daily_stock_prediction.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "GloveLSTM.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "TWU7zxhGVrxS",
        "colab_type": "code",
        "outputId": "31bc9d14-5d59-4668-b9ce-e96dcbe39bb2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 55
        }
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive/')"
      ],
      "execution_count": 107,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive/; to attempt to forcibly remount, call drive.mount(\"/content/drive/\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kbNX2bOPVxIW",
        "colab_type": "code",
        "outputId": "46e3739d-bd59-4c8d-8864-b160d859e620",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 217
        }
      },
      "source": [
        "# Import required Libraries\n",
        "!pip install fastparquet\n",
        "import math\n",
        "import pandas_datareader as web\n",
        "import numpy as np \n",
        "import pandas as pd\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from tensorflow.keras.models import Sequential \n",
        "from tensorflow.keras.layers import Dense, LSTM\n",
        "import matplotlib.pyplot as plt\n",
        "import datetime"
      ],
      "execution_count": 108,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: fastparquet in /usr/local/lib/python3.6/dist-packages (0.3.3)\n",
            "Requirement already satisfied: numpy>=1.11 in /usr/local/lib/python3.6/dist-packages (from fastparquet) (1.18.2)\n",
            "Requirement already satisfied: pandas>=0.19 in /usr/local/lib/python3.6/dist-packages (from fastparquet) (1.0.3)\n",
            "Requirement already satisfied: numba>=0.28 in /usr/local/lib/python3.6/dist-packages (from fastparquet) (0.48.0)\n",
            "Requirement already satisfied: thrift>=0.11.0 in /usr/local/lib/python3.6/dist-packages (from fastparquet) (0.13.0)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from fastparquet) (1.12.0)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.19->fastparquet) (2018.9)\n",
            "Requirement already satisfied: python-dateutil>=2.6.1 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.19->fastparquet) (2.8.1)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from numba>=0.28->fastparquet) (46.1.3)\n",
            "Requirement already satisfied: llvmlite<0.32.0,>=0.31.0dev0 in /usr/local/lib/python3.6/dist-packages (from numba>=0.28->fastparquet) (0.31.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S7CcdnvpWCAC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0_BwnKjgWGKf",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "stocks = pd.read_parquet('https://github.com/alik604/CMPT-419/blob/master/data/wsb.daily.joined.parquet.gz?raw=true', engine=\"fastparquet\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9ARCPWbzWqhO",
        "colab_type": "code",
        "outputId": "96d2f518-d3a0-4cf0-c384-60a1453ef7f3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        }
      },
      "source": [
        "stocks.head(5)\n",
        "stocks.tail(5)"
      ],
      "execution_count": 195,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>created_utc</th>\n",
              "      <th>score</th>\n",
              "      <th>gilded</th>\n",
              "      <th>wordcount</th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "      <th>15</th>\n",
              "      <th>16</th>\n",
              "      <th>17</th>\n",
              "      <th>18</th>\n",
              "      <th>19</th>\n",
              "      <th>20</th>\n",
              "      <th>21</th>\n",
              "      <th>22</th>\n",
              "      <th>23</th>\n",
              "      <th>24</th>\n",
              "      <th>AAPL_CPrc</th>\n",
              "      <th>AAPL_OPrc</th>\n",
              "      <th>AMZN_CPrc</th>\n",
              "      <th>AMZN_OPrc</th>\n",
              "      <th>BA_CPrc</th>\n",
              "      <th>BA_OPrc</th>\n",
              "      <th>SPY_CPrc</th>\n",
              "      <th>SPY_OPrc</th>\n",
              "      <th>TSLA_CPrc</th>\n",
              "      <th>TSLA_OPrc</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>708</th>\n",
              "      <td>2019-10-25</td>\n",
              "      <td>5.419186</td>\n",
              "      <td>0.000254</td>\n",
              "      <td>15.673863</td>\n",
              "      <td>-0.015092</td>\n",
              "      <td>0.086603</td>\n",
              "      <td>-0.009423</td>\n",
              "      <td>-0.018914</td>\n",
              "      <td>-0.037407</td>\n",
              "      <td>-0.022757</td>\n",
              "      <td>0.150014</td>\n",
              "      <td>-0.074765</td>\n",
              "      <td>-0.021758</td>\n",
              "      <td>-0.006170</td>\n",
              "      <td>-0.043355</td>\n",
              "      <td>0.060446</td>\n",
              "      <td>-0.667465</td>\n",
              "      <td>-0.029348</td>\n",
              "      <td>0.042688</td>\n",
              "      <td>0.020240</td>\n",
              "      <td>0.046672</td>\n",
              "      <td>-0.046239</td>\n",
              "      <td>-0.001512</td>\n",
              "      <td>-0.014393</td>\n",
              "      <td>0.006396</td>\n",
              "      <td>0.018746</td>\n",
              "      <td>0.037486</td>\n",
              "      <td>0.030975</td>\n",
              "      <td>-0.051407</td>\n",
              "      <td>246.58</td>\n",
              "      <td>243.16</td>\n",
              "      <td>1761.33</td>\n",
              "      <td>1698.01</td>\n",
              "      <td>339.83</td>\n",
              "      <td>342.99</td>\n",
              "      <td>301.60</td>\n",
              "      <td>299.74</td>\n",
              "      <td>328.13</td>\n",
              "      <td>297.72</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>709</th>\n",
              "      <td>2019-10-28</td>\n",
              "      <td>5.501885</td>\n",
              "      <td>0.000092</td>\n",
              "      <td>16.153627</td>\n",
              "      <td>-0.012493</td>\n",
              "      <td>0.088967</td>\n",
              "      <td>-0.010182</td>\n",
              "      <td>-0.019853</td>\n",
              "      <td>-0.039271</td>\n",
              "      <td>-0.024242</td>\n",
              "      <td>0.149202</td>\n",
              "      <td>-0.081592</td>\n",
              "      <td>-0.018510</td>\n",
              "      <td>-0.005629</td>\n",
              "      <td>-0.040943</td>\n",
              "      <td>0.060828</td>\n",
              "      <td>-0.669622</td>\n",
              "      <td>-0.026370</td>\n",
              "      <td>0.043762</td>\n",
              "      <td>0.021522</td>\n",
              "      <td>0.047934</td>\n",
              "      <td>-0.044034</td>\n",
              "      <td>0.000773</td>\n",
              "      <td>-0.018675</td>\n",
              "      <td>0.005929</td>\n",
              "      <td>0.016296</td>\n",
              "      <td>0.038137</td>\n",
              "      <td>0.029383</td>\n",
              "      <td>-0.054537</td>\n",
              "      <td>249.05</td>\n",
              "      <td>247.50</td>\n",
              "      <td>1777.08</td>\n",
              "      <td>1750.70</td>\n",
              "      <td>340.88</td>\n",
              "      <td>341.49</td>\n",
              "      <td>303.30</td>\n",
              "      <td>302.91</td>\n",
              "      <td>327.71</td>\n",
              "      <td>327.35</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>710</th>\n",
              "      <td>2019-10-29</td>\n",
              "      <td>4.831472</td>\n",
              "      <td>0.000162</td>\n",
              "      <td>15.053424</td>\n",
              "      <td>-0.013019</td>\n",
              "      <td>0.089879</td>\n",
              "      <td>-0.008358</td>\n",
              "      <td>-0.017215</td>\n",
              "      <td>-0.038393</td>\n",
              "      <td>-0.022216</td>\n",
              "      <td>0.151195</td>\n",
              "      <td>-0.074344</td>\n",
              "      <td>-0.021114</td>\n",
              "      <td>-0.003404</td>\n",
              "      <td>-0.043528</td>\n",
              "      <td>0.062134</td>\n",
              "      <td>-0.665775</td>\n",
              "      <td>-0.031624</td>\n",
              "      <td>0.043718</td>\n",
              "      <td>0.024438</td>\n",
              "      <td>0.046891</td>\n",
              "      <td>-0.044923</td>\n",
              "      <td>0.001078</td>\n",
              "      <td>-0.015523</td>\n",
              "      <td>0.009790</td>\n",
              "      <td>0.019005</td>\n",
              "      <td>0.038384</td>\n",
              "      <td>0.031873</td>\n",
              "      <td>-0.050217</td>\n",
              "      <td>243.29</td>\n",
              "      <td>248.97</td>\n",
              "      <td>1762.71</td>\n",
              "      <td>1775.00</td>\n",
              "      <td>348.93</td>\n",
              "      <td>340.15</td>\n",
              "      <td>303.21</td>\n",
              "      <td>303.03</td>\n",
              "      <td>316.22</td>\n",
              "      <td>319.99</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>711</th>\n",
              "      <td>2019-10-30</td>\n",
              "      <td>5.048346</td>\n",
              "      <td>0.000181</td>\n",
              "      <td>14.314462</td>\n",
              "      <td>-0.006399</td>\n",
              "      <td>0.089110</td>\n",
              "      <td>-0.016125</td>\n",
              "      <td>-0.019654</td>\n",
              "      <td>-0.033241</td>\n",
              "      <td>-0.019550</td>\n",
              "      <td>0.140450</td>\n",
              "      <td>-0.045746</td>\n",
              "      <td>-0.022562</td>\n",
              "      <td>-0.003709</td>\n",
              "      <td>-0.045547</td>\n",
              "      <td>0.050558</td>\n",
              "      <td>-0.656832</td>\n",
              "      <td>-0.030471</td>\n",
              "      <td>0.037854</td>\n",
              "      <td>0.018806</td>\n",
              "      <td>0.033313</td>\n",
              "      <td>-0.042526</td>\n",
              "      <td>-0.004206</td>\n",
              "      <td>-0.004015</td>\n",
              "      <td>0.004740</td>\n",
              "      <td>0.010254</td>\n",
              "      <td>0.037975</td>\n",
              "      <td>0.031479</td>\n",
              "      <td>-0.046712</td>\n",
              "      <td>243.26</td>\n",
              "      <td>244.76</td>\n",
              "      <td>1779.99</td>\n",
              "      <td>1760.10</td>\n",
              "      <td>346.06</td>\n",
              "      <td>346.63</td>\n",
              "      <td>304.14</td>\n",
              "      <td>303.43</td>\n",
              "      <td>315.01</td>\n",
              "      <td>313.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>712</th>\n",
              "      <td>2019-10-31</td>\n",
              "      <td>9.800924</td>\n",
              "      <td>0.000872</td>\n",
              "      <td>17.253552</td>\n",
              "      <td>-0.013783</td>\n",
              "      <td>0.088982</td>\n",
              "      <td>-0.012983</td>\n",
              "      <td>-0.012372</td>\n",
              "      <td>-0.036133</td>\n",
              "      <td>-0.023620</td>\n",
              "      <td>0.154619</td>\n",
              "      <td>-0.076849</td>\n",
              "      <td>-0.023294</td>\n",
              "      <td>-0.003048</td>\n",
              "      <td>-0.041251</td>\n",
              "      <td>0.061375</td>\n",
              "      <td>-0.663991</td>\n",
              "      <td>-0.028172</td>\n",
              "      <td>0.043660</td>\n",
              "      <td>0.026479</td>\n",
              "      <td>0.046957</td>\n",
              "      <td>-0.043210</td>\n",
              "      <td>0.002582</td>\n",
              "      <td>-0.017126</td>\n",
              "      <td>0.007530</td>\n",
              "      <td>0.019177</td>\n",
              "      <td>0.033687</td>\n",
              "      <td>0.031406</td>\n",
              "      <td>-0.051298</td>\n",
              "      <td>248.76</td>\n",
              "      <td>247.20</td>\n",
              "      <td>1776.66</td>\n",
              "      <td>1775.99</td>\n",
              "      <td>339.91</td>\n",
              "      <td>344.70</td>\n",
              "      <td>303.33</td>\n",
              "      <td>304.11</td>\n",
              "      <td>314.92</td>\n",
              "      <td>313.10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    created_utc     score    gilded  ...  SPY_OPrc  TSLA_CPrc  TSLA_OPrc\n",
              "708  2019-10-25  5.419186  0.000254  ...    299.74     328.13     297.72\n",
              "709  2019-10-28  5.501885  0.000092  ...    302.91     327.71     327.35\n",
              "710  2019-10-29  4.831472  0.000162  ...    303.03     316.22     319.99\n",
              "711  2019-10-30  5.048346  0.000181  ...    303.43     315.01     313.00\n",
              "712  2019-10-31  9.800924  0.000872  ...    304.11     314.92     313.10\n",
              "\n",
              "[5 rows x 39 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 195
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ABx22CiAXalK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Drop created date \n",
        "data = stocks.drop(['created_utc', 'AAPL_CPrc','AMZN_CPrc','BA_CPrc', 'SPY_CPrc', 'TSLA_CPrc' ] , axis=1)\n",
        "\n",
        "# Closing Price dataset\n",
        "plotData = stocks.filter(['created_utc','AAPL_CPrc', 'AMZN_CPrc', 'BA_CPrc', 'SPY_CPrc', 'TSLA_CPrc'])\n",
        "dataCPrc = stocks.filter(['AAPL_CPrc', 'AMZN_CPrc', 'BA_CPrc', 'SPY_CPrc', 'TSLA_CPrc'])\n",
        "datasetCPrc = dataCPrc.values\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d7BKqI90X45t",
        "colab_type": "code",
        "outputId": "48d1a67c-b937-4905-a9ac-cd7617d5b3ff",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 251
        }
      },
      "source": [
        "dataset = data.values\n",
        "print(dataset)"
      ],
      "execution_count": 197,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[4.62113587e+00 0.00000000e+00 1.72465852e+01 ... 1.56300000e+02\n",
            "  2.25070000e+02 2.15610000e+02]\n",
            " [5.84415584e+00 0.00000000e+00 1.69265951e+01 ... 1.57780000e+02\n",
            "  2.25640000e+02 2.14300000e+02]\n",
            " [6.36900369e+00 0.00000000e+00 1.79852399e+01 ... 1.58410000e+02\n",
            "  2.26280000e+02 2.26600000e+02]\n",
            " ...\n",
            " [4.83147159e+00 1.61890886e-04 1.50534240e+01 ... 3.40150000e+02\n",
            "  3.03030000e+02 3.19990000e+02]\n",
            " [5.04834621e+00 1.81071946e-04 1.43144616e+01 ... 3.46630000e+02\n",
            "  3.03430000e+02 3.13000000e+02]\n",
            " [9.80092391e+00 8.71611610e-04 1.72535518e+01 ... 3.44700000e+02\n",
            "  3.04110000e+02 3.13100000e+02]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XdJqirKIX9eP",
        "colab_type": "code",
        "outputId": "3cf847e2-b6ee-4661-97e4-8d8a6cb35a82",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "# set training set len\n",
        "training_data_len = math.ceil(len(dataset) * 0.8) \n",
        "training_data_len"
      ],
      "execution_count": 198,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "571"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 198
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EAOTCb4eYXlB",
        "colab_type": "code",
        "outputId": "a069ac13-afca-437d-d2da-c53fb0b2ffd1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 251
        }
      },
      "source": [
        "# Scale data \n",
        "scaler = MinMaxScaler(feature_range=(0,1))\n",
        "scaled_data = scaler.fit_transform(dataset)\n",
        "\n",
        "scalerCPrc = MinMaxScaler(feature_range=(0,1))\n",
        "scaledCPrc_data = scalerCPrc.fit_transform(datasetCPrc)\n",
        "scaledCPrc_data"
      ],
      "execution_count": 199,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[9.77223183e-04, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        1.84536233e-01],\n",
              "       [0.00000000e+00, 2.73748882e-03, 5.81702803e-03, 1.69835234e-02,\n",
              "        2.33072853e-01],\n",
              "       [4.43508983e-03, 2.08344675e-02, 6.13432047e-03, 1.47021546e-02,\n",
              "        2.31907975e-01],\n",
              "       ...,\n",
              "       [9.56701496e-01, 7.84733834e-01, 6.76749515e-01, 9.88212928e-01,\n",
              "        6.66165122e-01],\n",
              "       [9.56475983e-01, 7.98172415e-01, 6.66631412e-01, 1.00000000e+00,\n",
              "        6.60292190e-01],\n",
              "       [9.97820041e-01, 7.95582688e-01, 6.44949762e-01, 9.89733840e-01,\n",
              "        6.59855361e-01]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 199
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fo4YosvZYqu0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Create training set from scaled data \n",
        "train_data = scaled_data[0:training_data_len, :]\n",
        "train_y = scaledCPrc_data[0:training_data_len, :]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tR39LbGWY8lG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Split data into x_train, y_train\n",
        "x_train = []\n",
        "y_train = []"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FJOXMynkZ9L5",
        "colab_type": "code",
        "outputId": "6bf3cd55-99e3-4118-fee5-ae07ae91d148",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 107
        }
      },
      "source": [
        "for i in range(60, len(train_data)):\n",
        "  x_train.append(train_data[i-60:i, :])\n",
        "  y_train.append(train_y[i:i+1,:][0].tolist())\n",
        "  if(i < 65):\n",
        "    print(train_y[i:i+1,:][0], type(train_y[i:i+1,:][0]))"
      ],
      "execution_count": 202,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[0.20972713 0.09540771 0.07407016 0.1400507  0.47959035] <class 'numpy.ndarray'>\n",
            "[0.20777268 0.10334798 0.07012163 0.13307985 0.48211426] <class 'numpy.ndarray'>\n",
            "[0.20807337 0.10720535 0.06938128 0.1278834  0.5802553 ] <class 'numpy.ndarray'>\n",
            "[0.21604149 0.11914298 0.0766085  0.12978454 0.60539727] <class 'numpy.ndarray'>\n",
            "[0.21047884 0.121025   0.07089723 0.12091255 0.56205407] <class 'numpy.ndarray'>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ACnTHF1XdJpY",
        "colab_type": "code",
        "outputId": "4c3f1d62-3765-4043-ce88-8fc119233387",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 161
        }
      },
      "source": [
        "# Convert both x, y training sets to np array \n",
        "x_train, y_train = np.array(x_train), np.array(y_train)\n",
        "print(y_train, type(y_train))\n",
        "# Reshape the data // LSTM network expects 3 dimensional input in the form of \n",
        "# (number of samples, number of timesteps, number of features)\n",
        "x_train = np.reshape(x_train, (x_train.shape[0], x_train.shape[1], 33))\n",
        "x_train.shape"
      ],
      "execution_count": 203,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[0.20972713 0.09540771 0.07407016 0.1400507  0.47959035]\n",
            " [0.20777268 0.10334798 0.07012163 0.13307985 0.48211426]\n",
            " [0.20807337 0.10720535 0.06938128 0.1278834  0.5802553 ]\n",
            " ...\n",
            " [0.63188754 0.85251001 0.76696633 0.80544994 0.45736058]\n",
            " [0.62752763 0.84160672 0.74764675 0.78669202 0.45304082]\n",
            " [0.63594678 0.85054244 0.73319231 0.7991128  0.47124205]] <class 'numpy.ndarray'>\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(511, 60, 33)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 203
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IKVNbH_GeA8u",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Declare the LSTM model architecture \n",
        "model = Sequential()\n",
        "\n",
        "# Add layers , input shape expected is (number of timesteps, number of features)\n",
        "model.add(LSTM(75, return_sequences=True, input_shape=(x_train.shape[1], 33)))\n",
        "model.add(LSTM(50, return_sequences=False))\n",
        "model.add(Dense(25))\n",
        "model.add(Dense(5))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rfiNSgYqeHTG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Compile Model \n",
        "model.compile(optimizer='adam', loss='mean_squared_error')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AGFe80-GQ08x",
        "colab_type": "text"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R_3Q9445eJSt",
        "colab_type": "code",
        "outputId": "42909065-e3a1-4da0-91d1-8b8f539196d4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# Train the model \n",
        "model.fit(x_train, y_train, batch_size=16, epochs=100)"
      ],
      "execution_count": 206,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0604\n",
            "Epoch 2/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0106\n",
            "Epoch 3/100\n",
            "32/32 [==============================] - 2s 54ms/step - loss: 0.0076\n",
            "Epoch 4/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0061\n",
            "Epoch 5/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0056\n",
            "Epoch 6/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0049\n",
            "Epoch 7/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0044\n",
            "Epoch 8/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0043\n",
            "Epoch 9/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0039\n",
            "Epoch 10/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0041\n",
            "Epoch 11/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0039\n",
            "Epoch 12/100\n",
            "32/32 [==============================] - 3s 87ms/step - loss: 0.0037\n",
            "Epoch 13/100\n",
            "32/32 [==============================] - 3s 92ms/step - loss: 0.0036\n",
            "Epoch 14/100\n",
            "32/32 [==============================] - 3s 93ms/step - loss: 0.0031\n",
            "Epoch 15/100\n",
            "32/32 [==============================] - 2s 60ms/step - loss: 0.0030\n",
            "Epoch 16/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0030\n",
            "Epoch 17/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0030\n",
            "Epoch 18/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0033\n",
            "Epoch 19/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0028\n",
            "Epoch 20/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0029\n",
            "Epoch 21/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0029\n",
            "Epoch 22/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0029\n",
            "Epoch 23/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0026\n",
            "Epoch 24/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0029\n",
            "Epoch 25/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0025\n",
            "Epoch 26/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0023\n",
            "Epoch 27/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0025\n",
            "Epoch 28/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0027\n",
            "Epoch 29/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0023\n",
            "Epoch 30/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0021\n",
            "Epoch 31/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0021\n",
            "Epoch 32/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0020\n",
            "Epoch 33/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0020\n",
            "Epoch 34/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0019\n",
            "Epoch 35/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0022\n",
            "Epoch 36/100\n",
            "32/32 [==============================] - 2s 60ms/step - loss: 0.0019\n",
            "Epoch 37/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0019\n",
            "Epoch 38/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0019\n",
            "Epoch 39/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0017\n",
            "Epoch 40/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0017\n",
            "Epoch 41/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0018\n",
            "Epoch 42/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0020\n",
            "Epoch 43/100\n",
            "32/32 [==============================] - 2s 54ms/step - loss: 0.0019\n",
            "Epoch 44/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0018\n",
            "Epoch 45/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0017\n",
            "Epoch 46/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0017\n",
            "Epoch 47/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0018\n",
            "Epoch 48/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0015\n",
            "Epoch 49/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0014\n",
            "Epoch 50/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0014\n",
            "Epoch 51/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0016\n",
            "Epoch 52/100\n",
            "32/32 [==============================] - 2s 60ms/step - loss: 0.0014\n",
            "Epoch 53/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0014\n",
            "Epoch 54/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0014\n",
            "Epoch 55/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0015\n",
            "Epoch 56/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0014\n",
            "Epoch 57/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0013\n",
            "Epoch 58/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0013\n",
            "Epoch 59/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0013\n",
            "Epoch 60/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0014\n",
            "Epoch 61/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0014\n",
            "Epoch 62/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0013\n",
            "Epoch 63/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0012\n",
            "Epoch 64/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0012\n",
            "Epoch 65/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0012\n",
            "Epoch 66/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0011\n",
            "Epoch 67/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0012\n",
            "Epoch 68/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0011\n",
            "Epoch 69/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0011\n",
            "Epoch 70/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0011\n",
            "Epoch 71/100\n",
            "32/32 [==============================] - 2s 60ms/step - loss: 0.0011\n",
            "Epoch 72/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0011\n",
            "Epoch 73/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0012\n",
            "Epoch 74/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0012\n",
            "Epoch 75/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0011\n",
            "Epoch 76/100\n",
            "32/32 [==============================] - 2s 61ms/step - loss: 0.0011\n",
            "Epoch 77/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0011\n",
            "Epoch 78/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0012\n",
            "Epoch 79/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0012\n",
            "Epoch 80/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0012\n",
            "Epoch 81/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0012\n",
            "Epoch 82/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0011\n",
            "Epoch 83/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0011\n",
            "Epoch 84/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0011\n",
            "Epoch 85/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0010\n",
            "Epoch 86/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0011\n",
            "Epoch 87/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0011\n",
            "Epoch 88/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0011\n",
            "Epoch 89/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0011\n",
            "Epoch 90/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0012\n",
            "Epoch 91/100\n",
            "32/32 [==============================] - 2s 61ms/step - loss: 0.0011\n",
            "Epoch 92/100\n",
            "32/32 [==============================] - 2s 61ms/step - loss: 0.0011\n",
            "Epoch 93/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0010\n",
            "Epoch 94/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0010\n",
            "Epoch 95/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0010\n",
            "Epoch 96/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 9.7355e-04\n",
            "Epoch 97/100\n",
            "32/32 [==============================] - 2s 60ms/step - loss: 0.0011\n",
            "Epoch 98/100\n",
            "32/32 [==============================] - 2s 61ms/step - loss: 9.4772e-04\n",
            "Epoch 99/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0012\n",
            "Epoch 100/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 9.1165e-04\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f0d4ad17cf8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 206
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zMBz4Q63eKu9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Create the testing data set\n",
        "test_data = scaled_data[training_data_len - 60: , :]\n",
        "test_y = scaledCPrc_data[training_data_len-60:,:]\n",
        "# Create the data sets x_test and y_test\n",
        "x_test = []\n",
        "y_test = []\n",
        "\n",
        "for i in range(60, len(test_data)):\n",
        "  x_test.append(test_data[i-60:i, :])\n",
        "  y_test.append(train_y[i:i+1,:][0].tolist())\n",
        "\n",
        "print(len(test))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R8zY8JdFflgS",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Convert the data into a np array\n",
        "x_test = np.array(x_test)\n",
        "\n",
        "# Reshape data into 3 dimensions ( num samples, timesteps, num features )\n",
        "x_test = np.reshape(x_test, (x_test.shape[0], x_test.shape[1], 33))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tyGEr_xjfm_n",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Get the models predicted price values \n",
        "predictions = model.predict(x_test)\n",
        "\n",
        "# Inverse transform \n",
        "predictions = scalerCPrc.inverse_transform(predictions)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-aEJ2RyOfop3",
        "colab_type": "code",
        "outputId": "839ec772-bfde-42e8-e70c-3908c99ea41f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "# Evaluate model with root mean square error (RMSE)\n",
        "rmse = np.sqrt(np.mean((predictions-y_test)**2))\n",
        "rmse"
      ],
      "execution_count": 125,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "878.2621720539609"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 125
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3cpsmjVBjF8_",
        "colab_type": "code",
        "outputId": "f946aefb-a6f2-4068-f5f4-374df03ad5e2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 775
        }
      },
      "source": [
        "# Set up ploting dataset\n",
        "plotData = stocks.filter(['created_utc','AAPL_CPrc', 'AMZN_CPrc', 'BA_CPrc', 'SPY_CPrc', 'TSLA_CPrc'])\n",
        "plotData = plotData.set_index(\"created_utc\")\n",
        "print(plotData, type(plotData))\n",
        "\n",
        "# This is so it graphs nicely with dates \n",
        "AAPL = plotData[\"AAPL_CPrc\"]\n",
        "AAPL = AAPL.to_frame()\n",
        "AMZN = plotData[\"AMZN_CPrc\"]\n",
        "AMZN = AMZN.to_frame()\n",
        "BA = plotData[\"BA_CPrc\"]\n",
        "BA = BA.to_frame();\n",
        "SPY = plotData[\"SPY_CPrc\"]\n",
        "SPY = SPY.to_frame();\n",
        "TSLA = plotData[\"TSLA_CPrc\"]\n",
        "TSLA = TSLA.to_frame()\n",
        "\n",
        "appleTest = AAPL[0:training_data_len]\n",
        "amazonTest = AMZN[0:training_data_len]\n",
        "boeTest = BA[0:training_data_len]\n",
        "spyTest = SPY[0: training_data_len]\n",
        "teslaTest = TSLA[0: training_data_len]\n",
        "\n",
        "appleValid = AAPL[training_data_len:]\n",
        "amazonValid = AMZN[training_data_len:]\n",
        "boeValid = BA[training_data_len:]\n",
        "spyValid = SPY[training_data_len:]\n",
        "teslaValid = TSLA[training_data_len:]\n",
        "# print(appleValid)\n",
        "\n",
        "appleValid['Predictions'] = predictions[:,:1]\n",
        "amazonValid[\"Predictions\"] = predictions[:,1:2]\n",
        "boeValid['Predictions'] = predictions[:, 2:3]\n",
        "spyValid[\"Predictions\"] = predictions[:, 3:4]\n",
        "teslaValid[\"Predictions\"] = predictions[:, 4:5]\n",
        "print(len(appleValid), len(predictions))\n"
      ],
      "execution_count": 147,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "             AAPL_CPrc  AMZN_CPrc  BA_CPrc  SPY_CPrc  TSLA_CPrc\n",
            "created_utc                                                    \n",
            "2017-01-03      116.15     753.66   156.97    225.24     216.99\n",
            "2017-01-04      116.02     757.18   158.62    226.58     226.99\n",
            "2017-01-05      116.61     780.45   158.71    226.40     226.75\n",
            "2017-01-06      117.91     795.99   159.10    227.21     229.01\n",
            "2017-01-09      118.99     796.92   158.32    226.46     231.25\n",
            "...                ...        ...      ...       ...        ...\n",
            "2019-10-25      246.58    1761.33   339.83    301.60     328.13\n",
            "2019-10-28      249.05    1777.08   340.88    303.30     327.71\n",
            "2019-10-29      243.29    1762.71   348.93    303.21     316.22\n",
            "2019-10-30      243.26    1779.99   346.06    304.14     315.01\n",
            "2019-10-31      248.76    1776.66   339.91    303.33     314.92\n",
            "\n",
            "[713 rows x 5 columns] <class 'pandas.core.frame.DataFrame'>\n",
            "142 142\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:30: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:31: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:32: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:33: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:34: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T5YFXuL2jXN8",
        "colab_type": "code",
        "outputId": "7f2fd887-620e-4848-8f45-3c1cd50ee47c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 521
        }
      },
      "source": [
        "# Visualize data\n",
        "plt.figure(figsize=(16,8))\n",
        "plt.title('Prediction with reddit glove vectors')\n",
        "plt.xlabel('Date', fontsize=18)\n",
        "plt.ylabel('Close Price USD ($)', fontsize=18)\n",
        "plt.plot(appleTest)\n",
        "plt.plot(appleValid)\n",
        "plt.plot(appleValid['Predictions'])\n",
        "# plt.plot(amazonTest)\n",
        "# plt.plot(amazonValid)\n",
        "# plt.plot(amazonValid['Predictions'])\n",
        "# plt.plot(boeTest)\n",
        "# plt.plot(boeValid)\n",
        "# plt.plot(boeValid['Predictions'])\n",
        "# plt.plot(spyTest)\n",
        "# plt.plot(spyValid)\n",
        "# plt.plot(spyValid['Predictions'])\n",
        "# plt.plot(teslaTest)\n",
        "# plt.plot(teslaValid)\n",
        "# plt.plot(teslaValid['Predictions'])\n",
        "plt.legend(['Train', 'Validation', 'Prediction'])\n",
        "plt.show()"
      ],
      "execution_count": 210,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3cbZiMQYmrbR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Lets train a model without reddit posts to see how it performs \n",
        "opening_prices = stocks.filter(['created_utc','AAPL_OPrc', 'AMZN_OPrc', 'BA_OPrc', 'SPY_OPrc', 'TSLA_OPrc'])\n",
        "opening_prices = opening_prices.set_index('created_utc')\n",
        "# PlotData contains closing prices "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1T3BOippD7Bs",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "outputId": "8952f591-f01a-41f8-9642-360867aabb10"
      },
      "source": [
        "opening_dataset = opening_prices.values\n",
        "closing_dataset = plotData.values\n",
        "print(opening_dataset)\n",
        "print(closing_dataset)"
      ],
      "execution_count": 129,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[ 115.8   758.    156.3   225.07  215.61]\n",
            " [ 115.8   758.39  157.78  225.64  214.3 ]\n",
            " [ 115.93  761.55  158.41  226.28  226.6 ]\n",
            " ...\n",
            " [ 248.97 1775.    340.15  303.03  319.99]\n",
            " [ 244.76 1760.1   346.63  303.43  313.  ]\n",
            " [ 247.2  1775.99  344.7   304.11  313.1 ]]\n",
            "[[ 116.15  753.66  156.97  225.24  216.99]\n",
            " [ 116.02  757.18  158.62  226.58  226.99]\n",
            " [ 116.61  780.45  158.71  226.4   226.75]\n",
            " ...\n",
            " [ 243.29 1762.71  348.93  303.21  316.22]\n",
            " [ 243.26 1779.99  346.06  304.14  315.01]\n",
            " [ 248.76 1776.66  339.91  303.33  314.92]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h52vWQxYEqn1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# set training set len\n",
        "train_data_len = math.ceil(len(opening_dataset) * 0.80) "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Lt1W734XFBpD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "openScaler = MinMaxScaler(feature_range=(0,1))\n",
        "scaled_open_prices = openScaler.fit_transform(opening_dataset)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qOvvI5bmFLLE",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "closeScaler = MinMaxScaler(feature_range=(0,1))\n",
        "scaled_close_prices = closeScaler.fit_transform(closing_dataset)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PkBSzN9dFVBn",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 377
        },
        "outputId": "de7db11f-b372-4faa-ebf4-3c3a7d353bcd"
      },
      "source": [
        "o_train_data = scaled_open_prices[0:train_data_len, :]\n",
        "c_train_data = scaled_close_prices[0:train_data_len,:]\n",
        "print(o_train_data, len(o_train_data))\n",
        "c_train_data"
      ],
      "execution_count": 153,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
            "  1.68423621e-01]\n",
            " [0.00000000e+00 3.04152856e-04 5.10855683e-03 7.21153846e-03\n",
            "  1.62030259e-01]\n",
            " [9.76195840e-04 2.76857087e-03 7.28314521e-03 1.53087045e-02\n",
            "  2.22059541e-01]\n",
            " ...\n",
            " [2.75587595e-01 6.90247612e-01 6.56518588e-01 3.95242915e-01\n",
            "  7.49389946e-01]\n",
            " [2.78441090e-01 6.88282316e-01 6.62973318e-01 4.12196356e-01\n",
            "  7.85700342e-01]\n",
            " [2.63197417e-01 6.68364204e-01 6.60073867e-01 4.02327935e-01\n",
            "  7.87701318e-01]] 511\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.00097722, 0.        , 0.        , 0.        , 0.18453623],\n",
              "       [0.        , 0.00273749, 0.00581703, 0.01698352, 0.23307285],\n",
              "       [0.00443509, 0.02083447, 0.00613432, 0.01470215, 0.23190797],\n",
              "       ...,\n",
              "       [0.28399609, 0.70191702, 0.68972325, 0.42636248, 0.80570791],\n",
              "       [0.27264527, 0.68973831, 0.69074564, 0.42762991, 0.81682279],\n",
              "       [0.25543111, 0.67157911, 0.68179094, 0.40760456, 0.7544047 ]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 153
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GGIyaEwwFb5_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 71
        },
        "outputId": "237d975c-3f8c-41df-ed94-eeec1863b44f"
      },
      "source": [
        "o_x_train = []\n",
        "c_y_train = []\n",
        "print(len(o_x_train))\n",
        "\n",
        "for i in range(60, len(o_train_data)):\n",
        "  o_x_train.append(o_train_data[i-60:i, :])\n",
        "  c_y_train.append(c_train_data[i,:].tolist())\n",
        "  if(i<62):\n",
        "    print(c_train_data[i,:].tolist())\n"
      ],
      "execution_count": 134,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0\n",
            "[0.20972712921897307, 0.09540770696426493, 0.07407015688348317, 0.14005069708491735, 0.47959035091976887]\n",
            "[0.2077726828534916, 0.1033479799354513, 0.07012162876784778, 0.13307984790874539, 0.48211425520555273]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FhVJ3AoaGAqj",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "e46c1d20-3e2d-4b09-e1ea-591842392c6b"
      },
      "source": [
        "x_train2, y_train2 = np.array(o_x_train), np.array(c_y_train)\n",
        "x_train2 = np.reshape(x_train2, (x_train2.shape[0], x_train2.shape[1], 5))\n",
        "x_train2.shape"
      ],
      "execution_count": 135,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(511, 60, 5)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 135
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NeOf2hP1HWn2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Declare the LSTM model architecture \n",
        "model2 = Sequential()\n",
        "\n",
        "# Add layers , input shape expected is (number of timesteps, number of features)\n",
        "model2.add(LSTM(75, return_sequences=True, input_shape=(x_train2.shape[1], 5)))\n",
        "model2.add(LSTM(50, return_sequences=False))\n",
        "model2.add(Dense(25))\n",
        "model2.add(Dense(5))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EZIWdGiTHpRP",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Compile Model \n",
        "model2.compile(optimizer='adam', loss='mean_squared_error')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "saqgNNGNHryQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "e58db65d-2779-43ae-c354-ebcfa834fa6c"
      },
      "source": [
        "# Train the model \n",
        "model2.fit(x_train2, y_train2, batch_size=16, epochs=100)"
      ],
      "execution_count": 138,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0495\n",
            "Epoch 2/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0082\n",
            "Epoch 3/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0062\n",
            "Epoch 4/100\n",
            "32/32 [==============================] - 2s 54ms/step - loss: 0.0055\n",
            "Epoch 5/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0053\n",
            "Epoch 6/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0054\n",
            "Epoch 7/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0045\n",
            "Epoch 8/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0042\n",
            "Epoch 9/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0040\n",
            "Epoch 10/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0040\n",
            "Epoch 11/100\n",
            "32/32 [==============================] - 2s 54ms/step - loss: 0.0039\n",
            "Epoch 12/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0035\n",
            "Epoch 13/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0036\n",
            "Epoch 14/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0034\n",
            "Epoch 15/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0034\n",
            "Epoch 16/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0035\n",
            "Epoch 17/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0033\n",
            "Epoch 18/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0030\n",
            "Epoch 19/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0031\n",
            "Epoch 20/100\n",
            "32/32 [==============================] - 2s 54ms/step - loss: 0.0031\n",
            "Epoch 21/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0030\n",
            "Epoch 22/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0028\n",
            "Epoch 23/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0028\n",
            "Epoch 24/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0030\n",
            "Epoch 25/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0029\n",
            "Epoch 26/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0028\n",
            "Epoch 27/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0028\n",
            "Epoch 28/100\n",
            "32/32 [==============================] - 2s 54ms/step - loss: 0.0027\n",
            "Epoch 29/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0025\n",
            "Epoch 30/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0025\n",
            "Epoch 31/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0024\n",
            "Epoch 32/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0025\n",
            "Epoch 33/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0026\n",
            "Epoch 34/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0026\n",
            "Epoch 35/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0026\n",
            "Epoch 36/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0023\n",
            "Epoch 37/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0024\n",
            "Epoch 38/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0022\n",
            "Epoch 39/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0025\n",
            "Epoch 40/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0022\n",
            "Epoch 41/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0022\n",
            "Epoch 42/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0020\n",
            "Epoch 43/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0023\n",
            "Epoch 44/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0021\n",
            "Epoch 45/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0021\n",
            "Epoch 46/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0020\n",
            "Epoch 47/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0019\n",
            "Epoch 48/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0019\n",
            "Epoch 49/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0020\n",
            "Epoch 50/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0021\n",
            "Epoch 51/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0020\n",
            "Epoch 52/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0019\n",
            "Epoch 53/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0019\n",
            "Epoch 54/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0018\n",
            "Epoch 55/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0019\n",
            "Epoch 56/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0019\n",
            "Epoch 57/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0019\n",
            "Epoch 58/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0019\n",
            "Epoch 59/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0017\n",
            "Epoch 60/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0018\n",
            "Epoch 61/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0018\n",
            "Epoch 62/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0019\n",
            "Epoch 63/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0017\n",
            "Epoch 64/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0017\n",
            "Epoch 65/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0017\n",
            "Epoch 66/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0016\n",
            "Epoch 67/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0017\n",
            "Epoch 68/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0017\n",
            "Epoch 69/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0019\n",
            "Epoch 70/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0017\n",
            "Epoch 71/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0016\n",
            "Epoch 72/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0016\n",
            "Epoch 73/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0016\n",
            "Epoch 74/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0018\n",
            "Epoch 75/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0015\n",
            "Epoch 76/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0016\n",
            "Epoch 77/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0016\n",
            "Epoch 78/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0016\n",
            "Epoch 79/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0016\n",
            "Epoch 80/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0015\n",
            "Epoch 81/100\n",
            "32/32 [==============================] - 2s 55ms/step - loss: 0.0018\n",
            "Epoch 82/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0016\n",
            "Epoch 83/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0015\n",
            "Epoch 84/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0014\n",
            "Epoch 85/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0013\n",
            "Epoch 86/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0016\n",
            "Epoch 87/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0014\n",
            "Epoch 88/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0014\n",
            "Epoch 89/100\n",
            "32/32 [==============================] - 2s 60ms/step - loss: 0.0014\n",
            "Epoch 90/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0015\n",
            "Epoch 91/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0015\n",
            "Epoch 92/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0013\n",
            "Epoch 93/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0013\n",
            "Epoch 94/100\n",
            "32/32 [==============================] - 2s 59ms/step - loss: 0.0013\n",
            "Epoch 95/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0013\n",
            "Epoch 96/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0013\n",
            "Epoch 97/100\n",
            "32/32 [==============================] - 2s 58ms/step - loss: 0.0012\n",
            "Epoch 98/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0012\n",
            "Epoch 99/100\n",
            "32/32 [==============================] - 2s 56ms/step - loss: 0.0014\n",
            "Epoch 100/100\n",
            "32/32 [==============================] - 2s 57ms/step - loss: 0.0013\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f0d4d244160>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 138
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KiEpTH7MHvOQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_test_data = scaled_open_prices[train_data_len - 60:, :]\n",
        "y_test_data = scaled_close_prices[train_data_len - 60:, :]\n",
        "\n",
        "x_test2 = []\n",
        "y_test2 = []\n",
        "for i in range(60, len(x_test_data)):\n",
        "  x_test2.append(x_test_data[i-60:i, :])\n",
        "  y_test2.append(y_test_data[i:i+1,:][0].tolist())\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t4n2owTSJ4LM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_test2 = np.array(x_test2)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yOkonKz3K90T",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_test2 = np.reshape(x_test2, (x_test2.shape[0], x_test2.shape[1], 5))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lq3alJAILFJ1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Get the models predicted price values \n",
        "predictions2 = model2.predict(x_test2)\n",
        "\n",
        "# Inverse transform \n",
        "predictions2 = closeScaler.inverse_transform(predictions2)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "X5DZ7yM9LK3b",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "1dd9ca87-9d2b-4990-9f81-b2c243c2ef12"
      },
      "source": [
        "# Evaluate model with root mean square error (RMSE)\n",
        "rmse2 = np.sqrt(np.mean((predictions2-y_test2)**2))\n",
        "rmse2"
      ],
      "execution_count": 159,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "859.9505340314873"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 159
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fOvweIEXLQu7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 577
        },
        "outputId": "5cf8db63-566d-4001-a792-62a1913573c8"
      },
      "source": [
        "appleValid['Predictions2'] = predictions2[:,:1]\n",
        "amazonValid[\"Predictions2\"] = predictions2[:,1:2]\n",
        "boeValid['Predictions2'] = predictions2[:, 2:3]\n",
        "spyValid[\"Predictions2\"] = predictions2[:, 3:4]\n",
        "teslaValid[\"Predictions2\"] = predictions2[:, 4:5]"
      ],
      "execution_count": 209,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  \"\"\"Entry point for launching an IPython kernel.\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:2: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  \n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  This is separate from the ipykernel package so we can avoid doing imports until\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:4: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  after removing the cwd from sys.path.\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:5: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  \"\"\"\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZNUeklyYLSkz",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 536
        },
        "outputId": "42a7a929-8a9e-45fc-e638-35e3fba2e5b6"
      },
      "source": [
        "# Visualize data\n",
        "plt.figure(figsize=(16,8))\n",
        "plt.title('Stock Prediction with and without Reddit word vector data \\n (TSLA)')\n",
        "plt.xlabel('Date', fontsize=18)\n",
        "plt.ylabel('Close Price USD ($)', fontsize=18)\n",
        "# plt.plot(appleTest)\n",
        "# plt.plot(appleValid)\n",
        "# plt.plot(appleValid['Predictions'], color='red')\n",
        "# plt.plot(appleValid['Predictions2'], color='green')\n",
        "# plt.plot(amazonTest)\n",
        "# plt.plot(amazonValid)\n",
        "# plt.plot(amazonValid['Predictions'], color='red')\n",
        "# plt.plot(amazonValid['Predictions2'], color='green')\n",
        "# plt.plot(boeTest)\n",
        "# plt.plot(boeValid)\n",
        "# plt.plot(boeValid['Predictions'], color='red')\n",
        "# plt.plot(boeValid['Predictions2'], color='green')\n",
        "# plt.plot(spyTest, label='Train')\n",
        "# plt.plot(spyValid, label='Actual')\n",
        "# plt.plot(spyValid['Predictions'], label='Prediction with vector', color='red')\n",
        "# plt.plot(spyValid['Predictions2'], label='Prediction without vector', color='green')\n",
        "\n",
        "plt.plot(teslaTest, label='Train')\n",
        "plt.plot(teslaValid, label=\"Actual\")\n",
        "plt.plot(teslaValid['Predictions'], label=\"vector prediction\", color='red')\n",
        "plt.plot(teslaValid['Predictions2'], label=\"without vector prediction\", color=\"green\")\n",
        "plt.legend(('Train', 'Validation', 'Prediction with vector', 'Prediction without'), title=\"Legend\",)\n",
        "plt.show()"
      ],
      "execution_count": 193,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SlhPAN6yPPI2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}