DeepLearningForTimeSeriesForecasting/2_RNN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# One step univariate model\n",
    "\n",
    "In this notebook, we demonstrate how to:\n",
    "- prepare time series data for training a RNN forecasting model\n",
    "- get data in the required shape for the keras API\n",
    "- implement a simple RNN model in keras to predict the next step ahead (time *t+1*) in the time series\n",
    "- enable early stopping to reduce the likelihood of model overfitting\n",
    "- evaluate the model on a test dataset\n",
    "\n",
    "The data in this example is taken from the GEFCom2014 forecasting competition<sup>1</sup>. It consists of 3 years of hourly electricity load and temperature values between 2012 and 2014. The task is to forecast future values of electricity load. In this example, we show how to forecast one time step ahead, using historical load data only.\n",
    "\n",
    "<sup>1</sup>Tao Hong, Pierre Pinson, Shu Fan, Hamidreza Zareipour, Alberto Troccoli and Rob J. Hyndman, \"Probabilistic energy forecasting: Global Energy Forecasting Competition 2014 and beyond\", International Journal of Forecasting, vol.32, no.3, pp 896-913, July-September, 2016."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Please run this notebook after completing 0_data_setup notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import warnings\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import datetime as dt\n",
    "from collections import UserDict\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from IPython.display import Image\n",
    "%matplotlib inline\n",
    "\n",
    "from common.utils import load_data, mape\n",
    "\n",
    "pd.options.display.float_format = '{:,.2f}'.format\n",
    "np.set_printoptions(precision=2)\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load the data from csv into a Pandas dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>load</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2012-01-01 00:00:00</th>\n",
       "      <td>2,698.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 01:00:00</th>\n",
       "      <td>2,558.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 02:00:00</th>\n",
       "      <td>2,444.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 03:00:00</th>\n",
       "      <td>2,402.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 04:00:00</th>\n",
       "      <td>2,403.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        load\n",
       "2012-01-01 00:00:00 2,698.00\n",
       "2012-01-01 01:00:00 2,558.00\n",
       "2012-01-01 02:00:00 2,444.00\n",
       "2012-01-01 03:00:00 2,402.00\n",
       "2012-01-01 04:00:00 2,403.00"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_dir = 'data/'\n",
    "energy = load_data(data_dir)[['load']]\n",
    "energy.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create train, validation and test sets\n",
    "\n",
    "We separate our dataset into train, validation and test sets. We train the model on the train set. The validation set is used to evaluate the model after each training epoch and ensure that the model is not overfitting the training data. After the model has finished training, we evaluate the model on the test set. We must ensure that the validation set and test set cover a later period in time from the training set, to ensure that the model does not gain from information from future time periods.\n",
    "\n",
    "We will allocate the period 1st November 2014 to 31st December 2014 to the test set. The period 1st September 2014 to 31st October is allocated to validation set. All other time periods are available for the training set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "valid_start_dt = '2014-09-01 00:00:00'\n",
    "test_start_dt = '2014-11-01 00:00:00'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "energy[energy.index < valid_start_dt][['load']].rename(columns={'load':'train'}) \\\n",
    "    .join(energy[(energy.index >=valid_start_dt) & (energy.index < test_start_dt)][['load']] \\\n",
    "          .rename(columns={'load':'validation'}), how='outer') \\\n",
    "    .join(energy[test_start_dt:][['load']].rename(columns={'load':'test'}), how='outer') \\\n",
    "    .plot(y=['train', 'validation', 'test'], figsize=(15, 8), fontsize=12)\n",
    "plt.xlabel('timestamp', fontsize=12)\n",
    "plt.ylabel('load', fontsize=12)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data preparation\n",
    "\n",
    "For this example, we will set *T=6*. This means that the input for each sample is a vector of the prevous 6 hours of the energy load. The choice of *T=6* was arbitrary but should be selected through experimentation.\n",
    "\n",
    "*HORIZON=1* specifies that we have a forecasting horizon of 1 (*t+1*)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAt4AAADUCAYAAACrkCQQAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAADsMAAA7DAcdvqGQAABrKSURBVHhe7d3djxxVesfx/Ik2LA7LS4IgwYQNYdEi3hYJ0IogBBdAhIwUEZIILWBQLHFD4hU3kYKQpQiBN76IJS58gSGSr+wbX1X0tPpAbfN7putUnXqeMz3fj/SVGc/Y5zA13f10dc34zwYAAAAAq2PwBgAAAAIweAMAAAABGLwBAACAAAzeAAAAQAAGbwAAACAAgzcAAAAQgMEbAAAACMDgDQAAAARg8AYAAAACMHgDAAAAARi8AQAAgAAM3gAAAEAABm8AAAAgwPzB+38+pqlFUuuTXxS1NvllUXshHYBj7V+//I4m1hKDd0SR1PrkF0WtTX5Z1F5Il0XthfwyqH2QXxI1YJKuJQbviCKp9ckvilqb/LKovZAui9oL+WVQ+yC/JGrAJF1LDN4RRVLrk18UtTb5ZVF7IV0WtRfyy6D2QX5J1IBJupYYvCOKpNYnvyhqbfLLovZCuixqL+SXQe2D/JKoAZN0LTF4RxRJrU9+UdTa5JdF7YV0WdReyC+D2gf5JVEDJulaYvCOKJJan/yiqLXJL4vaC+myqL2QXwa1D/JLogZM0rXE4B1RJLU++UVRa5NfFrUX0mVReyG/DGof5JdEDZika4nBO6JIan3yi6LWJr8sai+ky6L2Qn4Z1D7IL4kaMEnXEoN3RJHU+uQXRa1NflnUXkiXRe2F/DKofZBfEjVgkq4lBu+IIqn1yS+KWpv8sqi9kC6L2gv5ZVD7IL8kasAkXUsM3hFFUuuTXxS1NvllUXshXRa1F/LLoPZBfknUgEm6lhi8I4qk1ie/KGpt8sui9kK6LGov5JdB7YP8kqgBk3QtMXhHFEmtT35R1Nrkl0XthXRZ1F7IL4PaB/klUQMm6Vpi8I4oklqf/KKotckvi9oL6bKovZBfBrUP8kuiBkzStcTgHVEktT75RVFrk18WtRfSZVF7Ib8Mah/kl0QNmKRricE7okhqffKLotYmvyxqL6TLovZCfhnUPsgviRowSdcSg3dEkdT65BdFrU1+WdReSJdF7YX8Mqh9kF8SNWCSriUG74giqfXJL4pam/yyqL2QLovaC/llUPsgvyRqwCRdSwzeEUVS65NfFLU2+WVReyFdFrUX8sug9kF+SdSASbqWGLwjiqTWJ78oam3yy6L2Qrosai/kl0Htg/ySqAGTdC0xeEcUSa1PflHU2uSXRe2FdFnUXsgvg9oH+SVRAybpWmLwjiiSWp/8oqi1yS+L2gvpsqi9kF8GtQ/yS6IGTNK1xOAdUSS1PvlFUWuTXxa1F9JlUXshvwxqH+SXRA2YpGuJwTuiSGp98oui1ia/LGovpMui9kJ+GdQ+yC+JGjBJ1xKDd0SR1PrkF0WtTX5Z1F5Il0XthfwyqH2QXxI1YJKupRM7eL/w5N8Mp06dmpR9rPo7JhdJrX9g/e1f/8XmuDz0l/fI91cVRa19jPrvC28O//C73wxPPfZXf3LbsGPw6m//bvjPD16Tf252WdRejlH/91//PLz/xm83x+mXf37Xj8fJ/tvux/7t3Evyz80qi9rLMc9uX3bc7L7N/lt9zOwyqH2QXxI1YB5aL//Tp8Mvztwt31dTSyd28B4/KO3rH199Wv4dk4uk1j+gbMgrx2XxEyIrilr7mGQD9/j24GUfp/78rLKovRyDbOCeepxsKLePV39PVVnUXo5Z9vn/9/f+fnN/Zk9ex8dHffyiMqh9kF8SNWAeUk+/+s6Ptyv1/ppa6mrwtjsjG3LtgaH5GbRR9nfbgbDhW72/eZHU+guzMzE25Nqv6v1R2dm6ciOymuwnilq7QeVMtJ0pU+9vkd0ebTiwz/fu2TgbHsorEM2OiZVF7WVhEcfIjkM5BjbM2dvj4dr2MH6V71g9ad2l9tKotY+V3YeNj8NudltTf25RGdQ+OihqxqguiRowa3r+jX8ZHv71M8NvfveGfH9W71z842Zf49uW+riaWupq8B4/eKj3t6qcGbIHKPX+5kVS6y/I7qjKMcm8o7IHxLKPpvuJotZuUPlaXuUBe9u+z7N9jZThu9mT2SxqLwuLOEY2TNhA97//8Y58f2n8itG+j91bFrWXRq19rMZnt+02Y+vZ7asM4/a2+nOLyqD20UFRM0Z1SdSAWVP5XPY0eNuTAbu0xPZ1/0OP/LhH9bE1tdTV4G0PHvYJWvMByipDQtPrHY8qklp/QfagUL5w1fsjsgGhXBpUHhib7SeKWrtBdluxz8Xiy6EWNn5AO1ZPiHapvSws4hjtvhLhNX4ivfjViSxqL41a81jZ/ZgN2PZ5333SU+7f7HY0/v0mZVD76KCoGaO6JGrArKncl2QP3naG2y4r+eX9D2z2Y4O3vW3Xd5c9qj9XU0tdDd4RD1B2h1cOxPil2FWLpNZfUPYd1fhsqp2tswct++9mLwVHUWs3qHwtNxl2F9Z0L1nUXhbW0zGyylnXxfezWdReGpVxrMaPSbsDeZMyqH10UMSMMaskasCsqXzdZg/etn7Zy6+efml488IXm99n8J5Q+QSteadXrhNuNrhNKZJaf0HZd1Rl6C6Df3mpvNlLslHU2guz24l9Liz1/ujKXhi8f6q3Y2QxeOuyjlV5tWi17znKoPbRQeX4rjljzCqJGjBrKp/PFoP3a7//w2ZQtrPX6v1HZWe3xwN3icFbtPvNcl72QKH+/JzK4BY6SEZS61c0Pvuyr1XOzowqx8qG7/LqRBkaml0mFEWtPaPxNbpHZR+n/vxaja/Bn3rZw5FlUXuprNdjVCrrL74NZVF7mVkPx6pcPtfkG15VGdQ+EsqYMWaVRA2YRzUeZPdlH6v+Dq8HH31i1p87KgZv0dQ7vZZ3SOVauibDwdQiqfUrmnpHtdrZmW3la2M8dI+vT2029EdRa8+oPPHYV7OfLjIxeyJr6zZ7AMui9lJZr8fIGl+Lv/g2lEXtZWY9HKvVX1XMoPaRUMaMMaskasA8KgbvNrq51KTcQNa8AZSzcuHPbiOp9WdWhvC1h+zdyteCrTt+glReCm66nyhq7YWVoSFjgBtnx6g8oW32zWFZ1F4W1MsxKu1eurWoLGovDco6VramtdrlDxnUPpKLmDFml0QNmDWVr90Wl5oweE+hvngWVB4Q1rzTK2fldrOhwW6MzS5f2C2SWn9mq78EKhoP+7uvSpTj12RoKEVRay9ofPY/9NWbnWxYKEO3PbCpj5lVFrWXmfVyjErjs39N9pNF7WVhWcfK1irrllf2mpdB7SO5iBljdknUgFlT+dpl8K7TzeBdPjlrftNDueEdlZ31aH7HG0mtP7Pob6w8aui2VtlPFLX2gsrZf0u9f81sbXvwKsej+TGxsqi9zCzzGO02HrqbnWDIovaysKxjZbcjW9Mem9T7m5RB7SO5cnzXnDFml0QNmDWVzymDd50uBu+oOz3vBme/P35gssGv6TcORlLrz6x8PiLuqPYN3VY5s9p0P1HU2gta5ey/6KhvtrUnqfaqyCrfZJtF7WVmUcfoqOzYjJ8gNX1VL4vay8KyjlV53Gn6atFuGdQ+EouaMWaXRA2YNZXPKYN3nS4G73Knt+qz/gnZwFeGu6Z3wJHU+jMa31Gt9hLotilDt/3+KvuJotZeUBmmmv1YRadybHazoduGhdWelGVRe5lZ1DHysuvty/3ZsX8lb0ztZWFZxyrkuvIMah+J9TJjuCVRA2ZN9jm1GLzrdDF42zXE9onZd6c3Hr6Oask1yeUGajU7kxdJrT+jmpdAy4P7UdnHqD87Zei2ysc1v+OMotZeUPmc7/tmxnIc9zXlgd+e8NigbR9rx6H82aZnUUtZ1F5mlnGMrN2z3Ha/usqT5yxqLwvLuj2Vj2/+pGhcBrWPxKbOGGklUQNmTeXrd+rgXT5+bjacq7/Xi8H7iMqz/n0P4FPv9Gx4Vn9+SuOX1pudzYuk1p/R1JdApz4ZUq8glDX2Dd3W1P1UF0WtPbPx1+i+J4flAWdfc77Wx7fHZj/NpJRF7WVGWcfI7vvKEGn3q6u9ImFlUXtZUNaxso8pH6/e36wMah+JTZ0x0kqiBsyaytcvg3ed9ME77Fl/RWU/zR60Iqn1Z1TOaE49y1ZbzdBtrbafKGrtmdmQWz536v2RleNoD2zq/bPLovYyo+hjZLeh8asQNoCvcpZ7XBa1lwVl3Z7Kq6vqpETTMqh9JNXjjPGzkqgBs6byeeVSkzrpg3e507PU+zMq+zmpg/f4jmqNM2bjMz1LW7y/KGrtmdnLpfb/vvoD9oTGx3Lf2cKqsqi9zCjyGNlZvHKW24bvsOEii9rLgrJuT7aerbv65Q8Z1D6S6nHG+FlJ1IBZU/m8MnjXSR+8w571T2yVoTOSWr+y8TCl3r+0csxbpP7+qqKotWdWHrDt86jeH9n4a6XZ7cXKovYyo6hjVF5xSPl6yKL2sqCs21N5smSDoXp/szKofSRVHm96mTFkSdSAWVO572HwrpM+ePc0RFirPDuOpNavrLc7qlX3E0WtPbPy9dl00J3ZePDmjPdPRRyj2su1mpdF7WVBEcdqNzteZd2mtxtVBrWPpHqbMWRJ1IBZU/kaZvCukz54hz3rn1i5kS75ySg/K5Jav7LevgN81f1EUWvPaPyAvfo1vBMqP22m+fWxWdReKos4RuXyiLSh28qi9jKzrNvTarcbVQa1j6R6mzFkSdSAWdMvzty9+dwyeNdJHbzXvpa4tnJn2Hw/kdT6lZVv0urlDMGq35EeRa09o56uV7Tbbzk2x/anzexSe6ls7WM0fpUh9X4zi9rLzLJuT+WJU9MTPF4Z1D4S6m3GcEuiBsyayrDM4F0ndfCOeACZehZj/KPRmp9ZjaTWr6x8HnoYvMd3nKuc2Yui1p5RuezGUu9f2tSfGmPHpTxBszNKU29nk8ui9lLZ2seoXGISMrQdVRa1l5mtfay8Qk9uZFD7SKibJ6n7SqIGzJrKsPzwr5+R76+JwXsK9cVT2XioKndA9nt2ZrPV2U17cLLLR+zv2x3c7G37/XLWzmp+5s6KpNavrDwo2OeuDFR2ZijkQWKn8R2nev/ioqi1ZzQ+Q1ceSOwaUTs2S5+YlM+13R7syae9PR6o7b/t9+x95eXb1S51yKL2Utmax8gqf/cqrwDVlEXtZWZrHyuv3TVXLYPaR0IRM0aTkqgBs6anX/3pZ+Dbf5fff+ncR8ObF774k4/d12u//8NmUH7n4h/l++fE4O1UrqnerdUdUhkQ9mUfZ3fC6u9YXCS1fmV2B6U+R+WOK7Kyl1W+sdKKotaekT1oeF/T6uNrsgeiqbcXy47Jat8YlkXtpbI1j5HdL6q/d2rq75xdFrWXma15rLzGx9DWVx/TtAxqH0mtPWM0KYkaMGuyIfn+hx6Rn9+WZ67nxuDtZHc85ZvnLDvb2nLAs2HaztCpG5+tZWe4V3/mG0mtP6PxWU07A2qfp5AHiZ3K10bzy39KUdTaM7MzceVVCcu+tls9abRjbLcHO967txn7erDfs9vnmmcDN2VRe5nRWsfIe1I8JduD+jtnl0XtZUFr3p5U5bJGW1O9v3kZ1D6SWnvGaFISNWDWZsP3r55+6cdvtLRf7dKT2jPea8TgfZKLpNYnvyhqbfLLovZCuixqL+SXQe2D/JKoAZN0LTF4RxRJrU9+UdTa5JdF7YV0WdReyC+D2gf5JVEDJulaYvCOKJJan/yiqLXJL4vaC+myqL2QXwa1D/JLogZM0rXE4B1RJLU++UVRa5NfFrUX0mVReyG/DGof5JdEDZika4nBO6JIan3yi6LWJr8sai+ky6L2Qn4Z1D7IL4kaMEnXEoN3RJHU+uQXRa1NflnUXkiXRe2F/DKofZBfEjVgkq4lBu+IIqn1yS+KWpv8sqi9kC6L2gv5ZVD7IL8kasAkXUsM3hFFUuuTXxS1NvllUXshXRa1F/LLoPZBfknUgEm6lhi8I4qk1ie/KGpt8sui9kK6LGov5JdB7YP8kqgBk3QtMXhHFEmtT35R1Nrkl0XthXRZ1F7IL4PaB/klUQMm6Vpi8I4oklqf/KKotckvi9oL6bKovZBfBrUP8kuiBkzStcTgHVEktT75RVFrk18WtRfSZVF7Ib8Mah/kl0QNmKRricE7okhqffKLotYmvyxqL6TLovZCfhnUPsgviRowSdcSg3dEkdT65BdFrU1+WdReSJdF7YX8Mqh9kF8SNWCSriUG74giqfXJL4pam/yyqL2QLovaC/llUPsgvyRqwCRdSwzeEUVS65NfFLU2+WVReyFdFrUX8sug9kF+SdSASbqWGLwjiqTWJ78oam3yy6L2Qrosai/kl0Htg/ySqAGTdC0xeEcUSa1PflHU2uSXRe2FdFnUXsgvg9oH+SVRAybpWmLwjiiSWp/8oqi1yS+L2gvpsqi9kF8GtQ/yS6IGTNK1xOAdUSS1PvlFUWuTXxa1F9JlUXshvwxqH+SXRA2YpGuJwTuiSGp98oui1ia/LGovpMui9kJ+GdQ+yC+JGjBJ1xKDd0SR1PrkF0WtTX5Z1F5Il0XthfwyqH2QXxI1YJKuJQbviCKp9ckvilqb/LKovZAui9oL+WVQ+yC/JGrAJF1LDN4RRVLrk18UtTb5ZVF7IV0WtRfyy6D2QX5J1IBJupbmD94AAAAAJmPwBgAAAAIweAMAAAABGLwBAACAAAzeAAAAQAAGbwAAACAAgzcAAAAQgMEbAAAACMDgDQAAAARg8AYAAAACMHgDAAAAARi8AQAAgAAM3gAAAEAABm8AAAAgAIM3AAAAEIDBGwAAAAjA4A0AAAAEYPAGAAAAAjB4AwAAAAEYvAEAAIAADN4AAABAAAZvAAAAIACDNwAAABCAwRsAAAAIwOANAAAABGDwBgAAAALMHrw//vjj7X8BAAAA2Gf24H3mzJnh9u3b27cAAABQw+Yom6dwcswevO+7777h5s2b27cAAABQw+Yom6dwcswevB988MHhxo0b27cAAABQw+Yom6dwcswevB955JHh+++/374FAACAGtevX9/MUzg5Zg/ejz322HDt2rXtWwAAAKhhc5TNUzg5Zg/eTz755HD16tXtWwAAAKhhc5TNUzg5Zg/ezz777HDlypXtWwAAAKhhc5TNUzg5Zg/eL7/88nDp0qXtWwAAAKhhc5TNUzg5Zg/eb7/99nDx4sXtWwAAAKhhc5TNUzg5Zg/e77///nD+/PntWwAAAKhhc5TNUzg5Zg/en3322XDu3LntWwAAAKhhc5TNUzg5Zg/eX3311fDiiy9u3wIAAEANm6NsnsLJMXvw/u6774azZ89u3wIAAEANm6NsnsLJMXvwNnffffdw69at7VsAAACYwuYnm6NwsiwavJ977rnhm2++2b4FAACAKWx+sjkKJ8uiwfu9994bPvzww+1bAAAAmOKjjz7azFE4WRYN3pcvXx6eeuqp7VsAAACYwuYnrho4eRYN3ubxxx8fLly4sH0LAAAAR7G5yeYnnDyLB+8vvvhieOCBB4bTp08TEaVn90d2vzTG/RQR9ZS6n8LJsHjwBoCeqDNJvDIHAOgBgzeAg2PXTtr3oBi+FwUA0AsGbwAHx37aUvlpAfar/fQAAACyMXgDODjjn4/LvzcAAOgFgzeAgzP+F+H4F3YBAL1g8AZwkM6ePTt8/fXXm18BAOgBgzeAg/TCCy8Mn3zyyeZXAAB6wOAN4CCdO3dueP311ze/AgDQAwZvAAfp/PnzwzPPPLP5FQCAHjB4AzhIFy9eHB599NHNrwAA9IDBG8BBunTp0uafZbZfAQDoAYM3gIN05cqV4Z577tn8CgBADxi8ARykq1evbn6Gt/0KAEAPGLwBHKRr164Nd9111+ZXAAB6wOAN4CBdv359uPPOOze/AgDQAwZvAAfpxo0bwx133LH5FQCAHjB4AzhIN2/eHE6fPr35FQCAHjB4AzhIt2/fHk6dOrX5FQCAHjB4AzhYNngDANALBm8AAAAgAIM3AAAAEIDBGwAAAAjA4A0AAAAEYPAGAAAAAjB4AwAAAAEYvAEAAIAADN4AAABAAAZvAAAAIACDNwAAABCAwRsAAAAIwOANAAAABGDwBgAAAAIweAMAAAABGLwBAACAAAzeAAAAQAAGbwAAACAAgzcAAAAQgMEbQLhvv/12+Pzzz2dlfzbDDz/8MHz66afDW2+9Nbzyyiub3n333eHLL7/cfgQAAEdj8AYQzgboU6dOzcoG30i3bt3aDNj33nuv3I9lHwMAwD4M3gDC2VlrG2ZVZZhV77NsaI9iA7Wd2bb9PPHEE5sz3rZ3+33LznbbEwEGbwDAFAzeALpSzizbpR3ZytD9/PPPM1wDABZj8AbQlV4G73I5jJ3pZugGALTA4A2gK70M3g8//PBmH5cvX97+DgAAyzB4A+hKD4O3Xbtte7BLTQAAaIXBG0BXehi87RsmbQ/8qEAAQEsM3gC60sPgXS4zyb7cBQBwWBi8AXRlzuBt12Hb2ena1D/GY99IaevbPsrbH3zwweabLO33LPspJ/ajDfmmSwBADQZvAF2ZM3iXH/tXmw3Qu2wYt/eVn2ZiH2Nv277s98rZ8PJ7fPMlAGAqBm8AXZkzeNvwW/5J+ZrsrPcu+7tsfRu4baC3/dg/nDM+u23DeRn27f1ckgIAmILBG0BX5gzeLdkwbuuXoVpdjlKUs+HR/4w9AOB4YvAG0JWeBm+7tvsotsfysVzvDQDYh8EbQFd6Gryn7IF/aAcAMBWDN4CuZA/e5RpvG6inKJeb2DXjAAAchcEbQFfmDN7lmyBrU/8ypV0yYuvb+6co32TJ4A0A2IfBG0BXbOCtHbztWmw781yb/SxupeYf0LG/xz5W/YQUAADGGLwBdGXO4N2aDeS2B/sxgkexs+Pl7HnmfgEAxwODN4Cu9DB4l39Ex/ZyFDvTbh+nLlkBAGAXgzeArvQweJty7bZdSqLY2fBytvuon/UNAEDB4A2gK70M3nYZif0T8bYX+9XObtt13PZNlOW6btsr13YDAKZi8AbQlV4Gb2PDt/2rlLaf3Wz45kw3AKAGgzeArtiwW+qF7cXOdNvlJRYDNwBgDgZvAAAAIACDNwAAABCAwRsAAAAIwOANAAAABGDwBgAAAAIweAMAAAABGLwBAACAAAzeAAAAQAAGbwAAACAAgzcAAAAQgMEbAAAACMDgDQAAAARg8AYAAAACMHgDAAAAARi8AQAAgNUNw/8DOEofgioOKPkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image('./images/one_step_forecast.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "T = 6\n",
    "HORIZON = 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our data preparation for the training set will involve the following steps:\n",
    "\n",
    "1. Filter the original dataset to include only that time period reserved for the training set\n",
    "2. Scale the time series such that the values fall within the interval (0, 1)\n",
    "3. Shift the values of the time series to create a Pandas dataframe containing all the data for a single training example\n",
    "4. Discard any samples with missing values\n",
    "5. Transform this Pandas dataframe into a numpy array of shape (samples, time steps, features) for input into Keras"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Data preparation - training set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>load_original</th>\n",
       "      <th>y_t+1</th>\n",
       "      <th>load_t-5</th>\n",
       "      <th>load_t-4</th>\n",
       "      <th>load_t-3</th>\n",
       "      <th>load_t-2</th>\n",
       "      <th>load_t-1</th>\n",
       "      <th>load_t</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2012-01-01 05:00:00</th>\n",
       "      <td>0.15</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 06:00:00</th>\n",
       "      <td>0.18</td>\n",
       "      <td>0.23</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 07:00:00</th>\n",
       "      <td>0.23</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 08:00:00</th>\n",
       "      <td>0.29</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.23</td>\n",
       "      <td>0.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 09:00:00</th>\n",
       "      <td>0.35</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.23</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.35</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     load_original  y_t+1  load_t-5  load_t-4  load_t-3  \\\n",
       "2012-01-01 05:00:00           0.15   0.18      0.22      0.18      0.14   \n",
       "2012-01-01 06:00:00           0.18   0.23      0.18      0.14      0.13   \n",
       "2012-01-01 07:00:00           0.23   0.29      0.14      0.13      0.13   \n",
       "2012-01-01 08:00:00           0.29   0.35      0.13      0.13      0.15   \n",
       "2012-01-01 09:00:00           0.35   0.37      0.13      0.15      0.18   \n",
       "\n",
       "                     load_t-2  load_t-1  load_t  \n",
       "2012-01-01 05:00:00      0.13      0.13    0.15  \n",
       "2012-01-01 06:00:00      0.13      0.15    0.18  \n",
       "2012-01-01 07:00:00      0.15      0.18    0.23  \n",
       "2012-01-01 08:00:00      0.18      0.23    0.29  \n",
       "2012-01-01 09:00:00      0.23      0.29    0.35  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1. Get the train data from the correct data range\n",
    "train = energy.copy()[energy.index < valid_start_dt][['load']]\n",
    "\n",
    "# 2. Scale data to be in range (0, 1). \n",
    "#   This transformation should be calibrated on the training set only. \n",
    "#   This is to prevent information from the validation or test sets \n",
    "#   leaking into the training data.\n",
    "scaler = MinMaxScaler()\n",
    "train['load'] = scaler.fit_transform(train)\n",
    "\n",
    "# 3. Shift the dataframe to create the input samples.\n",
    "train_shifted = train.copy()\n",
    "train_shifted['y_t+1'] = train_shifted['load'].shift(-1, freq='H')\n",
    "for t in range(1, T+1):\n",
    "    train_shifted[str(T-t)] = train_shifted['load'].shift(T-t, freq='H')\n",
    "y_col = 'y_t+1'\n",
    "X_cols = ['load_t-5',\n",
    "             'load_t-4',\n",
    "             'load_t-3',\n",
    "             'load_t-2',\n",
    "             'load_t-1',\n",
    "             'load_t']\n",
    "train_shifted.columns = ['load_original']+[y_col]+X_cols\n",
    "\n",
    "# 4.Discard any samples with missing values\n",
    "train_shifted = train_shifted.dropna(how='any')\n",
    "train_shifted.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now convert the target and input features into numpy arrays. X needs to be in the **shape (samples, time steps, features)**. Here we have 23370 samples, 6 time steps and 1 feature (load)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 5.Transform this Pandas dataframe into a numpy array\n",
    "y_train = train_shifted[y_col].as_matrix()\n",
    "X_train = train_shifted[X_cols].as_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image('./images/rnn_data_prep.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# This is an important step to reshape the X into 3 dimension array\n",
    "X_train = X_train.reshape(X_train.shape[0], T, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now have a vector for target variable of shape:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(23370,)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The target varaible for the first 3 samples looks like:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.18, 0.23, 0.29])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train[:3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The tensor for the input features now has the shape:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(23370, 6, 1)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And the first 3 samples looks like:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0.22],\n",
       "        [0.18],\n",
       "        [0.14],\n",
       "        [0.13],\n",
       "        [0.13],\n",
       "        [0.15]],\n",
       "\n",
       "       [[0.18],\n",
       "        [0.14],\n",
       "        [0.13],\n",
       "        [0.13],\n",
       "        [0.15],\n",
       "        [0.18]],\n",
       "\n",
       "       [[0.14],\n",
       "        [0.13],\n",
       "        [0.13],\n",
       "        [0.15],\n",
       "        [0.18],\n",
       "        [0.23]]])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train[:3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can sense check this against the first 3 records of the original dataframe:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>load_original</th>\n",
       "      <th>y_t+1</th>\n",
       "      <th>load_t-5</th>\n",
       "      <th>load_t-4</th>\n",
       "      <th>load_t-3</th>\n",
       "      <th>load_t-2</th>\n",
       "      <th>load_t-1</th>\n",
       "      <th>load_t</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2012-01-01 05:00:00</th>\n",
       "      <td>0.15</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 06:00:00</th>\n",
       "      <td>0.18</td>\n",
       "      <td>0.23</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 07:00:00</th>\n",
       "      <td>0.23</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     load_original  y_t+1  load_t-5  load_t-4  load_t-3  \\\n",
       "2012-01-01 05:00:00           0.15   0.18      0.22      0.18      0.14   \n",
       "2012-01-01 06:00:00           0.18   0.23      0.18      0.14      0.13   \n",
       "2012-01-01 07:00:00           0.23   0.29      0.14      0.13      0.13   \n",
       "\n",
       "                     load_t-2  load_t-1  load_t  \n",
       "2012-01-01 05:00:00      0.13      0.13    0.15  \n",
       "2012-01-01 06:00:00      0.13      0.15    0.18  \n",
       "2012-01-01 07:00:00      0.15      0.18    0.23  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_shifted.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Data preparation - validation set"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we follow a similar process for the validation set. We keep *T* hours from the training set in order to construct initial features."
   ]
  },
  {
   "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>load</th>\n",
       "      <th>y+1</th>\n",
       "      <th>load_t-5</th>\n",
       "      <th>load_t-4</th>\n",
       "      <th>load_t-3</th>\n",
       "      <th>load_t-2</th>\n",
       "      <th>load_t-1</th>\n",
       "      <th>load_t-0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2014-09-01 00:00:00</th>\n",
       "      <td>0.28</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-09-01 01:00:00</th>\n",
       "      <td>0.24</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-09-01 02:00:00</th>\n",
       "      <td>0.22</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.43</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.22</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     load  y+1  load_t-5  load_t-4  load_t-3  load_t-2  \\\n",
       "2014-09-01 00:00:00  0.28 0.24      0.61      0.58      0.51      0.43   \n",
       "2014-09-01 01:00:00  0.24 0.22      0.58      0.51      0.43      0.34   \n",
       "2014-09-01 02:00:00  0.22 0.22      0.51      0.43      0.34      0.28   \n",
       "\n",
       "                     load_t-1  load_t-0  \n",
       "2014-09-01 00:00:00      0.34      0.28  \n",
       "2014-09-01 01:00:00      0.28      0.24  \n",
       "2014-09-01 02:00:00      0.24      0.22  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1. Get the validation data from the correct data range\n",
    "look_back_dt = dt.datetime.strptime(valid_start_dt, '%Y-%m-%d %H:%M:%S') - dt.timedelta(hours=T-1)\n",
    "valid = energy.copy()[(energy.index >=look_back_dt) & (energy.index < test_start_dt)][['load']]\n",
    "\n",
    "# 2. Scale the series using the transformer fitted on the training set:\n",
    "valid['load'] = scaler.transform(valid)\n",
    "\n",
    "# 3. Shift the dataframe to create the input samples\n",
    "valid_shifted = valid.copy()\n",
    "valid_shifted['y+1'] = valid_shifted['load'].shift(-1, freq='H')\n",
    "for t in range(1, T+1):\n",
    "    valid_shifted['load_t-'+str(T-t)] = valid_shifted['load'].shift(T-t, freq='H')\n",
    "    \n",
    "# 4.Discard any samples with missing values\n",
    "valid_shifted = valid_shifted.dropna(how='any')\n",
    "valid_shifted.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 5.Transform this Pandas dataframe into a numpy array\n",
    "y_valid = valid_shifted['y+1'].as_matrix()\n",
    "X_valid = valid_shifted[['load_t-'+str(T-t) for t in range(1, T+1)]].as_matrix()\n",
    "X_valid = X_valid.reshape(X_valid.shape[0], T, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Prepare validation inputs in the same way as the training set:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1463,)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_valid.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.24, 0.22, 0.22])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_valid[:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1463, 6, 1)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_valid.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0.61],\n",
       "        [0.58],\n",
       "        [0.51],\n",
       "        [0.43],\n",
       "        [0.34],\n",
       "        [0.28]],\n",
       "\n",
       "       [[0.58],\n",
       "        [0.51],\n",
       "        [0.43],\n",
       "        [0.34],\n",
       "        [0.28],\n",
       "        [0.24]],\n",
       "\n",
       "       [[0.51],\n",
       "        [0.43],\n",
       "        [0.34],\n",
       "        [0.28],\n",
       "        [0.24],\n",
       "        [0.22]]])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_valid[:3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Implement the RNN"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will implement a simple RNN forecasting model with the following structure:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image('./images/one_step_univariate.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Model, Sequential\n",
    "from keras.layers import GRU, Dense\n",
    "from keras.callbacks import EarlyStopping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "LATENT_DIM = 5 # number of units in the RNN layer\n",
    "BATCH_SIZE = 32 # number of samples per mini-batch\n",
    "EPOCHS = 10 # maximum number of times the training algorithm will cycle through all samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /anaconda/envs/py35/lib/python3.5/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(GRU(LATENT_DIM, input_shape=(T, 1)))\n",
    "model.add(Dense(HORIZON))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use mean squared error as the loss function. The Keras documentation recommends the optimizer RMSprop for RNNs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer='RMSprop', loss='mse')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "gru_1 (GRU)                  (None, 5)                 105       \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1)                 6         \n",
      "=================================================================\n",
      "Total params: 111\n",
      "Trainable params: 111\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Specify the early stopping criteria. We **monitor** the validation loss (in this case the mean squared error) on the validation set after each training epoch. If the validation loss has not improved by **min_delta** after **patience** epochs, we stop the training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "earlystop = EarlyStopping(monitor='val_loss', min_delta=0, patience=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /anaconda/envs/py35/lib/python3.5/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Train on 23370 samples, validate on 1463 samples\n",
      "Epoch 1/10\n",
      "23370/23370 [==============================] - 25s 1ms/step - loss: 0.0263 - val_loss: 0.0019\n",
      "Epoch 2/10\n",
      "23370/23370 [==============================] - 6s 252us/step - loss: 0.0013 - val_loss: 8.3454e-04\n",
      "Epoch 3/10\n",
      "23370/23370 [==============================] - 6s 251us/step - loss: 7.6025e-04 - val_loss: 5.6926e-04\n",
      "Epoch 4/10\n",
      "23370/23370 [==============================] - 6s 253us/step - loss: 6.0118e-04 - val_loss: 5.4199e-04\n",
      "Epoch 5/10\n",
      "23370/23370 [==============================] - 6s 253us/step - loss: 5.6709e-04 - val_loss: 5.2899e-04\n",
      "Epoch 6/10\n",
      "23370/23370 [==============================] - 6s 254us/step - loss: 5.5480e-04 - val_loss: 4.9941e-04\n",
      "Epoch 7/10\n",
      "23370/23370 [==============================] - 6s 253us/step - loss: 5.4845e-04 - val_loss: 6.2813e-04\n",
      "Epoch 8/10\n",
      "23370/23370 [==============================] - 6s 252us/step - loss: 5.4434e-04 - val_loss: 5.2482e-04\n",
      "Epoch 9/10\n",
      "23370/23370 [==============================] - 6s 253us/step - loss: 5.3995e-04 - val_loss: 5.1850e-04\n",
      "Epoch 10/10\n",
      "23370/23370 [==============================] - 6s 253us/step - loss: 5.3340e-04 - val_loss: 4.7774e-04\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train,\n",
    "                    y_train,\n",
    "                    batch_size=BATCH_SIZE,\n",
    "                    epochs=EPOCHS,\n",
    "                    validation_data=(X_valid, y_valid),\n",
    "                    callbacks=[earlystop],\n",
    "                    verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_df = pd.DataFrame.from_dict({'train_loss':history.history['loss'], 'val_loss':history.history['val_loss']})\n",
    "plot_df.plot(logy=True, figsize=(10,10), fontsize=12)\n",
    "plt.xlabel('epoch', fontsize=12)\n",
    "plt.ylabel('loss', fontsize=12)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data preperation - test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 1. Get the test data from the correct data range\n",
    "look_back_dt = dt.datetime.strptime(test_start_dt, '%Y-%m-%d %H:%M:%S') - dt.timedelta(hours=T-1)\n",
    "test = energy.copy()[test_start_dt:][['load']]\n",
    "\n",
    "# 2. Scale the data\n",
    "test['load'] = scaler.transform(test)\n",
    "\n",
    "# 3. Shift the dataframe to create the input samples\n",
    "test_shifted = test.copy()\n",
    "test_shifted['y_t+1'] = test_shifted['load'].shift(-1, freq='H')\n",
    "for t in range(1, T+1):\n",
    "    test_shifted['load_t-'+str(T-t)] = test_shifted['load'].shift(T-t, freq='H')\n",
    "\n",
    "# 4.Discard any samples with missing values\n",
    "test_shifted = test_shifted.dropna(how='any')\n",
    "\n",
    "# 5.Transform this Pandas dataframe into a numpy array\n",
    "y_test = test_shifted['y_t+1'].as_matrix()\n",
    "X_test = test_shifted[['load_t-'+str(T-t) for t in range(1, T+1)]].as_matrix()\n",
    "X_test = X_test.reshape(X_test.shape[0], T, 1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1458,)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1458, 6, 1)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Make predictions on test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.21],\n",
       "       [0.3 ],\n",
       "       [0.38],\n",
       "       ...,\n",
       "       [0.54],\n",
       "       [0.46],\n",
       "       [0.42]], dtype=float32)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions = model.predict(X_test)\n",
    "predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compare predictions to actual load"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>timestamp</th>\n",
       "      <th>h</th>\n",
       "      <th>prediction</th>\n",
       "      <th>actual</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2014-11-01 05:00:00</td>\n",
       "      <td>t+1</td>\n",
       "      <td>2,673.13</td>\n",
       "      <td>2,714.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2014-11-01 06:00:00</td>\n",
       "      <td>t+1</td>\n",
       "      <td>2,947.12</td>\n",
       "      <td>2,970.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2014-11-01 07:00:00</td>\n",
       "      <td>t+1</td>\n",
       "      <td>3,208.74</td>\n",
       "      <td>3,189.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2014-11-01 08:00:00</td>\n",
       "      <td>t+1</td>\n",
       "      <td>3,337.19</td>\n",
       "      <td>3,356.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2014-11-01 09:00:00</td>\n",
       "      <td>t+1</td>\n",
       "      <td>3,466.88</td>\n",
       "      <td>3,436.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            timestamp    h  prediction   actual\n",
       "0 2014-11-01 05:00:00  t+1    2,673.13 2,714.00\n",
       "1 2014-11-01 06:00:00  t+1    2,947.12 2,970.00\n",
       "2 2014-11-01 07:00:00  t+1    3,208.74 3,189.00\n",
       "3 2014-11-01 08:00:00  t+1    3,337.19 3,356.00\n",
       "4 2014-11-01 09:00:00  t+1    3,466.88 3,436.00"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eval_df = pd.DataFrame(predictions, columns=['t+'+str(t) for t in range(1, HORIZON+1)])\n",
    "eval_df['timestamp'] = test_shifted.index\n",
    "eval_df = pd.melt(eval_df, id_vars='timestamp', value_name='prediction', var_name='h')\n",
    "eval_df['actual'] = np.transpose(y_test).ravel()\n",
    "eval_df[['prediction', 'actual']] = scaler.inverse_transform(eval_df[['prediction', 'actual']])\n",
    "eval_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compute the mean absolute percentage error over all predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# %load -s mape common/utils.py\n",
    "def mape(predictions, actuals):\n",
    "    \"\"\"Mean absolute percentage error\"\"\"\n",
    "    return ((predictions - actuals).abs() / actuals).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.015229265571591601"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mape(eval_df['prediction'], eval_df['actual'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot the predictions vs the actuals for the first week of the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "eval_df[eval_df.timestamp<'2014-11-08'].plot(x='timestamp', y=['prediction', 'actual'], style=['r', 'b'], figsize=(15, 8))\n",
    "plt.xlabel('timestamp', fontsize=12)\n",
    "plt.ylabel('load', fontsize=12)\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.5",
   "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": "3.5.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}