "       0.03906333, 0.05638668, 0.07648861, 0.04729964, 0.25924015,\n",
       "       0.06297491, 0.04067366, 0.04477013, 0.07663626, 0.03923129,\n",
       "       0.15700905, 0.16426432, 0.36571893, 0.12321281, 0.06550536,\n",
       "       0.13833351, 0.14892882, 0.09467307, 0.05824236, 0.05166679,\n",
       "       0.05271224, 0.27940974, 0.07990309, 0.1627896 , 0.0473761 ,\n",
       "       0.1782189 , 0.14624439, 0.08396647, 0.2341946 , 0.06830078,\n",
       "       0.11424285, 0.174566  , 0.08946941, 0.16659164, 0.12666763,\n",
       "       0.03904583, 0.04075967, 0.04503496, 0.08386146, 0.19496743,\n",
       "       0.28218612, 0.106359  , 0.08982681, 0.11313485, 0.04138222,\n",
       "       0.10504856, 0.23016027, 0.05186721, 0.1359144 , 0.3115    ,\n",
       "       0.2176362 , 0.05706485, 0.19324964, 0.05443583, 0.03916164,\n",
       "       0.07421615, 0.2074649 , 0.10354423, 0.04301214, 0.04023672,\n",
       "       0.05138351, 0.1404107 , 0.05081914, 0.26188523, 0.05625257,\n",
       "       0.17029831, 0.21004477, 0.0640847 , 0.04130775, 0.05556066,\n",
       "       0.12764552, 0.1129968 , 0.04375901, 0.05319725, 0.05779754,\n",
       "       0.09984026, 0.17898004, 0.0673922 , 0.09946589, 0.07079645,\n",
       "       0.0749694 , 0.11429054, 0.03956506, 0.08612087, 0.19976653,\n",
       "       0.13926375, 0.13520256, 0.0448415 , 0.06330276, 0.12247794,\n",
       "       0.17322487, 0.27627435, 0.22393744, 0.06736761, 0.1442805 ,\n",
       "       0.04328918, 0.35127914, 0.0744338 , 0.06209586, 0.04393955,\n",
       "       0.1970163 , 0.54251504, 0.18633106, 0.0602655 , 0.3384287 ,\n",
       "       0.40362096, 0.06321047, 0.04292741, 0.14054517, 0.13842621,\n",
       "       0.29144377, 0.05112391, 0.11599484, 0.15658696, 0.0423561 ,\n",
       "       0.12399867, 0.19926706, 0.06465083, 0.05566826, 0.2558288 ,\n",
       "       0.11260549, 0.06347404, 0.0592479 , 0.1322095 , 0.06763239,\n",
       "       0.07595536, 0.04754046, 0.12085115, 0.32286972, 0.14080746,\n",
       "       0.10829652, 0.13213758, 0.29845166, 0.21464154, 0.08965924,\n",
       "       0.0886182 , 0.04882848, 0.19890401, 0.09038493, 0.07977298,\n",
       "       0.38082987, 0.04234594, 0.04890163, 0.12241169, 0.06655428,\n",
       "       0.1523828 , 0.27295265, 0.26711705, 0.05222351, 0.15910006,\n",
       "       0.03930838, 0.15518692, 0.11270371, 0.09985731, 0.17041916,\n",
       "       0.22433954, 0.05150192, 0.05214962, 0.065662  , 0.44320637,\n",
       "       0.10856825, 0.12129713, 0.07197266, 0.06619185, 0.14717956,\n",
       "       0.20744097, 0.10215774, 0.044049  , 0.11437257, 0.1028975 ,\n",
       "       0.17466515, 0.04012553, 0.05688559, 0.11823516, 0.0390266 ,\n",
       "       0.05346045, 0.11076794, 0.10920667, 0.24501139, 0.2769828 ,\n",
       "       0.26970574, 0.04431426, 0.07186571, 0.05937576, 0.04336476,\n",
       "       0.0899944 , 0.03905579, 0.05447084, 0.64265496, 0.08629996,\n",
       "       0.11342337, 0.06962883, 0.10419587, 0.10695618, 0.04544956,\n",
       "       0.04542078, 0.23572397, 0.05525992, 0.56321853, 0.06436732,\n",
       "       0.12739047, 0.11502165, 0.13451613, 0.05941521, 0.04320174,\n",
       "       0.04557521, 0.16716781, 0.17802003, 0.04752253, 0.17790721,\n",
       "       0.13578877, 0.12973425, 0.1690934 , 0.20912412, 0.04021138,\n",
       "       0.06137719, 0.05895098, 0.0935679 , 0.10359865, 0.054229  ,\n",
       "       0.03908527, 0.03905275, 0.23121697, 0.3259227 , 0.05817346,\n",
       "       0.05603838, 0.04011051, 0.83375764, 1.2164607 , 0.06361629,\n",
       "       0.04951138, 0.23087534, 0.52698976, 0.10014047, 0.1024012 ,\n",
       "       0.05279292, 0.03902388, 0.04581282, 0.10970807, 0.06062356,\n",
       "       0.226128  , 0.24146244, 0.36140725, 0.04188195, 0.1539543 ,\n",
       "       0.28448513, 0.12447589, 0.05030377, 0.12555405, 0.09099882,\n",
       "       0.12084267, 0.38327402, 0.0620934 , 0.1352377 , 0.05891329,\n",
       "       0.2131604 , 0.07309241, 0.04634958, 0.04829334, 0.04455582,\n",
       "       0.07185476, 0.20556656, 0.13051037, 0.06563652, 0.03918429,\n",
       "       0.04097685, 0.16076955, 0.04445983, 0.04591167, 0.12170647,\n",
       "       0.33749643, 0.34363645], dtype=float32)"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predict.reshape(-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1f2de59aa88>"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,6))\n",
    "plt.plot(df_train.index[1:len(df_train.index)], y_predict, label = \"LSTM Predicted Value\")\n",
    "plt.plot(df_train.index, df_train['SP500_%Change'], label = \"Original Training Value\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In-sample prediction --- in test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.08055796]\n",
      " [0.12964767]\n",
      " [0.10198555]\n",
      " [0.03912818]\n",
      " [0.065107  ]\n",
      " [0.04224557]\n",
      " [0.14787018]\n",
      " [0.23594785]\n",
      " [0.25822207]\n",
      " [0.14566319]\n",
      " [0.04395886]\n",
      " [0.21865687]\n",
      " [0.3323489 ]\n",
      " [0.0428833 ]\n",
      " [0.04278328]\n",
      " [0.050108  ]\n",
      " [0.07869159]\n",
      " [0.23915121]\n",
      " [0.17968056]\n",
      " [0.17285338]\n",
      " [0.05163006]\n",
      " [0.40726265]\n",
      " [0.12934905]\n",
      " [0.09127562]\n",
      " [0.06120606]\n",
      " [0.1821869 ]\n",
      " [0.0713404 ]\n",
      " [0.17240307]\n",
      " [0.09202424]\n",
      " [0.12189733]\n",
      " [0.05788057]\n",
      " [0.10713873]\n",
      " [0.18535914]\n",
      " [0.05931533]\n",
      " [0.14147791]\n",
      " [0.11922828]\n",
      " [0.04456917]\n",
      " [0.22874728]\n",
      " [0.13868862]\n",
      " [0.08909868]\n",
      " [0.07226708]\n",
      " [0.12270256]\n",
      " [0.05536966]\n",
      " [0.03908725]\n",
      " [0.37609857]\n",
      " [0.06197071]\n",
      " [0.21877933]\n",
      " [0.12099478]\n",
      " [0.09532923]\n",
      " [0.07463141]\n",
      " [0.1978853 ]\n",
      " [0.09299889]\n",
      " [0.12899421]\n",
      " [0.12016179]\n",
      " [0.17876631]\n",
      " [0.06150211]\n",
      " [0.1868788 ]\n",
      " [0.06105186]\n",
      " [0.07605871]\n",
      " [0.05345339]\n",
      " [0.07116129]\n",
      " [0.16325141]\n",
      " [0.1419823 ]\n",
      " [0.20881373]\n",
      " [0.10503919]\n",
      " [0.07497792]\n",
      " [0.03920738]\n",
      " [0.04511365]\n",
      " [0.17247926]\n",
      " [0.07710302]\n",
      " [0.13784282]\n",
      " [0.11296924]\n",
      " [0.03963429]\n",
      " [0.24775356]\n",
      " [0.09070692]\n",
      " [0.10782973]\n",
      " [0.1153921 ]\n",
      " [0.03990555]\n",
      " [0.03927679]\n",
      " [0.06056724]\n",
      " [0.07398278]\n",
      " [0.18056837]\n",
      " [0.29187822]\n",
      " [0.1742896 ]\n",
      " [0.09179197]\n",
      " [0.04934623]\n",
      " [0.04015388]\n",
      " [0.04108975]\n",
      " [0.2270871 ]\n",
      " [0.07371918]\n",
      " [0.17357849]\n",
      " [0.10845967]\n",
      " [0.07387299]\n",
      " [0.04343185]\n",
      " [0.06537871]\n",
      " [0.11447133]\n",
      " [0.17812726]\n",
      " [0.05790534]\n",
      " [0.05899914]\n",
      " [0.04327422]\n",
      " [0.07176593]\n",
      " [0.13153218]\n",
      " [0.04815495]\n",
      " [0.08503833]\n",
      " [0.10171438]\n",
      " [0.04001878]\n",
      " [0.21610682]\n",
      " [0.32470867]\n",
      " [0.09815621]\n",
      " [0.10512843]\n",
      " [0.05082381]\n",
      " [0.04631846]\n",
      " [0.12020849]\n",
      " [0.06379408]\n",
      " [0.19564244]\n",
      " [0.335761  ]\n",
      " [0.07869112]\n",
      " [0.052127  ]\n",
      " [0.1043083 ]\n",
      " [0.27591932]\n",
      " [0.22853394]\n",
      " [0.06210747]\n",
      " [0.11576977]\n",
      " [0.1287211 ]\n",
      " [0.05721599]\n",
      " [0.24528417]\n",
      " [0.0702375 ]\n",
      " [0.19116223]\n",
      " [0.06735525]\n",
      " [0.20047879]\n",
      " [0.04636167]\n",
      " [0.13226086]\n",
      " [0.08994208]\n",
      " [0.20776078]\n",
      " [0.06706928]\n",
      " [0.16743648]\n",
      " [0.10435828]\n",
      " [0.06782638]\n",
      " [0.04682191]\n",
      " [0.07734302]\n",
      " [0.07000678]\n",
      " [0.0915049 ]\n",
      " [0.14033686]\n",
      " [0.04334749]\n",
      " [0.05540673]\n",
      " [0.2941926 ]\n",
      " [0.06171748]\n",
      " [0.04445173]\n",
      " [0.23253612]\n",
      " [0.17931625]\n",
      " [0.1643596 ]\n",
      " [0.08255424]\n",
      " [0.0495893 ]\n",
      " [0.04011615]\n",
      " [0.26708564]\n",
      " [0.04156394]\n",
      " [0.0390221 ]\n",
      " [0.3707887 ]\n",
      " [0.30180174]\n",
      " [0.06401101]\n",
      " [0.04924024]\n",
      " [0.0401462 ]\n",
      " [0.14897414]\n",
      " [0.12494475]\n",
      " [0.04296778]\n",
      " [0.26392445]\n",
      " [0.04328699]\n",
      " [0.8206141 ]\n",
      " [0.8441282 ]\n",
      " [0.05862468]\n",
      " [0.04215673]\n",
      " [0.04168926]\n",
      " [1.2086964 ]\n",
      " [0.04959729]\n",
      " [0.10865646]\n",
      " [0.5964617 ]\n",
      " [0.38726464]\n",
      " [0.16631882]\n",
      " [0.16349173]\n",
      " [0.10502797]\n",
      " [0.09391258]\n",
      " [0.1184344 ]\n",
      " [0.1395588 ]\n",
      " [0.7540614 ]\n",
      " [0.07678435]\n",
      " [0.10071672]\n",
      " [0.1189227 ]\n",
      " [0.08640493]\n",
      " [0.1576528 ]\n",
      " [0.15293346]\n",
      " [0.3066183 ]\n",
      " [0.04154981]\n",
      " [0.05756473]\n",
      " [0.04178538]\n",
      " [0.03905085]\n",
      " [0.5116389 ]\n",
      " [0.25455138]\n",
      " [0.06362435]\n",
      " [0.21009192]\n",
      " [0.93345875]\n",
      " [0.17711252]\n",
      " [0.14009503]\n",
      " [0.09615792]\n",
      " [0.13984403]\n",
      " [0.1699425 ]\n",
      " [0.0447375 ]\n",
      " [0.096984  ]\n",
      " [0.04700262]\n",
      " [0.1067852 ]\n",
      " [0.3116992 ]\n",
      " [0.22487095]\n",
      " [0.22545907]\n",
      " [0.07857653]\n",
      " [0.15740079]\n",
      " [0.2552079 ]\n",
      " [0.14814709]\n",
      " [0.04016707]\n",
      " [0.07122087]\n",
      " [0.04716424]\n",
      " [0.06356914]\n",
      " [0.15915208]\n",
      " [0.18399388]\n",
      " [0.11131997]\n",
      " [0.1271643 ]\n",
      " [0.15810817]\n",
      " [0.19079885]\n",
      " [0.40256166]\n",
      " [0.05568329]\n",
      " [0.11379069]\n",
      " [0.12796038]\n",
      " [0.09310699]\n",
      " [0.19842134]\n",
      " [0.06069349]\n",
      " [0.0543928 ]\n",
      " [0.44502777]\n",
      " [0.1882369 ]\n",
      " [0.05148494]\n",
      " [0.09922373]\n",
      " [0.04538801]\n",
      " [0.09425362]\n",
      " [0.07495689]]\n"
     ]
    }
   ],
   "source": [
    "#Predict on test set\n",
    "y_predict_test = model.predict(seq_test_x)\n",
    "print(y_predict_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.08055796, 0.12964767, 0.10198555, 0.03912818, 0.065107  ,\n",
       "        0.04224557, 0.14787018, 0.23594785, 0.25822207, 0.14566319,\n",
       "        0.04395886, 0.21865687, 0.3323489 , 0.0428833 , 0.04278328,\n",
       "        0.050108  , 0.07869159, 0.23915121, 0.17968056, 0.17285338,\n",
       "        0.05163006, 0.40726265, 0.12934905, 0.09127562, 0.06120606,\n",
       "        0.1821869 , 0.0713404 , 0.17240307, 0.09202424, 0.12189733,\n",
       "        0.05788057, 0.10713873, 0.18535914, 0.05931533, 0.14147791,\n",
       "        0.11922828, 0.04456917, 0.22874728, 0.13868862, 0.08909868,\n",
       "        0.07226708, 0.12270256, 0.05536966, 0.03908725, 0.37609857,\n",
       "        0.06197071, 0.21877933, 0.12099478, 0.09532923, 0.07463141,\n",
       "        0.1978853 , 0.09299889, 0.12899421, 0.12016179, 0.17876631,\n",
       "        0.06150211, 0.1868788 , 0.06105186, 0.07605871, 0.05345339,\n",
       "        0.07116129, 0.16325141, 0.1419823 , 0.20881373, 0.10503919,\n",
       "        0.07497792, 0.03920738, 0.04511365, 0.17247926, 0.07710302,\n",
       "        0.13784282, 0.11296924, 0.03963429, 0.24775356, 0.09070692,\n",
       "        0.10782973, 0.1153921 , 0.03990555, 0.03927679, 0.06056724,\n",
       "        0.07398278, 0.18056837, 0.29187822, 0.1742896 , 0.09179197,\n",
       "        0.04934623, 0.04015388, 0.04108975, 0.2270871 , 0.07371918,\n",
       "        0.17357849, 0.10845967, 0.07387299, 0.04343185, 0.06537871,\n",
       "        0.11447133, 0.17812726, 0.05790534, 0.05899914, 0.04327422,\n",
       "        0.07176593, 0.13153218, 0.04815495, 0.08503833, 0.10171438,\n",
       "        0.04001878, 0.21610682, 0.32470867, 0.09815621, 0.10512843,\n",
       "        0.05082381, 0.04631846, 0.12020849, 0.06379408, 0.19564244,\n",
       "        0.335761  , 0.07869112, 0.052127  , 0.1043083 , 0.27591932,\n",
       "        0.22853394, 0.06210747, 0.11576977, 0.1287211 , 0.05721599,\n",
       "        0.24528417, 0.0702375 , 0.19116223, 0.06735525, 0.20047879,\n",
       "        0.04636167, 0.13226086, 0.08994208, 0.20776078, 0.06706928,\n",
       "        0.16743648, 0.10435828, 0.06782638, 0.04682191, 0.07734302,\n",
       "        0.07000678, 0.0915049 , 0.14033686, 0.04334749, 0.05540673,\n",
       "        0.2941926 , 0.06171748, 0.04445173, 0.23253612, 0.17931625,\n",
       "        0.1643596 , 0.08255424, 0.0495893 , 0.04011615, 0.26708564,\n",
       "        0.04156394, 0.0390221 , 0.3707887 , 0.30180174, 0.06401101,\n",
       "        0.04924024, 0.0401462 , 0.14897414, 0.12494475, 0.04296778,\n",
       "        0.26392445, 0.04328699, 0.8206141 , 0.8441282 , 0.05862468,\n",
       "        0.04215673, 0.04168926, 1.2086964 , 0.04959729, 0.10865646,\n",
       "        0.5964617 , 0.38726464, 0.16631882, 0.16349173, 0.10502797,\n",
       "        0.09391258, 0.1184344 , 0.1395588 , 0.7540614 , 0.07678435,\n",
       "        0.10071672, 0.1189227 , 0.08640493, 0.1576528 , 0.15293346,\n",
       "        0.3066183 , 0.04154981, 0.05756473, 0.04178538, 0.03905085,\n",
       "        0.5116389 , 0.25455138, 0.06362435, 0.21009192, 0.93345875,\n",
       "        0.17711252, 0.14009503, 0.09615792, 0.13984403, 0.1699425 ,\n",
       "        0.0447375 , 0.096984  , 0.04700262, 0.1067852 , 0.3116992 ,\n",
       "        0.22487095, 0.22545907, 0.07857653, 0.15740079, 0.2552079 ,\n",
       "        0.14814709, 0.04016707, 0.07122087, 0.04716424, 0.06356914,\n",
       "        0.15915208, 0.18399388, 0.11131997, 0.1271643 , 0.15810817,\n",
       "        0.19079885, 0.40256166, 0.05568329, 0.11379069, 0.12796038,\n",
       "        0.09310699, 0.19842134, 0.06069349, 0.0543928 , 0.44502777,\n",
       "        0.1882369 , 0.05148494, 0.09922373, 0.04538801, 0.09425362,\n",
       "        0.07495689]], dtype=float32)"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predict_test.reshape(1, -1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1f2df9b5bc8>"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJMAAAFlCAYAAABfgKknAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOydd3wc5Z3/P6PeLNuy5CobueOCXLAderEBk4QAP0JzEjhKyoULEEhy3IXkLpdcckdIQghJCOEghEBiegfTTTPNNu69Wy6SJVl1tast8/vjO8/uaDW7O7s726zP+/Xya7Qr7czj3dmZ5/k8n+/n0XRdByGEEEIIIYQQQgghdsjLdAMIIYQQQgghhBBCSO5AMYkQQgghhBBCCCGE2IZiEiGEEEIIIYQQQgixDcUkQgghhBBCCCGEEGIbikmEEEIIIYQQQgghxDYUkwghhBBCCCGEEEKIbQoycdDq6mq9rq4uE4cmhBBCCCGEEEIIOSZZtWpVs67rNak+TkbEpLq6OqxcuTIThyaEEEIIIYQQQgg5JtE0bW86jsMyN0IIIYQQQgghhBBiG4pJhBBCCCGEEEIIIcQ2FJMIIYQQQgghhBBCiG0ykplECCGEEEIIIYQMdLxeLxoaGuB2uzPdFJJjlJSUoLa2FoWFhRk5PsUkQgghhBBCCCEkAzQ0NGDQoEGoq6uDpmmZbg7JEXRdR0tLCxoaGjB+/PiMtIFlboQQQgghhBBCSAZwu90YNmwYhSQSF5qmYdiwYRl1tFFMIoQQQgghhBBCMgSFJJIImT5vKCYRQgghhBBCCCEDlIqKin7Pbd26FWeddRZmz56NadOm4Zvf/CZeffVVzJ49G7Nnz0ZFRQWmTp2K2bNn4+qrr8by5cuhaRoeeOCB4D4+++wzaJqGX/3qV/32/5Of/ARjxozB7NmzMXPmTDz//PNJ/R/OOussrFy5EgDwhS98AW1tbRH/9tlnn8WmTZviPobV+3TWWWfh1Vdf7fPcb3/7W9xwww1x7yvXoJhECCGEEEIIIYSQIDfddBNuueUWrFmzBps3b8aNN96IxYsXY82aNVizZg3mzZuHRx99FGvWrMHDDz8MADjhhBPw2GOPBfexdOlSzJo1K+Ix1P6feOIJXHfddQgEAn1+7/P5Emr7yy+/jCFDhkT8faJikhVLlizB0qVL+zy3dOlSLFmyxJH9ZzMUkwghhBBCCCGEEBLk0KFDqK2tDT4+4YQTYr5m3LhxcLvdaGxshK7rWLZsGT7/+c/HfN20adNQUFCA5uZmXHPNNbj11ltx9tln47bbbkN3dzeuu+46zJ8/H3PmzMFzzz0HAOjp6cGVV16J+vp6XHHFFejp6Qnur66uDs3NzQCAhx9+GPX19Zg1axauuuoqrFixAs8//zx+8IMfYPbs2di5cyd27tyJ888/HyeeeCJOP/10bNmyBQCwe/dunHzyyZg/fz5+/OMfW7b90ksvxYsvvgiPxwMA2LNnDw4ePIjTTjsNXV1dWLRoEebOnYsTTjgh2HYzy5cvxwUXXBB8/J3vfAcPPfQQAGDVqlU488wzceKJJ2Lx4sU4dOhQzPcynXA1N0IIIYQQQgghJMP81wsbselgh6P7nD66Ev/5pRlxv+6WW27BwoULccopp+C8887DtddeG9Xto7j00kvxxBNPYM6cOZg7dy6Ki4tjvubjjz9GXl4eampqAADbtm3DG2+8gfz8fPzwhz/EwoUL8eCDD6KtrQ0LFizAOeecg/vuuw9lZWVYt24d1q1bh7lz5/bb78aNG/Hzn/8cH3zwAaqrq9Ha2oqqqipceOGFuOCCC3DppZcCABYtWoQ//elPmDx5Mj7++GPccMMNeOutt3DzzTfj29/+Nq6++mr84Q9/sGz7sGHDsGDBAixbtgwXXXQRli5diiuuuAKapqGkpATPPPMMKisr0dzcjJNOOgkXXnihrawjr9eLG2+8Ec899xxqamrw2GOP4fbbb8eDDz4Y87XpgmISIYQQQgghJHvw+4Cje4DqSZluCSEDlmuvvRaLFy/GsmXL8Nxzz+G+++7D2rVrY4pDl19+Oa644gps2bIFS5YswYoVKyL+7V133YVHHnkEgwYNwmOPPRYUWS677DLk5+cDAF577TU8//zzwdwlt9uNffv24d1338VNN90EAKivr0d9fX2//b/11lu49NJLUV1dDQCoqqrq9zddXV1YsWIFLrvssuBzymX0wQcf4KmnngIAXHXVVbjtttss/x+q1E2JSUrw0XUdP/zhD/Huu+8iLy8PBw4cQGNjI0aOHBnlHRS2bt2KDRs24NxzzwUA+P1+jBo1Kubr0gnFJEIIIYQQQkj2sP5x4IWbgR/sAEoGZ7o1hKSNRBxEqWT06NG47rrrcN1112HmzJnYsGEDTjzxxKivGTlyJAoLC/H666/j7rvvjiom3XLLLfj+97/f7/ny8vLgz7qu46mnnsLUqVP7/V0sh4+u6zH/JhAIYMiQIVizZo3l7+24iC6++GLceuutWL16NXp6eoIuqUcffRRHjhzBqlWrUFhYiLq6Orjd7j6vLSgo6JMVpX6v6zpmzJiBDz/8MObxMwUzkwghhBBCCCHZw9G9gL8X6HVluiWEDFiWLVsGr9cLADh8+DBaWlowZswYW6/96U9/ijvuuCPoLkqGxYsX45577oGu6wBkhTgAOOOMM/Doo48CADZs2IB169b1e+2iRYvw+OOPo6WlBQDQ2toKABg0aBA6OzsBAJWVlRg/fjyeeOIJACLirF27FgBw6qmnBsO11bGsqKiowFlnnYXrrruuT/B2e3s7hg8fjsLCQrz99tvYu3dvv9ced9xx2LRpEzweD9rb2/Hmm28CAKZOnYojR44ExSSv14uNGzfaes/SBcUkQgghhBBCSPbgkoEfAt7MtoOQAYLL5UJtbW3w329+8xu89tprmDlzJmbNmoXFixfjzjvvtFWeBQCnnHIKLr74Ykfa9uMf/xherxf19fWYOXNmMAj729/+Nrq6ulBfX49f/vKXWLBgQb/XzpgxA7fffjvOPPNMzJo1C7feeisA4Morr8Sdd96JOXPmYOfOnXj00UfxwAMPYNasWZgxY0YwKPvuu+/GH/7wB8yfPx/t7e1R27lkyRKsXbsWV155ZfC5r371q1i5cmVw5bvjjz++3+vGjh2Lyy+/HPX19fjqV7+KOXPmAACKiorw5JNP4rbbbsOsWbMwe/bsqC6vTKAphS+dzJs3T1+5cmXaj0sIIYQQQgjJcp64Ftj4NHDTZ0DVhEy3hpCUsnnzZkybNi3TzSA5itX5o2naKl3X56X62HQmEUIIIYQQQrIH5Uzy+zLbDkIIIRGhmEQIIYQQQgjJHlySa4IAxSRCCMlWKCYRQgghhBBCsodgZhLFJEIIyVYoJhFCCCGEEEKyA11nADchhOQAFJMIIYQQQggh2UFvN+D3yM8Bf2bbQgghJCIUkwghhBBCCCHZgXIlAYCfziRCCMlWKCYRQgghhBBCsgOzmMTMJELSQkNDAy666CJMnjwZEydOxM0334ze3l7Lvz148CAuvfTSmPv8whe+gLa2toTa85Of/AS/+tWv+jy3fPlynHzyyX2e8/l8GDFiBA4dOhTXvogzUEwihBBCCCGEZAdqJTeAYhIhaUDXdVxyySW4+OKLsX37dmzbtg1dXV24/fbb+/2tz+fD6NGj8eSTT8bc78svv4whQ4Y41s4zzjgDDQ0N2LNnT/C5N954AzNnzsSoUaMcOw6xD8UkQgghhBBCSHbgag79TDGJkJTz1ltvoaSkBNdeey0AID8/H3fddRcefPBBuFwuPPTQQ7jsssvwpS99Ceeddx727NmDmTNnAgBcLhcuv/xy1NfX44orrsDnPvc5rFy5EgBQV1eH5uZm7NmzB9OmTcM3vvENzJgxA+eddx56enoAAPfffz/mz5+PWbNm4ctf/jJcLlfEdubl5eGyyy7DY489Fnxu6dKlWLJkie19nXXWWcH2NTc3o66uDgDg9/vxgx/8APPnz0d9fT3uu+++JN/VgUFBphtACCGEEEIIIQBY5kYGNq/8G3B4vbP7HHkC8Pn/jfjrjRs34sQTT+zzXGVlJcaNG4cdO3YAAD788EOsW7cOVVVVfZxBf/zjHzF06FCsW7cOGzZswOzZsy2PsX37dvzjH//A/fffj8svvxxPPfUUvva1r+GSSy7BN77xDQDAj370IzzwwAO48cYbI7Z1yZIl+OY3v4nbbrsNHo8HL7/8Mu666y4AiHtfZh544AEMHjwYn376KTweD0499VScd955GD9+vK3XD1QoJhFCCCGEEEKyAwZwE5JWdF2HpmlRnz/33HNRVVXV72/ef/993HzzzQCAmTNnor6+3vIY48ePDwpNJ554YlCQ2rBhA370ox+hra0NXV1dWLx4cdS2zp8/H11dXdi6dSs2b96Mk046CUOHDk1oX2Zee+01rFu3Lli+197eju3bt1NMigHFJEIIIYQQQkh2QGcSGchEcRClihkzZuCpp57q81xHRwf279+PiRMnYtWqVSgvL7d8ra7rto5RXFwc/Dk/Pz9Y5nbNNdfg2WefxaxZs/DQQw9h+fLlMfd15ZVXYunSpdi8eXOwxM3uvgoKChAIBAAAbre7z//jnnvuiUuAIsxMIoQQQgghhGQLrhYg3xh4BvyZbQshA4BFixbB5XLh4YcfBiD5Qd/73vdwzTXXoKysLOprTzvtNDz++OMAgE2bNmH9+vhK9Do7OzFq1Ch4vV48+uijtl6zZMkSPPLII3jrrbdw4YUXxrWvuro6rFq1CgD6hIgvXrwY9957L7xecUNu27YN3d3dcf1fBiIUkwghhBBCCCHZgasVqBghPwdY5kZIqtE0Dc888wyeeOIJTJ48GVOmTEFJSQl+8YtfxHztDTfcgCNHjqC+vh533HEH6uvrMXjwYNvH/tnPfobPfe5zOPfcc3H88cfbes306dNRVlaGhQsX9nFM2dnX97//fdx777045ZRT0NwcCvv/+te/junTp2Pu3LmYOXMmvvWtb8HnozMyFppda5qTzJs3T1cp6oQQQgghhBACAPjD54DiQUDDp8CX7gZOvCbTLSIkpWzevBnTpk3LdDMSwu/3w+v1oqSkBDt37sSiRYuwbds2FBUVZbppAwar80fTtFW6rs9L9bGZmUQIIYQQQgjJDlwtwLBJ8jMDuAnJalwuF84++2x4vV7ouo57772XQtIAgmISIYQQQgghJPMEAmFlbsxMIiSbGTRoEFhxNHBhZhIhhBBCCCEk83jaAd0PDBopj7maGyGEZC0UkwghhBBCCCGZx9Uq24rhsmUAd25waB3Qtj/TrchpMpFjTHKfTJ83FJMIIYQQQgghmafbWF2pgs6knOKp64G3Y6/8RawpKSlBS0tLxoUBklvouo6WlhaUlJRkrA2OZSZpmpYPYCWAA7quX+DUfgkhhBBCCCEDAFeLbCtqZOunmJQTuNvlH0mI2tpaNDQ04MiRI5luCskxSkpKUFtbm7HjOxnAfTOAzQAqHdwnIYQQQgghZCCgxKTyGkDLozMpV/D2AL6eTLciZyksLMT48eMz3QxC4saRMjdN02oBfBHA/zmxP0IIIYQQQsgAQ4lJZcOAvEKKSbmC1wV43ZluBSEkzTiVmfRbAP8KIODQ/gghhBBCCCEDCVcLUFACFJYBeQUUk3IBv1c+JzqTCBlwJC0maZp2AYAmXddXxfi7b2qatlLTtJWsByWEEEIIIYT0wdUqriRNA/IpJuUEXkNEojOJkAGHE86kUwFcqGnaHgBLASzUNO2R8D/Sdf3Puq7P03V9Xk1NjQOHJYQQQgghhBwzuFqAsir5Oa9AXC8ku/EZIhKdSYQMOJIWk3Rd/3dd12t1Xa8DcCWAt3Rd/1rSLSOEEEIIIYRkPxueBjY9n/x+XC3iTAJY5pYreF3Gls4kQgYaTq7mRgghhBBCCBlofPwnCcyefmFy+3G1AEPGys95hUDAn3zbSGpRZW50JhEy4HBUTNJ1fTmA5U7ukxBCCCGEEJLFBPyA7sA6PK4WoKxafs7LBwIsc8t6mJlEyICFziRCCCGEEEJI4uh+wJ+ki8jvA9xtoTK3/EKWueUCSkzye4BAAMhzarFwQki2w287IYQQQgghJHECfsDfm9w+elplywDu3MJrKm/z0Z1EyECCYhIhhBBCCCEkcXQd8HmS24erRbZ9AriZmZT1+CgmETJQoZhECCGEEEIISRzdAWeSpZjEMresx+xM8jKEm5CBBMUkQgghhBBCSOIE/ClyJrHMLevxukI/05lEyICCYhIhhBBCCCEkcVLhTMqlAO71TwI73sh0KzKDeRU3OpMIGVBwNTdCCCGEEEJI4jjqTDIHcOeImPTur4ChxwGTzsl0S9IPnUmEDFjoTCKEEEIIIYQkjh6QpeF1PfF9uFqBokFAQbE8zsvPHWeS1zVwV55jZhIhAxaKSYQQQgghhJDE0QOyTabUzdUSciUBQF4Olbl5e5Iv88tVuJobIQMWikmEEEIIIYSQxAn4ZZtMqZurJZSXBORWALe3J3eEL6fxUkwiZKBCMYkQQgghhBCSOLohJiXtTDKJSfkFIZEqm9H1AV7m5rb+mRByzEMxiRBCCCGEEJI4TjmTyqtDj/MKckOg8XtFTMsVF5XTeF1AQan87GNmEiEDCYpJhBBCCCGEkMQJZiYlISZ1W5W55UDpmFrNLFdWnnMab08o64rOJEIGFBSTCCGEEEIIIYmjytx8CZa5eXsAb3duBnCrzKCBHMBdOjT0MyFkwEAxiRBCCCGEEJI4gSSdSa5W2fZxJuXniJhkOJMGbJmbSUyiM4mQAQXFJEIIIYQQQkjiJOtMcrXItk8Ad645k3KgranA6wKKKsRJRmcSIQMKikmEEEIIIYSQxFEB3Ak7kyzEpFwJ4FZi0oB1JrmBwlL5R2cSIQMKikmEEEIIIYSQxFEB3Imu5hZJTFIiVTYTDOAeqGJSjwhJBSV0JhEywKCYRAghhBBCCEkcVeaWaAi1ZWZSQW64fYJlbjnQ1lTgdRnOpBI6kwgZYFBMIoQQQgghhCSGrjvkTNKAkiGh5/IKciQzaYAHcPuMMreCUjqTCBlgUEwihBBCCCGEJIYSkoAknEktQOkQIL8g9JwK4NZ169esfhh4+38SO56TDGRnkq6LmFZAZxIhAxGKSYQQQgghhJDEMOcaJeNMMpe4AeJMCt+/mc0vAO/eCXQeTuyYTqGcSbo/svB1rOIzxCM6kwgZkFBMIoQQQgghhCRGH2dSMmJSdd/ngmJShFI3v1cEnDV/T+yYTuE1CSgDzZ2k/u+FZXQmETIAoZhECCGEEEIISQzd7ExKoswtojMpgkCjhJvVD2fWEdRHTErw/5+rBMWkEjqTCBmAUEwihBBCCCGEJIa5DC0pZ1JV3+diOZMCXkDLA47uBva8n9hxnUCVuak2DSToTCJkQEMxiRBCCCGEEJIYyTqTdN3amZRfKNtImUn+XuC4U4HiSuCzv8V/XKfo40zKgdXnnEQ5kQqUM4liEiEDCYpJhBBCCCGEkMQIJJmZ1NslwlC/Mrd8Y5+Rytx8IiSdcBmw6Tmg52j8x3YCOpNMziSWuREykKCYRAghhBBCCEkMcwB3Iqu5uVpk209MUs6kKGVu+QXA3KvFEbP+yfiP7QQM4Dat5kZnEiEDCYpJhBBCCCGEkMQwl7klEkAdUUyKFcDdC+QXAaNnAyPrgdV/jf/YTpDLYpKuA52HE3+9OYCbziRCBhwUkwghhBBCCCGJYc40SsiZ1CrbiGJSpMwkX8i9NPdq4PB64OCa+I+fLLlc5rb/Y+DXxyf+vqn/e2GZOJN0f+4JaoSQhKGYRAghhBBCCEkMx5xJYau55cdYzc3fG/qbEy6VEOjVD8d//GTJZWdS52EAOrD1lcRer8raCkvFmQTQnUTIAIJiEiGEEEIIISQxknYmxShziyTQBLxS5gYApUOB6RcB658Ael3Wf58qvC5Aywu1KZdQn9eONxJ7vXImFZSKmAcwN4mQAQTFJEIIIYQQQkhimAO4E3Um5RUAJYP7Ph8rgNtc5gYAc64CPB3A5ufjb0MyeHuA4kGhNuUSSvg5sCpUbhgP5gDuwtK+zxFCjnkoJhFCCCGEEEISw4nV3MqGAZrW9/k8O2VuJjGp7jSgakL6S928PUDx4FCbcong56UDu96O//VeU5kbnUmEDDgoJhFCCCGEEEISw1zm5k9ATOpu7l/iBgB5+cb+I4hJAW9fMUnTxJ209wOgeUf87UgUryvkqsq5MjclBpUDO96M//Vel4h++YV0JhEyAKGYRAghhBBCCEkMcwC3L5Eyt1ZrMSk/Spmbrsvz5jI3AJj9FUDLBz77W/ztSBRvD1BSKT/nXJmbIf5NWiS5Sboe3+u9PbKSG0BnEiEDEIpJhBBCCCGEkMRI1pnkaum/khsQPYBbPZcfJiYNGglMWQys+Xt6VlYLBABfD1BsiEm56EzS8oEp5wNdjUDjhjhf3xMSkehMIsmwdinw+/m5tyLiAIdiEiGEEEIIISQxVGaSlpegM6klQpmbcib5+/8uEEFMAoC5VwPdTcC2V+NvS7woF07QmZRjA2GfW8SgiQvlcbyrunl7QiISnUkkGZo2A83bgINrMt0SEgcUkwghhBBCCCGJocrcCsvjdyYFAkBPhDK3YGaSlTPJEK3yi/r/btK5QMXI9ARxKxdOca6KSR6goBioHAWMmBl/bpLXFSpzozOJJIP6Tu95N7PtIHFBMYkQQgghhBCSGAHDmVRYGr8zyd0mziZLMSnKam4qm0j9jZn8AslO2vE60HEwvvbEi9cl21wO4FaOokmLgH0fAZ5O+6/3uoFC4/V0JpFkUPldu9/LbDtIXFBMIrmJ1w08ewNwdG+mW0IIIYQQMnAJOpNK4ncmuVplG28Ad9CZZFHmBgBzviYi1ZpH42tPvCgXTs6WuXmAAsPdNekcEcPiGcybA7jpTMptAn6gszFzx1di0v6PEyuXJRkhaTFJ07Sxmqa9rWnaZk3TNmqadrMTDSMkKofXSwdh19uZbgkhhBBCyMBFZRoVloUGhHZxtcg2agC3hZgUzEyyKHMDgGETgbrTgdV/CzmnUoFyJuVyALdyFI09SUoV48lN8jEz6Zjhg98Cv5+XuRUJlRDtdQEHV2emDSRunHAm+QB8T9f1aQBOAvAvmqZNd2C/hETm6B7Z9hzNaDMIIYQQQgY0uqnMzR+noyAoJiVa5hbBmQRIEHfbXmBPCstm+jmTMjQQTxSVmQSIQ2n8GSIm6bq913u5mtsxw8ZnAU8H0NuVmeP7PEDFCAAaS91yiKTFJF3XD+m6vtr4uRPAZgBjkt0vIVFp2yPbnraMNmPAYbdzQQghhJCBge6EMymamBQtgDuKmDTtS0B+sWQnpYqgM2lw33blCmZnEiC5SW17gdZd9l5vDuDOLwKg0ZmUi7QfAA6vk597uzPTBn+viEkjZjKEO4dwNDNJ07Q6AHMAfOzkfgnpB51J6cfdDvzvOGB7nMvGEkIIISQ+eruBTc/nxiROsMytVIQl9dgOiTqTgmVuUcSkwlJg0Aigq8l+e+Il6EzK1QBukzMJkNwkwH6pmzmAW9PkPaczKffY/mro50yJST63nIvjTwf2fxK/ME0ygmNikqZpFQCeAvBdXdc7LH7/TU3TVmqatvLIkSNOHZYMVFTwNsWk9NHVJPbXfSsy3RJCCCHk2GbDU8DjVwH7Psx0S2KjytyCmTlxDAJdzUBBKVBU3v93wQBuC3FKBV1HK3MDgPLhaRKTcrTMze/p60yqGg9UTYhDTDIFcAOyLzqTco+ty0I/ezMlJvXK+VN3upxDDZ9mph0kLhwRkzRNK4QISY/quv601d/ouv5nXdfn6bo+r6amxonDkoEMxaT0ozpMLTsy2w5CCCHkWKdtn2w3PpvZdtjBHMANxCcmuFqtXUkAkJcvW6sV0vw2nEkAUJFqMckocyuqAKDlvjMJEHfSnvfFdRQLryuUlQQYziSKSTlFrwvY/Q4w3Ig8zliZm0dKJY87BYAm52A62fA08Oy/pPeYxwBOrOamAXgAwGZd13+TfJMIiYHfC3Q0yM/MTEofSkxqpphECCGEpJT2A7Ld/HxqVyNzAnMANxBfbpCrxXolNyDkOkq0zA0QMak7Dc6kwlJpi5Xwlc2EZyYBIiZ5XbFdcX6ffA4FJjGpoERWeCO5w+535DyYeYk8zliZW68Im6VDgFH16Q/hPviZOEJJXDjhTDoVwFUAFmqatsb49wUH9kuINe37peOSV0hnUjpRnYPWndnfsSWEDGz8vtwrNyHETEcDAA3oPATsz/IoUj3cmRRPmVtLFGeSnQDuouj7Lx8ux4gnxykelDOpsEzaknNikoUzqe40+b/EKnXzmYQ0BZ1Juce2ZUDRIGDyefI4k5lJ6vtcdzrQ8El687esvgskJk6s5va+ruuaruv1uq7PNv697ETjSBag68DbvwD2ZlHNvipxGzGdYlI6URd0nzvkDCOEkGzkpVuApUsy3QpCEqf9gDhE8ouBTVle6mYO4AYScCbFEpOsMpN8ff8mEhXDZQKyu9l+m+LB2wNAk0FoXkEOlrlZOJOKyqXUaMeb0V/rtRCT6EzKLXQd2PYqMGkhUDJEnstkmZs6F8efIdeR/Z+k+fgUk+LF0dXcyDHIx/cB79wBrH880y0JoVZyGzVbQuKY9p8ezLMDzdsz1w5CSHay+13gvSypdt//Ca9TJHfRdaDjAFAzFZh8LrDpuex2BOthYpJjzqQ8QMuzLnMLOpNslLkBqSt187rElaRpOVrm5ukvJgHAxEXAkc2hcksrrMQkOpNyi0Nrxf045Xwj9wsZLnMznEnjTga0fGBPGkvdfL0i3pO4oJhEInN4PfD6j+XnXldm22Kmba/M/oyYKY+Zm5QezIGaDOEmhITz8X3AWz8DPJ2ZbUcgALTuBtztmW0HIYnSc1TuuZVjgOkXZ3+pWyA8M8mmmOT3yvc0kpgESH/PSqAJZibFKHOrGCHbrkZ7bYoXb0/o/51XmKPOJIsB9KRzZLszijuJzqTcZ9syAJqUuKkVFTO1mpvfExJzSiqB0bPTm5sU6btAokIxiVjT2w08eZlhuIMAACAASURBVB1QWgVU1mbuwmLF0T3A4LFAudH5cFNMSgsqFwAaZ/wJIf05vF7KSQ6symw7Og5Ip9TdLg4PQnKNdqOUfPAYYOr52V/qFh7A7bNZ5qaiCiIFcAOGQGPlTLJZ5lZurCDddcRem+LF2xPKisovSH1WW1cT0LLTmX35ffLeWjmThk8DBo2OnptkzotSFJZkpzPJ1Qr8ehqwL4tF2UywbRlQOx8orxYhRcvLoDMprMys7jTpT6SrPf5eikkJQDGJWLPs30UwuOTPwKCRmbuwWHF0LzC0DigdKo+Zm5QeVOdg2EQ6kwghfelpE9coAOz/NLNtaTUGWro/u+5dhNilwygtqqwFigeFSt2yVRwND+C260xytcg2ljMpaplbLGeSUeaWMmeSKySi5RfFlxeVCG/+FPjHlc7sS31OVgNoTQMmLQJ2Lo8skCnHulmMKijNTmdSyw6g8yCw5YVMtyR76DgkK5hNPV8ea5qUumWNmHSGOP3S5cr0eWJfT0g/KCaR/mx8Flj9V+C0W4AJZ4rtMZs65Ef3AEOPo5iUbpSdeeQJFJNI+tF1yeTJ1sHUQKdxo/GDJiuwZBLzrD1L3QYO214D1mVRvmMymJ1JgIQhdx7K3v5OeAC3XWeSLTEp31pMCpa5xchMKqoQkas7lc6kNJa5udukjNeJDC2VbWXlTAJETPK0AwdWWv8+l5xJ6vPf91Fm25FNbH9NtlPODz2XqTFfwC+itDmzaNxJIianq9TNKoyexIRiEulL2z7ghZuAMfOAs38ozxWVZ09mkrsD6GkVZ5JadSBbO1fHGr4euajXTAPa92fPOUEGBns/AP76pej5DSRzHF4n28nnAQ2fZlb0a90V+pli0sDh43uB93+b6VY4Q8cBud+qEi3V3/F0ZK5N0VDOpII4M5PUCmvRxKT8SGVuhmiTF0NM0jR5H7tSHMANpKfMze8VwcqJQPGgsyhCac+Es6TsKdKqbsHMpBxwJqlz7eAa9l8V25YBg8cBw6eHnissy4yYFBQ2Tc6g4gpg9Nz0hXD7e/sen9iCYhIJ4fcBT31DZju+/H+h2Z6icqC3K7NtU6gyCpa5pR9vj3QSqifJY/OAjZBUc8gQKxo3ZbYdxJrD64Hy4cC0C+SanEn3opNiEu8vuYO7o+9CEblM+wHJq8nLl8clg2WbreJoP2dSnGVu5dWR/yYvgkDjt+lMAiSE+1gJ4FbvbbRV1mzvy/i+RFrBqnSo5OlEyk1SDqRccCa5DDEp4AUOrs5sW7IBbw+wazkwZbEIropMOZMinYt1pwEHVqdnYQ+fh6u5JQDFJBLi3TuB/R8BF9wFVI0PPV9YZgpfzjBH98h2yHFAcaXMmLCznx5Uh2mYISa1MISbpJEjm2XbvDWz7SDWHF4nJbC1C+Tx/gyWurXslIE4kNzg+9A64JcTpLySZD+ezmNHTOo4ECpxA2RlIyB7xaRgALfKTLJb5tYq29JoAdwRMpPslrkBkpuUsjI3c2ZSofXKc06i9t++P/l9+aJkJikmnSO5OsrZYyZY5mZeza1UnGlOlOE5SXdLyMW278PMtiUb2P2efH5Tz+/7fFFFZsZ86poRfi6OP12cj+koTwzPbCK2oJhEhL0rgHd/CcxaAtRf1vd3mQxjC+eoyZmUlyfWb4pJ6cHnlhknJSY1MzeJpJEmQ0w6si2z7SD98fUCTVtETKqeIi6KTOUmBfzA0d3AmLnyOJnBd/M2GSR/9Cdn2kZSiydLnUnPfBv47NH4XtPeAFSaxSTlTMrWMrfw1dzicCYVV0YvLckrsHb7BJ1JNspSKoanOIBblbkVpUFMMt7bDgedSdFyYiYtAqADO9+O8nqTmKRK3rLtu+hqBipHSVQDc5OkxK2wHDjutL7PF5VlpholkrA59iQRAdNR6uanmJQIFJOONToOAq/eHp/442qV8rahdcAX7uz/e2V5zIbg26N7pOOhStxKh1JMShdel3QYisqlk8sQbpIudF3ECkCcSdlwLSIhmrfKYG/kCSLyj5mXuRXdOg7IDOfoOfLY3Zb4vlTGyrZXgDYHXAAktbg77IsY6WTLS/G52wIB6csNthKTstSZFF7mZtuZ1AKURXElAVFWc/MC0EKlgNEoHy593VQIPX3K3CIIX06i3lsV0p4MdpxJo2aLc8wqrzCSMwnIPjGpu1mys8adJM5Zdc4mg7s9eybb40HXgW2vAhPP7pt3BWQuJze4OmPYuVhUBtTOS08IN8vcEoJi0rGEux145MvAh78H9nxg7zW6LoHbXY3Alx+QJWjDKSoDoIeC9jJJ215ZyU3V91JMSh9ed6jDMGwSy9yOZfze7BJs2huA3k5xvbjbU1euQBLj8HrZjqyX7dgFQNOmzLgo1EpuQTEpicF3VyOgGQPVlQ8m1y6SWvw+wNudHf2UcLzd8c30u5pFkKisDT1XnO1lbkpMMhw68TiTooVvA0bpmJWY1GuvxA0AKmoA6NalWsni7TE5k9JZ5uaEmGTDmZSXL0HcVoKo+r6ZX6/EiWz7LrqagbJqYNzJ4mJsciB/8dHLgRduTn4/6aZxA9DR0HcVN0WmqlGC56KF07DudODQmtRf/1jmlhAUk44VfL3AY1cBR4w8EbsD/VV/ATa/ACz6j1BZQDhFFbLNBvX96B7JS1KUDgV6kph5JvYx5wIMmyRlbtkkOBBn8PuA30zLrsHzEcOVNP1i4zFzk7KKw+tlMDVsojyunQ9ABw6sSn9bWg0xqeZ4aVNSYlITUDkamPJ5YPXDzrteXv/PzLxHxyK9Rjir7k/9alrx4PeKqyae8FglEpidSUpMytbV3Po5kxwUk/LyI2Qm+eyVuAESwA04swJaOOa+UV6EleecxOdkmZtyJsVYDr1qPNB5uH+fz9sjr80zDSez2plULc4kIPlSN79Xrt+5eA3fuky2Uxb3/11hpsrcVGaSxblYd5qU0u5NcdaVv5diUgJQTDoW0HXg+RuB3e8AF/1BcoSabYhJTZuBZf8OTFwInPydyH+nZly8DolJm18AXv5B/K8LBIC2fVKOpyhlZlLa8JmcSdWTAU97amb5SGbxdIjzZ8PTmW5JCDWDOP0i2TKEO7s4vB4YMSNUblI7D4AGNGSg1K11twxmBo2Se2FSZW6NkrUy/3qZ1d74rHPt9PYAH/wW2PiMc/scyJhdcNm0LLmahItncKZEAnNmUn4BUDQoi51JRmaSGgj64gjgjikmRRBo/F4pK7ND+XDZdjnsalViYdCZVGC/xC+ZYwLpK3MDZOJW9/cXRc0lfopsdCbphiutbBgwZJws0JBsCHfzdnEQHt2TnavXRWPbMmDMiXJ/C6eoPEMB3Ma5aCUQj10gz6c6N8nnti9QkyAUk44FGjcC65YCp38PmL1EBvqx8my8buDJ68V1dPGf+s4qhFNULlunnEmbngc++bMIQ/HQ1Shf9D5iEsvcADjTqYiFtyc04zRssmxZ6nbsoQY9+z/KnrDXpi0yszxihlyzGMKdPeh6aCU3RclgcQZlYkW3lp1A1QRjgYbByTuTKkYAE84GqiYCn97vXDvVfUutZkWSw+zYyabcJDUo88QhJqkl3wfX9n2+pDKLxSTDmZSXL5kjjjqTCuDp9eC1jYf7Ph9XmZsSkxwO4Q7PDEpnAHdXY/Lnul1nksopDe9vm/uFimx0JvV2yftWXi0xGeNOEpdLMu76xo2y1QMhR2wu0NUkbqopn7f+fVGFfLfsCsJOEU3YLCyVlWJTKSbpuuFMivFdIP2gmHQs0HlItqr2ddjk2M6k138MNG0E/t+fgEEjov+t02VuKu9k26vxva7NtJKbonSodK6cCNLLVfZ/Ctw1Q5axTiXentCMkypnseOAI7mFGvQEfMCu5RltSpAjm4Hh06QTWD05Nc6kv3wRWPMP5/d7rNO+X67BZjEJAMbOF2dSupeHbt0JDJsgPyctJhnOpLw8YN618v85useRZgbLs10tzuxvoGMWvrPJEaGCbONyJjWIIBMusiR7PqcS9T3X8mUwaGcg6u0Rx7uNAO7Gtm788yOr4PGZ+noBbxxlboaY5HSZmzrX0lnm5u+VVbgACWpPhmBOjQ1nEgD0hInfvhxxJikXfVm1bMedDHQeBJ66XibWn7gWePxqYOlXgX98Bfj7FcAjlwKv/TjyPps2hn7OpdL77a8B0K1L3AAjJxfOVaPYRYlJkQKwx58u45xUGQiUozDaypLEEopJxwLBi6TR8aieBHQdjuwq2P6GOINO+hdg8rmx968uLE6LSVtfie91qhMfnpkEPXs7WOlAiWypLikxl7kNGScX/GgOuKYtwPt3MVcp1zAPena8nrl2KAIB6ajVTJPH1VOdFzF9vcDe94G1f3d2vwOB8PBtRe0CKTFL56qPAb/cJ6ocEJMCfiltU1kr6vzrdMjZEHQmUUxyhKx1Jhn9prgykw5IVpdaaESRzWJS0JmUZ7hzbHwGypUXM4C7AH5fLwI6cKTTtN94ytyKymVitMtpMUk5k8xlbmkI4FbXuGRzk+wEcAPRnUnq/67IRmeSus6WG2LSlMXinm1YCRxcLe7api1A6y6pmug4KM6jFb+LLIw2bhTHKrTcEpO2viIltOETQIpgNUqaS938MUou604HoAN7V6Tm+Op85WpucWPzKkyyGiXOqItksARph3Wo9taXgeLBwDn/aW//Tpe5qfbueU9cEMUV9l531BBNhowLPWe+wcWa3TpWUZ3oxo3R/y5ZvK5QJyEvXzoz0QaK65aKmDTnqtC5SbIfNeipGAnseFPEwPBBTTpp2yvn3nBjMF8zRc4tT6f16pOJoP7P+z6SDlRRWfS/JyGaNstWfT6KsQtk2/CJfGbpoL1BZherDOdkyeDEXWzdzVK+oBwN6v7ilPjjpjPJUcxiTVZlJpmcSXavpR0H+pe4AXI+J+tESRUBP6AZ89N2nUmusInQSOQVIOATgaaxw4Paocb12R+HMwmQZeEdF5OsnEkpFpN8HgnEblyffMRBPJlJgIWY5Oq/tHwuOJOGHgf8y8fRX7PyL8CL35UxizkMX9G40QiG9udOjqPPA+x8G5h1ReRrUaYWXQoGcEc4F2vniei5+z3g+C+m//gkInQmHQu4muUGplb7qDaJSVa424HyYfa/MMpO60QgWyAgF/Wxn5NOfzxlNEf3SKiq+cYVvMEN4BXd1Exl44bUHsfr7mtnHjYxukNE3bxzacYmVbzxX8BDF2S6FfZQzqTpF8mgRokFmSJcrKg2hIlmB3OTlCDr75WsKGIfd5vMTKtJB8WwyTL4TWduksqtGGYSkxJ1cqhsFeVMUgNep8SfbHUmdbcAD54fmrzJFcyfczY6k/SA/T5U+4G+4duK4srsXc1ND0iJGxCHM8k4920EcAeMFfqOdJrcLgGv/cwkQIThlJW5KWdSYWqdSbou/28V95C0mGTXmWSI6eFiUq8rR5xJRn80nolNde23ytnqOSr9o+HTxS2dKzmOe96Ta1KkvCQg9Hmme0W3aAHcgIxZx6YwNymWM4pEhGLSsUB3i8y4KJW5aoLMEEUabLnbpZNtl6AzyYELS89RUfGnfUncUdviKHU7uqdvXhIQebZkIKHKGRs3pS6fRNf718ZXTwaO7o68DLOysCc76G9vSL1tPNVsfl5cL9m0ZHUkVGbSjItlm+lStyOGmFQzVbbVxtbJzpt5gLbrHef2OxBwd4QmMszk5QG189O7oluLISZVhYlJiZTaKgdDuJgUnhmSKGoCxN3u3PVt59vA6r8lt48Dq2SVo0yEpyeDJ8szkwB7IdwBv+RgWjkhsr3MTa3mWFBsT9ALlrnFGODn5Qe/I40dCZa5ASImpazMTQVwp1hMUrkuJYPlmpR0mZsHgBZblCsdItvwvrarub9Ak5XOpLAKDjsExSSLc6bRWGF2xExx3rbsyI3s1m2vitg3/vTIf1PkoIEgnIZVke/HdvK76s6QifPuFEzC2A2jJ/2gmHQs0H1EnEaKgmLJFYrkGklYTHLgwqJmhSpHA5MWAdtesy+AtO3tLyaVRLjBDSRUJ7q3E2iPc4U8u1jNXg2bLEGTbRFmsNVMUDJiUlcT8Lu5wKqHEt9HpulqMjoaXglWzXaUaFw9RWbdtmdYTGraDFTWhq5ZVeNlAOGoM8kokckvzp7Q8VwhWrlh7QL5/NI1AG7dLbOqg0bK45LB4phIZCIk6EwyytyKymXG1GlnUvjPdtB16wHrJ38Glv9Pcu1q3y9bpx0cqcadrZlJpn6TnfOwq1GEGStnUslg+X9mYw5hwB/mTLJT5mbPmeRDPjQjk6mxw+R2ibvMLRViUpgzKdVlbn5TKU7lGGecSQUlscsvC4qlSiG8CqC7pb8YmI3OpO5maVe4gzYa0VYAVLESI2bIBJff49ziDKlC14Gty4CJZ/cPTTeTqjK3Q2uB/1sI7Hnf+vd2ysyUCLb3A2fbBpgCwBnAHS8Uk44FXM39L+bVk6OXuSUkJjlwYQnODtTI6nPdTcDBz2K/zueRrABz+DYQcia5c6jMTded7QyaB2qHU1TqFt5hAkLllJFES9VRTGbQv/VluUmnqtTq6J7QMsypYt+HoZ9bd6X2WE6ghJWiCmDSOeKoiic81mmatgDDjw89zi8U54mTYpIaiE5cKB0eLtduH09HZDFp7HwAurhd0kFPa2jpZyB0n0tEzFIDiHJjQKFpMuh1OjMJiH+fn/4f8Oup/UWTrsbkS76VmOT0oDvVZG1mkqnfZOc6qu5HlplJlSI0pTvLxA56IAFnUgsALeR6iYA7kIcCKDHJ7EzqjbPMbYRcI5x0Dlk5kwK+1Al+qu35RcDgscn3X3we+2U9pUP7Ct8+D+Bpl/68GcOZdOBIFt1HXS3xZ3cGxSQrZ9IGeT8GjQy5pp3sk6SCg5/JhHOkVdwURSkqc1P3OXWPCSdY5hblfBw9V8YhqSh1Y5lbwlBMOhbobu5/MR82WSz/Vq6feMWkvHyZuXDiwqIuyuU1spKclgdsWxb7dW37AegWZW42nUlH9wLrn4y3tanhjZ9IJoVTuDuMHBktdSHcQTHJ7EyaJNuWCGJStwPOpC0vyTZVMz5PfR146Xup2bdi74ehYNJcEJN6u8T5U1As39GAF9j9bmba4vdJsGV4uHP1ZGezuNQgb9qXAOiZ+//mIp5OGeRaMWYeAA3Yn6ZSN29YfkdSYlKTlO+Zw9jLhjknNJrvWfGISYEAsOIeeU3n4b6/62qSPAw74ceRUE4HNfGTKzi5mpuTQkC8ziTlXo3kTAKys9Qt4A+JuPnF9p1JpUNDIlQE3H4NBZqISU19MpN8cYpJRj/ZyXPbKoBbtS0VBN0ThVIK6ZQzyQ6lQ/te/7qtc4jcEGfH2xv2Qc8WF113c+xsrnAKiqX6IZIzacRMOedVjmO680EbVgFLvyqTC3Z4/y65p02/OPrfpWo1N3XuRvr+2XEmFRTJKnSpmGC2G0ZP+kEx6VjASnGvniSzc1ZlNZ4IGRfRKCxzpn42ePMZLqvjjD0J+Ohe4NXbpUQhEkpMGBrmTMovBIoGxRaTPr0feOp6WfIz0+x+x9mwbHe7zI5UTUhdCLeyK5sHamVVEspo5YDz+2TmvaBEhMBEbkqezlDJkV0xadc7wP2L7M/ctu2LPEviFPtWAONOEYt1tHM8W/B0iStJ0+T7WVSRuVK3tr0yKKk5vu/zNVNFmEtm0GxGDUQnnCXXk90ZzE3yeYAnrgUOr89cG+IhWplbSaUIgQ1pyt/xhuW6JetMUjPTivDBVDL0tIXKQdR90Q473wyVFpsHOboeepyMU1cNTnPNmeTuCIUEJ1Ne07wD+OUEYO1SZ9oVb2ZS0JmUY2KSEcAdCOhwBfLR2tGJ/a0x7vuuFlsD/G6fhnwEML66HE39MpPidCYB1uJAogSdSSqA28hwsiOmJYLab36xuNc87X1LPOPF55EBuh1Kh/Tta5srDUzsau6BRy+At6sF722P49qWSqyynexQMaL/+RIIiJgxYoY8Lh0iq9+my5nUuElEpP9bCGx5EVj9sL3XbH4e+Ny3YjoBU1bmpoTXiGKSW0plY4jLKK9JzcIVPhvOKGIJxaRcx9sjs13hN+RhEUqQfL1y8yuJcTEJp6jCuTI3LT9UnnbR74HJ54ig9Ls5wPL/tX5d2x7ZhjuTgP7WWyvUDO6m5+y1c/kdIVeMkwT8chPq7XLuQq3EwZEzUycmqQ5T+AxW9WTpfIejQmprjTKXSCWX0djxhnScxn5OBB874YbrnwAOrLSXexPwyyAulYMmT6eIAsedIlk/TjmTOg8Dj32tvzPBCXq7QuJAQREw/kz5LDIxw6iEvvDy1prjpdzj0FpnjqPEpLIqoO7U2OdPwyrg7tnxiQB2ObAK2Pi0vOe5QKQAboUK4U7V4gBmvD3OOpPU4FPhdJmbWnUunn1++kAom6bzUN/9qYFmMhmCQWdSjolJno7Q55VM8G/bXrl/PfMt4JP7k29X3M6kA5JNY9VHU9+zbFzRTfej2wfU/9dr+GBPJw61tON/XonhHrApJrl8Ggrgw8wxg9FodibFW+amSla7UuhMUnkrqQrhDopJRSH3WjIh3PE6k/qISdbOpB1HurAqMAVX57+G/S/dkXjfodflnNveKtvJDlah7W17xAGqxCRAQriPbEmqiTFp3Q08/U3g3lPEPX327cC860UoijWx9t6vZBx30g2xj5Oq1dyUyB/p++fvtXculg1LTd/Lb8MZRSyhmJTrRLiYB/NswgfxqhMST5kbILZHR8SkJmlrnnHqDZsIXPYQcMsGEUMiOSCO7hG1uGJk/9+Fz5ZYoWYWNj4bu426Dnxwd2pCn1t3mS6oDnXW3e3S8RwxU242dmY/48Vr4UwCpNTNqsxNDY7GnSzbRGZstrwsN40TLpOLvHngFAkVyrft1dh/62oVQcLV7OwqHOZ97f9EZmyPO1mcY06JSfs/Bja/ALzzS2f2Z8bTGZqZAkTsbd+fmTwANbANn6WfslgGVh/+3pnjeDplhrugRNxJrbuiuxj3rZCVDFMRAqlW0XJywJNKPJ3RxaSxC+QaFakc1km8rtQ6k5wUk3qOyjUBsO92atsPbH8VmP0VedxpmjE3308SFZP8PskmBPqff4fWAo9e5nzpg1N4OkNlTMmUuakBxfDpwMvfl9KQZDCLSbYykxpkgRKrQGQlMGWjMyngR49PR+3QUhxfW43BhTp2NsXoM7pa7TmTvEABApgyvAJtLi/cXuMem2iZ25pHZWLRCadwhDK3Lrcbvb4UCOjBAW9RKFcrmdykeDKTyqrCynPV+KOvM2lHYye+4fs+dtUsxFfb78fRv3891IeMh5e+J05zJ/pnTjqTlMA13CQmVU+VFWZTMenWcQh48Vbg9/PkvD31JuDmtcCZ/yqTXwFvdCGreTuw4Wlg/vXyGcaioFgmLJxezS2mM8mmS07dh51+r+2sJkcsoZiU60S4mKNihJRrhDuTVCckbjGpzCExySLfCZDO0/AZkS8yR/cCQ8aFRCgztpxJxs3gwEojfykKPUdl1iEVswxm55BTdfvuDiknGTEDgJ6adget3GGzBsMmyY023GatRM6xCyQvKF4hwu8VQWjK50Oz90cjrBqn6DgkIkBegYiSsW40qoOgB5yb5Vj/JHDnpNAS5fs+lJty7QLDmbTbGYeGGtCuftj5PKneLqDYJCZNOle2mSh1U53k8PyQksHA/K9LxypSAHw8uI0QaU0TMQmQkslIqGtIKoKlG4x8ISdLMVJFIBA9gBuQcx9Iz1Lz/crckhh8R3ImuducGdz0tIlYVVxpX6Ba/Ve5rp3xfbmudJmciebzJdEQ7q7DIrCXDpX7k/kauuNNYPtr6QtTjxd3R8h5kkwAtxqs/78/ATMukYxDdT1PhF5XSGy160yyKnEDsrrMze/3oTeg4bzpIzC2egjKC/zY3dKNQCDKfdjVYmtw2+UFijQ/RgyW/seRTkMs9PfGV+ZWOUZKtzc/Dzx+NfC72dGv83bwuqQNStQyytyuvf8D/Pq1FGTomFecUmJSMqvEJuJMUteFYJlbf2dSTVUVRly/FH/QL8fQ7U8Cf70gPid16y5g3WPyXU52UYHebvmcEhaTwiZ+GzcC0PouDFIzVVZUtjPpaRdPJ/Daj+U8Xf1X4MRrgJvWAOf+NPS9GTVbttFc2u/9Wj7jk2+0d1xNc64axUyszCS/x16JWdkwuU85fR1kmVvCUEzKdbrV0qphF0lNk9yk8NlglaWQKWdSV5O1mATIhT6imLSnf16Swq4zaeJC+Xnz89H/Vrkh2vY7fzE1W3adcCYF/HIDKxkcstymotQtqNiHLScayQGnBkeVo6VEKd5gwj3vSxbA8V8MlTjFEk2US+TEa4HOg7EzZ8yDL6dKOg6slhKJZ78tn83eD4FR9SLOVE2Qm2XnweSPo5wMWh7wzp3J78+MykxSDBkrZWU7MiEm7ZcBotVM0Uk3yPMf/Db545hzf2qOlw5ktFI3dY04sDr5Y5vR9dwSk7zdAPToYtKwSSLqpCM3qV8AtzGIj7fT6e2R608/Z1KViM/JdmIDAbkXlw6VfdoRk/xeEY+nLJZy74rhzjuT1Hk9eq7Mdpv3o0ppDqxMbN+pxtMugwwtLzlnkioXKSwXBwAANG1KfH/e7lCfx25mUqXFSm5A4udzGuh29yKAPIyvKQcKilCsedHrC+BgewRhT9dtl7l1Gc6k4YPkPhAM4fZ741vGO78QuP5V4IcHga88Ls8lUyIG9C+tNcStg62dWLU3iXLTSARXcysUp76Wl1wIt88Tn5gU8Ib6xd1H5P0Pc6buaOrCpOGDUFlahM6Tvodve7+LwOENwJ/Ptn/PfO83IhgAodiERFGThYmWuXm7+353GzfK5KAKqgZSE8L93m9ksYXpFwPfWQl88ddA5ai+fzN0vBgHIolJR/cC6x4H5l0bcubZoagsBWVuypkUYfLWrjNJG5BooQAAIABJREFUiYJO5yaZXX8kLigm5ToRZgYAyMUtPM/GnWiZW4UxcEiS7iNRxKQaGQxYCThte63zkoDYziSfRzru404BRpwQOzcpGMisO+N6MNO4MZQX5cRgUZUtFleK6FI0CDicAjEp3MqtCK7oFi4mqZv3MJmxifd93PKSdNAmni3L32p59sSkokHA6bfK4+0xSt3Mgy+nBu5te6Vztf9j4N07ZeClSv1USYsTpW6uVhnszL8eWPsP69yqROnt7utMAoBJ5wB7V6SmhDIa7Q3WS2QD0jGae7UE5Sa7oo15RTJNk5yoXcsju8jajRK4g585WyLZts84F7XcCEBWZTuRVnMDxE1aOz89K7qFO5PyC+V7Eu/gW733Vs4kIPlObG+niFIlQ+yXzm15Uc6NedeF2maeBe/jTEpwEKscd2PmytY8uRMUULPQmaTroe9wQUlymUnB5aGLgCrDFZtI5p+i1yXX08Ly2IMzv1c+x0jOpGIHxKTm7cmvdmdBp7sXAV3DhOoKIL8YhbqsZra7OUK/0dMpgzcbYlJHL5APP0ZUiujRqEK4A75Q4HU8FJYCo2bJz8mcK0D/0lpD3CrUfNja2On8ambmAO78AmDQ6CTL3Nz2y3pU31VdX1Slgakk0+cPYHdzNyYNlz7EdafW4U2chD9O/KO4xv/y+dgrKx/dK/0atfBGstdbV4Q4EDtYhbY3b7dYFMR47GQcQPM26T9fcp+IV1bk5cmEZSQx6aN75fM5+TvxHbuo3PmSZlXqGO56VdgVNp26D/c7vpo0tymukiAUk3KdaBfJYZPF/moWZ4Jlbgms5uZUAHf4bK9CiUzh7qSeo9Lu8BBeRbj1Nhx1E6gYDsy4SAb60W6+5oGp0zkxjRuAutPlZyfK3Mxli5om7iSnAgvNBMWksIts1QQRevqJScZMUtkwcS+17LA/6NZ1EZMmLpROWkGRzNS2xShz2/MBMO4kcUONnhM7N6krwqx+MrTtlTKp6RcDy/9Hbk6pEJN6jKyJ026RjuA7EYLrE6G3S0Q5M5POkU7snvecO44dopV8AMAphm17RZLZSeErXE44S66tkRwJbftFCOjtcvYaoVxJtfNzw5mkJieiOZMAKXc9siX1jopwlwAg18Z4VzeLKCYZpQXJdmLVYKw0DjHp0weAwePkuwgAg0aFXcMajYGslvhqbmoiZbQhJpmvi+qe2ZCFYlJvt4hzxYaY5ERmUn6R9JMqRiYn1ntdIiQVV8TOTOo8BEDvX9arKCyR/1+i3yN3B3DvqZJF4zAudy/8QWdSMfJ1eR8jiknqnLclJukogA8jDGdSY4dyJtkrc2vqdMPnD5sYUAPGZFb+AywEbBG3CuBHp9uHwx1J7j8cv6nMDZD7YzIr0sbrTAJCTqHu5n6f395WF7x+PSgmDa8swSVzx+CejaVo+corcm156nrgzZ9Gnqx5/y7pVy76T3mc7AqakSo47KDGK+paGPBLH05NpJr/rmSIs0vWdxyIfC0wM7JenPjhfeyeo+JmnXlp9H6UFU5Vo5hRzqSA1/oe5e+1X+YGOB/C7TNd+0lcUEzKdbqb5WZqFYCqnDxmcSThzCQHVOpYdctBMSnsAqGycqI5kwK+yLN+6iYwaKQM8gEJL45E+365oGn5zuYPudvFeTBqlixhHC5gfPxnucHGUx8edJoZn78SkxwPplNiUthAraBYsqzCnUfdzUDxYHEGVE+VDlAsMUhx8DMpBTv+gtBzQ4+L7kzqOgI0b5UwQgCYvBhoWBn9ZtPVJDNlgIPOpH0iel5wV2ggqsSkyjFyk3LEmdQClA2VDsyCb8pMn1OdGE9nf2fScafIgCidK4zpuuFMGhv5b4aMA064XMLyk+lY9BOTzpTtbos8DU+ndITU+emkU2P/JyFHnrstJQ4CR1GD42gB3EBoVceGFJdIhbsEAENMiteZZJqAMBOcEU1ycKOu8aVDDTEpxv6ObBMhd941oWWTB43om0HS1STCR0llcmVupUND91pz+W9Hg9wXOw+GQrrTSWcj8Nmj1r/zmETNgpLkMpPCBxTDJiXpTOqWkpGiitjOJCXYRRv4lQxOfDW3A6vkXvzZI86thGngcnug5eWjsqQQyC+C5u9FeVE+dh2JJCYZ53wMt0hPrx/dXnG+DC3NR2G+hqZgZlLsMrcthztw+h1v49+eDit7V9eJpJ1J1mVuRZoM7LcethG6Hg+qzE2V4gyuTa5Uzx9HAHc/Z1L/SoMdTXKOKzEJAL5xxgT0+gP469pu4OrngLn/JDk+S7/S/x7X3iDn55yvyaI8gIPOpNjCZT9UP05dC9v2yXsWLiZpGjB6trNlwNGc2WZGzZJrXng/fOVfpKLklDhdSYDhpExRZhJg3V/zue0HcAMpdCYxMyleKCblOt3GCgVWK3+om7T5S5uUmJTkhUWJJ+WRnEmqvWGOHSUiRMxMUje4CCKM6nBXDBeXzPAZwKYoq7q1N0hOTNUEZ+uf1WB/xEzrFSI+uFtusPfMlSWJ7SwtG/55jpwp2RFOhzKrDpfVDNawyf2zuVwtoRt3sJbcpoNjy0si5E1ZHHoulpik8pKOM8SkKecB0KOLH12NcqMuLHdm5ayeNsNBN04cDFc8ApzzX6E69bx8GaQ5Veambqin3iwDlbd/Ye+1L30feOob1r/TdcOZFCYmFRQD48+wF2zuFD1HRRyI1Zk67bvSCfjo3sSP5Q4LkR5cK51Fq9wkVQo08WwRUZwUkxo+kZnbytHy2KmQ/lRhLrONxpgTAWgh51Uq8HtlUsFRMSnMmVTqsDPJbpnbygdlkDrn6tBzFSNlkKTuE2r1OTsLUkRCDV6Cs/HG+dfbLftUrqh0l7r5vcBjXwOeu8HaRWqeVClM1pmkytyMAYVV9mQ8qByv4orYZcJKFIiUmQTIdy1RZ1LDSgCaOOJevd3Ra7nL3Yv8AmNypqAYmr8X46vLknYmNXW64YcIqFrAh+GDSkzOJG/U1dzcXj9u+sdn8PgCeHJVAzYcML1vysWXjDOp1wUcXNNXEDPaM3u05Olsa3RYTPKFOZMqx4gImehnadOZpOs6eouMfmZ4mZsJKzFpYk0Fzp02An/9cC+6/XnAl+4GFv8C2PZK/xWW1/xdruOnfte5621SmUmqzM247qgwfpUXaqZ2gUzmOhEH4HXL/d+umAT0FYh9HuDj+8RlPfKE+I9fVO5MtIkZs3Br1bfxxelMSlVmEgO444ZiUiroOJi+AZerOfIF0qpszN0u9tHwwWIsiox6/2T+X+qCHi0zCeh/kWmz4UwCInegwwcGMy4G9n0kq39Z0bZfLuA1U50Vk1Qw9ogZIjCY/5++XpnxnXFJaEnie0+RUq1o73n4YC64ctLHzrUbiJyZBBgztzv7ttNlsj+rm67dcqAtL4kTxrzKy9A6+RzN7jjz8faukA776DnyeNQcES2jlbp1G6s1VQx3xpmklpIfMk62YxeI0GGmaoIzyxG7WkIdrbIq4OQbJFjezmzzlheB9Y/LdyAcn0c6cuZgScXkc+S7mMwsfTwoR2Usm3fNVGDal0SADV9V0C7mAG7FhLOkdFI5FYLtMsSkIcfJ+eaU28bbI1b1sfOtcxqcYv2TEnDvBB6bZW4llXJdS+WKbmqyozDs3E1ITGoCoPW/tzrViXWbnUlV0mmP5JDodQFr/w5Mv7BvgOqgsHNErT5nJSYtvwP46E+x26WcgKVVIuir2XjlmJl6vrg50y0mvf3zUIC71aRR8Dwc7EBmkoUzydWSuBut1yXX06JBNpxJxjUvljMpYTHpE8l2Oft2cbpteSmx/VjQ09uLApVfZLx3k4cV2xCToq/m1tjhgU8NVQI+DK8sRlMwMym6mPSLlzdjW2MXfv+VOagqL8J/v7QplGGkadKfSeZceecOceydeVvwqR6/tHX2mAqMqCzG1sMO5wwGA7iNAe/gWhFAE3Xm2shM0nUd331sDS5/2OgPq1iJ7iP9V3Jr6sKowSWoKO6bZfXPZ01Ee48Xj326X977z31brqc73+x7sB1viMNn6HHyvckvTj6A29VsBIXHuE9ZUWZcC9V1VgnLwyzEpLELpNz24GeJt1WhhGU7YlL1FLnumft/65+U1TlVFEC8pKTMzSTcWolJdl1yReXy/3Ul4Ua3wueRzzqRHLYBDsUkp9n6CvCbacAz33L+i2iFciZZYSXOeDpC+TrxUFQOQE9uFkd1TCOtKBDNmVQyJLKbSnXwIy3JqQYG6v2YfhEAPXKpm5qdrZkqLpLwwWSiNG6Uzu7gWhE6zDOsHQfkJjRxIfBPLwBX/kPqn/9+OfC3iyOHaoc7k4ZPl5/3rnCmzQpvj4iQVpby6kky+2oufXC1hAZiZVXy3jfbEOZadgJHNvctcQOAIXWyVYLN4Q3AL8YAy34oMzh7P5AbuepY5uUBk8+VjorfZ32sriYRkpwWkyI56ABDTNqVvNjc09p3RvekG+Q7Esud1N0S+p689d/9f68GO1adLuVK2J6mVd2CnakoZW6K028VR97KB+I/jjm818z4M2WQHz5wDoqGY4HaefK9TrZUApAZ7oBPBOHwnAan0HXglX8VF6QT2AngVoydL8JbpJyMZIkkeCfqTCqv7t+pjGdwo+sSOv7KvwG/mQG8/h+h34VnJgGRBaoNT0n7513f9/lBxqo+akU35UwqGdJfcFn1F2DZbcCGp6O3Wd378vLk/6/OP7X0+LBJ4qxNdbmimR1vSoaKcrha5Q6ZnUkFxQ6s5qaFSqDVoFE5EuLF221yJsUQuzsOSB8h2qC3ZHBiorlaKbJ2nqx4WnM88NqPHCmlbe/xwufzo7Aw5EwCgIlVhWg46oLHZ5GXaNOZ1Njhhg/GfgM+jFDOJF0XYSVCZtKbmxvx8Id7cf1p43FB/Wjccs5kfLSrFW9uNl1TC0oS79M2bgI+/D0w+2uh8noA+9pF7JlYVYQpIwY570wKip3G/1uJDeo7Gi82nEmPfLwPz605iM1tRoltz1EZ3/h6LJ1JZleSYu64oVhQV4UH3t8Nrz8g15iJC4Gdb4XuCT1tcm2ZuEgea5r91S6j0W30R+Md9wDiKC+vCfURm7fLd9Rq3DXmRNk6sXJp0KVoI+sov0Cuy4fXyWO/T1aBGz499F7Gix0xqXEj8PjV4iazg7cnNAlq6UyyJyZ9tr8N3uKhyZeb9zt+HGH0pA8Uk5zm/bvkQrPuceD+hc46W6xwRRGTrELK3O2xSxKsULO9yQhkwZXnIohJhaUye9cvM2lPZFcSAM/wE+DLK4FvW4SSpq7D8l6om2/NVKBmmnWpm88jfz/YWA5d9wOtRidSNwSoRC9gjRvFlaRpRpmbOdxUuR3Gye+P/wJww0fA+XfIIPO+04Hnb+y7FDTQf3W+vDxg7EnAvg/jb9+af0ReDt3nBgpKrW/Gwc62qRSgO2zJ3+op9lZ0UzOlx3+h7/Pq81elbuseEwHroz8A950h7+1xp/Z9zZTFcr5Hcml1NYacSU6UEykHXaSgeEDEJK8rOfHK75P/l3lGt3SILGO9bVn0VbMajdyIKZ+XmeldYZlAaqBm5VwcWiefdbpyk+zM0itGz5GO6Yd/jF/Y8bllhjt8ADf+dABa/+9E+34RVcuHS+dR9wOH1sV3TCtUB7R2fn9rvVN0NUnHPJIrM17sBnADIpJ52p1f1EDhNVyLlgHcCTiTwkvcgNiDG12X6/Xr/wH8th544BwpUevt7PtdC89MAiLvc+UDcr867pS+zwfPkcNyTehutnYmBfxyvdHygGdvkPZZ4W6Xz0cNTstN10WzS7B2niF8OriKYSQ6G2VirmYacL6xyIDH4rM0O3QLSpOb9PJ75Put7nXBFUsTLHXrNXK8imyUubUfCJW4RqIkwTK3lp1yXtTOl8Hnef8NHN0dPT/SJrubu5GHAIoKjD6W4ZqZMLQQAR3Y32qRt+lqEcEuRn+0qdPCmdTpMc4/3dKZ1NThxg+eXIdpoyrxr+dPBQBcuWAcJtSU4xevbBYxA5BrhTeBcyUQAF78rrT93J/2+dWuVhF76oYWYeqIQdje1Al/wMFKhfAAbiU2JLqiaYwB9KaDHfjZi5tw5pQaTB5TDTeKoLuOWi7+Ewjo2HmkCxNrrCsf/vmsCTjQ1oMX1xkTjxMXyTVG9Ut2vyP3UzVxBRhlwAmW7SosHFRxUWGa/G3ZIROoVn3hsirpIzmxcmmw/2PDmQRIqduhtXJuvvkTmZQ987bEBDQAKCpHwNONVXstxjtdTcALNwN/Ok1Wx17/hL19+jyh/pxVtITPEzMD7YMdzbj8vg+xy1WKgNMxAP5eikkJQjHJSfZ/IgPXhbcDVz0jnbs/nw2ss/lFS4TuKGVu+QWiAoeXucWblwSEyl6cEJOi1S2XV1s4k/ZGFZNe2tyOt70z0LPhBWvHR1eThG+bmX6RuHfCxRnlrhk81pT1YwiCB1ZLbsPqhyO3PxKBgMxkjZghjytqZMZSvZ9mt4OioAg46Z+Bmz4TS/Caf0ieknkwoDqV5g7ZcSfLgC0e23MgALxyW2THQviKJWaCne0d+GhXC25/eh10c2YSAAyfJm6iWDOqW16SlSlUqZhCuX3a9hqi3vPApEXA14xZe+hA3Wl9XzPhbJm13G5R6ubzSMe6YoR1flUitO2TAYMqu7RCLe+aTG6SGiiGz+gu+JZ8t97+eeTXKofbF38tndC3/rvvdyboTIpQBjv5XCmRcnrJWCvaG+Tzi5SxFs5pt4r78bNH4jtOpBDp0qEiUoWLSaoMNi8vNBPpRNnP/k+AoePl2qAEd6fFJFVq2+lQgHI08TGcsUYJbjyztq27gZ+PtrdCZSxnUjxuQOXyscIqMLtpM/DWz4F7TgT+fCbw4R+A4ccDF/8J+MF2WU3HnPnWc1QG3IWl0cWkA6ulZGLedf0HBeqe1nnYGNjpcu6Ei0ndzeJ6PeMHcqylX+l/3wNM4c/GPaiiJnT+tR8AoInQMeZEEcdSJQoqAgHgmW+KAHPZX0LimZUzqU8Ad3GSYpK374Bi6HEieiRS3hsIiHujyFjNLVaZW0dDRPF8T3M37np9G3ryKxITk1RemfoeqvulA/mKu5u7kI8AiooMYccI0a0bLE4WyxBulzHhFGOw29Thhq7cRwEfRlSWoL3HC7fH+IzDxKRAQMf3nlgLV68Pv7tyNooLpA2F+Xn44eenYdeRbiz9xOhvFSYQ1q7rwIq7pa9/3s/6hTrvahGxp6pEw5SRg+D2BqzFtGCD/cDapcBjV9nrswUDuFWZm/F9jbZCccRjB4wBtLUzqdvjw3f+vhpDSgvxm8tn4RunT8BRvQIHDx+yjK041OGGq9ePySOs7wdnTRmOKSMqcN87u6TccOJC+cUOo9RtxxtyH66dF3qRE86kaJPudjD3EVt29A/fNjN2gXzXknWft8fhTAJETPJ0yLm54h5g/tcl0iNB9MJy+D1d+Mr9H6PTbZxzXrcYJn43V/pZC74ln2GbzdUEfW6jj1yVUJnbxoPt+NbfVqGypBCNvnK0Nx+O+LcJ4fMwLylBKCY5yYp7pOM6+6sSzvrP7wGj6oGnvw68eKvzK/N43dI5iXaRLK9xSEyS2d6Hlm/EjqYEbbtdR8S1Fb68vJnw9gb8MkiPUjr08vrDeD0wF4PchxCwKgfrPNx/YDDjYkip2/N9nzfPBqisHyUmrTEGqYnMULbvkw64EpPKw8pY2vZDOusWsxBlVcD5vwC+vUI+b3ONuadDXGPmcoxxxgx2PO6ko7tlxjdSnk80MalyNFBYjobt6/BPD36CZz/ZBs3v6St2zP6KiGfRhLiuJumghZe4AXJeFJZJx/fwetlOu1BmsG74ELjy76FV0xQllSKsbXut//7UOVYxXDoKPUeT/36qldyidY6rJsg2GTFJdazCRaviCuC0W4Bdb0fOxGncIMG9g8fI4LLh/7P33uFRlWn4/+dMTTLpM0lISCeBEFrovRdFpK2gqLh2XV3LWtZdd11d17J2RUFBcQUVUBSVIkV6Dx1CCAkJIb33ZJLJtPP9453JTCpB+f6u3/W9fP5Jm5wp55z3fZ77ue/7Od5atubsnHcGDsRNExt+R8c31Yp17st510fKVFsgri1FN7ep6HGC/XLkw+6Z1zvD5MZqaBuxk8R0FvcCtjbflcD79BDJ3m8Fk5wSFGehp9IKudL19kxygjKNlddnP2quF2xS54SxrkIfh90jgJyz+8jsrvSjMkusG93xWmoBk1ozk2QPP5DtbDx+DeBHZ8wkaF/cnP4KPh4FB98R+8bsJfBsJtz5HSTdLvbbgGjhk+QEeUw1rvvX2VzpiPF68n/i/Qy6rf3fdMGAJK4Rd19AzwBxfOc96JS19hgAt68Tr+HbO9szMpzsWOe17c5MqisQ17pSDT0dhd7/bd+kw+8LIHfmm6IZ4ZRSdtSQaGXA7fnr2CbOaNsdV6rF+esOs7bdsdyuyW4zkzouHpfszmTJ7kzWpdRibaptP+r+alFwXKxxhj6O1+Qp8rHrAFhnlxtRSjIap8zNUZBFOMCkDn2TGiuvKnEDwUzy8nAUeDYLwT7i+/Iax2fZRub2v8NXOJhZwQuzEokPac2YnNo3mBExgXy4J4tGs1Ww2K6FydpUA9/dA7v+Ldi9SXe2e8ilSnHtSXYrfRzPn9HReifLkLZJeGP++LDIRVO+vfpraDHgdrxvr0ABBtV2s6B3j7Zm823iXxtTyak0smTRYPTeWm4aEIpR4UNRUaGb0sBVfzjX9bhOmEkKhcRDE3qRXlLPvkvlwvctZICQuskyZO0Rgz7cAULP6yFz67jpXmU0uzy0ugqnksBsFPKzjvySnBE+TIBX1b/RG7M235H3Xt0cHXCZcO/6t2Af3vDf3/T0mTV21FixW81CGpp/HJYNF8ePHieUEzPfEFK62vzugWeWJnGttq3znGHtnBmUX9XIPV+cwNdDxebHx2HRBtBUW96989fdsDZ3b5rc79EufgeTrldUZQtj22H3ubr6vmHC/2bM44Kq/vmM6ztlqwOaabvQBbWXuf0KMKlZIYCEjccv8ca29Gv+f6B7VFPv4Navt75YSFA6YSbVmywcyCwny1cAKLlHvm//IOfIZPcI7iuSqrSNrX/fklCHi26if6Tw+rE0wfkN4m+/xjvBWcSFOEadtpWx1OSJ66WrhSyot0j+3CUqppr25zNssFiwc68BTCo8Lb7W5nfsMWTtAkySJOp0UVxOP0uvIG96+zg0/e6bd8+hAuQ6trxzD6OMbYAMCbM6fA4CogVL7eImIdlwPs4rUHzfEYgTf4Og+zqZX85wL746M36/1qjObc+oaht+kaLL/ZuYSY6is6NEfPj94lrf81rHm3tJqmvc7uDFwnflrNu47a48k0BICdVekNXGNyn9Z1g2Uqxz2XtdBsO/JZxmwB1EdnkDr2+9yNbzxTSZHXIbSRLeSTV5wmemu9GViXTsROFj5O5BVpPfmkHYc8hvL6zri8U16WQ6wfVjzLlHWVrr5/yt0Vx7VYlbToWRzw5ks3DFUfYbozDnJPPImtPdk344GRjdYYW0yNxar1NHCsV68+HPJ6ht6gbIKMsdMpPsdpmMknpkzzbMpEvbxX3/dDrcvQmG3tPeVLhFpuuQwjZVC2kqdM5MaqoRJqoDFnS8ZytVYu2qL3btI94h4riyXTQvwG2t6yGaW/OXC+Byy19arxHuex+4mEmy7AB2HSCHPk6AEv83fZPyksUa1u8PMMQxwc55nXXkO9Rc5xos4mAmfZ2cyyf7LmO2XiPoYjO3l1ro43/dvu9kcGp04vVbmzrf/ywmkdN1IGtpMtvYcaGEqQnBePvpUdmbWbziwLXJpwpOiPXFHZz3uT5rTHaFEZ0KFE5Q2VEQ+qjs6HWaTsCkqm6BSaV1JnROMMnBTAKoqHPsVW7nKrWwlje3pzM9MYQ7R7bfiyVJ4rkb+lBe38yqIzmiUO8umJSXLGQ96Vtg6kuwaE27nMNis5Nd4TIHdzJ0MkrcwCTZMWX2s8mw/i5xry5cJRjZV/M0g/YTpyRJ3JsOj52yehPv7bzEyNd3sXTPVQDQllHo7QGL708V8MPpQh6fEs/oXuI8qZUKdH4G7E3V5OU71jI3ZlJHk9zaxpxBYYT6ebBiv+N+ipsiGp9FpwVo7S5xA3GN/GYD7sp2tcee9FKGvLKTYa/u4tE1p1h9JIeMknrsHd1TTpmbE1A2dMFMahmC8xulbnWdA8sdRnBfB5M7CBau/k2giMVm55dMcS6jfWBLSjHs+68AW/64Ee74xtVs948S11F38merSezPbes8Z9g6ZgZVGc3c/b/jmK12Vt83gjB/T8J7RuBtq+HYlevom2Tr3mTD36N9/A4mXa9I/kS4wI94uPXvlWqhTV+0VrA+VkyA9K3X5zk7kI3Z7TKHsyr4ZN9lHlt7mhPlCizulHZTreh4X8vTNFt5fZdINAeFqNmTXkZRza8wnDWWt0vQS2pNpJe4JYfupp/gAt868aHZk16G2Wrn2QUTOE88csa21g/opDAAHFK3w621u20nSBn6iJH26T+Lwikw9tcllc7JWcF9xVenCXnLtJz87hkN+4a2LgJNde3Mb20KR/c47xpMuJ3TJ+zWjjtczo5CB7H5XBH7K/3ooypl3YOjmN9HbAY1Upsic8xj4tgXN3ZwFMRn7B/pYm+1Df8ocT2kbRKgRndoy71vEF/bTnVrKb6Cr48/jSw7mElXAZOUKvGY68FM6mgKjtoTJjwrzn323tZ/s5qhPN0FaCrVojB0LyauJltSe0D0eCwZO/nsQDYmiw1OfC6kM56BMOYJ8bjrIc+qK+ywsEotrGXh8qN8eiCbR9ecZuirO3l83Rm2p5ZgipkGwf3g4HvdZ0d1ZSIdMUpc9w6pm2wxOTzV3M5zz2GiC9mRdKi74ZzCEprk+p27T8P1itJU1318PXyTOpiCZ7fLnMuv4e0d6cx4fz+T3tnHa1vZ7mQnAAAgAElEQVQv0tBsQxMzit6KQsrKStl0tkCsi1150DmBg+7cLx3I3A5cKmfNOQFIqSz1LnlLV2GqFUllG2bSB7szueGDA2y81ERTXTmVDY6isSxNAPg+nTCZoL3nW5MbM8nTHxmJC1nZnMqtdnVaz30jwIe2xtvu4RMirrsWwCi4/XRT537hlMUlzoVJ/4Bz6wSb2hlOWanzfeuCxefQXCcYM857UaEQ7/f/FjOpsQo2PCAA29kfuAp2J3OwQ5mb4zqUJFB5YjU38e9NF3hzezpzlx3mQtE1yMJsZmwKdeuOt76X8E68Vsalc7S2k5kE3Pj2NlYe7OB6dkpPOyggd10spdFs4/7xMSwcK/bHrLwil/fM1cJsFA2t8OGtf38VwHpvehlT3tnHhlMFXTIAssuNeKkVIgcGF7vA2kyMQUd2p8ykrie5gQCTvL0c97TdRrCvOHZlreOYDlZ2k9nGk9+cIcBLw5u3DETqhCE8LDqQKQnBLN93GauiG5JImxX2vQFfzBQMzPt+EU2LDtiYWWUNNNodZZXNgpdGRWSgl4uZlHsUVs2Cr28RvpJzP4ZHjkK/+dD/D4IF6wScO309bQy4AfzCaSzP5Zn15xj3xl4+2pOJzQ4rDmRjbO4EvAQXy6kNGySrrIF//ZTKqNhAnpjamoUTFBJKoMLIqYsOYMWt/rhc3kCAlxq9d+dSIY1Kwf3jYkjOruJsfo3wTbJbYedL4gFxbQyjvQLFWvZr2c5OBYcbcGm22nlly0Wi9V5M7BPEufxaXtp0gRs+OMDQV3fyp69O8cXhK1wsrhPgkneIaGo7paIdyNyazDZe33qRl4/Zsaq8sf3WicrOYQjdDZUW5i4T1g/d8ZnsItYeyyPfKK7jWQm+HLhUjr38kmCNxU5q/WBnY607UjerycFM6sDOBDo04G40W7lv1QkKa5pYefewFrZhbFQkvlITn++/jr7EVvPvMrdfGb+DSdcjGquEfnTAQlHsdxQJs+Dh/SKp/OZ22P/Wb39eo6OodBTVVUYz960+wZ0rj/Hm9nTO5NVw2ehBY3Up2eWOLo6prtvMpIZmK+tP5HPLJ0c4UyI2sD+PDUUGMd7zml9va2ZSncnCwhVHuPGDg9y/6gSphbUCVW+scG0czo21E2bStvMlBPtoGRWjpzZyGrHmDArz3BK1pmqxCbT1TAIhdZPtraVutfkiiXZSS4P6CG+I06sFCJB0pwCArmWaStFZATYmznOx1trJ3LrBagHBJHGfmtbGUH39yXz6v7SD86pE5OKUq9PqW17jGdci2hF7rhOZ2/enCnjymzOY/GIJsZfip7EzPUokd3vzXZv/+hP5PHYqBFtALBxZ2p4101wvCvaEmzuXiQVEI5dfFEyxxLnde1/6OOFDk9lG6ubOTHIUT/Jv6dA2VQsmQFeT3ACbXaZQ2wvLxW1cOb3r11F0G7tgJoHo5PtFtPdDqsgQ90KPAa7f6YJaAxZX8UxqMtvYbR2AuvYKX2/by/s/7BNJYMxEsb7FTxcP/K0sL7tNXOdtkqITOVXc/mkyWpWCXU9PYO2DI5k/uCdHsir409enGPrqblar/gAVGZjTtnTvubpiJqk9IHIUlsw9PP9DCrNedcgQ3JlJsZPE17Yjjq8lis8Bkos1BuK6NF5HMMlmFZJdp1fKtfommY1in3NP6tuA2VvPFzP6jd3MXXaY5fuz0eu0vHhzIgefm8y2J8czdtJMAO4LTCHk53vgfzfAgXc6f07nOntNzCQhc8sqa+DPa0/j6y/2nLHhKr44nHN1pkq5g3nrtmekFNSwbG8W4+MNWD0C0ZprmfLOHnKKy0WTKLgTANwZHYFJzqaOQolZ7cepi5e55ZMjTH13P8v2ZGI5vlIwScKSOjqiCO8eyA0lmGsdgJHOBSYVlxSTU2F0gZzu4NjE58R+tPNFlwy4razU0YD5bv8Z5LaFTcQIAU50Zub9a0OWHYMmimHB/1rnKgqlAGQ6k7k590GVFlOTEUmC1+cPoKKhmblLD/PezkvdYikVV9aSXW1l3sdH2Hq+WLB/DPGiELrWiVktzCSvljW1tqaaJbszqW1sw5Jr8axqXwhuPFtEiK+WkTF6JMd1M0Avs3RPVsdMirZReFrkOu3ApM4nmVptdl7ZkkZuVSPPfHeOxZ8fE9dTm7DbZXIqjHipcQEszlzCZibGoOuEmVTRbZmbt6eTmWQhxEfkZpVOZpJD5vbKz2lkVxh5/7YkAnVdszKendGHOpOV3Dp718ykmnxYfbNgZQxYCA8fhPChnT48ragOi3PynENq3TvEB0thCqxZCF/cKNaym96Bx0/C4DtdFgWJDn+btmz5tuFkzkkSNrvMzrRS9pVoqC3NYVtqMbePiGDPM5P49I9DqTdZ2XC6i2vWCaS5FdAmi43H1p7GU6NkyaLBKBWt8zGVLpBQdROVpYXYVF4tFhggGFjxwVcfxrBoRCS+HirBToocJdbsnIPCp7RtHuylF9fur2U7t0xFcxnbf3k0hysVRl6a04/3bk3i8N+ncPC5yby9YCBTEkJILarl5c1pzFxykPgXtvH4FnGMQ7sd5yawV6unKK0zcdunR/nsYDZrTxSS3BxFxsk9PL7uDJvOFXWPEeseTjbotYBJIOTQTrnbrwhZlimrN7FkdybhQeLenBbnjcLWhKKuoGN5X4tnVzcaNRZTi8xNNpZxLr+mdQ7cRmJstdl5fO0ZUgpqWLJoMMOjXeCz2kfs6+cysrsvm79aWE2/y9x+ZfwOJl2POPWFSGRH/7nrxwXGiK5G75mi0/Fbp6G0yNyCOJVbxawPD3Ikq5KXZidy9sXpHP77FKYM648fDdyx4hCXih0FbxdjnO12mSNZFTz97VmGv7qL5zakYLbaeW72EACCtDbGxwfx7Yn8a9fsN5S1gCiyLPOvn1IprG7injHRnMip4uaPDrH0eC3Idp74Yg/v/pJBZUEGIHXI2jE2W9mbUcbM/j1QKCQSJgpfiZQ9brpz945t2whOBH0cppQfWbH/MnOXHiI94yJ2t+eyBMaL7uyVAzDoDhe1s7vMEnMj/PCgANFuft/1e50BGYn1+05x12eHkWsLWxeonYVvWGtmUrMLHCytM/HK5jQkCd66GIgk22jOSb76Me02Ucw6gYCOtN5OeqpbfJ2cy7PfnWNsnIG5UyciIUNVNqFqkThuzGhGlmVO51Xzjx/Ps+V8KUuM0wWdua2fU9Zu8Tl3JHFzxMEKHZIsrrmGmBuu/r4AJAlTzDTs2ftZe+gi//oplUWfHmX3SYe3li6ohSX2/k+HePKbM79u8krLJDdXIpRf1ch3J/M5nVeNsdnKsexKZn90iHlX5pFnDcCwcTEPvPYJz/+Qwi8XSrruILpHi2dSJ11dlVYUi4WnMF/cxuGsCt7Yls6e/XvE30PcAYs2k+w68UySZZntqcVMe28/L6eLYue5Xnkkpf4Xm9UsGARKtZtk8DeCIPUlYqqLWzK1L6OMuz4/RpCvlu8fGUNcsA9jehl4bf4Ajv1jKmseGMmcpJ58VNyPXHswGd/9m3/+kEKd6SqJXGcG3EBNo5m95kTUlensP3WBGLX47CuUrvXkiDGMelUg9anb2v1/t6P4nEiknYMOAFkXhLWuhMUrj3Ey5zpQuSuzRCES57jP66/RvHLnS7Dxz60ZKW7MpILqRv763TkM3lreu3UQp16YxrqHRnHfuBgiAh0FR8+hICn4S+NHJFnP06z2d4E3HUULM+lK5/IgZ7gxk8rrm7lv1Qk0SgVPzRYeP/P7elNSZ7o6myP5YyEndnxOJouNp9efI8hby9Lbh7Bg3CAUkowfjSz9dgsgQ0hi18f08BVFkRNMcvNMqjaaKbJ40Utn4q0FAzF4azm46yfUVZl82jSZTeeKBAOwg7B5h1BTls93+05hU/uIws5x3Be/PcS8jw9jrCwQz+2eJEsSzPtEAMsb7hcgYxtZaZ1SgBa/7N+PZG3C7uMqxi7HLqZBrce4ZjFXCgqv37SqEyuFjGjav1tLPp2h9elc5ua4f42yGtliYsHQcO4YGcnOpyYwZ1AYH+7OZM7SQ6Jp1UnUmSxkFFUiqTTUNpp5dM1pxr25h/8eF2vI9v0HOZtf45oGdrVoATh1GBH75+AQFfUmK58ebMNwbil6WxeQNY1m9l8qY/bAMFHYO/K3e4YFklnWwPYL3biPnYwKd2NjwKYLQe6E/fj9qQKyK4wsu2Mwr87rT0p+LTd8cIBle7Navf+SOhNNFhueKklIDcF1rVmbiQnSUV7f7DLxBZFzNFVfFUxqMtuoN1nx1jmZSVb8vdRolAqqG5zMJA3bU0tYeyyPh8bHMjbu6ozlxDBf5gwKI6vaSmVNLcv2ZvHl0RwX2xDgwk+wfKzwaJy/Av7waZe5M8CFojqUzol2dvF++4ToeKPhn8j5x2Hay/DEWRjxYHtvmMAYwfi70LHUrdpoZl9GGUWVddgUalYdvsKUd/fx4JcnyTYHECLVcPS5Cbw8tz8xBh1DIgNIivDni8M5nQOOHTCTXtmSRnpJPe/eOqhFUtgqPAPQ2euJ1Bops/m0ALQbzxZyOq+GMXFXBwi9tSruGh3F9gslvL37CnU9Rok/dDDG3u4h1rO1+87y3s5LrSWDbaOxqj042EbdUGU0s2R3JhN6BzG5j2sfjwj0YuGwCN69dRCH/jaFQ3+bzLsLB/GnibGMHSTW996m81SqgpHdcuHzBbXMXXqYy2UNfHbXMM68OJ2w/uPpQx6nMwt4Yt0Zhr6ykztXJvPF4SsUVHdjeImpVjT2rgImNVttHLlcwdI9mTyw+gSj/7ubP685TUltx2y7mkYzuy+W8nVyLu/vvMQ/fjzPg1+eZN6yw4x9Yw8J/9rOiNd2U2U0M3u4qHUS9UqG+zhYrh3J+66JmdQEag/MWj2SqZYFy/bziVPuaLeLe8bBnJZlmX/+mMru9DJentufG/u3IQQ4GHE91MLywGaXYdMTLjP3XxNdmNH/Hl2H6uoP+T26DGszHFshHO3du8puIcsyFQ1mciuN5FQ2EmhJZIq8jT2nLzIwIR5DF5TQzkKWZcpLCwkGFq/L4nChlYgALzY8MoYB4a5uXnAPUfgFUM+Dn+5hP3TITMqtNLLhVAEbThdSWNOEj1bFvME9WTA0nCGR/kj1xbAdMDdw58hIHv7qFHvSy5jRrwPGT0dhswrds6PY/OF0IRvPFvH09N48MTWep2f0ZvXhHLSX0sAEzbWlLMtqJkZ1irEqA98fzGPOoDBXUQLsyyin2Wpn5gDBBjPEJlGhCkWX8wv7Mu5nTC8DGmfB1EayUNHQzNbzxXiZhjG/4ltWZB4nKKQnqvpCTsm9GGCx0dBs5Z0jVhwDiflP/iCeig3DByi5coHDxQYCdGoiA3VEBHq2TA1pFb+8IJhNf9zYQum22WU+P5zHAtkbjGVcqcpC0tqw+kS0uyHtdplLZfWcyKnmZE4Vwy+bucNSyqubzjEgQs/cploUDlPnlzdfoNlmZ9uT49lxqge2o2+ydv03nOwVSLTei2i9jmiDjii9F0HeWhcNvOKSoOP3uQkyf8Famc3etFJiDF7EObtMlsZW4MXnh67wypY0piYEs+zOIWjKHcljRWYL2HGyQsGRy5U8930KPfw8eHVef/65Hu5lLU1b30a1eB3Bji4j6T+L40eM6vDyWXX4CgcvyozXwGl7PC+szWXVvcF4apTsyyjnTF4N3lol/l4aVEqJrLIGMksbyCyrp2+jnq80zezcuoGT6uFEGbwoKM3BqPFBI6lpVATgB9BQzsazRXiolLxxy4BOafIdRss0PpGw7Eor5alvz1LvAIgkSTSbevp78uLtU/A1jIU1s/nI9Cr3nnuBdcej0CgVjIwNZFKfYKYkBBNj0HX8XE1VwjjUrSPoDFmWySxr4GD9SG5WhVH57fMsbn4VpUKJXpHMOJWaRq9I/IH0kjoyL1uY3VzHsp2pDIgKYbSpHjW0YulklTXw8uYLHMysIKGHD/+5bR5sXspNtd8iKYt537qIefYQYhAAiARUlhZQ4FdDoE7T6p7tdrTITUUytSWliKe+PUt8sA9f3j+i3ZqpUioYG2dgbJyB/8ztR86OJxhw/AXeOrWdmzIr+OC2JIZFdwK+dWDA3Wi28sXhHJbvv0yMOYzJGth8sw272Rf2wgv7avko0c6GUwW88FMqbyj7My1zNy/+eIYnpide+5pefE5INxHnMDm7ikupJu62NpKSXcg9X9Tw9QMjSYq4Nolyqyhz+LZFjQGVB8aKfJZuT0elkBgeHciQqAC8tZ2kBEVnRKEPrRlNzfXg1xNZlvn7BjHeefnioZ2fc60PJM5FtjTxZMWtLKxZxfTKLDq905znxm4RrFHnNETnny02LhTV0mS2M8bciAJoRMP9q09QVm9i3YOjCPEWiXD/QOgT4sOnB7KZP7hnx/d3RZaQ0Y5/uqVwfG/nJbLKGlh93wj8vNQt6/jL03uwdWsyqBGNiU6iodnKlXIjEdqeGK9c5MutF/mrsQqVwzPpjW3pLLT7kGSwM3ZYBLcOi8D49TJMOb6sMw7jyroz+HiomD0ojAVDwxkc4Y8kSVhsdnbkwExrNSFSFQUWH2zlDfjZPNEDekUjVpvMhYxLDPfv0f4z1nghL1qLbcVkGv93C5K5gfKQsegbLdhkmX/uKOUTYFHPciiHn/OUzB4DezPKeGxNOn0sj/Kt5hWyVtzFjfKz9OnhS0IPHxJ6+NI3VHwfcBWGSKsoToEd/4D4GTCqk8ac1peamkpe/e4cBdWNvHXLICL1Xg75vjhfZ4qaGIGZRyaKwsffS8N7tyVx04BQ/vHjeeYuO8yjk3rx2JS4dnv2uzsymGwzExbix+5HJvHLhRI2nSsitVzkD0eOH+PLo/54qpUkRfgzPDqAYdGBDI70x8ej/Yj6lmmtGi+2Z+ZxC/DXyT1RpPnwxeEc7h0b41orWta8sFaH2JZagsUmMzfJwVhy5G/jwjXEGpR8tCeLmf17dL1fFZwQLF03Wdn21BJSj9bwrKKBB1fuJzEqlLvHRBOo02Cy2FiyO5OkCH9u6CeOPT0xhJc3X+DtHRlsOlvE638YwNCogBbWkacKF5jUwkxqJtaxj+VUNLpyU5NoGl4NTCqrF0Wxn1PmZrMgSRJBPlpq6sXaUNMs8/ftKfTv6cszM/p0eTz3eGZGb9IyPTEaG3h7h5DKvLEtnXtG9uSJ5hV4pHwFYUPglpVC5thBmK12LpXWc6GoltTCOralFjMgyB+qaGEm9TWoCJQaKEv6J8Hj/tLlazL1mYPH3pc5euokiX0H4eelxm6X+fZkPm9uT6em0cLLqjzmKCX+vTmNoVEBPHdDAjeay1FsWY+ftQJwNbTuHxfD4+vOsCe9jGmJHchw23gm/ZxSzJpjeTw8IbYV0NIqPAOQbM2MMDRwpcyHnw5dYUJvA3/bkMKImED+PLkLPyG3eGBcLCkFtXyy7zINigheVsOGuj4MrTASGehF8pVKvj9VQGNaLsuB7w+e4yxNfLg7k76hvsxLCmNOUhihfq5rg+Xjoc9MmOXGdm3JzcTn8v7OSzSabbwwq2+Xry88wIvwoY59rEIFFyBYquFgc3/O77/MQ+NjWb7/Mkt2ZxLs48H3j4yhb6hYg2IHT4GLyzmw2J+zyn7sTCtjZ1oJL29O4+XNafz1hj48OqlX5/es+yAgt5BlmcvlDRy4VMHBzHKSs6tocjQaegXpSIrwZ9fFUvZfKueZGb3pE+JDblUj2eUNJGdXkVpU24qsrtdpCPLREuSjJdagw+CjJchby6AIf6IRDVeFxci8iEbIhnrvGHzavJ6CRjWhah9KczJI9S+hyWyjyWKj0WzDZLHRZLZhsdkZGO7PTRYTJlnNF6freRQYFwZvbc8gRq9jZoJDnu0Aot/flcm3J/N5bHIcd43qgPHvWDseGxnAQ4fK+WDjUZ45t1pcy22lkmmbhOVFF5PiAFHPazrJu3+PLuO6gEmSJN0ILAGUwEpZlt+4yr/8vxPnvxfsl3mfAIIdsi+jjJzKRgEeVYivRrOrszhLaWGKGt7ccJAMOZdBEf48O6M34+ODOnsWQGxcx69UsetiKbvTS7mj7gz3K5XU2D35y9RQ7hkbjZ9nm4TGAd6sXBDFsxsvQxOkVMgMRJhXbz1fzIZThRzPqUKSYFycgedu7MMN/XrgoXZLtJw3mNnI1MHBhPhqWXs8r/tgkoNFZdcFcSqnin9tTG218fh6qHl8ajzEjoTVsGJ+JGWG4fD5W5QZe/D2jgze3pHB4Eh/Zg8M4+aBoWxNLcbgrXFRHyUJEmYyMvUrBn9xEJWHN8+FnWMxUIY/2iYLOy6UsPlcEYezKrDLMNMwmgXSN+ycWY9+wnisr1Sxry6Jtz8/TkmdicY6f1BBnt8wVqfDL3mFHALWbN3DRzZXYSdJEOrrQaQDtInUezG8+RjDT35O84g/o42dRL3Jwg+nC1l9NIfsciOzfPTMjVKhN/hAMnyaYuHeJBvnC2s5kVPFyZwqTuVWU2cSgESwj5YRfmEoKuzsOpHK/474M8mzEnOQlpS0UraeL+GvN/ShV5A3j96YRP2lfowyXmJ1YS3bU0tadY91GiVReh0xBh2zbHu4CTgrxxGpCeVs8nEe3HsSpUJi8chInpreG3+Li5m0bG8Wb+/IYGb/HixZNBiNSuGi/VZmgakWWaHGrNTxwOqTWGx2vn9kDEkR/qx/fArblt/EopL1THn9a6z+scTptXxc9DNpfhP4ZUcmPloVPh4qfDzU+HioyK1s5LWtF7krbgAUgO/QBeScMjLjgwMYm61YbDJalQKzzd6yUXprVcQFezMlIZiEoPnYDn7I0oFleM2fgSRJXFn2PkWlvnzwzVmKa5v4QtaxqK8Guz6OpXuz8Nepee6GBIprmyiqMaFWSvh5qvH1VOProRbv2T0cCUuTLpxPdl7iw92Z9O/py6vzBlBe38zF4jq8NEoWj4py3VcPbYNVN/GN6U3OzfqKn8sN7Ekv45UtabyyJY1ovReTE4KZ3CeYkbGBrsKnsapVUVDR0MzhrAoOZooEo7ROdBvL/G/jeev7bJpaReyEO2n4dAkZ5eE8/skxBoT7syWliD9qVcwG1uw+RREGXve5yCKFBoVSTUOzlY92Z/L5oSt4apT8e3Yii0dFoVIqIH460rHlWPQJrKmay/crj+HnqSa3sp5zkoJ1e0/xzs44lAqJhybE8uTU+NbrydXCISepVAex5UgO/958gWFRAay8e3j7Na5NqJUK4mc8BGlLWea7n5trhnPriqPcNSqK24ZHkhjWprvcwkzywWKz882JfD7cnUl5fTPT+obw1+l3wZfvoC89KoATFOwqVDH/48OkFtYxoXcQE/vejv+OA1w8uZ/RJ4sZFatnSoKYHBTkoyXQSyM+t46ioRzqCrGEDODHE/l8mZxDamEd9+kEO+yXh/py6/oS/vj5MdY+OIr+PX/FNE4QsiSFilJtFGqlniMnzvGpVYxotsugVEgkhvoyPDqQETGiSDZ4a8FuQ97yFDaNDypzHd/vO8mXe/Q0mW1sbK7GI9yXb0/kcyirglfm9b86eLhwFRLwx8xyTq3ewrSag4x8+WdqLQrGxRn406RerjXdjYViq8giy2zgXH4NZwtqOJdfQ0ZJPVbHuva83wUeBp75MZPUwnpW3DWMwZEB0Cj+LplqeWD8OP76fQqrj+SQFBlAoJcGf50aH61KJPZHPhQ0+5F/Iqusgc8PXeGbE3ncMTKSib0d+7Pj3pscoUQRUk1TpYa0en+CpUYulzeQXW5s+Zpd0dByP36o1jFQymZVSSbPa4z8lN6I1KOQb0/mc39oCDqLg3FYX4ouexuMeJjdM24kObvSYYZbwNpjecQG6VgwNJwLRXUElKu5WS0zwb+c8zX+PL7yGLEe9XwNPD5az7iQAWi/LyPbFEQvRHPiYkkdx7KrOH6liuM5VUQ3/pl1mlfRSlZW5an46LVdBOjUKBs1oIIpvkVQDp+dM3NQdY4NpwvpE+LDB4sfpuKknelH/8Oy8EOsYi67L5ax/qRLVhNr0DG9XwgzEnswOMIfhZtkpsls42x+DTIy/hqZ+A13gzaA1KGvU5VRTp3JQr3JSl2ThTqTlXqThXtqoLQ0n61SMUqFxNxlh1i+eCgjm+vBO4SKhmZOF5kYp7ARGdC6cJiWGMLw6EBe3nKBj/Zk8cuFUt5ZOKgF4EgpqOHL5Fzu0Svx8vQAhcTMAaGiWSUPgf/68PckJSNjh4j9ObeKpXuzsMugkKBvqC/DogIYFatnTJxBrFMOZlKNVc1PaXXcooAYHztPTe/N1vPFLN93mRdudgCRdYWiodKmSbDpbBGxBh39ezrWLQeYpGyu5c+TR/DMd+fYdbGM6R2BBeCaFOlkJALfncznbxtSeELfAxrAWlvMR3saWHs8j7cXDCSrrIHiWhPvLhzUUvCG+Hrw8Z1D2ZlWyosbU1mw/AiLR0YR6i+ACK1SdsncWphJZmIMYh3LrmhwgUktvn9ds4ic946vzuWZJF6Llso6AWK9uTObZou/KxfpZkTpdUQlRSNfSif9yRvJq2pk2d4sso98j4f6K1bJs1hWfCfKFTlo1XloVQo81Eq0KgValZLaJgsZJfWYHSwtH62KxDBf7hreEzYhfICAPoHi8ztRZKY6OZfaJgu1TRZqGs2Or+LnSqMZTX0ghz1g/w+fcad9DkkR/tjsMucKahkRE8gTU+KJO7YRrwIvtjw0zrUXZDmlRoWt2NE39u9BqJ8H/zt8pRMwyclM8iCvspG/b0hhcKQ/z97QBSjnYD76G3NQePdhye5LfJ2ci7+nhmV3DEHd2T7XJgJ0Gr66fyRVRjN7z8fxxTE/XjtrwHpmH4E6DVVGMz5aFQ/0ioZs+N9tsVjjprHlXBE/nS3iv9vSeWN7OqNi9Mwf3JObNSfxqitALk5hX3oZ357IJ1FZMtcAACAASURBVFLvxWO2LHwVasoI5ONNF1hzLJc7R0bRO+TqcryWcFM2yPo43t6RwcYzRWSU1jNrYCivzO3fWlrpkJMqi04ydNx4hkYF8veZCWSXN/D+rkze3pFBtdHMP2f1RXLIFSsamgn2cTR53ViK1UYzhy9XcNABIBU5WEexBh23DgtnfHwQw2MCW/Ki3EojL/yUysubXcM2NEoFSRH+/GVqb0b30hOl9yJQp+n6XBU4aj5LI6P9RUNmZ4mOoT5GkrMrOZZdRXJ2JUW1JrZqAilKv8DD59v76CkkUClEfp6lbWLt6XIuWKNBCcvnR7JocyNPrT9LxD2J9AdQallzLJcPd2dy67BwnpnRu+PX57BMmRGt4h57NMlHt4MWzMYqvjmaw88pxQwM9+PBRCvB6+8S/mSDxfTFOpOFvMpGciqN5Dpq9ZpGC/+pqsHu40lFQQ39wvzaSTx/j87jN4NJkiQpgWXAdKAAOCFJ0iZZltO6/s//B0KW4ehS4ZfQawogjAj/tuE8aqVERKAAFkbECGZIlEFHtF5HeK0vfPUhH83pya7mPqw9lsddnx9nXJyB0b30nMmr5mx+DUqFRP8wP/qG+nKl0siBjHLqm61oVCLpnh6oQlkdxJYnJnT+Gh1gUk91A58siIOvYOnRcszlx0nOrsRksRMbpOO5G/swf3BPF8rfNtSuhUWlVHDbsAg+2pvF8v2XWxYkq82O1S5jttqx2u1YbDIWmx2LzU5QQyZPAs9sLeRH01ECdRqWLEpqf7O6TdYKjvEAawnB/aZxcMJktqQUs+lcEf/ZksarP6chSRKLhke0OoZhyFxI/R9rp1v4uqoH5WmbAZiy/CJNijxsdpnIQC8enRTH7EFh9Anxho+WoM/dCo23oLKbGDl4EP89VY2/p5qVD02FK/8kstcUfiSOt7ZnUF0SxB96NjN7zgTqTRbHYtRIXpVYlHZdLEVqKONW7d+5IEcx/8AIvE/tpNliw2i2MSjCn2V3DCHsTBSSqZKpocKP6rvLCt55aTtOzCc+2JtZA8MYHh3A8OhAwgM8kTLM8M0S9j6UwKHGCLzXGVmZVs/S9DP0CfHhwfGxLZ+FT+8J9D3xOftmV2NJmEthjYmcSiM5FYIhl1NpJK24jlG1x2hQeDB/fRmfqwOI8yjnsz8O48Clcr5KzuWns0XsVdZSpDCz7sfzrDmWx7ykMN5ZOMhVIHv4imlBlVkgKZB0Bqb1CmHr+RJemNW3hVHR09+ThY+8DEt+5OPYZJZ6DcZQdhQvu5F1tQPZciSH5g58LcbHG3jh7mFwZQNxMeNZN6SJ93ddIj7Ym+mJPRgaJRKcuiYLZpvdtSk7o2gyutw9LT/GeBgpCgzj5/PFKCRQGoIJU9XxzIzeVDeaWbE/m5UHr3Qq3/BUKx3gkgo/TzUP1B1jDF4MeF3I9xYMDefVef1bAJQOE33/CLh7M9IXs0jaezdJ92zhn7MmkVfZyN6MMvZmlLH2WB5fHM7BS6NkbJyB8fEGphQV4mH3ZsXPaRy5XMmFIlFw+3upGRdnYEJ8EOPiDYT53ggfb2XApY9h6mJ0TZko+0yl4YqNXWmlPDKxF4+G2eGHFfzycF+ONEbS9ONX1Ng0PP/VSc7m11Ba18ytw8J57saE1mybfvPh9Jeo533Ee00xvL0jA723hpGxgZjPBzKrh4qEUcP4Ja2ET/ZdZntqCQ9NiEXrSPadp0Zy8CUkCeyyTF5lExeL60jK38+fgIkrsmjAi4m9g1i+eCiemm4CUiotDLwV3+RP2PrsKF7dlsmaY3msPppLQg8fRsXq0es0BOg0DMstJE6hZXVyIV8ezSG3spER0YEsXzyEoVEOUCNmvPD0ipmA5BvKnYN7sfpoLneOjOTlOf1QmePhFwUrRlWxXBnD7oulrZI5gAAvNQZvLXpvDXpvLQadhkCdFt/C/dwLPLDDzH5LCr1DvHllXn9u9VfDNx/RQ1HL2gdHctuKZBZ9mkx4gGuNlmWQkd2+d/5efKfTqogI9CIy0Is5aUfxIIzJbx3iW40Pid4N7L1/EoHeGk7nVnMiRwAMa47l8r/DQuoaG6TjQY893F5+hmfMj/KOegWmqgL8QtVIkoS9to7vU2t57cxFRsfquXNEN3zfHDEuzkBtQhKKrB9Y3MdOpWcUm1OKWbj8KIMj/ekT4sO9uQWEqQ34WCp48+stfGoWa6WPh4qkCH/+NLEXgyL8aTRbqd0mPLJ2XKrlP/MGuu45J+PMVMuccWF8tCeLf7c5NyqFRC9PI5tta9jndQMrv77M8ZwqtCoFt4+I5J83uXWx3aavjfUpJbM6kluWt5YT+3qoiA3yZlxcELFBOnoF6RiaNRRDyglSnx4GH0BqlcTKb87S09+T2MhIuHxJ/POZL0UhOuxeFAqJMXEGxsQZeHluP7adL+H7UwW8tV0wKVYM6wepoKm5QlzsLOqzrZyuBzTQ06OZngPDqNtcx/aqCF5aeYyUgpqW5kR4gKcDqF5Avckf7a6/MH/qBKobojhyuYLn/zAJvpGQHMMZgnvGsv5kAVMSgvno9sHotCqY8TTUnmPaxeVMu3sWRE+nvL6Z9JI60orqOJRVwecHr7BifzaeaiW9e/jQO9ibvKpGzuTVtBTicVIBu7TZPGd5kPWrLrW7VjzUCnw91CyUdPQNtJD8yFQqG8zcv/oEiz8/RrJ3BQVNBl744gRj7Sph4GA1tesy+3mpee/WJGY5WErzPj5MYqgvMQYdF4pqCfLWEuGvbm1uDGKBMsThVXeFWQNDmTVQsKEbmq2czatpAZfWnyxg9dFclAqJIZH+3OaRwQLgg/2FVFu1oAHMDfQK8uYPQ8L5MjmXSL0XaqWCiXlZeKqD2H2qAIUECknCbLOTfKWSJ6fGu/YzJ7O8uY65SWEs2Z3Jk9+cIS7Ym/AATwJ1GjRKJWqVhFapQG8u4m5jOYebY7iSnEthTROf7LvMuDgDD48bA9+8xxcLIklT9+cv357hni9OoHXkmGM6kIxNTwxhdC897/6SwaojOcgyeGmUqBW4DLjdmElRei8kCT7cnUlydqUAcYypzAR25looNua4PuY2z5VWLIB+f28HwOYAaEJ8PUi/UAVa8PT0YNXtw+nVyTj6LkPtiWQx4aFW0jvEhyWLBlMWkgz7oS7pIWYqDDRb7DRbbTRb7TRb7YJtYbHh76Xm3nHR9A/zY0BPPyIDvQRYaqoTYJLDKDvK8bJ2Zhn56ZJge3ioFfh5qvH31ODnqSYi0IuB4X5EG6KpSRnEY5xHk/A0+zMrqGo0896tg1xsylQFaD1aNxWcDJbaAtxDrVRw95ho3tiWzsubL1Be30xRTVMLY3i4nMNNwP7sOt7dcVqcp0WDuwYZWgz+q4jrF4OUIlFe38z6P40myOfaVRaBOg23jOoDo15hZq2JH84UkFZUx/TEENHUbsiHJeAv14O3lnvGxnDP2BiuVBjZeLaQn84U8tyGFHpo3meCAmoK07l31QkM3hp2XixlgPIkwzUGJr17AItNZuHQCJ67sfsMNkAwalWeYG1i1PAR9DnhQ2mdiaV3DObmgWHtH+8VKBqsbSa6xQZ5s+S2JPQ6DSsPXSG/uhGlQuLI5UpqGi34eaoZGO7HLfZk5gHz1+RxtrYMWRZ7yrh4A4/HBzEuztBp0yZKr+PL+0aQnF2FLMtE6r0I9fO8dmCkhUDQQKg1nxIMPLfpMla78C/U6zSMitXzSGwgIRfiiDYW8PPCcXiqlXhpVHiqlXhoFGiUCmx2mTM55ai+shPg68NjY8fA1vfQNlfy6V3jmbfsMI+uPsoBBXx2tJD/VqQyuU8Qr83vQiHQsg9X8cKsvqzJ+Raq4EjqZV48dYHYIB3/O5zDhaMXWauC3YcOsvRoNHmVjVQaza0OFeyjxddTTb2xkUv1TTz1yVHS/nMD7Vek36OzuB7MpBFAlizL2QCSJH0DzAX+3weTLu8Wk1zmfdJSHQ2O9Ofgc5MJ8+/q5hVsnt5eTfQeHcf942JYk5zHR3syOZRVQWyQjom9g5FlmdSiWvZmlBGo03LTgFCm9g1mXLwBL40K1prA0jWbyQXOVODvI053/9hIVhfU8och4a1o812GUiUSBIdB76IRkaw8dIU3tnXsd6FUSKiVEmqFArVKwViEx1CfXr34oH8S4+MNHU98cHu9WJrE5KSAKCICvXhkUi8emdSLrLJ6Np0r5nBWBXeObEN/jBwNKk+Smk+TtHAR8nZf7Cc9eWrWUCqNZm7o14OB4X6t32/iPDi8BErERKUBif3YNHosBm+t0IxHPgfAQODrB0bCqgQCrCXg6Gy0FJzOkGWsX92CIreZi1OX8YwljNwq0aG8dViES6qSGSw6hg5WyyNzJnK52sqw6ECGRQV0LBFwGLwrG4qZGJcIWEmIjkBTpOCNWwa07syNelSMNP/+XtTxNxA96x2i+0RCm33U/ul/MEuDWTVxFEmpQ/BL/47IvsFivO6oSJbsykSRZeJsiYk1+XncPiKCV+cNaH99G+KFzE1nAC89z8/sy+hYPYvbUFTVfqEw6FYSz2/g46ffgr3r4Iwn7z73F97VeGG22ql3dKXrTVZMVhuDwv3Fe4sXY2MHRWhZde+Idh9Pp7KK3jMg42couyj8TRpKCQsfygcTk/DWqvA+1hMaypAkif/M7U+0XkdNk5mIAC/C/D2x2WXqTKKDWNtoafm+rslKbZMFvbWEWm0YT4/vzcBwPyb2DuqeTC4gWowTXzULVs+Be34mMjiBu8dEc/eYaJrMNo5mV7AnvYy96eXsTCuln6aQJlnDF4dzGBoVwF9v6MP4eEPHnZTJz8N398DRZdBYiaHXUHbPn4hdlsVnVSDuZ29LNTP6jcB20YeGTB/2pJeR0MOXTxYPZUhkQPvXHTkKni8AhZIJwITebutQQSgxHo3EJIYwLTGEOYN68vyPKTz/w/mrfhySBNF6Hbd51mKy6Xhu7nAiArwYF2/odsezJUL6g92Cd0Mub9wykOduTGBLShE/nink+1MFNDgkiK+rctErtbyyJY2+ob58ce9wJrU9f7GT4OJm4Z3mF8GLs/tx6/AIEkN9xeM8AyB8BIFF+/nHw//hHzf1JbfSSFpRHRVGM5UNzVQ0NFPZYKaywczFojoqGpqpM1l5ykOMWO87ZBx/GhjPqNhAccxih7dLQxnhkV6sfXBkC0Xf+Vm5g3Ftf4ckwNULhbXsSC1hsTqTfF1//jYhgX6FCXhXnAW9SEgn9A5qOYfNVhuphbUcv1JN+uXL3Jz/GemeQ5gy988o925kcYwHi+ePRLZZkV4xYVbqkCzw5i0DWzFPrn6uJW6ePB6y4IlBEvTtz99n9mX9yXzWHc9jT3oZCy1VXJSD6adoZGpwPQljBjEowp8Yva7dc8llwdiSNbx722DmD3aTByhVDuPmWrQqJT8/MY7MsgaqjWaqGwVDoMpoZuTlJajK7WzR3UKTxcZT03qzeFRk+33KLYlVVVykR/xkHg6IJVqvI9ago1ewN3qdpv39b+4DZ62oK0Rq9OebhmMqiWReUk9UmfsFW8NmhVOrxdQcpz+fI3w81Nw6PIJbh0eQU2GkoLqJcZ45CDWCjF9QOD/eOIaKBjOs9XAMn7DhY61G7R9KQXUjM/uHMjI2kJGxenr6uzeO7oX4kYQb4nnRHUjx0gtWsVLDO/dOY39mJbMGhLqaCJIEc5ZCSSp8fy88fJAgnxCCfIIYHx/EwxN7UdtkYV9GGWfza0gvrmdPehkhvh7cPSaK0b30eKiVSAVK2AsLJw5jdswIfD0EC1QwVFUuVub6L8Ua7iEYoj8+OpYnvzmDnFNHplmBzV9mQt8IyKRLycLUviH8EhXI8gOXSS2s5XRetaM4HILqsBlUHbD/9HGQ13pCk7dWFHjj4gXoYrHZOZtfw/6McvZfKieluJgFEvySWc/cAdHidTl86Z6cGs/PKcW8uFHIT7dpckiT9Tz73blWz6FUSMxLcjPldgNHVUoFnywewppjeRRUN5FeXE9tkwWz1U6zzY7ZameO4jB3a+C1czrSzgowY2b/HnywKAltpSN/ayglsd8YNj02jre2Z7D+ZD5/uzGhw8/O+b5fmt2PeUk9eeGnVEJ8tUhNNjdmknOamxkPtZInpsRzILOcHRdKqTKama5IYaYGPjhSxQX5QqfP43wug6/LgBvE+Uup8YIK+NfsgUixV/fp6TBUHsLHxS2CPQTA+cSNg1zAybWE8/5xyNw0NsGg+uucIfwjcSq+nuquWbrqW2Dnizw9ypenO5Lt2cztTYKdEwA7MIi/fXgky/df5sujuYQHeBLm50lBdRNHLldyxZrJTRr4YF8eqWj5+M4hV2eWun0muoAQVt49DEnit0mwHdHDz4NHJ7WRybmtt+4RY9Dxl2m9eXJqPBcvnqfv+hQq8UMv17L0D72YMaQ3lcZm+OxV8huDmdk/lCenxhPdmX1AVyFJgp1Uk4smpA8/PjoWuywLQL2ziBgBWbtEl8dtL1AoJF6anYifp5oP92TSw9eD6X1D6BvqS2ZZPefya6msvoIVJRGRUUwJ8WNcvIGB4f7dBoQkSWJ0r195TzjDTY0iVWShDIpjZmAoI6IF+zIu2Nu1x1XHwbkT9Av17XCAjkopMbyn2G/+MCIOejma3g3lBMVp+fL+EazfeRQugazSMm9wT16d1797oKaxApVSwe2xTVAF0Tozmx4cy8Bwf/KrGjm8JQeyQV19Gc9QJTP69SBK7yUIHnodkYFeLedRXqKmR1A4nw8f1jmT/PfoMK4HmNQTcHfeKgBGtn2QJEkPAQ8BREZ2v3v5/+s48pFgY/Rf0PIrD7Xy6otxG5NarUrJfeNiuGNkJM0Wu/BlcItmqw21QtE+UW+sAN1VFgzn9DRjeYtM6YlZw3jCfaJTd0Oja5lOEubvyel/TXexSGSxYKiVClQKqf1rPVcFP8KfZo4CQxdjKz0DhO7eWO7SOreZ5BYX7MPT0314enoH9EeVVkwrckxWkoxlSD4h3O/G2GkXiXPh0HvC+wrAL5x+YV1ISfRxkPZT538//hmq7N1w0zuMHzGe8Z09TucY/V2TC94h3Dq6g0kJbcNpglpfLHwHgMmDenH6/untixf/CHhgNxxfAXteg2UjYfI/YeSfXBNErGYUpal4jHxIyDiqekNKvShsdIYWQIHXrNw+KoEZY6e6fI7afS69hDZZksBLT0SgF3eNju74saMfE5OhTqwUfklxU1vo/RqVAr23tsvxstcc8TPE18wdDjCpDLxDmDfYcS1eCG4Zz65USDw4oYvrpaNY1gD6hHZjdLsV+l5w92YBKH05B+7Z2mJ06KlRMiUhhCkJIciyTFGtiZDV/0LuEU/GwplXTy76zoWQAbDnFfFzjwGt1xfnWuQwYVVajPj5BXLyyen4eKi6Bgc6GI0MCENzNwPucfEGdj89ieJakbQ7pYhtWTQy0MPXQ2zs6z6F6ij+2Nn1050IdrBJytIgOIFAnYY/jo5uOWaz1UZNowWfLevRlBk4dt9Ugry1Hb/n2Mnia10BRI1GqZDarxHx08QEvYYy8A4WHXh914mr2WpH/cO3UBzN3+e38Qtz+rw5hghE6XV8sGjwtXwCLWE1VqN6u5yeo8czanwv+CUCsre1S3RB7EdDowIFSG78GAqaSbhvBQlB4XAytGUAgORoLNwxsT/zh07tOrHuLPRu8ljE9e4EUgH45CUh26jXMtKzGoZ0bkgqWZtQarxaA0nO8PBrWS99PNTtAdKmGji9GfrN48OFt3T9mp3eceXpYCwnICaJ50d37b8BuPYxB9MnQB/Mq6Mc+3ChXhTKF34Q3lAzXu3yUNEG4X9HjVsh7B1MXLAPccGIfbSpGowVSLKN+ROGMX/E5K5fX0cm4t7BIs/wDcNf5+Hy7XEPD1+47Sv4bKow877rJ9f+Avh5qpmb1LPj/20JURwP7xMFUV00yLS+Llmq49ir7h2B/IqJBaMSWTB9PJy6LECbrqZ0IVhK7oCJLMtiD91vbjVRqCX08Q5bg7KOB3ogmCDDowMZHh0opELJqbAdNj41nQAPBbxPS0MuItCLEy9Mo7HZil2G4I/riOg9mYOTJ2N3SE/tsoy3VtXaCFmjEwwgx/XcL8yP1+d3nM/Jsox96z7kszrWPnsvZrsCmyzTw9dDvFdvh02BY/33UCt5cXYiL8zq2y1geFCEP5sfd0yH/Nju5pnk+PxsQkb11PTePOXI1+pNFqQzpbADvn58Jjbftr4wrZ9Dp1XiWSIAdyczacHQcBbo+8AqkH7L9CW1pwBn7G5AmHMCn/pXeqconAbcjoEBDt+snsFB0JGhddtwmuCbajqeDm3r4PrUeovpkLXtwSQ/LzXJz091NHhdBbIsyxjPNsFG+PjuMSh7Duw8t3MP98EfuqBuGZ7/ptB4i8/UKY1sE5IkkVi8ESQJ/Q3/gO1/4+aejaBSCKWFvZTQpFkMn9PFVMzuhHeIyNX18d1jSIcPh3PrhAF4G68/SZJ4anpvHp4Yi6da2T533/Al5PfkwztaG+b/fxpuYBKVWQQNvJWPZnWSf/hHCEm622CJduH051J7tlKgAPQK8ub5G2LhEjw0qS8M6sa5UqrFNe+4LjT/h707D4/kKu/F/z2tbqlba2u0zD6e3fbYeBmPDcZgjI0XsI2NwxrAZrMvxA6EBAhckpDL/XFvQhaS/LKAw2YCIRDCHvCGjdmxjfG+j7eZscczmk0z2luq+8dbp6u6uqq7quv0pvp+nkdPS61Wq1qqrjrnrfd9z37JaF3bPQesksDm6iXdeOML+oEngDMHD+DMK/17smqqMIO+nu6qLWeonInQm98Zp6w2xLKsay3L2mZZ1raRkUXwj9r/hJQ9vPCq6EsJZgcky8ezikY201EWSAJkcO97Yp/YW7XmHNk8kErLY+3Bh9+qRaF09jgNJfX25jLy0Z1BT1canWmfoBfgrEDWU2V7Ux1yJWJib9kqDKFtPEcmJweelhWLeqv0dVp+ogz0H71evh6oEuwc2iADdc+VEgBy1fTGPwE2nQ+c+q7Kz9M7Kj0V9jxcvhxqkJ5hGUiOP+s0p81WyCzrSMsqg1f/Sq523/hR4LNnFyc02POgDPhW2CcJPeHZ71rRzbKAuSl0dOYqDzaGNklz6LHHqi/5O3ossPEVwE//Vhr6VljFzYj+FbJy0aM3ypXhuYnSCUHvaNn7MTTLkkFG2P+hn+FNwOXflUHtdRf7rhaolMLKfA7p6f3I9A6Hu0qVSgFnf7SYco+lniXMPSd1zB4Gunox0J2JlGVS+pyj0gfIpTOdKgZX9ER4nf2xfqQX60d6sWGk1wlIHNrhXG2t1fBmmdwErBbWle7A0v4suhcmkM71Y2l/Nvg1L1nvDPJ9VpYE4PQkibCaSGc6BfXcPf7L+fYMy/bXul+6pPdJWVRxJb++FfK+nzoQ/EN7HwXu/AKw7R3AiB2471vmrJBpT+pT2f7aAkmAnAt7RoH92/2/r5d8X7KhGHAKNDcpS0wH/Z5KS0vf+XnZ96s0yAUg58GOLuCpn8nXFZpvlygGk+6WW/egWx8vf/LXcr4Kezx0Lyzh/jw3KAGyI/YCFH0hext66eNDf3AQD4AcVy76W1ne+8f/p/JjD+8uD/QU+5ZVKVXq6nfOe1phBmp+1llpK21nXOnJS0jFc6jfZB2Q0t5UWs7vYc3JeGlkyRKku3V5mhMM6+1KY7Q/i2W5eaRmDqJ3VLKwdT/DDSO95StqKWXvzz6r2vm8po5dd0Ct3Ip8bzdG+7NYPpBzXmtuUF6TZ2XHmo791rxrNTedmTRT9rC+bAa98zIWHRxejuHerpIP3RBYf3R3pmUbgdIVHfU5LVW5h15FeuUm974yNyHPWWuQSgel7MykoBVSA+mVwoKCoYXZ8jJMQErdDu3y/ZFspqMsy0Mphd60ZLkuH8qHCyQBpcetngbM5eyLk5gKWNF0fk4uTG48V8a4gDN+mjkiwfDBiHMIP72jctyvssJa0Wo7c37nnYEP6e5M+4/dx3dVP+bWmw4m7X9SAkVDFS6U6rFvpRXd9P6czkrZYEdX6UrCxWbwEd533UNOkHGvPcbxnuf1GOfAk857Msj8DFdzq5GJYNJOAO7R9SoAVdbeXQSWrAfe/XMZaEelUybjThIm9lU/mCslj3EHk3xWcwuls6d4VS2SiX2y3PLoceF+d89oaTDJk5lUld2/CttvqXgVsUgpyU4CZCDavaTy43Wzae+EvzAD/Ne7ZFB7yT/6pnuW0Nu1+97gCapXqkMmBq7MpFDBwfwa4E3/AbzuOhk4/uvZwPUfcSZEK7bK7aB9BeWAK5g0PwvAcgY5QXRZxtT+6kFDQLKTClMyAN18QfXHx7XpfGDHr4Ax+6RTMhEblcnk7KQEdL55lQSewpjcJxPZqEFPr9FjpOStMC0lbz5XGWVJ5YPVg3Vumy+QZbaXrAdynlT0zm4Z5OqT+syR8IPeID3D8nzeS8xRjO8KP2gLksnKa95TpeLatbx9IKWAdS+Tz/MB79VlJ8ix6/Gbwm/j1AE5zvkFk1IdcrFAB2/ieF7KW4qBD33Fe7zCqfrmj0lw5qwPO/fpYw/gNMeu9eKENrQB2BcQTJqxV+ka2igDVZ/JadHcVPAxKn+UBPp9f24a+NW/yHnD7//gpSc3u++Vr70B2iD9K2VSrAP5Wdd7UR8vxx4Btl7uP1n0k+50MgX8gknF1UxrDCbpc1SY9+JJvyvb/tO/AR65Pvhxn3kZ8PN/KL3P1QS/omy/TPYXnAVNylZjLAYyogWTiuZn/Ff9GdkMnPE+4N6vyYXEMGYn5fyW7pJJiurwH0Pp92HYCWS23zn/VzI3JUvb2w2By6RSToZ0XO7snmJm0qz/Yyf3Ba5I6r+d9vMuuINJhdLfVYti4Ma1r8xOht8uP0pJMMouycOsvW+bCibNzzo9qdwGVvmWuVVUbMAdIQu8gBQsSwAAIABJREFUJJhU56wkrXuJ/8VbAHj0Bgman/I2JwNIn088q+zGsuVS4LQrgzOyvUaOley2nbdH/12HdsQf/8TV0SnnKztjX2fL+9Lzl0MVgkl6X8vkXPPSsfLv++3bQXqGJVg4dVDGJR1dMqZyjz11MGmhIAkGlRRmov1+KjIRTLoDwCal1DqlVCeAN0Lazy1+y46vraYasN9IMU7gc9NykqpW5gbIhGRinz34UPEyk+zVSUKzLOD775MTwWWfqR5gAZzJ6IGnZcBRLRjkNbxZBmXbb5GJWK9P82OvLZfK7cCq6tvoKc0o+tHHZdJ2yT+H2+Ye+zHzs9GyWvqWy+BzJmJwUCnguEuBq2+XE++v/lkylXKDTsBOX8FxZybp/3m6SjBpyHWyCRPsWH+WZEStf3n1AJ4Jmy+Q5Yjv/qp8XZKZZO8jE3uAB74lE4Zf/XO45z1on6DiZCZpS48DLv+2/H/v+Fz596cOArBKU82rUUoCiW/+hv/3e0acycTskerZAdX0jkqQsJbAMyDHtsl9wEDMzCRAMuCCggjazOFw76H1Z8lt0P85lZJsu+23hA+k7bb7SAUFMXqXll69q9XzDwJdA84A1V0u6+epnwGP/AB46ftLJwy9y2RwNjcdPgBQzVBA1pFl2YE+O5gEq/S45DU3FZyZtO6lkv3kF6C959/lff+S94ff5u4l9tLmw+HPT6kO2Xf0Sj1+mUkqBZxyRfjtAOR8AJRuRzYv/6fDcTOTdDAp5HvxlX8lQdVvXeVcDHLTfRC9k16dvVFtbKL3Ndcqf2VBzUxtmUlFQZkfAHDmB+SCy/f/sDQAEUTvk0rJR1ev81rdikuBh/w7u8o2K3r2bplEBQWTAKBvqZmAtbXgNOCukJkEwF6RNMIFEW/pGOAEazpqzIoEXJlJrsDN7ETtJW5aR6eTBaGz+cMuOa6PYRWDSX79NFf6H98qKWaDRMjGyOScCXe1yghT3BkoXr/5ohwDN50n29a/ysl0NRlMOuF1wPmfCP/4jjSwciuwo0IwaaF8sRkszAPjz5kZ/8ShlLwP9BglbmaSfo/pY0PvSOnYRgeeowQ2u4fkWDJmL9qwcqs8j/u9486+3vdY5ecrzNSekZhwsYNJlmUVAFwD4AYADwH4umVV6ahHMkk4EmOSMGlHdMMczHVwRpcMpGr8t2e6S8rcQrn736Vx7Tl/KmVGYehMqgNPSXAjTADKTSlgw8vl6uH0QRksVbPiZAmoeGqbfQ2ulUG/+2r69ltkZb9Tr5Rmz2G4B/9B2Q5++u2+JcUyt4jBwVweuOhTwDtulIH/MRc6f+NMTiaa7omAHjRnqgw48kc5g74wA0WlgCu+L/02GmHlVtmu+74uX/sFk8afA277pHz+1M+qlxJYFnCv/XxRM+iCLD9R9oeDPldRdKp3lIE4IK9VB0H9vleSmRQzOKAnoLVe7a71NfoZ3SIZhJX6p8yMhwuIbLkEuPjvgXVnBT9m9akyeKl0hc5NX/VbFhRMGjEz0Rt7VDLf9PtcBxf8gkkLC1LK079Smvi76Z878rwrmBQ3M2mjPJ/3vTZ7RCao2X5gyO5hVqnUbW4yODNJlz88+ZPS+xfmJUtmxVZgbWB3u3I6+O3XZ6gS9zHCHcDUz7f5ldGvSOvzmzczadqdmRTiHOinV5e5hZzYZLLA66+TRgdfv6I8mKCzC7z/65mQ2Rt6X3OVijkZ157MpDDBHj9BmR+A7F8X/a1MVn/2qerPNTdRGuDs7AvITNJLgUcIJs1UL3PDTns1qUrBpN6lTjlkHJY7M0mv5lYhMynKBSRd5rbgKlMxUebml5k0NxEvMwmQQIK3zC3sRRq/AJebXwNuQI4bUweijdFryUzSC04AjSlzA+T3+WUmHdwhTa5PfqsTVHRnuuoxlIkyt1qsPk0uMM/6XIR/7l7gE8vKL3Yd2SP7ebMzkwA7gWBC9slK1RPdQ3KxudK4R7/H9EVpb0JFLfti9xLJbtKtDFbb7ZrdpW5TB5158liFYNLCgvzdWeZWEyPtyi3L+oFlWZsty9pgWVaE0G2C9cbMTNLpgWEO5u4yt1pL3AAZ6EU5Ue1/Evjhh4CjXiIlTWHp9MeDT9c+Qd9wtjPYCjOQVgp4yzeBC/+m+mPTXXJg1Vc/JvYB33oPMHw0cN7/Dr+NJcGkCCe7vhUS9IhbtrjmhcC7fwpc8k+l9y9ZV1rmpjOTgq76ax1pJxgXNhDQ1Rv+il1cqQ6pq9d/N/d+od9Ht39GSk1O+x9yYtleof/N/BzwnWuAX39aMr10w2cT8mucq2pu+upcd40ZkX56XFeIjGQmuVZkrEVxv46/OgxGj5WAhL5y5WdmPFxAJN0p/+dKV8J1T6Ld94fbvufukQlkb8BxvHepmRKUw7tLs1N0Nsu4TzDp/v+SUqyz/7Q8OKN/7vDu8kl8rXRGo7dvkrt8qVhaHFAOB0jAMGgSOHqcHJO8waQHvyPHupe8P9pFC318C9svSdPns67+0v1oYA1wwhuAs/442vMBdgmbKr2wlLMzk47stgf6NV5tLWYmRbjYsWQ98Jp/AZ67WxrSu+njlzcQMjMuAYhq26n3tekKmUk19kwqmp+tPKHZcLYsuvKzv608OQHKS6a6eksDYZruddPvs8y4n66QZW47b5d9Luj4AphpuQDIZExnJqVSEgAKykyaGIt2sUBnirnLG02UuflmJlXovRZWSZmbzkwKW+ZWY2aSDj4E9E3yVUtmEuAEk0xc8AkjKDPpt/aFyK1vde4b2uCcJw48LX/PRgW9vFadKtl0z91d/r2dt0tJ7WOedgrFwHKLBJMAOf9WSkJQyh6zVigj0+8xfVHaRJlb97DsF3seluP+8hPk/il3MOmAzEu6hytnJs3r38/MpFpw7btm0X2B3CfH+74B3BUyS0NnJoWpWdZv2tjBpJ7wwaT5AvCtd8vg4jWfDl9nDMhrmhmXqwu1pqeuPwvF3vBh+0UMbQhfqqSvflgW8L33SjbF73y2el8ht+5hZxujlEj1L5cSR91nIW5mgNfg2tJykigDDp0K26ha+qh01phKlQ6EdGDpgW9Jrfv5n5BSskd+6P88sxPAf/wucPeXgZd9GLjo76Jn0FUSGEwymLWj6TI3y5JgUuyeSaWrVUamBwJxjlXaiF7RLaDUrVhKFTMbS9PBhedDBJMW5iVzoFKfnt5RydqJ038KkHNNjyt4ne6U489hT8+kuWkp1132AglueOksmCO7DZa52cEkb9+kGVfmZS4v+1XVzKSASWAqJZlHT/7EtZygBfz87+T3R10AoOZgkh1s9wZKO9LAZdeG69nktfl84ITXlwancoPy9zjwdO39kgDJ5hxYEz6rWDvmQvl7P/3z0vunAjKTZo+E2498y9zs/dCbmVSpv1Yl8xXK3LTz/49MXv77Dyu/N+cmS0umOgOCSeM75f0Z9qp8Nl89mGRZwI47KmclAbJ/eMeitXA34AZkUlgxMylKmZunqTXgPHecMjffzKTJ+OfAjoyrzO2w/C3C9kErblNAS4lChTI3IFrfpFqyQQDJCMnmG1cS1D0kQQF3Wdh8QeZLG19ROn5e4logRy+MYnJsFoV+7/mVuo3Z57KnPMdInd3TEplJ9vm0Ur8kLb+6Spmb3td0ZpKnt6YO5kRtwL0wB+y6U3q26vYP055gUm5Qvj9WYfxQ63uBADCY1Dy9o3LF3J26+Yt/AG6tshKKNqEzFEKWuc1NyMQhzlXkzu7wPZN+/ilpdnzhX0cr4QKcyWhhqvbMpO4lMhAGovdcCmNoo0x+7voS8PD3gXP+zImKh9WRdgZUUU4cut/J2CMSrDOd2TO4TiaLOjVXXyELc7VOl1I16opVVBvOkb9Zz0hpgLPHFdh72Ydk4LfpPLlq5F5BBpD33nUXS3r1RZ8CXv4R84OV/FESRPCWauirc1F6JlXTOyrPO3NYjklxM5PilrnpgYC3WXgthjbIVeKgYNLshP2aDQWTunrl/RMmmPSLf5By0uMrLEXfu1QmTZVWIqumYP+89zjYv7w8M+n2zwCHngHO+4T/lUh3ZpKpYNLgOgCqPJhUzEyyg4pLKjTqBiqXuQFS6ja+y1k44YlbJTPsjPdFu9gBOO+/sM23NX0+M7Fva8ddKoEoN/38ex+pvV8SIJl977/Pf4nyavpXlGcnBmYmHQ533NH7QkmZm95P7P2wWplQJZZVucxN61sKvOJjEpy892vBj5udKM9M8itzO7QrfFYSIL0yJ/b6B6a08V1yHl91WuXn0mPRWjNJtYX50mNGutNgzyRd5ubXM8lAA25vz6S4ZW6pjLOtsxPRxmh+AS63qplJUYJJ07KvRx2/9I5G21/j6l4iwcoZVwD18ZtkTnPK20ofW+xpul2C6Sb6JdWqZ1gyNXW5qZu+MPLMr0oDufr81OyeSYATVB0KEUwaWF2lzM3TM6lnRPZlfS4oBnMiZMnpi9a77gJGjnHOe+7MpOmDEvgc2lglM6mGnk1UxGBSs+iBvb56b1nAvifk4OhXeuCly1LCNODWwZl9TzSmzG3XXcCP/wI47jLgBa+L/nvcKalxap03vkJu63HSW7JBrjj94IOyytOLrq7teXpHJSAYZbChB/Z7HpbgoOlAhi5V0ymrc5701ErWnC59IVrhqoqfXB5Ye0Z5JlhHRk5MI8c4zdiPfqVc1djxa+dxB54CPn8e8PwDwOv/rbbVHMPQ2+cdGJrsJ6T1jACwnEwoE6u5AbU3jjZZ5taRkYb8QcEkb1aDCcuOr17m9vwDcuHg2IurB5MAp7+eX8POaoKyWPtWlGYmTe4HfvI3EkRd/zL/58otkYnS4efsQaCKv79k7H4M3qwjd2YSYA8GK2UmVVjNDXBW49Mrcf3s7yQ45peBVc3yEyQTIGppaz2CSX50Gcr4znjBpDi6h32CSRV6JkXJTKpU5qbPU7VkJhUnFCECFKe8XTIPbvho8EpT3my5zoAG3FFXrzzmItnWB74V/BidDbFqW+XnKh5jYvZmczfgBuzMJJ//wfycBAWiZC/7NeDWmT9xeibpLIk5TzApbplbR8bZl2Yilo5Xy0wKbMC9AoCKWOZW41Lo534c+B2fBULqRY933O+z33xRsuo2n1/6WHdJ9MGnm9cvSVt1mgSTvBmM+x6T/WzmUOnFp0eul0zQWhd3MknPSyo139byq+ViQdAcUVc4ZFw9kwBnbFPMNIyYmQRIYHnkaGfM6G667c5Mmthb+j2/7eNqbjVhMKlZvFfvJ/Y6S4g+e1f1n58ck6s1YSZc+k07E7PMLdNtN0WtkNY9OynLqvculUaVtQQ6SoJJa6P/vHb6NcAbv1qnzCT7hNXZbZfx1fhWyh8lk90odGbS/u1mSoG8dCmGLnUrngRCDLCOfiXwoZhBy3q77LPAa79Qfv8l/wS89vPO/3LjOXJie+QH8vVz9wKfO0/eq2/9NnDsRfXbxuLqGJ4a9Ml9sk0ms9H0+01fEYubadKRkZN3rcGkYpmboQl3pRXdTC1v77b0BfK3DBxUzQLf/B/yHqlWHqn/N99+D/Cp46Vhp15dJSx9jumpkpl02yflHHTux4OfK5WSY/vh552V1kwEs/1WdNNBRf2/GdpQ2vjbq1KZm/75/pWSTbLrLuDJ26TBeC1XIo+9GPjDB6O/D/XEpt4TBffzNyuYpDOi3c1n9WSwLDPpSLj3oA4sun9+2tszyZ4cV2q6HyTKhCaVkvfv1AHg5o/5P2Z2snQf6QpqwP1s+ObbgASxhjcDv/1y8GN23il/i2oliu6m+nG4G3ADEpDzy64plmrX0oDbJ5gUq8zNZ1+Zm4x/fi0pc4tYOl5t/w1qwN2Rkf+lX5nbs3f7BzEL07Ud//Jroi8+EIfOBNX7zqGdkjV+8lvKywf1Ajm7fiPHCROr7Maxapu8t9xtCwoz8vVxl8nXutTt0C7ppbTlksZvpx99Ph0OEUwaqLKiWzEzSfdMsoPJ+mJXsZ1GxNXctJFjnPOezuRemJdxRG7QCYgFlboVmJkUB4NJzdLrCSa50/d3hQgmTYzJlb8wA3l3cCZuzyRrofIVv5v+VCLul/5z7QNm9xWrOCmq2X7gmFfV/vOVLD9JGuRd8k/xMp8u/nvgtRGv8OjB30LBfL8kwMlM0k249RWyMFewlGr9pTX7lvqXXm4+v7RspasPWPsS4NHrgSduA77wKhnUvuMG4KjT67uNuuGtt2+SLg8wmY2mj0X6/20iUNUTo6lr3MbyXqPHSumW38p8plYkc1t6HAArOIB1218Cz98n7/1qV+dHj5XsmblJYM2LZP/71aejbY8O6nmD6n0rZCBXmJHzzx3/KqviVMu26VvmrCZpqjxQlw27L1QUM5Ps/cBdvuCnWmaSUlLq9tRPpXly10B5iUS9ZQfk795X5xIRdyC2r4YSNRO8kwXAKXObmyztfxN2RUXfnknj0pdIBxXSMTKT9IQi7NXpZccDp18t5e5P/7L8+2Wrufn0TJoel9cQpaxFKXmv7vg1sDdgcYGdt8sqtdV69RTHojGDSQvzpZlJQxv9A9/FRSSiNOCuU5lbcV9xBb1MZCaVlLlFDCYpJb8/qEyzMBP8mvtXlmczz04CnztXSpj9nqsdJs/FzCR73/ntl+Vc4W68raU7Zfy0/Vb5upllboCs6AaUlrodeErmUuvOlOCX7i330HfldstrGrmFwaKUuRWz6QOCSfp4rM/R3v9p1GOv+zkACSZ19QNQzgVJPZbUmUlAcKnbPHsmxcFgUrN4m9Tq1QdygxJRr2ZyX/g0Yffj4pa5AcFX3B+9Ebjjs5IRtP6s2n+P/tt0D8fv31IvvSPAHz4QvXGrV9/S6MGorl5n8luPDKDcoEy0dKbKnCc9NUmOfpVkTHz5Mhnsv/NGs6u2BelbLoEDv2CSyX5JgCszSQeTDLznekdjlLkdlG2Ic8XZTf+/9j7i87t09ouhoAggE0zAfyK14w4JZJz05nDHjt5R4I8eBn7vl9Lg/8Q3APf9Z3BZjZ9iSbTnfNHv6n9085/LIO7l/7P68/UtszOExs2VBw5tkMxZd1mUd7W4YqNunyuLCwsyIaw2CVx3ppw7H/oecNq7zJY3hvX2HwBnfbi+v8N9ISfMaqb1UBzjuI4DU6791h1UmTkc7riT6ZaARUnPpIOl58E4PZNqaQJ71oflqvz33+9MiDS/1dy82d3F1Zsi9kg58Y1yjtArWrkVZqQfWLUSN8BwmZtrSrHuTGDvQ5LF6FZLMElnJvk14I5T5pbxKXMzkpmUdmUmReyZBMg+HJiZNBccTBpYWV7mNr5L/lZ6POdWa2ZSo+nVa6f2S9Dyri/JqopBlQtDG5ygQbPL3EaPk+OWuwm3XgVyeKOsdv30z+Uc9uB35PFhGl43wshm2Z4wZdn5gAugWsHTM8kbTKolmKOfo6NT9oVUSs4FOjNJl7Tl8vb308ErcLLMLRYGk5olOyA7rTszKZUGjr5Qytyqrd4zsTf8ydg9iYhzBV4PiuZ8gkkTY8B3rpYDz9l/WvvvAGTAlelu/kmglemrzfUIJiklUXw9+daZSUkMJm2+QAarq04F3nF943pBdaTtq4yeqzxT+6OVB4RhusxNP2etmUlTB82VuAFOMGnPg+XfM9VE2i1/lPQN8zbhnp0Evv1u+b9e8H9re+5Tr5RB111fCv8zQWVuOjvmwe/IFdEz3heuJEpnJplcBc8vUDQ9LoEDHSBasl5ufSdFuq9blWPUujPltqMLeOG7a9/eOJasa1zPJKB5mUl6cRC9WAhQury3ezWysKu5KSWPc2cZjj9b2iA8zmputfTt6OwBXvVXEjj55T+Wfs9vNbeFQum26cl/1HNL76icn+75j9IgCyAl2fOz1ZtvA/Ke6RooD/pEteApc9M9yp76aenj4gSTSsrc7M/DrpLmx5uZtDAvn8cOJnU6mVNReyYBcsyrVOYWGExaLZlJ7vmDHkP49VKqtWdSo7kDD4/fLAGySlmlum8S0PzMpI40sGJraWaSPs8NbZQenlMHgCdukWbcrVLiBsiY4D0/r/44QPpXpTLBmUn6orTuU1Y8P+gytxrKzLr65L0wtMm5+JjLO5lJ+jY3KMeJwbXBmUksc4uFwaRmUcpe9tmVmTS4Flh9qgyy/AbMbhNjpeVrlXT2OAPyuGVuQHlmkmUB33ufRIMvuzZco+ZqBlZL2iL5669jMAmQZnZjdvp8sZa5DQYdpuVXA79/J3D5dxvfEDG/xiczaZ/5YFJ2QE7IJjOTekZqXx1o+pDZyXZ+rQxg9j5c/r16NOBWSkrdnn+g9P6b/1wGkZf8U+3v26VbZMn1Oz4Xfinvib1y/PdOaHTg6NZPyEDwxdeEe76+ZTL4nRgzVx5YLGFzBZN05pMu6czk/Bt1A+FXnBxYBRx1BvCi99Snl16r6Op3skT6mpWZ5NOIf3IfiqtmlpSqRQhMZvtLM5O8K6EpVTmzo5JCDcEkADj6AumjddsnpYQFkHFR2Wpu9mt0903SPW6iZiYB0jNmYo/0j3HTE1e9NHk1vaPmM5OWnyjHOd3wXqspmOTTgHthToJMcUq+vZlJemxrosytJDMpajAp57//LizI665U5laYKm0yrMve/FZ5a5fMpK5++V9P7pfG2z2j0p8ziL44kR2of+A+jNWnArvvdf6n+x6T15AdkPMRANzwJwAsWZ2zlYR9f6VSkhlXKTOpo9PpSdrZLeOyYpnbNADlBI7DblvfstL+Xdm8T2aSPXYf3hzcM4llbrEwmNRMPSNOmdu+JySavvIU+bpa36QoZW6A89hYDbgDgkmPXg88/H3JSNIlHnG9+T+B8/4/M8+1GOmsgnr0TALkoHvkeYnsh52oLVaDa5vTByp/VHDPJJOUkoGNntSYKC3tHZGypaDljSvxlq3ElUoBo8eUZwoBrgbcBjOTADkOPv+Ac4X4iR9Lz4oXvjt4pbSwTrtSekA9ekO4x0/s9T9X6Al4YRo4+6Phr8b32kGo/dvN/d0G1shE1L2/T4+XH9+WrA8IJkXInnz7D4Bz/1ft29oOdLo/0Pwyt5KeSQec/U4HhObnZB8Muy919ZcGosaflf6FbulsjZlJMSYUF/ylZObcZDfjnp+VptTenklAeTBMpWrLINt4rvx/vY24d94ugdf+kM/Zt6z2TFLN24A71SGB7ydvK32cLtGNUq6dSgFQnsyk2XglbkB5ZpI+jnTGHOuUlLmFLOF0ywQEQ3W2U9B4RGe3ubND3MEkb8VDu2QmKSX7y/P3y3zDr/G2m7440eysJG3VabLvPnu3fL1vuxPwGjxK3qt7H5IL6CNHN28748qvCW7AXZhxspK0nuHSMrd0V/Tg8Bu+UrpoSC7vBJH0rc50H9ooiRp+F+L0+YJlbjVhMKmZekdlWUTLkh18yXpg5Fh5w1Xqm1SYkcFUd5Rgkj2wM52ZZFnAj/+vrAD2ot+r/bm9Bo8yn4GxmDQiMwmQ7KS5KQCKEftGy6+RciJ9kltYkDI30z2TADmpW/ay80Yyk+ysj1r6JpkucwNkMLfj9vIVberRgBsAlh4vx+iDT0um1bevloHMOQGrPkVx9IVyBfqOfw33+CN7ykvcALlal84Bo1ukh1NYetIbJQBQTUdaglTjrlIMv55MQxslmOSdFM2FLHNLktygHCuaddzu7JGJqjczSfc50aVqUUtNu/qdn50+JJN1b/PqdLbGnkk1ZiYBsg1bLgGesRtxFwMT7tXcdN9Jd2bSLtn3a+kR15EGXvA64LGbnJIOQFZyC9MvSesdBY7sjv773bwNuAEpdTv4jJOtBcg+0NUf/QJNR6a8zC1O822gPIutmJkUs8wtlXECP7X0TMp0O/uPW3HCW6FnElBa0qaDSd6MJaB9MpMAuYj22I0yTtl6eeXH6pLoZq/kpukMQZ0xOPZYaV+ko14st1taLCspqoE1Fcrcpsr3te4lpQ24a9kXl59QmpmadZW5TbvK3ABp3zE/4589VWBmUhwMJjVT76hkJh3eLX2IhjbI4GD5idI3KYguH4mUmVSnYNKj10ujxzM/YK5hLlVX7JlUx8wkQPomFaZkwGVyBTGqTjc01IPBmUMykDKdmQSUlvyYasANOJmXUUwfMh8kPfZiGTg/frPnd9krQaU6/H+uVkvtDM3nHwB++GHg8LPAaz4T/4o3IMfZU94ObL8luJmkW1BJtFLAaz4NvPYL0V6/u2zK5PFnwLMS0fS49HJxG9oo+4e3AXkxMymh2ZN+coPN65cE2BmPI07PpLlpGefoYNJMjcGkrCszSU+avYtYZGrMTKq1zE1zZ/TO+uyTxcwkVzDp0M54K8JuuUQCF7rUbfw5mdCF6Zek9S41k5mkPFMK3aPsCVd20uRYbRcKU+nyBtwmxpzprE9mUtyeSRkJdhVmZTsj90zKla4wp+nXH5Q9oVeBdQflS7KUPBP9dslMApx9Zv3LnRWHg+SPkvddmCXtG6F3RI57O2+XgN7kWOkKaRvOkffOcS2yilut8qtlPutdiACwF8jw7Gvdw858dn7GTFZQbtC/ATcgvZUA/+zmefZMioPBpGbqsVc80g3BdDR95SkSoPE2VdQmawkmGShz0ydYfcK1LODHfyEHyRPeUPvzUnR68FmvzKTBtXJgH3uk+pLbVB/6qpq+iqIn0PXI2NOZK+mcmQF6cSWnGvomTR803+fgqBfLwOWh75XeH3ZJ8qiWbgGggJ//PXDPvwMv/aNomQLVnHKFXP2+47PVHzuxRwazfo67VEoAo3AHKExmdPWvkJIlzTczyae3EsDMJD/HXQac8PrmbkP3kJOZpFdyG7Qngjq7SGfphA1id/U5waTiSmg+ZW5heiZ5M9zmq2R+VFOS0esTTPLtmbSrPLMqipXb5D354Hfk66j9kgAJJs0eKc/cDGvBzmr1BqVHjpasK3ep2+S+2i6IpNKl5SmVegdF4e5PNGuqzM3OTIq6b2vpnH/8otA7AAAgAElEQVRmUjFzLqDEq3tY/iYlAaRdTgDX24S7rTKT7HFPpcbbWkcaeMcNwBl/UNdNimTVabKa6z575e4hV6DrBa8Dfv+u6OfiVjOwGoDltExwm5sqL3PrHnJlJs2Y2Rd1A27LkmBSZ5/zftHBRd0P1q24mlsTWlosAgwmNVPvqGQa6JO/Hiiv3Co79p6H/H9OT9CaVuZmnyAfvQF47m7gpR+It6IGRacH5PXqh5HqkCsnex+VK8qcpDWeN5hUPE7UYdlYHWw20S8JcI43Ua92zxfk+GK6zC3VARzzKjlmuTMWZg7XJ7uvs0cuDuz4NbDsBODMD5l9/t5RuYp597+X9l/xWliwM5MMNpvOLXF6lZgMxPWvkom1nuD79UzS+/7+7aX3MzOp3IuvAV7S5MlUz4hz8UtPGoqZSfZqbrWUuemf0cEk3zK3KplJ04eAvzkGePi/nfv0BbxaJzXujF6dwe3XgFtvv2X593yKIpWSzMvHfyS/c+cdMiFafkL459DjiFqbcOsSaW+Zm1KSnfTkT5z3daxgkjszqRC/ZxJQmpmkx7YmytzmZ2sPJgU14K7W0yuVkqC8DhpZlmS+rX6RfO1twj3fRplJS9bL++ToV4V7/PITWqP5trb6NCkl3X6rfO0ex6VS1bOt2oF3zOpWmCnPTOoZdi6SFmbMBHKyeTlOzE1KMMm9D3QPyff9MrqLZW5t8n5oMQwmNZMuBXnmV/Im0imqK7fKbVDfpFrK3Na9TJo1xmrAbQ+KHv5vueL+o49LOumJb6z9Oak2S7cA77pF0mPrZWSznZk0yWBSM/StkMG5PjE/8C05Rugm/SbpY5GJEjfAlZkUMZiklwuvR8bdsa+W/iru1YXqlZkEyGC2o1PK2+rRwP20q2T77/1a8GOmDkj5SdiVP8NIpZzJp+nMpLlJJ0V95lB5oC+/RiaVzExqDz2uMgY9aehfLlmvZT2TQu5LXX3ys5blNK/WTeG1MD2Ttt8ikzv3VepqPWmqKcnorVDmpoMMUwfkcXEykwA5thWmpHfSzjukVUKUgJg+/tda6mbZGUMpnynFujMlO01fHJ3cH+1CqJZKlzfgNpFF6w7cGGvAbZe5FQOKtfRM8gsm6TK3CvvnwGonaDQxJgGjFSfLz3gzRkxlgzTCy/9ElqlvxmIoJujM5Hu+KuM6HVRfTHRrBr8m3Lpdhlv3EhmTFWbk/WwikKODR1MH5cMdTFJKspP8ytyKwaQ23b+ajMGkZtJXi5/5tRxYdIrw4Dqp+wzqm1RLmduGlwNv+Ua83iBdfZKa+diNwE1/Bux5ADjrI8xKapZVp/gP3kwZPho48LRM7rzpqVR/HWlnqdWpA3Llecsl9eldpY9FpjKTOrtl4nQkYgPuYsPEOlxRXHemTFgf+q5z38zh+q2IeO7Hgbf9oHTZWpNWbQOWnwTc/tnych1NB/OCytxq1WdP3k0G4tzNYy3L/3/TkZELGIHBJGYmtRQdTLIsJzMpt6S071ExmBTy2JPtlyvPhZng5tVheiY9ZvdPm3WVE8VpwA2UZvTO+vTf6fL0TCqW6cUMJq05Xa66P/BN4NnfRitxA5xMoan9lR8XRJefeTOTAGf1yps/JqvOTdTYM8nbgNtUmVtJZpIOAMbNTLKzqPT/OepxMmg1t2Kws8KYu3+ls1/pcrf8Grnfm5lUmG6fTIx0Z2tlGkW19HgZR+/fLgsMLcagRf9KCe77NeGe89nX9HFncp+9LxrKTAJkzDx1wGm+rQ1t8s9MKpY4t0lwtcUwmNRM+mrQzCFgyQbnfqUkoOTuH+E2MSYnK9OlINWkOoDfvxP4s/3AR3YBf/wUcNKbGrsN1DgjmwFYwO77eMW/WfJHyYn54R/I4PT4y+rze3RgutNgcKBnJPpqbjqYVI/MpHQXsPl8+VvO25OS6TpmJuXXAKsjTuqiUEqyk/Y+BDz1M//H6L+/yTI3oD7BJF3qM75LMjesBf8SxKGNTt8JrZgFwuNUS+kelivSsxNOoKJ7qHRFtlrK3AAJRgX1G3IHCPwsLACP3ySfz/kEk+JMaooZvXplMHdmkv0adWBNlyMNxChzAySYdsyFwIPfldcdNZiktzFMnyk/OjPJ24AbkOPgCW8Anv4F8J2rZX/oW1b+uGpSHc5xG5AsHRNlbpmcTHQB539mpAG3u2dS1MykoDI3HeysMOEdWCVzh4V5J3g0sEo+ynomtVFmUrvryDhVJ+5+SYtJR0b6twVlJnnPzzpDcWLMLnMz1DMJkLHk1IHyefLwRslI1eef4vaxAXccDCY1k7v0YGhD6fe6+oJ7YUyOyYCsWatrpTrkCps34kuLy7DdTHRib3mtMzXGwGrJTHrgmxJYWrG1Pr+nWOYWcxDtfc6oZW56Sdd6BcqPfbVMau/+CvCfb5NJX6ssH1yL4y+T4/Dt1/p/X5etmCxzA5zJoMl+U3pRgfFdzkDPL6g4tBHY/4TT9BdgmVurKpa77nXK3HKDAZlJEYNJ0+MyOfZbCS2ddQIEfnbf6/QHck/aCwauTuuMXv163SVTHWnJIrr/GzLZ12VHcTOTAODYSwDYGYqRg0n2+6bWYJLOTArKfL/sWuDDO4Br7gTe8BVg6xXRf0dZmducmax4d0mksQbcnfEacGe6ZZvKGsRXacANSHDVmpdVtcqCSa7MJMtqr8ykxUC/L+vR97JV5NcE90yqlJk0P2s4M+mgvZiLZ56q+9p5s5sL03KMMb2yb0IwmNRM2QFn0KJXctMqBZOClnomMmloo3OlkWVuzZFfI1cZn/ixNFyuVwDZdJkbIMeoyGVuds+keqWzbzxH9uXvvVeacZ/5ISnVbVeZHLD1culj5y1hAJzMpN56ZSYZDCb1LZMymUO7nECD3/MPrZdsksPPOffNGSpPIbN0xuPkPvno6pcJgzszqZbV3ADJ6B7f5d+8ulpmks5K6ur3z0yKUz5VzOi9V7727pOnXQUceEraBRzaJRMYE+/PdWcCXQNS9hc10yl2ZlJAA263VEr6lRx7UW1B6FSmtAH3gqFgkjszSfc4MlHmNj9Xe88kPen27sNhli93Z3ge2in/29ygBCwPP+dkd3Ep9MbTwaThRRxMGlgNHPIJJvkt5OM+P/gFm2rhzUzyK3MDyoNJ87MscYuBwaRmUsoZRJRlJvVXDibVshoGURSZrGTDALzi3yz5NQAsuSJ73Gvq93u6l0jg0FQDbkD2nbFHgds+6aQQV1PPMjdABvUv/wjwwncD770bOPujZgNozbDtnTKZu/ML5d+b2FufkuiVp0iwsJZylSCpDkmRH3/WlZkUUOYGlA4G56ZkIsv+fa1FTxZ0ZpLuleO+WDZzWCbvYa8I633i4I7g5tWZKsGkx26SpsR9y82XuemM3md/K7feLJdjL5bFFX79GZnw960wczU83SnHtjPeG/2iQzEzyWc5+jB0MKmeV/VTaScDCjBX5ubOTJqbkEBi3MbeuswtatadFhTcCxPs1IHEQzvkY2CV7A8DqyRj6chu+b5+fzAzqXHWnwWc8EZg0/nN3pL6ya92yizdClPlgcuSnkmGVnPTwaPxZ+X94r0wuWSdjHW9fZNY8hkLg0nNpjOMlviVuY2XPx6QMrcozbeJajViD4wZTGoOXYK1ZL2s0FMvqQ5Z7XH1C80955kfkADYrZ8APvNSYGfA6pRu9S5zA4Az3ge88i+BvqX1+x2NNHgUcPQrgd98sbzp8JE90pfAdKP+DWcDH3zcfL+p/hVS+qMz1LoCytwAaWSqzU3JBKxZpd/kr1jmNiYThpwdTMoOuMrcIvYt09lqex+WW78SsUqZSZP7ZcWzTeeV96YxUeamM3qff0C+9jaF78gAp74TeOJW6SPkV6ZXqxe9Bzj96ug/pwMKccvc/HommdJRpzI39z4wO2mmib/OoorTMwkoD+4VwgSTXAsZHNrpBJeKQSY7gzXuyoUUXVcvcNln4q/e2MoGVsv71J05DNgNuD3ziNwgAGWXuRkK5nT2yXFo/5Ou3+GS7rIX8fAGk6YZTIqBwaRm6x2VE7l3QNTVKytB+K3SMzFW29KqRFHp+mIGk5pDLx9bzxI37c1fB7a+1dzzdS8BXvs54M3fkGPZf7zJWdo4yPQhGdxyf4vmtCvlIsOD3ym9f2Kv+ZXc6mlgpVxRnKmQmdS3Qgal7ibcc5PcZ1pRtzszaZ9zJbqkAfeRaNmBOvCkl5r3DSZ1BfdM2n6LZNJsOk8m+SWruYVYer0andGrl7r2y9Y55W0SsDq0ozUmlqlUaYZOVJUacJuiS8e0+VnzwaS5STPZuXq7pg8BUNEDVMVgUkCZW6X9Mzsg769DO+0G9UHBJGYmUR3kV8uttwl3Ybq892qqQ4I9E2MSKDVRZpZKyXvgQEAwCZByW29m0vwsA6sxMJjUbJsvAE58Y/mV464+OUF7rxQVZmSgzZ5J1Ag6M4k9k5ojvxp4/b8BZ/xBs7ekdpvOBS78G2l4++j1lR87fVAGAswwiWbdWZIR4W3EPbG3vc4V/SvlinoxM8knmJRKSaaeu8xtlsGkltTZLSVsk/uk8b0uc8v2A7OHJaNl5nC0zCRdAquDSb6rueXkSrffxbjHbpIMqRUn24EEdzBpxm7CGnNoXMzoDQgi9AwDL3itfG6i+bYJQSuIhVGtAbcJ3gbcCwUzkz93FtvsRPzm24ATTJo6IMEpU2WHYXt69a+URQqOPC+ZIvo+oDwzicEkMkm3xnA34V6Yl0w9v3lEz7DZzCRAMtuDMpMA6Zu0b3vpIh6mejYlFINJzbbt7cDFf19+f7HJpKdvkl5Stoc9k6gBdP8HrubWPFtebXbVrGbY+ArJKPnNFys/bupgfUvcFqtUCjj1Sinf2XWXc/+RvU5z9XbQv1KyIw4+LV8H7fdDG3wykwxMAsm8niFXzyRXZhIg45uowST92P3bpU9Wr0+5qp6UeEvdFhak+fbGV0jgw6/MzcTVcZ3RW6m86bSr5Na7+EqzZLrj90yq1IA7rlSmvMwtFbO3ESD7QGFaAo+zE+bK3AAJJtXSky9odb2wPb0GVgE7b3c+B+RY2jUg2UqAKzOJpT1kUDEDzhVM0vux377WPeRqwG1oX8zlnd5gfuPJ4Y0yzhh3LVpSmDGzmlxCMZjUqtyDLbeJMbllmRs1wshmGbD5RfeJwupIAye/BXj8R/7LxmrTh+rXfHuxO+lNkgVyx2fla8tqzzI3QLJOVEfwxG5oo6Sx65WJ5qaYmdSqekakdHH2SGlmEiBZ1rNHpM9FWB0ZucK9UJAG2n7ZMHpf8AaTnv2tTFw2nWc/rkeaLmum+vBUy0wCgBUnAVfeCpz4pvi/z4Q4mUlWIzKTOjzBJENlbu6V0+Ymo/c38lPMTDpY2/PpDA5v2WHYzKSBlU52p3tlv4GVTmbSLruHocmeXUSZnBzz3WVu+jjsd452B5NMlZm5A0hBmUlAaanbvKELCQnFYFKrKmYmeZpw66We26l0gdpXdgB4503S44EoDt2P6bdfDn7M9MHy1TconOyAlEzf9w1gYp9M0gtT7XWu0KUYex6WgENQecjQBplY6gwm3YCbWk/3MLD3EflcN+DWF8umx6M34Aacxwf1G9JXuL09Zx6/CYACNp4jX3sDKKZKLXRGb7WSqZVbWyfrNx2nzK0BmUkdmfqUubmzgIxlJtkZU7rMLc42uYVpwA0A/a4AkruMcmCVBJMWFoBf/KMs6rHylOjbR1TJwGrpB6dV6s/VPSRJEibL3HJVgkk6c9RdKl+YZZlbDAwmtaqqZW7MTKIGWbnV/KpNlDz5NTKJ++2XnYwSr+lDLHOL47QrZVD22y/JSm5A+5W5AZIi79cvSSuu6PaE3LIBd+vqGQEm7H1Rl7m5M5Oilrm5fz6o31A6IDPpsRuBVducDKlMt6fMzVAT1hG9cIWBLJdGMZGZVM9ed34NuE2UudU9M6mWYJId0Kq1Z9JAQDCp385MevSHsprVi9/L/oRkXn5NaQb6XJXMJJ0kYSyYZAeQUmn/93PvqIwv3JlJhWmWucXAYFKrCgomFcvc2DOJiNrM1iukZ8PjN/t/f+ogy9ziGD0WWPtS4I7PS/NVoL3K3HpHnQlipT5hOpikryyyzK11ufs76iBOl/0enx6Pvpob4IyPgkp0/HomHdkr/cR0iRsgmUOzE06jblMr+mQHgN5l7bVPtkUD7nnn6/m5OmQmmQom2dtVc88kO8BVtprbTOnzB9EZez2jpZlvA6ukEf5tn5QJ/5ZLo28bUTX51RK01MfVQoWeST3DAOzHmSoz0xckc4P+wVKlZAyxj2VupjCY1KoCM5PGJJWYV++JqN0c/UoZ4N51Xfn3LEsyk1jmFs9pV0pmjy4nbKcyt1SH9MEBnICDn+4h+X4xmMQG3C3Lvf95M5Mm9sgqP5HL3Oyfd2dguPn1TNr+IwCWrC5Z8jjLWdnKZKnFyz7klPa2g5ZvwJ2WfUVbKJjvmTR7xGyZ2+zh2oJTgZlJ9uuv9rr1+8L7/tBfP3c3cPo10suQyLSBNfJ+0hlHOijqt5qbOzHCdJlbpV6vw5uAMW+ZG4NJtWIwqVUFNuDeK5HcuEvXEhE1WkcGOPnNwKM3AOPPlX5v9oiUSzAzKZ6jL5Ryhnu+Kl+3U5kb4JRlVNoPlCpd0Y2ZSa3LL5ikg0fjz9pfR1ytspiZVKVnkg4SAcBjN8m2LDvRuc87aTfVgBsATn0nsOUSM8/VCC3fgDtd3oDb1GpugEx45yar97kKw70PmeyZpDPnqpWm6fdFUDApNygLYhDVQ3613Oom3Dozya8/nHsxKdMNuCsFk4Y2yWpus/YCDIVpBpNiYESiVQU24N7HldyIqH1tvVwmH3d7GnFPHZRbZl3G05EGtr3dyRZot/56unSpUpkbYKepu4NJzExqSe7xircBt16mPOqEWwcaq/VM0pPxhXkprd14bumFOG8wqZDgUgcTZW6NbMBtqsxNZybNHpZgjYk+V6mYwaSgnl+F2XD7Z7oLWLkNWP3C0vsH18rtqe8yU85H5Ce/Rm4P2X2TdFDfNzNpifO5qQbYOjOp0lhy2FMqb6rEOaEYTGpV6S7Zsf3K3HrYL4mI2tSS9cC6M4G7vuSsAgQ4SxmzzC2+rW+T80du0FymRaPofh/VslWGNsiKMXNTbMDdynQws7PPaXCayUlWySE7mFRzmVvIzKSdd8pKke4SN8AVTLKDKPMJLnXI5MqXop+bArbfUv1ni2VudZxSpDqchRsWFuSChIljmz5uTO6XW5MNuIHaeiZ1pCUg5deAO+xrvvJHwIuvKb1vYBXw9uuBMz8UfZuIwhrQmUl2MGmuWs8km6kG2GEzkwCnCXdhhqu5xcBgUivr6gsoc2ujHhhERF5br5CBxhO3OvdN68wklrnF1jsCnPxWYNkJzd6S6PSy1mEyk2DJsvPWPINJrUpPFrpdA3ulJCBULHOLGEwa2gD0Lg0u4Sz2TLInMY/fJIGODS/3f1yxzC3CZH2x8ctMuv+/gH97TXlJslexAXc9g0muzKSFkL2DwtATSL1SsvEytxqDU96VBgEzPb2OOp2rVlF9ZftlHFcsc6uymptmKis0TM+koQ0AlJOZVDDYLy+B2H2tlfkGk1jmRkRt7tiLpeTlruuAjefIfSxzM+tVf92eyz7rMrcwmUkA8Pz9cssyt9akxyveFWiz/U6ZW9TsjW3vlJ4vQcELb2bSYzdKyY93cqEDB7OuMrekjq90A27Lco4bOltncgzoXx78s41uwK0bUadMZibZwaRWKHMD/IN7Jnt6EdXTwBrJHAZcmUk+mT+dPVL+Vpgy2IDbPs5XynLP5CSDSmcmzc+wzC0GZia1sq4+aUqrFWaBmUPt1wODiMgt3QWc9LvAwz+QJbsBlrmZlkq1ZzBJly5Vy0xaYgeTdt8nt8xMak2ZrJS45ZaU3t/V7/SEjNqAO5Wq/P9290w6vBt47p7yEjfApwH3bHKzNvxWwNPjTx3oD9KwBtz275mfldu2yEyqNZiULQ8mFTjhpTaRX+3KTLKD+kHHbH2hwVQwqWdU+jZVy8we3gTse0yOKwsFlrnFwGBSK+v0ZCbpk533Ch8RUbvZerlcab7n3+VrlrkRAIxuAY59NbD2pZUfl+2XQWMxmMTMpJY1vNHJJNPc7/OoZW7VuDOTHr9ZPt90XvnjvKtmJbkJq7dpOeCsdDRdJZjUkAbcrtXc9K3Rnkk6M8nAccS9ylwtPZP0dvj2TGIpDrWB/BppZWBZTrlxULBGN+E2tW9nssAf3Acc86rKjxveJIt46AB6Ui8kGMAyt1bW1Qcc2e18PWFfwWfPJCJqdyNHA2tOB35zHfDi99qZSQroYjAp0TI54A3/Fu6xQxuB3XaZm4mMAqqPy79THqRxZyPVmr0RxN0z6ZlfAn3LgaXH+zzOLmkqruaW4Mm633L0+mJm2MykujbgTjvlbTozyUSZW1lmkoF90b2v19wzKVe+mhvL3KhdDKyWFRKnDwJzOlgTEEzS1TaNDuYMbZTsywNPy9dJPfYbwMykVubtmTQ5JrcscyOixWDrFcD+7cBTP5MJS1d/fZu40uIytEFKvwGWubWy7ED5/6dYxqjML1OuJwUzR4DttwIbX+Ff8lnWgHsmuVenvSvbAU6ZW9XMJLtnUj3L3NI5yWQtzDpBJRNZZHofmKhXmVuNWXdpv55JbBJMbSKvV3TbIUH9VFqyC/0Uy9waXGY2bK/opvsu8r1VM47aW5k3mKRPdkltEElEi8txl8pE867rZMKSY1YSReAunWKZW3vRpW1d/eZ7e6VSElB66qcSbPQrcQNY5ubmDawBTplb1cykBjTg1r3UxneaLXNLdcj/XF+sNV3mFiczqazMbS65+ye1lwE7mHRoh71SWoWLPXpO2+h9e4jBJFMYTGplzEwiosUskwNOeAPw4HeBA09xJTeKZmij8zkzk9qLLnOrtadMNekssOPXMrFff5b/Y/REXwdNCgwmlZa56cykQ5V/ttiAu45TivxRcnvgaVeZm6FOHemcq8zNQJacO8hVc8+knFMepBVmWOZG7UG/Xw/ukGNKpkLWkekG3GH1r5BSZ10qzzK3mjGY1Mq6+qVmumCfOCf2ypUfTriIaLHYeoWk7+/4NZtvUzQlwSRmJrUVXeZmuvm2picva04PXhmwo1P6/BQzkxK8WpZ3ZTtAep4ArdGAe3Ct3B54ymyZGyD7ig5QGclMcpe5mcxMSnBPL2ov3UvkvXTwGZnHViphW36C9AL2rvhZb0pJdvPzD8jXzEyqGYNJrUwPsnTd+sSYRHDZU4SIFotlxwMrT5HPcwyUUwSD6wDYJVLMTGovXXUOJumJwaZzgx+jlL1q1pQERKyF5E4odPDN3fQ5dJlbAxpw96+QIM3Bp13BJENZOu6JrpHMJHcD7jiZSd6eSWzATW1CKSl1O/SM7MeVgkmbzwc++HhzFtEY3uQsdJXUY78BsY78Sqm/Uko9rJS6Vyn1LaUUZwIm6UHWzLjcTu5jiRsRLT6nvE1umXVJUWSyTm8GZia1F50tZHolN01PXjZWCCYBdjBpQkqIAGYmubNhZsI24NZlbnXMTEp1AAOrpMxtwXAwSQei01kzr0E3Gu7oqn0bM90+q7mxATe1kfxquwH3TOUyt2Ya3ux8ntRjvwFxLyPcBOB4y7JOAPAogI/E3yQqKgaT7FTjib0MJhHR4nPcZZLirOvsicLSTbiZmdReuuyS1rplJmWB/lXA6LGVH6czQHSZU1InFH49k3RWfCs04AaAwaNKM5NShjOTTAWk9XbF6QeWzrIBN7W3/Bq7AfdU5QbczeQulW/0anKLSKzudZZl3ej68lcAXhtvc6hEWTBpDFh+YvO2h4ioHrp6gd//Tf2yFGjxGtoAPHFr6w5WyV+xZ1JAP6O4Tn2nTA6qrRSX6ZZJuw4mpRM6WS9mJtnBpIUFJ5hULTNJB5PqmZkESN+kh75vvsxNB9JMlLgBznbFeb5Mt6xa5y5tKyS4pxe1n4HVUlEzua91s86HNzmfJ/XYb4ChpRAAAO8A8LWgbyqlrgJwFQCsWbPG4K9dxPQgSweTJseYmUREi1N3g5sv0uJwytslo429BNtLvVdz06Wz1WRywOykq8wtoWVExcykydLbVFpWc7Os4MDcQgN6JgHyPp8cc4JbpnsmGctMsqdWnTGy7tyZYvp1zid4tUFqP3l7rr9vO7D2Jc3dliDMTDKi6pFfKXWzUup+n49LXI/5KIACgK8EPY9lWddalrXNsqxtIyMjZrZ+sXNnJhVm5YTew78dERERAGngfsZ7m70VFFW9V3MLq7OHZW6Ak9mnl6PXWUn9K+Rv420G7daIBtyAlLkBwL7H5dZUmVsxM8lQMEkp2bZYmUn2xNb9d5+fZQNuah+6n+HcZOsGajp7gP6V8nlSLyQYUDUzybKsV1T6vlLqCgAXATjHsizL1IYRShtwT+6Tz7uHmrc9RERERHHlBoGMayDfLJkccGQPy9xSKZlM6Ywk3Xx7YLUs7z19MDjY0ogG3ACQXyu3OphkKvCnJ7omy6w7MvGy7vwaos/PsgE3tY/8aufzVu5pOLQRGN+V3GO/AXFXc7sAwB8DeLVlWZPVHk8RFYNJRyS1F2CZGxEREbW3TA645g7g5Lc0fzvmWOYGoHQ5ep2ZNLBKbis14S5mJjWgATcAjD0mtx2GOnXoia7JFSFTmXjBKW9D9IUF6aGU1Mw5aj+9y5zswVYOguq+Sa2aPdUG4uak/iOAPgA3KaXuVkp92sA2kdbZA0BJmduEHUzqZjCJiIiI2tzAyuaX7WR0mZtu6pzgybpuRg64ytzszLFKTbgb1YC7e0gCNPu227/PcM8kU2VugAS64gSTdNlhwQ4mJb0Mk9pPKuUEo1t5gYzVLwKyA80vuW5jcVdz21j9UVQzpWTnnjnslLmxZxIRERFRfDozad7OTEpyqYM7M6lY5mYHkyCe1eYAABZ7SURBVCplJi3YwaR6ZyYpJU249zwgX5sKrBQzkwyt5gYA294JrNxa+897M5MYTKJ2lF8NHHjS6QHWil7wWmDLq1s7e6rFmVzNjepBB5Mm9srXLHMjIiIiiq9sNbcET9Z9y9zsvifTh4J/rljmFrDam0mD7mBSC2cmnf3ReD9f7JnkCSZxwkvtZMBe0a2VM5OU4vsqJq6l2+q6+qQB98SYXPXJ5pu9RURERETtr7NHSokYTJJgUsETTApT5taoBtyAZCZpKVM9k3QwyWBmUlze1dyKmUlczY3aSF4HkxisWcwYTGp1xTK3MakXT/FfRkRERBSbLieaGZfbJE96/Mrc+lfIbSs04AacJtyAwdXc6lDmFldQZlKSg53UfvSKbq28mhvFxshEqyuWuY2xxI2IiIjIFD1p18GSJE/WSxpwT8htVz/QNdAaDbgBYHCt87mpLJ1MHcrc4ir2TLL/HwUGk6gN6TJZrpS2qLFnUqvr6gMO7ZJ03u6hZm8NERER0eKgg0nTDCaV9kw6LBk7HWkgN1ClAXcDM5N0mZtKmQteFTOTWjCYVJiWW2YmUTsaOVoCSbrcjRYlBpNanc5MsuaBZS9o9tYQERERLQ560q6DJUkuc0t7ytx0D6FsyMwk1YBiBz0pTRnsHaT3gVbqmZT2ZCYVVxtM8P5J7ad3FPjg9tZ6b5FxDCa1uq5+CSbNTQI9I83eGiIiIqLFgZlJjpLMpAmgq1c+z+Yrr+bWyAbcXb1A97CTsWNCSwaTugAoV8+kObllA25qN/o4QosWeya1uq4+STeePignUCIiIiKKr5M9k4pKgklHgM4++TyXD9GAW8kS240weJTZoEorBpOUsntYsQE3EbU2BpNaXVef83kPeyYRERERGeHNTEpyGZFuwG1ZkhFfLHPLVy9za0RWkja41mxQZc3pwDl/JretJJN1gknFBtwJ3j+JqCWxzK3VlQSTWOZGREREZIS3Z1IqwcPiTA6ABRRmpMyte4ncXy0zaWG+Mc23tTP+ADj2YnPPl+4CXvpH5p7PFN/MJJa5EVFrSfBZs024g0kscyMiIiIyw52Z1NHVuFKtVuRejn72iNPsOpsHClMSZPLL3LLmG9N8W1t+gnwsdpmc/N0BNuAmopbFMrdW1+nOTGIwiYiIiMiIjKtnUtIn6u7l6GeOAJ26AfeA3AZlJy00uMwtKdJZnwbc7JlERK2FwaRWx8wkIiIiIvOKAZQplhDpwNrcVOlqbrlBuQ1a0c1qcJlbUugeVoBkhQEMJhFRy2EwqdXpYJJKOSd0IiIiIorHvYJX0psb68Da7ISsIlzMTMrLbVATbmsBSHE6YVwmB8xNy+dczY2IWhSP/q1OB5O6h3iyJiIiIjKlI+M03U58ZpJuRr5fAkQ60Jazg0mBZW7MTKqLTM6nzC3h+ygRtRxGJ1pdMZjEEjciIiIio3R5V9J7JqXtYNLEmNzq8WfVzKQGN+BOikzOKXNjA24ialE8+rc6fTJn820iIiIis3QwiWVucjuxV251mVuYzCQ24DYvk5Nm6ADL3IioZTGY1Oo6MnK1iMEkIiIiIrN0ECXpJUQ6qHZkj9x2eVZzq9QziWVu5qVdmUmFWQDKKckkImoRDCa1g5HNwNLjmr0VRERERIsLy9xEWWaS3TOpIwNkeiqs5sYG3HXh7plUmJKsJKWau01ERB4McbeDq25r9hYQERERLT6duswt4SVEOqhWDCb1Od/L5dmAu9Ey3VLm9oULgad/DvQubfYWERGVYTCpHfBKBBEREZF5xTK3pAeTsnKrg0m6zA2QJtxswN1YfcvkdmIPcNZHgBPf2NztISLywWASERERESUTy9xEOqDMDaiemcQG3Oad/FZg4znAwGpeVCailsVLCURERESUTMxMEh1p+Rsc8azmBoTITGIwybiONJBfw0ASEbU0BpOIiIiIKJkyutF0woNJgL0cvd30uSSYNBCcmWRZzEwiIkooBpOIiIiIKJl0ZlKawaRiyV9HZ+nfI5cPXs1tgT2TiIiSikd/IiIiIkqmYplbwnsmAUDabsLtzkoCpMxt9jAwXyj/GTbgJiJKLB79iYiIiCiZOlnmVqQzk7o8waRcXm79spPYgJuIKLEYTCIiIiKiZGKZm0P/LfwykwD/JtxswE1ElFgMJhERERFRMrn7BCVdUDBJZyb5NeG2FpiZRESUUAwmEREREVEyMZjkCCpzyw7I7fSB8p9ZWGBmEhFRQjGYRERERETJVCxzYwNuZHQD7p7S+7MVeiZZ84BS9d0uIiJqSQwmEREREVEyMTPJof8WnX2l91cqc2MDbiKixGIwiYiIiIiSqZPBpCKdpVVW5sYG3EREVI7BJCIiIiJKJp2NwzI3V2aSJ5iUyQLpLBtwExFRCQaTiIiIiCiZdOAknW3udrSC4mpuPeXfy+b9M5MWmJlERJRUDCYRERERUTKNHA1c+i/A5guavSXNpwNqXX3l38sOBGcmKU4niIiSKN3sDSAiIiIiagqlgJN+t9lb0RqCytwAacLtt5rbwjyQYjCJiCiJePQnIiIiIkq6Wsrc2ICbiCixGEwiIiIiIko6nZnkXc0NkMwkNuAmIiIXBpOIiIiIiJKumJnk1zOJDbiJiKgUg0lEREREREm36lTgmIuApVvKv5fLA9PjwMJC6f3WPBtwExEllJGjv1LqA0opSyk1bOL5iIiIiIiogfqXA2/8SvBqbrCAGU8T7gWWuRERJVXsYJJSajWAcwE8E39ziIiIiIiopWTzcutd0Y2ZSUREiWXi6P8pAB8CYBl4LiIiIiIiaiU5O5jkbcLNBtxERIkVK5iklHo1gF2WZd0T4rFXKaXuVErduXfv3ji/loiIiIiIGqWYmeQJJrEBNxFRYqWrPUApdTOAZT7f+iiA/wngvDC/yLKsawFcCwDbtm1jFhMRERERUTsIzEximRsRUVJVDSZZlvUKv/uVUi8AsA7APUopAFgF4C6l1GmWZe02upVERERERNQcgZlJLHMjIkqqqsGkIJZl3QdgVH+tlHoKwDbLssYMbBcREREREbWC7IDc+mYmMZhERJREzEslIiIiIqJgnT1AKu2zmtsCkOJ0gogoiWrOTPKyLGutqeciIiIiIqIWoZSUurEBNxER2XgpgYiIiIiIKsvl2YCbiIiKePQnIiIiIqLKgjKT2ICbiCiRGEwiIiIiIqLKvJlJlgXAYpkbEVFCMZhERERERESVZQdKM5OsBbllZhIRUSIxmERERERERJVl86WruS3Myy17JhERJRKP/kREREREVJkuc7Ms+dpiMImIKMl49CciIiIiosqyeQkgzR6Rr3VmEsvciIgSicEkIiIiIiKqLJeXW92Eu5iZxGASEVESMZhERERERESVZe1gkm7CzQbcRESJxmASERERERFVlh2QW52ZtGAHk5iZRESUSAwmERERERFRZbrMTa/oVixzU83ZHiIiaioGk4iIiIiIqDJvmRsbcBMRJRqDSUREREREVBkbcBMRkQuDSUREREREVFlnH6BSbMBNREQAGEwiIiIiIqJqUilpwj3lKXNjZhIRUSIxmERERERERNVlB8ozkxSnE0REScSjPxERERERVZfNO6u5sQE3EVGiMZhERERERETV5fI+Dbg5nSAiSiIe/YmIiIiIqLpsng24iYgIAINJREREREQUhjsziQ24iYgSjcEkIiIiIiKqTjfgtiyWuRERJRyP/kREREREVF02D8zPAnNTwALL3IiIkozBJCIiIiIiqi6Xl9vpQ67MJAaTiIiSiMEkIiIiIiKqLquDSQddDbg5nSAiSqJ0szeAiIiIiIjagM5MmnIFk5iZRESUSLyUQERERERE1ZVkJrEBNxFRkvHoT0RERERE1WUH5HbqILBgB5PYgJuIKJEYTCIiIiIioupyg3JbkpnEYBIRURIxmERERERERNXpzKTpQ4BlyefMTCIiSiQGk4iIiIiIqLpUB9DVX1rmxp5JRESJxKM/ERERERGFk82zATcRETGYREREREREIeUG2ICbiIgYTCIiIiIiopDKMpMYTCIiSiIGk4iIiIiIKJysnZlkLcjXzEwiIkokBpOIiIiIiCicXF5Wc1uwg0nMTCIiSiQGk4iIiIiIKJyyMjfV3O0hIqKmYDCJiIiIiIjCyeWBuUlgbkq+ZpkbEVEiMZhEREREREThZPNyO7VfblnmRkSUSAwmERERERFRODqYNGkHk5iZRESUSAwmERERERFRODlPMImZSUREicRgEhERERERhVPMTNont4rTCSKiJIp99FdK/b5S6hGl1ANKqU+a2CgiIiIiImpBOU/PJJa5ERElUjrODyulXg7gEgAnWJY1o5QaNbNZRERERETUcpiZREREiJ+Z9B4Af2FZ1gwAWJa1J/4mERERERFRS8oOyK0OJjEziYgokeIGkzYDeKlS6tdKqduUUqcGPVApdZVS6k6l1J179+6N+WuJiIiIiKjh0p1AphuYPiRfswE3EVEiVS1zU0rdDGCZz7c+av/8IIAXATgVwNeVUusty7K8D7Ys61oA1wLAtm3byr5PRERERERtIJsH5iblc5a5ERElUtVgkmVZrwj6nlLqPQC+aQePbldKLQAYBsDUIyIiIiKixSiXBw4/K5+zzI2IKJHiXkr4NoCzAUAptRlAJ4CxuBtFREREREQtSjfhBljmRkSUULFWcwPweQCfV0rdD2AWwBV+JW5ERERERLRI5FzBpBTL3IiIkihWMMmyrFkAbzG0LURERERE1Op0ZhKzkoiIEouXEoiIiIiIKLzsgNyy+TYRUWLxDEBEREREROHpMjc23yYiSiwGk4iIiIiIKDyWuRERJR6DSUREREREFB4zk4iIEo/BJCIiIiIiCq+YmcSpBBFRUvEMQERERERE4eUYTCIiSjqeAYiIiIiIKDy9mhvL3IiIEovBJCIiIiIiCo8NuImIEo/BJCIiIiIiCo8NuImIEo/BJCIiIiIiCi+TAzq6mJlERJRgDCYREREREVE0uTygVLO3goiImoTBJCIiIiIiiiabZ5kbEVGCMZhERERERETRZAdY5kZElGAMJhERERERUTQ5ZiYRESVZutkbQEREREREbeaUtwNHnm/2VhARUZMwmERERERERNEc86pmbwERETURy9yIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0BpOIiIiIiIiIiCg0ZVlW43+pUnsBPN3wX1x/wwDGmr0RRCFwX6V2x32YFgPux7SYcH+mdsd9mBaDYQA9lmWN1PsXNSWYtFgppe60LGtbs7eDqBruq9TuuA/TYsD9mBYT7s/U7rgP02LQyP2YZW5ERERERERERBQag0lERERERERERBQag0lmXdvsDSAKifsqtTvuw7QYcD+mxYT7M7U77sO0GDRsP2bPJCIiIiIiIiIiCo2ZSUREREREREREFFqig0lKqc8rpfYope533XeiUuqXSqn7lFLfU0r12/d3KqW+YN9/j1LqLNfPvMm+/16l1PVKqeGA33eBUuoRpdTjSqkPu+6/xr7PCvpZSrYW2lc/Zz/nvUqpbyileuv4smmRaaH9+ItKqSeVUnfbHyfV8WXTItJC+/BPXfvvs0qpb9fxZdMi1UL789lKqbuUUvcrpa5TSqXr+LJpkWnCflz2++z7X6eUekAptaCU4opwFJrBffgN9v77gFLqkxV+3yn2zz+ulPoHpZSy74++D1uWldgPAGcC2Argftd9dwB4mf35OwD8b/vzqwF8wf58FMBvIMG4NIA9AIbt730SwJ/7/K4OANsBrAfQCeAeAFvs750MYC2Ap/Tz8IMf7o8W2lf7XY/7WwAfbvbfhh/t89FC+/EXAby22X8PfrTfR6vsw57H/ReAy5v9t+FH+320wv5sP8cOAJvtx30cwDub/bfhR/t8NHI/Dvp99v3HAjgawI8BbGv234Uf7fNhaB8eAvAMgBH7e9cBOCfg990O4HQACsAPAbzSvj/yPpzozCTLsn4CYL/n7qMB/MT+/CYAv2N/vgXAj+yf2wPgIIBtkH+CAtBjR/X6ATzr8+tOA/C4ZVlPWJY1C+A/AFxiP99vLct6ytDLokWohfbVcQCwfz4HgE3XKLRW2Y+JatVq+7BSqg/A2QCYmUSRtcj+PARgxrKsR31+J1FVDd6Pg34fLMt6yLKsR2K9GEokQ/vwegCPWpa1137czfA5liqllkOSA35pSQTpSwAutZ8v8j6c6GBSgPsBvNr+/HUAVtuf3wPgEqVUWim1DsApAFZbljUH4D0A7oMcdLYA+JzP866EXHnRdtr3EdWqKfuqUuoLAHYDOAbA/2/s1VBSNeuY+wk7FfhTSqkuY6+GkqiZ44bXAPiRDvQTGdDo/XkMQMZVUvFa1+8kqlW99mOiRom0DwN4HMAxSqm1dqnwpfA/lq6EHH+1WDEJBpPKvQPA1Uqp3wDoAzBr3/95yB/7TgB/B+AXAApKqQzk4HMygBUA7gXwEZ/nVT73MauD4mjKvmpZ1tvtn38IwBuMvBJKsmbsxx+BBENPBbAEwB8beSWUVM0cN7wJwFfjvgAil4buz/aV8TcC+JRS6nYAhwEUzL0cSqh67cdEjRJpH7Ys6wBkH/4agJ9C2uf4HUuNxiTY4M7DsqyHAZwHAEqpzQAutO8vAHi/fpxS6hcAHgNwkv397fb9XwfwYaXUagDfsx/+aUgU0R0dXIWA9EmiMJq5r1qWNa+U+hqADwL4gunXRsnRjP3Ysqzn7Ptm7Ey7D9TjtVEyNOtYrJQagpQOvaYer4uSqUnH5F8CeKn98+cB2FyfV0dJUa/92LKsTzfqNVCy1bAPw7Ks78HeX5VSVwGYV0p1QPoqAcB3AfwL5PirxYpJMJjkoZQatSxrj1IqBeBPICdAKKW6ASjLsiaUUudCIoAPKqVWANiilBqxaxTPBfCQZVk7YB+Y7J9PA9hkp6PtglyF+d3GvjpaTBq9r9o15Bssy3rc/vxiAA838CXTItSMY65SarllWc/Z+/GlkFRiopo0cdzwOgDftyxruhGvk5KhScdk/Tu7IJmin2jcK6bFqF77MVGjRN2HPT8zCOD3ALzesqx5ePZhpdRhpdSLAPwawOWI0bYk0cEkpdRXAZwFYFgptRPAxwD0KqWuth/yTThZF6MAblBKLUBOgm8FAMuynlVK/S8AP1FKzQF4GsDbvL/LsqyCUuoaADdAVrT4vGVZD9jb8V4AHwKwDMC9SqkfWJb1rjq8ZGpTrbCv2gez65QsTakgVxnfU4/XS4tTK+zH9re/opQagezHdwN4t+nXSotTC+3DgEzG/8LsK6QkaaH9+YNKqYsg7Tf+xbKsW4y/WFq0GrkfB/0+y7I+p5R6DWRSPgLgv5VSd1uWdb7p10uLj4l92Pb3SqkT7c8/7lrYwOs9kJWNc5DV3H5ob0fkfVhJqTIREREREREREVF1bMBNRERE9P/asQMBAAAAAEH+1isMUBgBALDJJAAAAAA2mQQAAADAJpMAAAAA2GQSAAAAAJtMAgAAAGCTSQAAAABsMgkAAACALYfrYQ/ypd8mAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plot the in-sample prediction results\n",
    "plt.figure(figsize=(20,6))\n",
    "plt.plot(df_test.index[1:len(df_test.index)], y_predict_test, label = \"LSTM Predicted Value\")\n",
    "plt.plot(df_test.index, df_test['SP500_%Change'], label = \"Original Value\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can't really work well due to very simple model structure (only 1 LSTM layer with bidirectional learning layer used). However the generic trend and fluctuation are roughly obtained, whilst having a limited volatility that is not enough to reflect how strong the real situation is. \n",
    "Ver. 2020-12-20"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note for model improvement:\n",
    "Due to noise of stock market data, outliers matters, especially in terms of robustness and preditability. It is hence suggested that we can scale the data regardless of X or Y (like sklearn RobustScale function) for better training and prediction. In this case, one should inversely transform the scaled data back to the original state for plotting, articulation or productionisation. \n",
    "Also, It is recommended to apply multivariate LSTM model to increase model complexity.\n",
    "2020-12-20"
   ]
  }
 ],
 "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
}