From regression to RNN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## From Regression to RNN\n",
    "\n",
    "Learning objectives:\n",
    "* [Objective 1] ARIMA\n",
    "* [Objective 2] RNN requires a 3-D tensor as input\n",
    "* [Objective 3] Stock data\n",
    "* [Objective 4] Create the training and test data for RNN/LSTM\n",
    "* [Objective 5] Run a simple RNN model\n",
    "* [Objective 6] RNN/LTSM requires normalized input data\n",
    "* [Objective 7] LSTM\n",
    "* [Objective 8] LSTM Regularization\n",
    "* [Objective 9] GRU\n",
    "* [Objective 10] GRU Regularization\n",
    "\n",
    "Note: Every run of the following code may vary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.graphics.tsaplots import plot_pacf, plot_acf\n",
    "from statsmodels.tsa.arima_process import ArmaProcess\n",
    "from statsmodels.tsa.stattools import pacf\n",
    "from statsmodels.regression.linear_model import yule_walker\n",
    "#from statsmodels.tsa.stattools import adfuller\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### [Learning Objective 1] ARIMA()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from statsmodels.graphics.tsaplots import plot_pacf, plot_acf\n",
    "from statsmodels.tsa.arima_process import ArmaProcess\n",
    "from statsmodels.tsa.stattools import pacf\n",
    "from statsmodels.regression.linear_model import yule_walker\n",
    "#from statsmodels.tsa.stattools import adfuller\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "%matplotlib inline\n",
    "\n",
    "# Generate the data\n",
    "import numpy as np\n",
    "ar = np.array([1, -0.8, 0.2])\n",
    "ma = np.array([1])\n",
    "my_simulation = ArmaProcess(ar, ma).generate_sample(nsample=100)\n",
    "\n",
    "plt.figure(figsize=[10, 5]); # Set dimensions for figure\n",
    "plt.plot(my_simulation, linestyle='-', marker='o', color='b')\n",
    "plt.title(\"Simulated Process\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot\n",
    "from pandas.plotting import autocorrelation_plot\n",
    "plot_acf(my_simulation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_pacf(my_simulation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                               SARIMAX Results                                \n",
      "==============================================================================\n",
      "Dep. Variable:                      y   No. Observations:                  100\n",
      "Model:                 ARIMA(2, 0, 0)   Log Likelihood                -143.183\n",
      "Date:                Thu, 03 Dec 2020   AIC                            294.366\n",
      "Time:                        23:41:06   BIC                            304.787\n",
      "Sample:                             0   HQIC                           298.583\n",
      "                                - 100                                         \n",
      "Covariance Type:                  opg                                         \n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const          0.4291      0.214      2.009      0.045       0.011       0.848\n",
      "ar.L1          0.8461      0.114      7.438      0.000       0.623       1.069\n",
      "ar.L2         -0.3351      0.100     -3.367      0.001      -0.530      -0.140\n",
      "sigma2         1.0184      0.177      5.753      0.000       0.671       1.365\n",
      "===================================================================================\n",
      "Ljung-Box (L1) (Q):                   0.02   Jarque-Bera (JB):                 2.20\n",
      "Prob(Q):                              0.88   Prob(JB):                         0.33\n",
      "Heteroskedasticity (H):               0.67   Skew:                             0.23\n",
      "Prob(H) (two-sided):                  0.25   Kurtosis:                         2.43\n",
      "===================================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n"
     ]
    }
   ],
   "source": [
    "import warnings\n",
    "import statsmodels.api as sm\n",
    "from statsmodels.tsa.arima_model import ARMA\n",
    "\n",
    "mod = sm.tsa.arima.ARIMA(my_simulation, order=(2, 0, 0))\n",
    "mod_fit = mod.fit()\n",
    "print(mod_fit.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### [Learning Objective 2] RNN requires a 3-D tensor as input\n",
    "* Tensor: One tensor is a vector that enters the model\n",
    "* Time Step: One time step is one observation in the tensor.\n",
    "* Feature: One feature is one observation at a time step."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.12509907, 0.2133423 , 0.52879158, 0.20713005, 0.74325534,\n",
       "       0.74410956, 0.56170084, 0.97857636, 0.03467591, 0.69219908,\n",
       "       0.8078287 , 0.20574229, 0.34024462, 0.80735351, 0.46677125,\n",
       "       0.31449383, 0.61349645, 0.69445249, 0.47836441, 0.2243877 ])"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "array = np.random.random(20) #.astype(np.float32\n",
    "array.reshape((1,20,1))\n",
    "array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=[10, 5]); # Set dimensions for figure\n",
    "plt.plot(array, linestyle='-', marker='o', color='b')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0.12509907],\n",
       "        [0.2133423 ],\n",
       "        [0.52879158],\n",
       "        [0.20713005],\n",
       "        [0.74325534]],\n",
       "\n",
       "       [[0.74410956],\n",
       "        [0.56170084],\n",
       "        [0.97857636],\n",
       "        [0.03467591],\n",
       "        [0.69219908]],\n",
       "\n",
       "       [[0.8078287 ],\n",
       "        [0.20574229],\n",
       "        [0.34024462],\n",
       "        [0.80735351],\n",
       "        [0.46677125]],\n",
       "\n",
       "       [[0.31449383],\n",
       "        [0.61349645],\n",
       "        [0.69445249],\n",
       "        [0.47836441],\n",
       "        [0.2243877 ]]])"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "array.reshape((4,5,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0.12509907, 0.2133423 ],\n",
       "        [0.52879158, 0.20713005],\n",
       "        [0.74325534, 0.74410956],\n",
       "        [0.56170084, 0.97857636],\n",
       "        [0.03467591, 0.69219908]],\n",
       "\n",
       "       [[0.8078287 , 0.20574229],\n",
       "        [0.34024462, 0.80735351],\n",
       "        [0.46677125, 0.31449383],\n",
       "        [0.61349645, 0.69445249],\n",
       "        [0.47836441, 0.2243877 ]]])"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "array.reshape((2,5,2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0.40228021, 0.57090883],\n",
       "        [0.84224461, 0.17685027],\n",
       "        [0.42273716, 0.97394303],\n",
       "        [0.87404443, 0.14958933],\n",
       "        [0.4071778 , 0.66398901]],\n",
       "\n",
       "       [[0.18919906, 0.26362842],\n",
       "        [0.96476798, 0.37230784],\n",
       "        [0.45283844, 0.14503324],\n",
       "        [0.01276874, 0.13518748],\n",
       "        [0.85953952, 0.20941405]]])"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Example 1\n",
    "array = np.random.random(20) #.astype(np.float32\n",
    "array.reshape((1,20,1))\n",
    "\n",
    "# Example 2\n",
    "array.reshape((4,5,1))\n",
    "\n",
    "# Example 3\n",
    "array.reshape((2,5,2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* use numpy to store and manipulate image data. \n",
    "* We will use the Python Imaging library (PIL) to read and write data to standard file formats.\n",
    "* 4 lines of pixels, each line of pixels contains 5 pixels, and each pixel contains 3 bytes (red, green, blue)\n",
    "* The shape of the array is: height x width x 3."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[107, 237, 128],\n",
       "        [187, 103, 176],\n",
       "        [226, 134, 158],\n",
       "        [156, 241, 215],\n",
       "        [252, 143, 117]],\n",
       "\n",
       "       [[168, 227, 125],\n",
       "        [110, 135, 233],\n",
       "        [204, 123, 140],\n",
       "        [127, 160, 239],\n",
       "        [180, 133, 236]],\n",
       "\n",
       "       [[226, 174, 182],\n",
       "        [173, 196, 164],\n",
       "        [123, 109, 233],\n",
       "        [192, 126, 139],\n",
       "        [167, 161, 134]],\n",
       "\n",
       "       [[234, 235, 182],\n",
       "        [128, 128, 110],\n",
       "        [225, 118, 170],\n",
       "        [144, 181, 196],\n",
       "        [244, 153, 144]]], dtype=uint8)"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# IF it is image data:\n",
    "# Create an RGB image that is 3 dimensional arrays of 8-bit unsigned integers.\n",
    "width = 5\n",
    "height = 4\n",
    "RGB = 3\n",
    "p = width * height * RGB\n",
    "img_data = np.random.randint(100,high=255, size=p, dtype=np.uint8) # Generate values in (100,255)\n",
    "img_data = img_data.reshape((height, width, RGB))\n",
    "img_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "img = Image.fromarray( img_data)\n",
    "img.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### [Learning Objective 3] Stock data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "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>Adj Close</th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2013-01-02</th>\n",
       "      <td>257.31</td>\n",
       "      <td>256.08</td>\n",
       "      <td>258.10</td>\n",
       "      <td>253.26</td>\n",
       "      <td>257.31</td>\n",
       "      <td>3271000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-01-03</th>\n",
       "      <td>258.48</td>\n",
       "      <td>257.27</td>\n",
       "      <td>260.88</td>\n",
       "      <td>256.37</td>\n",
       "      <td>258.48</td>\n",
       "      <td>2750900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-01-04</th>\n",
       "      <td>259.15</td>\n",
       "      <td>257.58</td>\n",
       "      <td>259.80</td>\n",
       "      <td>256.65</td>\n",
       "      <td>259.15</td>\n",
       "      <td>1874200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-01-07</th>\n",
       "      <td>268.46</td>\n",
       "      <td>262.97</td>\n",
       "      <td>269.73</td>\n",
       "      <td>262.67</td>\n",
       "      <td>268.46</td>\n",
       "      <td>4910000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-01-08</th>\n",
       "      <td>266.38</td>\n",
       "      <td>267.07</td>\n",
       "      <td>268.98</td>\n",
       "      <td>263.57</td>\n",
       "      <td>266.38</td>\n",
       "      <td>3010700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-01-09</th>\n",
       "      <td>266.35</td>\n",
       "      <td>268.17</td>\n",
       "      <td>269.50</td>\n",
       "      <td>265.40</td>\n",
       "      <td>266.35</td>\n",
       "      <td>2265600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-01-10</th>\n",
       "      <td>265.34</td>\n",
       "      <td>268.54</td>\n",
       "      <td>268.74</td>\n",
       "      <td>262.30</td>\n",
       "      <td>265.34</td>\n",
       "      <td>2863400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-01-11</th>\n",
       "      <td>267.94</td>\n",
       "      <td>265.10</td>\n",
       "      <td>268.43</td>\n",
       "      <td>264.11</td>\n",
       "      <td>267.94</td>\n",
       "      <td>2413300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-01-14</th>\n",
       "      <td>272.73</td>\n",
       "      <td>268.00</td>\n",
       "      <td>274.26</td>\n",
       "      <td>267.54</td>\n",
       "      <td>272.73</td>\n",
       "      <td>4275000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-01-15</th>\n",
       "      <td>271.90</td>\n",
       "      <td>270.68</td>\n",
       "      <td>272.73</td>\n",
       "      <td>269.30</td>\n",
       "      <td>271.90</td>\n",
       "      <td>2326900</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Adj Close    Open    High     Low   Close   Volume\n",
       "Date                                                          \n",
       "2013-01-02     257.31  256.08  258.10  253.26  257.31  3271000\n",
       "2013-01-03     258.48  257.27  260.88  256.37  258.48  2750900\n",
       "2013-01-04     259.15  257.58  259.80  256.65  259.15  1874200\n",
       "2013-01-07     268.46  262.97  269.73  262.67  268.46  4910000\n",
       "2013-01-08     266.38  267.07  268.98  263.57  266.38  3010700\n",
       "2013-01-09     266.35  268.17  269.50  265.40  266.35  2265600\n",
       "2013-01-10     265.34  268.54  268.74  262.30  265.34  2863400\n",
       "2013-01-11     267.94  265.10  268.43  264.11  267.94  2413300\n",
       "2013-01-14     272.73  268.00  274.26  267.54  272.73  4275000\n",
       "2013-01-15     271.90  270.68  272.73  269.30  271.90  2326900"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import yfinance as yf\n",
    "from yahoofinancials import YahooFinancials\n",
    "AMZN = yf.download('AMZN', \n",
    "                      start='2013-01-01', \n",
    "                      end='2019-12-31', \n",
    "                      progress=False)\n",
    "# AMZN = yf.download('AMZN') for all \n",
    "all_data = AMZN[['Adj Close','Open', 'High', 'Low', 'Close', 'Volume']].round(2)\n",
    "all_data.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Date'>"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD8CAYAAACb4nSYAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAA3dUlEQVR4nO3dd3gc1fXw8e9RL5bk3m1kG+OGjYsAGzCYUExLgJBQQwtgCC0JJS8QCARCIA1+EEpiSigBA6EEAwZsDKG6IFfcO7ZsWbYsW8Xq0nn/mNnVrrTqK+2udD7Po8czd2Z2jtarM3fv3LlXVBVjjDGdQ1SoAzDGGNN+LOkbY0wnYknfGGM6EUv6xhjTiVjSN8aYTsSSvjHGdCIxoQ6gMT179tT09PRQh2GMMRFjyZIluaraK9C2sE/66enpZGZmhjoMY4yJGCLyfX3brHnHGGM6EUv6xhjTiVjSN8aYTsSSvjHGdCKW9I0xphOxpG+MiSi780upqKoOdRgRy5K+MSZi5BaVMfmh+fxt7oZQhxKxGk36IjJIRD4TkTUislpEfumWdxeReSKy0f23m1suIvK4iGwSkZUiMtHntS53998oIpe33a9ljOmI1uwqAGDZ9v0hjsTf7vxSVu3MD3UYTdKUmn4lcKuqjgYmAzeIyGjgDmC+qg4H5rvrAKcDw92fGcDT4FwkgHuBo4GjgHs9FwpjjGmKzXuLABjaK5miskqe/GwTlSFu6tmYU8jkh+Zz1t+/CmkcTdVo0lfVbFVd6i4XAmuBAcDZwIvubi8C57jLZwMvqWMh0FVE+gHTgXmqmqeq+4F5wGnB/GWMMR3b1tyDAKQmxvLkZ5v4y8fruWnWspDG9OXGXO/yyqwDoQukiZrVpi8i6cAEYBHQR1Wz3U27gT7u8gBgh89hWW5ZfeWBzjNDRDJFJHPv3r3NCdEY04F5avqLt+aRFBsNwIacwlCGRLfkWO/yj574OoSRNE2Tx94RkS7AW8CvVLVARLzbVFVFJGiT7arqTGAmQEZGhk3ia4wBoKi0EoBl2w+wZa9T64/yyUWhkBAT7beuqkiIY2pIk2r6IhKLk/BfUdW33eIct9kG9989bvlOYJDP4QPdsvrKjTGmSSqqauqA+SUVAFRpaOuFJRVVfuulFeHdnbQpvXcEeA5Yq6qP+GyaDXh64FwOvOtTfpnbi2cykO82A30MnCoi3dwbuKe6ZcYY0ySV1XUTanytmnawFZZW8MwXW+p9NqB2ki8ur2zTeFqrKc07xwKXAt+JyHK37C7gYeANEbkK+B443902BzgD2AQUA1cCqGqeiDwAfOvud7+q5gXjlzDGdA6VVXVr9WW1atrBdt2/l/D1pn2IwNVTh9bZXrumX3s93DSa9FX1K6C+BqqTAuyvwA31vNbzwPPNCdAYYzwqq+sm/eLytkuyG3IK+XrTPgAWbc0LmPS/yzpASnwMD5xzOL96fTklbRhPMNgTucaYiBGoT35bNqeUV9acr7C0os72gtIKZq/YxY8nDqB7chwA+4vr7hdOLOkbY8JeQWkFm/cWse9gOVcdN8RvW1s2p2Ruq2mBPljmf57yymrW7CqgWmHKsB70TUsAYHdBaZvFEwxhP12iMcb89OkFrHf744/om+K3raJKqaiqJjY6+HXY+95b413OL6mgqlqJjnJau6968Vvvg1mDuid5k35OfngnfavpG2PC3nqfB7BG9k3h6CHdAbjk6MFA+9w83Z5XzAl/+cy77vsk7qDuSaTEx5AUF012mCd9q+kbY8LaxlpP3I7ql8rzVxzJrgMlLHabX0rKq0hNiA10eKsM65XMZvchMICs/SXe5V4p8ewtLAPwnrtvagI5heGd9K2mb4wJa1kHahJtTJQQGx1FcnwMw/ukkOImW8+DWsE2pn8ag7sncXWt+wiqyv6D5XX2T0mI4WBZePfTt6RvjAlr6vPEbZ/UBL9tg7olArBoy742OffGPUUM6ZlMaqL/t4jVuwq83Ud/MLK3tzwpLjhJf9XO/Da7kFnSN8aErUfmbWDemhwA/vyTcbx81VF+29N7JANwz7urg37unQdKWJtdQFpiLKkJNS3h+cUV3P7mSgCevSyDpy7xThlCcnx0nV4+zVVUVslZf/+K+2YH/3cCS/rGmDD2+PyNzFrsDM47fUxfhvbq4re9m9s3flS/1KCet6paecDtuTNxcFe/mv5j8zeyNtuZzOWkUb1JiK0ZBiIpLqbVzw14BpV7Z1nbDE1mSd8YE5Z8a7op8TGkJQa+UTukZzLDe3cJuK2lHnh/DR+t3g3Aj8YPIDm+pqbfNSmWhFgnddYeTTM5PpqDPk/k5hdXcNc73/G3ueubfG7fB8LagiV9Y0xYeuGbbd7lwgbayaOjhNkrdtXp5dMaby3N8i53T44jNromuT8ybwPVCudNHFjnuOS4GPYWljHnO2eqkXtnr+LVRdv5+6ebmtzWX1bpXDROGtnb735GsFjSN8aEneYkuxj3YalTHv2i1ectKa9i9a58Ct0mlnUPOJP71R6zv7yymu7Jdb95JLnfCK5/ZSlnPPYlvr/Fg3PWNnjenW4vpTK3pn/BkYPaZFx+S/rGmLCzdPsBv/XJQ7vXu6/nCdlg+PXryznz8Zq5bj3t9ZOH9uDkUX389u2eHF/n+BSfZqA12QW8u3yXd/3VRdvrPe+Nry7l2Ic/pbpavTX9+Ni2GTLakr4xJuys3+3fVPOjIwLOrApAMCvD32yuecr2tlMP8y4nxEbz7OUZ3HJKTVmgmn5qYtOfd/1+30HOfuIrduQVM3+dMwfVnsIyb00/PqZt0rMlfWNM2KndA6ahYXU8TTHB4Dsz17QRvetsP3dCzcWnS3zdpJ8S4Kng1IQYRvVLZUBX55mC6mrnwa5/L/yeFVn5vLp4u7dLaNb+Yi5+ZhEQ3G8wvizpG2PCTqk7ls7lUw4BGh6u+Pt9xd7lN77d0arzlvsM3ewZQM2X71APvjd3PbrE163pF5RW8uMJA9h5oIRtuQd59JMNTHhgHs98uRWAz9bt8TblfOz2GALo1aVu81EwNGW6xOdFZI+IrPIpe11Elrs/2zwzaolIuoiU+Gz7h88xk0TkOxHZJCKPSzjPHGyMCani8iqio4TJQ3sAeGvJjfnNWyu9y3kHy/nD+2vqneYwkCqfSVq6J8XV2d7F5yGt2ADNLzEBaudDeyUzur/zHMGO/cU8/9VWv+3rdhd6x/DZmutcwM4e35/0nslNjrs5mlLTfwE4zbdAVS9Q1fGqOh5nwvS3fTZv9mxT1et8yp8GrgGGuz9+r2mMMR7F5VUkxUZz+th+vH/TcZw1rl+zX+P3763m2a+28qnbXt5cUQESuG+TS3yANqfkADX9py+Z5B2y4devL/frx1/bp+ucp4+vCTBDV7A0mvRV9Qsg4Fy2bm39fGBWQ68hIv2AVFVd6E6n+BJwTrOjNcZ0aB9+l83BskpKyqtIjHOaPA4fkNasrosTH5jHrgMlFLhj1wRqhqnP2AFpALxz/TGN7hsTIOmPG5jGn84b69eHf2ivZO+MX7lF/oO0+d4YBqhW6NklnjH9g/uEsa/WtulPBXJUdaNP2RARWSYin4vIVLdsAJDls0+WWxaQiMwQkUwRydy7d28rQzTGRIJVO/P5xStLuW/2avYdLPNOP9hceQfL2bSnyKcXTNO7PkZFCScc1osJg7s1um+g9nsR4YIjB/O3848AnHsSsdFRfs1Go/ulEh8TxbaHz+TSyYfUeY3corI26Z/v0drx9C/Cv5afDQxW1X0iMgn4r4iMae6LqupMYCZARkZG8B9JM8aEnf3FTi04a38JO/YXc8TAri1+rcrq6hZ1fdxXVEZ6j6Qm7ds1qeHx+7c+dIZ3efKwHt7lNdkFTB3eE6gZOwigW1Is+4sruOKY9CbH2xItTvoiEgP8GJjkKVPVMqDMXV4iIpuBw4CdgO8zywPdMmOMAeD/uSNXRkVBdn4p54xv+Y3Miir19gAKVGuurlaW7djPpENqHvqqqlZ255c2+aZxSkLD6dP3vLUneJk+pq93ef0fTuPlBd9z2ZR0oqTtump6tKZ552Rgnap6m21EpJeIRLvLQ3Fu2G5R1WygQEQmu/cBLgPebcW5jTEdzC53msGvN+2jqloZ0K1pyddTa/ZVWaXemn51gCEdXlywjfOeXsBnPjd59xSWUtmM8ybHNa/O/LPJg73LvqOCxsdEc/XUocTFRBETHdWmTTvQtC6bs4AFwAgRyRKRq9xNF1L3Bu7xwEq3C+ebwHWq6rkJfD3wLLAJ2Ax82PrwjTEdQUmAHi0Dm5h8X77qaN678Ti/svKqKu9wBrlud0iPiqpqNriDs3nGu1FV79y2PQIMrxBIoN49Dbn7zNHe5ba8UduYRi9VqnpRPeVXBCh7C6cLZ6D9M4HDmxmfMaYTmPLw/Dplw3q1fLjk4vIqSiucmv4vXlnKtofP9G4b/tua+qZnILWZX2zhoQ/XAZAU1zZj3iTERrPk7pNZk13gNwZ/e7OJ0Y0xIXeg1hO3g7sn0b+JbetQd/ydkvIq7wNPvjzt/LWPe2x+TQfExsaz/9eVR7J5T1GTY/PVo0s8U4f3atGxwWJJ3xgTFrrEx1Dkjjl/8dGDG9m7YUX1jF3v6SHk4WmhKfZpXvKd8zaQE0f05sQA4/JECht7xxgTUp6292uPr3kKNa6hEdYCqN0ts6jWIGw78pzhDf633v+5ny25B+u8VnPb6iONJX1jTEjtyHNupg7olsj/XTAegFPH9GngiLqG90nxWz9YXkmyT9v8c+54N3e+/Z3ffv/8fEtzw4141rxjjAmpLXud9vGhvbowflBXzplQ/9j5TVVYWukd7wacbwKV9Qy8tnDLvlafL5JYTd8YE1Lb3aaXpj4J2xQHyyr9+uf3TUugoJ5x9y+cuTBo540EVtM3xoRU1v4SUuJjSEtseFiD5vhs/V58m+ajRDhQ6yZuZ2U1fWNMSHy+YS85BaXMX5fDgG6JrX4StfZomtUKVx6bDjgPZDU0EcvQnslMGdqDf/xsYqtiiARW0zfGtJnMbXmM6JtSZxrB7PwSLn9+sXf9jtNHtvpcsdFRVFT598P3jLBZVa3klzg1/ScvnsgNry712+/DX01t1mickcxq+saYNpFbVMZP/rGAsffNRWuNf3Pdy0v81i+fkt7q88UG6OYZ53blrKxW7wNgY/qn+vXsOWlk706T8MGSvjGmjazLLvQuP/D+Wu9yVbWyIivfb9/EIAx94En6vv39E2KjEHGexPUk/a5JsXz4y+O9D4CdO7H1vYUiiSV9Y0yb2Lav5sGnN5fUTFj+3Fdt0zfe84BWqs8N4ZgoITnOedL3gDuTVkpCLIN7JPHHc8ey9v7TOGtc/zaJJ1xZm74xpk0Ul9d0kfR0lyytqOJPH60HYPygrhx3aE/vzdbW8tzI9e0FNLJvKsnx0RwsqyQhNpq46Ci/8eqD8Q0j0ljSN8a0ieJawyXvPFDC7/67yjt14Nu/OCaoQx542u99JzdJjo8mp6CMNzKzmHH80DafoCQSWPOOMaZN7HLHqvfYnV/KfHfSkm5JsUEf48bTpu97QzcxtuYCsGz7fkoq6o7b39lY0jfGtIms/SVMGNzVu57oM4Z8ZVXwp74OlPST42vO+e22/UE/ZyRqysxZz4vIHhFZ5VN2n4jsFJHl7s8ZPtvuFJFNIrJeRKb7lJ/mlm0SkTuC/6sYY8LFrMXb+WbzPvqn+Y+J72leKaxn6OPW+P2PxjB2QBpHpnfzliU1c0rDzqApNf0XgNMClD+qquPdnzkAIjIaZxrFMe4xT4lItDtv7pPA6cBo4CJ3X2NMB+QZzXKwz3g6ldXVHD2ke32HtNoRg7ry3k3H+d3ITYqL5tDeLZ+BqyNqNOmr6hdAXmP7uc4GXlPVMlXdijMf7lHuzyZV3aKq5cBr7r7GmA6mymd0y94pNfPNVlSpXxNPW/EdziExNpoLMga1+TkjSWva9G8UkZVu84/n+9QAYIfPPlluWX3lAYnIDBHJFJHMvXv31rebMSYM7S4o9S4f4lvTr6qmojr4bfkNiYoSYqKtx46vlib9p4FhwHggG/hbsAICUNWZqpqhqhm9eoV2PkljTPNku712Th3dhxNH9ObvF00AnKEQyivbp/fM/26bxtJ7TgEgxufG7r+uPLJdzh/OWnSXQ1VzPMsi8gzwvru6E/D9LjXQLaOBcmNMB7Ir36np3zZ9BCLineC8oqq60UnHgyW9Z7J3Oc6npn9I9+CN2R+pWlTTF5F+PqvnAp6ePbOBC0UkXkSGAMOBxcC3wHARGSIicTg3e2e3PGxjTDhSVW6etQyAAW6y9zwpW1mllLVT0vcVE1WT5ronx7X7+cNNozV9EZkFTAN6ikgWcC8wTUTGAwpsA64FUNXVIvIGsAaoBG5Q1Sr3dW4EPgaigedVdXWwfxljTOi8umi735j2yfFOevEk3atfyvRuu/vMUe0Wl2+bfjAnaolUjSZ9Vb0oQPFzDez/IPBggPI5wJxmRWeMiRh3vVMz6bhvcq2uNaxyr5R4rp46lPYS59Om39qJWjoCeyLXGBN09589xrtcWmvogz6p8bV3b1MxAcbZ78zs3TDGtFp1ra6YyT5Pwk4Y3M1v29OXTGqXmDysy6Y/S/rGmFbz7ZsPNe35QJ2RLQe1cw+aOKvp+7F3wxjTasu2H/BbT00MnzFvYmw4ZT+W9I0xrZad7z+McmpC4F4y7990XHuE48fa9P3Zu2GMabWiWqNmpiUFTvq+D021F2ve8WfvhjGm1Q6WVZLkM/Vgl1pDGnv67yeHYHrCxDhLc77Cp+HNGBOxnvlyq9967VmxPvzlVJZ+fyAk/eR7pSS0+znDmSV9Y0ybO7R3Cof2TgnJuVMTLM35su89xphWG9Enhelj+oQ6jIDsKVx/dgk0xrRaWWUVCbHR3H3mqDpP4IaDT245nkSbOhGwpG+MCYLC0kqS4mLadUyd5ghV01I4suYdY0yrrM0uYN/Bcvql2Q3TSGBJ3xjTYkVllZz+2JcA9E21pB8JLOkbY1rsQHG5d7lXO4+eaVrGkr4xpsUOljk3bc8a148Thtt81pGg0aQvIs+LyB4RWeVT9hcRWSciK0XkHRHp6pani0iJiCx3f/7hc8wkEflORDaJyONi/aiMiXgHy53hF86bNLDOA1kmPDWlpv8CcFqtsnnA4ao6DtgA3OmzbbOqjnd/rvMpfxq4Bmfe3OEBXtMYE2GK3Zp+snWHjBiNJn1V/QLIq1U2V1U9IywtBAY29BruROqpqrpQVRV4CTinRREbY8JGntum3yXekn6kCEab/s+BD33Wh4jIMhH5XESmumUDgCyffbLcsoBEZIaIZIpI5t69e4MQojGmLSzbvp+E2CgO7d0l1KGYJmpV0heR3wKVwCtuUTYwWFUnALcAr4pIanNfV1VnqmqGqmb06mU3h4wJV5nb9jN+UFfiYqxPSKRo8f+UiFwBnAVc4jbZoKplqrrPXV4CbAYOA3bi3wQ00C0zxkSwzXuLGNWv2fU6E0ItSvoichrwG+BHqlrsU95LRKLd5aE4N2y3qGo2UCAik91eO5cB77Y6emNMyBSXV1JcXkVvG7o4ojSly+YsYAEwQkSyROQq4AkgBZhXq2vm8cBKEVkOvAlcp6qem8DXA88Cm3C+AfjeBzDGhLncojKWbd9fs17o3MTt2SUuVCGZFmj0lruqXhSg+Ll69n0LeKuebZnA4c2KzhgTNq564VtWZOWz4t5TuXnWMu9MWT1T7EncSGL9rIwxTbIiKx+AD1Zm8/mGml516T3af95b03J2y90Y06iKqmrv8l3vfOe3Lb1HUnuHY1rBkr4xplEHiisClv/x3LE2M1WEseYdY0yj8g6W+63/57opHJnePUTRmNawmr4xpkEVVdWc9/Q3fmWTBncLUTSmtaymb4xp0PDf1u1dbSNqRi5L+saYepWU+09y/uVvTvR21TSRyZK+MSYgVWXU7z4C4JWrj+aYYT3spm0HYG36xpiANuQUeZcnHdLNEn4HYUnfGBNQblGZdzkh1pp0OgpL+saYgKqqFYB/XjopxJGYYLKkb4wJqNoZMZ2eXWxsnY7Ekr4xnYiqkl8S+Onauvs6/0Zb98wOxZK+MZ3I7BW7OOL3c1m/u7DRfT3NO5bzOxZL+sZ0InNX5wCwamd+o/t6mneirNdOh9KkpC8iz4vIHhFZ5VPWXUTmichG999ubrmIyOMisklEVorIRJ9jLnf33ygilwf/1zHG1KessooPvssGaNKctm5F35J+B9PUmv4LwGm1yu4A5qvqcGC+uw5wOs40icOBGcDT4FwkgHuBo4GjgHs9FwpjTPBVe7K2a2VWTe2+uLyy8eM9NX1rD+hQmvTfqapfAHm1is8GXnSXXwTO8Sl/SR0Lga4i0g+YDsxT1TxV3Q/Mo+6FxBgTBPe+u4qhd83ho1XZLN9xAICH5qz1bs8tKq/nSMfXm3K5/pWlAERbTb9Dac01vI874TnAbqCPuzwA2OGzX5ZbVl+5MSaI1u8u5MUF3wNw3b+Xcs6TX/OfzB0s3X7Au8/f5q73+ybwv/V7uPS5RagqZZVVXPLsIu82exK3YwnKFzdVVUAb3bGJRGSGiGSKSObevXsbP8AY4/X4pxvrlN3+5koAfjppIOC011/y7CLKKp0B1W56dRlfbsxlR14JV7+Y6Xes9d7pWFqT9HPcZhvcf/e45TuBQT77DXTL6iuvQ1VnqmqGqmb06tWrFSEa0/lszCnkuEN7Btz2h3MP58IjnT/DBVv2cdfbqzj83o8pLHPa+OesymaZzzcCsH76HU1rkv5swNMD53LgXZ/yy9xePJOBfLcZ6GPgVBHp5t7APdUtM8YEyX8yd7Ahp4gpw3oE3B4fE83D543z9t55a2kWRWU1N3Uf/nAdAFcck+4tS7ShlDuUpnbZnAUsAEaISJaIXAU8DJwiIhuBk911gDnAFmAT8AxwPYCq5gEPAN+6P/e7ZcaYIFi8Nc/bjPODkb29E5ZfkDGozr7lldV1yjyKyipJjIvm0smHMHFwV3qnJLRNwCYkmjSevqpeVM+mkwLsq8AN9bzO88DzTY7OGNMkc1fvZsbLS7zrI/um8Nlt0yitqOZASTmvZ+7grjNGerc/dclEb+8cgI9+NZWVO/L5zVvORePMsf04fEBa+/0Cpt3YJCrGdAD3zV7tXe6aFOvtcZMYF01iXCLL7jmFtMRY7z5njO3HtofP5IJ/LmB7XjGHdE8mNcHZnpYYawm/A7Okb0wHcMAdRO3PPxnH+QGac7olxwU87vVrp6CqiAiJcYn85rQRHD0k8P0A0zFY0jcmwlVWVVNRVc2Vx6YHTPiN8e2Hf/20Q4MZmglDlvSNiVCPfbKRRz/ZwHkTB1JRpYzulxrqkEwEsFE1jIlQj36yAXC6XQJMG9E7lOGYCGFJ35gI5Dt/LcA1U4fQK8VmuDKNs+YdYyLQgs37AOidEk+/tAR+e+boEEdkIoUlfWMi0M4DJQDMv/UEUhJiG9nbmBqW9I2JAG8uyWJfURlj+qdx//ur2ZBTRGpCjCV802yW9I1pY4WlFcxbk8Pph/dr0Tg2ewpKue0/K+qU90tLDEZ4ppOxpG9MGxt731wAbnljBT8Y2ZtbTjmsSU+8XvzMQr5x2+4DOXNcv6DFaDoPS/rGtKHv9x30W/903R5yCkr54OapDR5XWlHll/CPGJjGuzceB0BVtVJcXmlNO6ZFrMumMW1oS66T9KeP6eMta2yi8RU7DnDuU9/4ld02fYR3OTpKLOGbFrOkb0wb+rc7beFtp9Yk7cba9Wd+uYW12QV+ZZMO6Rb84EynZM07xrSh+eucCeV6pybw8a+O5+wnv2pwLPt9RWV8sNKZevr26SMY2C2R3KJykuLsT9UEh32SjGkDv3xtGX1SnclHBnZLJC0xlrTEWJLiYli+4wA/+Ov/uP/sw9lfXM4Pj+jvPW7RVmdeoZ8fO4QbTrTBz0zwtTjpi8gI4HWfoqHA74CuwDWAZ0bzu1R1jnvMncBVQBVws6radImmw9iRV0xSXDRFZZW8u3yXt/y3Z4zyLucdLAectv6fPbcIgFH9Uli2/QCTh/Zg6ff7Abjm+CHtGLnpTFqc9FV1PTAeQESicSY5fwe4EnhUVf/qu7+IjAYuBMYA/YFPROQwVa1qaQzGhIP8kgqO+P3cerf7tsf/9oxRPDhnrd/2kx/5os4xdqPWtJVg3cg9Cdisqt83sM/ZwGuqWqaqW3Hm0D0qSOc3pt1tzT3Iku/zeP3b7fXuM7RXMr1Ta+aYvfLY9Ca9drJNRm7aSLDa9C8EZvms3ygilwGZwK2quh8YACz02SfLLTMmIp341//5rd8+fQTrdheyYPM+3r/pOGav2Ml5Ewf67RMTHcWJI3rx2fq9fuWpCTGcMKI3763YRUJslN/EJsYEU6uTvojEAT8C7nSLngYeANT992/Az5v5mjOAGQCDBw9ubYjGBN2OvOI6ZbVvvM44fljAY5+9/EiG3TUHcCYwX7e7kC9+cyJdk+I4b+IAjhjYNejxGuMRjJr+6cBSVc0B8PwLICLPAO+7qzsB37ncBrpldajqTGAmQEZGhgYhRmNaZfmOA5zz5NcALLjzB94JTABE4O5mDG0cHVVTi3/92ims311I1yRnDlubCMW0tWAk/YvwadoRkX6qmu2unguscpdnA6+KyCM4N3KHA4uDcH5j2syDH6xBRJj5xRZv2ZSHPgVgRJ8UPvzlVKKiWt4Uk5YYy1FDurc6TmOaqlVJX0SSgVOAa32K/ywi43Gad7Z5tqnqahF5A1gDVAI3WM8dE+6e+XJrvdumDu/Z4oT/yS0nkFNQ2tKwjGmxViV9VT0I9KhVdmkD+z8IPNiacxrTHF9u3MuArokM7dWl2cdWVfu3LH5x+4nkFJby038sAOCYQ3sEOqxJDu3dhUN7Nz8mY1rLxt4xHdqlzy3mB3/7vEXHvu1OOA5wzLAeDOqeyJHp3bn5pOEADOyWFJQYjWlPNgyD6RCy9hezaEset/pMNnLVcS1/qrW4vJLb31wJwGMXjufs8TW9i3998nDOGd+/Rd8ejAk1S/qmQzjlkS8oqfC/RfTcV/W3xzdmW25Nl8wTR/r3qBERS/gmYlnzjol4O/KK6yT82vKLK5r1mtvzaiY/SbUhEUwHYknfRLz1uwv91j+99QTv8pMXTwTg+zz/GawaU1zuXEQeveCIVkZnTHixpG8i3m6fro+nH97XO6Tx8Yf1Yngfpxlm054i8osr+OOctVRU1T+ePcC7y3dyyxvOvYGjh7S8h44x4cja9E1EKy6v5O7/Os//rfr9dLrEOx/p9248jkN7dyEqynkCdtXOAr7dtp9Zi7czql8K504YGPD1VJVfvrbcux4TbWPgmI7Fkr6JaCt25APOk62ehA8wdmCad7mqWnn+661kuEMcV1bVP7JHTkGZ33pslH0ZNh2LfaJNRPti416io4T/3Tat0X0z3QlKGvKfzB1+61bTNx2NJX0T0eavzeGYYT3olhxX7z63Tx/ht15fPT/9jg/42zxnILWzxztTGCbE2rj2pmOxpG8iWt7BcgZ1b/jJ2Olj+vqtl/lMTO6ZvrB2D6BHzx/PugdOIzba/kRMx2Jt+iZi5RSUkltUTnxMw4m5W5J/P/ttuU73zXlrcrjmpUyOGJjGiizn3kCUwKK7TiYqSkiIslq+6Xgs6ZuIdfEzzkRs3++rO6GJrx5d4v3WPfuv310A4E34MVHChj+c3qqhko0Jd5b0TcTavNepsZc28jQuwOe3TwPg1jdWcLCskg9WZvPXuRv89jlldB9L+KbDs6RvIsp1Ly+hsrqagtJKb9lDPx7b6HGH9EgGICUhhtyicr7ZnOvdds3UITzz5VbOGNsv+AEbE2Ys6ZuwU1Bawbj75gJw9XFDGN6nCxccOZicglI+Wr3bb9/HL5rgTehNsTIrn30Hy1mb7TTtbP7jGURHCXecPspvGkNjOqpgTIy+DSgEqoBKVc0Qke7A60A6zuxZ56vqfhER4DHgDKAYuEJVl7Y2BtOxeNrqAZ51R8o8P2MQR/9xfp19f3RE/2a99j63t06lO0GKJ9FbwjedRbD6o52oquNVNcNdvwOYr6rDgfnuOjiTqA93f2YATwfp/KYDCTQ0zpA759Qp2/Tg6c1+7d+eMcq7fPiA1GYfb0yka6tOyGcDL7rLLwLn+JS/pI6FQFcRsYZU4ychNoqpw3uy7eEz+fXJh/lte+sXU7zLMS3oQ5+SUPPl9qmLJ7U8SGMiVDCSvgJzRWSJiMxwy/qoara7vBvo4y4PAHyfc89yy4zxKquoJs5N6KmJ/i2Qkw7p3qrX7urTZ39wD5vu0HQ+wbiRe5yq7hSR3sA8EVnnu1FVVUTqH+EqAPfiMQNg8ODBQQjRRIpVO/NZk13AGvdGa3ef4RVmXTMZgH/8bBLV2qyPlNeEwd1aH6QxEazVSV9Vd7r/7hGRd4CjgBwR6aeq2W7zzR53953AIJ/DB7pltV9zJjATICMjo2V/3SYivbp4OwA/nuh8AezfNRFwRtGcMswZ2/60w/sGPrgJ+qQmkJYY2+wbwMZ0FK1q3hGRZBFJ8SwDpwKrgNnA5e5ulwPvusuzgcvEMRnI92kGMoZXFzlJ/5HzxwM1ST+/pHnTHTZkxb2n8sA5hwft9YyJJK2t6fcB3nF6YhIDvKqqH4nIt8AbInIV8D1wvrv/HJzumptwumxe2crzmw6kMkC3nb6pCRw1pDtXHJPe/gEZ0wG1Kumr6hagziSiqroPOClAuQI3tOacJjL9b/0evtiQyz1njcKtJHh9siaHFxds46YfDAfgbz+t+UhFRwlvXDsFY0xw2BO5pl08NGcd63MK2V1Qwvf7ijln/ABe+3Y7USJs3FMEwJcbnaERjh7auh46xpj6WdI37aJ/1wTW5xQy5ztnGIXVuwrq3XeA245vjAk+S/qmXXRLqjuzVVx0FBXV1fRLTeCVayYTJZAYG12n+ccYEzyW9EPsyc828ci8DfzriiPJ2l/CE59uJCO9O/f+cHSdceAjWXlVNUN6JvPYheMZ1S+VaBEbxtiYELCkHwJrswt4e2kWP80YxF8+Xg/AZc8v9m6fvWIX+SUVvPjzo0IVYoup+9BU7dr6t9vyqFYYN7BrCKIyxnhY0g+BR+dtYO6aHJ75cmu9+3y+Ya/felW1UlRaSVqtqf/CxVcbc7nj7ZVk7S8B8M4vu6+ojKPc0TEHdrO2emNCzZJ+O9uYU8iCLfu8612TYll2zymICKUVVXy4Kptfv76Ca48f6t1n1uLt3Pn2dwBs+MPpxDUyJ2x7u/fdVby44Hu/spH3fFRnv7t8Rrg0xoSGJf128Nm6PVz5wrdcfdwQ7/jw71x/DOMGdiVKappCEmKjOWtcf379+grmrcnhhBG9KCqt9CZ8gA05hRw+IK1Z58/aX8za7EJOGd2n8Z2bKe9guTfhx0YLj5w/nptmLfPb57A+XTj98H5MH9Py4ROMMcFhSb8dXPnCt0DNhCD90xLqHfgrxr25uSX3IBc/s8hbfvMPDuXxTzfxs+cWsfx3pzb53N9szvW+zpybpzK6f3DHkP9PpjNo6hMXT+Cscc54Nsnx0XSJj+VAcTkj+6baaJbGhJHwaifooI4e4v+w0ee/ObHefQN1V/z9j8Zwy6kjADhQXMEXtdr766OqPD5/o3d9d4HT3l5aUcWPn/qa9Ds+YPaKXY2+Tm5RmfcGbW1b9h6kd0q8N+ED/GBkH44a0p1Tx/S1hG9MmLGk38YqqqpZtDWPvqkJTBvRi5evOorYRib/ePmqozhmWA8eOf8Irj1hKJdNOQSA3501GnB6+ny2fg/pd3zAO8uy/I79ePVu0u/4gP/35kp+9fpyFm7JIy3RuflbWlHNVxtzGXnPRyzdfgCAm2ct443MHVQEGPemsqqaR+auJ+MPn/Dywu/rbAenK2Z8rH2MjIkUUl8NLlxkZGRoZmZmqMNoti17i1i4JY/9xeXebpnbHj6zVa9ZUl7FEffPpbyyJkH3Toln8W9PBpyafaBpBT3GDUwjJkpYuv0A/dMSqKhW9haW+e2z4nenensInf+PBSzelgdAlMCWh+rGf8MrS1m3u4D5t05r1e9mjAkeEVniM32tH2vTb4H84go27S1iW+5Bzps00G9bfYn3g5uPa/V5E+Oi+fN54/jV68u9ZaP7p7I7v5S9hWVc+3Lgi+OsayZz0TMLWZmVD8CNJx7KbdNHUFpRVaeXzRH3z2XK0B68cvXR3oQ/sm8K63YX8v7KXYwdkMa8NTlcfPRgDpZV8cF32Qztmdzq380Y0z6spt9MO/KKmfrnz7zrI/um8N8bjiUhNhpV5c0lWdz+5so6x7W2lu+hqlwwcyG780vZnlcccJ9//Gwipx3uP/XwfzJ3eONa98BpJMRG+23/elMulzy7iNquO2EYU4f3rLMtLTGWy6ccwuOfbgKC9/sZY1qvoZq+NcY207IdB/zW1+0uZOQ9H/H1plz+9fU2b2L9z3VTGNk3BYD3b2p9Ld9DxBlq+IsAN4OnDu/Jn84by7QRvetsO3v8AB6/aAKbHjy9TsIHOPbQngGfAL72+KEce2hPfjFtmF95fkmFN+E/duH4Fv42xpj21uFr+gWlFdz9ziruOH2kdxam5vpoVTYbc4oYMyCVO95y+sy/e+OxxMdE8+Rnm3juq5ona/unJTDj+KFcNiWdwtJKcg+WMaxXlxbH35CNOYXc+Ooy1ucUMqpfKq/NmOy9adtSX2zY6x0SYtY1k71TFALc/94axg5MZcKgbkz76/8AOGpIdxvv3pgw01BNv8VJX0QGAS/hzJ6lwExVfUxE7gOuATz9Cu9S1TnuMXcCVwFVwM2q+nFj52lp0ldV5q7J4dqXl3jLPv7V8Xy/7yB7Csu4+7+rmD6mD/+8NANV5dqXl7AyK5+M9G78/aIJLNi8j/jYKBZtzePPH633vkbXpFhenzGFEW4tXlW58dVlfPCdM+vjh7+cyqh+we0L35Cyyio27SliTP/mPbBVn/LKaj5clc30MX0DfiPwKC6v5K2lO/nxhAEkx9utIWPCSVsl/X5AP1Vd6s6TuwQ4B2dqxCJV/Wut/UcDs3AmTu8PfAIcpqpVDZ2npUn/jcwd/CZA23pto/ulsibbf2z3K45J54VvtgXc/4UrjwzYfALOBcCGBTbGhFqbtOmraraqLnWXC4G1wIAGDjkbeE1Vy1R1K848uW0yjOSB4nLuf28NAP+64siA+9z3Q6fPe+2ED9RJ+E9cPIFtD5/JpgdPrzfhQ+AHq4wxJpwE5Xu5iKQDE4BFwLHAjSJyGZAJ3Kqq+3EuCAt9DsuinouEiMwAZgAMHjy42fF0TYrjnrNGccywngzqnsTWh86goKSSRVv3cfxhvSitqKJrUhxLth/gs3V7uPXUwxg3sCsxUcLZT34NwLJ7TqFbsv/EHzGNPFRljDHhrtU3ckWkC/A58KCqvi0ifYBcnHb+B3CagH4uIk8AC1X13+5xzwEfquqbDb1+e3fZ3LSniNSEGHqnJrTbOY0xJpja7OEsEYkF3gJeUdW3AVQ1x2f7M8D77upOYJDP4QPdsrByaO+26WljjDHhoMXtFeI0YD8HrFXVR3zKfZ8KOhdY5S7PBi4UkXgRGQIMBxZjjDGm3bSmpn8scCnwnYgsd8vuAi4SkfE4zTvbgGsBVHW1iLwBrAEqgRsa67ljjDEmuFqc9FX1KyBQd5V6R/xS1QeBB1t6TmOMMa1j3VGMMaYTsaRvjDGdiCV9Y4zpRMJ+wDUR2QsEnrapcT1xnhmIBJEUK0RWvJEUK0RWvJEUK0RWvK2J9RBV7RVoQ9gn/dYQkcz6HlAIN5EUK0RWvJEUK0RWvJEUK0RWvG0VqzXvGGNMJ2JJ3xhjOpGOnvRnhjqAZoikWCGy4o2kWCGy4o2kWCGy4m2TWDt0m74xxhh/Hb2mb4wxxoclfWOM6UQs6RtjGiQ2JVyHEvFJX0Tqn707zIhImvtv2L/vItLX/Tci/uBFZIyIRMzMNyJyrIgMC3UcTZQY6gCaw5MTIuGzG4pYwz751EdEMkTkZeB34fzHIyJRIpIqIu8DjwOoanWIw6qXiEwQkfk4s56hYX6nX0TGichXwB+AHqGOpzEiMlFE5gKfAmmhjqchIjJZRN4CnhSRU8O9giUiU9yJm34tIinh/Nl1L/ovAneLSPf2jDXikr6bRJ8A/gnMB/oB94lIUmgjC8xN8IVALDBARC6A8Kvti+NR4CXgRVW9JtQxNdHdwJuqeq6q7oTwrOGJSKyI/BOnG97jwMfANHdbWH0WAERkGvAU8DawHvgZ0C2EITVIRE4AnsC5mPYH7hKR6aGNKjARGYrz3n4GHAI8ICJnttf5w+7D1hg3iX4KnKSqLwB/xpmwpTKUcTViJM4YGv8HXOLWQqrDKTm5NY0uwDJVfQlARIaFY0IC78V/KFCkqv/nlp0iIl2BcPx6H48zl/RUVX0fZ5rRUSISE6bf/MYC36rqK8DLOJWWotCG1KCJwNeqOgvnW2ofnJn6+oY2rICOxJlx8AXgVmA5cJaIDGrooGAJyz/o2tyvmYd51lX1bVU9ICKnAJk4tf0/isiokAXp8o3VJ+lsAsqBre7P5SIyONRfP2u/rzgfwKNF5B4R+Rr4C/CCiEwKTYT+fON1E2UuMFVEzhSR/wK34dSib3f3Caf396CqvqqqJe56DFClqpXhcGEN8Fn4EvipiPwOWIrzN/aUiPw0JAHWEiDeDUBXEemnqvtxLlBxwDmhiM+XiPxQRG4Ukclu0bfAIBEZ5Mb6NXAA+HF7xBPyD1tDRKSriHwAzAPOF5Fkt9yTTPcDF6vqKcBBnGTaJ1xi9Uk6GUCBqq4GVgP3Ak+7X/nb/f+gvvdVVQuAJ4GfAHcCFwHZwHkiEnDEvjCI9184NbvnVXU68Cww2ecPLCziVVV1m9A8/9+fA+eKSLdQ1vQDxNoFQFWXA6cB6cD1qjoNJzmdFsrKVX3x4iT9AuBF9z7EIGAZzrfXkHzrE5F+IvIe8BucprF/ich0Vd0CLADOd3ddjzONbHdph84IYZ30gWScts+b3OXjoaYGp6qZquqZnvFDYAJQHII4oZ5YXduBFBF5HecDsATYoKoVIfqDrzdWVX0cmKaqX6hqGfBfnItWqN5XaPi9fR8nMXnamzOBHKCsHeOrrd7PrdusF4Uzf/THwAmhCtJVO9apng2quhjohRMrOM2qKTgVrFCp773dCNwCPAT8R1XPxalgTXO3h+JbXwbwpapOVdUHgMcAz72yL4GxInKUO1f4TuBYVS1t66DCLumLyGUicoKIpLo35mYCbwClOE0P/es5dBJOrbTd2vabEWs3nD+e3TgXpl8AI9qzxtSc99X9yukxCcgC2nUS+ybEO8CNdSVOc86NItIT54bj4cC+MIu3v7ufuBf6ePfQUk95GMYaD3wD3OAeehJOD6k2T0zNiPcoT7yqWq6qn6nqa+6hk4CPQhDrNPe9m49zP8RjH7DRXV6E803kUffbyhhgu7RDh5SwGHvH/cD3BV4FqoHNOFfxX6pqrrvPsThfh75V1X+7ZanA0cAfcRLqraq6IYxizVTVl92ynj7buwBxqpoXRrH6vq/xwBTgrzgX0jZ/X1sQr/e9dctvAYYCw4Ffq+qaMIvX9/2NVtUqEfk3sElV7wuzWH0/t2NwmiP7AhXAjaq6Nszi9b63bvlxOLXqXOBaVd0WylhFJFZVK0TkZmC0ql7nc+wjwECcXjyXqer6towVAFUN6Q8Q7f57GPBvTxnwd+DtWvv+Gqc/dhqQ4JadDJwd5rEm++wbFeaxJrplxwDnRMDnIMWnPDYC4k3yKY8L41i7+nwWEoGhEfDeev7O+gNnhEusPvu8B5zsLvd2/43x/Qy3S8ztebLabxZODf1POO2aP8TpH+7ZHoVTez/Bp6wLTrfHb3HabftHQKyLIyjWdn1fI+29jbR4g/RZGBBh7+3AcIwVpyfR88Bg4EFgBdCtvd5b35+QtOmL8yDFEpy27k04vS8qgBNF5Cjwdsm7z/3xOBO4Hqdf61hV3RUBsa6IoFiXt1esQYq33d7bSIs3iJ+FnW0da5Di9by3WWEW6+/dwxKAK3Da+VNwavz7CYVQXGlweghc6rP+FM7NzSuAJT5Xyr44N2zS3bKzgeMt1siP1eK1WCM13hbEOhA4Cudp9/Ht/d7WiT8kJ4UknN4LnrauS4CH3OXlwE3ucgYwK6RvkMVq8UZgvJEUa6TF28xYXwv1e1v7JyTNO6parKpl6vRPBTgF2OsuX4nzePr7wCycpwFD9ki9xdp2LN62E0mxQmTF28xYl0B4DQkSE8qTizNqn+KMkzHbLS4E7sLpa71V3TZFdS+doWKxth2Lt+1EUqwQWfFGUqy+Qv1wVjXOQE65wDj36ngPUK2qX2k73URqIou17Vi8bSeSYoXIijeSYq0R6vYlYDLOm/cVcFWo47FYLd6OFm8kxRpp8UZSrJ6fkD+RKyIDgUuBR9QZ6yVsWaxtx+JtO5EUK0RWvJEUq0fIk74xxpj2E+o2fWOMMe3Ikr4xxnQilvSNMaYTsaRvjDGdiCV9Y3yISJWILBeR1SKyQkRulUamtBSRdBG5uL1iNKY1LOkb469EVcer6hicx+tPx5lEpCHpgCV9ExGsy6YxPkSkSFW7+KwPxRlbvifO7EYv48yKBM4sUt+IyEJgFLAVeBF4HHgYZ37WeOBJVf1nu/0SxjTAkr4xPmonfbfsADACZ1yValUtFZHhOKM9ZojINOA2VT3L3X8GzsxIf3Cnnvwa+Kmqbm3HX8WYgEI64JoxESYWeEJExuNMFH9YPfudijMWy0/c9TScuXst6ZuQs6RvTAPc5p0qYA9O234OcATO/bDS+g7DGVP943YJ0phmsBu5xtRDRHoB/wCeUKcdNA3IVmcqvEtx5kkFp9knxefQj4FfiEis+zqHiUgyxoQBq+kb4y9RRJbjNOVU4ty4fcTd9hTwlohcBnwEHHTLVwJVIrICeAF4DKdHz1J38oy9wDntE74xDbMbucYY04lY844xxnQilvSNMaYTsaRvjDGdiCV9Y4zpRCzpG2NMJ2JJ3xhjOhFL+sYY04lY0jfGmE7k/wPPx/zVuFpmpwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "all_data['Adj Close'].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 1510 observations in the training data\n",
      "There are 251 observations in the test data\n"
     ]
    }
   ],
   "source": [
    "print(\"There are \"+ str(all_data[:'2018'].shape[0]) + \" observations in the training data\")\n",
    "print(\"There are \"+ str(all_data['2019':].shape[0]) + \" observations in the test data\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## [Learning objective 4] Create the training and test data for RNN/LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1504, 5)"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def ts_train_test(all_data,time_steps,for_periods):\n",
    "    '''\n",
    "    input: \n",
    "      data: dataframe with dates and price data\n",
    "    output:\n",
    "      X_train, y_train: data from 2013/1/1-2018/12/31\n",
    "      X_test:  data from 2019 -\n",
    "      sc:      insantiated MinMaxScaler object fit to the training data\n",
    "    '''\n",
    "    # create training and test set\n",
    "    ts_train = all_data[:'2018'].iloc[:,0:1].values\n",
    "    ts_test  = all_data['2019':].iloc[:,0:1].values\n",
    "    ts_train_len = len(ts_train)\n",
    "    ts_test_len = len(ts_test)\n",
    "\n",
    "    # create training data of s samples and t time steps\n",
    "    X_train = []\n",
    "    y_train = []\n",
    "    y_train_stacked = []\n",
    "    for i in range(time_steps,ts_train_len-1): \n",
    "        X_train.append(ts_train[i-time_steps:i,0])\n",
    "        y_train.append(ts_train[i:i+for_periods,0])\n",
    "    X_train, y_train = np.array(X_train), np.array(y_train)\n",
    "\n",
    "    # Reshaping X_train for efficient modelling\n",
    "    X_train = np.reshape(X_train, (X_train.shape[0],X_train.shape[1],1))\n",
    "\n",
    "    inputs = pd.concat((all_data[\"Adj Close\"][:'2018'], all_data[\"Adj Close\"]['2019':]),axis=0).values\n",
    "    inputs = inputs[len(inputs)-len(ts_test) - time_steps:]\n",
    "    inputs = inputs.reshape(-1,1)\n",
    "\n",
    "    # Preparing X_test\n",
    "    X_test = []\n",
    "    for i in range(time_steps,ts_test_len+time_steps-for_periods):\n",
    "        X_test.append(inputs[i-time_steps:i,0])\n",
    "        \n",
    "    X_test = np.array(X_test)\n",
    "    X_test = np.reshape(X_test, (X_test.shape[0],X_test.shape[1],1))\n",
    "\n",
    "    return X_train, y_train , X_test\n",
    "\n",
    "X_train, y_train, X_test = ts_train_test(all_data,5,2)\n",
    "X_train.shape[0],X_train.shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>257.31</td>\n",
       "      <td>258.48</td>\n",
       "      <td>259.15</td>\n",
       "      <td>268.46</td>\n",
       "      <td>266.38</td>\n",
       "      <td>266.35</td>\n",
       "      <td>265.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>258.48</td>\n",
       "      <td>259.15</td>\n",
       "      <td>268.46</td>\n",
       "      <td>266.38</td>\n",
       "      <td>266.35</td>\n",
       "      <td>265.34</td>\n",
       "      <td>267.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>259.15</td>\n",
       "      <td>268.46</td>\n",
       "      <td>266.38</td>\n",
       "      <td>266.35</td>\n",
       "      <td>265.34</td>\n",
       "      <td>267.94</td>\n",
       "      <td>272.73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>268.46</td>\n",
       "      <td>266.38</td>\n",
       "      <td>266.35</td>\n",
       "      <td>265.34</td>\n",
       "      <td>267.94</td>\n",
       "      <td>272.73</td>\n",
       "      <td>271.90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>266.38</td>\n",
       "      <td>266.35</td>\n",
       "      <td>265.34</td>\n",
       "      <td>267.94</td>\n",
       "      <td>272.73</td>\n",
       "      <td>271.90</td>\n",
       "      <td>268.93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1499</th>\n",
       "      <td>1591.91</td>\n",
       "      <td>1520.91</td>\n",
       "      <td>1551.48</td>\n",
       "      <td>1495.08</td>\n",
       "      <td>1460.83</td>\n",
       "      <td>1377.45</td>\n",
       "      <td>1343.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1500</th>\n",
       "      <td>1520.91</td>\n",
       "      <td>1551.48</td>\n",
       "      <td>1495.08</td>\n",
       "      <td>1460.83</td>\n",
       "      <td>1377.45</td>\n",
       "      <td>1343.96</td>\n",
       "      <td>1470.90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1501</th>\n",
       "      <td>1551.48</td>\n",
       "      <td>1495.08</td>\n",
       "      <td>1460.83</td>\n",
       "      <td>1377.45</td>\n",
       "      <td>1343.96</td>\n",
       "      <td>1470.90</td>\n",
       "      <td>1461.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1502</th>\n",
       "      <td>1495.08</td>\n",
       "      <td>1460.83</td>\n",
       "      <td>1377.45</td>\n",
       "      <td>1343.96</td>\n",
       "      <td>1470.90</td>\n",
       "      <td>1461.64</td>\n",
       "      <td>1478.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1503</th>\n",
       "      <td>1460.83</td>\n",
       "      <td>1377.45</td>\n",
       "      <td>1343.96</td>\n",
       "      <td>1470.90</td>\n",
       "      <td>1461.64</td>\n",
       "      <td>1478.02</td>\n",
       "      <td>1501.97</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1504 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            0        1        2        3        4        0        1\n",
       "0      257.31   258.48   259.15   268.46   266.38   266.35   265.34\n",
       "1      258.48   259.15   268.46   266.38   266.35   265.34   267.94\n",
       "2      259.15   268.46   266.38   266.35   265.34   267.94   272.73\n",
       "3      268.46   266.38   266.35   265.34   267.94   272.73   271.90\n",
       "4      266.38   266.35   265.34   267.94   272.73   271.90   268.93\n",
       "...       ...      ...      ...      ...      ...      ...      ...\n",
       "1499  1591.91  1520.91  1551.48  1495.08  1460.83  1377.45  1343.96\n",
       "1500  1520.91  1551.48  1495.08  1460.83  1377.45  1343.96  1470.90\n",
       "1501  1551.48  1495.08  1460.83  1377.45  1343.96  1470.90  1461.64\n",
       "1502  1495.08  1460.83  1377.45  1343.96  1470.90  1461.64  1478.02\n",
       "1503  1460.83  1377.45  1343.96  1470.90  1461.64  1478.02  1501.97\n",
       "\n",
       "[1504 rows x 7 columns]"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Make the 3-D shape to a data frame so we can see: \n",
    "X_train_see = pd.DataFrame(np.reshape(X_train, (X_train.shape[0],X_train.shape[1])))\n",
    "y_train_see = pd.DataFrame(y_train)\n",
    "pd.concat([X_train_see,y_train_see],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1343.96</td>\n",
       "      <td>1470.90</td>\n",
       "      <td>1461.64</td>\n",
       "      <td>1478.02</td>\n",
       "      <td>1501.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1470.90</td>\n",
       "      <td>1461.64</td>\n",
       "      <td>1478.02</td>\n",
       "      <td>1501.97</td>\n",
       "      <td>1539.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1461.64</td>\n",
       "      <td>1478.02</td>\n",
       "      <td>1501.97</td>\n",
       "      <td>1539.13</td>\n",
       "      <td>1500.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1478.02</td>\n",
       "      <td>1501.97</td>\n",
       "      <td>1539.13</td>\n",
       "      <td>1500.28</td>\n",
       "      <td>1575.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1501.97</td>\n",
       "      <td>1539.13</td>\n",
       "      <td>1500.28</td>\n",
       "      <td>1575.39</td>\n",
       "      <td>1629.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>244</th>\n",
       "      <td>1760.33</td>\n",
       "      <td>1760.94</td>\n",
       "      <td>1769.21</td>\n",
       "      <td>1790.66</td>\n",
       "      <td>1784.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245</th>\n",
       "      <td>1760.94</td>\n",
       "      <td>1769.21</td>\n",
       "      <td>1790.66</td>\n",
       "      <td>1784.03</td>\n",
       "      <td>1792.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>1769.21</td>\n",
       "      <td>1790.66</td>\n",
       "      <td>1784.03</td>\n",
       "      <td>1792.28</td>\n",
       "      <td>1786.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>247</th>\n",
       "      <td>1790.66</td>\n",
       "      <td>1784.03</td>\n",
       "      <td>1792.28</td>\n",
       "      <td>1786.50</td>\n",
       "      <td>1793.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>248</th>\n",
       "      <td>1784.03</td>\n",
       "      <td>1792.28</td>\n",
       "      <td>1786.50</td>\n",
       "      <td>1793.00</td>\n",
       "      <td>1789.21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>249 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           0        1        2        3        4\n",
       "0    1343.96  1470.90  1461.64  1478.02  1501.97\n",
       "1    1470.90  1461.64  1478.02  1501.97  1539.13\n",
       "2    1461.64  1478.02  1501.97  1539.13  1500.28\n",
       "3    1478.02  1501.97  1539.13  1500.28  1575.39\n",
       "4    1501.97  1539.13  1500.28  1575.39  1629.51\n",
       "..       ...      ...      ...      ...      ...\n",
       "244  1760.33  1760.94  1769.21  1790.66  1784.03\n",
       "245  1760.94  1769.21  1790.66  1784.03  1792.28\n",
       "246  1769.21  1790.66  1784.03  1792.28  1786.50\n",
       "247  1790.66  1784.03  1792.28  1786.50  1793.00\n",
       "248  1784.03  1792.28  1786.50  1793.00  1789.21\n",
       "\n",
       "[249 rows x 5 columns]"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Make the 3-D shape to a data frame so we can see: \n",
    "X_test_see = pd.DataFrame(np.reshape(X_test, (X_test.shape[0],X_test.shape[1])))\n",
    "pd.DataFrame(X_test_see)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 1504 samples in the training data\n",
      "There are 249 samples in the test data\n"
     ]
    }
   ],
   "source": [
    "print(\"There are \" + str(X_train.shape[0]) + \" samples in the training data\")\n",
    "print(\"There are \" + str(X_test.shape[0]) + \" samples in the test data\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## [Learning objective 5] Run a simple RNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[38.548107, 40.074337],\n",
       "       [38.548107, 40.074337],\n",
       "       [38.548107, 40.074337],\n",
       "       [38.548107, 40.074337],\n",
       "       [38.548107, 40.074337],\n",
       "       [38.548107, 40.074337],\n",
       "       [38.548107, 40.074337],\n",
       "       [38.548107, 40.074337],\n",
       "       [38.548107, 40.074337]], dtype=float32)"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def simple_rnn_model(X_train, y_train, X_test):\n",
    "    '''\n",
    "    create single layer rnn model trained on X_train and y_train\n",
    "    and make predictions on the X_test data\n",
    "    '''\n",
    "    # create a model\n",
    "    from keras.models import Sequential\n",
    "    from keras.layers import Dense, SimpleRNN\n",
    "    \n",
    "    my_rnn_model = Sequential()\n",
    "    my_rnn_model.add(SimpleRNN(32, return_sequences=True))\n",
    "    #my_rnn_model.add(SimpleRNN(32, return_sequences=True))\n",
    "    #my_rnn_model.add(SimpleRNN(32, return_sequences=True))\n",
    "    my_rnn_model.add(SimpleRNN(32))\n",
    "    my_rnn_model.add(Dense(2)) # The time step of the output\n",
    "\n",
    "    my_rnn_model.compile(optimizer='rmsprop', loss='mean_squared_error')\n",
    "\n",
    "    # fit the RNN model\n",
    "    my_rnn_model.fit(X_train, y_train, epochs=100, batch_size=150, verbose=0)\n",
    "\n",
    "    # Finalizing predictions\n",
    "    rnn_predictions = my_rnn_model.predict(X_test)\n",
    "\n",
    "    return my_rnn_model, rnn_predictions\n",
    "\n",
    "my_rnn_model, rnn_predictions = simple_rnn_model(X_train, y_train, X_test)\n",
    "rnn_predictions[1:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3072268.0, <AxesSubplot:xlabel='Date'>)"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def actual_pred_plot(preds):\n",
    "    actual_pred = pd.DataFrame(columns = ['Adj. Close', 'prediction'])\n",
    "    actual_pred['Adj. Close'] = all_data.loc['2019':,'Adj Close'][0:len(preds)]\n",
    "    actual_pred['prediction'] = preds[:,0]\n",
    "\n",
    "    from keras.metrics import MeanSquaredError\n",
    "    m = MeanSquaredError()\n",
    "    m.update_state(np.array(actual_pred['Adj. Close']),np.array(actual_pred['prediction']))\n",
    "    \n",
    "    return (m.result().numpy(), actual_pred.plot() )\n",
    "    \n",
    "actual_pred_plot(rnn_predictions)  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## [Learning objective 6] RNN/LTSM requires normalized input data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3852.171, <AxesSubplot:xlabel='Date'>)"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def ts_train_test_normalize(all_data,time_steps,for_periods):\n",
    "    '''\n",
    "    input: \n",
    "      data: dataframe with dates and price data\n",
    "    output:\n",
    "      X_train, y_train: data from 2013/1/1-2018/12/31\n",
    "      X_test:  data from 2019 -\n",
    "      sc:      insantiated MinMaxScaler object fit to the training data\n",
    "    '''\n",
    "    # create training and test set\n",
    "    ts_train = all_data[:'2018'].iloc[:,0:1].values\n",
    "    ts_test  = all_data['2019':].iloc[:,0:1].values\n",
    "    ts_train_len = len(ts_train)\n",
    "    ts_test_len = len(ts_test)\n",
    "\n",
    "    # scale the data\n",
    "    from sklearn.preprocessing import MinMaxScaler\n",
    "    sc = MinMaxScaler(feature_range=(0,1))\n",
    "    ts_train_scaled = sc.fit_transform(ts_train)\n",
    "\n",
    "    # create training data of s samples and t time steps\n",
    "    X_train = []\n",
    "    y_train = []\n",
    "    y_train_stacked = []\n",
    "    for i in range(time_steps,ts_train_len-1): \n",
    "        X_train.append(ts_train_scaled[i-time_steps:i,0])\n",
    "        y_train.append(ts_train_scaled[i:i+for_periods,0])\n",
    "    X_train, y_train = np.array(X_train), np.array(y_train)\n",
    "\n",
    "    # Reshaping X_train for efficient modelling\n",
    "    X_train = np.reshape(X_train, (X_train.shape[0],X_train.shape[1],1))\n",
    "\n",
    "    inputs = pd.concat((all_data[\"Adj Close\"][:'2018'], all_data[\"Adj Close\"]['2019':]),axis=0).values\n",
    "    inputs = inputs[len(inputs)-len(ts_test) - time_steps:]\n",
    "    inputs = inputs.reshape(-1,1)\n",
    "    inputs  = sc.transform(inputs)\n",
    "\n",
    "    # Preparing X_test\n",
    "    X_test = []\n",
    "    for i in range(time_steps,ts_test_len+time_steps-for_periods):\n",
    "        X_test.append(inputs[i-time_steps:i,0])\n",
    "        \n",
    "    X_test = np.array(X_test)\n",
    "    X_test = np.reshape(X_test, (X_test.shape[0],X_test.shape[1],1))\n",
    "\n",
    "    return X_train, y_train , X_test, sc\n",
    "\n",
    "def simple_rnn_model(X_train, y_train, X_test, sc):\n",
    "    '''\n",
    "    create single layer rnn model trained on X_train and y_train\n",
    "    and make predictions on the X_test data\n",
    "    '''\n",
    "    # create a model\n",
    "    from keras.models import Sequential\n",
    "    from keras.layers import Dense, SimpleRNN\n",
    "    \n",
    "    my_rnn_model = Sequential()\n",
    "    my_rnn_model.add(SimpleRNN(32, return_sequences=True))\n",
    "    #my_rnn_model.add(SimpleRNN(32, return_sequences=True))\n",
    "    #my_rnn_model.add(SimpleRNN(32, return_sequences=True))\n",
    "    my_rnn_model.add(SimpleRNN(32))\n",
    "    my_rnn_model.add(Dense(2)) # The time step of the output\n",
    "\n",
    "    my_rnn_model.compile(optimizer='rmsprop', loss='mean_squared_error')\n",
    "\n",
    "    # fit the RNN model\n",
    "    my_rnn_model.fit(X_train, y_train, epochs=100, batch_size=150, verbose=0)\n",
    "\n",
    "    # Finalizing predictions\n",
    "    rnn_predictions = my_rnn_model.predict(X_test)\n",
    "    from sklearn.preprocessing import MinMaxScaler\n",
    "    rnn_predictions = sc.inverse_transform(rnn_predictions)\n",
    "\n",
    "    return my_rnn_model, rnn_predictions\n",
    "\n",
    "\n",
    "X_train, y_train, X_test, sc = ts_train_test_normalize(all_data,5,2)\n",
    "my_rnn_model, rnn_predictions_2 = simple_rnn_model(X_train, y_train, X_test, sc)\n",
    "rnn_predictions_2[1:10]\n",
    "actual_pred_plot(rnn_predictions_2)   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## [Learning objective 7] LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2267.276, <AxesSubplot:xlabel='Date'>)"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def LSTM_model(X_train, y_train, X_test, sc):\n",
    "    # create a model\n",
    "    from keras.models import Sequential\n",
    "    from keras.layers import Dense, SimpleRNN, GRU, LSTM\n",
    "    from keras.optimizers import SGD\n",
    "    \n",
    "    # The LSTM architecture\n",
    "    my_LSTM_model = Sequential()\n",
    "    my_LSTM_model.add(LSTM(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    #my_LSTM_model.add(LSTM(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    #my_LSTM_model.add(LSTM(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    my_LSTM_model.add(LSTM(units=50, activation='tanh'))\n",
    "    my_LSTM_model.add(Dense(units=2))\n",
    "\n",
    "    # Compiling\n",
    "    my_LSTM_model.compile(optimizer=SGD(lr=0.01, decay=1e-7, momentum=0.9, nesterov=False),loss='mean_squared_error')\n",
    "    # Fitting to the training set\n",
    "    my_LSTM_model.fit(X_train,y_train,epochs=50,batch_size=150, verbose=0)\n",
    "\n",
    "    LSTM_prediction = my_LSTM_model.predict(X_test)\n",
    "    LSTM_prediction = sc.inverse_transform(LSTM_prediction)\n",
    "\n",
    "    return my_LSTM_model, LSTM_prediction\n",
    "\n",
    "my_LSTM_model, LSTM_prediction = LSTM_model(X_train, y_train, X_test, sc)\n",
    "LSTM_prediction[1:10]\n",
    "actual_pred_plot(LSTM_prediction)   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## [Learning objective 8] LSTM Regularization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2152.7368, <AxesSubplot:xlabel='Date'>)"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def LSTM_model_regularization(X_train, y_train, X_test, sc):\n",
    "    # create a model\n",
    "    from keras.models import Sequential\n",
    "    from keras.layers import Dense, SimpleRNN, GRU, LSTM\n",
    "    from keras.optimizers import SGD\n",
    "    \n",
    "    # The LSTM architecture\n",
    "    my_LSTM_model = Sequential()\n",
    "    my_LSTM_model.add(LSTM(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    #my_LSTM_model.add(LSTM(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    #my_LSTM_model.add(LSTM(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    my_LSTM_model.add(LSTM(units=50, activation='tanh'))\n",
    "    my_LSTM_model.add(Dense(units=2))\n",
    "\n",
    "    # Compiling\n",
    "    my_LSTM_model.compile(optimizer=SGD(lr=0.01, decay=1e-7, momentum=0.9, nesterov=False),loss='mean_squared_error')\n",
    "    # Fitting to the training set\n",
    "    my_LSTM_model.fit(X_train,y_train,epochs=50,batch_size=150, verbose=0)\n",
    "\n",
    "    LSTM_prediction = my_LSTM_model.predict(X_test)\n",
    "    LSTM_prediction = sc.inverse_transform(LSTM_prediction)\n",
    "\n",
    "    return my_LSTM_model, LSTM_prediction\n",
    "\n",
    "my_LSTM_model, LSTM_prediction = LSTM_model_regularization(X_train, y_train, X_test, sc)\n",
    "LSTM_prediction[1:10]\n",
    "actual_pred_plot(LSTM_prediction)  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## [Learning objective 9] GRU"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1232.82, <AxesSubplot:xlabel='Date'>)"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def GRU_model(X_train, y_train, X_test, sc):\n",
    "    # create a model\n",
    "    from keras.models import Sequential\n",
    "    from keras.layers import Dense, SimpleRNN, GRU\n",
    "    from keras.optimizers import SGD\n",
    "    \n",
    "    # The GRU architecture\n",
    "    my_GRU_model = Sequential()\n",
    "    # First GRU layer with Dropout regularisation\n",
    "    my_GRU_model.add(GRU(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    #my_GRU_model.add(GRU(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    #my_GRU_model.add(GRU(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    my_GRU_model.add(GRU(units=50, activation='tanh'))\n",
    "    my_GRU_model.add(Dense(units=2))\n",
    "\n",
    "    # Compiling the RNN\n",
    "    my_GRU_model.compile(optimizer=SGD(lr=0.01, decay=1e-7, momentum=0.9, nesterov=False),loss='mean_squared_error')\n",
    "    # Fitting to the training set\n",
    "    my_GRU_model.fit(X_train,y_train,epochs=50,batch_size=150, verbose=0)\n",
    "\n",
    "    GRU_prediction = my_GRU_model.predict(X_test)\n",
    "    GRU_prediction = sc.inverse_transform(GRU_prediction)\n",
    "\n",
    "    return my_GRU_model, GRU_prediction\n",
    "\n",
    "my_GRU_model, GRU_prediction = GRU_model(X_train, y_train, X_test, sc)\n",
    "GRU_prediction[1:10]\n",
    "actual_pred_plot(GRU_prediction)  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## [Learning objective 10] GRU Regularization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1232.82, <AxesSubplot:xlabel='Date'>)"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def GRU_model_regularization(X_train, y_train, X_test, sc):\n",
    "    '''\n",
    "    create GRU model trained on X_train and y_train\n",
    "    and make predictions on the X_test data\n",
    "    '''\n",
    "    # create a model\n",
    "    from keras.models import Sequential\n",
    "    from keras.layers import Dense, SimpleRNN, GRU\n",
    "    from keras.optimizers import SGD\n",
    "    from keras.layers import Dropout\n",
    "    \n",
    "    # The GRU architecture\n",
    "    my_GRU_model = Sequential()\n",
    "    # First GRU layer with Dropout regularisation\n",
    "    my_GRU_model.add(GRU(units=50, return_sequences=True, input_shape=(X_train.shape[1],1), activation='tanh'))\n",
    "    my_GRU_model.add(Dropout(0.2))\n",
    "    # Second GRU layer\n",
    "    my_GRU_model.add(GRU(units=50, return_sequences=True, activation='tanh'))\n",
    "    my_GRU_model.add(Dropout(0.2))\n",
    "    \n",
    "    # Third GRU layer\n",
    "    my_GRU_model.add(GRU(units=50, return_sequences=True, activation='tanh'))\n",
    "    my_GRU_model.add(Dropout(0.2))\n",
    "    # Fourth GRU layer\n",
    "    my_GRU_model.add(GRU(units=50, activation='tanh'))\n",
    "    my_GRU_model.add(Dropout(0.2))\n",
    "    # The output layer\n",
    "    my_GRU_model.add(Dense(units=1))\n",
    "    # Compiling the RNN\n",
    "    my_GRU_model.compile(optimizer=SGD(lr=0.01, decay=1e-7, momentum=0.9, nesterov=False),loss='mean_squared_error')\n",
    "    # Fitting to the training set\n",
    "    my_GRU_model.fit(X_train,y_train,epochs=50,batch_size=150, verbose=0)\n",
    "\n",
    "    GRU_predictions = my_GRU_model.predict(X_test)\n",
    "    GRU_predictions = sc.inverse_transform(GRU_predictions)\n",
    "\n",
    "    return my_GRU_model, GRU_predictions\n",
    "\n",
    "my_GRU_model, GRU_predictions = GRU_model_regularization(X_train, y_train, X_test, sc)\n",
    "GRU_predictions[1:10]\n",
    "actual_pred_plot(GRU_prediction)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}