stock_price_prediction.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "stock-price-prediction.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xuGY57DXrQu4",
        "colab_type": "text"
      },
      "source": [
        "# Description\n",
        "This program uses an artifical recurrent neural network called Long Short Term Memory (LSTM) to predict the closing price of Apple Inc. using the past 60 day stock price.\n",
        "\n",
        "### Reference:\n",
        "https://youtu.be/QIUxPv5PJOY\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DpPB1xwKsEFd",
        "colab_type": "text"
      },
      "source": [
        "## Import Libraries\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bvNVy6DurKzp",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4685036e-c80c-4d2a-80d0-5e2d96b99082"
      },
      "source": [
        "import math\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import pandas_datareader as web\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, LSTM\n",
        "import matplotlib.pyplot as plt\n",
        "plt.style.use('fivethirtyeight')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/pandas_datareader/compat/__init__.py:7: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  from pandas.util.testing import assert_frame_equal\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EySqaP1ptVDO",
        "colab_type": "text"
      },
      "source": [
        "## Get Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6fkpE5_-sB2V",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "510f891a-468b-4b06-a2b8-a40e5e76ef44"
      },
      "source": [
        "# Get the stock quote\n",
        "df = web.DataReader('AAPL', data_source='yahoo', start='2012-01-01', end='2020-07-31')\n",
        "\n",
        "df.head()"
      ],
      "execution_count": null,
      "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>High</th>\n",
              "      <th>Low</th>\n",
              "      <th>Open</th>\n",
              "      <th>Close</th>\n",
              "      <th>Volume</th>\n",
              "      <th>Adj Close</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>2012-01-03</th>\n",
              "      <td>58.928570</td>\n",
              "      <td>58.428570</td>\n",
              "      <td>58.485714</td>\n",
              "      <td>58.747143</td>\n",
              "      <td>75555200.0</td>\n",
              "      <td>50.765709</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2012-01-04</th>\n",
              "      <td>59.240002</td>\n",
              "      <td>58.468571</td>\n",
              "      <td>58.571430</td>\n",
              "      <td>59.062859</td>\n",
              "      <td>65005500.0</td>\n",
              "      <td>51.038536</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2012-01-05</th>\n",
              "      <td>59.792858</td>\n",
              "      <td>58.952858</td>\n",
              "      <td>59.278572</td>\n",
              "      <td>59.718571</td>\n",
              "      <td>67817400.0</td>\n",
              "      <td>51.605175</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2012-01-06</th>\n",
              "      <td>60.392857</td>\n",
              "      <td>59.888573</td>\n",
              "      <td>59.967144</td>\n",
              "      <td>60.342857</td>\n",
              "      <td>79573200.0</td>\n",
              "      <td>52.144630</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2012-01-09</th>\n",
              "      <td>61.107143</td>\n",
              "      <td>60.192856</td>\n",
              "      <td>60.785713</td>\n",
              "      <td>60.247143</td>\n",
              "      <td>98506100.0</td>\n",
              "      <td>52.061932</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                 High        Low       Open      Close      Volume  Adj Close\n",
              "Date                                                                         \n",
              "2012-01-03  58.928570  58.428570  58.485714  58.747143  75555200.0  50.765709\n",
              "2012-01-04  59.240002  58.468571  58.571430  59.062859  65005500.0  51.038536\n",
              "2012-01-05  59.792858  58.952858  59.278572  59.718571  67817400.0  51.605175\n",
              "2012-01-06  60.392857  59.888573  59.967144  60.342857  79573200.0  52.144630\n",
              "2012-01-09  61.107143  60.192856  60.785713  60.247143  98506100.0  52.061932"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OuwqagUMtIjU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "d0d7034c-d1ac-4f73-dbfc-f1e12392d1dc"
      },
      "source": [
        "df.tail()"
      ],
      "execution_count": null,
      "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>High</th>\n",
              "      <th>Low</th>\n",
              "      <th>Open</th>\n",
              "      <th>Close</th>\n",
              "      <th>Volume</th>\n",
              "      <th>Adj Close</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>2020-07-27</th>\n",
              "      <td>379.619995</td>\n",
              "      <td>373.920013</td>\n",
              "      <td>374.839996</td>\n",
              "      <td>379.239990</td>\n",
              "      <td>30303500.0</td>\n",
              "      <td>378.557434</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-07-28</th>\n",
              "      <td>378.200012</td>\n",
              "      <td>372.989990</td>\n",
              "      <td>377.470001</td>\n",
              "      <td>373.010010</td>\n",
              "      <td>25906400.0</td>\n",
              "      <td>372.338684</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-07-29</th>\n",
              "      <td>380.920013</td>\n",
              "      <td>374.850006</td>\n",
              "      <td>375.000000</td>\n",
              "      <td>380.160004</td>\n",
              "      <td>22582300.0</td>\n",
              "      <td>379.475800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-07-30</th>\n",
              "      <td>385.190002</td>\n",
              "      <td>375.070007</td>\n",
              "      <td>376.750000</td>\n",
              "      <td>384.760010</td>\n",
              "      <td>39532500.0</td>\n",
              "      <td>384.067535</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-07-31</th>\n",
              "      <td>425.660004</td>\n",
              "      <td>403.299988</td>\n",
              "      <td>411.540009</td>\n",
              "      <td>425.040009</td>\n",
              "      <td>93584200.0</td>\n",
              "      <td>424.275024</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  High         Low  ...      Volume   Adj Close\n",
              "Date                                ...                        \n",
              "2020-07-27  379.619995  373.920013  ...  30303500.0  378.557434\n",
              "2020-07-28  378.200012  372.989990  ...  25906400.0  372.338684\n",
              "2020-07-29  380.920013  374.850006  ...  22582300.0  379.475800\n",
              "2020-07-30  385.190002  375.070007  ...  39532500.0  384.067535\n",
              "2020-07-31  425.660004  403.299988  ...  93584200.0  424.275024\n",
              "\n",
              "[5 rows x 6 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "03jun49HtN4l",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "abb1ba02-f4d6-4b94-c431-7dcc6ffd4d62"
      },
      "source": [
        "# Get the number of rows and columns in the data set\n",
        "df.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2159, 6)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1RMmMrK2togG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gn6J4zgptrbw",
        "colab_type": "text"
      },
      "source": [
        "## Visualize the data\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rKVE89-ztsNd",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "019fbfd5-15cc-43b2-e22a-d8067295f099"
      },
      "source": [
        "plt.figure(figsize=(16,8))\n",
        "plt.title('Close Price History', fontsize=20)\n",
        "plt.plot(df['Close'])\n",
        "plt.xlabel('Date', fontsize=18)\n",
        "plt.ylabel('Close Price USD ($)', fontsize=18)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "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": "LGFFC6AgtsQE",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "quk-ptpU6lVy",
        "colab_type": "text"
      },
      "source": [
        "## Create the Training Set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-g7wiA59tsbI",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "b21e9d8d-5fbd-4165-de58-9116246905b2"
      },
      "source": [
        "# Create a new dataframe with only the 'Close' column\n",
        "data = df.filter(['Close'])\n",
        "# Convert the dataframe to a numpy array\n",
        "dataset = data.values\n",
        "# Get the number of rows to train a model on\n",
        "training_data_len = math.ceil(len(dataset) * .8)\n",
        "\n",
        "training_data_len"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1728"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c2ZPiUYpvqaW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        },
        "outputId": "557955ea-4484-4429-ec74-9b6ef09cb62b"
      },
      "source": [
        "# Scale the data to be (0, 1) inclusively\n",
        "\n",
        "scaler = MinMaxScaler(feature_range=(0,1))\n",
        "scaled_data = scaler.fit_transform(dataset)\n",
        "\n",
        "scaled_data"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.00800851],\n",
              "       [0.00886353],\n",
              "       [0.01063932],\n",
              "       ...,\n",
              "       [0.87845632],\n",
              "       [0.89091402],\n",
              "       [1.        ]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "POnRoX4zwAX2",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6003c6b0-d914-471e-842e-acefa2ab48ac"
      },
      "source": [
        "# Create the training dataset from the scaled dataset\n",
        "train_data = scaled_data[:training_data_len, :]\n",
        "\n",
        "# Split the data into x_train and y_train dataset\n",
        "x_train = []\n",
        "y_train = []\n",
        "\n",
        "for i in range(60, len(train_data)):\n",
        "  x_train.append(train_data[i-60: i, 0])\n",
        "  y_train.append(train_data[i,0])\n",
        "  \n",
        "  if i <= 61:\n",
        "    print(x_train)\n",
        "    print(y_train)\n",
        "    print()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[array([0.00800851, 0.00886353, 0.01063932, 0.01233001, 0.0120708 ,\n",
            "       0.01265499, 0.01238804, 0.01193925, 0.01132798, 0.01321984,\n",
            "       0.01492601, 0.01439984, 0.01151755, 0.0142683 , 0.01156011,\n",
            "       0.02171583, 0.02093045, 0.0219557 , 0.02417255, 0.02551503,\n",
            "       0.02540284, 0.02498888, 0.02675307, 0.02841279, 0.03029307,\n",
            "       0.03333011, 0.03970984, 0.03980655, 0.04335816, 0.04601218,\n",
            "       0.0414508 , 0.04320727, 0.04317245, 0.0480975 , 0.04739723,\n",
            "       0.04869328, 0.05102233, 0.05231841, 0.05605184, 0.05877165,\n",
            "       0.05955701, 0.0598317 , 0.05518136, 0.05405938, 0.05422575,\n",
            "       0.05859753, 0.05982784, 0.06247025, 0.0686991 , 0.07700938,\n",
            "       0.0754541 , 0.07545796, 0.08146629, 0.08334654, 0.08200792,\n",
            "       0.08078538, 0.07951253, 0.08374116, 0.08664281, 0.08785763])]\n",
            "[0.08485539492813565]\n",
            "\n",
            "[array([0.00800851, 0.00886353, 0.01063932, 0.01233001, 0.0120708 ,\n",
            "       0.01265499, 0.01238804, 0.01193925, 0.01132798, 0.01321984,\n",
            "       0.01492601, 0.01439984, 0.01151755, 0.0142683 , 0.01156011,\n",
            "       0.02171583, 0.02093045, 0.0219557 , 0.02417255, 0.02551503,\n",
            "       0.02540284, 0.02498888, 0.02675307, 0.02841279, 0.03029307,\n",
            "       0.03333011, 0.03970984, 0.03980655, 0.04335816, 0.04601218,\n",
            "       0.0414508 , 0.04320727, 0.04317245, 0.0480975 , 0.04739723,\n",
            "       0.04869328, 0.05102233, 0.05231841, 0.05605184, 0.05877165,\n",
            "       0.05955701, 0.0598317 , 0.05518136, 0.05405938, 0.05422575,\n",
            "       0.05859753, 0.05982784, 0.06247025, 0.0686991 , 0.07700938,\n",
            "       0.0754541 , 0.07545796, 0.08146629, 0.08334654, 0.08200792,\n",
            "       0.08078538, 0.07951253, 0.08374116, 0.08664281, 0.08785763]), array([0.00886353, 0.01063932, 0.01233001, 0.0120708 , 0.01265499,\n",
            "       0.01238804, 0.01193925, 0.01132798, 0.01321984, 0.01492601,\n",
            "       0.01439984, 0.01151755, 0.0142683 , 0.01156011, 0.02171583,\n",
            "       0.02093045, 0.0219557 , 0.02417255, 0.02551503, 0.02540284,\n",
            "       0.02498888, 0.02675307, 0.02841279, 0.03029307, 0.03333011,\n",
            "       0.03970984, 0.03980655, 0.04335816, 0.04601218, 0.0414508 ,\n",
            "       0.04320727, 0.04317245, 0.0480975 , 0.04739723, 0.04869328,\n",
            "       0.05102233, 0.05231841, 0.05605184, 0.05877165, 0.05955701,\n",
            "       0.0598317 , 0.05518136, 0.05405938, 0.05422575, 0.05859753,\n",
            "       0.05982784, 0.06247025, 0.0686991 , 0.07700938, 0.0754541 ,\n",
            "       0.07545796, 0.08146629, 0.08334654, 0.08200792, 0.08078538,\n",
            "       0.07951253, 0.08374116, 0.08664281, 0.08785763, 0.08485539])]\n",
            "[0.08485539492813565, 0.08086662151222263]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3pwS2Rbkw9Rm",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "74288378-f1fd-40f8-b90d-3a84b7c17942"
      },
      "source": [
        "# Convert the x_train and y_train to numpy arrays\n",
        "x_train, y_train = np.array(x_train), np.array(y_train)\n",
        "\n",
        "x_train.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1668, 60)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ltWjUZO6xvA2",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "9c40a244-dbbc-4653-9997-d39f4208940f"
      },
      "source": [
        "# Reshape the x_train\n",
        "x_train = np.reshape(x_train, (x_train.shape[0], x_train.shape[1], 1))\n",
        "x_train.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1668, 60, 1)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zNEqiCosyHxk",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "u-WQe_zTyUAR",
        "colab_type": "text"
      },
      "source": [
        "## LSTM Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4qPkR5kPyWOD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# build the LSTM model\n",
        "model = Sequential()\n",
        "model.add(LSTM(50, return_sequences=True, input_shape=(x_train.shape[1],1)))\n",
        "model.add(LSTM(50, return_sequences=False))\n",
        "model.add(Dense(25))\n",
        "model.add(Dense(1))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SBDl5fKCyaBx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Compile the model\n",
        "model.compile(optimizer='adam', loss='mean_squared_error')\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vzMYHkVhzCju",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "d781932c-b81c-48df-99d5-2e9592cdaf1d"
      },
      "source": [
        "# Train the model\n",
        "model.fit(x_train, y_train, batch_size=1, epochs=1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1668/1668 [==============================] - 39s 23ms/step - loss: 3.6290e-04\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7efff12a5fd0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D0lvyET7zNRf",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0L6ymmfLzcSR",
        "colab_type": "text"
      },
      "source": [
        "## Testing data set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ePlSYZg_zdU0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Create a new array containing caled values from index ... to ...\n",
        "test_data = scaled_data[training_data_len-60:, :]\n",
        "# Create the data sets x_test, and y_test\n",
        "x_test = []\n",
        "y_test = dataset[training_data_len:, :]\n",
        "\n",
        "for i in range(60, len(test_data)):\n",
        "  x_test.append(test_data[i-60:i, 0])\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z92fXylyzrFs",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Convert to numpy array\n",
        "x_test = np.array(x_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZLTngBw50b3n",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "ccb2b109-19af-4e0b-bb59-b66842d365f8"
      },
      "source": [
        "# Reshape the data\n",
        "x_test = np.reshape(x_test, (x_test.shape[0], x_test.shape[1], 1))\n",
        "x_test.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(431, 60, 1)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A7cl-VMV0j9d",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Get the model predicted price values\n",
        "predictions = model.predict(x_test)\n",
        "predictions = scaler.inverse_transform(predictions)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6VMxc6Zd1Ete",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "52794de0-958a-483d-9778-3b34148959db"
      },
      "source": [
        "# Get the root mean squared error (RMSE)\n",
        "rmse = np.sqrt(np.mean(((predictions- y_test)**2)))\n",
        "rmse"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "11.168985852343347"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Tw8kZsJI1nrM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 139
        },
        "outputId": "df0717d2-0e9f-4b43-e7d4-f3c481182cce"
      },
      "source": [
        "# Plot the data\n",
        "\n",
        "train = data[:training_data_len]\n",
        "valid = data[training_data_len:]\n",
        "valid['Predictions'] = predictions\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/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": "markdown",
      "metadata": {
        "id": "lOLE6pFF602h",
        "colab_type": "text"
      },
      "source": [
        "## Results"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gQlNm1Yt9yQ1",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 425
        },
        "outputId": "3a2dcf6a-ddc5-447d-d750-3a8fe0d74ddc"
      },
      "source": [
        "\n",
        "# Visualize the data\n",
        "plt.figure(figsize=(16,8))\n",
        "plt.title('Model',  fontsize=20)\n",
        "plt.xlabel('Date', fontsize=18)\n",
        "plt.ylabel('Close Price SD ($)', fontsize=18)\n",
        "plt.plot(train['Close'])\n",
        "plt.plot(valid[['Close', 'Predictions']])\n",
        "plt.legend(['Train', 'Validation', 'Predictions'], loc='lower right')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "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"
          }
        }
      ]
    }
  ]
}