02 Algorithm spot check.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ML algorithm spot check \n",
    "\n",
    "Table of Contents\n",
    "1. [Read in cleaned data](#cleaned)\n",
    "1. [Spot check XGBoost](#xgboost)\n",
    "1. [Spot check Multilayer Perceptron](#mlp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read in the cleansed data <a name=\"cleaned\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "prepared_data_dir = 'data/processed'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "processed = pd.read_pickle(\"{}/cooked_v3.pkl\".format (prepared_data_dir))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 1502293 entries, 2017-07-03 08:00:00 to 2017-07-31 20:00:00\n",
      "Data columns (total 14 columns):\n",
      "StartPrice        1502293 non-null float64\n",
      "MaxPrice          1502293 non-null float64\n",
      "MinPrice          1502293 non-null float64\n",
      "EndPrice          1502293 non-null float64\n",
      "TradedVolume      1502293 non-null float64\n",
      "NumberOfTrades    1502293 non-null float64\n",
      "HourOfDay         1502293 non-null int64\n",
      "MinOfHour         1502293 non-null int64\n",
      "MinOfDay          1502293 non-null int64\n",
      "DayOfWeek         1502293 non-null int64\n",
      "DayOfYear         1502293 non-null int64\n",
      "MonthOfYear       1502293 non-null int64\n",
      "WeekOfYear        1502293 non-null int64\n",
      "Mnemonic          1502293 non-null category\n",
      "dtypes: category(1), float64(6), int64(7)\n",
      "memory usage: 161.9 MB\n"
     ]
    }
   ],
   "source": [
    "processed.info ()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>StartPrice</th>\n",
       "      <th>MaxPrice</th>\n",
       "      <th>MinPrice</th>\n",
       "      <th>EndPrice</th>\n",
       "      <th>TradedVolume</th>\n",
       "      <th>NumberOfTrades</th>\n",
       "      <th>HourOfDay</th>\n",
       "      <th>MinOfHour</th>\n",
       "      <th>MinOfDay</th>\n",
       "      <th>DayOfWeek</th>\n",
       "      <th>DayOfYear</th>\n",
       "      <th>MonthOfYear</th>\n",
       "      <th>WeekOfYear</th>\n",
       "      <th>Mnemonic</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CalcDateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-07-03 13:39:00</th>\n",
       "      <td>34.535</td>\n",
       "      <td>34.535</td>\n",
       "      <td>34.535</td>\n",
       "      <td>34.535</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>13</td>\n",
       "      <td>39</td>\n",
       "      <td>819</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>DEQ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-13 17:08:00</th>\n",
       "      <td>71.900</td>\n",
       "      <td>71.900</td>\n",
       "      <td>71.890</td>\n",
       "      <td>71.900</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>17</td>\n",
       "      <td>8</td>\n",
       "      <td>1028</td>\n",
       "      <td>3</td>\n",
       "      <td>194</td>\n",
       "      <td>7</td>\n",
       "      <td>28</td>\n",
       "      <td>OSR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-06 14:22:00</th>\n",
       "      <td>1.059</td>\n",
       "      <td>1.059</td>\n",
       "      <td>1.059</td>\n",
       "      <td>1.059</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>14</td>\n",
       "      <td>22</td>\n",
       "      <td>862</td>\n",
       "      <td>3</td>\n",
       "      <td>187</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>TINA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-10 09:51:00</th>\n",
       "      <td>5.488</td>\n",
       "      <td>5.490</td>\n",
       "      <td>5.488</td>\n",
       "      <td>5.490</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9</td>\n",
       "      <td>51</td>\n",
       "      <td>591</td>\n",
       "      <td>0</td>\n",
       "      <td>191</td>\n",
       "      <td>7</td>\n",
       "      <td>28</td>\n",
       "      <td>NOA3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-21 08:35:00</th>\n",
       "      <td>4.541</td>\n",
       "      <td>4.541</td>\n",
       "      <td>4.541</td>\n",
       "      <td>4.541</td>\n",
       "      <td>373.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8</td>\n",
       "      <td>35</td>\n",
       "      <td>515</td>\n",
       "      <td>4</td>\n",
       "      <td>202</td>\n",
       "      <td>7</td>\n",
       "      <td>29</td>\n",
       "      <td>SNH</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     StartPrice  MaxPrice  MinPrice  EndPrice  TradedVolume  \\\n",
       "CalcDateTime                                                                  \n",
       "2017-07-03 13:39:00      34.535    34.535    34.535    34.535         100.0   \n",
       "2017-07-13 17:08:00      71.900    71.900    71.890    71.900           0.0   \n",
       "2017-07-06 14:22:00       1.059     1.059     1.059     1.059           0.0   \n",
       "2017-07-10 09:51:00       5.488     5.490     5.488     5.490           0.0   \n",
       "2017-07-21 08:35:00       4.541     4.541     4.541     4.541         373.0   \n",
       "\n",
       "                     NumberOfTrades  HourOfDay  MinOfHour  MinOfDay  \\\n",
       "CalcDateTime                                                          \n",
       "2017-07-03 13:39:00             1.0         13         39       819   \n",
       "2017-07-13 17:08:00             0.0         17          8      1028   \n",
       "2017-07-06 14:22:00             0.0         14         22       862   \n",
       "2017-07-10 09:51:00             0.0          9         51       591   \n",
       "2017-07-21 08:35:00             1.0          8         35       515   \n",
       "\n",
       "                     DayOfWeek  DayOfYear  MonthOfYear  WeekOfYear Mnemonic  \n",
       "CalcDateTime                                                                 \n",
       "2017-07-03 13:39:00          0        184            7          27      DEQ  \n",
       "2017-07-13 17:08:00          3        194            7          28      OSR  \n",
       "2017-07-06 14:22:00          3        187            7          27     TINA  \n",
       "2017-07-10 09:51:00          0        191            7          28     NOA3  \n",
       "2017-07-21 08:35:00          4        202            7          29      SNH  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "processed.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>StartPrice</th>\n",
       "      <th>MaxPrice</th>\n",
       "      <th>MinPrice</th>\n",
       "      <th>EndPrice</th>\n",
       "      <th>TradedVolume</th>\n",
       "      <th>NumberOfTrades</th>\n",
       "      <th>HourOfDay</th>\n",
       "      <th>MinOfHour</th>\n",
       "      <th>MinOfDay</th>\n",
       "      <th>DayOfWeek</th>\n",
       "      <th>DayOfYear</th>\n",
       "      <th>MonthOfYear</th>\n",
       "      <th>WeekOfYear</th>\n",
       "      <th>Mnemonic</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CalcDateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:00:00</th>\n",
       "      <td>15.830</td>\n",
       "      <td>15.830</td>\n",
       "      <td>15.820</td>\n",
       "      <td>15.825</td>\n",
       "      <td>6594.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>480</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>DBK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:01:00</th>\n",
       "      <td>15.835</td>\n",
       "      <td>15.835</td>\n",
       "      <td>15.795</td>\n",
       "      <td>15.800</td>\n",
       "      <td>120179.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>481</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>DBK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:02:00</th>\n",
       "      <td>15.800</td>\n",
       "      <td>15.805</td>\n",
       "      <td>15.775</td>\n",
       "      <td>15.785</td>\n",
       "      <td>32404.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>482</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>DBK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:03:00</th>\n",
       "      <td>15.785</td>\n",
       "      <td>15.815</td>\n",
       "      <td>15.785</td>\n",
       "      <td>15.805</td>\n",
       "      <td>34169.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>483</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>DBK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:04:00</th>\n",
       "      <td>15.805</td>\n",
       "      <td>15.805</td>\n",
       "      <td>15.795</td>\n",
       "      <td>15.805</td>\n",
       "      <td>26589.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>484</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>DBK</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     StartPrice  MaxPrice  MinPrice  EndPrice  TradedVolume  \\\n",
       "CalcDateTime                                                                  \n",
       "2017-07-03 08:00:00      15.830    15.830    15.820    15.825        6594.0   \n",
       "2017-07-03 08:01:00      15.835    15.835    15.795    15.800      120179.0   \n",
       "2017-07-03 08:02:00      15.800    15.805    15.775    15.785       32404.0   \n",
       "2017-07-03 08:03:00      15.785    15.815    15.785    15.805       34169.0   \n",
       "2017-07-03 08:04:00      15.805    15.805    15.795    15.805       26589.0   \n",
       "\n",
       "                     NumberOfTrades  HourOfDay  MinOfHour  MinOfDay  \\\n",
       "CalcDateTime                                                          \n",
       "2017-07-03 08:00:00             9.0          8          0       480   \n",
       "2017-07-03 08:01:00            29.0          8          1       481   \n",
       "2017-07-03 08:02:00            29.0          8          2       482   \n",
       "2017-07-03 08:03:00            58.0          8          3       483   \n",
       "2017-07-03 08:04:00            31.0          8          4       484   \n",
       "\n",
       "                     DayOfWeek  DayOfYear  MonthOfYear  WeekOfYear Mnemonic  \n",
       "CalcDateTime                                                                 \n",
       "2017-07-03 08:00:00          0        184            7          27      DBK  \n",
       "2017-07-03 08:01:00          0        184            7          27      DBK  \n",
       "2017-07-03 08:02:00          0        184            7          27      DBK  \n",
       "2017-07-03 08:03:00          0        184            7          27      DBK  \n",
       "2017-07-03 08:04:00          0        184            7          27      DBK  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "processed.head (5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected = processed[processed['Mnemonic'].isin(['EOAN', 'BMW', 'SAP'])]\n",
    "selected.Mnemonic.cat.remove_unused_categories (inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>StartPrice</th>\n",
       "      <th>MaxPrice</th>\n",
       "      <th>MinPrice</th>\n",
       "      <th>EndPrice</th>\n",
       "      <th>TradedVolume</th>\n",
       "      <th>NumberOfTrades</th>\n",
       "      <th>HourOfDay</th>\n",
       "      <th>MinOfHour</th>\n",
       "      <th>MinOfDay</th>\n",
       "      <th>DayOfWeek</th>\n",
       "      <th>DayOfYear</th>\n",
       "      <th>MonthOfYear</th>\n",
       "      <th>WeekOfYear</th>\n",
       "      <th>Mnemonic</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CalcDateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:00:00</th>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>480</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:01:00</th>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>264.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>481</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:02:00</th>\n",
       "      <td>8.300</td>\n",
       "      <td>8.300</td>\n",
       "      <td>8.284</td>\n",
       "      <td>8.285</td>\n",
       "      <td>5180.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>482</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:03:00</th>\n",
       "      <td>8.286</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.282</td>\n",
       "      <td>8.285</td>\n",
       "      <td>20331.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>483</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:04:00</th>\n",
       "      <td>8.282</td>\n",
       "      <td>8.285</td>\n",
       "      <td>8.279</td>\n",
       "      <td>8.282</td>\n",
       "      <td>11333.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>484</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     StartPrice  MaxPrice  MinPrice  EndPrice  TradedVolume  \\\n",
       "CalcDateTime                                                                  \n",
       "2017-07-03 08:00:00       8.293     8.301     8.293     8.298       13532.0   \n",
       "2017-07-03 08:01:00       8.297     8.297     8.297     8.297         264.0   \n",
       "2017-07-03 08:02:00       8.300     8.300     8.284     8.285        5180.0   \n",
       "2017-07-03 08:03:00       8.286     8.293     8.282     8.285       20331.0   \n",
       "2017-07-03 08:04:00       8.282     8.285     8.279     8.282       11333.0   \n",
       "\n",
       "                     NumberOfTrades  HourOfDay  MinOfHour  MinOfDay  \\\n",
       "CalcDateTime                                                          \n",
       "2017-07-03 08:00:00            12.0          8          0       480   \n",
       "2017-07-03 08:01:00             1.0          8          1       481   \n",
       "2017-07-03 08:02:00             9.0          8          2       482   \n",
       "2017-07-03 08:03:00            29.0          8          3       483   \n",
       "2017-07-03 08:04:00            27.0          8          4       484   \n",
       "\n",
       "                     DayOfWeek  DayOfYear  MonthOfYear  WeekOfYear Mnemonic  \n",
       "CalcDateTime                                                                 \n",
       "2017-07-03 08:00:00          0        184            7          27     EOAN  \n",
       "2017-07-03 08:01:00          0        184            7          27     EOAN  \n",
       "2017-07-03 08:02:00          0        184            7          27     EOAN  \n",
       "2017-07-03 08:03:00          0        184            7          27     EOAN  \n",
       "2017-07-03 08:04:00          0        184            7          27     EOAN  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selected.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Forecast with XGBoost <a name=\"xgboost\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting xgboost\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/54/21/8b2ec99862903a6d3aed62ce156d21d114b8666e669c46d9e54041df9496/xgboost-0.81-py2.py3-none-manylinux1_x86_64.whl (16.6MB)\n",
      "\u001b[K    100% |████████████████████████████████| 16.6MB 2.9MB/s \n",
      "\u001b[?25hRequirement already satisfied: scipy in /home/ec2-user/anaconda3/envs/mxnet_p36/lib/python3.6/site-packages (from xgboost) (1.1.0)\n",
      "Requirement already satisfied: numpy in /home/ec2-user/anaconda3/envs/mxnet_p36/lib/python3.6/site-packages (from xgboost) (1.14.6)\n",
      "Installing collected packages: xgboost\n",
      "Successfully installed xgboost-0.81\n",
      "\u001b[33mYou are using pip version 10.0.1, however version 19.0.3 is available.\n",
      "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "!pip install xgboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_xgb_target (df):\n",
    "    return df.MaxPrice.shift(-1).fillna (method='ffill')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_xgb_features (df, horizon):\n",
    "    n_df = df.copy ()\n",
    "    \n",
    "    for offset in range(1, horizon+1):\n",
    "        min_price = n_df['MinPrice'].shift (offset).fillna(method='bfill')\n",
    "        max_price = n_df['MaxPrice'].shift (offset).fillna(method='bfill')\n",
    "        start_price = n_df['StartPrice'].shift (offset).fillna(method='bfill')\n",
    "        end_price = n_df['EndPrice'].shift (offset).fillna(method='bfill')\n",
    "        trade_vol = n_df['TradedVolume'].shift (offset).fillna(method='bfill')\n",
    "        num_trades = n_df['NumberOfTrades'].shift (offset).fillna(method='bfill')\n",
    "        \n",
    "        n_df[\"h{}_MinPrice\".format (offset)] = min_price\n",
    "        n_df[\"h{}_MaxPrice\".format (offset)] = max_price\n",
    "        n_df[\"h{}_StartPrice\".format (offset)] = start_price\n",
    "        n_df[\"h{}_EndPrice\".format (offset)] = end_price\n",
    "        n_df[\"h{}_TradeVolume\".format (offset)] = trade_vol\n",
    "        n_df[\"h{}_NumberOfTrades\".format (offset)] = num_trades\n",
    "        \n",
    "    return n_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>StartPrice</th>\n",
       "      <th>MaxPrice</th>\n",
       "      <th>MinPrice</th>\n",
       "      <th>EndPrice</th>\n",
       "      <th>TradedVolume</th>\n",
       "      <th>NumberOfTrades</th>\n",
       "      <th>HourOfDay</th>\n",
       "      <th>MinOfHour</th>\n",
       "      <th>MinOfDay</th>\n",
       "      <th>DayOfWeek</th>\n",
       "      <th>DayOfYear</th>\n",
       "      <th>MonthOfYear</th>\n",
       "      <th>WeekOfYear</th>\n",
       "      <th>Mnemonic</th>\n",
       "      <th>h1_MinPrice</th>\n",
       "      <th>h1_MaxPrice</th>\n",
       "      <th>h1_StartPrice</th>\n",
       "      <th>h1_EndPrice</th>\n",
       "      <th>h1_TradeVolume</th>\n",
       "      <th>h1_NumberOfTrades</th>\n",
       "      <th>h2_MinPrice</th>\n",
       "      <th>h2_MaxPrice</th>\n",
       "      <th>h2_StartPrice</th>\n",
       "      <th>h2_EndPrice</th>\n",
       "      <th>h2_TradeVolume</th>\n",
       "      <th>h2_NumberOfTrades</th>\n",
       "      <th>h3_MinPrice</th>\n",
       "      <th>h3_MaxPrice</th>\n",
       "      <th>h3_StartPrice</th>\n",
       "      <th>h3_EndPrice</th>\n",
       "      <th>h3_TradeVolume</th>\n",
       "      <th>h3_NumberOfTrades</th>\n",
       "      <th>h4_MinPrice</th>\n",
       "      <th>h4_MaxPrice</th>\n",
       "      <th>h4_StartPrice</th>\n",
       "      <th>h4_EndPrice</th>\n",
       "      <th>h4_TradeVolume</th>\n",
       "      <th>h4_NumberOfTrades</th>\n",
       "      <th>h5_MinPrice</th>\n",
       "      <th>h5_MaxPrice</th>\n",
       "      <th>h5_StartPrice</th>\n",
       "      <th>h5_EndPrice</th>\n",
       "      <th>h5_TradeVolume</th>\n",
       "      <th>h5_NumberOfTrades</th>\n",
       "      <th>NextMaxPrice</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CalcDateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:00:00</th>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>480</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:01:00</th>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>264.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>481</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:02:00</th>\n",
       "      <td>8.300</td>\n",
       "      <td>8.300</td>\n",
       "      <td>8.284</td>\n",
       "      <td>8.285</td>\n",
       "      <td>5180.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>482</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>264.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:03:00</th>\n",
       "      <td>8.286</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.282</td>\n",
       "      <td>8.285</td>\n",
       "      <td>20331.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>483</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "      <td>8.284</td>\n",
       "      <td>8.300</td>\n",
       "      <td>8.300</td>\n",
       "      <td>8.285</td>\n",
       "      <td>5180.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>264.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.285</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-03 08:04:00</th>\n",
       "      <td>8.282</td>\n",
       "      <td>8.285</td>\n",
       "      <td>8.279</td>\n",
       "      <td>8.282</td>\n",
       "      <td>11333.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>484</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>EOAN</td>\n",
       "      <td>8.282</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.286</td>\n",
       "      <td>8.285</td>\n",
       "      <td>20331.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>8.284</td>\n",
       "      <td>8.300</td>\n",
       "      <td>8.300</td>\n",
       "      <td>8.285</td>\n",
       "      <td>5180.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>8.297</td>\n",
       "      <td>264.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.301</td>\n",
       "      <td>8.293</td>\n",
       "      <td>8.298</td>\n",
       "      <td>13532.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>8.283</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     StartPrice  MaxPrice  MinPrice  EndPrice  TradedVolume  \\\n",
       "CalcDateTime                                                                  \n",
       "2017-07-03 08:00:00       8.293     8.301     8.293     8.298       13532.0   \n",
       "2017-07-03 08:01:00       8.297     8.297     8.297     8.297         264.0   \n",
       "2017-07-03 08:02:00       8.300     8.300     8.284     8.285        5180.0   \n",
       "2017-07-03 08:03:00       8.286     8.293     8.282     8.285       20331.0   \n",
       "2017-07-03 08:04:00       8.282     8.285     8.279     8.282       11333.0   \n",
       "\n",
       "                     NumberOfTrades  HourOfDay  MinOfHour  MinOfDay  \\\n",
       "CalcDateTime                                                          \n",
       "2017-07-03 08:00:00            12.0          8          0       480   \n",
       "2017-07-03 08:01:00             1.0          8          1       481   \n",
       "2017-07-03 08:02:00             9.0          8          2       482   \n",
       "2017-07-03 08:03:00            29.0          8          3       483   \n",
       "2017-07-03 08:04:00            27.0          8          4       484   \n",
       "\n",
       "                     DayOfWeek  DayOfYear  MonthOfYear  WeekOfYear Mnemonic  \\\n",
       "CalcDateTime                                                                  \n",
       "2017-07-03 08:00:00          0        184            7          27     EOAN   \n",
       "2017-07-03 08:01:00          0        184            7          27     EOAN   \n",
       "2017-07-03 08:02:00          0        184            7          27     EOAN   \n",
       "2017-07-03 08:03:00          0        184            7          27     EOAN   \n",
       "2017-07-03 08:04:00          0        184            7          27     EOAN   \n",
       "\n",
       "                     h1_MinPrice  h1_MaxPrice  h1_StartPrice  h1_EndPrice  \\\n",
       "CalcDateTime                                                                \n",
       "2017-07-03 08:00:00        8.293        8.301          8.293        8.298   \n",
       "2017-07-03 08:01:00        8.293        8.301          8.293        8.298   \n",
       "2017-07-03 08:02:00        8.297        8.297          8.297        8.297   \n",
       "2017-07-03 08:03:00        8.284        8.300          8.300        8.285   \n",
       "2017-07-03 08:04:00        8.282        8.293          8.286        8.285   \n",
       "\n",
       "                     h1_TradeVolume  h1_NumberOfTrades  h2_MinPrice  \\\n",
       "CalcDateTime                                                          \n",
       "2017-07-03 08:00:00         13532.0               12.0        8.293   \n",
       "2017-07-03 08:01:00         13532.0               12.0        8.293   \n",
       "2017-07-03 08:02:00           264.0                1.0        8.293   \n",
       "2017-07-03 08:03:00          5180.0                9.0        8.297   \n",
       "2017-07-03 08:04:00         20331.0               29.0        8.284   \n",
       "\n",
       "                     h2_MaxPrice  h2_StartPrice  h2_EndPrice  h2_TradeVolume  \\\n",
       "CalcDateTime                                                                   \n",
       "2017-07-03 08:00:00        8.301          8.293        8.298         13532.0   \n",
       "2017-07-03 08:01:00        8.301          8.293        8.298         13532.0   \n",
       "2017-07-03 08:02:00        8.301          8.293        8.298         13532.0   \n",
       "2017-07-03 08:03:00        8.297          8.297        8.297           264.0   \n",
       "2017-07-03 08:04:00        8.300          8.300        8.285          5180.0   \n",
       "\n",
       "                     h2_NumberOfTrades  h3_MinPrice  h3_MaxPrice  \\\n",
       "CalcDateTime                                                       \n",
       "2017-07-03 08:00:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:01:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:02:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:03:00                1.0        8.293        8.301   \n",
       "2017-07-03 08:04:00                9.0        8.297        8.297   \n",
       "\n",
       "                     h3_StartPrice  h3_EndPrice  h3_TradeVolume  \\\n",
       "CalcDateTime                                                      \n",
       "2017-07-03 08:00:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:01:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:02:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:03:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:04:00          8.297        8.297           264.0   \n",
       "\n",
       "                     h3_NumberOfTrades  h4_MinPrice  h4_MaxPrice  \\\n",
       "CalcDateTime                                                       \n",
       "2017-07-03 08:00:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:01:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:02:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:03:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:04:00                1.0        8.293        8.301   \n",
       "\n",
       "                     h4_StartPrice  h4_EndPrice  h4_TradeVolume  \\\n",
       "CalcDateTime                                                      \n",
       "2017-07-03 08:00:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:01:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:02:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:03:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:04:00          8.293        8.298         13532.0   \n",
       "\n",
       "                     h4_NumberOfTrades  h5_MinPrice  h5_MaxPrice  \\\n",
       "CalcDateTime                                                       \n",
       "2017-07-03 08:00:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:01:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:02:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:03:00               12.0        8.293        8.301   \n",
       "2017-07-03 08:04:00               12.0        8.293        8.301   \n",
       "\n",
       "                     h5_StartPrice  h5_EndPrice  h5_TradeVolume  \\\n",
       "CalcDateTime                                                      \n",
       "2017-07-03 08:00:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:01:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:02:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:03:00          8.293        8.298         13532.0   \n",
       "2017-07-03 08:04:00          8.293        8.298         13532.0   \n",
       "\n",
       "                     h5_NumberOfTrades  NextMaxPrice  \n",
       "CalcDateTime                                          \n",
       "2017-07-03 08:00:00               12.0         8.297  \n",
       "2017-07-03 08:01:00               12.0         8.300  \n",
       "2017-07-03 08:02:00               12.0         8.293  \n",
       "2017-07-03 08:03:00               12.0         8.285  \n",
       "2017-07-03 08:04:00               12.0         8.283  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb_data = create_xgb_features (selected, 5)\n",
    "xgb_data['NextMaxPrice'] = create_xgb_target (xgb_data)\n",
    "pd.options.display.max_columns=150\n",
    "xgb_data.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/anaconda3/envs/mxnet_p36/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2069: FutureWarning: From version 0.21, test_size will always complement train_size unless both are specified.\n",
      "  FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "train_data, test_data = train_test_split (xgb_data, train_size=0.8, shuffle=False)\n",
    "train_data, validate_data = train_test_split (train_data, train_size=0.8, test_size=0.2, shuffle=True)\n",
    "\n",
    "train_labels = train_data.NextMaxPrice\n",
    "validate_labels = validate_data.NextMaxPrice\n",
    "test_labels = test_data.NextMaxPrice"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/anaconda3/envs/mxnet_p36/lib/python3.6/site-packages/ipykernel/__main__.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  app.launch_new_instance()\n"
     ]
    }
   ],
   "source": [
    "train_data.drop (\"NextMaxPrice\", axis=1, inplace=True)\n",
    "validate_data.drop (\"NextMaxPrice\", axis=1, inplace=True)\n",
    "test_data.drop ('NextMaxPrice', axis = 1, inplace=True)"
   ]
  },
  {
   "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>StartPrice</th>\n",
       "      <th>MaxPrice</th>\n",
       "      <th>MinPrice</th>\n",
       "      <th>EndPrice</th>\n",
       "      <th>TradedVolume</th>\n",
       "      <th>NumberOfTrades</th>\n",
       "      <th>HourOfDay</th>\n",
       "      <th>MinOfHour</th>\n",
       "      <th>MinOfDay</th>\n",
       "      <th>DayOfWeek</th>\n",
       "      <th>DayOfYear</th>\n",
       "      <th>MonthOfYear</th>\n",
       "      <th>WeekOfYear</th>\n",
       "      <th>h1_MinPrice</th>\n",
       "      <th>h1_MaxPrice</th>\n",
       "      <th>h1_StartPrice</th>\n",
       "      <th>h1_EndPrice</th>\n",
       "      <th>h1_TradeVolume</th>\n",
       "      <th>h1_NumberOfTrades</th>\n",
       "      <th>h2_MinPrice</th>\n",
       "      <th>h2_MaxPrice</th>\n",
       "      <th>h2_StartPrice</th>\n",
       "      <th>h2_EndPrice</th>\n",
       "      <th>h2_TradeVolume</th>\n",
       "      <th>h2_NumberOfTrades</th>\n",
       "      <th>h3_MinPrice</th>\n",
       "      <th>h3_MaxPrice</th>\n",
       "      <th>h3_StartPrice</th>\n",
       "      <th>h3_EndPrice</th>\n",
       "      <th>h3_TradeVolume</th>\n",
       "      <th>h3_NumberOfTrades</th>\n",
       "      <th>h4_MinPrice</th>\n",
       "      <th>h4_MaxPrice</th>\n",
       "      <th>h4_StartPrice</th>\n",
       "      <th>h4_EndPrice</th>\n",
       "      <th>h4_TradeVolume</th>\n",
       "      <th>h4_NumberOfTrades</th>\n",
       "      <th>h5_MinPrice</th>\n",
       "      <th>h5_MaxPrice</th>\n",
       "      <th>h5_StartPrice</th>\n",
       "      <th>h5_EndPrice</th>\n",
       "      <th>h5_TradeVolume</th>\n",
       "      <th>h5_NumberOfTrades</th>\n",
       "      <th>Mnemonic_BMW</th>\n",
       "      <th>Mnemonic_EOAN</th>\n",
       "      <th>Mnemonic_SAP</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CalcDateTime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-07-19 10:52:00</th>\n",
       "      <td>8.483</td>\n",
       "      <td>8.485</td>\n",
       "      <td>8.483</td>\n",
       "      <td>8.484</td>\n",
       "      <td>10544.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10</td>\n",
       "      <td>52</td>\n",
       "      <td>652</td>\n",
       "      <td>2</td>\n",
       "      <td>200</td>\n",
       "      <td>7</td>\n",
       "      <td>29</td>\n",
       "      <td>8.481</td>\n",
       "      <td>8.483</td>\n",
       "      <td>8.483</td>\n",
       "      <td>8.482</td>\n",
       "      <td>6805.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>8.484</td>\n",
       "      <td>8.485</td>\n",
       "      <td>8.485</td>\n",
       "      <td>8.484</td>\n",
       "      <td>22069.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>8.483</td>\n",
       "      <td>8.486</td>\n",
       "      <td>8.483</td>\n",
       "      <td>8.485</td>\n",
       "      <td>22961.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>8.484</td>\n",
       "      <td>8.488</td>\n",
       "      <td>8.488</td>\n",
       "      <td>8.484</td>\n",
       "      <td>8802.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>8.488</td>\n",
       "      <td>8.491</td>\n",
       "      <td>8.488</td>\n",
       "      <td>8.490</td>\n",
       "      <td>15191.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-17 19:07:00</th>\n",
       "      <td>91.010</td>\n",
       "      <td>91.060</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.050</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>19</td>\n",
       "      <td>7</td>\n",
       "      <td>1147</td>\n",
       "      <td>0</td>\n",
       "      <td>198</td>\n",
       "      <td>7</td>\n",
       "      <td>29</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.060</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.050</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.060</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.050</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.060</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.050</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.060</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.050</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.060</td>\n",
       "      <td>91.010</td>\n",
       "      <td>91.050</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-18 09:50:00</th>\n",
       "      <td>8.537</td>\n",
       "      <td>8.537</td>\n",
       "      <td>8.537</td>\n",
       "      <td>8.537</td>\n",
       "      <td>996.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>9</td>\n",
       "      <td>50</td>\n",
       "      <td>590</td>\n",
       "      <td>1</td>\n",
       "      <td>199</td>\n",
       "      <td>7</td>\n",
       "      <td>29</td>\n",
       "      <td>8.533</td>\n",
       "      <td>8.535</td>\n",
       "      <td>8.533</td>\n",
       "      <td>8.535</td>\n",
       "      <td>4392.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>8.534</td>\n",
       "      <td>8.534</td>\n",
       "      <td>8.534</td>\n",
       "      <td>8.534</td>\n",
       "      <td>2550.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>8.535</td>\n",
       "      <td>8.537</td>\n",
       "      <td>8.537</td>\n",
       "      <td>8.535</td>\n",
       "      <td>7720.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>8.538</td>\n",
       "      <td>8.542</td>\n",
       "      <td>8.542</td>\n",
       "      <td>8.538</td>\n",
       "      <td>5890.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>8.542</td>\n",
       "      <td>8.545</td>\n",
       "      <td>8.545</td>\n",
       "      <td>8.543</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-05 15:02:00</th>\n",
       "      <td>81.390</td>\n",
       "      <td>81.400</td>\n",
       "      <td>81.370</td>\n",
       "      <td>81.400</td>\n",
       "      <td>1910.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>902</td>\n",
       "      <td>2</td>\n",
       "      <td>186</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>81.300</td>\n",
       "      <td>81.380</td>\n",
       "      <td>81.300</td>\n",
       "      <td>81.370</td>\n",
       "      <td>3509.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>81.280</td>\n",
       "      <td>81.310</td>\n",
       "      <td>81.280</td>\n",
       "      <td>81.310</td>\n",
       "      <td>2622.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>81.270</td>\n",
       "      <td>81.320</td>\n",
       "      <td>81.310</td>\n",
       "      <td>81.280</td>\n",
       "      <td>2579.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>81.290</td>\n",
       "      <td>81.310</td>\n",
       "      <td>81.290</td>\n",
       "      <td>81.310</td>\n",
       "      <td>2675.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>81.290</td>\n",
       "      <td>81.310</td>\n",
       "      <td>81.310</td>\n",
       "      <td>81.290</td>\n",
       "      <td>1684.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-07-06 17:55:00</th>\n",
       "      <td>8.088</td>\n",
       "      <td>8.089</td>\n",
       "      <td>8.074</td>\n",
       "      <td>8.080</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>17</td>\n",
       "      <td>55</td>\n",
       "      <td>1075</td>\n",
       "      <td>3</td>\n",
       "      <td>187</td>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>8.074</td>\n",
       "      <td>8.089</td>\n",
       "      <td>8.088</td>\n",
       "      <td>8.080</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.074</td>\n",
       "      <td>8.089</td>\n",
       "      <td>8.088</td>\n",
       "      <td>8.080</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.074</td>\n",
       "      <td>8.089</td>\n",
       "      <td>8.088</td>\n",
       "      <td>8.080</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.074</td>\n",
       "      <td>8.089</td>\n",
       "      <td>8.088</td>\n",
       "      <td>8.080</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.074</td>\n",
       "      <td>8.089</td>\n",
       "      <td>8.088</td>\n",
       "      <td>8.080</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     StartPrice  MaxPrice  MinPrice  EndPrice  TradedVolume  \\\n",
       "CalcDateTime                                                                  \n",
       "2017-07-19 10:52:00       8.483     8.485     8.483     8.484       10544.0   \n",
       "2017-07-17 19:07:00      91.010    91.060    91.010    91.050           0.0   \n",
       "2017-07-18 09:50:00       8.537     8.537     8.537     8.537         996.0   \n",
       "2017-07-05 15:02:00      81.390    81.400    81.370    81.400        1910.0   \n",
       "2017-07-06 17:55:00       8.088     8.089     8.074     8.080           0.0   \n",
       "\n",
       "                     NumberOfTrades  HourOfDay  MinOfHour  MinOfDay  \\\n",
       "CalcDateTime                                                          \n",
       "2017-07-19 10:52:00            10.0         10         52       652   \n",
       "2017-07-17 19:07:00             0.0         19          7      1147   \n",
       "2017-07-18 09:50:00             1.0          9         50       590   \n",
       "2017-07-05 15:02:00            14.0         15          2       902   \n",
       "2017-07-06 17:55:00             0.0         17         55      1075   \n",
       "\n",
       "                     DayOfWeek  DayOfYear  MonthOfYear  WeekOfYear  \\\n",
       "CalcDateTime                                                         \n",
       "2017-07-19 10:52:00          2        200            7          29   \n",
       "2017-07-17 19:07:00          0        198            7          29   \n",
       "2017-07-18 09:50:00          1        199            7          29   \n",
       "2017-07-05 15:02:00          2        186            7          27   \n",
       "2017-07-06 17:55:00          3        187            7          27   \n",
       "\n",
       "                     h1_MinPrice  h1_MaxPrice  h1_StartPrice  h1_EndPrice  \\\n",
       "CalcDateTime                                                                \n",
       "2017-07-19 10:52:00        8.481        8.483          8.483        8.482   \n",
       "2017-07-17 19:07:00       91.010       91.060         91.010       91.050   \n",
       "2017-07-18 09:50:00        8.533        8.535          8.533        8.535   \n",
       "2017-07-05 15:02:00       81.300       81.380         81.300       81.370   \n",
       "2017-07-06 17:55:00        8.074        8.089          8.088        8.080   \n",
       "\n",
       "                     h1_TradeVolume  h1_NumberOfTrades  h2_MinPrice  \\\n",
       "CalcDateTime                                                          \n",
       "2017-07-19 10:52:00          6805.0                4.0        8.484   \n",
       "2017-07-17 19:07:00             0.0                0.0       91.010   \n",
       "2017-07-18 09:50:00          4392.0                7.0        8.534   \n",
       "2017-07-05 15:02:00          3509.0               28.0       81.280   \n",
       "2017-07-06 17:55:00             0.0                0.0        8.074   \n",
       "\n",
       "                     h2_MaxPrice  h2_StartPrice  h2_EndPrice  h2_TradeVolume  \\\n",
       "CalcDateTime                                                                   \n",
       "2017-07-19 10:52:00        8.485          8.485        8.484         22069.0   \n",
       "2017-07-17 19:07:00       91.060         91.010       91.050             0.0   \n",
       "2017-07-18 09:50:00        8.534          8.534        8.534          2550.0   \n",
       "2017-07-05 15:02:00       81.310         81.280       81.310          2622.0   \n",
       "2017-07-06 17:55:00        8.089          8.088        8.080             0.0   \n",
       "\n",
       "                     h2_NumberOfTrades  h3_MinPrice  h3_MaxPrice  \\\n",
       "CalcDateTime                                                       \n",
       "2017-07-19 10:52:00               21.0        8.483        8.486   \n",
       "2017-07-17 19:07:00                0.0       91.010       91.060   \n",
       "2017-07-18 09:50:00                4.0        8.535        8.537   \n",
       "2017-07-05 15:02:00               15.0       81.270       81.320   \n",
       "2017-07-06 17:55:00                0.0        8.074        8.089   \n",
       "\n",
       "                     h3_StartPrice  h3_EndPrice  h3_TradeVolume  \\\n",
       "CalcDateTime                                                      \n",
       "2017-07-19 10:52:00          8.483        8.485         22961.0   \n",
       "2017-07-17 19:07:00         91.010       91.050             0.0   \n",
       "2017-07-18 09:50:00          8.537        8.535          7720.0   \n",
       "2017-07-05 15:02:00         81.310       81.280          2579.0   \n",
       "2017-07-06 17:55:00          8.088        8.080             0.0   \n",
       "\n",
       "                     h3_NumberOfTrades  h4_MinPrice  h4_MaxPrice  \\\n",
       "CalcDateTime                                                       \n",
       "2017-07-19 10:52:00               22.0        8.484        8.488   \n",
       "2017-07-17 19:07:00                0.0       91.010       91.060   \n",
       "2017-07-18 09:50:00               10.0        8.538        8.542   \n",
       "2017-07-05 15:02:00               22.0       81.290       81.310   \n",
       "2017-07-06 17:55:00                0.0        8.074        8.089   \n",
       "\n",
       "                     h4_StartPrice  h4_EndPrice  h4_TradeVolume  \\\n",
       "CalcDateTime                                                      \n",
       "2017-07-19 10:52:00          8.488        8.484          8802.0   \n",
       "2017-07-17 19:07:00         91.010       91.050             0.0   \n",
       "2017-07-18 09:50:00          8.542        8.538          5890.0   \n",
       "2017-07-05 15:02:00         81.290       81.310          2675.0   \n",
       "2017-07-06 17:55:00          8.088        8.080             0.0   \n",
       "\n",
       "                     h4_NumberOfTrades  h5_MinPrice  h5_MaxPrice  \\\n",
       "CalcDateTime                                                       \n",
       "2017-07-19 10:52:00                6.0        8.488        8.491   \n",
       "2017-07-17 19:07:00                0.0       91.010       91.060   \n",
       "2017-07-18 09:50:00               11.0        8.542        8.545   \n",
       "2017-07-05 15:02:00               16.0       81.290       81.310   \n",
       "2017-07-06 17:55:00                0.0        8.074        8.089   \n",
       "\n",
       "                     h5_StartPrice  h5_EndPrice  h5_TradeVolume  \\\n",
       "CalcDateTime                                                      \n",
       "2017-07-19 10:52:00          8.488        8.490         15191.0   \n",
       "2017-07-17 19:07:00         91.010       91.050             0.0   \n",
       "2017-07-18 09:50:00          8.545        8.543             0.0   \n",
       "2017-07-05 15:02:00         81.310       81.290          1684.0   \n",
       "2017-07-06 17:55:00          8.088        8.080             0.0   \n",
       "\n",
       "                     h5_NumberOfTrades  Mnemonic_BMW  Mnemonic_EOAN  \\\n",
       "CalcDateTime                                                          \n",
       "2017-07-19 10:52:00               10.0             0              1   \n",
       "2017-07-17 19:07:00                0.0             0              0   \n",
       "2017-07-18 09:50:00                0.0             0              1   \n",
       "2017-07-05 15:02:00                7.0             1              0   \n",
       "2017-07-06 17:55:00                0.0             0              1   \n",
       "\n",
       "                     Mnemonic_SAP  \n",
       "CalcDateTime                       \n",
       "2017-07-19 10:52:00             0  \n",
       "2017-07-17 19:07:00             1  \n",
       "2017-07-18 09:50:00             0  \n",
       "2017-07-05 15:02:00             0  \n",
       "2017-07-06 17:55:00             0  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_features = pd.get_dummies(train_data)\n",
    "validate_features = pd.get_dummies(validate_data)\n",
    "test_features = pd.get_dummies(test_data)\n",
    "train_features.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "       colsample_bytree=1, gamma=0, learning_rate=0.1, max_delta_step=0,\n",
       "       max_depth=3, min_child_weight=1, missing=None, n_estimators=500,\n",
       "       n_jobs=1, nthread=None, objective='reg:linear', random_state=0,\n",
       "       reg_alpha=0, reg_lambda=1, scale_pos_weight=1, seed=None,\n",
       "       silent=True, subsample=1)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = xgboost.XGBRegressor (n_estimators=500)\n",
    "model.fit (train_features, train_labels, \n",
    "           eval_set=[(train_features, train_labels), (validate_features, validate_labels)],\n",
    "           early_stopping_rounds = 50, verbose=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f6348e96dd8>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x1800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "from matplotlib import pyplot as plt\n",
    "from xgboost import plot_importance\n",
    "fig, ax = plt.subplots(1,1,figsize=(25,25))\n",
    "plot_importance(model, height=0.8, ax=ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 2.087893778186076\n",
      "MAE: 0.9517232656610102\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
    "pred = model.predict (test_features)\n",
    "print (\"RMSE: {}\".format (mean_squared_error (y_true = test_labels, y_pred = pred)))\n",
    "print (\"MAE: {}\".format (mean_absolute_error (y_true = test_labels, y_pred = pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "433"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_labels.index.get_loc('2017-07-14 08:00:00')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f6346d1d748>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "start_loc = test_labels.index.get_loc('2017-07-14 08:00:00')\n",
    "end_loc = test_labels.index.get_loc('2017-07-14 20:00:00')\n",
    "data = pd.DataFrame ()\n",
    "data['True_Y'] = test_labels[start_loc:end_loc]\n",
    "data['Pred_Y'] = pred[start_loc:end_loc]\n",
    "\n",
    "data.plot (figsize=(25,10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Forecasting with an MLP <a name=\"mlp\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "import mxnet as mx\n",
    "from mxnet import nd, autograd, gluon"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_hidden_1 = 64\n",
    "num_hidden_2 = 24\n",
    "num_output = 1\n",
    "learning_rate = 0.001\n",
    "batch_size = 100\n",
    "num_examples = 33233\n",
    "\n",
    "model_ctx = mx.cpu ()\n",
    "data_ctx = mx.cpu ()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dataset = gluon.data.dataset.ArrayDataset (nd.array(train_features), nd.array(train_labels))\n",
    "train_dataloader = gluon.data.DataLoader (train_dataset, batch_size=batch_size)\n",
    "\n",
    "validate_dataset = gluon.data.dataset.ArrayDataset (nd.array(validate_features), nd.array(validate_labels))\n",
    "validate_dataloader = gluon.data.DataLoader (validate_dataset, batch_size=batch_size)\n",
    "\n",
    "test_dataset = gluon.data.dataset.ArrayDataset (nd.array(test_features), nd.array(test_labels))\n",
    "test_dataloader = gluon.data.DataLoader (test_dataset, batch_size=batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = gluon.nn.Sequential ()\n",
    "with model.name_scope ():\n",
    "    model.add (gluon.nn.Dense (num_hidden_1, activation='relu'))\n",
    "    model.add (gluon.nn.Dense (num_hidden_2, activation='relu'))\n",
    "    model.add (gluon.nn.Dense (num_output))\n",
    "    \n",
    "model.collect_params().initialize(mx.init.Normal(sigma=.1), ctx=model_ctx)\n",
    "l2loss = gluon.loss.L2Loss ()\n",
    "trainer = gluon.Trainer(model.collect_params(), 'adam', {'learning_rate': learning_rate})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_accuracy(data_iterator, model):\n",
    "    rmse = mx.metric.RMSE ()\n",
    "    for i, (data, label) in enumerate(data_iterator):\n",
    "        data = data.as_in_context(model_ctx).reshape((-1, 46))\n",
    "        label = label.as_in_context(model_ctx)\n",
    "        output = model(data)\n",
    "        predictions = output # predictions = nd.argmax(output, axis=1)\n",
    "        rmse.update(preds=predictions, labels=label)\n",
    "    return rmse.get()[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0. Loss: 1962.963911236102, Train_acc 27.4207754987212, Validate_acc 28.85848260905645\n",
      "Epoch 1. Loss: 205.47536626750426, Train_acc 22.250806126807564, Validate_acc 22.777545471713967\n",
      "Epoch 2. Loss: 176.13024672036266, Train_acc 19.08598838557083, Validate_acc 19.697416801975198\n",
      "Epoch 3. Loss: 133.25862023646252, Train_acc 14.094971853433195, Validate_acc 14.351780519093552\n",
      "Epoch 4. Loss: 112.6757434788381, Train_acc 13.860201052374038, Validate_acc 13.439384375533013\n",
      "Epoch 5. Loss: 96.95291852768827, Train_acc 12.094614399257804, Validate_acc 11.578186414013171\n",
      "Epoch 6. Loss: 61.42664280017521, Train_acc 18.92711680861273, Validate_acc 19.512030157324386\n",
      "Epoch 7. Loss: 55.95249971701932, Train_acc 10.628919929163562, Validate_acc 11.119008782791765\n",
      "Epoch 8. Loss: 27.38206881068951, Train_acc 9.78239513017058, Validate_acc 9.988344989410818\n",
      "Epoch 9. Loss: 27.586493060787777, Train_acc 10.753636950070096, Validate_acc 10.927772711401117\n",
      "Epoch 10. Loss: 20.077051043514025, Train_acc 9.000411697269715, Validate_acc 9.131902884130609\n",
      "Epoch 11. Loss: 17.749630135181015, Train_acc 7.078944565094623, Validate_acc 7.350342384756428\n",
      "Epoch 12. Loss: 24.382146596399206, Train_acc 6.92872441347522, Validate_acc 7.221549298665295\n",
      "Epoch 13. Loss: 43.366821501187815, Train_acc 4.260954897837951, Validate_acc 4.319326524865137\n",
      "Epoch 14. Loss: 9.322199358796786, Train_acc 4.458651005197637, Validate_acc 4.35183767096637\n",
      "Epoch 15. Loss: 6.413949864672499, Train_acc 3.6975829912624816, Validate_acc 3.8763565755870246\n",
      "Epoch 16. Loss: 7.054128294054508, Train_acc 4.459622103733705, Validate_acc 4.603357912742928\n",
      "Epoch 17. Loss: 6.634850076127826, Train_acc 3.063769644887996, Validate_acc 3.0940538840751124\n",
      "Epoch 18. Loss: 8.040041456485682, Train_acc 8.541903025505878, Validate_acc 8.893051062544732\n",
      "Epoch 19. Loss: 22.032662131068715, Train_acc 4.506830437486524, Validate_acc 4.668174145972892\n",
      "Epoch 20. Loss: 6.722301576016514, Train_acc 3.482580663412297, Validate_acc 3.549021629438008\n",
      "Epoch 21. Loss: 3.693864390044625, Train_acc 2.516801206106992, Validate_acc 2.596136989658826\n",
      "Epoch 22. Loss: 2.621646590119928, Train_acc 2.490343363834001, Validate_acc 2.60356118417766\n",
      "Epoch 23. Loss: 2.940316162118558, Train_acc 3.4869507899399066, Validate_acc 3.587617874145508\n",
      "Epoch 24. Loss: 6.264379913713006, Train_acc 2.696434367153653, Validate_acc 2.870456953571267\n",
      "Epoch 25. Loss: 4.133939272360256, Train_acc 2.2928761747694506, Validate_acc 2.383450057408581\n",
      "Epoch 26. Loss: 2.013715528851401, Train_acc 1.7361330527210563, Validate_acc 1.8361108123439631\n",
      "Epoch 27. Loss: 2.3730003930406576, Train_acc 3.111601471491286, Validate_acc 3.1882997914536357\n",
      "Epoch 28. Loss: 1.9921448390511765, Train_acc 1.606532769924177, Validate_acc 1.7156490634565484\n",
      "Epoch 29. Loss: 1.4371542345335742, Train_acc 1.7471711129667014, Validate_acc 1.8906883115637791\n",
      "Epoch 30. Loss: 3.0570023950802856, Train_acc 2.046717331991163, Validate_acc 2.1650015899579818\n",
      "Epoch 31. Loss: 2.03922623985097, Train_acc 1.1310608935929656, Validate_acc 1.2817988591651395\n",
      "Epoch 32. Loss: 2.0367172976877135, Train_acc 2.4975585331212202, Validate_acc 2.4888533582426096\n",
      "Epoch 33. Loss: 1.717098002665789, Train_acc 1.6669976309812355, Validate_acc 1.7701150538170174\n",
      "Epoch 34. Loss: 1.4648004848728278, Train_acc 1.3967091217893095, Validate_acc 1.4588835712981552\n",
      "Epoch 35. Loss: 1.4609978908600458, Train_acc 2.1885979228822636, Validate_acc 2.328723046877613\n",
      "Epoch 36. Loss: 1.6142885931686113, Train_acc 2.5056543743487487, Validate_acc 2.5993878106548363\n",
      "Epoch 37. Loss: 2.3912201472969894, Train_acc 1.4016802009847975, Validate_acc 1.445337219597542\n",
      "Epoch 38. Loss: 1.1455760289245775, Train_acc 1.2966766736351747, Validate_acc 1.379949729736537\n",
      "Epoch 39. Loss: 0.6799923252522043, Train_acc 1.1937618499359315, Validate_acc 1.2750733225312951\n",
      "Epoch 40. Loss: 0.6205892997619796, Train_acc 1.2822608472555364, Validate_acc 1.36080938169401\n",
      "Epoch 41. Loss: 1.093138289481013, Train_acc 0.8384495702191317, Validate_acc 0.9050780306123707\n",
      "Epoch 42. Loss: 0.7147912937515898, Train_acc 0.8509322643484857, Validate_acc 0.9141144082970816\n",
      "Epoch 43. Loss: 0.8293690052284427, Train_acc 3.005745856212996, Validate_acc 3.1113428024396503\n",
      "Epoch 44. Loss: 1.463594924086062, Train_acc 0.9986532244485679, Validate_acc 1.061176116335882\n",
      "Epoch 45. Loss: 0.6783624224395471, Train_acc 0.8301978873223373, Validate_acc 0.8645965620262982\n",
      "Epoch 46. Loss: 0.5599620064112782, Train_acc 0.7727686698289261, Validate_acc 0.7953796452038908\n",
      "Epoch 47. Loss: 0.497813637650555, Train_acc 0.9324112838691043, Validate_acc 0.9449141188843609\n",
      "Epoch 48. Loss: 0.41284191418672644, Train_acc 0.7631581287408612, Validate_acc 0.7833885742376928\n",
      "Epoch 49. Loss: 0.5693547427878458, Train_acc 0.8419443741696807, Validate_acc 0.8500441386275095\n"
     ]
    }
   ],
   "source": [
    "epochs = 50\n",
    "smoothing_constant = .01\n",
    "\n",
    "for e in range(epochs):\n",
    "    cumulative_loss = 0\n",
    "    for i, (data, label) in enumerate(train_dataloader):\n",
    "        data = data.as_in_context(model_ctx).reshape((-1, 46))\n",
    "        label = label.as_in_context(model_ctx)\n",
    "        with autograd.record():\n",
    "            output = model(data)\n",
    "            loss = l2loss(output, label)\n",
    "        loss.backward()\n",
    "        trainer.step(data.shape[0])\n",
    "        cumulative_loss += nd.sum(loss).asscalar()\n",
    "\n",
    "    validate_accuracy = evaluate_accuracy(validate_dataloader, model)\n",
    "    train_accuracy = evaluate_accuracy(train_dataloader, model)\n",
    "    print(\"Epoch %s. Loss: %s, Train_acc %s, Validate_acc %s\" %\n",
    "          (e, cumulative_loss/num_examples, train_accuracy, validate_accuracy))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 0.2554271180677173\n",
      "MAE: 0.2600659035917296\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
    "pred = model(nd.array(test_features)).asnumpy ().reshape(-1)\n",
    "print (\"RMSE: {}\".format (mean_squared_error (y_true = test_labels, y_pred = pred)))\n",
    "print (\"MAE: {}\".format (mean_absolute_error (y_true = test_labels, y_pred = pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f62dcedc6d8>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "start_loc = test_labels.index.get_loc('2017-07-14 08:00:00')\n",
    "end_loc = test_labels.index.get_loc('2017-07-14 20:00:00')\n",
    "data = pd.DataFrame ()\n",
    "data['True_Y'] = test_labels[start_loc:end_loc]\n",
    "data['Pred_Y'] = pred[start_loc:end_loc]\n",
    "\n",
    "data.plot (figsize=(25,10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "Extra credit\n",
    "\n",
    "## Forecasting with CNN\n",
    "\n",
    "## Forecasting with LSTM"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Forecasting with DeepAR"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "conda_mxnet_p36",
   "language": "python",
   "name": "conda_mxnet_p36"
  },
  "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
}