notebooks/Exponential Smoothing and ARIMA on Real Data.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exponential Smoothing Real Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: seaborn in /opt/app-root/lib/python3.6/site-packages (0.8.1)\n",
      "Requirement already up-to-date: statsmodels in /opt/app-root/lib/python3.6/site-packages (0.9.0)\n",
      "Python version 3.6.3 (default, Mar 20 2018, 13:50:41) \n",
      "[GCC 4.8.5 20150623 (Red Hat 4.8.5-16)]\n",
      "Spark version: 2.2.1\n"
     ]
    }
   ],
   "source": [
    "# install and load necessary packages\n",
    "!pip install seaborn\n",
    "!pip install --upgrade --no-deps statsmodels\n",
    "\n",
    "import pyspark\n",
    "from datetime import datetime\n",
    "import seaborn as sns\n",
    "import sys\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "print('Python version ' + sys.version)\n",
    "print('Spark version: ' + pyspark.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "env: DH_CEPH_KEY=DTG5R3EEWN9JBYJZH0DF\n",
      "env: DH_CEPH_SECRET=pdcEGFERILlkRDGrCSxdIMaZVtNCOKvYP4Gf2b2x\n",
      "env: DH_CEPH_HOST=http://storage-016.infra.prod.upshift.eng.rdu2.redhat.com:8080\n",
      "env: METRIC_NAME=kubelet_docker_operations_latency_microseconds\n",
      "env: LABEL=\n",
      "kubelet_docker_operations_latency_microseconds\n"
     ]
    }
   ],
   "source": [
    "%env DH_CEPH_KEY = DTG5R3EEWN9JBYJZH0DF\n",
    "%env DH_CEPH_SECRET = pdcEGFERILlkRDGrCSxdIMaZVtNCOKvYP4Gf2b2x\n",
    "%env DH_CEPH_HOST = http://storage-016.infra.prod.upshift.eng.rdu2.redhat.com:8080\n",
    "%env METRIC_NAME = kubelet_docker_operations_latency_microseconds\n",
    "%env LABEL =\n",
    "\n",
    "label = os.getenv(\"LABEL\")\n",
    "where_labels = {}#\"metric.group=route.openshift.io\"}\n",
    "metric_name = str(os.getenv(\"METRIC_NAME\"))\n",
    "print(metric_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Establish Connection to Spark Cluster\n",
    "\n",
    "set configuration so that the Spark Cluster communicates with Ceph and reads a chunk of data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Application Name:  G2T0 - Ceph S3 Prometheus JSON Reader\n"
     ]
    }
   ],
   "source": [
    "import string \n",
    "import random\n",
    "\n",
    "# Set the configuration\n",
    "# random string for instance name\n",
    "inst = ''.join(random.choices(string.ascii_uppercase + string.digits, k=4))\n",
    "AppName = inst + ' - Ceph S3 Prometheus JSON Reader'\n",
    "conf = pyspark.SparkConf().setAppName(AppName).setMaster('spark://spark-cluster.dh-prod-analytics-factory.svc:7077')\n",
    "print(\"Application Name: \", AppName)\n",
    "\n",
    "# specify number of nodes need (1-5)\n",
    "conf.set(\"spark.cores.max\", \"88\")\n",
    "\n",
    "# specify Spark executor memory (default is 1gB)\n",
    "conf.set(\"spark.executor.memory\", \"400g\")\n",
    "\n",
    "# Set the Spark cluster connection\n",
    "sc = pyspark.SparkContext.getOrCreate(conf) \n",
    "\n",
    "# Set the Hadoop configurations to access Ceph S3\n",
    "import os\n",
    "(ceph_key, ceph_secret, ceph_host) = (os.getenv('DH_CEPH_KEY'), os.getenv('DH_CEPH_SECRET'), os.getenv('DH_CEPH_HOST'))\n",
    "ceph_key = 'DTG5R3EEWN9JBYJZH0DF'\n",
    "ceph_secret = 'pdcEGFERILlkRDGrCSxdIMaZVtNCOKvYP4Gf2b2x'\n",
    "ceph_host = 'http://storage-016.infra.prod.upshift.eng.rdu2.redhat.com:8080'\n",
    "sc._jsc.hadoopConfiguration().set(\"fs.s3a.access.key\", ceph_key) \n",
    "sc._jsc.hadoopConfiguration().set(\"fs.s3a.secret.key\", ceph_secret) \n",
    "sc._jsc.hadoopConfiguration().set(\"fs.s3a.endpoint\", ceph_host) \n",
    "\n",
    "#Get the SQL context\n",
    "sqlContext = pyspark.SQLContext(sc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Read the Prometheus JSON BZip data\n",
    "jsonUrl = \"s3a://DH-DEV-PROMETHEUS-BACKUP/prometheus-openshift-devops-monitor.1b7d.free-stg.openshiftapps.com/\"+metric_name+\"/\"\n",
    "\n",
    "jsonFile = sqlContext.read.option(\"multiline\", True).option(\"mode\", \"PERMISSIVE\").json(jsonUrl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pyspark.sql.functions as F\n",
    "from pyspark.sql.types import StringType\n",
    "from pyspark.sql.types import IntegerType\n",
    "from pyspark.sql.types import TimestampType\n",
    "\n",
    "# create function to convert POSIX timestamp to local date\n",
    "def convert_timestamp(t):\n",
    "    return datetime.fromtimestamp(float(t))\n",
    "\n",
    "def format_df(df):\n",
    "    #reformat data by timestamp and values\n",
    "    df = df.withColumn(\"values\", F.explode(df.values))\n",
    "    \n",
    "    df = df.withColumn(\"timestamp\", F.col(\"values\").getItem(0))\n",
    "    \n",
    "    df = df.withColumn(\"values\", F.col(\"values\").getItem(1))\n",
    "\n",
    "    # drop null values\n",
    "    df = df.na.drop(subset=[\"values\"])\n",
    "    \n",
    "    # cast values to int\n",
    "    df = df.withColumn(\"values\", df.values.cast(\"int\"))\n",
    "    #df = df.withColumn(\"timestamp\", df.values.cast(\"int\"))\n",
    "\n",
    "    # define function to be applied to DF column\n",
    "    udf_convert_timestamp = F.udf(lambda z: convert_timestamp(z), TimestampType())\n",
    "\n",
    "    df = df.na.drop(subset=[\"timestamp\"])\n",
    "    # convert timestamp values to datetime timestamp\n",
    "    df = df.withColumn(\"timestamp\", udf_convert_timestamp(\"timestamp\"))\n",
    "\n",
    "    # drop null values\n",
    "    df = df.na.drop(subset=[\"values\"])\n",
    "    \n",
    "    # calculate log(values) for each row\n",
    "    #df = df.withColumn(\"log_values\", F.log(df.values))\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def extract_from_json(json, name, select_labels, where_labels):\n",
    "    #Register the created SchemaRDD as a temporary variable\n",
    "    json.registerTempTable(name)\n",
    "    \n",
    "    #Filter the results into a data frame\n",
    "\n",
    "    query = \"SELECT values\"\n",
    "    \n",
    "    # check if select labels are specified and add query condition if appropriate\n",
    "    if len(select_labels) > 0:\n",
    "        query = query + \", \" + \", \".join(select_labels)\n",
    "        \n",
    "    query = query + \" FROM \" + name\n",
    "    \n",
    "    # check if where labels are specified and add query condition if appropriate\n",
    "    if len(where_labels) > 0:\n",
    "        query = query + \" WHERE \" + \" AND \".join(where_labels)\n",
    "\n",
    "    print(\"SQL QUERRY: \", query)\n",
    "    df = sqlContext.sql(query)\n",
    "\n",
    "    #sample data to make it more manageable\n",
    "    #data = data.sample(False, fraction = 0.05, seed = 0)\n",
    "    # TODO: get rid of this hack\n",
    "    #df = sqlContext.createDataFrame(df.head(1000), df.schema)\n",
    "    \n",
    "    return format_df(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SQL QUERRY:  SELECT values FROM kubelet_docker_operations_latency_microseconds WHERE metric.quantile='0.9' AND metric.hostname='free-stg-master-03fb6'\n"
     ]
    }
   ],
   "source": [
    "if label != \"\":\n",
    "    select_labels = ['metric.' + label]\n",
    "else:\n",
    "    select_labels = []\n",
    "    \n",
    "where_labels = {\"metric.quantile='0.9'\",\"metric.hostname='free-stg-master-03fb6'\"}\n",
    "# get data and format\n",
    "df = extract_from_json(jsonFile, metric_name, select_labels, where_labels)\n",
    "select_labels = []\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_pd = df.toPandas()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>values</th>\n",
       "      <th>timestamp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:00:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:01:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:02:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:03:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:04:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:05:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:06:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:07:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:08:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:09:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:10:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>175890</td>\n",
       "      <td>2018-05-30 15:15:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:16:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:17:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:18:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:19:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:20:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:21:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:22:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:23:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:24:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>31374</td>\n",
       "      <td>2018-05-30 15:25:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>31374</td>\n",
       "      <td>2018-05-30 15:26:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>27546</td>\n",
       "      <td>2018-05-30 15:27:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>27546</td>\n",
       "      <td>2018-05-30 15:28:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>25720</td>\n",
       "      <td>2018-05-30 15:29:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>41214</td>\n",
       "      <td>2018-05-30 15:30:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>41214</td>\n",
       "      <td>2018-05-30 15:31:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>45085</td>\n",
       "      <td>2018-05-30 15:32:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>41214</td>\n",
       "      <td>2018-05-30 15:33:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621736</th>\n",
       "      <td>1882</td>\n",
       "      <td>2018-02-12 12:23:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621737</th>\n",
       "      <td>2219</td>\n",
       "      <td>2018-02-12 12:47:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621738</th>\n",
       "      <td>1933</td>\n",
       "      <td>2018-02-12 13:11:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621739</th>\n",
       "      <td>1920</td>\n",
       "      <td>2018-02-12 13:35:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621740</th>\n",
       "      <td>1977</td>\n",
       "      <td>2018-02-12 13:59:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621741</th>\n",
       "      <td>2386</td>\n",
       "      <td>2018-02-12 14:23:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621742</th>\n",
       "      <td>1984</td>\n",
       "      <td>2018-02-12 14:47:22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621743</th>\n",
       "      <td>1856</td>\n",
       "      <td>2018-02-12 15:11:21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621744</th>\n",
       "      <td>1750</td>\n",
       "      <td>2018-02-12 15:35:20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621745</th>\n",
       "      <td>2101</td>\n",
       "      <td>2018-02-12 15:59:19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621746</th>\n",
       "      <td>2016</td>\n",
       "      <td>2018-02-12 16:23:18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621747</th>\n",
       "      <td>1995</td>\n",
       "      <td>2018-02-12 16:47:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621748</th>\n",
       "      <td>1920</td>\n",
       "      <td>2018-02-12 17:11:16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621749</th>\n",
       "      <td>3108</td>\n",
       "      <td>2018-02-12 17:35:15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621750</th>\n",
       "      <td>1798</td>\n",
       "      <td>2018-02-12 17:59:14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621751</th>\n",
       "      <td>2466</td>\n",
       "      <td>2018-02-12 18:23:13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621752</th>\n",
       "      <td>1808</td>\n",
       "      <td>2018-02-12 18:47:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621753</th>\n",
       "      <td>1569</td>\n",
       "      <td>2018-02-12 19:11:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621754</th>\n",
       "      <td>3027</td>\n",
       "      <td>2018-02-12 19:35:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621755</th>\n",
       "      <td>1780</td>\n",
       "      <td>2018-02-12 19:59:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621756</th>\n",
       "      <td>2233</td>\n",
       "      <td>2018-02-12 20:23:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621757</th>\n",
       "      <td>1516</td>\n",
       "      <td>2018-02-12 20:47:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621758</th>\n",
       "      <td>1281</td>\n",
       "      <td>2018-02-12 21:11:06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621759</th>\n",
       "      <td>1292</td>\n",
       "      <td>2018-02-12 21:35:05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621760</th>\n",
       "      <td>4393</td>\n",
       "      <td>2018-02-12 21:59:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621761</th>\n",
       "      <td>6346</td>\n",
       "      <td>2018-02-12 22:23:03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621762</th>\n",
       "      <td>4209</td>\n",
       "      <td>2018-02-12 22:47:02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621763</th>\n",
       "      <td>3781</td>\n",
       "      <td>2018-02-12 23:11:01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621764</th>\n",
       "      <td>3014</td>\n",
       "      <td>2018-02-12 23:35:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621765</th>\n",
       "      <td>2432</td>\n",
       "      <td>2018-02-12 23:58:59</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>621766 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        values           timestamp\n",
       "0        36762 2018-05-30 15:00:59\n",
       "1        36762 2018-05-30 15:01:59\n",
       "2        36762 2018-05-30 15:02:59\n",
       "3        36762 2018-05-30 15:03:59\n",
       "4        36762 2018-05-30 15:04:59\n",
       "5        36762 2018-05-30 15:05:59\n",
       "6        36762 2018-05-30 15:06:59\n",
       "7        36762 2018-05-30 15:07:59\n",
       "8        36762 2018-05-30 15:08:59\n",
       "9        36762 2018-05-30 15:09:59\n",
       "10       36762 2018-05-30 15:10:59\n",
       "11      175890 2018-05-30 15:15:59\n",
       "12      165059 2018-05-30 15:16:59\n",
       "13      165059 2018-05-30 15:17:59\n",
       "14      165059 2018-05-30 15:18:59\n",
       "15      165059 2018-05-30 15:19:59\n",
       "16      165059 2018-05-30 15:20:59\n",
       "17      165059 2018-05-30 15:21:59\n",
       "18      165059 2018-05-30 15:22:59\n",
       "19      165059 2018-05-30 15:23:59\n",
       "20      165059 2018-05-30 15:24:59\n",
       "21       31374 2018-05-30 15:25:59\n",
       "22       31374 2018-05-30 15:26:59\n",
       "23       27546 2018-05-30 15:27:59\n",
       "24       27546 2018-05-30 15:28:59\n",
       "25       25720 2018-05-30 15:29:59\n",
       "26       41214 2018-05-30 15:30:59\n",
       "27       41214 2018-05-30 15:31:59\n",
       "28       45085 2018-05-30 15:32:59\n",
       "29       41214 2018-05-30 15:33:59\n",
       "...        ...                 ...\n",
       "621736    1882 2018-02-12 12:23:28\n",
       "621737    2219 2018-02-12 12:47:27\n",
       "621738    1933 2018-02-12 13:11:26\n",
       "621739    1920 2018-02-12 13:35:25\n",
       "621740    1977 2018-02-12 13:59:24\n",
       "621741    2386 2018-02-12 14:23:23\n",
       "621742    1984 2018-02-12 14:47:22\n",
       "621743    1856 2018-02-12 15:11:21\n",
       "621744    1750 2018-02-12 15:35:20\n",
       "621745    2101 2018-02-12 15:59:19\n",
       "621746    2016 2018-02-12 16:23:18\n",
       "621747    1995 2018-02-12 16:47:17\n",
       "621748    1920 2018-02-12 17:11:16\n",
       "621749    3108 2018-02-12 17:35:15\n",
       "621750    1798 2018-02-12 17:59:14\n",
       "621751    2466 2018-02-12 18:23:13\n",
       "621752    1808 2018-02-12 18:47:12\n",
       "621753    1569 2018-02-12 19:11:11\n",
       "621754    3027 2018-02-12 19:35:10\n",
       "621755    1780 2018-02-12 19:59:09\n",
       "621756    2233 2018-02-12 20:23:08\n",
       "621757    1516 2018-02-12 20:47:07\n",
       "621758    1281 2018-02-12 21:11:06\n",
       "621759    1292 2018-02-12 21:35:05\n",
       "621760    4393 2018-02-12 21:59:04\n",
       "621761    6346 2018-02-12 22:23:03\n",
       "621762    4209 2018-02-12 22:47:02\n",
       "621763    3781 2018-02-12 23:11:01\n",
       "621764    3014 2018-02-12 23:35:00\n",
       "621765    2432 2018-02-12 23:58:59\n",
       "\n",
       "[621766 rows x 2 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pd = df_pd[[\"values\",\"timestamp\"]]\n",
    "df_pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "values                int64\n",
       "timestamp    datetime64[ns]\n",
       "dtype: object"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pd.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_pd.sort_values(by='timestamp')\n",
    "#df_pd = df_pd.set_index(\"timestamp\")\n",
    "#df_pd.set_index(\"timestamp\")\n",
    "train_frame = df_pd[0 : int(0.7*len(df_pd))]\n",
    "test_frame = df_pd[int(0.7*len(df_pd)) : ]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>values</th>\n",
       "      <th>timestamp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:00:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:01:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:02:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:03:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:04:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:05:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:06:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:07:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:08:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:09:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>36762</td>\n",
       "      <td>2018-05-30 15:10:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>175890</td>\n",
       "      <td>2018-05-30 15:15:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:16:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:17:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:18:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:19:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:20:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:21:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:22:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:23:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>165059</td>\n",
       "      <td>2018-05-30 15:24:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>31374</td>\n",
       "      <td>2018-05-30 15:25:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>31374</td>\n",
       "      <td>2018-05-30 15:26:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>27546</td>\n",
       "      <td>2018-05-30 15:27:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>27546</td>\n",
       "      <td>2018-05-30 15:28:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>25720</td>\n",
       "      <td>2018-05-30 15:29:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>41214</td>\n",
       "      <td>2018-05-30 15:30:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>41214</td>\n",
       "      <td>2018-05-30 15:31:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>45085</td>\n",
       "      <td>2018-05-30 15:32:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>41214</td>\n",
       "      <td>2018-05-30 15:33:59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621736</th>\n",
       "      <td>1882</td>\n",
       "      <td>2018-02-12 12:23:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621737</th>\n",
       "      <td>2219</td>\n",
       "      <td>2018-02-12 12:47:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621738</th>\n",
       "      <td>1933</td>\n",
       "      <td>2018-02-12 13:11:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621739</th>\n",
       "      <td>1920</td>\n",
       "      <td>2018-02-12 13:35:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621740</th>\n",
       "      <td>1977</td>\n",
       "      <td>2018-02-12 13:59:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621741</th>\n",
       "      <td>2386</td>\n",
       "      <td>2018-02-12 14:23:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621742</th>\n",
       "      <td>1984</td>\n",
       "      <td>2018-02-12 14:47:22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621743</th>\n",
       "      <td>1856</td>\n",
       "      <td>2018-02-12 15:11:21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621744</th>\n",
       "      <td>1750</td>\n",
       "      <td>2018-02-12 15:35:20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621745</th>\n",
       "      <td>2101</td>\n",
       "      <td>2018-02-12 15:59:19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621746</th>\n",
       "      <td>2016</td>\n",
       "      <td>2018-02-12 16:23:18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621747</th>\n",
       "      <td>1995</td>\n",
       "      <td>2018-02-12 16:47:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621748</th>\n",
       "      <td>1920</td>\n",
       "      <td>2018-02-12 17:11:16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621749</th>\n",
       "      <td>3108</td>\n",
       "      <td>2018-02-12 17:35:15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621750</th>\n",
       "      <td>1798</td>\n",
       "      <td>2018-02-12 17:59:14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621751</th>\n",
       "      <td>2466</td>\n",
       "      <td>2018-02-12 18:23:13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621752</th>\n",
       "      <td>1808</td>\n",
       "      <td>2018-02-12 18:47:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621753</th>\n",
       "      <td>1569</td>\n",
       "      <td>2018-02-12 19:11:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621754</th>\n",
       "      <td>3027</td>\n",
       "      <td>2018-02-12 19:35:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621755</th>\n",
       "      <td>1780</td>\n",
       "      <td>2018-02-12 19:59:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621756</th>\n",
       "      <td>2233</td>\n",
       "      <td>2018-02-12 20:23:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621757</th>\n",
       "      <td>1516</td>\n",
       "      <td>2018-02-12 20:47:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621758</th>\n",
       "      <td>1281</td>\n",
       "      <td>2018-02-12 21:11:06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621759</th>\n",
       "      <td>1292</td>\n",
       "      <td>2018-02-12 21:35:05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621760</th>\n",
       "      <td>4393</td>\n",
       "      <td>2018-02-12 21:59:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621761</th>\n",
       "      <td>6346</td>\n",
       "      <td>2018-02-12 22:23:03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621762</th>\n",
       "      <td>4209</td>\n",
       "      <td>2018-02-12 22:47:02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621763</th>\n",
       "      <td>3781</td>\n",
       "      <td>2018-02-12 23:11:01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621764</th>\n",
       "      <td>3014</td>\n",
       "      <td>2018-02-12 23:35:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>621765</th>\n",
       "      <td>2432</td>\n",
       "      <td>2018-02-12 23:58:59</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>621766 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        values           timestamp\n",
       "0        36762 2018-05-30 15:00:59\n",
       "1        36762 2018-05-30 15:01:59\n",
       "2        36762 2018-05-30 15:02:59\n",
       "3        36762 2018-05-30 15:03:59\n",
       "4        36762 2018-05-30 15:04:59\n",
       "5        36762 2018-05-30 15:05:59\n",
       "6        36762 2018-05-30 15:06:59\n",
       "7        36762 2018-05-30 15:07:59\n",
       "8        36762 2018-05-30 15:08:59\n",
       "9        36762 2018-05-30 15:09:59\n",
       "10       36762 2018-05-30 15:10:59\n",
       "11      175890 2018-05-30 15:15:59\n",
       "12      165059 2018-05-30 15:16:59\n",
       "13      165059 2018-05-30 15:17:59\n",
       "14      165059 2018-05-30 15:18:59\n",
       "15      165059 2018-05-30 15:19:59\n",
       "16      165059 2018-05-30 15:20:59\n",
       "17      165059 2018-05-30 15:21:59\n",
       "18      165059 2018-05-30 15:22:59\n",
       "19      165059 2018-05-30 15:23:59\n",
       "20      165059 2018-05-30 15:24:59\n",
       "21       31374 2018-05-30 15:25:59\n",
       "22       31374 2018-05-30 15:26:59\n",
       "23       27546 2018-05-30 15:27:59\n",
       "24       27546 2018-05-30 15:28:59\n",
       "25       25720 2018-05-30 15:29:59\n",
       "26       41214 2018-05-30 15:30:59\n",
       "27       41214 2018-05-30 15:31:59\n",
       "28       45085 2018-05-30 15:32:59\n",
       "29       41214 2018-05-30 15:33:59\n",
       "...        ...                 ...\n",
       "621736    1882 2018-02-12 12:23:28\n",
       "621737    2219 2018-02-12 12:47:27\n",
       "621738    1933 2018-02-12 13:11:26\n",
       "621739    1920 2018-02-12 13:35:25\n",
       "621740    1977 2018-02-12 13:59:24\n",
       "621741    2386 2018-02-12 14:23:23\n",
       "621742    1984 2018-02-12 14:47:22\n",
       "621743    1856 2018-02-12 15:11:21\n",
       "621744    1750 2018-02-12 15:35:20\n",
       "621745    2101 2018-02-12 15:59:19\n",
       "621746    2016 2018-02-12 16:23:18\n",
       "621747    1995 2018-02-12 16:47:17\n",
       "621748    1920 2018-02-12 17:11:16\n",
       "621749    3108 2018-02-12 17:35:15\n",
       "621750    1798 2018-02-12 17:59:14\n",
       "621751    2466 2018-02-12 18:23:13\n",
       "621752    1808 2018-02-12 18:47:12\n",
       "621753    1569 2018-02-12 19:11:11\n",
       "621754    3027 2018-02-12 19:35:10\n",
       "621755    1780 2018-02-12 19:59:09\n",
       "621756    2233 2018-02-12 20:23:08\n",
       "621757    1516 2018-02-12 20:47:07\n",
       "621758    1281 2018-02-12 21:11:06\n",
       "621759    1292 2018-02-12 21:35:05\n",
       "621760    4393 2018-02-12 21:59:04\n",
       "621761    6346 2018-02-12 22:23:03\n",
       "621762    4209 2018-02-12 22:47:02\n",
       "621763    3781 2018-02-12 23:11:01\n",
       "621764    3014 2018-02-12 23:35:00\n",
       "621765    2432 2018-02-12 23:58:59\n",
       "\n",
       "[621766 rows x 2 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "sc.stop()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "values                int64\n",
       "timestamp    datetime64[ns]\n",
       "dtype: object"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pd.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_pd_trimmed = df_pd[df_pd.timestamp > datetime(2018,6,17,3,14)]\n",
    "df_pd_trimmed = df_pd_trimmed[df_pd_trimmed.timestamp < datetime(2018,6,21,3,14)]\n",
    "#df_pd_trimmed = df_pd_trimmed[[\"values\"]]\n",
    "train_frame = df_pd_trimmed[0 : int(0.7*len(df_pd_trimmed))]\n",
    "test_frame = df_pd_trimmed[int(0.7*len(df_pd_trimmed)) : ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>values</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>82766</th>\n",
       "      <td>34347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82767</th>\n",
       "      <td>34347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82768</th>\n",
       "      <td>34347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82769</th>\n",
       "      <td>34347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82770</th>\n",
       "      <td>34347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82771</th>\n",
       "      <td>34347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82772</th>\n",
       "      <td>34347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82773</th>\n",
       "      <td>34347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82774</th>\n",
       "      <td>34347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82775</th>\n",
       "      <td>5856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82776</th>\n",
       "      <td>5856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82777</th>\n",
       "      <td>5856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82778</th>\n",
       "      <td>5856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82779</th>\n",
       "      <td>5856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82780</th>\n",
       "      <td>5856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82781</th>\n",
       "      <td>5856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82782</th>\n",
       "      <td>5856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82783</th>\n",
       "      <td>5856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82784</th>\n",
       "      <td>2024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82785</th>\n",
       "      <td>2024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82786</th>\n",
       "      <td>2024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82787</th>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82788</th>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82789</th>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82790</th>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82791</th>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82792</th>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82793</th>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82794</th>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82795</th>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220110</th>\n",
       "      <td>53722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220111</th>\n",
       "      <td>53722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220112</th>\n",
       "      <td>53722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220113</th>\n",
       "      <td>57187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220114</th>\n",
       "      <td>53289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220115</th>\n",
       "      <td>57187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220116</th>\n",
       "      <td>65487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220117</th>\n",
       "      <td>63789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220118</th>\n",
       "      <td>59313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220119</th>\n",
       "      <td>58443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220120</th>\n",
       "      <td>58443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220121</th>\n",
       "      <td>58443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220122</th>\n",
       "      <td>55172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220123</th>\n",
       "      <td>55120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220124</th>\n",
       "      <td>50567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220125</th>\n",
       "      <td>50775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220126</th>\n",
       "      <td>47531</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220127</th>\n",
       "      <td>48099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220128</th>\n",
       "      <td>50046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220129</th>\n",
       "      <td>50046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220130</th>\n",
       "      <td>51074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220131</th>\n",
       "      <td>51483</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220132</th>\n",
       "      <td>64351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220133</th>\n",
       "      <td>64351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220134</th>\n",
       "      <td>65219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220135</th>\n",
       "      <td>65219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220136</th>\n",
       "      <td>67439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220137</th>\n",
       "      <td>65278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220138</th>\n",
       "      <td>64181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220139</th>\n",
       "      <td>64181</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15701 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        values\n",
       "82766    34347\n",
       "82767    34347\n",
       "82768    34347\n",
       "82769    34347\n",
       "82770    34347\n",
       "82771    34347\n",
       "82772    34347\n",
       "82773    34347\n",
       "82774    34347\n",
       "82775     5856\n",
       "82776     5856\n",
       "82777     5856\n",
       "82778     5856\n",
       "82779     5856\n",
       "82780     5856\n",
       "82781     5856\n",
       "82782     5856\n",
       "82783     5856\n",
       "82784     2024\n",
       "82785     2024\n",
       "82786     2024\n",
       "82787     1903\n",
       "82788     1903\n",
       "82789     1903\n",
       "82790     1903\n",
       "82791     1903\n",
       "82792     1903\n",
       "82793     1903\n",
       "82794     1903\n",
       "82795     1903\n",
       "...        ...\n",
       "220110   53722\n",
       "220111   53722\n",
       "220112   53722\n",
       "220113   57187\n",
       "220114   53289\n",
       "220115   57187\n",
       "220116   65487\n",
       "220117   63789\n",
       "220118   59313\n",
       "220119   58443\n",
       "220120   58443\n",
       "220121   58443\n",
       "220122   55172\n",
       "220123   55120\n",
       "220124   50567\n",
       "220125   50775\n",
       "220126   47531\n",
       "220127   48099\n",
       "220128   50046\n",
       "220129   50046\n",
       "220130   51074\n",
       "220131   51483\n",
       "220132   64351\n",
       "220133   64351\n",
       "220134   65219\n",
       "220135   65219\n",
       "220136   67439\n",
       "220137   65278\n",
       "220138   64181\n",
       "220139   64181\n",
       "\n",
       "[15701 rows x 1 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_frame"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exponential Smoothing\n",
    "\n",
    "inspiration: https://github.com/statsmodels/statsmodels/blob/master/examples/notebooks/exponential_smoothing.ipynb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: patsy in /opt/app-root/lib/python3.6/site-packages (0.5.0)\r\n",
      "Requirement already satisfied: six in /opt/app-root/lib/python3.6/site-packages (from patsy) (1.11.0)\r\n",
      "Requirement already satisfied: numpy>=1.4 in /opt/app-root/lib/python3.6/site-packages (from patsy) (1.14.5)\r\n"
     ]
    }
   ],
   "source": [
    "!pip install patsy\n",
    "from statsmodels.tsa.api import ExponentialSmoothing, SimpleExpSmoothing, Holt\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_frame = train_frame.reset_index()\n",
    "df_series = pd.Series(train_frame[\"values\"])\n",
    "#df_series.index = pd.DatetimeIndex(train_frame.index, freq=\"N\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        34347\n",
       "1        34347\n",
       "2        34347\n",
       "3        34347\n",
       "4        34347\n",
       "5        34347\n",
       "6        34347\n",
       "7        34347\n",
       "8        34347\n",
       "9         5856\n",
       "10        5856\n",
       "11        5856\n",
       "12        5856\n",
       "13        5856\n",
       "14        5856\n",
       "15        5856\n",
       "16        5856\n",
       "17        5856\n",
       "18        2024\n",
       "19        2024\n",
       "20        2024\n",
       "21        1903\n",
       "22        1903\n",
       "23        1903\n",
       "24        1903\n",
       "25        1903\n",
       "26        1903\n",
       "27        1903\n",
       "28        1903\n",
       "29        1903\n",
       "         ...  \n",
       "15671    53722\n",
       "15672    53722\n",
       "15673    53722\n",
       "15674    57187\n",
       "15675    53289\n",
       "15676    57187\n",
       "15677    65487\n",
       "15678    63789\n",
       "15679    59313\n",
       "15680    58443\n",
       "15681    58443\n",
       "15682    58443\n",
       "15683    55172\n",
       "15684    55120\n",
       "15685    50567\n",
       "15686    50775\n",
       "15687    47531\n",
       "15688    48099\n",
       "15689    50046\n",
       "15690    50046\n",
       "15691    51074\n",
       "15692    51483\n",
       "15693    64351\n",
       "15694    64351\n",
       "15695    65219\n",
       "15696    65219\n",
       "15697    67439\n",
       "15698    65278\n",
       "15699    64181\n",
       "15700    64181\n",
       "Name: values, Length: 15701, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_series"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax=df_series.plot(title=\"Original Data\")\n",
    "ax.set_ylabel(\"value\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Simple Exponential Smoothing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0,'timestamp')"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fit1  =  SimpleExpSmoothing(df_series).fit()\n",
    "fcast1 = fit1.forecast(600)\n",
    "ax = df_series.plot(marker='o', color='black',legend=True, label=\"Prom data\", figsize=(12,8))\n",
    "fcast1.plot(marker='o', markersize=0.2, ax=ax, color='blue', legend=True, label=\"Predicted\")\n",
    "fit1.fittedvalues.plot(marker='o', markersize=0.2, ax=ax, color='green',legend=True, label=\"ES fitted values\")\n",
    "plt.title(\"Simple Exponential Smoothing on Real Data\")\n",
    "plt.ylabel(\"value\")\n",
    "plt.xlabel(\"timestamp\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Holt Winters Method (Additive)\n",
    "The additive method gives a prediction by adding the seasonality, trend, and value components of the model together. This method is ideal for time series when the seasonality is relatively constant."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fit2 = ExponentialSmoothing(df_series, seasonal_periods=4, trend='add', seasonal='add').fit(use_boxcox=True)\n",
    "results=pd.DataFrame(index=[r\"$\\alpha$\",r\"$\\beta$\",r\"$\\phi$\",r\"$\\gamma$\",r\"$l_0$\",\"$b_0$\",\"SSE\"])\n",
    "params = ['smoothing_level', 'smoothing_slope', 'damping_slope', 'smoothing_seasonal', 'initial_level', 'initial_slope']\n",
    "results[\"Additive\"] = [fit2.params[p] for p in params] + [fit2.sse]\n",
    "ax = df_series.plot(figsize=(10,6), marker='o', color='black',legend=True, label=\"Prom data\")\n",
    "fit2.fittedvalues.plot(ax=ax, style='--', color='green', legend=True, label=\"HW fitted values\")\n",
    "fit2.forecast(5000).plot(ax=ax, style='--', marker='o', color='blue', legend=True, label=\"Predicted\")\n",
    "\n",
    "plt.title(\"Holt Winters Additive method on Real Data\")\n",
    "plt.ylabel(\"value\")\n",
    "plt.xlabel(\"timestamp\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Holt Winter's Method (Multiplicative)\n",
    "This method is used when the trend and seasonality vary with time. This is effective for when the trend is non-stationary (ie. changes with time)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fit2 = ExponentialSmoothing(df_series, seasonal_periods=4, trend='add', seasonal='mult').fit(use_boxcox=True)\n",
    "results=pd.DataFrame(index=[r\"$\\alpha$\",r\"$\\beta$\",r\"$\\phi$\",r\"$\\gamma$\",r\"$l_0$\",\"$b_0$\",\"SSE\"])\n",
    "params = ['smoothing_level', 'smoothing_slope', 'damping_slope', 'smoothing_seasonal', 'initial_level', 'initial_slope']\n",
    "results[\"Additive\"] = [fit2.params[p] for p in params] + [fit2.sse]\n",
    "ax = df_series.plot(figsize=(10,6), marker='o', color='black',legend=True, label=\"Prom data\" )\n",
    "fit2.fittedvalues.plot(ax=ax, style='--', color='green', legend=True, label=\"HW fitted values\")\n",
    "fit2.forecast(5000).plot(ax=ax, style='--', marker='o', color='blue', legend=True, label=\"Predicted\")\n",
    "plt.title(\" Holt-Winters' Multiplicative method\")\n",
    "plt.ylabel(\"value\")\n",
    "plt.xlabel(\"timestamp\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from __future__ import print_function\n",
    "import statsmodels.api as sm\n",
    "from statsmodels.tsa.arima_process import arma_generate_sample\n",
    "np.random.seed(12345)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "arparams = np.array([.75, -.25])\n",
    "maparams = np.array([.65, .35])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "arparams = np.r_[1, -arparams]\n",
    "maparams = np.r_[1, maparams]\n",
    "nobs = 250\n",
    "y = arma_generate_sample(arparams, maparams, nobs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "#dates = sm.tsa.datetools.dates_from_range('1980m1', length=nobs)\n",
    "y = df_series#pd.Series(y, index=dates)\n",
    "arma_mod = sm.tsa.ARMA(y, order=(2,2))\n",
    "arma_res = arma_mod.fit(trend='nc', disp=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                              ARMA Model Results                              \n",
      "==============================================================================\n",
      "Dep. Variable:                 values   No. Observations:                15701\n",
      "Model:                     ARMA(2, 2)   Log Likelihood             -159650.441\n",
      "Method:                       css-mle   S.D. of innovations           6305.157\n",
      "Date:                Fri, 06 Jul 2018   AIC                         319310.881\n",
      "Time:                        16:39:30   BIC                         319349.189\n",
      "Sample:                             0   HQIC                        319323.563\n",
      "                                                                              \n",
      "================================================================================\n",
      "                   coef    std err          z      P>|z|      [0.025      0.975]\n",
      "--------------------------------------------------------------------------------\n",
      "ar.L1.values     1.4139      0.158      8.924      0.000       1.103       1.724\n",
      "ar.L2.values    -0.4237      0.156     -2.719      0.007      -0.729      -0.118\n",
      "ma.L1.values    -0.4332      0.159     -2.732      0.006      -0.744      -0.122\n",
      "ma.L2.values     0.0110      0.008      1.334      0.182      -0.005       0.027\n",
      "                                    Roots                                    \n",
      "=============================================================================\n",
      "                  Real          Imaginary           Modulus         Frequency\n",
      "-----------------------------------------------------------------------------\n",
      "AR.1            1.0175           +0.0000j            1.0175            0.0000\n",
      "AR.2            2.3196           +0.0000j            2.3196            0.0000\n",
      "MA.1            2.4629           +0.0000j            2.4629            0.0000\n",
      "MA.2           36.7783           +0.0000j           36.7783            0.0000\n",
      "-----------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "print(arma_res.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15696    65219\n",
       "15697    67439\n",
       "15698    65278\n",
       "15699    64181\n",
       "15700    64181\n",
       "Name: values, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Time Series Forecasting using the ARIMA model')"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10,8))\n",
    "fig = arma_res.plot_predict(start=8000, end=16000, ax=ax)\n",
    "legend = ax.legend(loc='upper left')\n",
    "plt.title(\"Time Series Forecasting using the ARIMA model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Time Series Forecasting using the ARIMA model')"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10,8))\n",
    "fig = arma_res.plot_predict(start=0, end=25000, ax=ax)\n",
    "legend = ax.legend(loc='upper left')\n",
    "plt.title(\"Time Series Forecasting using the ARIMA model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting fbprophet\n",
      "  Downloading https://files.pythonhosted.org/packages/83/a1/a39be1675a62597e4c0d4fdaeb65b14752b8aa8afc92e90edd91614353ab/fbprophet-0.3.post2.tar.gz\n",
      "Collecting Cython>=0.22 (from fbprophet)\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/6f/79/d8e2cd00bea8156a995fb284ce7b6677c49eccd2d318f73e201a9ce560dc/Cython-0.28.3-cp36-cp36m-manylinux1_x86_64.whl (3.4MB)\n",
      "\u001b[K    100% |████████████████████████████████| 3.4MB 16.6MB/s ta 0:00:01\n",
      "\u001b[?25hCollecting pystan>=2.14 (from fbprophet)\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/46/37/801a5a932e7f1f038542e7c5e4c4010aac19a26ea6bde9534505465f8c8c/pystan-2.17.1.0-cp36-cp36m-manylinux1_x86_64.whl (68.1MB)\n",
      "\u001b[K    100% |████████████████████████████████| 68.1MB 8.9MB/s ta 0:00:0111                      | 4.5MB 94.9MB/s eta 0:00:01\n",
      "\u001b[?25hRequirement already satisfied: numpy>=1.10.0 in /opt/app-root/lib/python3.6/site-packages (from fbprophet) (1.14.5)\n",
      "Requirement already satisfied: pandas>=0.20.1 in /opt/app-root/lib/python3.6/site-packages (from fbprophet) (0.23.0)\n",
      "Requirement already satisfied: matplotlib>=2.0.0 in /opt/app-root/lib/python3.6/site-packages (from fbprophet) (2.2.2)\n",
      "Requirement already satisfied: python-dateutil>=2.5.0 in /opt/app-root/lib/python3.6/site-packages (from pandas>=0.20.1->fbprophet) (2.7.3)\n",
      "Requirement already satisfied: pytz>=2011k in /opt/app-root/lib/python3.6/site-packages (from pandas>=0.20.1->fbprophet) (2018.4)\n",
      "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /opt/app-root/lib/python3.6/site-packages (from matplotlib>=2.0.0->fbprophet) (2.2.0)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in /opt/app-root/lib/python3.6/site-packages (from matplotlib>=2.0.0->fbprophet) (1.0.1)\n",
      "Requirement already satisfied: cycler>=0.10 in /opt/app-root/lib/python3.6/site-packages (from matplotlib>=2.0.0->fbprophet) (0.10.0)\n",
      "Requirement already satisfied: six>=1.10 in /opt/app-root/lib/python3.6/site-packages (from matplotlib>=2.0.0->fbprophet) (1.11.0)\n",
      "Requirement already satisfied: setuptools in /opt/app-root/lib/python3.6/site-packages (from kiwisolver>=1.0.1->matplotlib>=2.0.0->fbprophet) (39.2.0)\n",
      "Installing collected packages: Cython, pystan, fbprophet\n",
      "  Running setup.py install for fbprophet ... \u001b[?25ldone\n",
      "\u001b[?25hSuccessfully installed Cython-0.28.3 fbprophet-0.3.post2 pystan-2.17.1.0\n"
     ]
    }
   ],
   "source": [
    "!pip install fbprophet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fbprophet import Prophet"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Forecasting __list_images__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_frame = get_filtered_op_frame(OP_TYPE)\n",
    "temp_frame = temp_frame.set_index(temp_frame.timestamp)\n",
    "temp_frame = temp_frame[['timestamp','value']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Separating train and test frame\n",
    "Separation is being done to check the forecast quality and residual plot and its distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_frame = temp_frame[0 : int(0.7*len(temp_frame))]\n",
    "test_frame = temp_frame[int(0.7*len(temp_frame)) : ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(len(train_frame), len(test_frame), len(temp_frame))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/app-root/lib/python3.6/site-packages/ipykernel_launcher.py:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \n",
      "/opt/app-root/lib/python3.6/site-packages/ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    }
   ],
   "source": [
    "train_p=train_frame\n",
    "train_p['y'] = train_frame['values']\n",
    "train_p['ds'] = train_frame['timestamp']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_p=test_frame\n",
    "train_p['y'] = train_frame['values']\n",
    "train_p['ds'] = train_frame['timestamp']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Initialisation of model\n",
    "We can add seasonality when initialising the model. Its little hard to understand but, inital approach will show some seaonality in forecast plot. Seasonality can be week, daily month or yearly\n",
    "* We can add additinal features as well for better forecasting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "m = Prophet()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model fitting\n",
    "It is being trained on training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:fbprophet.forecaster:Disabling yearly seasonality. Run prophet with yearly_seasonality=True to override this.\n",
      "INFO:fbprophet.forecaster:Disabling weekly seasonality. Run prophet with weekly_seasonality=True to override this.\n",
      "/opt/app-root/lib/python3.6/site-packages/pystan/misc.py:399: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  elif np.issubdtype(np.asarray(v).dtype, float):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<fbprophet.forecaster.Prophet at 0x7f38f4eeaef0>"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m.fit(train_p)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Make future frame wholding the timestamps of future for forecasting. \n",
    "For example, if we run the model every 12 hours, we can make the future frame for next 11 hours 59 minutes, taking frequency as 1 minute. As the time capture is being done for every one minute"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "future = m.make_future_dataframe(periods= int(len(test_p) * 1.1),freq= '1MIN')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "forecast = m.predict(future)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ds</th>\n",
       "      <th>trend</th>\n",
       "      <th>yhat_lower</th>\n",
       "      <th>yhat_upper</th>\n",
       "      <th>trend_lower</th>\n",
       "      <th>trend_upper</th>\n",
       "      <th>additive_terms</th>\n",
       "      <th>additive_terms_lower</th>\n",
       "      <th>additive_terms_upper</th>\n",
       "      <th>daily</th>\n",
       "      <th>daily_lower</th>\n",
       "      <th>daily_upper</th>\n",
       "      <th>multiplicative_terms</th>\n",
       "      <th>multiplicative_terms_lower</th>\n",
       "      <th>multiplicative_terms_upper</th>\n",
       "      <th>yhat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018-06-17 23:59:59</td>\n",
       "      <td>23541.156486</td>\n",
       "      <td>-12123.271112</td>\n",
       "      <td>55904.790931</td>\n",
       "      <td>23541.156486</td>\n",
       "      <td>23541.156486</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>22433.652491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018-06-17 23:59:59</td>\n",
       "      <td>23541.156486</td>\n",
       "      <td>-13011.160637</td>\n",
       "      <td>59179.004816</td>\n",
       "      <td>23541.156486</td>\n",
       "      <td>23541.156486</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>22433.652491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018-06-17 23:59:59</td>\n",
       "      <td>23541.156486</td>\n",
       "      <td>-13642.472200</td>\n",
       "      <td>60539.777282</td>\n",
       "      <td>23541.156486</td>\n",
       "      <td>23541.156486</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>-1107.503995</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>22433.652491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018-06-18 00:00:59</td>\n",
       "      <td>23542.708790</td>\n",
       "      <td>-14833.892816</td>\n",
       "      <td>57834.224556</td>\n",
       "      <td>23542.708790</td>\n",
       "      <td>23542.708790</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>22479.700509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018-06-18 00:00:59</td>\n",
       "      <td>23542.708790</td>\n",
       "      <td>-10373.433107</td>\n",
       "      <td>57276.211722</td>\n",
       "      <td>23542.708790</td>\n",
       "      <td>23542.708790</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>-1063.008281</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>22479.700509</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   ds         trend    yhat_lower    yhat_upper   trend_lower  \\\n",
       "0 2018-06-17 23:59:59  23541.156486 -12123.271112  55904.790931  23541.156486   \n",
       "1 2018-06-17 23:59:59  23541.156486 -13011.160637  59179.004816  23541.156486   \n",
       "2 2018-06-17 23:59:59  23541.156486 -13642.472200  60539.777282  23541.156486   \n",
       "3 2018-06-18 00:00:59  23542.708790 -14833.892816  57834.224556  23542.708790   \n",
       "4 2018-06-18 00:00:59  23542.708790 -10373.433107  57276.211722  23542.708790   \n",
       "\n",
       "    trend_upper  additive_terms  additive_terms_lower  additive_terms_upper  \\\n",
       "0  23541.156486    -1107.503995          -1107.503995          -1107.503995   \n",
       "1  23541.156486    -1107.503995          -1107.503995          -1107.503995   \n",
       "2  23541.156486    -1107.503995          -1107.503995          -1107.503995   \n",
       "3  23542.708790    -1063.008281          -1063.008281          -1063.008281   \n",
       "4  23542.708790    -1063.008281          -1063.008281          -1063.008281   \n",
       "\n",
       "         daily  daily_lower  daily_upper  multiplicative_terms  \\\n",
       "0 -1107.503995 -1107.503995 -1107.503995                   0.0   \n",
       "1 -1107.503995 -1107.503995 -1107.503995                   0.0   \n",
       "2 -1107.503995 -1107.503995 -1107.503995                   0.0   \n",
       "3 -1063.008281 -1063.008281 -1063.008281                   0.0   \n",
       "4 -1063.008281 -1063.008281 -1063.008281                   0.0   \n",
       "\n",
       "   multiplicative_terms_lower  multiplicative_terms_upper          yhat  \n",
       "0                         0.0                         0.0  22433.652491  \n",
       "1                         0.0                         0.0  22433.652491  \n",
       "2                         0.0                         0.0  22433.652491  \n",
       "3                         0.0                         0.0  22479.700509  \n",
       "4                         0.0                         0.0  22479.700509  "
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forecast.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Useful features in forecasted frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "forecasted_features = ['ds','yhat','yhat_lower','yhat_upper']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Plotting the future with history\n",
    "Confidence interval will be also printed for expectation. After completion of historical plot, future plot will be in right part of graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsgAAAGoCAYAAABbtxOxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3XuQlNWd//H38/RlrjDcHK4GYwYRDIkxKGTzC8FQhJDKYqIWmqQWXEyRVas07m7E2lrduH9EYDeW2Y3WOinKgi0jmzKupHYVSeFijMUl4mVVvKABF2aG+9x6+vY855zfH09P0wM9gArTM/J5VaVCP9Pd58yZQT5z5vt8j+ecc4iIiIiICAB+pScgIiIiIjKYKCCLiIiIiJRQQBYRERERKaGALCIiIiJSQgFZRERERKSEArKIiIiISAkFZBERERGREgrIIiIiIiIlFJBFRERERErEKz2BoWLMmDFcdNFFAzpmEAQkEokBHXMo0LqUp3UpT+tSntalPK1LeVqX/mltyhus67J3716OHDly2ucpIJ+hiy66iJdeemlAx2xtbWXChAkDOuZQoHUpT+tSntalPK1LeVqX8rQu/dPalDdY12XmzJln9DyVWIiIiIiIlFBAFhEREREpoYAsIiIiIlJCAVlEREREpIQCsoiIiIhICQVkEREREZESCsgiIiIiIiUUkEVERERESiggi4iIiIiUUEAWERERESmhgCwiIiIiUkIBWURERESkhAKyiIiIiEgJBWSRCps1axa+71NdXc2KFSsqPR0REZHzXrzSExA5n82aNYsdO3YAkMvlWL16NQCrVq2q5LRERETOa9pBFqmgl19++aRrTz75ZAVmIiIiIr0UkEUq6Iorrjjp2rXXXluBmYiIiEgvBWSRCtq+fTtXXXUVnudRVVXFXXfdpfIKERGRClMNskiFbd++vdJTEBERkRLaQRYRERERKaGALCIiIiJSQgFZRERERKSEArKIiIiISAkFZBERERGREgrIIiIiIiIlFJBFREREREooIIuIiIiIlFBAFhEREREpoYAsIiIiIlJCAVlEREREpIQCsoiIiIhICQVkEREREZESCsgiIiIiIiUUkEVERERESiggi4iIiIiUUEAWERERESmhgCwiIiIiUkIBWURERESkhAKyiIiIiEiJigbkiy66iBkzZnD55Zczc+ZMAI4dO8b8+fOZMmUK8+fPp729HQDnHLfffjtNTU187nOf4+WXXy6+z9q1a5kyZQpTpkxh7dq1xes7d+5kxowZNDU1cfvtt+OcO+UYIiIiIiIV30H+n//5H1599VVeeuklAFauXMm8efPYvXs38+bNY+XKlQA888wz7N69m927d9Pc3Mwtt9wCRGH3vvvuY/v27ezYsYP77ruvGHhvueUWfvnLXxZft3HjxlOOISIiIiJS8YB8og0bNrB06VIAli5dylNPPVW8vmTJEjzPY/bs2XR0dNDW1sazzz7L/PnzGTVqFCNHjmT+/Pls3LiRtrY2urq6mD17Np7nsWTJkj7vVW4MEREREZF4JQf3PI+vf/3reJ7HD3/4Q5YvX87BgwcZP348AOPGjePgwYMAtLS0cOGFFxZfO2nSJFpaWk55fdKkSSddB/od40TNzc00NzcDcODAAVpbW8/iZ396hw8fHtDxhgqtS3lal/K0LuVpXcrTupSndemf1qa8ob4uFQ3If/jDH5g4cSKHDh1i/vz5XHrppX0+7nkenued0zmcaozly5ezfPlyAGbOnMmECRPO6VzKqcSYQ4HWpTytS3lal/K0LuVpXcrTuvRPa1PeUF6XipZYTJw4EYDGxka+853vsGPHDsaOHUtbWxsAbW1tNDY2Fp+7b9++4mv379/PxIkTT3l9//79J10H+h1DRERERKRiAbmnp4fu7u7inzdt2sRnP/tZFi1aVOxEsXbtWq655hoAFi1axLp163DOsW3bNhoaGhg/fjwLFixg06ZNtLe3097ezqZNm1iwYAHjx49n+PDhbNu2Decc69at6/Ne5cYQEREREalYicXBgwf5zne+A0AYhnzve9/jG9/4BldeeSWLFy9mzZo1TJ48mV//+tcAfPOb3+Tpp5+mqamJ2tpaHn30UQBGjRrFPffcw5VXXgnAvffey6hRowB4+OGHuemmm8hkMixcuJCFCxcCcPfdd5cdQ0RERESkYgH54osv5rXXXjvp+ujRo9m8efNJ1z3P46GHHir7XsuWLWPZsmUnXZ85cyZvvPHGGY8hIiIiIjLo2ryJiIiIiFSSArKIiIiISAkFZBERERGREgrIIiIiIiIlFJBFREREREooIIuIiIiIlFBAFhEREREpoYAsIiIiIlJCAVlEREREpIQCsoiIiIhICQVkEREREZESCsgiIiIiIiUUkEVERERESiggi4iIiIiUUEAWERERESmhgCwiIiIiUkIBWURERESkhAKyiIiIiEgJBWQRERERkRIKyCIiIiIiJRSQRURERERKKCCLiIiIiJRQQBYRERERKaGALCIiIiJSQgFZRERERKSEArKIiIiISAkFZBERERGREgrIIiIiIiIlFJBFREREREooIIuIiIiIlFBAFhEREREpoYAsIiIiIlJCAVlEREREpIQCsoiIiIhICQVkEREREZESCsgiIiIiIiUUkEVERERESiggi4iIiIiUUEAWERERESmhgCwiIiIiUkIBWURERESkhAKyiIiIiEiJigdkYwxf+MIX+Na3vgXAnj17mDVrFk1NTdxwww3k83kAcrkcN9xwA01NTcyaNYu9e/cW3+P++++nqamJqVOn8uyzzxavb9y4kalTp9LU1MTKlSuL1/sbQ0RERESk4gH55z//OdOmTSs+XrFiBXfeeSfvvfceI0eOZM2aNQCsWbOGkSNH8t5773HnnXeyYsUKAHbt2sX69et588032bhxI7feeivGGIwx3HbbbTzzzDPs2rWLxx9/nF27dp1yDBERERGRigbk/fv389///d/84Ac/AMA5x3PPPcf1118PwNKlS3nqqacA2LBhA0uXLgXg+uuvZ/PmzTjn2LBhAzfeeCNVVVV8+tOfpqmpiR07drBjxw6ampq4+OKLSSaT3HjjjWzYsOGUY4iIiIiIxCs5+I9+9CNWr15Nd3c3AEePHmXEiBHE49G0Jk2aREtLCwAtLS1ceOGFAMTjcRoaGjh69CgtLS3Mnj27+J6lr+l9fu/17du3n3KMEzU3N9Pc3AzAgQMHaG1tPZuf/mkdPnx4QMcbKrQu5WldytO6lKd1KU/rUp7WpX9am/KG+rpULCD/13/9F42NjXzxi19ky5YtlZrGKS1fvpzly5cDMHPmTCZMmDDgc6jEmEOB1qU8rUt5WpfytC7laV3K07r0T2tT3lBel4oF5BdffJHf/va3PP3002SzWbq6urjjjjvo6OggDEPi8Tj79+9n4sSJAEycOJF9+/YxadIkwjCks7OT0aNHF6/3Kn1NueujR4/udwwRERERkYrVIN9///3s37+fvXv3sn79er72ta/x2GOPcfXVV/PEE08AsHbtWq655hoAFi1axNq1awF44okn+NrXvobneSxatIj169eTy+XYs2cPu3fv5qqrruLKK69k9+7d7Nmzh3w+z/r161m0aBGe5/U7hoiIiIhIxbtYnGjVqlU88MADNDU1cfToUW6++WYAbr75Zo4ePUpTUxMPPPBAsW3bZZddxuLFi5k+fTrf+MY3eOihh4jFYsTjcX7xi1+wYMECpk2bxuLFi7nssstOOYaIiIiIiOecc5WexFAwc+ZMXnrppQEds7W1dUjX75wrWpfytC7laV3K07qUp3UpT+vSP61NeYN1Xc40zw26HWQRERERkUpSQBYRERERKaGALCIiIiJSQgFZRERERKSEArKIiIiISAkFZBERERGREgrIIiIiIiIlFJBFREREREooIIuIiIiIlFBAFhEREREpoYAsIiIiIlJCAVlEREREpIQCsoiIiIhICQVkEREREZESCsgiIiIiIiUUkEVERERESiggi4iIiIiUUEAWERERESmhgCwiIiIiUkIBWURERESkhAKyiIiIiEgJBWQRERERkRIKyCIiIiIiJRSQRURERERKKCCLiIiIiJRQQBYRERERKaGALCIiIiJSQgFZRERERKSEArKIiIiISAkFZBERERGREgrIIiIiIiIlFJBFREREREooIIuIiIiIlFBAFhEREREpoYAsIiIiIlJCAVlEREREpIQCsoiIiIhICQVkEREREZESCsgiIiIiIiUUkEVERERESiggi4iIiIiUUEAWERERESlRsYCczWa56qqr+PznP89ll13GP/zDPwCwZ88eZs2aRVNTEzfccAP5fB6AXC7HDTfcQFNTE7NmzWLv3r3F97r//vtpampi6tSpPPvss8XrGzduZOrUqTQ1NbFy5cri9f7GEBERERGpWECuqqriueee47XXXuPVV19l48aNbNu2jRUrVnDnnXfy3nvvMXLkSNasWQPAmjVrGDlyJO+99x533nknK1asAGDXrl2sX7+eN998k40bN3LrrbdijMEYw2233cYzzzzDrl27ePzxx9m1axdAv2OIiIiIiFQsIHueR319PQBBEBAEAZ7n8dxzz3H99dcDsHTpUp566ikANmzYwNKlSwG4/vrr2bx5M845NmzYwI033khVVRWf/vSnaWpqYseOHezYsYOmpiYuvvhikskkN954Ixs2bMA51+8YIiIiIiLxSg5ujOGLX/wi7733Hrfddhuf+cxnGDFiBPF4NK1JkybR0tICQEtLCxdeeCEA8XichoYGjh49SktLC7Nnzy6+Z+lrep/fe3379u0cPXq03zFO1NzcTHNzMwAHDhygtbX1LK/AqR0+fHhAxxsqtC7laV3K07qUp3UpT+tSntalf1qb8ob6ulQ0IMdiMV599VU6Ojr4zne+w9tvv13J6Zxk+fLlLF++HICZM2cyYcKEAZ9DJcYcCrQu5WldytO6lKd1KU/rUp7WpX9am/KG8roMii4WI0aM4Oqrr2br1q10dHQQhiEA+/fvZ+LEiQBMnDiRffv2ARCGIZ2dnYwePbrP9dLX9Hd99OjR/Y4hIiIiIlKxgHz48GE6OjoAyGQy/O53v2PatGlcffXVPPHEEwCsXbuWa665BoBFixaxdu1aAJ544gm+9rWv4XkeixYtYv369eRyOfbs2cPu3bu56qqruPLKK9m9ezd79uwhn8+zfv16Fi1ahOd5/Y4hIiIiIlKxEou2tjaWLl2KMQZrLYsXL+Zb3/oW06dP58Ybb+Tv//7v+cIXvsDNN98MwM0338xf/MVf0NTUxKhRo1i/fj0Al112GYsXL2b69OnE43EeeughYrEYAL/4xS9YsGABxhiWLVvGZZddBsCqVavKjiEiIiIi4jnnXKUnMRTMnDmTl156aUDHbG1tHdL1O+eK1qU8rUt5WpfytC7laV3K07r0T2tT3mBdlzPNc4OiBllEREREZLBQQBYRERERKaGALCIiIiJSQgFZRERERKSEArKIiIiISAkFZBERERGREgrIIiIiIiIlFJBFREREREooIIuIiIiIlFBAFhEREREpoYAsIiIiIlJCAVlEREREpIQCsoiIiIhICQVkEREREZESCsgiIiIiIiUUkEVERERESiggi4iIiIiUUEAWERERESmhgCwiIiIiUkIBWURERESkhAKyiIiIiEgJBWQRERERkRIKyCIiIiIiJRSQRURERERKKCCLiIiIiJQ4bUA+ePAgN998MwsXLgRg165drFmz5pxPTERERESkEk4bkG+66SYWLFhAa2srAJdccgkPPvjgOZ+YiIiIiHwy7WvPsHNfBz25sNJTKeu0AfnIkSMsXrwY34+eGo/HicVi53xiIiIiIjK0WetwzgGwvyPD0Z48e4+lOdCdJbSOtw+lKjzD8uKne0JdXR1Hjx7F8zwAtm3bRkNDwzmfmIiIiIgMXcY6Xm3pLD52ODw8rHPkQovnQdwfnJuupw3IDzzwAIsWLeL999/ny1/+MocPH+aJJ54YiLmJiIiIyBD1aksn1jkKe6xYB845QuvoyRuSMY+q+ODsF3HagHzFFVfw/PPP88477+CcY+rUqSQSiYGYm4iIiIgMMYGxHEsHuCoXhWILngfOQTY05I0jExjyxqOu6rRRtCJOO6t169b1efzyyy8DsGTJknMzIxEREREZso725DnYnWV4cjihtYQm2jWO+R49eUM6bwidxcer9FT7ddqA/Mc//rH452w2y+bNm7niiisUkEVERETkJC2dWZyLSipSOQNAJjB4HmQDiwOMcVjPVXaip3DagPyv//qvfR53dHRw4403nrMJiYiIiMjQ4pyjKxuy51gaR7RjHFpHT2CIeZALLcb1drTwmDZ2GAe6czTWV1V66mV96MKPuro69uzZcy7mIiIiIiIVZqyjrSvL8Oo4w6rixU5m5fTedLf3WJqOTEBoo7rjnrzBZgNyoSHm+VgctlBmYQtt3xL+EC6x+PM///Piwlhr2bVrF4sXLz7nExMRERGRgZfKhRzoznKwO8p/V0xqwPM8MoEhVvj/4dVRcD7QnaO1MwuAsZANDKZwM57NGax14FusdXxmdB2BtQRm8JZW9DptQP7bv/3b40+Ox5k8eTKTJk06p5MSERERkYEVGMsbbd1MbKjGOvCIguzL+zv59Oha/nS0Bw8P4xwxz2NiQzUtnZnCax350JIOLNkwCsGetVEtso06WCTjPsnTn1E3KJw2IH/1q18diHmIiIiISAUExtKRCfi/9gzWOf6vI01gHPFCCYTD8f6RHgLjSMSisGtwtHRmCaN78EgHhlxoyRT+3xRO0KtJxKhOxBhbn6zgZ/jh9RuQhw0bVrbmxDmH53l0dXWd04mJiIiIyLmVCQy7DnQzdlhV4SY6CI0jMBbnot3iuO8R933SgaHOixVqiD2cD7nQEBZu0MuHltDawjs7LhpVS3V8cJ6Udzr9BuTu7u6BnIeIiIiIDDBjHdY5DnbnomBsLaF1BGF0Q52x0Ul4nmfJhQZjewOzRzZw5IwlMJZcaAmNZfzwatq6oprkoRqO4UN0sTh06BDZbLb4+FOf+tQ5mZCIiIiInDuhiULwWwdTWOfIhpZkzCcXGjqyIR5gHMQ8yIaWuO/hAYF15HxLwveIxzyMjdq3OcAUdo5H1CToyobkKvkJngWnDci//e1v+Zu/+RtaW1tpbGzkgw8+YNq0abz55psDMT8REREROQuMjUon3jzQjaNQTmEd2SDaAe7Jh2QLRcU+HrlCO7a8iXaRrQXrHNb5BNZhrCuUVHg4B1MuqAdgQkM11fnqSn2aZ8VpbyW855572LZtG5dccgl79uxh8+bNzJ49+2MPvG/fPq6++mqmT5/OZZddxs9//nMAjh07xvz585kyZQrz58+nvb0diGqfb7/9dpqamvjc5z5XPPIaYO3atUyZMoUpU6awdu3a4vWdO3cyY8YMmpqauP322wvNqfsfQ0REROST4P/a0wDsa89wOJXDWserLZ28eaAb6xyBif7XlQ3JBIbObEg2iNqxWQuhjf7sCnXJ0QZx9DgwhnxhF3rcsGoguj+t96a+uO+RjA2NbhX9Oe3sE4kEo0ePxlqLtZarr76al1566WMPHI/H+dnPfsauXbvYtm0bDz30ELt27WLlypXMmzeP3bt3M2/ePFauXAnAM888w+7du9m9ezfNzc3ccsstQBR277vvPrZv386OHTu47777ioH3lltu4Ze//GXxdRs3bgTodwwRERGRocpaR3c2xNioprgnF3IwleWD9jSvtHRinMMUwnHeWLqyIalcSN5E9cWhjW6+84CLRtZyaWM9nxldV9hgjIJybSLOiOoEuGjzckRNgsb6KsYNG5wn4n1Upy2xGDFiBKlUiq985St8//vfp7Gxkbq6uo898Pjx4xk/fjwQdcyYNm0aLS0tbNiwgS1btgCwdOlS5s6dy6pVq9iwYQNLlizB8zxmz55NR0cHbW1tbNmyhfnz5zNq1CgA5s+fz8aNG5k7dy5dXV3F3e4lS5bw1FNPsXDhwn7HEBERERkqQmOJx3za03n+dDTaMbbO4XteVGN8qBvT21SCKBRDVDecCw09eVsomTi+U3zJBfV4HviFTmaJmMclF9TTlQ1IB5YJw6vwPI9xw6OSDYBRtUOrhduZ6Dcg33bbbXz3u99lw4YN1NTU8OCDD/LYY4/R2dnJvffee1YnsXfvXl555RVmzZrFwYMHi8F53LhxHDx4EICWlhYuvPDC4msmTZpES0vLKa+XHmjSex3od4wTNTc309zcDMCBAwdobW09i5/16R0+fHhAxxsqtC7laV3K07qUp3UpT+tSntalf5Vcm7cOdlMVj5ENorph46Kb5eK+T2As8ZhHPrRYHB5gXdTWDSAfumJLtrjvU5PwGV6doPtYuuxYHlAHdB49s7mlOo6d/jmZkKOkSGYHX71yvwH5kksu4cc//jFtbW0sXryY7373uyxduvSsTyCVSnHdddfx4IMPMnz48D4f8zzvlOd/nw2nGmP58uUsX74cgJkzZzJhwoRzOpdyKjHmUKB1KU/rUp7WpTytS3lal/K0Lv0bqLXpygbsPtzD5yYM539bu2gYU4NzELMOD49caKKOFHE/SsuAbyz5sBCEPfBCi+95JKwlZh2fGllDXfKMm5p9KCPGjD3lx01PntGN9UwYVXtOxv84+q1BvuOOO9i6dSvPP/88o0ePZtmyZVx66aX84z/+I+++++5ZGTwIAq677jq+//3vc+211wIwduxY2traAGhra6OxsRGAiRMnsm/fvuJr9+/fz8SJE095ff/+/SddP9UYIiIiIoNFdEOcpSsbsHNfB+8d6cE6x/+2dhFah7FRh4l8aMkbSyofksoburIh3bmQrmxQrC/OhYaewBAai3EW5xxNY+rOWTge6k57k97kyZNZsWIFr7zyCo8//jj/+Z//ybRp0z72wM45br75ZqZNm8Zf//VfF68vWrSo2Ili7dq1XHPNNcXr69atwznHtm3baGhoYPz48SxYsIBNmzbR3t5Oe3s7mzZtYsGCBYwfP57hw4ezbds2nHOsW7euz3uVG0NERESkkpxzHEnleOtgN+8eTvFaaye7D/cUa4WNjVqzpXJR94l03pAODD35kHTekjeGTGDIBqYQjI93o4gVfmNubXQTXmKId5o4l077Y0MYhjzzzDOsX7+ezZs3M3fuXH7yk5987IFffPFF/v3f/50ZM2Zw+eWXA/DTn/6Uu+++m8WLF7NmzRomT57Mr3/9awC++c1v8vTTT9PU1ERtbS2PPvooAKNGjeKee+7hyiuvBODee+8t3rD38MMPc9NNN5HJZFi4cCELFy4E6HcMERERkYEUGstrrV1cMamBl/d3AuCIyiOcA0cUmkPr8D2immLnyASGnHHEvegADw/IGxO1ZPOid6lNxOnJh8R9n9BamsZEHSnePpSiOjF0T7kbCP0G5N/97nc8/vjjPP3001x11VXceOONNDc3n5UOFgD/7//9v2Jf4hNt3rz5pGue5/HQQw+Vff6yZctYtmzZSddnzpzJG2+8cdL10aNHlx1DREREZKBY63ittQuH4+X9nVE3CaJg7HkQGFfcOc4GlmTMoysb4nkeuUJZRdyPOlZ4XvS6T4+uZU+ho8WnRtYAEBiLKWQuz/OY2lhfqU95yOg3IN9///1873vf42c/+xkjR44cyDmJiIiIfOL05ELeO9rD58YPx/M89nVkiifamUIQBkgHhlhJCPY9j2xoiXnHPxZaWzwJrzoeIxsafM+jOh5jamN9sQUbQCLmkyiZh3+OGyB8EvQbkJ977rmBnIeIiIjIJ87uwymSMZ8xdUm6ciGBsX1KKUzh+OZ8aAkLdcKpbFj4eBSArYt2gaPjnh0WxwX1VdQlY/TkDWPqkrR1ZYkVTrLzPQ8/phD8cejWRREREZGzLDCW/y2UTzgHR3ryAIWDO6KSCIBUPiTmeXTnQnojbTo0+HiYwuEdkegI6Mb6KkbXHT+Yo6ZQSzx++ODrJTyUKSCLiIiInCXWOrKhYXehJZsj6hqBHx3xHBQO8siFhmTMpysbkvA9evJRiYRxFmMc1qN4r1ZvffFFo2qLgVjOLQVkERERkY/haE+evcfSjK5LcrQnj6P34A5L7wZwJrDkTdSNIuZZegLD8Kp4VGdM1M3C86A+GWfcyCreO9pDIuYzfngVh1N5Jo2oIe6rbGKgKCCLiIiIfESdmQDjZXE4jvTkijXF4MgEltBaYr5PJh9iiXaYPd8jG4QY6zDWAh6XNtaTDW1xh/iiUbX4eCTjPnWjFNcGmlZcRERE5CMIjKWlM8uIpCnUFkN3LsT3olrjVD4kNI5kzJIODR4e1jk8G5VMBMZyQV0Vh1I5PM/rUz5RHVcpRSUpIIuIiIh8BLnQYhyEBoyzWAdduZAYUfeJbBgd6WydjzEOz3PUV8UJjOPiUcdLJkpvupPBQQFZRERE5CN451CKfGhIByHWQWiidm0eUUCuivtUxXy6cgFVcZ9PjaxVHfEQoYAsIiIi8iHs3NcBRId75EKLHxgC4/rUFDvnGDesiuq4zxiTJOZ7xT7FMvgpIIuIiIgU5ENLMu73uWatoysX8v6RHiaPrCn2J86Flmxo8AJLzkQn29VXxenOhkxoqC7WFCfjCsZDjQKyiIiInLcygaErG+KcY39nBoBLG4dxOJVnTF2SXGj4oD06Eto62NueJpULifkeqZwhsA7fWKx1XNpYj+d50FDhT0o+NgVkEREROS+k8yEdmZC477GvI0NDTYKOTB7rIOZ5BMbhex5vH+rGOTiazuHhEVqH70VdJ0Lj6M6FxR1kax0J32PSyNooHMsnggKyiIiIfGI55+jKhjTUJNh1sLtPGD7ak4+CrjHEfY9saPGBRMzHOqiK+1hnyQTRqXeZICqpyIUu6lpho2NALh5dV9lPUs46BWQRERH5xOrOhew+kmJiQw3GRrvAcd+nMxt1lgiMI50PSRb+nDeWmkQUkOuSMRyQyoYk4j750Eat23AYE92E1xOoRdsnkQKyiIiIfKJY6zDO8b+tXUB0aMf/tWfIBCY6/tkZUvmQ7lx0YEfeWKo2uJt5AAAgAElEQVRs1IUiMJbQumKrNojqlL3QRO9lHJ4HExqqaahO4KUVpT6J9FUVERGRT4T2dJ6RtUleaekEiHZ6bRRwjXOk84ZsYLEuCtDWOjwvasmWCy3gGD+8mrauLL7vkQ0N1kZlGhCVXkwaWUMy5hGP+aeYiQx1CsgiIiIyZL11sBtjo4DrcIzKBNhCoM2baEc4G1h68iFBYZcYojZtkxpqqE1Grdisc6QDw7CqOCNqhtGdDTmWyZM2hs+Mrjup9Zt8sikgi4iIyKAVGAtEu7e98qElbyxvH+ouXgsN+D4cTuUJrSO0UTDOhZbAuqhMwvPwPbjkgmEnjRPDo6FkjGHVcYZVKyadr/SVFxERkUHrrYMprHPUJmN0Z0MgKp0AyIWOuO8R2igIx3yPuO/Rk7fkw6jeODrAwzGsKk7eWHWckDOigCwiIiKDgrGO9470UF8Voz0dUF8VJzBRzXBHxhafY2x0o1xP3pLwIZU3xWvJmB/tGhc6UsR9j0kja6iK+epTLGdMAVlEREQqKjCWzkzA3vY0zkEqFxLaqCYYICzUEjsgG1iMiw7uSOUMsZhHEEYfi0Jy9BrnopvrGuurqI7HKvfJyZCkgCwiIiIDbue+DmqTMariPsfSeQACEx3n7OHIhQbfj3Z8e0NxLrTkw2gn2fM8AmPAeFAoufDwmNRQTXfOUJ+MUV+lmCMfjb5zRERE5KzrbY1WWtbgnKOtK0dbVxbrHKlcSDrvkQstvueRDy0540jGPNKBIbSueKRzMu6TzhvAkYz55I3FOZg8sqa40zymLjq0oy6peCMfj76DRERE5Kz7oD1DVzYkMLbYWxjAuOgQjryJArRzlp68IR7zyAaWVC6kOuGXHOcM4DBBdEPehOFRa7b3j/YwqaGWqrhfbNUmcrYoIIuIiMhZl8qF5EKDA9q6ssVT6axzBIWaYls4xS5bKJvIG4O1jnQQHeLR66JRtew5mmZCQ3UxDH9G3SjkHFJAFhERkQ/NOUdoHb7n0ZkNqE3EeONAFxfUVRGPeWSCKByHxpG1lsBYPI/CjXZR+URYqDnOm6je2BR2la11jKhO0JENcc5RHY8xeWQtNQkd1iEDQwFZREREPhTnHK+1dhFai4eHdVGLtXzoOJjKAdFhHrkw2iXOhaYYpqNrURAOrcX3PZyD+kSc6lqfmkSMdN4wui7JuOEUTr5DZRQyoBSQRURE5IxlAsM7h1IExpLOG6riMYyLgnI6MMRCyBZOuitUThAWWrR5HD8Zz/O8YunEyNok44ZVFceoSRwPwzFfvYtl4Ckgi4iIyBmx1vFGWxfGRjfbdWQCqhNR4M0GBgPEvai8whR2la2NdpxrEjGyYXTc86WN9cX3nNpYj68DPGSQUUAWERGRfhnreOtgN22mAyjcVBdEO8SZ0BIUWrH11hEHeBgX3WTnedGBHRDdaBfaqINFKYVjGYwUkEVERKRfmcDgHFG49eBoT0C+0IHCWIt1HtY6hlcn6M4FWBzDqhIMr4pTk4xFN+cV3iuucgkZIhSQRUREzgNd2YC6ZPyManp7ciHpwHCwO0c2NORCC7kQ6yAdGEzUnJjaRJyefMiljfU4YIxJ0pMPGVWbLL5X3NfNdTL0qF+KSIXNmjUL3/eprq5mxYoVlZ6OiHyCdGQC0vmQTGB493CKV1s6aenMlH1u78l3AG8fSvFBe5pMYMiFjlQ+JJULOZbOExrLhOHVTGqoYWJDNU1j6vA8D9/zqIr7fcKxyFClHWSRCpo1axY7duwAIJfLsXr1agBWrVpVyWmJyBDWkQkAGFGT4N3DKWKeVzycIxGLDu040JVj+rhhJHyPtw+laBpTxxsHuvAKxRAOR2iiNmztmaikgtASGsu0scP6jBc7qapYZOjTDrJIBb388ssnXXvyyScrMBMR+SSw1vHekRTvHUnx8v4OAuMwzpENLd25kHTeYCwExvF6WxevtnaSyhneONBFYKKDP/LGks5bOrMBndmArmxIPow+NqGhutKfosiA0A6ySAVdccUVxR3kXtdee22FZiMiQ90rLZ1YF3WOcJ4jGxjSeQqlEpaemMfIQmCuTcaIeR49eUNDdYJcaDE22mlOh4ag0MS4PhmjoT7JsIZq6qsUG+T8oO90kQravn07s2bN4o9//CPJZJI77rhD5RUi8qG8cyhFKhcybWw91jnyoSW0Ducc3bkwOrkOhzGOmPM4lonqjY93ooh2h0MbHewRmuOvB2gaU0dH2KVwLOcVfbeLVNj27dsrPQURGSSMdVjnSMSOV0D25EK6ciHjh0flDaGxeJ5HYCyBcXRlAxyw62A3oY1CcU++0JrN2eJBHSNrk7Sn82RtiO975EIAh+d50bhEvYudi+qXL6hLEo+pElPOTwrIIiIig8Tuwyl68oYvXjgC5xythRvqHFFwbuvKFm+kg+hmumxosdaBB6mcIZULCa0DHMmYz8WNdcXnjxtWhXWOdw+n+NSIWvLG0taVxXnRbvGFI2qoSyoaiOhvgYiIyCBwJJWjJ2+wztHSmeFAVw7rHMaC78P+jiyBsSTjfvF0utBaenKGdGiw1hVqiC0XjaolEfPLHszhex5x3ycZ86hNJhhWFSedNwTWKhyLFOhvgoiISAV1Z0PeOdyNc2ALwXd/RzaqJy7UA8d9j2xgifkeYaEMw1gIrCWVMwTGAlF5xIlt2MppGnN8VznmewyrVhwQKVXR4qJly5bR2NjIZz/72eK1Y8eOMX/+fKZMmcL8+fNpb28Hovqp22+/naamJj73uc/1aY+1du1apkyZwpQpU1i7dm3x+s6dO5kxYwZNTU3cfvvtxRsO+htDRETk48oXuj8AvNbaSWiix0GhnKH3wI1j6Tw9uSgc50JHLrSk8yHt6ai1Wnsm4EhPQHs64EBXjo5MQHcupCMT0FH4eBSODeOHVzFlTP0ZhWMROb2KBuSbbrqJjRs39rm2cuVK5s2bx+7du5k3bx4rV64E4JlnnmH37t3s3r2b5uZmbrnlFiAKu/fddx/bt29nx44d3HfffcXAe8stt/DLX/6y+LresfobQ0RE5ONI50Neb+vizQPd9ORCAmN5tbWTnfs6eK21k30dGd4+1M17R3rYfbiHNw5005MzdGQCjqUD2rMhHdmAI+kc3dmQbGDIhaZwQ54lExjSgSGTD8mHhnxoAI+G6sQZHSEtImemogF5zpw5jBo1qs+1DRs2sHTpUgCWLl3KU089Vby+ZMkSPM9j9uzZdHR00NbWxrPPPsv8+fMZNWoUI0eOZP78+WzcuJG2tja6urqYPXs2nuexZMmSPu9VbgwZ2rZu3cr999/P1q1bKz0VEfkEsiWtz060c18H6XwY3TDnHKlcyK6D3VgHudAWdoct+dDSmQmLO8EHu3McTOWjQzwCQyZvCI2lsa6KumQM5xyXNg5j2thhXNpYj7WOqpjPxaPqmDb2+HURObsGXdHRwYMHGT9+PADjxo3j4MGDALS0tHDhhRcWnzdp0iRaWlpOeX3SpEknXT/VGCdqbm6mubkZgAMHDtDa2noWP9PTO3z48ICON1SUW5eXXnqJG264gSAISCQS/Md//AczZ86swOwq53z5fvne975X7B/9q1/96rTPP1/W5cPSupR3qnV5+1BUJ+x7HhePriXuexxLBxxK5TAWth6J/i0JLVgX7fg6Fz3GQU8Qkoz5ZEJD3Pei9muAsb0lGR4jahKMqI5DT5oaoCYGHUfSxTmMjwEW0h3dpBk4qY5jAzja0KK1Ke9M1iWVCTlKimR28J3QOOgCcinP8/C8c/sro1ONsXz5cpYvXw7AzJkzmTBhwjmdSzmVGHMoOHFd3nzzTYIgwBhTfLxo0aJKTK2iPunfLwsWLOD5558H4Pnnn+cv//IvefbZZ0/7uk/6unxUWpfyTlyX/R0ZhlXFGRbWFG+iO2AhBpgkVI+IDtmojsfIBoZ8EP13KAii9mueF/U39o3FxTwSJnqTEVVx8Ch2kejIBFw0iGuIR4wZW+kpDFpam/JOty6mJ8/oxnomjKodoBmduUEXkMeOHUtbWxvjx4+nra2NxsZGACZOnMi+ffuKz9u/fz8TJ05k4sSJbNmypc/1uXPnMnHiRPbv33/S8081hgxdc+fOxfd9jDH4vs/cuXMrPSU5B1544YVTPhY5m0Jj2X2kh1Qu5IAHxkJHJqAqHh3PnIzHiHmQM5bQONpdQN5EZRgx3yNXqA+OOD5dCAHVidhJYzVUJxhZmxi4T05ETmnQHZGzaNGiYieKtWvXcs011xSvr1u3Ducc27Zto6GhgfHjx7NgwQI2bdpEe3s77e3tbNq0iQULFjB+/HiGDx/Otm3bcM6xbt26Pu9VbgwZul5//XWCIAAgCAJef/31Cs9IzoUZM2ac8rHI2eCco7Uzy6utnXRmQvLG0ZM3HOnJ0ZkN6MyGdOdDOgs31vXkDJkg6l8cmOiGulxoGFGT5OJRtYyuTXDRyFqqE7Gy4bhXdbz/j4nIwKroDvJ3v/tdtmzZwpEjR5g0aRL33Xcfd999N4sXL2bNmjVMnjyZX//61wB885vf5Omnn6apqYna2loeffRRAEaNGsU999zDlVdeCcC9995bvPHv4Ycf5qabbiKTybBw4UIWLlwI0O8YMnQ9+OCDfR6vWbOmWB4jnxzf/va32bFjR5/HImfLzn0ddBzpptXUYCz05ENyoSVvomLhVD4EPEwQHcphveO9hz0PmsYMIx9afC+qLe49LvqC+qpKfloi8hFUNCA//vjjZa9v3rz5pGue5/HQQw+Vff6yZctYtmzZSddnzpzJG2+8cdL10aNHlx1DhqatW7fy1ltv9bmWz+crNBs5lzo6Ok75WKRXTy4EoCYRwz+D9mdHUtGpdaF19OQMndmQniDExyvcSOeI+9HJdDWJGHXJGO2ZgAtH1ADHex8n44PuF7Mi8hEMuhpkkQ+rtAa914EDBwZ+IueRrVu3smXLFubOncuXvvSlARv3xK91ua+9nJ/yoeX1ti4AHH1bsU0fO4zakiOUd+7rYPq4YTjn+NPRNCNrExzoypE3jnTe0JXKEzqLMQ7rRTvEo+uSNJ6wE1xfdfw9FYxFPlkUkGXIK3dDXiqVGviJnCe2bt3KvHnzyOfzJJNJNm/ePGAhubq6+pSP5fxirOPVlk4gCsX50JENDTHPIxn3cS4qdXjjQDe1iRi50PKFiQ04HF3ZkH0daYyFdGAIjKMjE51gF09GXSgubawnbxzJ2LnvqCQig4t+5JUh70tf+hL19X0b5adSKeLxOLNmzTrla5ubm1mwYEGx3/WH8XFeO5Rt2bKFbDaLMYZsNjugu7jTp08/5WM5P+w9lmbnvg5e3t9BNrSk8iGdmZDDqRwHUzk6syGHunOkA8ORVI5caAtHMjv+uK8D62BfR5pc6Ejlo0M72tN5unIBMR8uGlXLmLoqPM+jKu4rHIuch7SDLGfFuf6V+/jx4zlw4ADjxo1j586dJ3283I6xMYYdO3Ywa9Ystm/fftLHm5ub+eEPfwjApk2bAE55Y1/p5/j6669/qNeebZUqcYCo7rf3NDHn3DmpA25ubuY3v/kN1113XZ91XbJkCY8++mhx93rJkiVnfWwZeNa66PS5vKEq7lOTiHG0J8/w6jipXMjI2iR/OtrDpIYa/nS0h85sVF/cng5IhwZno9phi8MYR8qFOAeW6BS7jkyA70FVPIZ1EPch5nukciE54whCS2gdo2sTJKimOh5TRwmR85wC8hBRyUB0Ouf6V+694Rii2uLLL7+cQ4cOFT++YMGCU76+XKAG+PnPf97n8W9+8xuWL19eNpyd+DmOHdu3+fmDDz44YAG5ubmZW2+9FWMMsViMhx9++LRjn/j9018APROvvvrqKR+fzvTp03n77bcZM2YMGzZsOOl75VQ/uHzpS1/iX/7lX4pzH2x/F6R/PbmQupKa3dBYjvTk2d+ZwcPrUzfs4WGdw/c8QuuI+2kcjvZ0QGgdndkQ5xxduZDAHO81XJ+MUV8XheruXEg6HwXp7lz03nHfEit0mHAO8sbinKMq7jN5ZA01iRgd2QFbEhEZxBSQh4BK1nyeiS1btpDP5zHGkM/n2bJly1md34k33J14FOzpOpIMG3byyVTNzc3s2rWrz7ULLrjgpHD2/vvvs2rVKrZs2UIul8NaSy6XO2nXNJPJFP98Ln+Y2bp1a3F+EO2S//CHP2TGjBlMnjy57GtWrFjB6tWri4+///3v89hjjwHHA+iMGTPOeM6XX3558XW9j8/U9OnTix1HDh8+zJe//GVefPHFPmP+5je/6fOa3h9cIPr8f/SjH5HP53nhhReYMWNG8bWD+YfI88HOfR1c2liP58FbB1NMbazHOkd1PMbrbV1YF50oB8cDsCPa4QXwvegIZ+scHh6BtdQkYqTzYbFlmrGObGDpyAbR651j4vAasqHp00ptRE0C5xx54/CAD9ozDKuKuk70lks45/j06FrtFItIWQrIQ8CZBNCtW7eybt06IPo19EcNCP3tnp4qeMydO5dYLIa1llgsdtZPsRs3blyfkHzBBRf0+XgsFiseMV1OR0cHzc3NfXZKTwxhEAW2E6//0z/9E11dXQwfPhxro3/IrbXU1dX1CckjRowofg2am5uLa/HCCy+c9muxYsUKnnzySa699lpWrVrV52PTp0/nnXfeYerUqezatYtbb7217HvcfffdZdsmNjc39wnHQDEc91qzZg0vv/wyYRgWx7zjjjv63Vnetm3bKR+fyjvvvNPnsXOuWMPc+z12YgC/7rrrin/u7+/CYP8hcqjqLX3o3dsNrSMZ88mFhppEjJf3d9JQk6AzEwXWtw+lcDiMhbcPdWNs7+ss+dCSjPuEhZDreR7WOnLGEFpI+B4x3yNvLNZFgTkZCwmMIxdaYr6HsQ5TCMbjh1dTm4iRiPkMK/NPWVQ/HIXhKRfUAdA4rIrDqTx1yVifDhQiIifSfyGGgLlz5xKPx7HWEo/HTwqgW7duZe7cucXev2vWrOH555/vExD6C7m9gfjyyy/n3Xff5amnngL67ix+5StfKQbQq666qmw9bxhGv/LsDVkn+jgBvvdY8NIa5NIgP3369NP+mr83+Pa+5rrrrusTwoA+R5n3cs7xb//2byfdpJNOp/s8zuVyzJs3r89OsjGGu+++m+eff77snOrq6vq8T2+QXbVqFStWrOCBBx4orudbb73FRRdd1G/7uhP7QPeuT+lx6/1paWnp83XbtWtXcZe6NCT3vueLL77Y5/V//OMfTztGr6lTp/aZq+d5jB49mq9+9avFkxBLTZo0qc9peXPnziWZTBaDcO/fhdLgnMlkmD9/Pl/+8pcBmDdvHnfdddcZz/F89f6RHpIxnxE10XHH+zoyZILo731vuYPD9Sl/yIaGsCcK0dZR6Bh8vHQhjM7XoDsXYqyjvipGKmcKu8SR0EaBN+uB70ehGaISiEzhsedBYGB4dYJEzGN4VfyUJ9L1x/c8xg7ToR0icnqe673bRk5p5syZvPTSSwM65m9/+1seffRRfv/733Ps2DEgChQ//vGPizuNW7du5Sc/+clJYe/yyy8nlUpx7bXX8u1vf7vs7lppOUE5X//618lms/z+978/6fqzzz5bfPyFL3yhT0CdMmUK7777bvHxiQHe933+8Ic/FHf+PuyvxVevXs2KFSuKj3sD06lMmzatTzB75JFH+Ku/+itKv/09z+Oj/nWYM2cOf/jDH4q7zKU8z+NTn/oUe/fuZevWraxevbr4g8iJkskkvu+TzX64QshkMskPfvADHnroodN+Xc/UuHHjaGtrAzjleyYSCZ5//nlWr17NK6+8Ql1dXdkd6N6v9SOPPMIHH3yA53nMnz+f2traftcDjn+/QBSEOzo62LJlCxMmTOCuu+7iqaee4rHHHqO1tbXfr98jjzzyiTlZMTSWeOzjNyBqbW1lwoQJvLy/A+fAlIRWY6OT4UJrAa8wpofveQTGFcslUrmwEJwhNMfDcSHjFoKzK4beWMzDmOi1vV+qC0fUUF8V51Aqx9GePOOHR637GqrjtHblmNhQzZGePA3V8eLJdOdSx5GDjBgz9vRPPM9oXfqntSnvTNblaE+eSxrruWhU7QDN6szznALyGRrogLx161bmzJnT747sI488wvvvv88///M/lw1lpebMmcMLL7xQDA9z5szhG9/4Bs3Nzezdu7ff133+85/n3Xff7bMr2qumpoavfOUrPPvsszQ0NNDV1XXS/HoDyf3338/f/d3f9fl4PB7nhhtu4Fe/+hXOuTMuRwBobGw8qQ75dKqqqsjlcsXHHycMf1Tjxo3jyJEj/X5Nz4a77rqLLVu29DmO+eOYM2cOf/rTn0in08Uf0sopt57f/va3ueuuu04qgbDWfui1nzNnDlu3bj1pl9n3/dN+/8PJP9QNZTv3RaU9l1xQT3XC/8ihsbW1lYO2jlxosc7Rkzc4otvd0vmQeMwnbyw+FK/HYz45YwmNw/eiPsShtcWT6pyLyjLqq6Ib5RIxn+HVcRrrq3DOFQNvb1eJYVXxQddCTWGnPK1L/7Q25SkgnycGOiCXC5WlJk+ezAcffHBG71UuvJytgDht2jTg5F/xQ1Qb/MUvfpEHH3yQP/uzPzvte82ZM6ffcoReJ95wJn1VV1d/6N3nc6n3Nx4jRozgnnvuOWWt+Lk01HeQWzoztHVlGV4d1fsaC/FYFEhrk3GyhVKIC0fU0HiaEoKeXMj+ziz7W1qoGXEBudCQDqKjlT2P4m5vablD72Pnju/8Uii3+MyYOnYfTlGbjOEcTBpRQ/wMjnYerBR2ytO69E9rU95QD8iqQR6k3nzzzVN+/EzDMVA2CJ+tn4veeustYrHytYC9fYjPJBzDmdWyPvnkkx9qfuebwRSOIfo+W716NZMnTx7wHftSP/3pTz92QO7IBAyvipfsljq6cyHDqxMnPTc00U1lpT+IpnKGfOGutVG1ieKNaKfaQe3JhcUb37KBJTB58qElG1qSMQ+/cFNbzPPxPNjbni6MC93ZkMkl/+gYG+36vnWom0xgyQaWfDYgnTfkjKWhKiphGF2XJB9aPmjPMG5YFcmYz7FMnu5C7+G47zN+eBXxmFfsADFt7MmdYkREhjIF5EFoxYoVJ3UaGMzO1q7gmQSoU/2aXwavD/MD3bkav7eTiXOOl/d3Up2IMX1sPem84e1DKUbUJJjQUE1govZixjrivseB7hxjh1Xx3pEUo2urSMY94r5PYCwHurNMvWAYobW0dGa5oL6KfR3RjZcefYNvb92usRD3j9/wBtHOb0NNnDcPdBd3aHt7R+RCS2Cio5F7gzJENcNV8WiedckYNQmfuO/z7uEUvheF5CM9+T7jA+SN40hPnq6ePPFYHudgygX1fXZ9k3G/2PkBoDZZAw1n+YsiIjKIKSAPQufrLmlpO6/+nItT2+T8cM+9/8BXrvku6bwhtI50PmTn/g48PIyLQuOxdL5YbwtR1wPjHK1dWULjONh9vI495kNgHK+3dRHzo84O2dCQzlsSsSj6lt7olg0MybhPOm+oTkR1w1WxGDlj+KA9jWuPxjWW4k5zOh/tOmdDS2BMMTwPq47TnQ3JhQZrHYExpHI+ibhPwo92dpMx7/+z9+ZRcl31ve9n7zPU2LO61ZNkSZYsyfIsMGAbQwzEYILDHPNCSOAGAi/rQnJ5b13g3qzLHUKSl9yXFcJbBJIQSJwYMCaAE2zCTAwYg4wtT5I1WXOr5+qaz7D3+2OfU92t7tZgNLSk/Vnqpa7qqlOndg3ne377u78/MkmFV6OphTFos7Cu2jTV4E19tvJrsVgsi2EF8jLkjW984yXns+3t7eXuu+8+6e36+vqWjDoDWLthI9dedz1fuffzZ3L3LBcBpZkZZhoRjSgmiDS+axIZPEfQjEzUWGpPkECsTaU3iDX1MMKVMklfMJFmGdc0sRBA1nOSjm9J9JkylgY3EcpRkvwghSCIY3K+SxRr8p6kESmyrsR3HbNALoyI4tkmGlpr2jIuA2158v6snamejxmvBvTkfTKuqRwrrWkAlSAi7znEOsSTZh+qQdxqxbyht0jFqS02TBaLxXJG2TVW5aF9k2wfmeFIqUElMLPOnhRkPcn/fPWmc+pBPlWsQF6G/Mmf/MklJ5DDMKQZxa2K11JUq9UT/v2ud/0OpampM7lrlouEW175GiZrATOJl9aIVYXvSOIkjsyTkkgrJKbFccaVrUxfU9+drS8LEaE1OFJQDWeru+lCt7asSyNUuFIgxWzHuE19RXaOVdCaVlW4EcY4TpQsiksfB4Y7cks2tMh5Dqs6c63LqQ84bd1cCaJWhjCYxXVdeZ+VRX/ZJUdYLJbzSxApHj1c4ofPTXKs3EQKwUB7lhuG2nnhqs7Tzh3XWrPtaJUvfX87Pz1YAmBdd55VnTnas+Y7LYwVk7WQrHv24xufD1YgL0PmZvxeKpQrVZ48WuaG4Y4lD95vf/vbKZfLJ9zOVz9/N7/7X/4QISX6FOK/LJcOa6+5kal6iFKmUpz3HNb15NkxWqY775N1JUdmmjhCMNSZRQD7JmvkPIcVBX+eUN0xWkFrzWVdeaZqIYWMg1KarrzP4VKDoY7sCffleGuD0prJWkhnziOK1fNqgpEy3GFE877JGh0Fl+6833oMaYWxxXLRkbZVd6Q4rQSZ0UqTH+6b5KF9UzxycIp6qMg4ksGODLGGH+yd4B+2HaLgO7x2cx9vvmaAdT2FE24zVprv7B7nsz89yM6xKr0Fnw+8dC2v3dzX+i6aS5pisRyxAnkZ8ulPf/p878I5J45CHvvZT3Dki7mqv23RRgj33HNy28TIyBEGNl3L0JrLObR319nYVcsFyhf/vz/mt9/9bpzjkiPaMx4rCj5SCDxHknVl6+9LpTOsTaYDM64k1zFfzJ5MHC+GFIIVBXPwcOXzF8dzWXvclKUVxxbL2WdkpsGOsSqHSnUaoVMc8L4AACAASURBVMkYz/sORd+l6JvW6G7SVr0RKapBRDWIqTZj6lFMI1TUwzhZd2AW6IaxIlTp7+ZyM1Y0wphqEFMPY5I+PeQ8SVvGpSvn0ZP36cp7dOc9unI+XTmPahCxf6rOtkMl9k4am1V/W4bXbl7JLWu7ecFwR+sEPYgUPz9S4l+fGeWfnxzhi48f5Yahdt549QAvX9/TSrEBmKoFPLhzjM8/doTDpQaru3J88MUredML1uMv0wrxybACeRlSKpXO9y6cF77yxXu46oYb+fnhElIIuvKeiaBKpqaVOnlaxtTYKLUg4pfvehef+diHn/e+CCHR2lagLyaqlfKi1ZXBOYI2d4qV28wF+oVvsVigHsbsn6ozWQtwpWB1Z46VbZnnbT2qR4qvP3qYrz41wp6J5+/tl8J8B2VdSdYzC209aZoBeY7Ac8zMl+sIfEeS9x3ynkPOMyk2kTINf2YaEVP1kMlawN7JGpO1gCCeTYnKew7XDrbzK1eu5Oa1Xazrzi/63H1X8qLVXbxodRe//9K13P/0KF9+4ij/9cGduFKwfkWBou8wWQvZN1lDA9cOtvOBl67lZet6KE+OXrDiGKxAtiwj6mHMeLVJxjVn22OVoNWSthkpPD9DGDRPuA2tNJv62tjwnt9Ba/jyp/5fSpOn1nVvRf8wsYq56fZf5Z0f/AP+89tfy54nHyf1g55JPM+n2NbO9PTkklaQvv5BRkeOnPHHPlN4mQxh88Svx1Lc/JrX88MHFraXfs3r38oDX/niL7prOJ5PHM5vP/7Cm279hbdrsVguTI6Vm3x/7wTf2zPBtkMlYjX/e32wPcMrNqzgDVcNsLort8RW5jPTCPni40f5p0cPMtNUXDPQzu/dupbrBjtY3Zkl7zkIIagFMdUgohLEhLEyHSi1Jus6FHyHou+Q9118R5yV9QFamxSbqVpI3nfoynmn/ThdeZ93vGCYt28d4mcHp/nJgWmeHatSDWJWd+V41RUreNnlPVzRuzztEs8HK5CXIZs2bZrXma69s4ue3j727dp5Hvfq7HPzHW9iuh7hytjEV2nwk3a3sdL0Da/m8ElsE+2dnYBZOPW773svv/u+93LzxiGa9ROf1buex//zyb/lmq03tq77wte/y/Ztj/Af3vjqk2c0iyTWS0rUKeRCv+0/vI/3f+SjbN/2CJ/+8z/m4R98d8FtXvqK29l41TV87CP/aW77MtZdeR35tnbWbLySB+/5DNFxQhBMF8OT5VMPDK9m45ar+d43/nXh/V2Xl77i9kX/lnLVdVu5+voX8uBX7qWzq4dndzwJWiOkREpJvERL7Ztf+Roq0wsXUr7jvR/g/R/5KNe/6CXc85lPsm/Xs0s+tuN6vP6ut/Powz9k3+75t7v6RbeycvgyvnXfP7Su6+sf5BN337fk9iwWy/Oj0oxas3yONJVN35VLemGV1jQjRTNpeNOM4tbl9CeIk99jYzeoB4p6FFMPYuqROR505lw6ssZGsKJgfnqLPh1Z0768EcbsGq/y6OES3909wZMjZv3KZV053n7DEFeubGNFwSeMFXsna/xw3yT3/PwId287zM1ru3nrtQO8aHUXziLPY7TS5J6fH+a+7SPUwpgXDxV4983ruXawfdHn3JZ1acueP7klhKDguxT8X3wfpBDcuLqLG1d3nYE9W97YVtOnyLluNb1p85U8u3MHa9Zfwb3ffhiA37zzlTy9/efkCwXe9Ovvoq29na//8xeXFM5CCF700l+i2Wiwe8dTNBqNeRXYQlsbjuMyM0esvOO9H+AfPvXxBYKwUGzj1a9/M4cP7OeZ7T+n0ajTPMWubUJIPvyx/01papKtL7mF3Tue5mMf/v15t0nF0Z6JKkGkkpX3ZjW/1iYZ4Nv3/SN/90cntk185I/+nDf++m8tuP51N13L0cMHGRhaRaNeY2pivPW3tes38gd/+vF54ngu27c9wl/+0X/nuT3PMjM9bawXQqCVRmuF4zj85//1Z63n971vfJ2//6u/aN3/Na9/K+uu2MjeZ3fy5GM/45de/Tre/5GPztv++972epqNeus6z/f51Bfu55qtN7J92yP865c+DwJe+6a7WLPleg6XGgx35Njz5DYeuO8LPPLD73Pwub3JfTP83//9j/nRd7/F9/5tcYHruB5/fe+/APDut7x2gZjdct1WPvjfPsbv/NrrCIOFAhzgM//8jXljtn3bI2z78UOLvsaO46CUwvU8PvWF+xf8/TWvfyv/8+OfWjDu2378EOWZGf7xrz+BUgrpOLz+rt/gtW+6i2u23siX//Gz87azdsNG7v32w2zf9gi/82uvIwpDXNfjU1+8f8nX91LFtsddHDsui5OOy0Q14P6nj/Hw/il2jFZakV3HIwV4jiTjSFxHECSCN4yfn+RwhIlTzHsOUsB0PaIZL5x9c6WxItTD2b9t6ity2/oefmn9igXe/LmMVwPu236U+544ymQtpLfg8/LLe9jYVyTrSkYrgameHpwGrXnFhl5+64XD9Imqfc8swoXeatoK5FPkXAvkciPi3x7dyZpVQye97W/e+UqeeeIxOjq7KE1NoVSM4zj89Ze+vkAU/MH7f4cffe+b3PTyV7UEyVxhkwqyP/7I/8WzzzwBgJSSv7nvgQVi6N1v+RXiKGxdt3bDxnliPRWG6Xbn8uV//Cz3fOaTgOBt73rvAlF7uNQgVppVncYfWg1i8r7DV/7pc3zjq1/i9l99Mz//yY/57jf+Bcd16OsfXHQ7i5EK0igMcD2fT97zlVMWT3PHCpg3bsc/v+888DVue82dp7xP2378EB1d3S2hfbqC7kfffpCdzzw9775z92P9pivnCe30NqkA/84D91MqTbH56uv43Ne+teD53vu5v+UH33qAodVr+NAf/tlJ9+/4xz5+rE5njI5/jy71OHO3k95n4+YruekVrz7VYbxksEJwcS6GcYlixWNHZvjhc5P85MA0pXpIxnNY3ZnjxlWd3LZhBf1tmdPa5tP7DnL/viZfe+oYzVhxRW+BawfaGWjPkvckGpNgEMSaIDZV4DBWBMmiMt+RZNz5P1nXmb183N/T2+cSj+1i9oN6GDNRDRhPfsaS/4NY0ZXzWNud56r+NnqLp/dcg0jx7/sm+ddnjvGzgyXTZCdhuCPLbetX8KZrBloLci+G98zZwArkS4TlLJDnciIhcbqcbFvHVzav2XojH//YR/nug/cvqJKeSc7El9GZHKflgv2SXhw7Lotjx2VxzvW4aK2JlLEd1MKYWhBTC02aQdaTFHyHjqxHe9Y9YRLJWKXJj/dP8dA+I4qrQYznCK4f7GBlW4ZaGLN7vMr+qTpSwK3renjzNQPcuLrzhNt9dqzC5352iG8+O4YjBa/dvJLf2DrEZV3Lr7HD2SBWpoNmM1b05D3as96C29jP0uJYgXyJcKEI5EsB+2W0OHZcFseOy+LYcVmcueMymojOPeM1jlWaBJECAcUktivvO3iOaHVZrAcmdqsSGKFrYrti6qHx1IZJZTVSJq4r/f1UcAR05jy60+iunBHNk7WQXYnwBegt+Ny8tptb1hif6NzuiwCHput89aljfOXJEabqIUMdWd5wVT+3rutmTXceAUzXQ35yYJqvPDnCzw6VyHsOv7KhnXfetOG0q7GXAvaztDgXukC2i/Qsy57bt25iYuwY+UKRT9x9H6svu+x879IZ5TfvfCVPPf4onufxtne976xV3i2Wi5FaEPP4kRn2TdaohTFZT7KuO8/1Qx2nHNs3F601P3puks/+9BCPHjaRmxlXMtCWIeNKtIa9QUwlya+dK3BdKShmHAq+m8RvmbiurpxP1jO2gdQj6zkST4pWhNfc2K68b6wHzUhRacbMNEIm6yGTNRPdNVULeaLUoFQP6c77XNaV4/VX9XPjqk6u6C2cMKFguDPH7968hne/aDXf3TPOfdtH+MQPn+MTP3wON2k00UgW3a1sy/Afb1nDG67qR1Um6bTi2HIJYQWyZVmTimOAWrXCu95wOx//u3suGk/pb975Sp56bBsAYRC0FvdZkWyxLM10PTSRXbsn+MmBqXkZrykZV3Lr2m7eet0g1w22nzTWKlaa7+4Z529+dIDdU01WFn3ed9NlvGxdD+t68kvaELTWKG3SGbxFGhwtV3xXcvvGPm7f2MehUp1HD5XYP1UniBUrixmuG2xn88q2VorDdOU877DFco6xAtmyrEnF8Vwe2/bIRSOQdzz5+ILrvvvg/VYgW845jSjm0HSD0UqzlRE7t7qZxnd5jsAVIonomo3jMrFdCyO70tiuIIn1qoezHttaaLqAOVLQmfPozHqtyK40tmtFwSdWmrFKwDOjZX56cJpHD5WINQy0ZXjT1QPcsq6bDSsKtGc9Ks2InWMVvrdngm/sGOObu8bZ2FvgzdcM8MoNvQvithphzAM7RvmHRw9zYKrOcJvHH7xqA3ds6jslwSuEwBHgcOF2KhzuyLValFssFoMVyJZlTU/vygUi+bqLZFEdwKarrm1VkFN+6dWvO097Y7kQqAYR9VDRlnF/4Y5+Uax4cOcY9z99jO1HZ553BNfJSEV2NkklyHsOWc80SOgt+ERaU6qHPF1qMFEL5kV0Hc+67jzveMEqXrFhBRsXsRN05rxW968P3LKWr+8Y5fOPHeEPv72bP/3eHq4b7GBNdw4pBAemTcvdZqTY3Ffkj+7YxA2dMT19/WdlHCwWy4WDFciWZc03tu24qD3In/vat6wH2XJCIqV54ugMP9w3yUPPTbJ7fLbpzXBHlpes6eK1m1dyVX/bKW+zHsZ8eccU9+3cz7FykzVdOe66bpBNfUVWtmXwHeO1DZXJrQ2SBWZpjm2kNFn3+NguZ+F1SdX5RCkJi1ENotnoroppB9yV91pV4lMl6zm88eoB3nBVP08fq/DAjlEePzLD158ZRQMD7RnuvHIlr7qil+uHjA1jenzhrJXFYrn0sAL5AiFWmtFKk5FykzBWSGGmJLtyHh05b8muRRcD39i2Y97li+UAFsWKJ0bK/PJHP8dLw5ir+tvYOtxxvnfLchK0NlmvzSgmTDJeQ6Vbma9RcnmuhaAWxgSxIu+Z9IPOnGvSCHIenTlvQbeuY+UmPz04zY+em+Lh/VPMNCMcKbhusJ333XQZHRmX6UbEUyNlvvbUMe59/Cib+4q86ZoBbt/Yu+TitMlawJe2H+ULjx2h1Ii4fqidD9+2npvXdJ2VFrfPl7Tr15mKEhNCsKW/jS2ncRJhsVgubaxAXsbEWvO93eM8uHOMh/dPLdmxCKDgO7RlXAq+0wpsV1oTK/MTKU2sNbFJKUII0zJSCnPwSP8X0LpfpHTS7lnQnffpLfis7sqxpjvP2u4ca7vzdOf9Ez6HWhAzUm4QxKbiNNyZu6jF/KlwuNTgK0+O8LWnRpiohfP+tqLg87brB/m1awfJPo8V+JalmaoF/Pu+SX56sMS+yRpT9ZAgUviuyZoteCY5wE+aFMRKU00yaWth0uY2Ebxn0okgMLaArrxHznUYKTda74vuvMfLLu/h5rVdvHh1F8XMwq/sSjPigR2j3Lv9KP/rW7v48x/s5TWb+njR6k5WdRpf6d6JGv++b5Jv7xojiDW3ruvmTRsK3Lx5zZl7IhaLxXIRYQXyMqTajPjLh/bxF98/yGgtoifvcduGFVzd38ZAe5aMK4mUZroeMlUPKdVDZpoR5YaJHRICHCGQSWSPIwWOML/LRJwqbVo4z/1faVMdc9L7JPdrRIqJWsBoucmjh0utCCCA/rYMV/W3MdCeoeC7NCPFeDVg/1SN/VN1So35LYwzjmTrqg7u3LKS29avOO2p1wuVIFL8YO8EX3lyhJ8cmEYIuHltN7+yeSXXD7WT9xwePjDNvY8f4S8feo57Hz/K7926llesX7GsKntnmyhWPHq4xE8OTHO41CCIFR1Zj429BW5Z281w5+kvJNo71eSr23byjZ1jRErTk0zVr+/Jk3ElQaypBhGVICaIFLV6SBArHCEo+A49eY9VXpa877S6euVTMe0I3DkL2Tw5eznrOhT89PYSz5FUg5hyM6LUCJmqmeiuqVrIRBLdVQtjbl7RzeU9BV4w3MGG3sJJPyPFjMtbrh3kzdcM8PiRGe574ihfe2qEL20/Ou927RmXO7f0c9d1g6zpzl80MzEWi8VyNrCNQk6Rc9kopNKMWPU/vslA0eW3XrSWl13es2yqrkprRstN9k3W2TNZ5amjZZ4erTBeCWjGCkcKunIel3XluKwrx2B7lv62DDnPoZysLv/O7gmOlZtc3pPn/7xpDbeu6z4tETg3fDyMFfsmaxwqNfivH/wAQbMBQR0aZa7ZspmP/e+P4yd+SID7vnAPP/jud3jBy17Fy179ulYb1DA2Ffa5kU3pCUN6OV2J30w8mRptsk7T7NJEPHlJS9RyI+JQqc5jR2Z4eP8UpUbEyrYMr9+ykju39LNyiVavjx4q8aff28Ou8So3DHXwn162jk19xZOOy76Dh9k+4/DQ3kn2TNSYqAU4wuSyDnfkGOrIMtyZnf29I7toRfIXJT3pOt42sBTVIOLh/dN8b88ED+2bpNyMcKVgsD1L1pVM1IJWRXXLyjbu3LKS2zf2nnTfd4xW+PTD+/nB3klynuTOLf287sqViy7sutgIIsWO0QrHKk0AhjqyXNFbnPc9YpsbLI4dl8Wx47I0dmwW50JvFGIF8ilyrjvp7Rmv8tizz520k96X//GzfOeBr3Hba+7kjb/+W63r3/KKF/Pcnl2suXwD93774bO7s8y2S3WlOKn4UFrz7V3j/NWP97N/qs4Lhjv4vVuXFoHHt4WeGhthVz3DvY8f5ScHpqmFS1tPlgMdHqyMxvmVLQM8+PE/YOeTj7Hpqmv53Ne+teC26ev58lffCVf/Mn/18H5K9YjXbO7jrdcOsmVlsTW+Smt2jVX50XNT/GDvBE+OlNGYcP9r+tvoKfhoYP/hEfaNzTCtfZpiviWm6EIhl0GllhytUWr2BEFpjcbMSKQzE9XyDHEU4DoOnd09RMlCrijWREq17AedOZfeQobeJK6rr5iht+iT9xzqYcyhUoNnRiv8/HCJMNZ0ZF1uXdfDyy7v4cWrO+dZTI6UGnxr1zhf33GM3eM1Mq7ktvUr+NUtK7lhuKNVZY1ixcMHprnviaP8+95J2jIub9rYwW/ctIGO01jcdSlgD+qLY8dlcey4LI0dm8WxAvkS4VwL5P6BAY6NjJyzxzsvSAeu/mV4yf8BuTZ46jvwo7uhMrn47YWEK26GF7wBVq6H6hTsfhgOPgFTRyCoggb8HOTaIVsEPw+ub34AwiZE6U8IcQgqMr/rGLQGreb8n/6uIQrM7aMA4sBsz8uClwM//T9nnpcQppJdOgaViec/RpkCvOgtcO0d5rFq0zB91IxF54B5ngAju2DPI7DnJzD+3NLb83PQ0Q+d/eb/jn5wvNnnqtSc5538mBcLpDTPS8g5P0AcmR8Vz46nkJDvhGI3FLqh2AOFTnN9ShzC5CHY/xjsfQQOPzPn8U7AyvVw1atg061mfCoTZpwdD3pWzY7TY/8Kj94PQY2B4dXc/6OFmdOXMvagvjh2XBbHjsvS2LFZHCuQLxHOpUAeGBhg5GIXx3PJFODGt8D1rzMi9dH7Yc/D84Xg8FVw9e1G2E0egp/9MzzzXSPMLgX8PGy4CYauhLYVRrDPjMKRZ2D/z83JwnJHOpDvAi9jTlRq00ZMP19cH9a/BNZuNUI8jmDqMBzYDvu2Ldi2FcnzsQf1xbHjsjh2XJbGjs3iXOgC2S7SW4ZcUuIYoFmFf/8sPP4A3PIbpmL6orcsvN2hp+D7f2sqpVxi53VBDZ76lvm5UFExVMbP3PaiAHZ83/ycAkcPHThzj22xWCyWixorkC3Lh5lj8PU/g+/9zWylVMVQHoejO03F0WKxWCwWi+UsYwXyMmTz5s0888wz53s3zh+1adj1o+d99/bOLqIoolYpn8GdslzovPjW2873LlgsFovlAkGe/CaWc83TTz/Nxk2bf/ENXeRRVktRKc/we//lf/CzA1MMDK8+ya3nj5GQEs/36eld3Df14ltvY8t1WxFS4nre7P2TsZbSWfK+ywkpF29Cks3n6esfJJvPk88XTvgekvLcfH1IZ+G+Oo7Teg6e7/PiW2/D8xePzQPzun3i7vvO2j5aLBaL5eLikq4gP/jgg3zgAx8gjmN++7d/mw996EPne5da/PTn2/m3R3eyZtUQ27c9wt9/8uP84FsPoJOs3pRMLs9v/8cPtiLQ5jI3Hg2Y9/uH3vdORkeOIB2HG29+GQ//4Dut+73jvR/g/R/5aOvyxz/2Ue7+9CdQajZOTQjBldfewAf/28fY9uOH6OjqpjQ1ydaX3MK9n/tbvvuNfyGTzXD9jTexf+9u9u/djSMdELD1xbew7eGHCAOTBOE4Du/94EfY+pJb2PbjhyjPzPDs09tb0XUvWN21YHze8d4P8IW/+xTNZmPB31Qc87EP/z4f+/Dvt65ra++gPFMy+y4l+XyBaqXC8V5mrRTZTI6JsWP09K7krne+h6//8xd5bs8uMtks23/2E+r1GmvWXzEvPu/jH/so9/ztJwnDgKmJscVf1FNESonSmkKxSKPeII5mu+0NDK/mD//yr/n7T36csdERVq+5nKnJMW57zZ18+e7PsGvnM2y66lq2vvgW/v6v/qJ1v/bOLgrFNvqHhlm3YSOFYjtf/fzfU5qev7ivUatRKLQxMDRMo97gxi0v5x3vfT+7dzzNV7/wD4RBiOf7jBw+xOQJnqeUktvvfDPXv+gl3POZT9KoN+gfGqajs4ue3j4e/cmPeG7PLvKFAkEzIAyarfvmC0U+cfd9/N4772JmegrHcXjD297Bxi3XtN5j6Xt9+7ZH+ND73skjP/w+l61bzx1veCvf+7evs3vHU6zftIWX//IdbNx8JTe94tW/0GtisVgslkuLSzbFIo5jrrjiCr75zW8yPDzMC1/4Qu655x6uvPLKRW9/rmPeHMdBqVOIvLJYLKfEzw5cAEkf5xC78n5x7Lgsjh2XpbFjszgXeorFJWuxeOSRR1i/fj3r1q3D933uuusuvvrVr57v3QKsOLZYzgaLzURYLBaLxbIYl6xAPnz4MKtWrWpdHh4e5vDhw+dxj2ax4thisVgsFovl/HFJe5BPxqc//Wk+/elPAyab+MiRI+fkcaWUViRbLGeB6fFj53sXlg2V6SU6Vl7i2HFZHDsuS2PHZnFOZVwq9YgJKviN7DnYo9PjkhXIQ0NDHDx4sHX50KFDDA0NzbvNe97zHt7znvcAxrMyODh4TvYtjmNrs7BYzjAXgwe53IiohTF9RR+loRkplNZEStOZ8057e8f7A4+UGvQWfapBjBDQlnGpBTH1MKbgu+T92UQRpTWNUOFIwWilyWB7llhppBS48sJO0LF+0sWx47I0dmwW52TjElcDevqKDJ5DD/KpcskK5Be+8IXs2rWLffv2MTQ0xOc//3n+6Z/+6XzvVovparOVYpHSiGJ8RyKAHaMVALryPiuLPpHSOFIgk1iuRhhTC2OyrkPOkygN9TBGCkHWk63bzUVpTazMms3DpQZruvMcnK6T8xyakWKoY3mc4V0sCyJGyk2yruTozGwSx5ruPDlv8Qi2k3GxjMuZ5kyNSxgrpushvcWl4+QmqgE5zyHvO0RKzxOKjTDGkQLPMc42pTUCkwizFEprmpHiwFQdjQYEQsBUPWjdRgiB1hqlNd15/5SeSxAplIJ9kzUaYdxK83OkpDZVRwNSCI6UGgghkBImagFazz7eXKQUPDtWQQjTBX3DigJCCGpBjOuY53wqovn4MbNYLJbzxSUrkF3X5ROf+AS33347cRzzrne9iy1btpzv3VrA4VKDmYaJ+Zp7HJXSHKSmagFTtaB1YDqeudcvdZvjbw/mdjtGTaONSjMCaO1HxpUMtGcp1UO68j4aTdZdKOqakamAazTj1QA0DLRnCWOFhuctBC90IqXZlYgJSAWSeWGem6wBsKozRzFzyX48UVqjlMZ1zuwyCaU1B6fraA29RZ+RctO8HzUMd+RoRDHVIKa/PUMzUhwpNVjVmWO8GlAPY4QQjFfnf94yrsR3JN15n7Fqc8HnreC71MJojrA0/8/dRkfWo9QIWd2VY7oekXElY5Vmazue45BxBE4iHl0pqUUxDuadE8aa8VpAI1T0t2cWnABHsWqNZTWIODhdpzFdI99dwHdnt+05EiepADtSMF0PcWQqwCFUGlcIwthUjmOlCZQi65qTaE8KGpFi90QNrTWeI4mPi6acOz7FjEu5EVHwXTpzLodLp36yqLSm2jTRk4dKdQA2r2yjEcVUmjErCqd2smCxWCyLcekegYE77riDO+6443zvxpLUw5h6I0QmlWHflUjATQ5MMvm/GSlcKdBArDRxcvCVmPtJmRyJhbm9IwUqqRQrrVvVIgAnOa6mFWlXCpqRal0WQBBr9k8ZITdVn83o7cp5ZFxJMzbCHVLBnQpAwa7xyjyR3t+WoesUq15nm3IjotQMGe7InfB2jTDGcyRpoUuZoV20Kp9ydKaBFOY1Mic0RoBkXYnvSjwpaUYxzVjTjGIOlYyIu6K32BJFp4rWmljDdD1krNJk88q2eX8PIoXniBNWLs8X49WAnrzHztHKoid061cUWhXYU6EZKdBwcLpOxpVM1kKyriRQiiMzTQSQdR1iDUcrDSRmXJ6brLcE49FyE08KMq7T+ixpSGZsIIg0jUhxZKZBxnVwpCCIEkGqNaHSdGS91vORUlAPFWjdEpmVMEJKwcHpOkIIyk3dEqttGZec5yCFEcYquV+bcnGEIIgVUkA1iJmoBZTHI7TWFDMuwx05tNbsnqhSzLgoZQSy50g8R9KZ9ci6kozr4LuCKAbPmX1f+HPGOp1h0pj3UNaThLGmHsbkEyFr9j1EJvuV81wipQgjZcYOUErTjDWuIwhjTSHjadIYggAAIABJREFUEkSKY5UAz5EoNErB/qkaV/QWFxX7042IsUqz9f3iOAKl4Jlj5WQ/zPen0prB9uxpvWcsFosFLnGBvJwZKTeQAnIZNzmAmQOak4gsKUAkXdwiZSqyUayI9WxvOFeag316fFHKTMP6rqQRqkTMmuqZK83BKutJdOJt9F1BxnFoxjGRMtPBjhDUwhiNuU0Ym20qrSk1I2QgUHr24A7JATeZmlVaE8aaIFaAZqTcREpBR/b0/ZMpU7WAjpzHaCVoCfMURwrWdOVxHcGzY1UcIcj5kqzrMFZp0lvMMFZpcnlPoVWFOqDqVANTNS9m3FYF/WR0ZD16Cj7T9ZCMKxmvBoTxQh+5n9hefEdS8B2EEGRcQaxcIqWpBjG1IKIRKZ4dM1aa40Xu8URK0wjjlkBIEUIsuC5l88q2luBxBMQ63ZbClWZKvBHG7JuskfcdhjpyuFIQJdXDahC3qtxhrIi1Zt9ErbX9nOcw2J7Fd2VSMV1akI+Um63XLhU+rpQIIYiUQgrzuLvHq637bOxbKJ6OH5O9E1WapRqF7iK1MKY969KecamGMRlX4iYnKunrZE7joBYqcp4RVRlXpueXaE1LADvJ5yt97p6UxErhuw6NKE5mVTRam20EsU4qs0ZUx1rjO5JGGBMkJ0ZpxTWKNQXfoeC7uM7sZ73gO1QDUzV1peCK3iI7x8poTXLSZr4fyk3jVd4xWibrmvdYNYiTE22HjqxLM3DpzvvI5ORu/YoCkdL0JJXXKFYcmK7TnnHZP1UHBNoxlWQ/sUy4UrfEu3lsTdYzFhQzVhArYw9zpfn+irVGClrfHb4jqUcxSkEzMnaPIDYzCM+OVZNtzVbchTAnGUJA0TfvP4X5TtTa/K6VNt9TGmphTEE8P1+01pqJWkh7xsV1xAnfbxaL5eLCCuRlSjWIKWQcegq+ObAkBZDWtPycFsle0opXJ17jtF47V0QDiYfRXOc7EiHMgZvk9qlQVlqT92Xrvp7jopJSnhCQ9x2UNgIkSsRBLTAHdyFMddpUn41AM9UbjcCIkSBWrYU/tTDm6EwTR4hTthQEiRc09e4KAccqAaAT68nsM1Ya9k5WWwfWSGnKDUWZCCFgvGrE2J6JamtszXQ4iaiY/T2dHjeXSarwIqkCKmaaITNNU1GfeyBPxYvAjHfeNwJ5bhVaYMbLlUYsZ11JPYwpN6N5wvd4odyMFPUw5uhMg2a5SbazHUcKnGTmQAiIY/PKR0q1RIbWelHhvJglRwjMtPlx1f/F7zNrF2lEMXsmqvNuu6Lg05XzUNpUi0uJbUdKgZsIPEcKfMe8Rz1HECszM1IJIgQkVgfYOVqhO+/TW/Q5MFWnJ+/jJX7XahBxOPHPCiHIew5Zz4x71pVkE1GX7nvGlfOeRyEz+/nRyXs3pbfoM1YJWp8n33FanzetzXPIuCJ5Xc3na6gjy+FSg4Lv0NeWaZ1IrGzLcKzcRKOJYrdVHZbCzEys6c7TV/QXnFzMPeHYOtzJkZkGx8pNOnOuEcGOoBk5xoYQxGRch6wrcKWkmHFxpaBSdxnqyDLcufiMietI1vUUzOuW+K6V0jw5UmaoI4srBbvHTWXadyRDHVk8xwjxeM4MlBSC7UeTLpYIPASdOa/loxaIVtVaabd1Et2IFDmliGJzMpFWy4UQeIkFpJhJLBjJiUt6O6Wh3AyphyrxURvbUsE/+XdM+pkSAvZN1JBSMJ5YZ9qzHv1tGcJkvy5lG5TFcrFjP93LlO68x5jntjx4ed+ht+Az3YhY1ZklM8fzW6qHFDMujhTsGqsw04haB96+YoauvIeTWCWcRPTlfYdyMzIHZz1bKb2sK8f+qTr97RlWFHxK9Yi+tgwHp8yUc9aT5qAhEquHYw7mec+8lebOZC4m5s1tTEW8Q3lM1Jo0QsWhUoO85zDcmV20SpOKcTAr7bOiieMIHCHJJAfXdNo7Fej10FTBI7VQ+EnmGD8ERApcYe4fJvaTVKQ4yf4YG4URgb5jLBGOkEgpCGNjj0hflSDW+I7AdWYFmhSCnCfnnbxkPYfNfUU0UAti2rIu2w5Ok/fNGOU8h4laQCOKUWppUSsE+K6kmEmqjomASL28aR07iJQ5uUl84EGs51lFtE4WSYnZSjEYARvFxkuaWnJcR6CTsQZTgXaT6qiZ8jbVU6WNwAGYrIdM1ALSeQ4hjMXBc2afb7o93zECP3k5KGaMEJ2uhwjMSeRUPWQyqTynMwBmu8YSkfUkYeDRU/BadoHBjiw9ed9UKiNTaYyVpjvvIZMZkowrUcpM/y9W/V7dlW/5aqfrIUGsWVHwUcn4LVYt72+fXeTamfValoPhzhxKaX5+uIQGNve1ESWV/aWSKeZuXwjBUEeO3kKGelIl/+nB6dZ3R0fWpFxkXHOinb7vVnTnlxTHSyGl4JrB9tblras6F9xmMdH4glWLNWkx4jue83krNyN2jVVxpW6dtISxsXZIYcSrECRVa1ONT0/iWuOBSE40fIoZxVglwEvsKzpZB9GZ83jmWHmez9nYPJpMlptMyVoytqZqb6rdphq9a7xKejIeT88/Y1zbkyfrOuhE5PuubH2OlqOlyWKxLI0VyMuU1V159h4wU6ieYw5oMFvJmUvHnIPoht5i6/fjD37pgaAta1729qzH0CJ+2/ash59UlvvazH1Wdc3eLl0pP1UL2JtUwtK1NIUkBioViGt78jRCRVvG5eB0ndVdOWKlefzIDI4DKwoZakHEdCOiGSl2jlYoZlw8R9Keccl5piq9K7EapFPvBd8h486KqlT8ptO8AkEzNtO6sTLTuGESmyeFqbAa8aVbJw9zPZ7mgCwS+4GYdwBOhXN71p0z9e62qmaR0oC5v6nIz544CAQr2zJkXbngtUxfl62rOqkFEc8cq+D6Do7M0AjN4rEgVsTKCJ60Kp0KcZRPTzHbeqx0LOaSOj7SfWxGykzrJ883jBWuM+s9n5uMEivdqsK7SYXaLKQzzzlSmmxiR4i1BvOPWGvCSLfGcO74hLGJJ8u4sjVNfzypJTZ9/2ZcSRRrcn7MdD0yQhxTIc845gSkmESSZT1JqemSdR029hVbVqWUjDs77inpez9lKWGTXj/XQ+9waiIoFcdzL1872I6G5+2X9RM/O5j3rOdKVnflODBVJ4xNCs1ckX7kyOLWm3PNXI99e9bjhuEORspNjs402LCiyLNjlWQNhHkPpO/pvmKG/rYsk7Vg3vMCM8vw9EgZjUAXfDKupDFlrDoj5QYj5WbiNa/RlfeJlKLcMAspXSnwHQfPEQSxpi3jttJG0oXH9TBGASEKldhoHCHYN1HDcyRhPDtbk/6/sa/YsuikJ3MnW+9gsVjOH1YgL2Ou6C0uOHifC44XCEvRlffZklT8TnRQT6vdlyU5h64j2Lqqk20Hp3GloC3r4kjjU52oh4nHMmY6mYJNDzKONNPj5NyWqEqncOdPg4vkcZx5U+RzExGOv20+OfCa643XcrHp9RQ5/+6QVNSBBV5HKcxJzNru/CkvuMv7LltXdfLcZI2JaoArBQXfTaaYSSwtZs88RyZCUM5bYJXzHOqh8auu7cmzb6JG+tKm+5hWVUXLTjBrrUltOOk4zK3Updelv6cWnFTgzhXmWoPOzF9wN3dbx4/XtYPtS6ZXHCs3OTRdx3HNSUHWdYiUWYgX69lqdtZ15jxHsWilc7lxJhM75j7f55OPfD4RQjDQnmUgEb1bV3VydKZB1pV05f0FcXnHi2Mw7/30RPPJkTKOMCemQawpN9KZI0EzMjYmMGsDHAEi49JR8Mg4xrKW8yRxMguVnqQVMya1I1ZGLOtkbcdMIrIzrrG3CCGItUIgWusJ+tuyVJoRWsOzQYU1XfnWd67S2vqcLZZlghXIll+I7C8Q1XbDcAdaw0wzmvW3Cqgm09vNZNWYk1gfXCloz3rUQpdixojfK3qLFHyHqSSOaq4YSA+qed/hyaNl1nTlmW5ErE4q61KY2KpmpCj4Zpp/ZKZBf3uWZhQTxppSI1xQZS83olbDhGYUkz/O1xgnCxqfOlamLeMy0J6ZZ4k5HdZ05+nOe+xKFiu5jhGw6UF7rod5oD3LQE9+XkVTJ9VwN40hqzSZqoes7c4jgKePVQhjxcY+k5bx9EiZawbbCWOzAPPITIONvUUOJNPTlSAiSDKx0wVt3XmfJ47O4DqClcUMvcUMu8er+I409gdhZPeJiqvrVxTmzYQsxcq2DCvbMhwu1RmZaeK4ApDoxG/uHHeytK4nT90pnnijlmXPwBwRfDrfOXnfZVNfkUhp9oxXySVrNJpRjO86Zh2GIyj6Ls1Yk3EFkXIp+m7LviGFQDiay3sK5JNFknsnzPs7Vsb+ltooUv+3wszWREoRROb6SJuZj2OVJhnXIUhsTukaCTAifLA924q7E8IsnrTZ0BbLuccKZMt5I03Y6Mx5bB3u5OB0nbFK0LJMpPFV6eK+dAq+kVx31UBbS3j2LJJ5OvegmlbUjrc1uM78POa0GpVxHTLu4n7KuVX948UxzE4ZXz3QvuBvz4f2rNfa/yeOztDflmG8GlALYjpyHqV6yFUDbUyM1hZE5onEK5zSmwjYlFQMp9W49HHSy6loXXOSLkfHV2g39hlRellXznRhW2Qcw9hMeQsWWg5OxlBHrnXiEivNY4dLLSuGYH7FuD59Wpu2XGS0Jwk5cbc54QOXWCVrO7zZ3OdsurA26yGk4Mr+Ng6XGrQl7930s5VxHbrzPjON0PilmV2omTZEiRL/fagUgaNaJ7LNSCcnc8arHmvz/k397EIIjib2DzC2pF1jlZZI3j9V57KunPUzWyznACuQLcsCIQSru/IMtmeZroc8N1UjJ2en+1O7wMa+IjNOjcHB5T9lfjZIRfeJurmdLmczI1ZKsag4PpOP68gLw0JhOb+kJ4c/OzjVWmsghCDnuXRkXY6VTWOWrrzHuuT9tH5FYcnttWc9rh/qoBbGhLGi3IwYaM8igMeOmNQOR5o4x3TdQs4zJ/5OEpcYa52sf5iNGwxjSRCrZO2CIIhj9rTiEzU7x6ps6ps/K3K8xen4joTlRkTOd2wl2mI5DaxAtiwrXMcsXptpRhR9lxUFf0F1ceY87ZvFYrnwEQhWJTarrCdbFeaB9ixSwNGjtRPdfR5SzsZTzp292dzXRt53aETKdC0MVStrWwjZSgDSGvLebBKNUhBrRTNSyWJXSHsxBZFuLTRO02zmrjNI6W/LMFJu0pnzmK6H5mQAk2azpjtPI4yXTXMmi2U5YwWyZVmS5q9aLBbLmWSp2YbT7Vh5ItJZk5zncEXv/GrvTw9M0Vv0mWlERtzOSW9xHPBwcOXsor1MEhUZJFXqNKYRktbfaQpNEuE3Wg1wHUk1iHGShbyukMRC8dxkzTRYSZKCLBbL0liBbLFYLBbLOaI779Od91ndlWfbQWOQX7+i0MqybyRZyxlXUm5ErbjInJZ4STMUzZxMc6Xm5birxEvdnnWZqoWtpJdACQIUAtgzUaO/LUPRd85oeorFcjFhBbLFYrFYLOeIy+f4mherZmc9p7XWQCnNjtEKV/a3sf3IDEonHRuB0FVEShMrk+2dNv9xpfE8Z1yJyJvKtOe4VIOYyNXEWlENFCPlJmke+fFV7rmMV4NWvn2UJPRM1UOqQcRgu+momLad11pzrBLQ33bm1khYLOcLK5AtFovFYlmGyCRNA2BLfxuHpuvkPJPBfKzSwGs1SIJ4TiVYCpL0jYhYgecIZEa0uhY6IiJSimasaYQxM42Qgu+2OoXOZbzapBG6FIHd45VWIxSAQ6UGjaQtd9qRtRpEViBbLgqsQLZYLBaLZZnjSNFqtgSm4VIYK0YrTXzHYU1vrpWXvqozR3feo9SI2DdZpTvv04hiaoFZ0FfMOISxQEYK3zHRclo3Wg182jNm4WJ71gVMbN1ouYnTViSMtelQqjXNyPicwQjkahAjhGDHaIWh9uy8SMwoVkzWw1aXS4tluWMFssVisVgsFxhpznusZtM4bhjuAGarwD0Fn+cma/QV/dbCwWrTdBf0XYkjZxucVIOYSClcKZlphoCg1AjxHEkjimlGim5XIoVo5ToHyiwijGJNDOjEAB3FisMzdXTJLLh2pWD/dJ0w1lSDmJ68R6Q03XPSNCrNqGXT0Czedt5iOZdYgWyxWCwWywXK2p7ZqvJiDUSO9zkXMi7rVxR4brJGzpOtBX0AYWxSL7KuRzMyNgopBVlPks14tCXpF5FSFHy31fa7ESmEwAhlrYmkoBmZFvT7JmcznF0paYQxh0um0nys3OSyrjx53+FQqc5wR45KEDNVC9i8su2sjJfFcqpYgWyxWCwWyyVET8GnI+sSa00jVOwer6K0JlIStGni04wUkVL4rsSTknrTJec5uI5AKZPlnPMdBEnHTmk8zhpNI1JkXGPDiJUmiNNGJhBr0yHQFZJQKQ5MGwEthOBQqZ7YPIzQj5XmUKnOZV0n7uRpsZwNrEC2WCwWi+USw3UkLqZ1dm/RJy4bEauU8SFnXNPRL+NK0zZbgu8a4Sod6CtmGK00EytE2g0QtBb4jpwTQaepNGM6si7NSCV/02Rdh0oQEaSVamEi7FQioCtNkxNdC+LkfoJ6qJhphHTlfTKujaeznF2sQLZYLBaL5RJmVWfONBmpNHGS9XPtWZfpemgEsIDOnMf1Q50cKtUpZly68z6rukxHwjBWSdU55smjZTb0Fjg03aAaRLjSiN+c5+C7krzSRMrEy/muIIg0QazQWhMojSStHDcA0w5832QVEOikPfhUPWRTX3FRS4nFcqawAtlisVgslksYIQSrunKs6sqhlEZKI0bLzYii7yKl4MiRGlIKVi9id0i78mVcp+V5vrLfY7IWsG+iBhivs4NAOuAmDU18x8GVijwSEARRTJxkPVeaMfUwxk0ynl3H+Jc9RxJrxY7RCpf3FBirBqwo2Iqy5cxjBbLFYrFYLBbALMoDI5rbs94vtK3uvE8zUhwpNegrZhjuzBHGiu1HZhBCIxC4UuIkbbJdKdAaZFJ1TuPgmlGMK00nQUcKmrGgSczeySpSCmYmQgDyvtPyKyutW/7oWhCT9ST1MKbgz8qe8WpAZ9a13QQti2IFssVisVgslrPCQHu2FUkHptp8w3AHSsN0PeS5yRpXD7Tju7LVehvAdyTZxPOc84yAzboSDRRxKDUEodK4EspxhJN4lJ85VmawI8uRUqP1+EdnzO9CwKa+NpTW7BytIIRphLKpr40wVrhSWNuGpYUVyBaLxWKxWM4ZQggcYdI0Mq7ET+wRqT0jihXbDpVmK8tIYgW+Y9psg6kQa0wDlaRBIFGsCZWpWJs4Os1IudGyjJgmJuWkSm3yn2OlqTQjDk7XWVHw6S3aLoAWgxXIFovFYrFYzgvFzEIZIpPYONcRXDvYgdaa/VMmAm6yFrTuF8YaRwq6cxBpTRApgkgQA44ApSGITFRdnHirAeJYm0QODWESJec4gvFqQKyxrbItgBXIFovFYrFYlhFSilZXQDAV5zVJm+21PXm01hydaXJ0poFGk0u8yqGrCGON1qakrKGVyZw2McmYsjSuFMltYaZpugPWdcx0PaAtY7Y3169sufSwr77FYrFYLJZlxYm8wEIIBjuyDHZk2TlaodwMGWg3vmNHgMYIXw04QuA5EuUYgZzzHBL9jBQKJ2md7TmSmUZIPVQcnK4DgpwnbZOSSxgrkC0Wi8VisVyQrF9RIIwVWc9hZKaJcDRaC0RisXCEg+uYBibmskALI6B9RyIFONJIobzv4EpJuWm2XQ8Vz45V2LCigBDGxxzEOmmeoheI+GakKDVCJqpBa9+cxNYh7eK/Cw4rkC0Wi8VisVyQOFLgSGOJuHqgnX2TNXKeTDr7CY6Vm3TlPCZqAShNb9FnrBKQ8SQZV7Ki4LN3oorWkHUdMq7xKjfCGI3pBrhjtMLqrhxhbBb9eY4kjBX9bVnjj56skVVlALQGxxEIBPun6kRKoTW0ZV2GO3LncaQsp4sVyBaLxWKxWC54fFeysa8477o0Ym6wI8v2IzOs7sovaHZS9F0qQcSe8SpCmHSLjCtwmjG1ACJhbBdaG3uHWRwoGSk3AfO77zhoTJMTT5oIuiBWOFISK0WlGbFjtEJP3qMtY7oUrmzL2Fi5ZYwVyBaLxWKxWC5qPEdy7WD7on/zXUm367NP1BjqyHK41MBLqsCuFDSimGakkQIipfFdiSOgESqkkDiuJO87KK0p+g5KQzWIyXmSINYEEcRaIwRM1EJTzQbKzZjuvEdH0qykGZks5tSWYTm/WIFssVgsFovloudkHfPSHOb+9iyHS3VGZpq4UpB1JeVmhEhykzOuJOc5TBGQ91yCwG3ZOgq+qSQLYewfrpRUmhG1MMYRgnoYI4REAJFWjFUDRitN1q8osG+ySlvGoxHFFH2XzpxnW2ifR6xAtlgsFovFYpnDUEeOkZkmvms8zkIIQqXIuqYBSdaTdGoPz5FoR5L3jJxKF+OlEXFGkzu4jsARAt+RxFohhCCIzG3rUcyeiSqONEkaQsBUPWSyFrB5ZRs6aYoCoJKuKLY99tnHCmSLxWKxWCyW47huqINnjpVpRgrpSdxY4LsCGWscIci4Do40IthzjNi9fEWBWGn2TlSRQnDNYDuPHS4B4EqJ55imJQDK00ghGKsGIDWuI6ihUcpUn2MFzxwzi//SFA3zu1kM6LuSyzpzViyfJaxAtlgsFovFYjkORwquGjC+5SOlBkdnGmzsK7Jj1IhWzxEMtGfp0XnWJfYMgGozQgrRsmyYzoASV5rMZRIxnbbILvqmQp11Ja6MCGKF0iBQaJLuf0olLbONQAYIY82u8SqbV7adg9G49LAC2WKxWCwWi+UEpI1JAF6wqmve345UnXmXCxm3JY4BNq8s8syxCltXdbLt4DQAV/QWGSk3mGlEdOY8tAbXAVd6VIIYrTWRkoBOWmabxXthrFFak+hmokRlp5XmTX1Fm4xxhrAC2WKxWCwWi+UskfdnBfPWVZ3Uw5ic59CWLbLt4DSuFGi0Sc1wIO/JlqXCeJVjlGcW/9XDmCDWuEY7Uwk0O0bLreryjtEK63oKVIOI7rzf2ocwVq1saMupYQWyxWKxWCwWyzki581WnK8b6mgtwhPAY0dKZD2J1qC0wJHgSpdY6aR1NtTCmKzngIYwqSB7UtCIFAqM/1maBYGR0rhScKhURyDoKfhIwTzxbFkcK5AtFovFYrFYzgOmojtb1RWIVrc/0FzeU+BQqUEtiFtiWUpB1nVaojmKNRlXoBsRsdJEGFF8cLreWtCX9RwaYcx4tYnWRiArrYmVWSgoBcRKL1jwN1UL6LpExbQVyBaLxWKxWCzLgNSKsaozx87RCgXfZU1XnolawEB7hieOzpD3HIQwi//yOMSuxnckzcg0I4m1RmIqyqmlwnckQaySRX6aZ46V8V1JEKmWiE4ft5gx0rARxoyUm3TkvFZ83aWEFcgWi8VisVgsywghBJuSdArflbRljVy7qr8dVwpGKwFHZxr4LsTKVIBzvknKiJQmUpq2jIvnCGpBjO9IMo4kUAqBqRIbcSyYq30PTtdpz3r0FX32TdYQwviejSi/tESyFcgWi8VisVgsFwDZxL882JGlr+jz5EgZUBT9/7+9ew+K6rzfAP6cc3bZZQFRUBNcTLgsoiAXUUSjNiJFa23MjOIticGpie3ENqkZjTY2xj/qmExtNRozTlKSMLWjU40JtjUyVRubqTWUmNhfRAckYLiohUW5ymX3fH9/LNkusipoFFifz4wj5+w55z3nGdGvL+95X4NrajgFMKiul/IMZldBq0BxjWuGBr0NsPipaO1wFcqePcNOcb0o2Njm6FywxFU8V1y9BhHcd9PJsUAmIiIiGmAMmgprsBlNbU5EhlpQWHEFADoLWxW63lkwaypUxTVvs1MXWIwadAHMiutYTVHQ2uGE2jlXc2ObEwZNQYdDBwCIosCp6335qH2iT5Zf2bdvH+Lj46GqKgoLC7t8tnnzZthsNsTGxiI/P9+9//Dhw4iNjYXNZsNrr73m3l9WVoa0tDTYbDYsWrQI7e3tAIC2tjYsWrQINpsNaWlpKC8vv2UbRERERAPFsEATIkMtAICIIRaMGOSPiBALNEWB6rHKn+uXCn8/DSaDBoufhmCzEUEmAyx+Gix+BgwyGxDgZ0CQScMgkwH+RtdxRlWBoig4998mnL3cCHtzex8/9b3RJwXy2LFjceDAAXzve9/rsr+oqAh79+7FmTNncPjwYTz33HNwOp1wOp1YuXIlPv74YxQVFWHPnj0oKioCAKxduxarVq3C+fPnMWTIEOTk5AAAcnJyMGTIEJw/fx6rVq3C2rVrb9oGERER0UA1NNCEEcFmhAb4wTY0AJqiYLC/HzRFgYLOccoGDQZNQaCfAWajCpNBhdmgYpDZgEA/15hlf6MGi1HDYH8jAvw0BJmNMKoqVBVQVQW1PSyQnbrA4dTR5tDdC5kMJH1SII8ZMwaxsbHd9ufl5WHx4sUwmUyIjIyEzWZDQUEBCgoKYLPZEBUVBT8/PyxevBh5eXkQERw7dgxZWVkAgOzsbHz00Ufua2VnZwMAsrKycPToUYjIDdsgIiIi8gXB/kaMHzkYtqEBSAkPRvTQgM6FSFxDKgyqq2jWFAVRoQGubQUwdE4hp6muccv+Rg2Bfprra4MGTVGgi6CpzdGlPacuaPcohJvaHKiqb0WpvQVf25tds2vogobWDjh1QYdTh3w7dUY/1a/GIFdVVWHSpEnu7fDwcFRVVQEARo4c2WX/Z599BrvdjsGDB8NgMHQ7vqqqyn2OwWBAcHAw7Hb7TdsgIiIi8iWKomCwvxGhAX6obW5DSniwe6W+VocOf6OG0AA/OHXBF1VXXUMzOl/eUzXXFHCWzpX8HHoHHLqCyvpWiAhMBhUPD7HgwpUWtDtds2J8c+Uamtsd0EUgEPcqf8U1Te4p5RSgdZTWAAAR00lEQVQFncV3vypDu7hrd/b9738fly5d6rZ/06ZNePzxx+9Ws9+pt99+G2+//TYA4NKlS6iurr6n7dfU1NzT9gYK5uIdc/GOuXjHXLxjLt4xlxsbKNkoTh3OxjZcvHity/4rnb+LCOprmwC4Zqxw6gJdBOftzXDqrvHMaqsD0u4A4HrhrxnA/9m/fYHPtZR1na4DUNDeVA9VVWBUFTjF1YOsKQp0uOZpvqLruAIgFMHwazXfmxB64a4VyEeOHOn1OVarFRUVFe7tyspKWK1WAPC6PzQ0FFevXoXD4YDBYOhy/LfXCg8Ph8PhQH19PUJDQ2/axvVWrFiBFStWAAAmTJiAESNG9PqZ7lRftDkQMBfvmIt3zMU75uIdc/GOudzYQMnm4Vt8brW6CmXPOY9DhztQ39oBVVFQXteChlYHHLqgzalDA3DNobvXAjSoCjp0gQKgw6AgIGQ4jKrS2fss0AWAuFb6u+bQIRCEDg/BiBDLXXneO9EnY5BvZO7cudi7dy/a2tpQVlaGkpISTJw4EampqSgpKUFZWRna29uxd+9ezJ07F4qiID09Hfv37wcA5Obmunun586di9zcXADA/v37MWPGDCiKcsM2iIiIiO531y8IEmgywBrsj6EBfjCoCoLNRgSaNJg0FYFmA8wGFcFmo2ucsp+GAD8NZoMKk0FDiL8R/n6uccyDTK5jzUbXTBr+ncfUtfTPWTH6ZPDHhx9+iJ///OeoqanBnDlzkJycjPz8fMTHx2PhwoWIi4uDwWDAzp07oWmuSbHffPNNzJo1C06nEz/+8Y8RHx8PAHj99dexePFi/OpXv8K4ceOwfPlyAMDy5cuxdOlS2Gw2hISEYO/evQBw0zaIiIiIqDuj5lpYRNUARdEAAUwGDboO+GkKVFWByeAaeuHUAc2owuKnod0p/xvTrLhe7tNU12fXOpzor+/qKdLfXyPsJyZMmNBtzua7rbq6esD82OZeYi7eMRfvmIt3zMU75uIdc7mx+y2bNocTX11shN5ZPrY7dWiKinanDqOmQBfX3MtXa/6L0OEPQBdxL2ftcLpW7DOormnn2jpfEhw/cvA9u/+e1nP99/VBIiIiIupXTAYNgSYDmtocUBXX+GKtc/U+VYF7W1OB6KEBqG1ux4hBJnQ4BSW1TTDgf0tgGzUVBk25aXt9hQUyEREREfVYzNAAdw/yl9X1AFxTt6mdXcWRoRa0akEY7G/EYH8jAEDXBWGDzPA3aiiztyDIbED9tQ6MfXBQ3zzELbBAJiIiIqIeU1UFamdPcNKIYHxtb0FDawfGWYOhqq791Ve7n2MN9gcAhFj84HDq6NAFmsoeZCIiIiLyIUZNhaJ09iD3otg1aCoM/XiOBBbIRERERHTbwoLMCDL5VknpW09DRERERPdUkNmAILNvlZT9aqEQIiIiIqK+xgKZiIiIiMgDC2QiIiIiIg8skImIiIiIPLBAJiIiIiLywAKZiIiIiMgDC2QiIiIiIg8skImIiIiIPLBAJiIiIiLywAKZiIiIiMgDC2QiIiIiIg8skImIiIiIPLBAJiIiIiLywAKZiIiIiMgDC2QiIiIiIg+KiEhf38RAMHToUERERNzTNmtqajBs2LB72uZAwFy8Yy7eMRfvmIt3zMU75nJjzMa7/ppLeXk5amtrb3kcC+R+bMKECSgsLOzr2+h3mIt3zMU75uIdc/GOuXjHXG6M2Xg30HPhEAsiIiIiIg8skImIiIiIPGgbN27c2Nc3QTc2fvz4vr6Ffom5eMdcvGMu3jEX75iLd8zlxpiNdwM5F45BJiIiIiLywCEWREREREQeWCATEREREXlggdwLFRUVSE9PR1xcHOLj4/HGG28AAOrq6pCZmYmYmBhkZmbiypUrAIBz585h8uTJMJlM2LJlS5drbd26FfHx8Rg7diyWLFmC1tZWr23m5uYiJiYGMTExyM3Nde9vb2/HihUrMGrUKIwePRoffPCB1/PXr1+PkSNHIjAwsMv+b775Bunp6Rg3bhwSExNx6NAh5gLgwoULyMjIQGJiIqZPn47Kysr7JpeWlhbMmTMHo0ePRnx8PNatW+f+rK2tDYsWLYLNZkNaWhrKy8uZC4B//OMfSElJgcFgwP79+287E1/L5Xe/+x3i4uKQmJiIjIwMXLhwgbkA2LVrFxISEpCcnIypU6eiqKiIuXj44IMPoCjKHU0N5ku5vP/++xg2bBiSk5ORnJyM3//+98yl05/+9Cf3szzxxBO3nctNCfVYdXW1fP755yIi0tDQIDExMXLmzBlZs2aNbN68WURENm/eLC+99JKIiFy+fFkKCgrk5Zdflt/85jfu61RWVkpERIS0tLSIiMiCBQvkvffe69ae3W6XyMhIsdvtUldXJ5GRkVJXVyciIhs2bJD169eLiIjT6ZSamhqv9/yvf/1LqqurJSAgoMv+Z599Vt566y0RETlz5ow8/PDDt5mKb+WSlZUl77//voiIHD16VJ566qnbjWXA5dLc3CzHjh0TEZG2tjaZOnWqHDp0SEREdu7cKT/5yU9ERGTPnj2ycOFC5iIiZWVlcvr0aVm6dKns27fvtjPxtVyOHTsmzc3NIiLy1ltv8c9LZy719fXu4/Ly8mTWrFnMpVNDQ4NMmzZN0tLS5N///jdzEZH33ntPVq5cedtZePKlXIqLiyU5Odl9vcuXL99xPt6wB7kXwsLCkJKSAgAICgrCmDFjUFVVhby8PGRnZwMAsrOz8dFHHwEAhg8fjtTUVBiNxm7XcjgcuHbtGhwOB1paWjBixIhux+Tn5yMzMxMhISEYMmQIMjMzcfjwYQDAu+++i1/+8pcAAFVVMXToUK/3PGnSJISFhXXbrygKGhoaAAD19fVe2+8pX8qlqKgIM2bMAACkp6cjLy+vt3G4DbRcLBYL0tPTAQB+fn5ISUlx96B73nNWVhaOHj0Kuc33e30pl4iICCQmJkJV7/yvUl/KJT09HRaLBYDre+1OfhLjS7kMGjTIfVxzczMURWEunV555RWsXbsWZrP5tjMBfC+X74ov5fLOO+9g5cqVGDJkiPte7wYWyLepvLwcX3zxBdLS0nD58mV3sfXggw/i8uXLNz3XarVi9erVeOihhxAWFobg4GDMnDmz23FVVVUYOXKkezs8PBxVVVW4evUqANdfKCkpKViwYMEt27zexo0bsXv3boSHh+OHP/whduzY0avzb2Sg55KUlIQDBw4AAD788EM0NjbCbrf36hreDLRcrl69ij//+c/IyMjodm2DwYDg4GDmchf5Ui45OTmYPXv2LZ+5J3whl507dyI6OhovvfQStm/f3uNnv5mBnsupU6dQUVGBOXPm9Oq5b2Wg5wK4hp0kJiYiKysLFRUVPX72mxnouRQXF6O4uBhTpkzBpEmT3IX3d40F8m1oamrC/PnzsW3bti49AoCrZ/ZWvQJXrlxBXl4eysrKUF1djebmZuzevbvH7TscDlRWVuKRRx7BqVOnMHnyZKxevbpXz7Bnzx4sW7YMlZWVOHToEJYuXQpd13t1jev5Qi5btmzB8ePHMW7cOBw/fhxWqxWapvXqGtcbaLk4HA4sWbIEzz//PKKionrcTm8xF+98KZfdu3ejsLAQa9as6XH7N+IruaxcuRKlpaV4/fXX8etf/7rH7d/IQM9F13W8+OKL+O1vf9vjNntioOcCAI899hjKy8vxn//8B5mZme6e3jvhC7k4HA6UlJTgk08+wZ49e/Dss8+6C+/vEgvkXuro6MD8+fPx5JNPYt68eQCABx54ABcvXgQAXLx48Zbd/UeOHEFkZCSGDRsGo9GIefPm4cSJE/jss8/cg/EPHjwIq9Xa5X+MlZWVsFqtCA0NhcVicbe/YMECnDp1Ck6n033+hg0bbnoPOTk5WLhwIQBg8uTJaG1tRW1t7X2fy4gRI3DgwAF88cUX2LRpEwBg8ODB91UuK1asQExMDH7xi1+493le2+FwoL6+HqGhofd9Lt81X8rlyJEj2LRpEw4ePAiTycRcrrN48WL3j7Pv51waGxvx1VdfYfr06YiIiMDJkycxd+7cO3pRzxdyAYDQ0FD3984zzzyDzz///LYz8aVcwsPDMXfuXBiNRkRGRmLUqFEoKSm5o2y8uisjm32UruuydOlSeeGFF7rsX716dZdB7mvWrOny+auvvtplkPvJkyclLi5OmpubRdd1efrpp2X79u3d2rPb7RIRESF1dXVSV1cnERERYrfbRURk0aJFcvToURFxDeTPysq66b1f/zLaD37wA/fA+qKiIgkLCxNd13uQQne+lEtNTY04nU4REXn55ZfllVde6UkEXg3EXNavXy/z5s1zZ/CtN998s8tLegsWLOhNFF34Ui7fys7OvuOX9Hwpl1OnTklUVJQUFxf3MoXufCkXzzwOHjwo48eP72kM3fhSLp4effTRO3pJz5dyqa6udn994MABSUtL62kM3fhSLh9//LE8/fTTIuL6Nzs8PFxqa2t7E0ePsEDuhU8//VQASEJCgiQlJUlSUpL89a9/ldraWpkxY4bYbDbJyMhw/yG4ePGiWK1WCQoKkuDgYLFare63mDds2CCxsbESHx8vTz31lLS2tnptMycnR6KjoyU6Olreffdd9/7y8nKZNm2aJCQkyIwZM+TChQtez1+zZo1YrVZRFEWsVqu8+uqrIuKaueKRRx6RxMRESUpKkvz8fOYiIvv27RObzSYxMTGyfPnyG7bvi7lUVFQIABk9erT7ft955x0REbl27ZpkZWVJdHS0pKamSmlpKXMRkYKCArFarWKxWCQkJETi4uKYi4hkZGTI8OHD3fsfe+wx5iIizz//vMTFxUlSUpJMnz5dvvrqK+ZynTstkH0pl3Xr1klcXJwkJibK9OnT5ezZs8xFXMX+qlWrZMyYMTJ27FjZs2fPbedyM1xqmoiIiIjIA8cgExERERF5YIFMREREROSBBTIRERERkQcWyEREREREHlggExERERF5MPT1DRARkXd2u929vOqlS5egaRqGDRsGALBYLDhx4sRdabe8vBwnTpzAE088cVeuT0TU33GaNyKiAWDjxo0IDAzs9fLpt+OTTz7Bli1b8Je//OWut0VE1B9xiAUR0QAUGBgIwFXMPvroo3j88ccRFRWFdevW4Y9//CMmTpyIhIQElJaWAgBqamowf/58pKamIjU1Ff/85z8BAMePH3cv8Tpu3Dg0NjZi3bp1+PTTT5GcnIytW7eivLwc06ZNQ0pKClJSUtw91z1te9myZfjpT3+KCRMmYNSoUSy8iajf4xALIqIB7vTp0zh79ixCQkIQFRWFZ555BgUFBXjjjTewY8cObNu2DS+88AJWrVqFqVOn4ptvvsGsWbNw9uxZbNmyBTt37sSUKVPQ1NQEs9mM1157rUsPcktLC/72t7/BbDajpKQES5YsQWFhYY/bBlzDNgoKClBaWor09HScP38eZrO5zzIjIroZFshERANcamoqwsLCAADR0dGYOXMmACAhIQF///vfAQBHjhxBUVGR+5yGhgY0NTVhypQpePHFF/Hkk09i3rx5CA8P73b9jo4O/OxnP8OXX34JTdNQXFzcq7YBYOHChVBVFTExMYiKisK5c+eQnJz8HSdBRPTdYIFMRDTAmUwm99eqqrq3VVWFw+EAAOi6jpMnT3brtV23bh3mzJmDQ4cOYcqUKcjPz+92/a1bt+KBBx7A6dOnoet6l2v0pG0AUBSlyzWv3yYi6k84BpmI6D4wc+ZM7Nixw7395ZdfAgBKS0uRkJCAtWvXIjU1FefOnUNQUBAaGxvdx9bX1yMsLAyqquIPf/gDnE5nr9vft28fdF1HaWkpvv76a8TGxt75QxER3SUskImI7gPbt29HYWEhEhMTERcXh127dgEAtm3bhrFjxyIxMRFGoxGzZ89GYmIiNE1DUlIStm7diueeew65ublISkrCuXPnEBAQ0Ov2H3roIUycOBGzZ8/Grl27OP6YiPo1TvNGRER31bJly/CjH/0IWVlZfX0rREQ9wh5kIiIiIiIP7EEmIiIiIvLAHmQiIiIiIg8skImIiIiIPLBAJiIiIiLywAKZiIiIiMgDC2QiIiIiIg//DzVFF59wDrfMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "m.plot(forecast,xlabel=\"Timestamp\",ylabel=\"Value\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}