00_-_DL_keras_proof_of_concept_t-1_forecast.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.278104Z",
     "start_time": "2017-11-18T07:58:00.027595Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.281277Z",
     "start_time": "2017-11-18T07:58:00.279299Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Display only top 5 and bottom 5 rows\n",
    "pd.set_option('display.max_rows', 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.290334Z",
     "start_time": "2017-11-18T07:58:00.282329Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def nl():\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.356863Z",
     "start_time": "2017-11-18T07:58:00.291611Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "Size of training set: (48120, 4)\n",
      " Size of testing set: (11808, 3)\n",
      "\n",
      "\n",
      "Columns in train: ['DateTime', 'Junction', 'Vehicles', 'ID']\n",
      " Columns in test: ['DateTime', 'Junction', 'ID']\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>DateTime</th>\n",
       "      <th>Junction</th>\n",
       "      <th>Vehicles</th>\n",
       "      <th>ID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015-11-01 00:00:00</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>20151101001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015-11-01 01:00:00</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>20151101011</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015-11-01 02:00:00</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>20151101021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015-11-01 03:00:00</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>20151101031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015-11-01 04:00:00</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>20151101041</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             DateTime  Junction  Vehicles           ID\n",
       "0 2015-11-01 00:00:00         1        15  20151101001\n",
       "1 2015-11-01 01:00:00         1        13  20151101011\n",
       "2 2015-11-01 02:00:00         1        10  20151101021\n",
       "3 2015-11-01 03:00:00         1         7  20151101031\n",
       "4 2015-11-01 04:00:00         1         9  20151101041"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train = pd.read_csv('./data/train_aWnotuB.csv', parse_dates=[0], infer_datetime_format=True)\n",
    "df_test = pd.read_csv('./data/test_BdBKkAj.csv', parse_dates=[0], infer_datetime_format=True)\n",
    "\n",
    "nl()\n",
    "print('Size of training set: ' + str(df_train.shape))\n",
    "print(' Size of testing set: ' + str(df_test.shape))\n",
    "\n",
    "nl()\n",
    "print('Columns in train: ' + str(df_train.columns.tolist()))\n",
    "print(' Columns in test: ' + str(df_test.columns.tolist()))\n",
    "\n",
    "df_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Quick EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.364493Z",
     "start_time": "2017-11-18T07:58:00.358040Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_tmp = df_train.set_index(['Junction', 'DateTime'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.367572Z",
     "start_time": "2017-11-18T07:58:00.365656Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "level_values = df_tmp.index.get_level_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.383143Z",
     "start_time": "2017-11-18T07:58:00.368553Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "time_targets = df_tmp.groupby([level_values(0)] + [pd.Grouper(freq='1M', level=-1)])['Vehicles'].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.389541Z",
     "start_time": "2017-11-18T07:58:00.384543Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Junction  DateTime  \n",
       "1         2015-11-30    14736\n",
       "          2015-12-31    15487\n",
       "          2016-01-31    17940\n",
       "          2016-02-29    20813\n",
       "          2016-03-31    22215\n",
       "                        ...  \n",
       "4         2017-02-28     5564\n",
       "          2017-03-31     4931\n",
       "          2017-04-30     4454\n",
       "          2017-05-31     4877\n",
       "          2017-06-30     6097\n",
       "Name: Vehicles, Length: 66, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "time_targets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.393919Z",
     "start_time": "2017-11-18T07:58:00.390593Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "del df_tmp\n",
    "del time_targets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature engineering, lag_features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Convert each junction as a column\n",
    "We actually have 4 predictions to do, one per junction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.416099Z",
     "start_time": "2017-11-18T07:58:00.394937Z"
    }
   },
   "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>Junction</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-11-01 00:00:00</th>\n",
       "      <td>15.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 01:00:00</th>\n",
       "      <td>13.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 02:00:00</th>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 03:00:00</th>\n",
       "      <td>7.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 04:00:00</th>\n",
       "      <td>9.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 19:00:00</th>\n",
       "      <td>105.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>11.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 20:00:00</th>\n",
       "      <td>96.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>30.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 21:00:00</th>\n",
       "      <td>90.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>16.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 22:00:00</th>\n",
       "      <td>84.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 23:00:00</th>\n",
       "      <td>78.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14592 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Junction                 1     2     3     4\n",
       "DateTime                                    \n",
       "2015-11-01 00:00:00   15.0   6.0   9.0   NaN\n",
       "2015-11-01 01:00:00   13.0   6.0   7.0   NaN\n",
       "2015-11-01 02:00:00   10.0   5.0   5.0   NaN\n",
       "2015-11-01 03:00:00    7.0   6.0   1.0   NaN\n",
       "2015-11-01 04:00:00    9.0   7.0   2.0   NaN\n",
       "...                    ...   ...   ...   ...\n",
       "2017-06-30 19:00:00  105.0  34.0  33.0  11.0\n",
       "2017-06-30 20:00:00   96.0  35.0  31.0  30.0\n",
       "2017-06-30 21:00:00   90.0  31.0  28.0  16.0\n",
       "2017-06-30 22:00:00   84.0  29.0  26.0  22.0\n",
       "2017-06-30 23:00:00   78.0  27.0  39.0  12.0\n",
       "\n",
       "[14592 rows x 4 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train = df_train.pivot(index='DateTime',columns='Junction', values='Vehicles')\n",
    "train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Fill NaNs with 0 vehicles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.419658Z",
     "start_time": "2017-11-18T07:58:00.417380Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train = train.fillna(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate lag features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.440771Z",
     "start_time": "2017-11-18T07:58:00.420653Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def gen_lag_features(df, n_in=1, n_out=1, dropnan=True):\n",
    "    \"\"\"\n",
    "    Arguments:\n",
    "        data: Dataframe of observations.\n",
    "        n_in: Number of lag observations as input (X).\n",
    "        n_out: Number of forecast observations as output (y).\n",
    "        dropnan: Boolean whether or not to drop rows with NaN values.\n",
    "    Returns:\n",
    "        Dataframe.\n",
    "    \"\"\"\n",
    "    n_vars = df.shape[1]\n",
    "    cols, names = list(), list()\n",
    "    # input sequence (t-n, ... t-1)\n",
    "    for i in range(n_in, 0, -1):\n",
    "        cols.append(df.shift(i))\n",
    "        names += [('Junction %d (H-%d)' % (j+1, i)) for j in range(n_vars)]\n",
    "    # forecast sequence (t, t+1, ... t+n)\n",
    "    for i in range(0, n_out):\n",
    "        cols.append(df.shift(-i))\n",
    "        if i == 0:\n",
    "            names += [('Junction %d (H)' % (j+1)) for j in range(n_vars)]\n",
    "        else:\n",
    "            names += [('Junction %d (H+%d)' % (j+1, i)) for j in range(n_vars)]\n",
    "    # put it all together\n",
    "    agg = pd.concat(cols, axis=1)\n",
    "    agg.columns = names\n",
    "    # drop rows with NaN values\n",
    "    if dropnan:\n",
    "        agg.dropna(inplace=True)\n",
    "    return agg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.457335Z",
     "start_time": "2017-11-18T07:58:00.441972Z"
    }
   },
   "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>Junction 1 (H-1)</th>\n",
       "      <th>Junction 2 (H-1)</th>\n",
       "      <th>Junction 3 (H-1)</th>\n",
       "      <th>Junction 4 (H-1)</th>\n",
       "      <th>Junction 1 (H)</th>\n",
       "      <th>Junction 2 (H)</th>\n",
       "      <th>Junction 3 (H)</th>\n",
       "      <th>Junction 4 (H)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DateTime</th>\n",
       "      <th></th>\n",
       "      <th></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>2015-11-01 01:00:00</th>\n",
       "      <td>15.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 02:00:00</th>\n",
       "      <td>13.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 03:00:00</th>\n",
       "      <td>10.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 04:00:00</th>\n",
       "      <td>7.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 05:00:00</th>\n",
       "      <td>9.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 19:00:00</th>\n",
       "      <td>95.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>105.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>11.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 20:00:00</th>\n",
       "      <td>105.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>96.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>30.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 21:00:00</th>\n",
       "      <td>96.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>16.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 22:00:00</th>\n",
       "      <td>90.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>84.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 23:00:00</th>\n",
       "      <td>84.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>78.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14591 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Junction 1 (H-1)  Junction 2 (H-1)  Junction 3 (H-1)  \\\n",
       "DateTime                                                                    \n",
       "2015-11-01 01:00:00              15.0               6.0               9.0   \n",
       "2015-11-01 02:00:00              13.0               6.0               7.0   \n",
       "2015-11-01 03:00:00              10.0               5.0               5.0   \n",
       "2015-11-01 04:00:00               7.0               6.0               1.0   \n",
       "2015-11-01 05:00:00               9.0               7.0               2.0   \n",
       "...                               ...               ...               ...   \n",
       "2017-06-30 19:00:00              95.0              34.0              38.0   \n",
       "2017-06-30 20:00:00             105.0              34.0              33.0   \n",
       "2017-06-30 21:00:00              96.0              35.0              31.0   \n",
       "2017-06-30 22:00:00              90.0              31.0              28.0   \n",
       "2017-06-30 23:00:00              84.0              29.0              26.0   \n",
       "\n",
       "                     Junction 4 (H-1)  Junction 1 (H)  Junction 2 (H)  \\\n",
       "DateTime                                                                \n",
       "2015-11-01 01:00:00               0.0            13.0             6.0   \n",
       "2015-11-01 02:00:00               0.0            10.0             5.0   \n",
       "2015-11-01 03:00:00               0.0             7.0             6.0   \n",
       "2015-11-01 04:00:00               0.0             9.0             7.0   \n",
       "2015-11-01 05:00:00               0.0             6.0             2.0   \n",
       "...                               ...             ...             ...   \n",
       "2017-06-30 19:00:00              17.0           105.0            34.0   \n",
       "2017-06-30 20:00:00              11.0            96.0            35.0   \n",
       "2017-06-30 21:00:00              30.0            90.0            31.0   \n",
       "2017-06-30 22:00:00              16.0            84.0            29.0   \n",
       "2017-06-30 23:00:00              22.0            78.0            27.0   \n",
       "\n",
       "                     Junction 3 (H)  Junction 4 (H)  \n",
       "DateTime                                             \n",
       "2015-11-01 01:00:00             7.0             0.0  \n",
       "2015-11-01 02:00:00             5.0             0.0  \n",
       "2015-11-01 03:00:00             1.0             0.0  \n",
       "2015-11-01 04:00:00             2.0             0.0  \n",
       "2015-11-01 05:00:00             2.0             0.0  \n",
       "...                             ...             ...  \n",
       "2017-06-30 19:00:00            33.0            11.0  \n",
       "2017-06-30 20:00:00            31.0            30.0  \n",
       "2017-06-30 21:00:00            28.0            16.0  \n",
       "2017-06-30 22:00:00            26.0            22.0  \n",
       "2017-06-30 23:00:00            39.0            12.0  \n",
       "\n",
       "[14591 rows x 8 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Xy_train = gen_lag_features(train)\n",
    "Xy_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that the dataset does not start at 2015-11-01 midnigh anymore (there is no H-1 for midnight)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Normalize features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.589943Z",
     "start_time": "2017-11-18T07:58:00.458767Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler, StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.593494Z",
     "start_time": "2017-11-18T07:58:00.591076Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# scaler = MinMaxScaler(feature_range=(-1, 1))\n",
    "# scaler = StandardScaler()\n",
    "scaler = MinMaxScaler(feature_range=(0, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.602557Z",
     "start_time": "2017-11-18T07:58:00.594672Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Xy_train[Xy_train.columns] = scaler.fit_transform(Xy_train[Xy_train.columns])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.616610Z",
     "start_time": "2017-11-18T07:58:00.604230Z"
    }
   },
   "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>Junction 1 (H-1)</th>\n",
       "      <th>Junction 2 (H-1)</th>\n",
       "      <th>Junction 3 (H-1)</th>\n",
       "      <th>Junction 4 (H-1)</th>\n",
       "      <th>Junction 1 (H)</th>\n",
       "      <th>Junction 2 (H)</th>\n",
       "      <th>Junction 3 (H)</th>\n",
       "      <th>Junction 4 (H)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DateTime</th>\n",
       "      <th></th>\n",
       "      <th></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>2015-11-01 01:00:00</th>\n",
       "      <td>0.066225</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.044693</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.052980</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.033520</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 02:00:00</th>\n",
       "      <td>0.052980</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.033520</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.033113</td>\n",
       "      <td>0.085106</td>\n",
       "      <td>0.022346</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 03:00:00</th>\n",
       "      <td>0.033113</td>\n",
       "      <td>0.085106</td>\n",
       "      <td>0.022346</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.013245</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 04:00:00</th>\n",
       "      <td>0.013245</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.026490</td>\n",
       "      <td>0.127660</td>\n",
       "      <td>0.005587</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 05:00:00</th>\n",
       "      <td>0.026490</td>\n",
       "      <td>0.127660</td>\n",
       "      <td>0.005587</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.006623</td>\n",
       "      <td>0.021277</td>\n",
       "      <td>0.005587</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 19:00:00</th>\n",
       "      <td>0.596026</td>\n",
       "      <td>0.702128</td>\n",
       "      <td>0.206704</td>\n",
       "      <td>0.472222</td>\n",
       "      <td>0.662252</td>\n",
       "      <td>0.702128</td>\n",
       "      <td>0.178771</td>\n",
       "      <td>0.305556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 20:00:00</th>\n",
       "      <td>0.662252</td>\n",
       "      <td>0.702128</td>\n",
       "      <td>0.178771</td>\n",
       "      <td>0.305556</td>\n",
       "      <td>0.602649</td>\n",
       "      <td>0.723404</td>\n",
       "      <td>0.167598</td>\n",
       "      <td>0.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 21:00:00</th>\n",
       "      <td>0.602649</td>\n",
       "      <td>0.723404</td>\n",
       "      <td>0.167598</td>\n",
       "      <td>0.833333</td>\n",
       "      <td>0.562914</td>\n",
       "      <td>0.638298</td>\n",
       "      <td>0.150838</td>\n",
       "      <td>0.444444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 22:00:00</th>\n",
       "      <td>0.562914</td>\n",
       "      <td>0.638298</td>\n",
       "      <td>0.150838</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>0.523179</td>\n",
       "      <td>0.595745</td>\n",
       "      <td>0.139665</td>\n",
       "      <td>0.611111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 23:00:00</th>\n",
       "      <td>0.523179</td>\n",
       "      <td>0.595745</td>\n",
       "      <td>0.139665</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.483444</td>\n",
       "      <td>0.553191</td>\n",
       "      <td>0.212291</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14591 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Junction 1 (H-1)  Junction 2 (H-1)  Junction 3 (H-1)  \\\n",
       "DateTime                                                                    \n",
       "2015-11-01 01:00:00          0.066225          0.106383          0.044693   \n",
       "2015-11-01 02:00:00          0.052980          0.106383          0.033520   \n",
       "2015-11-01 03:00:00          0.033113          0.085106          0.022346   \n",
       "2015-11-01 04:00:00          0.013245          0.106383          0.000000   \n",
       "2015-11-01 05:00:00          0.026490          0.127660          0.005587   \n",
       "...                               ...               ...               ...   \n",
       "2017-06-30 19:00:00          0.596026          0.702128          0.206704   \n",
       "2017-06-30 20:00:00          0.662252          0.702128          0.178771   \n",
       "2017-06-30 21:00:00          0.602649          0.723404          0.167598   \n",
       "2017-06-30 22:00:00          0.562914          0.638298          0.150838   \n",
       "2017-06-30 23:00:00          0.523179          0.595745          0.139665   \n",
       "\n",
       "                     Junction 4 (H-1)  Junction 1 (H)  Junction 2 (H)  \\\n",
       "DateTime                                                                \n",
       "2015-11-01 01:00:00          0.000000        0.052980        0.106383   \n",
       "2015-11-01 02:00:00          0.000000        0.033113        0.085106   \n",
       "2015-11-01 03:00:00          0.000000        0.013245        0.106383   \n",
       "2015-11-01 04:00:00          0.000000        0.026490        0.127660   \n",
       "2015-11-01 05:00:00          0.000000        0.006623        0.021277   \n",
       "...                               ...             ...             ...   \n",
       "2017-06-30 19:00:00          0.472222        0.662252        0.702128   \n",
       "2017-06-30 20:00:00          0.305556        0.602649        0.723404   \n",
       "2017-06-30 21:00:00          0.833333        0.562914        0.638298   \n",
       "2017-06-30 22:00:00          0.444444        0.523179        0.595745   \n",
       "2017-06-30 23:00:00          0.611111        0.483444        0.553191   \n",
       "\n",
       "                     Junction 3 (H)  Junction 4 (H)  \n",
       "DateTime                                             \n",
       "2015-11-01 01:00:00        0.033520        0.000000  \n",
       "2015-11-01 02:00:00        0.022346        0.000000  \n",
       "2015-11-01 03:00:00        0.000000        0.000000  \n",
       "2015-11-01 04:00:00        0.005587        0.000000  \n",
       "2015-11-01 05:00:00        0.005587        0.000000  \n",
       "...                             ...             ...  \n",
       "2017-06-30 19:00:00        0.178771        0.305556  \n",
       "2017-06-30 20:00:00        0.167598        0.833333  \n",
       "2017-06-30 21:00:00        0.150838        0.444444  \n",
       "2017-06-30 22:00:00        0.139665        0.611111  \n",
       "2017-06-30 23:00:00        0.212291        0.333333  \n",
       "\n",
       "[14591 rows x 8 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Xy_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Split train and valid (and normalize for real)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.630042Z",
     "start_time": "2017-11-18T07:58:00.618220Z"
    }
   },
   "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>Junction 1 (H-1)</th>\n",
       "      <th>Junction 2 (H-1)</th>\n",
       "      <th>Junction 3 (H-1)</th>\n",
       "      <th>Junction 4 (H-1)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-11-01 01:00:00</th>\n",
       "      <td>0.066225</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.044693</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 02:00:00</th>\n",
       "      <td>0.052980</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.033520</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 03:00:00</th>\n",
       "      <td>0.033113</td>\n",
       "      <td>0.085106</td>\n",
       "      <td>0.022346</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 04:00:00</th>\n",
       "      <td>0.013245</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 05:00:00</th>\n",
       "      <td>0.026490</td>\n",
       "      <td>0.127660</td>\n",
       "      <td>0.005587</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 19:00:00</th>\n",
       "      <td>0.476821</td>\n",
       "      <td>0.574468</td>\n",
       "      <td>0.178771</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 20:00:00</th>\n",
       "      <td>0.496689</td>\n",
       "      <td>0.531915</td>\n",
       "      <td>0.156425</td>\n",
       "      <td>0.222222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 21:00:00</th>\n",
       "      <td>0.483444</td>\n",
       "      <td>0.638298</td>\n",
       "      <td>0.156425</td>\n",
       "      <td>0.222222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 22:00:00</th>\n",
       "      <td>0.403974</td>\n",
       "      <td>0.574468</td>\n",
       "      <td>0.150838</td>\n",
       "      <td>0.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 23:00:00</th>\n",
       "      <td>0.423841</td>\n",
       "      <td>0.553191</td>\n",
       "      <td>0.162011</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12407 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Junction 1 (H-1)  Junction 2 (H-1)  Junction 3 (H-1)  \\\n",
       "DateTime                                                                    \n",
       "2015-11-01 01:00:00          0.066225          0.106383          0.044693   \n",
       "2015-11-01 02:00:00          0.052980          0.106383          0.033520   \n",
       "2015-11-01 03:00:00          0.033113          0.085106          0.022346   \n",
       "2015-11-01 04:00:00          0.013245          0.106383          0.000000   \n",
       "2015-11-01 05:00:00          0.026490          0.127660          0.005587   \n",
       "...                               ...               ...               ...   \n",
       "2017-03-31 19:00:00          0.476821          0.574468          0.178771   \n",
       "2017-03-31 20:00:00          0.496689          0.531915          0.156425   \n",
       "2017-03-31 21:00:00          0.483444          0.638298          0.156425   \n",
       "2017-03-31 22:00:00          0.403974          0.574468          0.150838   \n",
       "2017-03-31 23:00:00          0.423841          0.553191          0.162011   \n",
       "\n",
       "                     Junction 4 (H-1)  \n",
       "DateTime                               \n",
       "2015-11-01 01:00:00          0.000000  \n",
       "2015-11-01 02:00:00          0.000000  \n",
       "2015-11-01 03:00:00          0.000000  \n",
       "2015-11-01 04:00:00          0.000000  \n",
       "2015-11-01 05:00:00          0.000000  \n",
       "...                               ...  \n",
       "2017-03-31 19:00:00          0.166667  \n",
       "2017-03-31 20:00:00          0.222222  \n",
       "2017-03-31 21:00:00          0.222222  \n",
       "2017-03-31 22:00:00          0.250000  \n",
       "2017-03-31 23:00:00          0.166667  \n",
       "\n",
       "[12407 rows x 4 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train = Xy_train[Xy_train.index < '2017-04-01'].iloc[:,0:4]\n",
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.643078Z",
     "start_time": "2017-11-18T07:58:00.631507Z"
    }
   },
   "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>Junction 1 (H)</th>\n",
       "      <th>Junction 2 (H)</th>\n",
       "      <th>Junction 3 (H)</th>\n",
       "      <th>Junction 4 (H)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-11-01 01:00:00</th>\n",
       "      <td>0.052980</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.033520</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 02:00:00</th>\n",
       "      <td>0.033113</td>\n",
       "      <td>0.085106</td>\n",
       "      <td>0.022346</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 03:00:00</th>\n",
       "      <td>0.013245</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 04:00:00</th>\n",
       "      <td>0.026490</td>\n",
       "      <td>0.127660</td>\n",
       "      <td>0.005587</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01 05:00:00</th>\n",
       "      <td>0.006623</td>\n",
       "      <td>0.021277</td>\n",
       "      <td>0.005587</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 19:00:00</th>\n",
       "      <td>0.496689</td>\n",
       "      <td>0.531915</td>\n",
       "      <td>0.156425</td>\n",
       "      <td>0.222222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 20:00:00</th>\n",
       "      <td>0.483444</td>\n",
       "      <td>0.638298</td>\n",
       "      <td>0.156425</td>\n",
       "      <td>0.222222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 21:00:00</th>\n",
       "      <td>0.403974</td>\n",
       "      <td>0.574468</td>\n",
       "      <td>0.150838</td>\n",
       "      <td>0.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 22:00:00</th>\n",
       "      <td>0.423841</td>\n",
       "      <td>0.553191</td>\n",
       "      <td>0.162011</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31 23:00:00</th>\n",
       "      <td>0.417219</td>\n",
       "      <td>0.553191</td>\n",
       "      <td>0.162011</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12407 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Junction 1 (H)  Junction 2 (H)  Junction 3 (H)  \\\n",
       "DateTime                                                              \n",
       "2015-11-01 01:00:00        0.052980        0.106383        0.033520   \n",
       "2015-11-01 02:00:00        0.033113        0.085106        0.022346   \n",
       "2015-11-01 03:00:00        0.013245        0.106383        0.000000   \n",
       "2015-11-01 04:00:00        0.026490        0.127660        0.005587   \n",
       "2015-11-01 05:00:00        0.006623        0.021277        0.005587   \n",
       "...                             ...             ...             ...   \n",
       "2017-03-31 19:00:00        0.496689        0.531915        0.156425   \n",
       "2017-03-31 20:00:00        0.483444        0.638298        0.156425   \n",
       "2017-03-31 21:00:00        0.403974        0.574468        0.150838   \n",
       "2017-03-31 22:00:00        0.423841        0.553191        0.162011   \n",
       "2017-03-31 23:00:00        0.417219        0.553191        0.162011   \n",
       "\n",
       "                     Junction 4 (H)  \n",
       "DateTime                             \n",
       "2015-11-01 01:00:00        0.000000  \n",
       "2015-11-01 02:00:00        0.000000  \n",
       "2015-11-01 03:00:00        0.000000  \n",
       "2015-11-01 04:00:00        0.000000  \n",
       "2015-11-01 05:00:00        0.000000  \n",
       "...                             ...  \n",
       "2017-03-31 19:00:00        0.222222  \n",
       "2017-03-31 20:00:00        0.222222  \n",
       "2017-03-31 21:00:00        0.250000  \n",
       "2017-03-31 22:00:00        0.166667  \n",
       "2017-03-31 23:00:00        0.166667  \n",
       "\n",
       "[12407 rows x 4 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train = Xy_train[Xy_train.index < '2017-04-01'].iloc[:,4:]\n",
    "y_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reshape the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.647835Z",
     "start_time": "2017-11-18T07:58:00.644636Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(12407, 4) (12407, 4)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape, y_train.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have the following dimensions:\n",
    "(Batch_size, junctions, features) with Batch_size = 12407, junctions = 4, features = 1\n",
    "\n",
    "junctions can be thought as \"Colors\" and \"channels\" in image predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:00.654520Z",
     "start_time": "2017-11-18T07:58:00.649386Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(12407, 4, 1)\n",
      "(12407, 4)\n"
     ]
    }
   ],
   "source": [
    "X_train = np.expand_dims(X_train.values, axis=2)\n",
    "print(X_train.shape)\n",
    "\n",
    "y_train = y_train.values\n",
    "print(y_train.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Defining and training a model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:02.395681Z",
     "start_time": "2017-11-18T07:58:00.655592Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import LSTM\n",
    "from keras.initializers import he_normal"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:02.400266Z",
     "start_time": "2017-11-18T07:58:02.397258Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import keras.backend as K\n",
    "\n",
    "def root_mean_squared_error(y_true, y_pred):\n",
    "        return K.sqrt(K.mean(K.square(y_pred - y_true), axis=-1)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:02.667185Z",
     "start_time": "2017-11-18T07:58:02.401591Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Initialising the RNN\n",
    "regressor = Sequential()\n",
    "\n",
    "# Adding the input layer and the LSTM layer\n",
    "regressor.add(LSTM(units = 50,\n",
    "                   activation = 'relu', # default is tanh\n",
    "                   kernel_initializer = he_normal(seed=0),\n",
    "                   input_shape = (None, 1)))\n",
    "\n",
    "# Output for 4 junctions\n",
    "regressor.add(Dense(units = 4))\n",
    "\n",
    "# Compiling the RNN\n",
    "regressor.compile(optimizer = 'adam', loss = root_mean_squared_error)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:16.393125Z",
     "start_time": "2017-11-18T07:58:02.668458Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 76s - loss: 0.2951\n",
      " 1408/12407 [==>...........................] - ETA: 7s - loss: 0.2626 \n",
      " 2816/12407 [=====>........................] - ETA: 3s - loss: 0.2227\n",
      " 4096/12407 [========>.....................] - ETA: 2s - loss: 0.1876\n",
      " 5504/12407 [============>.................] - ETA: 1s - loss: 0.1558\n",
      " 6912/12407 [===============>..............] - ETA: 1s - loss: 0.1353\n",
      " 8320/12407 [===================>..........] - ETA: 1s - loss: 0.1209\n",
      " 9600/12407 [======================>.......] - ETA: 0s - loss: 0.1111\n",
      "10880/12407 [=========================>....] - ETA: 0s - loss: 0.1035\n",
      "12160/12407 [============================>.] - ETA: 0s - loss: 0.0972\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0961\n",
      "Epoch 2/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0425\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0426\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0417\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0412\n",
      " 5504/12407 [============>.................] - ETA: 0s - loss: 0.0408\n",
      " 6784/12407 [===============>..............] - ETA: 0s - loss: 0.0407\n",
      " 8192/12407 [==================>...........] - ETA: 0s - loss: 0.0406\n",
      " 9600/12407 [======================>.......] - ETA: 0s - loss: 0.0404\n",
      "11008/12407 [=========================>....] - ETA: 0s - loss: 0.0402\n",
      "12288/12407 [============================>.] - ETA: 0s - loss: 0.0400\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0400\n",
      "Epoch 3/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0403\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0383\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0382\n",
      " 4096/12407 [========>.....................] - ETA: 0s - loss: 0.0379\n",
      " 5376/12407 [===========>..................] - ETA: 0s - loss: 0.0379\n",
      " 6784/12407 [===============>..............] - ETA: 0s - loss: 0.0382\n",
      " 8064/12407 [==================>...........] - ETA: 0s - loss: 0.0380\n",
      " 9472/12407 [=====================>........] - ETA: 0s - loss: 0.0379\n",
      "10752/12407 [========================>.....] - ETA: 0s - loss: 0.0376\n",
      "12160/12407 [============================>.] - ETA: 0s - loss: 0.0374\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0374\n",
      "Epoch 4/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0347\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0365\n",
      " 2944/12407 [======>.......................] - ETA: 0s - loss: 0.0363\n",
      " 4352/12407 [=========>....................] - ETA: 0s - loss: 0.0364\n",
      " 5760/12407 [============>.................] - ETA: 0s - loss: 0.0366\n",
      " 7040/12407 [================>.............] - ETA: 0s - loss: 0.0364\n",
      " 8448/12407 [===================>..........] - ETA: 0s - loss: 0.0363\n",
      " 9856/12407 [======================>.......] - ETA: 0s - loss: 0.0363\n",
      "11136/12407 [=========================>....] - ETA: 0s - loss: 0.0361\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0360\n",
      "Epoch 5/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0374\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0362\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0359\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0356\n",
      " 5632/12407 [============>.................] - ETA: 0s - loss: 0.0353\n",
      " 6912/12407 [===============>..............] - ETA: 0s - loss: 0.0350\n",
      " 8320/12407 [===================>..........] - ETA: 0s - loss: 0.0347\n",
      " 9728/12407 [======================>.......] - ETA: 0s - loss: 0.0347\n",
      "11136/12407 [=========================>....] - ETA: 0s - loss: 0.0346\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0346\n",
      "Epoch 6/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0335\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0344\n",
      " 2944/12407 [======>.......................] - ETA: 0s - loss: 0.0342\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0340\n",
      " 5632/12407 [============>.................] - ETA: 0s - loss: 0.0338\n",
      " 7040/12407 [================>.............] - ETA: 0s - loss: 0.0339\n",
      " 8448/12407 [===================>..........] - ETA: 0s - loss: 0.0339\n",
      " 9856/12407 [======================>.......] - ETA: 0s - loss: 0.0338\n",
      "11264/12407 [==========================>...] - ETA: 0s - loss: 0.0337\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0337\n",
      "Epoch 7/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0357\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0338\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0340\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0340\n",
      " 5632/12407 [============>.................] - ETA: 0s - loss: 0.0337\n",
      " 7040/12407 [================>.............] - ETA: 0s - loss: 0.0336\n",
      " 8448/12407 [===================>..........] - ETA: 0s - loss: 0.0334\n",
      " 9856/12407 [======================>.......] - ETA: 0s - loss: 0.0333\n",
      "11264/12407 [==========================>...] - ETA: 0s - loss: 0.0333\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0333\n",
      "Epoch 8/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0321\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0330\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0333\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0331\n",
      " 5632/12407 [============>.................] - ETA: 0s - loss: 0.0329\n",
      " 7040/12407 [================>.............] - ETA: 0s - loss: 0.0331\n",
      " 8448/12407 [===================>..........] - ETA: 0s - loss: 0.0331\n",
      " 9856/12407 [======================>.......] - ETA: 0s - loss: 0.0330\n",
      "11264/12407 [==========================>...] - ETA: 0s - loss: 0.0330\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0330\n",
      "Epoch 9/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0320\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0325\n",
      " 2688/12407 [=====>........................] - ETA: 0s - loss: 0.0337\n",
      " 4096/12407 [========>.....................] - ETA: 0s - loss: 0.0336\n",
      " 5504/12407 [============>.................] - ETA: 0s - loss: 0.0335\n",
      " 6912/12407 [===============>..............] - ETA: 0s - loss: 0.0333\n",
      " 8320/12407 [===================>..........] - ETA: 0s - loss: 0.0332\n",
      " 9600/12407 [======================>.......] - ETA: 0s - loss: 0.0330\n",
      "11008/12407 [=========================>....] - ETA: 0s - loss: 0.0329\n",
      "12288/12407 [============================>.] - ETA: 0s - loss: 0.0328\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0328\n",
      "Epoch 10/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0334\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0337\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0329\n",
      " 4096/12407 [========>.....................] - ETA: 0s - loss: 0.0326\n",
      " 5504/12407 [============>.................] - ETA: 0s - loss: 0.0329\n",
      " 6912/12407 [===============>..............] - ETA: 0s - loss: 0.0329\n",
      " 8192/12407 [==================>...........] - ETA: 0s - loss: 0.0328\n",
      " 9472/12407 [=====================>........] - ETA: 0s - loss: 0.0328\n",
      "10752/12407 [========================>.....] - ETA: 0s - loss: 0.0329\n",
      "12032/12407 [============================>.] - ETA: 0s - loss: 0.0328\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0328\n",
      "Epoch 11/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0316\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0320\n",
      " 2688/12407 [=====>........................] - ETA: 0s - loss: 0.0323\n",
      " 4096/12407 [========>.....................] - ETA: 0s - loss: 0.0327\n",
      " 5376/12407 [===========>..................] - ETA: 0s - loss: 0.0326\n",
      " 6656/12407 [===============>..............] - ETA: 0s - loss: 0.0325\n",
      " 7936/12407 [==================>...........] - ETA: 0s - loss: 0.0326\n",
      " 9344/12407 [=====================>........] - ETA: 0s - loss: 0.0327\n",
      "10624/12407 [========================>.....] - ETA: 0s - loss: 0.0326\n",
      "11904/12407 [===========================>..] - ETA: 0s - loss: 0.0326\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0327\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0315\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0327\n",
      " 2688/12407 [=====>........................] - ETA: 0s - loss: 0.0329\n",
      " 3968/12407 [========>.....................] - ETA: 0s - loss: 0.0325\n",
      " 5376/12407 [===========>..................] - ETA: 0s - loss: 0.0326\n",
      " 6784/12407 [===============>..............] - ETA: 0s - loss: 0.0326\n",
      " 8064/12407 [==================>...........] - ETA: 0s - loss: 0.0326\n",
      " 9472/12407 [=====================>........] - ETA: 0s - loss: 0.0326\n",
      "10752/12407 [========================>.....] - ETA: 0s - loss: 0.0326\n",
      "12032/12407 [============================>.] - ETA: 0s - loss: 0.0326\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0326\n",
      "Epoch 13/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0329\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0323\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0326\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0328\n",
      " 5504/12407 [============>.................] - ETA: 0s - loss: 0.0326\n",
      " 6784/12407 [===============>..............] - ETA: 0s - loss: 0.0327\n",
      " 8192/12407 [==================>...........] - ETA: 0s - loss: 0.0327\n",
      " 9600/12407 [======================>.......] - ETA: 0s - loss: 0.0327\n",
      "11008/12407 [=========================>....] - ETA: 0s - loss: 0.0326\n",
      "12288/12407 [============================>.] - ETA: 0s - loss: 0.0326\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0326\n",
      "Epoch 14/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0326\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0324\n",
      " 2944/12407 [======>.......................] - ETA: 0s - loss: 0.0324\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0320\n",
      " 5632/12407 [============>.................] - ETA: 0s - loss: 0.0320\n",
      " 6912/12407 [===============>..............] - ETA: 0s - loss: 0.0322\n",
      " 8320/12407 [===================>..........] - ETA: 0s - loss: 0.0324\n",
      " 9728/12407 [======================>.......] - ETA: 0s - loss: 0.0326\n",
      "11136/12407 [=========================>....] - ETA: 0s - loss: 0.0325\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0325\n",
      "Epoch 15/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0298\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0316\n",
      " 2944/12407 [======>.......................] - ETA: 0s - loss: 0.0324\n",
      " 4352/12407 [=========>....................] - ETA: 0s - loss: 0.0325\n",
      " 5760/12407 [============>.................] - ETA: 0s - loss: 0.0325\n",
      " 7168/12407 [================>.............] - ETA: 0s - loss: 0.0325\n",
      " 8448/12407 [===================>..........] - ETA: 0s - loss: 0.0325\n",
      " 9856/12407 [======================>.......] - ETA: 0s - loss: 0.0325\n",
      "11136/12407 [=========================>....] - ETA: 0s - loss: 0.0326\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0326\n",
      "Epoch 16/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0309\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0331\n",
      " 2944/12407 [======>.......................] - ETA: 0s - loss: 0.0322\n",
      " 4352/12407 [=========>....................] - ETA: 0s - loss: 0.0321\n",
      " 5760/12407 [============>.................] - ETA: 0s - loss: 0.0322\n",
      " 7168/12407 [================>.............] - ETA: 0s - loss: 0.0323\n",
      " 8448/12407 [===================>..........] - ETA: 0s - loss: 0.0324\n",
      " 9856/12407 [======================>.......] - ETA: 0s - loss: 0.0325\n",
      "11264/12407 [==========================>...] - ETA: 0s - loss: 0.0325\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0325\n",
      "Epoch 17/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0320\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0321\n",
      " 2944/12407 [======>.......................] - ETA: 0s - loss: 0.0321\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0321\n",
      " 5632/12407 [============>.................] - ETA: 0s - loss: 0.0322\n",
      " 6912/12407 [===============>..............] - ETA: 0s - loss: 0.0322\n",
      " 8320/12407 [===================>..........] - ETA: 0s - loss: 0.0323\n",
      " 9728/12407 [======================>.......] - ETA: 0s - loss: 0.0322\n",
      "11008/12407 [=========================>....] - ETA: 0s - loss: 0.0323\n",
      "12288/12407 [============================>.] - ETA: 0s - loss: 0.0323\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0323\n",
      "Epoch 18/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0343\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0323\n",
      " 2688/12407 [=====>........................] - ETA: 0s - loss: 0.0320\n",
      " 3968/12407 [========>.....................] - ETA: 0s - loss: 0.0323\n",
      " 5376/12407 [===========>..................] - ETA: 0s - loss: 0.0322\n",
      " 6784/12407 [===============>..............] - ETA: 0s - loss: 0.0323\n",
      " 8192/12407 [==================>...........] - ETA: 0s - loss: 0.0323\n",
      " 9600/12407 [======================>.......] - ETA: 0s - loss: 0.0324\n",
      "11008/12407 [=========================>....] - ETA: 0s - loss: 0.0324\n",
      "12288/12407 [============================>.] - ETA: 0s - loss: 0.0324\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0324\n",
      "Epoch 19/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0316\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0325\n",
      " 2944/12407 [======>.......................] - ETA: 0s - loss: 0.0329\n",
      " 4352/12407 [=========>....................] - ETA: 0s - loss: 0.0325\n",
      " 5632/12407 [============>.................] - ETA: 0s - loss: 0.0325\n",
      " 7040/12407 [================>.............] - ETA: 0s - loss: 0.0324\n",
      " 8448/12407 [===================>..........] - ETA: 0s - loss: 0.0325\n",
      " 9856/12407 [======================>.......] - ETA: 0s - loss: 0.0323\n",
      "11136/12407 [=========================>....] - ETA: 0s - loss: 0.0323\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0323\n",
      "Epoch 20/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0331\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0315\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0318\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0319\n",
      " 5504/12407 [============>.................] - ETA: 0s - loss: 0.0320\n",
      " 6784/12407 [===============>..............] - ETA: 0s - loss: 0.0322\n",
      " 8064/12407 [==================>...........] - ETA: 0s - loss: 0.0321\n",
      " 9472/12407 [=====================>........] - ETA: 0s - loss: 0.0321\n",
      "10880/12407 [=========================>....] - ETA: 0s - loss: 0.0321\n",
      "12288/12407 [============================>.] - ETA: 0s - loss: 0.0322\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0322\n",
      "Epoch 21/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0314\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0321\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0324\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0324\n",
      " 5632/12407 [============>.................] - ETA: 0s - loss: 0.0321\n",
      " 7040/12407 [================>.............] - ETA: 0s - loss: 0.0320\n",
      " 8448/12407 [===================>..........] - ETA: 0s - loss: 0.0323\n",
      " 9856/12407 [======================>.......] - ETA: 0s - loss: 0.0322\n",
      "11136/12407 [=========================>....] - ETA: 0s - loss: 0.0323\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0323\n",
      "Epoch 22/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0333\n",
      " 1408/12407 [==>...........................] - ETA: 0s - loss: 0.0333\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0328\n",
      " 4096/12407 [========>.....................] - ETA: 0s - loss: 0.0324\n",
      " 5376/12407 [===========>..................] - ETA: 0s - loss: 0.0325\n",
      " 6656/12407 [===============>..............] - ETA: 0s - loss: 0.0325\n",
      " 8064/12407 [==================>...........] - ETA: 0s - loss: 0.0323\n",
      " 9472/12407 [=====================>........] - ETA: 0s - loss: 0.0321\n",
      "10752/12407 [========================>.....] - ETA: 0s - loss: 0.0321\n",
      "12160/12407 [============================>.] - ETA: 0s - loss: 0.0322\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0322\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0339\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0323\n",
      " 2944/12407 [======>.......................] - ETA: 0s - loss: 0.0327\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0324\n",
      " 5504/12407 [============>.................] - ETA: 0s - loss: 0.0322\n",
      " 6912/12407 [===============>..............] - ETA: 0s - loss: 0.0321\n",
      " 8192/12407 [==================>...........] - ETA: 0s - loss: 0.0323\n",
      " 9600/12407 [======================>.......] - ETA: 0s - loss: 0.0322\n",
      "11008/12407 [=========================>....] - ETA: 0s - loss: 0.0322\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0322\n",
      "Epoch 24/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0346\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0323\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0322\n",
      " 4224/12407 [=========>....................] - ETA: 0s - loss: 0.0324\n",
      " 5632/12407 [============>.................] - ETA: 0s - loss: 0.0326\n",
      " 7040/12407 [================>.............] - ETA: 0s - loss: 0.0325\n",
      " 8320/12407 [===================>..........] - ETA: 0s - loss: 0.0323\n",
      " 9728/12407 [======================>.......] - ETA: 0s - loss: 0.0322\n",
      "11008/12407 [=========================>....] - ETA: 0s - loss: 0.0323\n",
      "12288/12407 [============================>.] - ETA: 0s - loss: 0.0322\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0322\n",
      "Epoch 25/25\n",
      "\n",
      "  128/12407 [..............................] - ETA: 1s - loss: 0.0361\n",
      " 1536/12407 [==>...........................] - ETA: 0s - loss: 0.0323\n",
      " 2816/12407 [=====>........................] - ETA: 0s - loss: 0.0323\n",
      " 4096/12407 [========>.....................] - ETA: 0s - loss: 0.0323\n",
      " 5504/12407 [============>.................] - ETA: 0s - loss: 0.0323\n",
      " 6912/12407 [===============>..............] - ETA: 0s - loss: 0.0323\n",
      " 8320/12407 [===================>..........] - ETA: 0s - loss: 0.0325\n",
      " 9728/12407 [======================>.......] - ETA: 0s - loss: 0.0324\n",
      "11136/12407 [=========================>....] - ETA: 0s - loss: 0.0322\n",
      "12407/12407 [==============================] - ETA: 0s - loss: 0.0322\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7f6277c358d0>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Fitting the RNN to the Training set\n",
    "regressor.fit(X_train, y_train, batch_size = 128, epochs = 25, verbose = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T06:49:22.553588Z",
     "start_time": "2017-11-18T06:49:22.551687Z"
    }
   },
   "source": [
    "# Validating"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:16.407666Z",
     "start_time": "2017-11-18T07:58:16.394464Z"
    }
   },
   "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>Junction 1 (H-1)</th>\n",
       "      <th>Junction 2 (H-1)</th>\n",
       "      <th>Junction 3 (H-1)</th>\n",
       "      <th>Junction 4 (H-1)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-04-01 00:00:00</th>\n",
       "      <td>0.417219</td>\n",
       "      <td>0.553191</td>\n",
       "      <td>0.162011</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-01 01:00:00</th>\n",
       "      <td>0.384106</td>\n",
       "      <td>0.510638</td>\n",
       "      <td>0.122905</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-01 02:00:00</th>\n",
       "      <td>0.317881</td>\n",
       "      <td>0.574468</td>\n",
       "      <td>0.078212</td>\n",
       "      <td>0.138889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-01 03:00:00</th>\n",
       "      <td>0.238411</td>\n",
       "      <td>0.361702</td>\n",
       "      <td>0.083799</td>\n",
       "      <td>0.111111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-01 04:00:00</th>\n",
       "      <td>0.225166</td>\n",
       "      <td>0.361702</td>\n",
       "      <td>0.055866</td>\n",
       "      <td>0.111111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 19:00:00</th>\n",
       "      <td>0.596026</td>\n",
       "      <td>0.702128</td>\n",
       "      <td>0.206704</td>\n",
       "      <td>0.472222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 20:00:00</th>\n",
       "      <td>0.662252</td>\n",
       "      <td>0.702128</td>\n",
       "      <td>0.178771</td>\n",
       "      <td>0.305556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 21:00:00</th>\n",
       "      <td>0.602649</td>\n",
       "      <td>0.723404</td>\n",
       "      <td>0.167598</td>\n",
       "      <td>0.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 22:00:00</th>\n",
       "      <td>0.562914</td>\n",
       "      <td>0.638298</td>\n",
       "      <td>0.150838</td>\n",
       "      <td>0.444444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 23:00:00</th>\n",
       "      <td>0.523179</td>\n",
       "      <td>0.595745</td>\n",
       "      <td>0.139665</td>\n",
       "      <td>0.611111</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2184 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Junction 1 (H-1)  Junction 2 (H-1)  Junction 3 (H-1)  \\\n",
       "DateTime                                                                    \n",
       "2017-04-01 00:00:00          0.417219          0.553191          0.162011   \n",
       "2017-04-01 01:00:00          0.384106          0.510638          0.122905   \n",
       "2017-04-01 02:00:00          0.317881          0.574468          0.078212   \n",
       "2017-04-01 03:00:00          0.238411          0.361702          0.083799   \n",
       "2017-04-01 04:00:00          0.225166          0.361702          0.055866   \n",
       "...                               ...               ...               ...   \n",
       "2017-06-30 19:00:00          0.596026          0.702128          0.206704   \n",
       "2017-06-30 20:00:00          0.662252          0.702128          0.178771   \n",
       "2017-06-30 21:00:00          0.602649          0.723404          0.167598   \n",
       "2017-06-30 22:00:00          0.562914          0.638298          0.150838   \n",
       "2017-06-30 23:00:00          0.523179          0.595745          0.139665   \n",
       "\n",
       "                     Junction 4 (H-1)  \n",
       "DateTime                               \n",
       "2017-04-01 00:00:00          0.166667  \n",
       "2017-04-01 01:00:00          0.166667  \n",
       "2017-04-01 02:00:00          0.138889  \n",
       "2017-04-01 03:00:00          0.111111  \n",
       "2017-04-01 04:00:00          0.111111  \n",
       "...                               ...  \n",
       "2017-06-30 19:00:00          0.472222  \n",
       "2017-06-30 20:00:00          0.305556  \n",
       "2017-06-30 21:00:00          0.833333  \n",
       "2017-06-30 22:00:00          0.444444  \n",
       "2017-06-30 23:00:00          0.611111  \n",
       "\n",
       "[2184 rows x 4 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_valid = Xy_train[Xy_train.index >= '2017-04-01'].iloc[:,0:4]\n",
    "X_valid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:16.412942Z",
     "start_time": "2017-11-18T07:58:16.408758Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_valid = np.expand_dims(X_valid.values, axis=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:16.536919Z",
     "start_time": "2017-11-18T07:58:16.414098Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_pred = regressor.predict(X_valid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:16.540641Z",
     "start_time": "2017-11-18T07:58:16.538119Z"
    },
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# We rescale y in the integer count range\n",
    "# To do that we must first reconcatenate with the X data as scaler expects a shape of 8\n",
    "\n",
    "y_pred = scaler.inverse_transform(np.concatenate((X_valid.squeeze(), y_pred), axis = 1))[:, 4:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:16.547786Z",
     "start_time": "2017-11-18T07:58:16.541795Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  65.91468313,   25.39552939,   23.70431116,    7.06105739],\n",
       "       [  61.52063251,   23.44733906,   17.40749882,    6.80152202],\n",
       "       [  48.55064473,   23.78885856,    8.41054807,    5.18977618],\n",
       "       ..., \n",
       "       [ 107.75789869,   36.12657791,   22.51998513,   15.98233831],\n",
       "       [  91.32769555,   29.06533414,   23.49875508,   12.69209504],\n",
       "       [  89.90044576,   28.50936747,   19.0962181 ,   13.76738727]])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:16.558781Z",
     "start_time": "2017-11-18T07:58:16.548838Z"
    }
   },
   "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>Junction</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-04-01 00:00:00</th>\n",
       "      <td>63.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-01 01:00:00</th>\n",
       "      <td>53.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-01 02:00:00</th>\n",
       "      <td>41.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-01 03:00:00</th>\n",
       "      <td>39.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-01 04:00:00</th>\n",
       "      <td>31.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 19:00:00</th>\n",
       "      <td>105.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>11.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 20:00:00</th>\n",
       "      <td>96.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>30.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 21:00:00</th>\n",
       "      <td>90.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>16.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 22:00:00</th>\n",
       "      <td>84.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-06-30 23:00:00</th>\n",
       "      <td>78.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2184 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Junction                 1     2     3     4\n",
       "DateTime                                    \n",
       "2017-04-01 00:00:00   63.0  25.0  23.0   6.0\n",
       "2017-04-01 01:00:00   53.0  28.0  15.0   5.0\n",
       "2017-04-01 02:00:00   41.0  18.0  16.0   4.0\n",
       "2017-04-01 03:00:00   39.0  18.0  11.0   4.0\n",
       "2017-04-01 04:00:00   31.0  15.0  10.0   1.0\n",
       "...                    ...   ...   ...   ...\n",
       "2017-06-30 19:00:00  105.0  34.0  33.0  11.0\n",
       "2017-06-30 20:00:00   96.0  35.0  31.0  30.0\n",
       "2017-06-30 21:00:00   90.0  31.0  28.0  16.0\n",
       "2017-06-30 22:00:00   84.0  29.0  26.0  22.0\n",
       "2017-06-30 23:00:00   78.0  27.0  39.0  12.0\n",
       "\n",
       "[2184 rows x 4 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_truth = train[train.index >= '2017-04-01']\n",
    "y_truth"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:17.029466Z",
     "start_time": "2017-11-18T07:58:16.559941Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f61f0052d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f61f0077550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDAaDwZAUIyAMBoPBkBQjIAwGg8GQlP8HpY4UsSoOzlUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f61f071f390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzsnXeYFFXWh3+nu2cYMhJEBBFMoAgM\niAoqyiq6riKrqy6r6KeriGFdXQPKqruCqy4iuogJURETIAYEFJAkSTKSg8Qhh2FgApO7+3x/VFXX\nreqq6uqe7p6Znvs+zzzTlW7dSvfcE+65xMyQSCQSSc3FU9kVkEgkEknlIgWBRCKR1HCkIJBIJJIa\njhQEEolEUsORgkAikUhqOFIQSCQSSQ1HCgJJGETUk4h2ENFJIupDRC2IaDERFRDRa0T0LyIaXdn1\nrEyIqBcRbarselhBRB8R0XOVcN5ZRNQ/2eeVVByS4wiqN0R0UlisA6AUQEBdfpCZv4yhzAUAJjHz\nu+ryUADnA+jHFXhhiGgxgG4A/MLq3zHzyljLTAZE5ANQDqAtM2dVcl16A/iImdtUcj1eBtCKme9N\n4jlrAdgIIK2yrz/V8FV2BSQVg5nrab+JKAvAAGaeY7c/EfmY2W+3XeVMAJtMy5srIgQEHmLmcbEe\nTEQeAGDmYBzqIqleDAZwCEDryq5IqiFNQykOEb1MRF8R0QQiKgBwFxH1IKJlRJRLRIeIaBQRpan7\nZ0H50GaopqHPAfQH8Jy63Estc5xwjivV8vKIaB8R3R1DPa8golVqGSuI6FJh22Ii+g8RLQVQCKA1\nETUiok/U+u8nopc0IaEe8yARbVXNWRuJqLO6/gUi2qWu30REfYVjziOihWodjhHReHXTQvX/JvUe\n3EpEvdV7pR27n4ieJKIN6vET1B6stv2fRHSYiA4Q0QNExETUxuZeDCCiLWoddxLRAHV9QwDT1Os/\nqf6danH8F0Q0RChrvrDNJ55b3XcUEc1Qz7eUiNoK+3ckojlEdFyt/zNE1AfAMwD6q3VYLTyne9Xf\nHiL6NxHtIaKjRDSOiBqo285R6/B/6n3LJqLBVvdCqMc5APoBGO60nyRGmFn+pcgfgCwAvU3rXgZQ\nBuAmKIK/NoCLAVwKRSM8C8A2AI8Kx+wH0EtY/gLAEFOZ49TfbQEUAPizWl5TAJk29VsM4F6L9U0B\n5AG4Qy3jLgA5AE4RjsuCYp5KU/f5AcB7UMxhpwFYDeB+df87AOwDcBEAAnAegDPUbX8G0EK9F3cC\nOAmgubrtawDPqtsyAFyurvcBYABthDr3BpBlumfL1Lo0Ue/pAHVbHwAH1frXBTDBXJ7pftykPhcC\ncDWAYgCdrM5rc3zoeQEYAGC+sM1wLeq+x6CY7NIAfAXgC3VbQwBHADwOoBaABgAuMb8DVs8XwED1\nHrQFUB/AFACfqNvOUeswWr3PXaGYNM91uKaZ6n2JeP3yL/o/qRHUDBYz8zRmDjJzMTOvZOblzOxn\n5l0AxgC4Ksay7wIwk5knqeUdY+a1Dvu/p2oiuUS0Ql13E4BNzDxBLeMLALsA3CgcN5aZtzBzOYDm\nAK4B8AQzFzHzYQAjAfxF3XcAgGHMvJoVtjHzPgBQ63lIvRfjoQiYbupx5QDaAGjBzCXM/EuU92Ik\nMx9m5hwogipTXf9nAB+r9S8EMNSpEPVZ7VLrPg/AXAA9o6xLNHzDzKvUe/ulUO++APYx81vMXMrM\n+cy8wr4YA/0BjGDm3cxcAOA5AHeKWhsUYVXCzL9CMUV2tiqIiG4H4GfmabFcnCQyUhDUDPaJC0TU\nnoh+VFX9fAAvQemVx8IZAHZGsf8jzNxI/btEXXc6gD2m/fYAaCksi9dwJpQe6hFNqAB4F4qAcKwT\nEd1LROuE49pDv/anoPSKV6kmnnuiuC4AOCz8LgKg+W9ON9Xf8Dws6tiHiJar5phcANch9ufjBrt6\nnwFgR4xlmp/pHgDpAJppK1QBbnXeEERUD8B/oWglkgQhBUHNwOzk/QBK9MU5zNwAwL+hmCFiYR+A\nsytQN0Axm5xpWtcawAFhWbyGfVAajsaCUGnAzJ2c6kREZwF4H8DDAJowcyMAW6Feu6opDGDmFgD+\nBmCMai+vqJP8EIBWwvIZdjsSUW0A30Bp/JqrdZwF/flEW5dCKOYzjdOiONbp2Uaqh/mZtoZiosyO\n4vyAIqhbA/iFiA4DmATgDLUTY3sfJdEhBUHNpD4Um3whEZ0P4MEKlPUFgOtVB6qPiJpqjtko+AFA\nByLqp5ZxJxQ78nSrnVUzzwIAI4iogeqYPIeIrlR3+QjAM0TUhRTOVRuNelAasGwApDph22vlEtGf\niUjTQnLVfQPMHIDiszgryuvSmATgfiJqR0R1APzLYd9aUHrO2QACqmP2GmH7EQBNiai+y3OvA9BJ\ndfrWBvBiFPWeCsUx/SgRpav3WtPijgBoQ0R2HYgJAJ4kojZqXV8BMIGjj/ZaC0UQZKp/D0IRMpnq\nf0kckIKgZvIUgHugOHk/gOIgjAlm3g3Fxv8sgOMAfgXQMcoysqHYo5+F0uA+AaAPMx93OOwuKI7X\nzQBOQHH0nqaWNwHAa1CuKx/Ad1Acz+sBjAKwAkovvT2A5UKZlwJYSUSF6jF/Y+a96rYXAYxXTUp/\nivL6pkHRRBYC2A5A8z2UWuybq17/ZCj38zYoglLbvhHAtwCy1LqERQ2ZytsM4FUA8wH8Bj0Cyk29\n8wBcC+BWAEehOH81X9JXUATWccHXI/Khus8iKP6eAsRg3lF9Roe1PyjPOqAuByIdL3GHHFAmkSQZ\nIuoIRWDWiqGH7Kb88QA2MvOr8S5bkppIjUAiSQJEdItqXmkCYBiAKQkSAj4A7QDsjnfZktRFCgKJ\nJDn8DUq8/nYAJepyIjgGxYwzOUHlS1IQaRqSSCSSGo7UCCQSiaSGUy2SzjVt2pTbtGlT2dWQSCSS\nasXq1auPMXOzSPtVC0HQpk0brFq1qrKrIZFIJNUKIjKP2LdEmoYkEomkhiMFgUQikdRwpCCQSCSS\nGk618BFIqh/l5eXYv38/SkpKKrsqkgqSkZGBVq1aIS0trbKrIkkQUhBIEsL+/ftRv359tGnTBvZ5\nySRVHWZGTk4O9u/fj7Zt20Y+QFItkaYhSUIoKSlBkyZNpBCo5hARmjRpIjW7FEcKAknCkEIgNZDP\nMfWRgiBF+WHbD9ifv7+yqyGRSKoBUhCkKDdNuAmXfnRpZVejUvF6vcjMzMSFF16Im266Cbm5uTGX\n1aZNGxw7dixs/auvxpbpedGiRejQoQMyMzNRXFyMQYMGoUOHDhg0aBBGjx6Nzz77LOa6SiTRIp3F\nKczBgpo9gVPt2rWxdu1aAMA999yDd999F88//3xcz/Hqq6/iueeeC1vPzGBmeDzWfa0vv/wSTz/9\nNP76178CAD744ANkZ2ejVq1aca2fROIGqRFIagQ9evTAgQP6FMivv/46Lr74YnTq1AkvvqjP3njz\nzTfjoosuQocOHTBmzBjHMgcPHozi4mJkZmaif//+yMrKwvnnn49HHnkEXbt2xb59+/Dwww+jW7du\n6NChQ+g8H330ESZNmoSXXnoJ/fv3R9++fVFYWIhLL70UX331FYYMGYIRI0YAAHbs2IHevXujc+fO\n6Nq1K3bu3JmAuyOp6SRMIyCiDCjT4tVSz/MNM79IROOgTHeXp+56LzOvTVQ9JJXPP2b+A2sPx/cR\nZ56WiZHXj3S1byAQwNy5c3H//fcDAGbNmoXt27djxYoVYGb07dsXCxcuxJVXXomxY8eicePGKC4u\nxsUXX4xbb70VTZo0sSx32LBheOedd0JaR1ZWFn777Td88skneO+99wAAr7zyCho3boxAIIBrrrkG\n69evx4ABA7B48WL06dMHt912GwCgXr16oXKGDBkSOkf//v0xePBg3HLLLSgpKUEwGPe5bCSShGoE\npQCuZubOUCaavp6IuqvbBjFzpvonhYAkIWi99SZNmuD48eO49tprASiCYNasWejSpQu6du2KrVu3\nYvv27QCAUaNGoXPnzujevTv27dsXWu+WM888E927dw8tT5o0CV27dkWXLl2wadMmbN682XVZBQUF\nOHDgAG655RYAysCuOnXqRFUficQNCdMIWJnx5qS6mKb+yVlwaiBue+7xRvMR5OXloU+fPnj33Xfx\n2GOPgZnxz3/+Ew8++KBh//nz52POnDlYunQp6tSpg169ekUdP1+3bt3Q7927d2PEiBFYuXIlTjnl\nFNx7771RlScnjZIki4T6CIjIS0RroUydN5uZl6ubXiGi9UT0PyKy9I4R0UAiWkVEq7KzsxNZTUmK\n07BhQ4waNQojRoxAeXk5fv/732Ps2LE4eVLppxw4cABHjx5FXl4eTjnlFNSpUwdbt27FsmXLIpad\nlpaG8vJyy235+fmoW7cuGjZsiCNHjmDGjBlR1btBgwZo1aoVvv/+ewBAaWkpioqKoipDInFDQgUB\nMweYORNAKwCXENGFAP4JoD2AiwE0BvCszbFjmLkbM3dr1izivAoSiSNdunRB586dMXHiRFx33XW4\n88470aNHD3Ts2BG33XYbCgoKcP3118Pv96NTp07417/+ZTDx2DFw4EB06tQJ/fv3D9vWuXNndOnS\nBR06dMB9992Hyy+/POp6f/755xg1ahQ6deqEyy67DIcPH466DIkkEkmbs5iIXgRQyMwjhHW9ADzN\nzH2cju3WrRvLiWmig4Yqo0H5xcoxL2zZsgXnn39+pZxbEn/k86yeENFqZu4Wab+EaQRE1IyIGqm/\nawPoDWArEbVQ1xGAmwFsTFQdJBKJRBKZRA4oawHgUyLyQhE4k5j5ByKaR0TNABCAtQAeSmAdJBKJ\nRBKBREYNrQfQxWL91Yk6p0QikUiiR44slkgkkhqOFAQSiURSw5GCQCKRSGo4UhBIUhYxDfXtt99e\nocFY8+fPR58+SpTz1KlTMWzYMNt9c3NzQ7mGokFMNify/fffR5WaQqO0tBS9e/dGZmYmvvrqK0Pq\n6wMHDoTyHEkkUhBIUhYtxcTGjRuRnp6O0aNHG7Yzc0xJ3Pr27YvBgwfbbo9VENjhJAj8fr/tcWvW\nrEF5eTnWrl2Lfv36hVJfr127Fi1btsQ333wTtzrWBMoD5Ri5bCTKA9Yjyd3w8+6fsWx/5BHryUYK\nAkmNoGfPntixY4dlquhZs2ahR48e6Nq1K26//fZQ6omZM2eiffv2uOKKK/Ddd9+Fyho3bhweffRR\nAMCRI0dwyy23oHPnzujcuTOWLFmCwYMHY+fOncjMzMSgQYMA2Ke9fuWVV9CuXTv07t0bv/32W1i9\nlyxZgqlTp2LQoEHIzMzEzp070atXLzz33HO46qqr8NZbb2HatGm49NJL0aVLF/Tu3RtHjhzB0aNH\ncdddd2Ht2rXIzMzEBx98YEh9nZWVhQsvvBCAkp316aefRseOHdGpUye8/fbbCXsO1Zn3Vr6HJ356\nAiOXxZ476+rPrkaPj3vEsVbxQU5MI0k8//gHsDbOSWYzM4GR7j5Iv9+PGTNm4PrrrwcAQ6roY8eO\n4eWXX8acOXNQt25dvPbaa3jzzTfxzDPP4IEHHsC8efNwzjnnoF+/fpZlP/bYY7jqqqswefJkBAIB\nnDx5EsOGDcPGjRtDaaXt0l7XrVsXEydOxJo1a+D3+9G1a1dcdNFFhvIvu+wy9O3b15CyGlC0jgUL\nFgAATpw4gWXLloGI8NFHH2H48OF444038NFHH2HEiBH44YcfAABLly4NlZOVlRUqa8yYMdi9ezfW\nrFkDn8+H48ePu3sGNYz80nwAQF5pXoQ9qx9SEEhSFi0NNaBoBPfffz8OHjxoSBW9bNkybN68OZQH\nqKysDD169MDWrVvRtm1bnHvuuQCAu+66y3Kimnnz5oWmlfR6vWjYsCFOnDhh2EdMew0AJ0+exPbt\n21FQUIBbbrkllFq6b9++rq9NFEz79+9Hv379cOjQIZSVlaFt27auywGAOXPm4KGHHoLPpzQHjRs3\njur4moKSDCE1s8JKQSBJPC577vFGnKpSREwVzcy49tprMWHCBMM+a9euDX34FcUu7fXIkSNjPod4\nDX//+9/x5JNPom/fvpg/f75hYhu39YvXtaYyhNS9R9JHkIKkYo8lUXTv3h2//PILduzYAQAoKirC\ntm3b0L59e+zevTs0NaRZUGhcc801eP/99wEotvb8/HzUr18fBQUFoX3s0l5feeWVmDx5MoqLi1FQ\nUIBp06ZZnsNcnpm8vDy0bNkSAPDpp59GeQeA6667DqNHjw45nqVpqOYhBYGkRtOsWTOMGzcOd9xx\nBzp16oTu3btj69atyMjIwJgxY3DjjTfiiiuuwJlnnml5/FtvvYWff/4ZHTt2xEUXXYRNmzahSZMm\nuPzyy3HhhRdi0KBBtmmvu3btin79+iEzMxO33norevbsaXmOv/zlL3j99dfRpUsXyzmLhwwZgttv\nvx09e/ZE06ZNo74HAwYMQOvWrdGpUyd07twZ48ePj7qMmgSn4PxaSUtDXRFkGuroYGZ4XlJkvExD\nLYkH8nkCry56Fc/Pex6DLx+M//b+b0xlJDs9fKWnoZZIJJJUQvoIJNWKVFRdJZKqQip+X1IQSBJG\ndTA7SiIjn6NCKkdWSUEgSQgZGRnIycmRjUg1h5mRk5ODjIyMyq6KJIHIcQSShNCqVSvs378f2dnZ\nlV0VSQXJyMhAq1atKrsaVYZU7NxIQZCCVIUXNS0tLeoRrhJJVUZzFksfgUQikdRQpI8gBogog4hW\nENE6ItpEREPV9W2JaDkRbSeir4goPVF1kEgkEklkEqkRlAK4mpk7A8gEcD0RdQfwGoD/MfO5AE4A\nuD+BdZBIJJK4UhVMr/EmYYKAFU6qi2nqHwO4GoA2I8anAG5OVB0kEokkXiTbR7B8/3J8venrpJwr\noc5iIvICWA3gHADvAtgJIJeZtWmV9gNoaXPsQAADAaB169aJrGbKkYrOLImkskm2j6D7x0qqdO6Q\n+O85oc5iZg4wcyaAVgAuAWCVrMTyKpl5DDN3Y+ZuzZo1S2Q1JRKJpEaTlKghZs4FMB9AdwCNiEjT\nRFoBOJiMOkgkEkk8kD6CKCCiZkTUSP1dG0BvAFsA/AxAm3PvHgBTElUHiUQiiRepnHQukT6CFgA+\nVf0EHgCTmPkHItoMYCIRvQxgDYCPE1iHGkkq9lgkkqpCKvrgEiYImHk9gC4W63dB8RdIJACAJ396\nEjN2zMCWv22p7KpIJLbIOYslkgTyv2X/q+wqSCQRSWXTkEwxIZFIJDUcKQhSkFS0YUokVYVU/L6k\nIJBIJBIXpLKPQAoCiUQicYHmIygoK6jkmsQfKQgkEokkCj5Z+0llVyHuSEGQgqSi6iqRSBKHFAQS\niUTiAjkxjUQikdRw5DgCiSQJSJOWRFI5SEEgqTIEOVjZVZBIbEnF8QMaUhCkINX1hZWCQFKVSeX3\nUwoCiUQicYEUBBJJEqiumoykZpDKPiwpCCQSicQFqdxRkYIgBUnlnotEUllI05BEkgSkAJNUZaQg\nkEgkkhpOKndUpCCQVBlS2QYrqf5IjSAGiOgMIvqZiLYQ0SYielxdP4SIDhDRWvXvhkTVoaYiG1SJ\nJP6ksiBI5JzFfgBPMfOvRFQfwGoimq1u+x8zj0jguSUSiSSupHIHK2GCgJkPATik/i4goi0AWibq\nfJLqTyrbYCXVn1TWCJLiIyCiNgC6AFiurnqUiNYT0VgiOsXmmIFEtIqIVmVnZyejmhKJRGJL64at\nK7sKCSPhgoCI6gH4FsA/mDkfwPsAzgaQCUVjeMPqOGYew8zdmLlbs2bNEl3NlKK69qxTWfWWVH9O\nrXsqACDztMxKrkn8SaggIKI0KELgS2b+DgCY+QgzB5g5COBDAJcksg4SiUQST6prR8uJREYNEYCP\nAWxh5jeF9S2E3W4BsDFRdZBIJJJ4kYoCQCORUUOXA7gbwAYiWquuew7AHUSUCYABZAF4MIF1kFQj\nUvlDk6QOqWjCTGTU0GLAcm636Yk6p0QiqXyYGU/PehoPXPQA2jdtX9nViTup2GGRI4tTkOraY6mu\n9ZYY2Zu3F28uexPXf3F9ZVclrqTy+ykFgUQiiStag5mqDWcqXpcUBBKJRBIF0jQkkSSQVPzAJKmH\n1Agk1QLZoEok8SeVvyspCCRVhlTsaUlSh/QT+eAhwG2LjsetzMlbJoOGEnYc3xG3MmNBCgKJRJIQ\nyDJ6vPpS58BRAMDtv+TGrcyJmyYCAFYfXB23MmNBCgKJRCJxQVCVa5QAzbWytWEpCFKQyn6pYiWV\nbbA1iVR/jhTHy6sqWpMUBBKJROIGUhrteAqCqoIUBJIqQ3XVZFINf9CPL9Z/gbySvMquiit+2PYD\nSv2llts+XfspZu+cbbktakKCIP7vaWXfaykIJBKJgbm75uLuyXfj1UWvVnZVIrJ472LcNOEmDJ4z\nOGzbTzt+wr1T7sV1X1yH8kB5hc8VVDsq8TTm5BTnAAAe+vEh231WHFgRxzNaIwVBCpLqNlpJYskr\nVXqnu3J3VXJNInO8WAnl3HliZ9i27KI4z2yoagTxVFztNBmRE8Un4ndCG6QgkFQZpABLLYgS7wjV\nnK1WZsW4O2I101B8S60SSEEgkUhSElEQxcP/FAofjWfUkAthmQyBKgWBpMogncVVi6oS2uiE1kha\naZMJ0wjiqLlWlXssBUEKIhtUSTyoDu+Rk2lIJC5mxwT0zJPR23dDREFARK2JKENYrk1EZyS2WhKJ\npLqSTAHiqBHEuZFltThPMH5lutEIkqE1uNEIvgMgXnoQwLeJqY6kJiOdxVWLqmK2cMKtszgewkkr\nIZ4pJqqNRgDAx8xl2gIzlwKoFekgIjqDiH4moi1EtImIHlfXNyai2US0Xf1/SuzVl0gkksgaQVxN\nQ0nur1QVZ3EOEd2gLRBRHwBu8rD6ATzFzOcD6A7gb0R0AYDBAOYy87kA5qrLEkm1sElL3JMMjSJk\nGqqm4aNVRQv2udjnYQDjiehddTkbwF2RDmLmQwAOqb8LiGgLgJYA/gigl7rbpwDmA3g2qlpLHKkq\nL5ekeuIP+iu7Cq5xauzjHj6qjSyO4+cVZN3qXlxejAxfRpgGUCV8BMy8jZm7AegCoAszX8LM26I5\nCRG1UY9fDqC5KiQ0YXFqtJWWSCSJo/93/QFUHfu1G6pr+KhInVfr4I2lb1icNvHPwVYjIKI7mHkC\nET1mVSlmHuXmBERUD4pz+R/MnO/2oohoIICBANC6dWtXx0iqN1KTkUSLo2ko3j6CBLyf5npP2DgB\nT1/2dNzPEwknjaCR+r+ZzV9EiCgNihD4kpm/U1cfIaIW6vYWAI5aHcvMY5i5GzN3a9bM1ekkEkkN\nIxQ1lAyNQCs3jvLAXO/K6gw5+Qi0sQJrhEbcNaSI448BbGHmN4VNUwHcA2CY+n9KtGVLnKmuTtfq\nWm+JkWQ2Zk4Whnj7CLQSPPEUBKZ6JcXpbYGTRnATEXkBvBBj2ZcDuBvA1US0Vv27AYoAuJaItgO4\nVl2WxJlTioC6kRMbSiQJo7KTzonEQzhpvoFEOYuBqqkRzIYSJlqXiMRwUQLAzNzYqWBmXgz7SKtr\noqqlJGqODwcO1wVQ9VPKS2oYs3bOQvum7dG6Yey+v+X7l6NOWp3k5hpSSQ8krrFed2QdghyEh5Kb\n/cfpbM9A8RP8BKNvoClc+ggklctphcCe3D2VXQ3XSGdx1SJRDenvv/g9Or7fsUJldP+4OzqN7hRa\njugsjodpSO291yuJX44Jq3d+7JqxhuXKHlC2nJVaZjNzwPyX8JpJYkZ8uUr8JZVYE4nEmvzS/LiU\nUxnO4ow4DrOwElDaZDvJxMk0VIuI+gPoSUR9zRuZeWriqiWpiUhnsSRaXDuL4+IjqHARYZh9BACS\nbhYCnAXB36CMIG4E4HbTNoYS/SORSCSVTmVF2yQCsyBIxnXYCgJmXgBgARGtYuYPEl4TiUSSEiQ1\nDbWTaSjePoJgHPNPO1AZAsxNiokPiKg9Ef2JiO7U/pJROUlsVFcTi3QWV022HtuKbmO6Ibckt7Kr\nEobbpHNV9d2qKvVyMzHNCwDGABgN4A8ARgK4LcH1kkgkVYShC4Zi9aHVmL59emVXJQy3GkFVxcpH\nYBZqlR01pNEPwO8AHGLmuwF0hruspZIqQHX4GDSqqyaT6qR50gAA5YHyqI6rSmmo4/NuJT7XEFA5\nWoIbQVCshov6iag+gMMAzkpstSTxoqqonpLqS0gQBN0JgqSmmEimPT1J1xWmEVSms1hgDRE1AjAW\nwCoA+QB+TWitJBJJpaP1ttO8iiBwO09BZWh2SZmhLAFUlXo5agRq4rghzJzLzO8CuBHAg8z8f0mp\nnSQmxJerWpmGkvxRHC8+jkBQjo2MhM+j9Ber0oQ1074Ejr2WXNMQB8PL2JO7x9LO75ac4pyKVClu\nOAoCdWTxD8LyDmaW2oCk2pNbkosmw5vg2TlycrxIaILArY/ASaDHS9j32Q40KXY/Q1ki2Jy9GW3e\naoPhvwx3fczAVcD/Zii/S/2l2Ju3N2yfJrWbGJarirN4BRF1TXhNJDWeZJoUThSfAAB8s/mbpJ2z\nuqINcIpPKufEPONIKSbimX1U41DBIQDA7F2zXZfxwQ/AP5Yrv8sCZaH1Tes0Df1u1aBVBWoZG04z\nlPmY2Q/gCgAPENFOAIXQs49K4SCptlQnk1lloTWkTiGaVjg19oky/0VKOpdIYr0msc4VMS/FAydn\n8QoAXQHcnKS6SOKEDMN0j7xXkXGyw0dLpWkEccw+Giq/goJG9E+Jvxls9PNVctQQAQAz70x4LSQS\nVJ0ICok1rjUCh/3i3fPVGvjK0Agq+r4GhCTO4n1hZsO2ZODkI2hGRE/a/SWthpKEsebQGtw4/kaU\n+q2nMjtWdAyfrPkkybVKHDlFOaFc79GaO2oybmcB09D2s2qIY73f2YXZju/iqoOrUFReZF+nGM87\nc8dMbDiyAYBxlq28kryYyhMxaASmhr+wrBBNC4Hf7arwaVzhJAi8AOoBqG/zJ6kGOKmVXcd0xfTt\n0/Hfxf+13N7vm364b+p92HF8R6KqZyDRZpo7v7sT90+9H1uPbZU+AhdovVSnWcCiJdZnfOukW3Hf\n1PuQlZtlLE+o07Oz7SPAYj2A38EyAAAgAElEQVTvH778Q2gCHPFcD/7wYMVNQ3YaARg7ju/AnM+A\neZ+FO6kTgZNp6BAzv5TwGkjijvjCuvkAsguzLdcfLDgIwBjdUJ05fPIwACVsr05anUquTdUnJAii\n1QgSED4qPjtDeUKdsous3+O4IdT9aOHRChcnjsswm8yCHETHI+ppk5D11EkjkF2mGoLdQCHto62u\ned3dIJ3F9phNPJWpEbhxWDv5HxJpAoz1mmydxcyGayGLgWzxxkkQVGiCeSIaS0RHiWijsG4IER0g\norXq3w0VOYckMt68goj72OWQcbL1xpNTioCrdqdwjppqCA8BHnxvJYDYfQSW2+L8jMXyzIIg3ucS\nZyhjcIXfISfTUJCDYLV4TsLod1tBwMwVnThzHIDrLdb/j5kz1b+ql9c2xWgwcXLEfWwFQZI0gikT\ngfmfAlRUnNDzWCGdxfb0nr8HQHw1gkTGyztqBFUw9NUcMmrYJgiJyjYNVQhmXggg+bMwS6J+Ye1S\nByRLI2h3TP1xMrL2Ei/iGRvvhnWH16HZ683iYluOBwcLDqLp8KbYdHQTAKVRavdOO3y96euwfePq\nIzCV8ej0RzFw2sCIZdpFeTkNyor7s41jp+GDVR/Yhoje8e0d6PlJz1Dtk+EsTv4sycCjRLReNR2d\nYrcTEQ0kolVEtCo7O8FOoBSGXTTidj2pZGkEhenKf08SNYJkm4ZeX/I6jhUdw087fkrqee2YsnUK\ncopz8PaKtwEAReVF2JazDfdNvS9s32g1glAHwuIem8t4d+W7+PDXDyOWWaV9BDGU/dCPD9nWN780\n31h+ZZqGEsT7AM4GkAngEIA37HZk5jHM3I2ZuzVr1ixZ9Us9XLR3toIgSRpBqVf5TyXW4xniRVUw\nA1WVsFVz425oYE23KVqNwImYncV2GoFLH0FctIN4vj+MiJlvNR9BZTuL4w4zH2HmACtjtT8EcEky\nzy+xprI1Aq30pE0OXgl56qtadJLdMyUQvKbHEK3wSkT4qF0dxPuazNG4FU317uHI9dXOkHIaARG1\nEBZvAbDRbl9J8rBrpJKlEegnTFKjXAmaQVUNxQ2laBDuiddOI4jivt22CTjnULiGV1GBGI1pSNw3\nLs/cpoxYyvYGIzvONY0A1dlZTEQTACwF0I6I9hPR/QCGE9EGIloPZR7kJxJ1/pqMQSVmxq4TzuPU\nK10jUKub6F6zQRNwyFGTCOIhVHef2B23qButHrkluQCM9bPTCPxBf9jIXo3i8uLQAEQG4+uvgen/\nDc+179RoFpcXY0v2Fuv6xmAaipWcopywFBJHTh4xnhccui9HC4+iuLw4lJbaDV43piH1f7U2DTHz\nHczcgpnTmLkVM3/MzHczc0dm7sTMfZnZ/Z2TxMT8PQtw9qiz8esh+/mE7F7IZDSWKw+sDP1ORnSE\nRrK1gooK1c3Zm3HWqLPw2uLX4lIfrR5fb/46bL2dRvDyopfR9q222Hk8PA/lDeNvQMs3WwIw3tut\nx7Ya9nNqrB+b8RgueO+C0FwRhjoIzmK7yCvzexyrj6Dp603ReHhjw7rT3jjNoBGI17Hl2BZc98V1\nOP3N012fw41GoMGBFDMNSZKPlifI6uPViKQRJLLR3JazTW8ak+QjEEm2QIhVI9iTq8T0L9izICH1\nMJiGIvgIDhQcCCtvftZ8y/OYZ+ByapC/3fItAKCwvDC8voIAPV6sR6XHI6TVikiNdCAYMJS5eO/i\nqMr3BcN9BNfsBH78AmiuRlFnqJsbL1qJRCMFgSRi1FAiNQKDHTeJgiDZztuKni/R4x7EBHM+syCI\nQosJcjDmkcXaTGhOJpNoxigkchxBRc1QXtbLuHYH4AkCcz4HbtgBzPiyQkXHhBQEKYihcaXwdWbs\nXmptfaJ7zedqHTxpGrIl3mmzzfUQ3w+zaciMUx0izWvs9B5qws7qfRQFod0UlNEIiVgwRyhVpExv\nUBF4gxcBs74ABgsKRUvjMAIEPYlvpqUgSHHYRbtjGzVkFWMeZwwfb4IbZ6uGoro4i+OtEZjrIWYa\nNZuGohmxWxYoc2yczcut8oABq5XfIY3AIqzSjSAM0wiYcf12gIcA3iMVH5Qq5hoym4aiRdMI/jtX\nWe4guD1KfTCM5Qj6pCCoMszaOQtzd82t7GrYMnrVaMuIjuzCY4blj3/9GNtythnWRTQNCS/8ybKT\nGDxnMA7kh9uJK0wwCGbG8F+Gh+zAby9/O+7nIiLLa0skVU0j2H1it2FZLPfG7cp/7a14b9V7hn3N\nETRi/UoD4SGjP+/+Gb3G9UJuSS6OFR1DRjmQria8nfsp8OE0AHl58JIystDKNLTuyDqlnubGPhhA\n9mvA/auVaxi5bGQoemdP3h4884uyn2/LbwCAbzd/ixUHVhjKyCvJw1M/PRXm2Daz6uCq0O+Kjlnw\nBoE1h9eElusIilS5B2hUoi+z11uhc7lBCgKX/P6L36P3570ruxqWFJYV4uEfH8ZV464K29bquB88\nBGj6q/KSD5g2AF0/6GrYJ6KzWPj4ftj2A1775TW8vuT1ONXecEIs2rsIz855Fg9MewBZuVl4bOZj\n+OPEPybgVJUzwCtWjUDrLcdLI3h50cuGZdFH8OE0fX1BaUFYlM4d394RVp7Po0xtomgExvfp6s+u\nxoI9C/DYjMcwYNoAFL8C7BilbDvtpL6fk2lIxHAPy/1oWgx8NE1x2D7x0xPo900/AMrkMQ3VBjXY\nsAEA4Lavb8OlH11qKG/p/qV4c9mbeHH+i47nnb5dz5EZCFbQNMTAoNmDQsutBHMQweiwD3qlRiBx\ngfZCHitSev9iI3e12vFrM3l+aJ05KiMajUCbGESM3IgXzByaBCe3JDc0T4IW6x5vJk0C+q1KTn6j\neDmLE5W90zwJDaA0Dlbns7oWr0fvzQdtnL35pfmh0NAz1IbPEwqWJ0fTUOjcZgFuESKaV6qPAdAc\n35xmPweXdo3RvGcBrqBpyHRbuwmB9B6GwWEfyEiP+TxukYIgBXAyG4R8BOpkF0eGA99NNO4TeWCL\nXm4iRxlTMGi4lnj32jPKgnhpHkClZWAwbt8MvD3pZOQD40hVMQ2ZcRJUvgDw8lygQYntLqFG3B/0\nA34xhbIptt90npAgCARCZTgJuyAHjffQ4t0Vt3dSlRk3aS+szttjL3Blllqu6CMI+MFg3LNGD/eM\nBi8DPbOst/1wnlEQHO+RGf0JokQKghTAyZEorglyEKcWAbeYTKGRTEOJzCFvTAMQtBz9Gy/hc/9P\nR/GvhUDjz76utKihWElW+KjVeW/fBDy/CPjvHPvjQ/Z9DiAY0Ge8O/Nz3c6kTbgiEhIEwaBj+Ojl\ne4D22eH1tBp1q92reqK7wuH+a2Va3YMlY4EF48KPaZVdivTs4xg3BZg2QTyNu+fjDRr9AiKbmin3\nO1RmElK8SEGQAjg12KEMhhz+EYb2sYsasjANxTMTZfgJ2VB+vFNc1C5Tyyt3DnFMBEnP2xSBU4qA\nf88HSDOdODRgWjhpfYepqzXTkD/oBwuCoPYB3bFspeW50QiYGYs/Aba8a7HNNOr2+QVA151FYfVz\nStymPZtIHR4xrPb7EftDX0BrIRuF206TLwgUpVlve2868LDul05KZJsUBAlk1PJRaPxa48g7VhBz\ngy2+OG0Es6fdS5rhywhb97+l/9N9DhamIe1c7d5ph3/N+1cFaq8zd+ccQ/mxNp5XjbsKD0x9AADw\n4s8vgoYSlu1fposTIWpI4+IPL8bjMx4HoESI0VBCTlFOWNm5JbmgoWRwHMaLzdmbQUMJNJRQWGb0\n45hNQ9d9fh3u+f6eqMpvPqI53lz6JgDgnenA0PnA9crAc7R5qw0Ao30dUO6fX20lzAPNRDSNwB/0\nA0Lj/N0WfYa8adumGd7BIyePWGoE3T7sZojQE48JchBzdumqyfebvzPU4+WfgY9fV6LiPMIj3np0\ni+XzBOw7UuYJc16cr/9uUMK4/TvFaV5LmPLbqtH2veQDDSWU+HXbmpeBk4k3/btGCoIE8vjMx3Gi\nJDxvSrxxivdvLrQn5he9awsleujyMy4PO+7JWU+i7XHg1TnhvS7xXNtytoVFoEQDlem9809+HWup\ncUSrESzcsxAfrfkIAPDSwpcAKEJZK5JBYT3TVQdXYdQKJZRl2OJhAIC1h9eGlb3xqJIw99VFr0ZV\nJzfazfgN40O/9+XvM2wzm4Zm75qNz9Z9FlUdjhYexVOzngIA1FVvey3To9Uc9Bojl48MCYI0B1dS\ns2LCxK+BYO4Jg2nIfLXiO7h472JLjQAARiwdEfptNh++9oueb2n8en0YrtdUP9GmP23rFEO4pgiD\n4Q2Efx/mCXOuM+Vu1K4tXTivVWdLc34fPnk4tO6hVUBLl76FNo3auNuxAkhBkAK4iom3MA3VS6/n\neNxnk4F/LgYyNv0WWhfvTKS+Al1SEYwaR/wnH2f9RG72t9BE3Dg0rXCj3Yhlmu9zPJzFtcv0+P16\nqpln8leRj3OjETw5twj9NgGnjJ9s0AjMhDmLtR+CRgAYr1+85ibTjGN5RHPNez8azyVqBD7yIs1j\nbYvJOHAU/v8A1y9UQncoqKR8iIRWvigInJ6P6Pt4eJXRt+BE+6bt3e1YAaQgSAKJdkwyM/ptALoe\ncH577WZwsk0xoX6LnoLwyJp4XZOh58/G9fG0qzM4VD55PK7srlbXGKsgSCv147PvgNpH7TVEp3sa\nD2dx0avA5neV36eG53WzpHERcKM6/tBJENRVBYs/I92gQZKpugZhJz7XQMDQ+FsFDQBA2tFjhuXO\neicbd683nksUBGnkRZrXWhDUzVIGLF63TBl9vPRjIPCS5a6W5Ytvp9N7Yda2qhJSECSQv2xQhrcH\nDyc+2/bEb4EVY5wbCfNLesn6HEwZb9+4FKvfDRXrsfbxdnYaHdHGnm+i5kNgxC7IYhUEl604hLvX\nAx1GfGq7j20jifiFj56tyiG3pXwzCRigWlTSHC45Te01+L1kcBabsQo8ABCWedY2n5BJ2xCdqrVN\np3WrEQTSlTEGaX6lDpdaDGRvb5GhItSxEM4TKfS1qiIFgYlDBYcMtrxtOdtQWFaIL78BCqM0hQ9U\nc6gENyVmIjbOzUUwoxa8c4zqcnZh+Ft7pPBoWFjeiJGb0XcbQP5I4wj0yc0B4E+bgW177Oc3iArh\nw/awboM/WHAQReVK9Ee8hE9oAhxy17OOZBoqLi/GzB0zwyYxARQzwIYjG5TzMWPfCSWNdCnbRyw5\nmobUuuzO3R1x3Icbgg63dHlL/XdbQYERNYJdJ3YhyEGsP6J0w4NedcBbwG+I0DGfxqgB6r/zi42D\nuTzkCd1D8Zj9uXtdC8NbhTlu0uBFutfaOxtUB5uVFRfaNtZmsxMAjJ0Svs7pvZIaQTXi9DdPR4s3\n9Bk1273TDn+c+EfcuRGoE+Vz1D62RM05Ovf7/8FTWoaTLzxrWN91TNewfZuv3ho2YtOvfryiw1ZE\nbDj//PWf0e6ddqi7LQvfTgKe/OQ3PDv7WTy4ErjX2gfnDpMZ4ZHpjwBQGrzbv75d2aUC9y/dD5S9\nBFy2cHdoHZuap01HN7kuTxQEd0++G3/48g9o/264DfelBS+h0+hOWH9kPUavGo0TRcpI7MX7ltiW\n3eTAcfAQoPOhcCGknffwycN4aYFut7DK++MGJ9Eq9ojLhDQ3oiA4e9TZGP7LcHQe3RkrD6xEQM2Q\nyf5ysN/+Q6lTrG/zFevxqLd82TcsOu3lhS+j0+hOWHNIf8FmbJuOYr+uoTaqfYrtuf4sPNazGrYJ\npcEwE/ApF3nJASXPlRXmkcAAcL06xYfYiFoJkrNzgNmfAuWFMYw8SxI1WxAUFgI+H/CdHoKW7gfq\na/lJ1Ic6d3dsyeY0QRB0UJUrwm51spnCoDFNAlm8tG1zAc8qYy8+oD59j8WH2y5bDz1lQA/ZK1Fu\nTvtjwE87f8LoH4FPLHpGrgmKvWAje/L2oONhoGG+QwB7BJoVKiaNO79YDwo1NEZH9NHCo3j7R+CL\nb0NbbcvTeupBDmLe7nkAjNEgGqsOKTaLvXl7sergqtC1OfXEz16TBQB4dIW9sxgAluz7BX/ZoETJ\nWE3iYoWhFx0IGMwmoX2GABONE5YZ6pseAO5crztSVx5UJkzZm7cX7FFNV+XlBuH++HIArHxXFATu\nWSLE+AvC7qApsSCBQuWLEVQe1tOcAECr+i1hh3iNZzZsbftcgx69HnYz+Z0eoQ13Go+x422g926g\n8bcWakUVoWYLgh07lJd2yJDQqhUfAvlK9GCFVfCAphE49JAqdgKlfkzGx/jp99a7c47RZORXPwCP\nhUaw9V3gPDWdEHso1EMN1FLsrLUCMER5xIwwuMscntgqD1g/Gpj2mjDL1bp1Uc1k1mO/8t8bCBpN\nQ0LPzefx4dGVQP8NxmOtfBNi8je7HqZWJqCYA8qD5aFGycmoUZ6m9Ex9QQsfgbDca+khTPgWGLTE\nvblB7KnyunWWPVwA6GdSjsSImB77gS+/UwSVmYCaKpkD/jA7/sRvgNKXgTdmAZ6gsdev4TQrmti4\nethkQnN4F0RBwIGgvY0+aO+b0TgnQhS4Fr3k6Acojb5DUzhjatTHxEIiJ68fS0RHiWijsK4xEc0m\nou3qf3u9LhloTtAMfUBVZ0HT9lzVC3ev1QfLRIumOjo5zyoCBdSJYzzGl9ccPaERMDVD2lGZC34L\n31mAhRG/QfVDyfAD9UorHjmkXQOghNNdeARorWoiX6m901Pz1YZlxQogMxMYbq2+W/G1WkajvFKD\nIBBTDthFk1jhIQ/OOwbUKi53FASaY1ITBNq5nTQCDaV6xnsrLtcrUBqUJ5YqOW/cIPaG/aXFESef\n0Ui36AudZpGeiTXTUHl52PuuCZeHVxoFgfgM+hgzoxudxTAKArcpoA2CwJQkztBgx2FmPE2QOQqC\n8ugFAdWpE2ONoiORGsE4ANeb1g0GMJeZzwUwV12uPFQzB2rXttzs/WUJPvseaFZUMWdlokxDmu/B\nLAhs62FarlOqrPnT6AURTsR6+KJ6LbX8eshghRC0JS8DG94H9oxUls8/Zto3K0v5/2tsjupQ/j2Q\nYVpMXxSCnojw2zvA6HeyQmkVrGhUGMCVWcqMXf6yEoyaoZ7b4VEF1QbvnnWAd7/RVGJoyNUneWoR\nUOenea7qLTaC/oP7w3rgvXbDEvOAMyA8JBTQUyVn7D9k27BmBIzHig31MJP1lUjXQsXG1cPAqbn6\nO+Mkz8Rr9BUWK/eQlYPsBEGsEWqaYHU0K/oDOOm+zwEA8BQ7ZPqLIwkTBMy8EIA5V/EfAWjxc58C\nuDlR54/E8F+GY+1eVcdduBCH1yxCbrGu/4kfzvY39RcvFnMRBwLwB8qx+rqOyJ//k+U+Jf4SPDP7\nGWzJ3oKh84eGwidfWfhK2CQiGp0/VcpiYSq7VVvsG4aCYvs0u1qEzpStU/DN5m8M256e+STqHy/C\nqOkAShX7bIYf2HZ8u152qWJEnb59OiZvmYxSfykaDmuIBVnhQoaZcd+U+5Qh/Db38w/bgNqCxWrX\n8Z1APyXPvD8YwPsr3w9NFKKdE0AoisUS9ZF+/9v3KCvXP7C8se8b6qY9e802PX37dDQf0Rz+oB+z\nd8wCAHTdVWzQFIMcxAvzXgj5C/7z2PdYMA54ZEJ/dF59IBRooGkE4zeMx7zd87Bs/zIMmjUIr//y\nemhCFQCoNWU67p9yP2gohfLsP7tI0ZgCQmu6d7sQP6le/8hlI8MuXWz4at/2F0NzN3vnbPwsRLUW\nCYqOlUZg6V9Q38Gzx8/EoV32z+CBX8QZV4wFpfkZHVSN3C581MPArev173Fbh9Nsz9VdkKVdBr+F\nn7bPBA8F3vwJ6PFxDzR+rTH+8OUfEJippws5e4nz5DR23PQbkDMMGL/sw1Cdh8wfgiVCcMCiPQtt\n8wvZUpScNOnJ9hE0Z+ZDAKD+P9VuRyIaSESriGhVdnbFp5kTKSgtwL9nPIvnf9KjbU7ceiMmvKJP\nuLH60OrQ73pCgzRrjbGRdAMHA5i+cgIumr0R3htvstxnzOoxeH3J67jgvQswZMEQbM7ejL15e/HC\nzy+gz4Q+4QecPIlLtoXr6F06XGNbj7w8+wgTLQfNzV/djPs/v92wrXZJAJO+Bv6+Ami9QEm7kOEH\n8v26o1Kb1OPG8TfiT5P+hLFrxiK/NB+9Pu0Vdq7Vh1bjkzWfYPPkD5GdZz3GYvp4pQepccGb54R+\nHy85jkemP4KLP7zYcE4A6Dy6c2i//FLj5K8etUHZcnwbvlirp2dYMueT0O99+fuwKVuxZWi5fG4c\nfyOOFh7F5C2T8a8fngrte+3mUvAQ4KzjSrqEVxa9gr9O+SsAoIX6aIbMY4PzVGvS+n/XH9d8dg16\nfNwDI5aOwDNznkFZUH/R1hxdh7FrxwIAen7SEyv2LsOwuYrGFBBa8U/XGscldPmgC5746QmYMfdU\nF/dsHfp93efXme6T/ttKEFxg8TmKWumXi98L30GlqdCumQXKnQtPYOP7wMX7VY0A1hqBaNZambfZ\n9lwi9QvL8c85/wQAPLFMSSlyouQEZu6YiRs+1Rvr49vWhX7PsR/yEcbEb4HGJcDYSc8BUMKghy4Y\nisvH6ulbNh3eYClEnUhveUZ0B8RIlXUWM/MYZu7GzN2aNWsW37KP56DkFeBHPbULKBDEw6/ovXW/\nzSTc5731RdTnS5s2HX51wpWQaYAIeOCB0D7ahCwi2geg9dYNvK/3YtPKjOYVO6jE3pYjRmI0NGmj\n33wNXKn5a8tV01DAqJab61getI+X9wf9uP9XYPEnQIf57j7kekLVI8WRb3hXmbrQ7MzWTUNGc53Y\nsAY5aFu+P+jHTOHx91mrXPP4bxVNMTAEuHmq0djdoNSYKtnJRxAQjHd+CuKZxcDVan6bJsLtFRtd\nc8OivTNmzdV8TQ3r69+U2UzkYWBTeyVZopUg6LPduMxgcCyzaJnq1DFLkRJLPwZOO6pfsCjEfEFj\nkjendBZm3PhFxHDZaywU8UWtgb2N7B+idg7NiS9G8P1ud3hk3IHmdTD/r7+zLGvWWYDnkkstt8Wb\nZAuCI0TUAgDU/0cj7B8/5swBPlF6fvUu6Rm2uf1eYxhe3V+tB4F5yqO399cePwlN1ygO2XqFQgP5\n0Ue2x4iNmGXDJHwAJ1o0clUPKgufT1aj7Rr9rS93MJmTX6l/WtA0tD6K1M7MjHNVo2HTfboj4OsL\n7I/JFCI0nXwi6X7gwmxl6kIzorNYtFX/3zphJ1POGzNXCLngSmopN+rSAwCY4QHw4CRjZjIl57w7\nQSBGALHXg9fmAHNVxeWonoMNJPgmHl9uXZa5YxGW40cQTuYRw+lBoLi2YsNw7FgYZjMTnbv2TD1P\nqINpx3Ifhc750gvzkFYWwIifAF+BLhSeW2wcQVy71EYQWHwzdpFSIqUR3EWlXmc/jzbWQnNoDxDc\nWdftCr9m9hCCPuuT/trCcnVCSLYgmArgHvX3PQAqEoEeHddeC9x3HwDAs39/xN2bfzHZcn2sbuOz\nvnFw6u3Zg6cvH4QdbwGdDgMdjij53d2OqM1u3dRyfW4t43K3f75tW8a9g/Qsjk7qKwmCUNzv1nFK\ni9S0ECE7rxPaoeR35/gTJ/EIOux5imB6CJ/WUP8nvvhnCa6TupN/dHQCi5TWEvazcZD222RMYHah\nQ9fHL0TDpAkN3IWm+ynGvXewsZqaJ5E3R7O03K2nZK5v0T8ore2cI3n9e4DPr5fpFe6os/DQMb9n\nZV59a/2CUvReuA9PLQU6jTZ+i5cKn++AeeGjugEAI8P9JG40gt/vdN5e6tMj56yYoFqOtfvdyKRd\nm+81E2wnp3cSOPEmkeGjEwAsBdCOiPYT0f0AhgG4loi2A7hWXa6SpB8xh6xEweuvA9uMJgJyMmfM\nUEJKzj4BrBsNbHwfqDtrvuvTdZseni4ZiByqOLtLw9Bv7aP0BoxO2jCEhltU0S9dqHxBG99T6u8U\nfcFg/EHNgy+mt/izg5VI/CgCDiF64sdunkRdxE7YefLy4SEPWuRbbxcpzdC9qk5pOsRn32uPvWlL\nFAQ3vD879HvD+6YdheCAn9sYNzXwe9HnNwuNwHTO8zbrKtaTS8PrUlrHWRB0PAo0Pab31MXe9sDV\nFgeoiKYmMXQY0Ee6AwAThQSot8x4LaJG0KzA5r5PmhS2yo1GYBd6rVHqdf6utE6FJggKTbfRZ+6b\nEIFtNAI3ocbxIpFRQ3cwcwtmTmPmVsz8MTPnMPM1zHyu+j/+M6BbcMmHl0R9TKNFKy3XlwbKQEMJ\nX20Mz9+7fP9y1H2egGeeQellRtve3rw9od8nFusTa3y35TtsPhT+9g3/7EHL88/YPgM0lDBiiW4r\naHrAerRLwOnpHj2KggzjDp4XCf7/KCMh7RBn9/qvEPKXfrIE2YXZhvkP7Bi6YGhovEb2CYsMXyoH\n6+m/xY944d5F4TtDibwSG/hjrzxv2K6ZHm7YDszfZa2hPT/vBfTcVoqDbwI3bzHOSfDeKqMTtDRd\nFwQ3fm6OlNYxh1t2fL+jYTmjXElQ6HZMwMbjemSLZopo/FpjPPTDQxj1QxDTJgCHl8zCntw9oYlu\nrv38WmMhgmnoZotAmdKMyOEtM3YqPrXbv74dB/J0m9lN2+yOMAqCt965y7CN0vVzMgHrjqoj/Eza\nVgcXBuXjW415T0ZeauwkpKm3+pLIxgEDpb7IPfXsk0fR4+Me2PyOMnbCiaCDaajio3TcU2WdxfFE\nG6oOAMHCik1WXnf9Fty9Fvh2yn/Dtk3YKCQYLzCOSe+xTpd5p/TUP8pRy0dhye6FYWWV2Vgn3lw8\nHP+eD5SdsJ5tSSTg9MLm5YWiaDRu2G6zr4DY820vKE11/MD2Lb/o+xGh3wYljbGZWTtnhX47TXay\nVogMPK/R2aHf/TcAp+cDzUyPMqcoxyAIGo82hn2Upiuv+5+2AsHt1hc7+kfgkixF2HXfj1BYKqBE\nBhkQ7u9FB42b1gh1N8XCH3MAACAASURBVGuDWlSSxrA5wIRvgR6/WdwsC8TSTvMpWt3S10/ghqc+\nQIdcpTEtO55tqO/yAyZnguBjam/xKtVubAzoW9oqfB+xx+o2GkYUBFMmGrc1rqObOJlI78iY5iVu\nZO/q0svKNe5UnGbsTNQtB27dBCy3d9NZcvZp5xtMc1YcmPABAGUczIURAh6ZyNY01P2M7tFVrgLU\nCEEg4l8c3uhGQ6vfDuGz74FJL6xz3M/RFCTAYIO5RUMUBKKj78pV2Rg6X3GaRSLodXhhV682jvKE\nc755DdGmbxY0mXfqYYsNDp3AxG+BSabcNQDwf4Ilyym1sVifuib15sCbRgeqhtggmT/Ycp9exsPG\n8HsDDYqVQhi6ieWWzc6+j4XjjMtH6uq/mx0zxoL33QpcJUSkNFcFWtM8dw53Q579siDwn/+gXQ7Q\ndxvg9SpaSkB1erc9rk9GI+K1mPRdhGploFy9XSVeIMdizKVYgpt3BwAO17Pf1n6jbq5iD+nvV5xG\n/ooagS8I/C1Cb92KBg1OjWiyST/mflZC9hhNQ3Nf+mvo96n1k+ctrnmCoNjljBxuOGrUUQmkT1YR\n4UPTOO1YCWoXhndxhs+2trP7/O4VRkcV9o47wkwWxS4Gu4gawSWmXnCd7Vmh3141f9EZFr68Qbri\nENII1jUHDtQ37icKif4z7U1IoboRGUeummzQXXbqz97KHKKfVylk8C/AuUuUaK/vJim+DwMOwl5s\nrJvlGAXBlInAfEFZ0e57rTJ3oZBnCvc0rTwA/Pvf+gptNG7QD19JGXaNAj6xyD0VsQfv8YQa4iBZ\n26vFdU6anUiBg+tBFNTEHCqfAoGwUfFu2NQM+OZ84GSaIgReFqyBH00FfpcVfZnBdB84UhAHkaNd\nR7wHio9ANzGyYB6L8zQcjkhBUBG6dw+NtNXQPjC3qvIHQ39F/6lZYeubFAPpvyxF/qvA81P0HkYk\ntVQkEGFfs0YQKXQOsM5Uaon6sZxn4QWqL/j+uqnjyU47GT6Hq2hGaFAY+bwEwjnC+RrlWE8ADzg7\nDj0B/b5cM94+bbTT3TVEqNgIjKt2K855bYxE7VJ3zd35grkhvdzYAmsRLVxeBp86vuTOjUoqZBHz\nszdDHo+eRt1GEDCpE7aw9XgDK5widzTTnVY/TUDWzsmPqaHyBRU/md+jpKR+QAjl/KNzei1bgj6f\n4V6M6GGxE7OjhvRFJ2FXIgQF0xDXEkL94jwJlBM1QhDcKDx0f4SEVaJKH5Hdu1Ey+h0UdTgPePBB\ntF+VhQ/U+HW3pqFGJ+0buNo/zkb9MmDgfKWF3Je3Dw3WuBuABTiHuQHAteuMLe9Ve2x2FIg0iY1G\noTDA7IApxbBVuKKVk9kQYeLQcLXMA3rsBZ7+egA+s476BQCUpbkLcSwrEeZR9hu/6HXN9d9p5dZf\n+4KsBa4iVOZ/Crw9A7hdfaRpfneC4P/U2IIfzw2vQ04DpXe5ZfZ4BMr02EVzAEAkQeCBbqO30wh6\nZQFb3gUeWO1s4hPp6fCO7S3TJZw3wKF0DGcudf/Oi7TLUeod8ABnuIgCc4XXY9C0F50ZvktRfo7j\ne1gutLpBAiB22NIFdSEe2X1dYp8+MYX4QfDhlkdQMl/tCbw1033Z/MwzqFMWBDZvhxjnE+1QcisO\nnzwEcahY65GtwTYDiKzY18SHNsftG+50k5lp6PzIZXpcCoKP543AQPX34wNa4ZtJ+rkyXCoVYg74\noN/afr764GrsGKWmoxg73XIfjTTSX3enHpto5vD4/QY1f18DPUNtq2xrr2WvT3thgRjGKmzLr5eG\nBif1axEnT6lFPgDuByyeTA83Jx0uV+IX7x+/FfNXPWV1GIDINv0zV27XNQJYC4LHlyn/Lzrk3jTU\nxibsHzA2kJ4ghzmTYyFAEYImosXrNWjl5RZtdbdhn6GbQxGi/+94yQmQ6CwWNYIotP+KUiM0AhH2\nEMocrrrEB8ftZhI5D+kZM+zNEpH4JBM4Ud+9nC+KEDeu4VYQiNEY35gcxpF6j/n1lK6gaCqyC0s9\ndPKQISeRGfFZ1mJBI3Aw7NQyxbr/SZjyUGxA7ZyG+94ATi0RRoYLoqAow2h/E8urxdGlOy/1Gs1Y\nZnptcxFeYwOxMJ8GAbdtCd9HM+s9uBq40oU2GYm+DmGnALDFetykI+ee2g7+OLZy7PWGNIIgEFPZ\nbU49N/Tb7zHOJ0IGjUAKgoTBfr+zWcBrH7ppWZ7Dtj0NrdcXusxAWPeYsfvU1eSczakNfGkMSQ8R\nJABe94KguI67Snlcmi8sCQSA7dvDBIG5xBUX2886FS3ip+QV7OlehwZU7N16/cHQvAiAseG2M1e1\nKgDaHxXSXDvcs4ZCW+0NRHdvS32Ro39ESqKRM16PwTQUidYVML3MtzCvWLEphpRjHo/PeTxNlLDX\nEzK5BjwRxuoIiLPHcpr+XZZ7Yej5k090FktBkDA44CwIyj3OuXbMOD0qq7ztALDRNueqA+++i9Vj\njKtKvfaCKECAxyY+2QoxcZ3G5PCpeHHWtgqkhxozBjjvvLDV5o/Jn+5egEVKQOdhYK1q1/cKsfNO\n74AoqDyBgGE0qCgI3Nr0b1gkJkpyqGuUMrbUqzpEhZfQKRoqKnOl4Cw+JcEp8d32qo86+O/W23xT\n7PW4Mg0tah15HwCAoBFEY3YSfQlBIUFfuUdxzGtQmhQE8cfvBw4b55ONNG2k3xOdRuCE3WN0G2Fh\n4NFHw1YpeU+sd49WI2iQH25GmN8mfL90lyGOljzyiOVqc0MQTIvTA4AijLXyvS4bbvH5mHvpoiC4\n7lcHg7ct9q1xNL17QHn+gPEdaOAwWZDbWckAZX6BitrWjwljD34RsimvON24X2+LLJ9WTLHomGh8\neJHNBq/XVa/d/A5OC++vhMrTfARBci/EDPsJDb/fY1wm4ZvlKL7fipLaguDhh4EWpkEZRc7ho4Eo\nBUFdhzFAdhqBW8daJJzynlx6AChj9xlBrYink80pzYf5POVp7l/LvhP7Om73QP8IfeXubny6wVls\nLwhiwe+QATZqQaC+p3bvWVj5pv12OSWt9XoN79bOGCaVndZO/335Pj3iKlZTTYlPnzTnfZM31vYW\neDyuGmvzO5ibYb0fezwoVyMPozEN2QmCG3YYlz1pgo/AJvVEIkhtQWAWAgBaPD00bJ1oowxQ/DQC\nO1Xcavq/WHDKe9LlMHCsxP0IRyvi6WRbMdB+GKffA+QL7//eYhfpS1W8Lu5lSBC41AhEQX1KvrGL\nfbHJTxMtTr1yJ7+FFXXrK3MGmBOZaWyMYFPf4rRd8BEA4Q2vG8zVOqk+Y/N7le8uTgEQ5u/1e4DF\nqpYx82z7DhF5fa46NOYGXSzvH7/Xf7PPizJWrAoBAu7ucg/cYIwuIqMpS3AWewQfgV0yukSQ2oKg\nceOwVZ6y8F7yd+frv+/t8te4CYJ6ZcD4C8PXx2QasiBSJsSKOsnuzLy7YgW4hHw+dB+gL+c0dD+f\nn/8/kfdpHlRsFKcfC++N/+XW8P3FOQfMRGNesaKFQ6qrSLH9Zpqecrrl+jWnARtOjdzh8Na2nxid\nPUaN4LIzlZm2SqKwVojv371/1IMkzA3zpA7GZTun9r//MCw05effVwBB1eka8ADpvlqWx3h9aa6+\ngwZ1dJUnP91YxzUthMGWXn1AWZCADi30GfGceKT730O/2eMxmo291hoBpbmVkBUntQVBursbKfZQ\nfOm1LGOD/TGYSRqUAQUW7+eZsZiWLSj1GXtdz11t3K69sGabrGuS1CMJeMngCMyrpwuCr0yNxKfu\nvjsDDR1GJS90GbGSDKI1OwV81q1ymVdpyOpYWAbFUzia4Mw+AtV84dS7Njfg4vy8uRl6SmZzw2wu\ncqtNmKi3fgPDsl+tv9/joBH40lxptuzxYKaa1zDb5JQu9wjfmVd3ogc8AIRn8JbDZGJipBA8eiqU\n7Dow+gjENitN+gjig0tBIEYJpaVlWH6QseYGNx+3Ko55pEq9xggXc0KvrmqctzknkGuiiDqqCEEi\nwzMQncXmhmex2+gOAadUG7GYv8zCKV5E6yMI2kRXlau267oWjuPpegg7/E6C3kPGd1dtrJy+A7Fs\nwCgI/B7dl9HElGR1XKZx2e4c6XWMyaj8Pl04NS62vnduTUPsNY6kPiYoSwZBI/hO/B4YBoNtb2w9\nwAyAQWBA0Ahe7QmjIBBMQ4ZQ0gST0oLATe4cwNgYpPlq4QKLOWnSY3QSml/PHHttPGpKfMA9QhJU\nc9irODI3JsevTY8z3gS8ZHgGYu4Vc6Ng+6E50LhA1wh+a2I6dwzlzT7LOhtnRYnGZLipGXD2Lmsf\nULmqETgFMgCAP934wjzYR1jweAwahZZozSnhmvn9m9fWuO1mNdVLpskFdMLkmLV7JmnpxpuuaY5+\nD1C/VP/SRO2dPB5bx69IUAiXDRLwlpABulyYnpIFQXC4nrGxdooiEhPLpfk5NLNcgGB0FqfrJgQp\nCOLE51vDJ4+xQsx9U6tWHFtqhDtz3TTIG1yOMyg1tdNiIzmwD1AsbI9Jo3GpEdzSL4ayBYIeoyBo\n0kBP6OPkxHOLGAJqtnHHohEU1DIeZzbJJYMgAdthPa9TmVe5b1YjuMX7ebBMP/5oHeBbwVfGXq8h\nOWCosXLQrsy+tTn6FBKO99m8ze4bqZNhNA3l1VE1Ao9xkvh7bxZ28npxyJTV1gryetFJFVCnFBu/\npXKPIFS9Xpym+nm6HFb8WxqiIDhUT49wAowJDztt0Mfi+D0AiTmFxHEEaVIQxAVKt3YgAcaJNjoK\nY6Tq1ja+bE5pc91gbrjc9ECz3M1FH6bxNG+i200KahkbvZg0ApdxzFe1ucqwnGUzotqOoIfwfX99\ntvnr298U+n3FXmD6Ofq+dTMcEtrbkNVQtcUCSEs3dg9jEgTpxuOSObesRpCAFzKtNYJTG7W0fd7i\n+2iY84KM7yZ7PKEO0gu/UwZmAYDHYT5ns7Y29S9TQ787t+wKO+7tNsCwbPeNnNrQZFdVTb+N6jUx\nTLJk6CB5va60SI8vLeS7O7UI+M/VehSCOfTTMO+0IAjqZdQP7buiJfDaFUI1hDkV6vn0zub93R4I\nCdmcBmmG6Tp95e7zTlWUlBYETj4CcdSsYaYlkzpW0RBKs2lIs9s74fac9eobo6L6dNRDYIJknF8g\nlsZKtH/+8S/2+7VuZPS4fh2lDb3piVK0bSq09kJO9rNPAG8LTrjfnXVNdIUDOFEbaKbapS/YZxwm\nG8vzLfEZj0vm3LIap50Mnw9Xw5Ney9BTFRHrKv5mmDoLXk+ocVjeSs+HE41p6KZ2ukAPej04cbZ1\n6hCvqcNm22kxaaiN6iuqc/P6pxsSKBo0E4/HYG6x46KVxgy5F5+hv3TidRHI4PsQB4Ddcv6fQu9F\nTm2jYDzWRR9UwYIpqG6dhiFB4PX4EBDmZPBGMfdIRakUQUBEWUS0gYjWEpHDXFEVw1PL2PsTH0zA\n5kP2mHrBFR1UVZwG7BZ6+Oac+1a4bZyC6SbVUchcGCSjxiBex2i7UZgAPhRzogiC1NHOamocom0Y\ngx6Cl/TKmsPmDOYcl8PuxwnRRQ/cZL+f2Gj0/5OrokM57jWSObesRvNCe39JMM2LoN024fb12WZc\nb9QI9B1PpiNkEmIH05BTz7vMA6x8SulNHGtqNL+GmYZsyhGFW8BDoai2BsdPIk1oNA2astdraFzt\nCJ8UyCbDqNcbeh+/vsBYJ296rdC1FKWZBKPwbYr3kLw+XTCQKQNrknx0QOVqBL9j5kxmjmGoijtE\nQTA2UzcPAMaGUfyQPV53GoHbyJG9DYGp7ay32WVTdOsQrWP6Yqi27kwz55AXf+8IH14RYoXQafPX\n1+PoomncoxWetUsD8IgZGAWt7LXLjeWdtjvCJLAqr1+u/97exH4/MXZx1tn2u4kEKDbT0Ms93e0n\nsra5/Ta7nFjss59Fy6wFhH6b8uaIvVbRoclEOOMJWOKUo6ucgig8VYvVJ8MANfPANa0eP55rHNEs\nxtgHPYRztyuz7ZyzejfShcGCZWZB4I2+mSNzGohQJTyhgaLvdwM8Pr1OXl96aN8Sn/E7Fv0A4rPx\ngFArT3E6NDpRbLgXlMTmOaVNQyRI4bFd9I9gbXP7hs1r6o3a9U6O1AX2NrDeJlLusY7nBoCZ51iv\nd6sRXLf4kGGGJMrQBYG5MTY0AA4NlyF3/gV6y2i+X2J6Z3OKg7a59uXb4RVsz2KyvL0Njc+g9fq9\nrsoT76Hb6DG3AwkDHuP9dSMkX73CekyJm3PZYfeeBH3GnPmzzxLKE+oq2tJbFtinQQiKgsBD2G/z\n3lvVR2sMyz16Mm4mYIE4ml+o665TdIHi9wAjhegdEt4R9ngQFOoozsts8BF4PCghfWObx63rHoZw\nLoNpiCgUXl7mBVBLEATCGKRSn/E4saNj0Kp8PtQ5ojvtA0FdM6kJGgEDmEVEq4looNUORDSQiFYR\n0ars7P9v78zDrCjOPfx+3efMygyzMAwMso2yiyCbsoiICALGHcWguMXdGMUNNVE0N0ZjTGJyiWa5\nJuZeo1nUGzUajDsxV5FEERcQ4wq4i2wyMDOn7h/VS1WfhTMTYGSmf88zz/Tprq6urq6qb/8qPy4w\nCrcwXBgb3dA9b2ld9gkWVQ1lm2y+Z8b20OhmzwWzyuDM3zdsoC3RW5t+9U5RKPJM7TfNKmcuVrly\n05jlmgtDzrxZ4HDDTmAO8ihhOeHlnE3OCEsiMCZhNMPj/5aGYb83Z9om0IPZh6dGjJEmqotDcSEX\nwZhvmCaaxY7azdadNxpSSaObO3tmNuSSrk4YekLG85+/+ZrlMWYG4WUzFkO6sTg4FnjufZ0iZHPT\nFkuKWmuM25TAg5FYAr9PDxt8ZLgAilgR2qPqQ6vq+HeBIk0xKxpsN12TS0+5wqrN4VhY1zlckM33\ncpuaaWgK3QJNFWeu7y3GYn3O2DAqWMQJCEFxcRlSEn7UhOHe2pCw2+FkkQjETfDRgNBzZe+uezN9\njo4O33Dk9OwN3MFoK0IwXik1ApgOnCciE6MFlFI/V0qNUkqNqqlpRSJybNVQk6O9AUAP2Ml7Tgmu\n/aOn4cebp7G40c1vwZ417ARKC8PBMmtWeM3cu9Ssq09NFlEhA8zJmywJZ+Wh/WcExxdNixAC4/7o\nngkWITDiUP984kOWkczyOLHut1eufNVcpo3AJMYXjZvHdw75bvB7oZG77sNS+Lvh/WXC7M9fHP4L\n69q408LjTy4Lg0ZySQSm4f2xU59kT8NhZ1yvcUC6h9nrhkoqJWGunZYg1xi75sBrMp6f8kYzmwrD\n72C6T47qMTo47lJWyzXHhI186ZyXwoLGd0wJrN2m1TDlkSwdS42o9YmF/dPiZHzu/CtDjrY4YXOX\nugF9QqNVIgXTX9YPOfAd+/1NF0wlYnH+N84MDXFXHBz2S6Kh0coPZTIwU+aSFaYqZ8Eh3wmOHdcN\nCMF9c/9Msjic225BYUDgjtv3RIb3DPVfQ7qGumTlSJAnKVFYRM2UMHFiz849eX1MPSPOTg+g25lo\nE0KglFrr/f8IuA/Inpry30CqMhwcafuEGuqHe4YYlvqIasjMQ/Rjo5UpyU8Xnkq4mEuvGbEY1aku\n9Oao5Om22RgxgiUM7iS5LtwpZFltdokgaiA1y5n7O7uuvcGH2Z+f9Q4DH7597UFWfeNPIy9YqiHj\nWBIJyyBnLtbNTiiNrLj5Co41iGyuBXRdlmAwleOeJ/sYbTX8u5d3JVg0NxTZA8LkPhWtC4bLpXZK\nONnHySZDDbXB3A/dcGNMJRxmvLQl+O1mcQ1VQKUXuVsWiVY21V1jH3+di6fa133DqkTsFoUGIZBI\nkolUt9AwYn5Hk6tOOWJx9F8Y6QBMJwe3OWV56pn15Zq/psecyVA6biIgBE5hEUk3fFaysTl4r+aC\nhDf3NQqMcogExDmhJO07+v2R6/vuaOxyQiAipSJS5h8DU4FWKBO2D9Ul5HbMSTlmTegXDbbOzuRG\nF/eyOZ5vhEx23rnIVSJhcczdjKRj5kDcnNQcLpB3jp8HZ9mx+cpYoIpef9O6lk0iiAZYmWqeplQT\n6/2JbnhERO/bWlwQqLY2di7mOkO+y8eQevFUe5Kb30DchEUYs01kp6DQJmI5vk1r3D1f6hYem+3Z\nnAzHT7NrV/y/houyynO8RBHtv0f7hsfZFor7BsJGQyKwVSDhaGx2HSu62JTKzEGSEijbkjm0PpqV\n9NOI+st/tqNsydHaDtQgEEqgccrk4LdpbzLLvVdfbUkEpm5dGXp7SalgDt820ibGucemrcf34brJ\ngOt3C4utNiXXb6Kbl+W+y9JXrQh5kxB8OGxP7vaSUar+/Uk4CU48Co79lvYqaUppatKuCQFQC/xN\nRJYBS4A/K6VasF18/kgmw9Xf5MRHfACuMQrMBcU1vAAclZ1rSAkU5xHvkUq41ph6zJjI5sIwc04Y\nCVrQlN92ln85bl/rd3NDyN2ZqW399vr4meE+GiUExxskubG5MUx7UFJse5VYC4UKOXXX5U5P5fVJ\ncX6ulYqIaigqERgLrzmRm5ywfpVI5E0I/l2XYHOM6ARlnmulCPMMjth04VRsf69mHwtD7U1a/5mb\nuCTdzIOk0bHVVBbhU4oFXvzfpvJi7jo01K3ZC4+9OGdL9re5ILeu3WfAnKZma8zkkgi2HX9scNxp\nW+ZyP7t8ii0dquZA3YKh3nVTqeD97xlsS31RNZcJ0x5hqslMJsWXFH7oGbQTxi5/hZ+tDxiEzUmb\nEKyaOZZ7B0Pn+cCIESSdJHcOg5Xdk8G7QHYJbWdglxMCpdSbSqlh3t8QpdR3tn9X65BMhnJrdMEr\n/CxMAWraBUzVkJtK5x6/69m1Xq+2B3M2aFc+fXzZlEiQl9H7b1bB+Uv08fi/vMLMr+aut8ulkEgW\nWgvFlurQnWPzVDva11TrrDdUI9F+MTmwlEqFhKCoKKgjLU2wOMECXZIsCRbaaAqMbFASkQgihmNT\nNZRNKkkl7J2odrREYMLyZxdCH3vX4XcZ0o6DfkdzvLyVI3r8/Jnh8d1Gfd0utsuZC/c3DY2cymWP\nSCn+5TkpKEdIGHPEXHjSXIfLM7sKbXNzb4F51Gz49kSQgYMslaRpIzC/d7NAaoq2zL/azbWlPqPc\nF6UFwXf82YGlOOIwYw7sc7b9faQ5lfcGUaceYTwri3rW7CO3qBhBgnHoGoRAIShPInhkT5sQ+O3e\nUKTPRwm6zxRZEtpORrt2H60qCVVDj5/6hHXtxU9C1vf3x/8xOHaNj+KodPHxyoO13vu3Q7MvNusN\nvWlt5R4BJ7OxECb3OyRre82kY9FUuFE0unYYPMD+fUNH9Yaxo4OF+IcTv2MtfofudWhwPO+gK4KY\niHfL9fv5OGnYSYEYnCgto2dVH0BvaGJOrtNHnRHYO07fey6HGSki8tkn96hBR1kTLGEMy6iN4OaZ\ntwTHo/qOCxa8ytIuJI2c9D+c+eOszzt4r8zf4JsHfNP6bbrI3jbztuDYZBbqyusC6atJFFtrwzH3\np9l/surrRDi2Rp+RtXmMrgtFglu9w+fr4MOI7dAkBKYdYN/uI2hwwsFkDmF3ayPvjhvCimp45Dg7\nstBceF4/57jgeOHMhXxv+g+C3z+d8dPguMkJc/u8MH1fDu5rR36/WwFXT9YcduBAKjYhMNUrJRVd\nKC8s5zt//AapZ56xiLv5vsrIjvpxmctTpzzFxiJY3g07WWJzc/D+Cnj29GeDS/tU2PtRfjYi1OWJ\nZJ7cjmNLBINqBnGXR6w3TZsUXKs68YxAxZtIQdIx1hXHZfbes6krq6OmtIb+1f25eOzF3HPcPQA8\nPOdhLht3GXVlrc0f33K0a0LQ3ejISX0mWddMH+S9qkOfN0siUNCrcyTvscDfe8FDcx7Kyln2D73N\n6FPTL+DaRcE1B10XXLvrmLuC49l7z84aAGZiZTVMnwPrv6uoKq5idF2opC0rDFeKplRT4J7at3Nv\nq7754+cHqoNzD5jHHwbrYyV2LviiRKhaK6iqYeaAzCG6pUVlgUTQu6Q7Y3trsakkWcLF+4fRR2ZS\nMxMDugyyuL1CsW0EZpDf3nWhXWRk/QS2lOkVMLFpM6eOCC3TY/tkj976yqAjMp6fO8x2I3nH4NrP\nGnWW1SZfBVRT2jVQHRQVlFheSIcPCL1BZvWdyfCq0HOkule2TXHhsbmPhT8E9rwAJp8MpUmbO0g4\nibRsqqDHweDuoUvaMUeHBC65ZStDB09i0NehYsQ46it1rMjTvWxuN2Xo2Sf0mcjoXqFD/zmjzwmO\nx/edyG/3gRFnwoMXzeThOQ9nfS/zpdb1qTV+hYOzV8+9ERGuOuZH7F2/H6XJUKcb1Zn7Y/XdKpd9\nag0XPDMVRVNzkMTxgQv+j72qQo+8r5/9q+D4qslw9BDD2yASQe2ral1TUigowBGH1NAhyALYOij8\npok5JwaxCG7KlmZcN8ldx9zFmnlrcMTBEYfvT/0+/av1/YNqBnHjITdaBHJno10TAicLVQfbWBz9\nSMFxDp1uLkOO6S8uBQV2JGGeeyRkIwSfFsNfDF9tx7Hf0VfbNDY3cvkhcO4MaDx0qp0fRcKNMSgI\nRexMLo7T52h//WQkK+sLZv4vx+FqXzVRX2+9r2rWnOmzPeCrGXYDA+2zbXKjBRiEICIRJIzYEFWQ\nZKu3OYnTsM3yfZccmVMly7ezPDuAY4/LWAxJJAIioUQo3ajzF5VvyK50HnjPUywelznXDnj6Yr9+\nES6fAkeeqNv5ZpX2AmpM2ZGJCScRRENbkquIxeiYzM3KyftYxkhz8fcX5Ns9WutH9YrjZlWV+EnS\nXqiD5qSbU6+tjMRw94wOJ4m54L34tcOse/57dEHGcqCDRMefBg8Mtb+bmFk7m5u57BCYeArI8OG2\nd1p9GDAZTQcdlQh82565Pvi5zPw6m5W52bUTMAiOiq4xu84InC86FCF42mDuTU8hcxFyTdEvh1oj\nm6EOsGRxx9DBaNHFSwAAG6xJREFUaq/V7PeZz8tm0IyqqqJ6xF4XQf0FWiJoSMKtY7S+87OIy+R1\nvgmhpCQQvxWwNqIK/ks/uGSaPQlFRdxOHYc/DQJZAJSXW21UKhW0e1siDMx6eC94xjfuiVjfqsBJ\n8IgXDSuOa9lwTJ12qrCAzcX6e6nmJmvLP9Ogm4YsKQcK3AJrQ/esAWCuGxBSAcY+tBzITQicphTb\nio1NR6Iuk5ZXl/C9CbBooL1gNDY3WvsQJ5xEsBmR+X0/nDqeHqtDG5i5+L14+JiAECTdZNockQVw\n+pGR7+04WbOOWqk2lMrJfJlQkff10dDJHqgflmSehEqpQDpvVjbHZjoXSHMzjQlY3EfPFcspwbRL\nIdbiH2Uk/BZai7hHcPw6m1PNYeqZwsKAsEjkWYlcY7ON0KEIgQmTe7QMQMY9pu94FAknkbbFXsY2\nFBZagz4Xx2QSAnNhMJOmRaeFROr7uBO8VWVzj9F+SKkUN03wFu5EuIPTdo2oxuKwudDYdFxlp5i+\nRODX7Usmq6p09KQPs18KJOxbJ5G0CYFBSFVhIT+eXsUt+8H6446wJBE3x6Ye2fLwFLgFWZMRWnDd\nUO8sYqVAzgqxOeIoMhH+KLFQKHtbVSfB98bDSUd5Niuv+Kf770Pp5tDdxopqJRwbegyL37zsTc8x\nj8z0EGo7PmLm9ZSxeFucfkTCtbjsLEhFCEHURuDDdUKJ5YMIkXcQi0GIvrNP+C2CGGQNDSWCU46E\ngecBZWUhIYglgrZFGrdjXcyiGjKOL54W6t3fLbcnZk1JjbVw/9DIiWLCLSgKCIEom4PvUhIq5Ad3\nGWxLBEbTFxnJ0KKLk5NlClcUVQSGu6j0UlNiR2oHEkGGqvpW9E0/CXTr1C1sSyrF4JrBGculvIls\n5pkBTfSCx0UlAhLhxEskbT9u07Ml6TKk3zgunA5FZRVZ1X1pbXIz95np8QR5pkMW2LL1iywFjWei\nGNp1KLWXQPVlMKJ7mOb1lRr7u/rvOL7XeOtbja4bnRZg1eTC/wzT7fCr6NKpK2sd3aYH+2HpxRWK\ngdXaKNq7c2/rmxcnQ268W6eQSufSVQ/oFro11VfWp103n21iZPfQUG3Oq+oye1emqG0veG6XAUEd\nE3vbiQlsr6Fm9uuhU0o74pB0ktRcCkPPhULDuaCurM5idMTJvHa4GdSKo7prO11FUQUNSVhZY9cR\nVQ3VlWVXEbYVvnykaQciuhj4EYEfVCatwW2phgzaOL7+QPZYVQS8wgvdYc28NXz8xcds3raZAV0G\ngLEJven3/4dZf4AF2vDkFBSyvkTXf8Hk+RZHMbTr0OD4ygOupFldHfz+2sgzAJ0aYb0RDJesrOLx\nuaGXk28j+CIBJcBr573Gsg+WMbhmMHcefSevfvwqJUlbvz+k6xAe+upDASGas+/JcOcdFDpJNsz/\nFBaE+qFH5z7KJ1/Ye3dWFlWw/JzlnL6khu8+Cv3LyvjbqX/jvQ1e7hczPYEX6FNeXMEHF6/gO7Pr\ngJRF9MQzmPko6DcgmHiSSFgLvPmtetX257/GnceF+19IXVmdLRHkcL3LxpkXJ4utBfnPJz4EN81I\nL+i6DKoeAKz0K8z6LB8pFNdMuoYe5T3o1bmXt3hpZ4EZc+CQflOBRwBtpF96xlL6V/enWTXzu5d/\nx7BuwxjYZSArr63MWP9DX30IrtVtHVa3L694EcRdjziB0YNC44xSisvGX8aEXhM4oPcBrJRfArBH\n+R5UFFXw99P+zsZtGxnTYwxvGPVn67NvHXwt00bPpinVlLYgAzz3tedYvWE1ABvqqrllP/jspGn8\n7LDbAO0lY87F6QNmWvffc9w98NX0bH3zxs7j/DHnc+vzt3LemPMAWDtvLRu2bmDlJyuCcu9MGc2i\nE6/nzXU6wDLpJrnjjD+TUinKCkLnioknXM5znbSkdOsomBx5X3+8ZpI0f3Toj5g7bC79q/vz1ClP\nBcFtvlTRu3NP6x1H9dwpiRT+LXRIQnD1mf3Y453PM5aTyko+KIVum9GcxKpwU+DuZd3pXpZ59/nH\n+8I3F+vjYweHATFOQSG3Ta3iDVnH+XNPDhaoz4tsg7PruGwogIIGWF9VwkF7HoxPCEzXwBe/fT7n\n9A2dxv3tQ57sAzOAgV0GMrCL5vhqO9VS2ylzHuPp/cKEVjX9hwN30HVjiqJC20exvrI+4PR87jHp\nJulS0oV7B8O9g/W2hpXFlVQWe4uUMYl81VBZcTm1nWoZWjsUWGbLMVGOs2vXUAfvuJYaz/xWxZ0q\nKE4WM6aHN7EMophTV51j4TYJQU1ZlhzQrkuPTnXAShDh3Jlw7++zP85Hwklw9qizATsSNiXwzQOv\nxicEACPrQo7Z9Fi6dxDstwZeGF7LvkD/6v68/unr9K3sGzoAOE6QSqJHZBFKqRSu43JAb9urKuEx\nKGN7ZsjkZ/TXI/uUYmaRSBaVML5ivFV8Y4GOL6gGqoqrqCr23NdEuHA6nNmritICw1hsjIao/rzA\nLeDmsVC/Do4CltTBmLX6+xYlirhobOiV5s/PVZ+tYsKpsLILfH/aGA4o6sy+3cPgyxn9bOK+KQkN\nE/anae0/qJunbUMvRgzzSRwghZtIMuxsGP4B3OFdK0wUBv1mEkNfIih2I5t57EJvoHzRrlVDUX38\npYfoQJ43e5dbeiKrXFUV/b+ug3dMnWYunm/UGfBEvQ4Y2/sc+5pbUMS2pPDj/bWnkuu4DDoPxp5b\nmKayuWaS/n/bjcfZye+Mtn5RaW/VGPUa2h4OOSn9nCrTdSa8nO4DztcG5yi2legB/VGvXAn+CXKs\nbCwv4t0he3D3EHjwYu1K6U/6lMBPR2tJ6ovpU9LqcEzVkJvFwNfTdu0te1+n8y1pSv/2Q841X1hX\nHt00HWyVnJPNM8x1eX8Pna1v7aiBVhqSbIhOfRFhjUdzFbm9nEx8b7zWQd8wXy/kPqdu2RMch8+9\n/XyVl29rU9J/lj2Sg/syDPCoqrDoKvj6ad3skxlSJVddDt0uyVCf8RCfIVoRHUoZFslLpsHRXubb\nCadBpyvS67aqQHimN3xSmsF+EEHZFVBzWdgP75dDs6sN870uhOKrdLlbxnvfZ+RIXuoGvxmepULr\nVXyda6RzWzhndwW+fC3agYhyhU/1hfoLobG4wA4bj5TbWKSDd3IZ90z8w1P53TQBXutm1+UUFtoc\nj5NgRQ28VZnugvqT/aDkSljXpzb7IhSZKC3dvCJTOgA/R5Gk9Pu+3kUbnKNYX1fFwXPhjxcdmn7R\nQEN1Z874Ctxy9VS2uXDCLFjXq8Zrr26/Al6uhU5XQXMPLWWl0PpyXU7DcRNpuv8RZ8KYr0FBZ7uR\n74wbYpUzYW4C1FBWzEtd4eQjSYNpFzA9lCy4Lqv7VlM3D1Yce2Ba1tLAG8pA1EAdNQRHjf5ZIZ4O\n2usTf3EVMUaC47DwkArOmwGbZumYiT3mQe0l6WM6r011lEKh2JrE2ucASNs+EqDJ1X/p1RhtFYfy\n+TD8bG28Pnw23J89tCJAY0I7KuSClZxuO4RgUyE0eETSlJy2NW/jvYrw2qI9tXNFojazRiAj/Pqi\n68iXUCLoUKohE+ZEzKZPNjmYfD9dmjrK9HtXKniWSHrWQQS2FHjeHNk4xMggaqlEkGniNxfodrh5\n0L3H62FU8fbd3345Ek6rKqHIm4j+e/utzdSO5NWaQKTAUA05aaqhFzwOvNCN7HVrtCv6Tc10HinX\nYdi5ZIRltM3meeS6KKV4vxwQScsiO+kUnb4glwlZRMJAwxzlsiEnk+I4bE3AT8fAhZ7U6acVySoR\nZHpGhuellW/FoiZoQrDRk8i2NW/jgYHwwMD8clNtDy0hBNnui8Zs+N5LLUoE54/bVCwRtCmii/LZ\nI7V+9oIxF9i+/pFyxQk9a2YNnpX3NoTDu2lZceGMhdZ5t6CIKyZoWbZHeQ86F3WmoqiCbx/07bRB\nddTAowA4ZtAxWVNRm+khzLYXJ3KzSSuuOIMXukGPSenRwWbWUl1XMd/YL30rp5n9tSHPt4GcOeJM\nKorSk+b4nh4nDTuJowfpgAN/I3N/gbcysnoeKiknXLADiUBcSyKoNtKGRFVrZtrf6Dc1I8Qn9NKR\nz/2qIruoYC/KWf29XTeIQp7UZxLj6u28TmP6jGPkXhPsezIsmP7Ymj34uLyjSC8Yo3V2Z43UdgN/\nbFnpCBwn83nsaOeWoEe5FnvnT5i/nZJQXlgezDUTB9drL7Y5+8wJ3vf6yddTW5pjP05gQPWAYG6M\n7zk+o1HahD8XIbvXURT1lfUc1Efb3oZ2HRrY2fwgw+snXw9oBmP6XtMZVpslC58B3zFEosToSygR\noJT60v+NHDlStQbrtqxTSgtmadeum9MzuLalcUvWcj+9appSoN6cuE/6A7LcY11raEi7lEqltlvH\n8688al/LUm7hNTOVAvX22EGZ25EH7n/lvtzvsgNx+ynDlAK1cs70nOUW99TtWbfofrX47afT2mf1\noYen3n4qKLe+Yb11z7PvPZv9HY3zL9QS/P5s/YeZv0Hk2YtXPrrdujd3Lkm79E65d8/bb6ula5b+\n+9/Av7+pKe9bbvvhiUqBenfvXmnXVlZ59a1Ykf1ZMbLiP2+erRSo1UN76xN+nzU27rI2AEtVHmts\nh5IIIheDw1yuhp911Ra9j4b0aV0jkunqhXy4v3yNh1ZGrVbCSSR5pzPccnqW1Jk7EL476fZUWqGN\nwM3IQWXqQzOxV6Zv+rshcOm56VLAoXPgBM/DUuVQDT3g67Cj+v480obsCh7QNwi3SPXwZeRO2wk+\n7q0NXs8fH8l7le/c3oXoEITg0fQ4F96r1eqfecamKI/3SS+3ur4Lg8+Ff548Ne3aEbP1puS5G5G7\niw+fDcdkymlTXMyKarh9v9yLzBcV2g3PN8a2Bo449LkIHp608wNdGpJ64WkuKdlOSQ1RoV76iZEZ\nLNgGfFXR6i4FGZmA2bPgqeHpqqxF/eBuL6Tj+3rXSRZce1CgGvrc07rNmgU95qU/N1seHoD/1XuN\n8GnvdPXHY/647NQpIGzvV+axEUUWjDlDb0vaksW9qUi/46aqTmnXgvZVpPdZU0w/toutnUuRBfDa\nRB1suarK24P5S0h8272xOPEtSCSSNESuLRvQmcHn6pTKN4uj0y2QmbF+rSsZF/T7B+q/KzPcM/ws\n6LkBHthOGx8YmPm8kyxg0NehsqiU04CRZ8LY9+A/I+Xe69+NaSfCYWd8hTw82jI/y1s0W2JYay0W\nTezB2reWM/7c4xmQo9y/KmH8e+A26FQJlZfD0Pr+HJTjngK3gH3Ohs71e/JYKzf1+O0+WnI4YWgP\nEm6SaSfC2jJYDmxNwtoM63TAcIysJOoIe/TxcNViqL14LudHrp19GNwwAVZWV+N88B77fQ1K+vXj\niVa1XI/T17rCD1twzwf9unPa4TDwjCOIJoe9YDr8YCysqk0nYnt+A3qth8WtbGtHQHRejTtd70mx\nui0blQXtWiIA7RNcXpzO0ZQmS/UCL6GaIZNHgB+G3tJt45Z1hwdzrXTbge8H37VUh9z/sw4W7pde\nrihRxCN7gRRlcIrPE/47Fm7H4LwjkCgs5vqJen+DXLjm2CounwJqysG4jsvnxTp4KRcK3UKWd4Mv\nOusNQy49BI6fq+/x+zMaZZ0Jza4eHwknwSN7wVt75N4cosAtoNMVcP056ao15cB/HAhN1eljcFtC\nu+qCVmUt2QO2VO26DctBt/1XIyBVkb7xTJMLb2QJGXm3Av7Weyc3bjeHn7LDXzs+KYU1nduyRTmQ\njyGhrf9aayxWSqkbFt+gVnycbuz69ItPFQtQKz9ZqZRS6gd//4Fa/uHytHLrG9aryx65TG1t2pp2\n7f4V96v7Xrsv43PvfOlO9ZPnfrLd9v3qhV+pp99+Ou18Q2ODOuuBs7QBVCm16I1F6u7ld6eV27h1\no7r0kUtVQ2O6UTpfNKea1VWPXaXe3/h+q+vIFx9u+lBd+eiVqqk5t0HzybeeVPP/Ol+lUinVnGpW\n33r8W2rNhjU570mlUuq6J69Tb697Wyml1E3P3KRe/ejV4NrVj1+tVq9fnXbf/yz7H/Xovx5VSim1\n7INlShaINjYrpb7/zPfVKx+9opRS6vE3H1e/efE3GZ979eNXq/fWv5d2bfE7i1XxfxRrh4QIlqxe\nom59/tagjgVPLFDvfv5uznfMhSffelL9+oVft+iezds2q0sfuVR9se2LtGtL1yxVC5cszHjfAysf\nUPe+em+r2tlREO3bm/9+c15rwo4EeRqLRZf9cmPUqFFq6dKlbd2MGDFixNitICL/UEqN2l65NlEN\nicihIrJSRN4Qke07JseIESNGjJ2GXU4IRMQFFgLTgcHACSKSOYdxjBgxYsTY6WgLiWAM8IZS6k2l\n1DbgbiDzJrIxYsSIEWOnoy0IQQ/gPeP3au9cjBgxYsRoA7QFIcgUTZFmsRaRM0VkqYgs/fjjj3dB\ns2LEiBGjY6ItCMFqwEzUuwewNlpIKfVzpdQopdSomprWR83GiBEjRozcaAtC8DzQT0T6ikgBMBu4\nvw3aESNGjBgxaIMUE0qpJhE5H1gEuMDtSqlXdnU7YsSIESOGxm4RUCYiHwPvtPL2LsAn2y3V8RD3\nS2bE/ZKOuE8yY3fol95Kqe3q1ncLQvDvQESW5hNZ19EQ90tmxP2SjrhPMqM99Uu7TzoXI0aMGDFy\nIyYEMWLEiNHB0REIwc/bugFfUsT9khlxv6Qj7pPMaDf90u5tBDFixIgRIzc6gkQQI0aMGDFyICYE\nMWLEiNHB0a4JQUfe90BE3haR5SLyoogs9c5VichfRWSV97/SOy8i8mOvn14SkRFt2/odBxG5XUQ+\nEpGXjXMt7gcROdkrv0pETm6Ld9mRyNIvC0RkjTdmXhSRGca1K7x+WSki04zz7WaOiUhPEXlCRF4T\nkVdE5Bve+fY/XvLZxmx3/ENHLf8LqAcKgGXA4LZu1y58/7eBLpFz3wPme8fzgRu94xnAw+iEgPsD\nz7V1+3dgP0wERgAvt7YfgCrgTe9/pXdc2dbvthP6ZQFwSYayg735Uwj09eaV297mGNAdGOEdlwGv\ne+/e7sdLe5YI4n0P0nEEcId3fAdwpHHe34z3WaBCRLq3RQN3NJRSTwOfRU63tB+mAX9VSn2mlFoH\n/BU4dOe3fuchS79kwxHA3UqprUqpt4A30POrXc0xpdT7Sql/escbgdfQKfLb/Xhpz4Sgo+97oIBH\nROQfInKmd65WKfU+6EEPdPXOd7S+amk/dKT+Od9Tc9zuq0DogP0iIn2AfYHn6ADjpT0Tgrz2PWjH\nGK+UGoHeEvQ8EZmYo2xH7ysf2fqho/TPrcCewHDgfeBm73yH6hcR6QTcA1yolNqQq2iGc7tlv7Rn\nQpDXvgftFUqptd7/j4D70GL8h77Kx/v/kVe8o/VVS/uhQ/SPUupDpVSzUioF/AI9ZqAD9YuIJNFE\n4E6l1L3e6XY/XtozIeiw+x6ISKmIlPnHwFTgZfT7+x4MJwN/8o7vB+Z6XhD7A+t9UbidoqX9sAiY\nKiKVnrpkqneuXSFiFzoKPWZA98tsESkUkb5AP2AJ7WyOiYgA/wW8ppT6gXGp/Y+XtrZW78w/tFX/\ndbRnw1Vt3Z5d+N71aA+OZcAr/rsD1cBjwCrvf5V3XoCFXj8tB0a19TvswL64C63maERzaqe3ph+A\n09BG0jeAU9v6vXZSv/y3994voRe57kb5q7x+WQlMN863mzkGTECrcF4CXvT+ZnSE8RKnmIgRI0aM\nDo72rBqKESNGjBh5ICYEMWLEiNHBEROCGDFixOjgiAlBjBgxYnRwxIQgRowYMTo4YkIQo0UQESUi\nNxu/LxGRBTuo7l+LyLE7oq7tPGeWl2Hyicj5SSLy4M5+fkvbtROeUysiD4rIMhF5VUQe8s73EZGv\ntqK+ChE5d8e3NMauQkwIYrQUW4GjRaRLWzfEhIi4LSh+OnCuUuqgndUeEyKSyLNoxna14P58cR06\nKdowpdRgdEZNgD5AiwiB1+8VQEwIdmPEhCBGS9GE3qv1ouiFKEcvIpu8/5NE5CkR+b2IvC4iN4jI\nHBFZInrPhD2NaqaIyGKv3GHe/a6I3CQiz3sJ0c4y6n1CRH6LDuiJtucEr/6XReRG79zV6MCh20Tk\npgzv10lE/igiK0TkTi/aFBE5WERe8Oq7XUQKvfNv+0RRREaJyJPe8QIR+bmIPAL8RkSGeO/7ovcO\n/SJttdolIqeIyB9E5AF08kDxzr/steH4VvStj+7oIDIAlFIveYc3AAd4bbzIkxAWi8g/vb9xWfr9\nBmBP775MfRrjy462jmiL/3avP2ATUI7e76AzcAmwwLv2a+BYs6z3fxLwOXoBKgTWANd6174B/Mi4\n/y9oBqUferEqAs4EvumVKQSWovPiTwI2A30ztLMOeBeoARLA48CR3rUnyRA97dW3Hp0bxgH+D704\nF6GzSfb3yv0GnZAMjH0fgFHAk97xAuAfQLH3+yfAHO+4wD8feX7QLuAU7/39KNZj0OmMXaDWe7fu\n+fZt5DnTvHueQEcM1xnv/6BRrgQo8o77AUuNckG/oyWJl6PPif92n79YIojRYiidkfE3wAUtuO15\npfO9b0WH5D/inV+OXkh8/F4plVJKrUJv6DEQnatlroi8iE4LXI1emACWKJ0jP4rR6EX5Y6VUE3An\nejOW7WGJUmq10onXXvTaNgB4Syn1ulfmjjzrul8ptcU7/j/gShG5HOhtnM+Fvyql/D0DJgB3KZ0U\n7kPgKe8dIf++BUAptQidhuQX6P59QURqMjw/CfxCRJYDf0Bv0uIjW7/H2A0RE4IYrcWP0DrtUuNc\nE96Y8lQqBca1rcZxyvidQnPsPqI5T/y0vl9XSg33/voqpfzFbnOW9mVKBZwPzHY2e23LVVfwzmjJ\nwUTQNqXUb4HDgS3AIhGZnEdbzHfL1YZ8+zaA0pum/FYpdRI6eVwmwnYR8CEwDC3tmN8zW7/H2A0R\nE4IYrYLHqf4eTQx8vA2M9I6PQHOULcUsEXE83XY9OsnZIuAc0SmCEZH+orOq5sJzwIEi0sUzaJ6A\n5qJbgxVAHxHZy/t9klHX24TvfEy2CkSkHnhTKfVjdEK3fVrYhqeB4z17SQ164V7Swjr8tkwWkRLv\nuAy9B8G7wEb0Fo0+OgPve9LRSWi1VCZE74uxmyEmBDH+HdwMmN5Dv0AvvkuA/Wgd17gSvcg+DJyt\nlGoAfgm8CvxT9GbrPyMLp+tD6XTAV6D14MuAfyql/pTrnhx1NQCnAn/w1CQp4Dbv8rXALSKyGC1B\nZMPxwMueemsgWrXWEtyHzoq5DG3vuEwp9UEL6/AxElgqIi+hVVa/VEo979Xf5LmVXgT8FDhZRJ4F\n+pPleyqlPgWe8QzZsbF4N0ScfTRGjBgxOjhiiSBGjBgxOjhiQhAjRowYHRwxIYgRI0aMDo6YEMSI\nESNGB0dMCGLEiBGjgyMmBDFixIjRwRETghgxYsTo4Ph/HvN4tv6EC48AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f61f17dcbe0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualising Result for the junctions\n",
    "for junction in range(4):\n",
    "    plt.figure\n",
    "    plt.plot(y_truth.values[:,junction], color = 'green', label = 'Real traffic')\n",
    "    plt.plot(y_pred[:,junction], color = 'red', label = 'Predicted traffic')\n",
    "    plt.title('Traffic Forecasting at junction %i' % (junction+1))\n",
    "    plt.xlabel('Number of hours from Start')\n",
    "    plt.ylabel('Traffic')\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:17.057560Z",
     "start_time": "2017-11-18T07:58:17.030796Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "from math import sqrt\n",
    "\n",
    "def rmse(y_true, y_pred):\n",
    "    return sqrt(mean_squared_error(y_true, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-18T07:58:17.061791Z",
     "start_time": "2017-11-18T07:58:17.058889Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5.812450286303113"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rmse(y_truth, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For predictions we will have to do a different pipeline since we don't have the previous hour traffic like in validation.\n",
    "\n",
    "- We predict step by step and previous prediction is input of the next one\n",
    "- We create 6 months min lag features\n",
    "- We predict 3 months of traffic at each point (rolling forecast)"
   ]
  }
 ],
 "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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}