material/deep_learning/financial_data_predict_commodity_price.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "financial_data_predict_commodity_price.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "t3J2t4Ymf0UF",
        "ESYLyhWeiRrf"
      ]
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "data from https://www.kaggle.com/mattiuzc/commodity-futures-price-history\n"
      ],
      "metadata": {
        "id": "ohP561iSdS78"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 데이터 압축 풀기"
      ],
      "metadata": {
        "id": "xIEVVEF1frW-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://github.com/dhrim/deep_learning_data/raw/master/commodity.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZnvrQDxheBkc",
        "outputId": "0a8d3190-363a-43ff-c50b-39f0caf70981"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2022-05-20 07:48:59--  https://github.com/dhrim/deep_learning_data/raw/master/commodity.zip\n",
            "Resolving github.com (github.com)... 140.82.112.3\n",
            "Connecting to github.com (github.com)|140.82.112.3|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://raw.githubusercontent.com/dhrim/deep_learning_data/master/commodity.zip [following]\n",
            "--2022-05-20 07:48:59--  https://raw.githubusercontent.com/dhrim/deep_learning_data/master/commodity.zip\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 1835488 (1.8M) [application/zip]\n",
            "Saving to: ‘commodity.zip’\n",
            "\n",
            "commodity.zip       100%[===================>]   1.75M  --.-KB/s    in 0.06s   \n",
            "\n",
            "2022-05-20 07:48:59 (30.7 MB/s) - ‘commodity.zip’ saved [1835488/1835488]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TJ7oZpj0dDrn",
        "outputId": "9c7cbfee-127b-4eff-a877-6ba497e96bef"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  commodity.zip\n",
            "   creating: commodity/\n",
            "  inflating: commodity/Natural Gas.csv  \n",
            "  inflating: commodity/Soybean.csv   \n",
            "  inflating: commodity/Gold.csv      \n",
            "  inflating: commodity/Platinum.csv  \n",
            "  inflating: commodity/Heating Oil.csv  \n",
            "  inflating: commodity/Palladium.csv  \n",
            "  inflating: commodity/Silver.csv    \n",
            "  inflating: commodity/Lumber.csv    \n",
            "  inflating: commodity/Lean Hogs.csv  \n",
            "  inflating: commodity/Soybean Meal.csv  \n",
            "  inflating: commodity/Copper.csv    \n",
            "  inflating: commodity/Live Cattle.csv  \n",
            "  inflating: commodity/Soybean Oil.csv  \n",
            "  inflating: commodity/Brent Crude Oil.csv  \n",
            "  inflating: commodity/Cotton.csv    \n",
            "  inflating: commodity/RBOB Gasoline.csv  \n",
            "  inflating: commodity/Oat.csv       \n",
            "  inflating: commodity/Coffee.csv    \n",
            "  inflating: commodity/Wheat.csv     \n",
            "  inflating: commodity/Sugar.csv     \n",
            "  inflating: commodity/Corn.csv      \n",
            "  inflating: commodity/Crude Oil.csv  \n",
            "  inflating: commodity/Cocoa.csv     \n",
            "  inflating: commodity/Feeder Cattle.csv  \n"
          ]
        }
      ],
      "source": [
        "!rm -rf commodity\n",
        "!unzip commodity.zip"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "JRjPtpGXdW0L"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 파라매터 설정"
      ],
      "metadata": {
        "id": "37eayrO3y-3K"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "SEQUENCE_LENGTH = 10\n",
        "OFFSET = 1\n",
        "BATCH_SIZE = 32"
      ],
      "metadata": {
        "id": "F2iE0M7Hy6Q6"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 데이터 로딩"
      ],
      "metadata": {
        "id": "Zu_4OtD_ftsY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv(\"commodity/Silver.csv\")\n",
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "g4lWKXiCdcP8",
        "outputId": "56c3ba23-20c8-4fda-8c22-06258b035236"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         Date   Open   High    Low  Close  Adj Close  Volume\n",
              "0  2000-08-30  4.950  4.950  4.895  4.930      4.930  2319.0\n",
              "1  2000-08-31  4.920  5.045  4.920  5.003      5.003  1612.0\n",
              "2  2000-09-01  5.035  5.035  4.985  5.004      5.004  1375.0\n",
              "3  2000-09-04    NaN    NaN    NaN    NaN        NaN     NaN\n",
              "4  2000-09-05  4.990  5.005  4.975  4.998      4.998   856.0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a1e71c0d-2165-4286-8e2c-4122b1a261fb\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>Date</th>\n",
              "      <th>Open</th>\n",
              "      <th>High</th>\n",
              "      <th>Low</th>\n",
              "      <th>Close</th>\n",
              "      <th>Adj Close</th>\n",
              "      <th>Volume</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2000-08-30</td>\n",
              "      <td>4.950</td>\n",
              "      <td>4.950</td>\n",
              "      <td>4.895</td>\n",
              "      <td>4.930</td>\n",
              "      <td>4.930</td>\n",
              "      <td>2319.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2000-08-31</td>\n",
              "      <td>4.920</td>\n",
              "      <td>5.045</td>\n",
              "      <td>4.920</td>\n",
              "      <td>5.003</td>\n",
              "      <td>5.003</td>\n",
              "      <td>1612.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2000-09-01</td>\n",
              "      <td>5.035</td>\n",
              "      <td>5.035</td>\n",
              "      <td>4.985</td>\n",
              "      <td>5.004</td>\n",
              "      <td>5.004</td>\n",
              "      <td>1375.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2000-09-04</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2000-09-05</td>\n",
              "      <td>4.990</td>\n",
              "      <td>5.005</td>\n",
              "      <td>4.975</td>\n",
              "      <td>4.998</td>\n",
              "      <td>4.998</td>\n",
              "      <td>856.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a1e71c0d-2165-4286-8e2c-4122b1a261fb')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-a1e71c0d-2165-4286-8e2c-4122b1a261fb button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-a1e71c0d-2165-4286-8e2c-4122b1a261fb');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(len(df))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "er07xkbpdnLf",
        "outputId": "1a0e4687-1670-4dca-fa17-208d163946fd"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "5291\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 결측치 처리"
      ],
      "metadata": {
        "id": "kh7oOvIifwgm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "suMkYKdTduGY",
        "outputId": "bdb6c943-6e50-4bca-87bf-ce7cbe506af6"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 5291 entries, 0 to 5290\n",
            "Data columns (total 7 columns):\n",
            " #   Column     Non-Null Count  Dtype  \n",
            "---  ------     --------------  -----  \n",
            " 0   Date       5291 non-null   object \n",
            " 1   Open       5181 non-null   float64\n",
            " 2   High       5181 non-null   float64\n",
            " 3   Low        5181 non-null   float64\n",
            " 4   Close      5181 non-null   float64\n",
            " 5   Adj Close  5181 non-null   float64\n",
            " 6   Volume     5181 non-null   float64\n",
            "dtypes: float64(6), object(1)\n",
            "memory usage: 289.5+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.isnull().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8VlYB3fEd0bM",
        "outputId": "299f1236-f4fc-4d7e-a25f-b1701fba914d"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Date           0\n",
              "Open         110\n",
              "High         110\n",
              "Low          110\n",
              "Close        110\n",
              "Adj Close    110\n",
              "Volume       110\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df = df.dropna()"
      ],
      "metadata": {
        "id": "uMf9RDKZd9WC"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df.isnull().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dnO_h4QJeH8t",
        "outputId": "d16ec8ee-3bcd-4226-adae-8a3db1853a97"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Date         0\n",
              "Open         0\n",
              "High         0\n",
              "Low          0\n",
              "Close        0\n",
              "Adj Close    0\n",
              "Volume       0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YMIpK7ebeROK",
        "outputId": "52cab458-b6a5-4ef3-9b2d-15fe1730d008"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 5181 entries, 0 to 5290\n",
            "Data columns (total 7 columns):\n",
            " #   Column     Non-Null Count  Dtype  \n",
            "---  ------     --------------  -----  \n",
            " 0   Date       5181 non-null   object \n",
            " 1   Open       5181 non-null   float64\n",
            " 2   High       5181 non-null   float64\n",
            " 3   Low        5181 non-null   float64\n",
            " 4   Close      5181 non-null   float64\n",
            " 5   Adj Close  5181 non-null   float64\n",
            " 6   Volume     5181 non-null   float64\n",
            "dtypes: float64(6), object(1)\n",
            "memory usage: 323.8+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 정렬 순서 확인"
      ],
      "metadata": {
        "id": "t3J2t4Ymf0UF"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df.sort_values(by=[\"Date\"])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "_ms5GTSxew67",
        "outputId": "c0e335dc-43f9-4a2c-adee-a1c7ad6d071d"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "            Date       Open       High        Low      Close  Adj Close  \\\n",
              "0     2000-08-30   4.950000   4.950000   4.895000   4.930000   4.930000   \n",
              "1     2000-08-31   4.920000   5.045000   4.920000   5.003000   5.003000   \n",
              "2     2000-09-01   5.035000   5.035000   4.985000   5.004000   5.004000   \n",
              "4     2000-09-05   4.990000   5.005000   4.975000   4.998000   4.998000   \n",
              "5     2000-09-06   5.000000   5.000000   4.980000   4.983000   4.983000   \n",
              "...          ...        ...        ...        ...        ...        ...   \n",
              "5286  2021-06-03  27.190001  27.459999  27.190001  27.459999  27.459999   \n",
              "5287  2021-06-04  27.879000  27.879000  27.879000  27.879000  27.879000   \n",
              "5288  2021-06-07  27.615000  28.000999  27.615000  28.000999  28.000999   \n",
              "5289  2021-06-08  28.010000  28.010000  27.705000  27.715000  27.715000   \n",
              "5290  2021-06-09  27.695000  27.985001  27.695000  27.985001  27.985001   \n",
              "\n",
              "      Volume  \n",
              "0     2319.0  \n",
              "1     1612.0  \n",
              "2     1375.0  \n",
              "4      856.0  \n",
              "5      232.0  \n",
              "...      ...  \n",
              "5286    20.0  \n",
              "5287     0.0  \n",
              "5288     4.0  \n",
              "5289    27.0  \n",
              "5290    27.0  \n",
              "\n",
              "[5181 rows x 7 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-cdb4ec21-4e36-43c6-a941-7d4e7c077c8f\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>Date</th>\n",
              "      <th>Open</th>\n",
              "      <th>High</th>\n",
              "      <th>Low</th>\n",
              "      <th>Close</th>\n",
              "      <th>Adj Close</th>\n",
              "      <th>Volume</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2000-08-30</td>\n",
              "      <td>4.950000</td>\n",
              "      <td>4.950000</td>\n",
              "      <td>4.895000</td>\n",
              "      <td>4.930000</td>\n",
              "      <td>4.930000</td>\n",
              "      <td>2319.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2000-08-31</td>\n",
              "      <td>4.920000</td>\n",
              "      <td>5.045000</td>\n",
              "      <td>4.920000</td>\n",
              "      <td>5.003000</td>\n",
              "      <td>5.003000</td>\n",
              "      <td>1612.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2000-09-01</td>\n",
              "      <td>5.035000</td>\n",
              "      <td>5.035000</td>\n",
              "      <td>4.985000</td>\n",
              "      <td>5.004000</td>\n",
              "      <td>5.004000</td>\n",
              "      <td>1375.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2000-09-05</td>\n",
              "      <td>4.990000</td>\n",
              "      <td>5.005000</td>\n",
              "      <td>4.975000</td>\n",
              "      <td>4.998000</td>\n",
              "      <td>4.998000</td>\n",
              "      <td>856.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>2000-09-06</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>4.980000</td>\n",
              "      <td>4.983000</td>\n",
              "      <td>4.983000</td>\n",
              "      <td>232.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5286</th>\n",
              "      <td>2021-06-03</td>\n",
              "      <td>27.190001</td>\n",
              "      <td>27.459999</td>\n",
              "      <td>27.190001</td>\n",
              "      <td>27.459999</td>\n",
              "      <td>27.459999</td>\n",
              "      <td>20.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5287</th>\n",
              "      <td>2021-06-04</td>\n",
              "      <td>27.879000</td>\n",
              "      <td>27.879000</td>\n",
              "      <td>27.879000</td>\n",
              "      <td>27.879000</td>\n",
              "      <td>27.879000</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5288</th>\n",
              "      <td>2021-06-07</td>\n",
              "      <td>27.615000</td>\n",
              "      <td>28.000999</td>\n",
              "      <td>27.615000</td>\n",
              "      <td>28.000999</td>\n",
              "      <td>28.000999</td>\n",
              "      <td>4.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5289</th>\n",
              "      <td>2021-06-08</td>\n",
              "      <td>28.010000</td>\n",
              "      <td>28.010000</td>\n",
              "      <td>27.705000</td>\n",
              "      <td>27.715000</td>\n",
              "      <td>27.715000</td>\n",
              "      <td>27.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5290</th>\n",
              "      <td>2021-06-09</td>\n",
              "      <td>27.695000</td>\n",
              "      <td>27.985001</td>\n",
              "      <td>27.695000</td>\n",
              "      <td>27.985001</td>\n",
              "      <td>27.985001</td>\n",
              "      <td>27.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5181 rows × 7 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-cdb4ec21-4e36-43c6-a941-7d4e7c077c8f')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-cdb4ec21-4e36-43c6-a941-7d4e7c077c8f button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-cdb4ec21-4e36-43c6-a941-7d4e7c077c8f');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sorted_index = df.sort_values(by=[\"Date\"]).index"
      ],
      "metadata": {
        "id": "ZRs14Udwe5SR"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(sorted_index)\n",
        "print(sorted_index - df.index)\n",
        "print(sum(sorted_index - df.index))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ybOz2otyfTjE",
        "outputId": "dafe7da2-4173-4006-9cc8-a55f77248000"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Int64Index([   0,    1,    2,    4,    5,    6,    7,    8,    9,   10,\n",
            "            ...\n",
            "            5281, 5282, 5283, 5284, 5285, 5286, 5287, 5288, 5289, 5290],\n",
            "           dtype='int64', length=5181)\n",
            "Int64Index([0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
            "            ...\n",
            "            0, 0, 0, 0, 0, 0, 0, 0, 0, 0],\n",
            "           dtype='int64', length=5181)\n",
            "0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(df['Close'] - df['Adj Close'])\n",
        "print(sum(df['Close'] - df['Adj Close']))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bzUbYVEJhVab",
        "outputId": "1de93209-88d5-4dba-ea2c-2fcbf35309f9"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0       0.0\n",
            "1       0.0\n",
            "2       0.0\n",
            "4       0.0\n",
            "5       0.0\n",
            "       ... \n",
            "5286    0.0\n",
            "5287    0.0\n",
            "5288    0.0\n",
            "5289    0.0\n",
            "5290    0.0\n",
            "Length: 5181, dtype: float64\n",
            "0.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.drop(columns=[\"Date\", \"Adj Close\"], inplace=True)"
      ],
      "metadata": {
        "id": "RU6dxvFnhqTB"
      },
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for i in range(len(df.columns)):\n",
        "    print(df.columns[i])\n",
        "    plt.hist(df.iloc[:,i], bins=1000)\n",
        "    plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "LF73-1VEitmC",
        "outputId": "23cf2525-9feb-4afb-f3e5-bb2c60d12189"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Open\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "High\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Low\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Close\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Volume\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAD4CAYAAAAZ1BptAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAV4UlEQVR4nO3dfayed33f8fenzgMVT3HIWeTZZjbUXWUm1WT3QiZoxWBNnLTDYUPIqGo8GsntlkigdWuTIi0UOgm2QTZUCApLhoMojhtAsRhdcEM0tD/ycAwmxAlpDiRRbDnxKQ4BlCmb0+/+uH8numP8s4/Puc+DrfdLunWu63s9fa8rx/mc6+GcK1WFJEnH8wtL3YAkafkyJCRJXYaEJKnLkJAkdRkSkqSus5a6gRO54IILat26dUvdhiSdVvbu3fs3VTUxjnUt65BYt24dk5OTS92GJJ1Wkjw5rnV5uUmS1GVISJK6DAlJUpchIUnqMiQkSV2GhCSpy5CQJHUZEpKkLkNCktRlSEiSugwJSVLXrEMiyYok30nytTa+Psl9SaaS3J7knFY/t41PtenrRtZxfas/muSyce+MJGm8TuVM4gPAIyPjHwdurKpfAp4Frm71q4FnW/3GNh9JNgJbgTcBm4HPJFkxv/YlSQtpViGRZA3wm8B/a+MB3gHc0WbZAVzZhre0cdr0d7b5twA7q+qFqnocmAIuHsdOSJIWxmzPJP4L8IfA37bx1wE/rqqjbfwAsLoNrwaeAmjTn2vzv1Q/zjIvSbI9yWSSyenp6VPYFUnSuJ00JJL8FnC4qvYuQj9U1c1VNaiqwcTEWN6ZIUmao9m8dOitwLuSXAG8AngN8F+B85Kc1c4W1gAH2/wHgbXAgSRnAa8FfjRSnzG6jCRpGTrpmURVXV9Va6pqHcMbz9+sqt8G7gHe02bbBtzZhne3cdr0b1ZVtfrW9vTTemADcP/Y9kSSNHbzeX3pHwE7k/wp8B3glla/BfhCkingCMNgoar2J9kFPAwcBa6pqhfnsX1J0gLL8If85WkwGJTvuJakU5Nkb1UNxrEuf+NaktRlSEiSugwJSVKXISFJ6jIkJEldhoQkqcuQkCR1GRKSpC5DQpLUZUhIkroMCUlSlyEhSeoyJCRJXYaEJKnLkJAkdRkSkqSuk4ZEklckuT/Jd5PsT/Inrf75JI8n2dc+m1o9ST6VZCrJg0kuGlnXtiSPtc+23jYlScvDbF5f+gLwjqr6WZKzgf+d5C/btH9XVXccM//lDN9fvQF4C3AT8JYk5wM3AAOggL1JdlfVs+PYEUnS+J30TKKGftZGz26fE73zdAtwW1vuXuC8JKuAy4A9VXWkBcMeYPP82pckLaRZ3ZNIsiLJPuAww//R39cm/Yd2SenGJOe22mrgqZHFD7Rar37strYnmUwyOT09fYq7I0kap1mFRFW9WFWbgDXAxUn+AXA98CvAPwLOB/5oHA1V1c1VNaiqwcTExDhWKUmao1N6uqmqfgzcA2yuqkPtktILwH8HLm6zHQTWjiy2ptV6dUnSMjWbp5smkpzXhn8R+A3g++0+A0kCXAk81BbZDVzVnnK6BHiuqg4BdwGXJlmZZCVwaatJkpap2TzdtArYkWQFw1DZVVVfS/LNJBNAgH3A77f5vw5cAUwBzwPvB6iqI0k+CjzQ5vtIVR0Z365IksYtVSd6UGlpDQaDmpycXOo2JOm0kmRvVQ3GsS5/41qS1GVISJK6DAlJUpchIUnqMiQkSV2GhCSpy5CQJHUZEpKkLkNCktRlSEiSugwJSVKXISFJ6jIkJEldhoQkqcuQkCR1GRKSpK7ZvL70FUnuT/LdJPuT/Emrr09yX5KpJLcnOafVz23jU236upF1Xd/qjya5bKF2SpI0HrM5k3gBeEdV/SqwCdjc3l39ceDGqvol4Fng6jb/1cCzrX5jm48kG4GtwJuAzcBn2itRJUnL1ElDooZ+1kbPbp8C3gHc0eo7gCvb8JY2Tpv+ziRp9Z1V9UJVPc7wHdgXj2UvJEkLYlb3JJKsSLIPOAzsAX4A/LiqjrZZDgCr2/Bq4CmANv054HWj9eMsM7qt7Ukmk0xOT0+f+h5JksZmViFRVS9W1SZgDcOf/n9loRqqqpuralBVg4mJiYXajCRpFk7p6aaq+jFwD/CPgfOSnNUmrQEOtuGDwFqANv21wI9G68dZRpK0DM3m6aaJJOe14V8EfgN4hGFYvKfNtg24sw3vbuO06d+sqmr1re3pp/XABuD+ce2IJGn8zjr5LKwCdrQnkX4B2FVVX0vyMLAzyZ8C3wFuafPfAnwhyRRwhOETTVTV/iS7gIeBo8A1VfXieHdHkjROGf6QvzwNBoOanJxc6jYk6bSSZG9VDcaxLn/jWpLUZUhIkroMCUlSlyEhSeoyJCRJXYaEJKnLkJAkdRkSkqQuQ0KS1GVISJK6DAlJUpchIUnqMiQkSV2GhCSpy5CQJHUZEpKkrtm8vnRtknuSPJxkf5IPtPqHkxxMsq99rhhZ5vokU0keTXLZSH1zq00luW5hdkmSNC6zeX3pUeAPqurbSV4N7E2yp027sar+8+jMSTYyfGXpm4C/C/xVkl9ukz/N8B3ZB4AHkuyuqofHsSOSpPE7aUhU1SHgUBv+aZJHgNUnWGQLsLOqXgAeb++6vrhNm6qqHwIk2dnmNSQkaZk6pXsSSdYBbwbua6VrkzyY5NYkK1ttNfDUyGIHWq1XP3Yb25NMJpmcnp4+lfYkSWM265BI8irgy8AHq+onwE3AG4FNDM80PjGOhqrq5qoaVNVgYmJiHKuUJM3RbO5JkORshgHxxar6CkBVPTMy/XPA19roQWDtyOJrWo0T1CVJy9Bsnm4KcAvwSFV9cqS+amS2dwMPteHdwNYk5yZZD2wA7gceADYkWZ/kHIY3t3ePZzckSQthNmcSbwV+B/hekn2t9sfA+5JsAgp4Avg9gKran2QXwxvSR4FrqupFgCTXAncBK4Bbq2r/GPdFkjRmqaql7qFrMBjU5OTkUrchSaeVJHurajCOdfkb15KkLkNCktRlSEiSugwJSVKXISFJ6jIkJEldhoQkqcuQkCR1GRKSpC5DQpLUZUhIkroMCUlSlyEhSeoyJCRJXYaEJKnLkJAkdRkSkqSu2bzjem2Se5I8nGR/kg+0+vlJ9iR5rH1d2epJ8qkkU0keTHLRyLq2tfkfS7Jt4XZLkjQOszmTOAr8QVVtBC4BrkmyEbgOuLuqNgB3t3GAy4EN7bMduAmGoQLcALwFuBi4YSZYJEnL00lDoqoOVdW32/BPgUeA1cAWYEebbQdwZRveAtxWQ/cC5yVZBVwG7KmqI1X1LLAH2DzWvZEkjdUp3ZNIsg54M3AfcGFVHWqTngYubMOrgadGFjvQar36sdvYnmQyyeT09PSptCdJGrNZh0SSVwFfBj5YVT8ZnVZVBdQ4Gqqqm6tqUFWDiYmJcaxSkjRHswqJJGczDIgvVtVXWvmZdhmJ9vVwqx8E1o4svqbVenVJ0jI1m6ebAtwCPFJVnxyZtBuYeUJpG3DnSP2q9pTTJcBz7bLUXcClSVa2G9aXtpokaZk6axbzvBX4HeB7Sfa12h8DHwN2JbkaeBJ4b5v2deAKYAp4Hng/QFUdSfJR4IE230eq6shY9kKStCAyvJ2wPA0Gg5qcnFzqNiTptJJkb1UNxrEuf+NaktRlSEiSugwJSVKXISFJ6jIkJEldhoQkqcuQkCR1GRKSpC5DQpLUZUhIkroMCUlSlyEhSeoyJCRJXYaEJKnLkJAkdRkSkqSu2by+9NYkh5M8NFL7cJKDSfa1zxUj065PMpXk0SSXjdQ3t9pUkuvGvyuSpHGbzZnE54HNx6nfWFWb2ufrAEk2AluBN7VlPpNkRZIVwKeBy4GNwPvavJKkZeyk77iuqm8lWTfL9W0BdlbVC8DjSaaAi9u0qar6IUCSnW3eh0+5Y0nSopnPPYlrkzzYLketbLXVwFMj8xxotV795yTZnmQyyeT09PQ82pMkzddcQ+Im4I3AJuAQ8IlxNVRVN1fVoKoGExMT41qtJGkOTnq56Xiq6pmZ4SSfA77WRg8Ca0dmXdNqnKAuSVqm5nQmkWTVyOi7gZknn3YDW5Ocm2Q9sAG4H3gA2JBkfZJzGN7c3j33tiVJi+GkZxJJvgS8HbggyQHgBuDtSTYBBTwB/B5AVe1PsovhDemjwDVV9WJbz7XAXcAK4Naq2j/2vZEkjVWqaql76BoMBjU5ObnUbUjSaSXJ3qoajGNd/sa1JKnLkJAkdRkSkqQuQ0KS1GVISJK6DAlJUpchIUnqMiQkSV2GhCSpy5CQJHUZEpKkLkNCktRlSEiSugwJSVKXISFJ6jIkJEldJw2JJLcmOZzkoZHa+Un2JHmsfV3Z6knyqSRTSR5MctHIMtva/I8l2bYwuyNJGqfZnEl8Hth8TO064O6q2gDc3cYBLmf4XusNwHbgJhiGCsPXnr4FuBi4YSZYJEnL10lDoqq+BRw5prwF2NGGdwBXjtRvq6F7gfOSrAIuA/ZU1ZGqehbYw88HjyRpmZnrPYkLq+pQG34auLANrwaeGpnvQKv16j8nyfYkk0kmp6en59ieJGkc5n3juqoKqDH0MrO+m6tqUFWDiYmJca1WkjQHcw2JZ9plJNrXw61+EFg7Mt+aVuvVJUnL2FxDYjcw84TSNuDOkfpV7SmnS4Dn2mWpu4BLk6xsN6wvbTVJ0jJ21slmSPIl4O3ABUkOMHxK6WPAriRXA08C722zfx24ApgCngfeD1BVR5J8FHigzfeRqjr2ZrgkaZnJ8JbC8jQYDGpycnKp25Ck00qSvVU1GMe6/I1rSVKXISFJ6jIkJEldhoQkqcuQkCR1GRKSpC5DQpLUZUhIkroMCUlSlyEhSeoyJCRJXYaEJKnLkJAkdRkSkqQuQ0KS1GVISJK65hUSSZ5I8r0k+5JMttr5SfYkeax9XdnqSfKpJFNJHkxy0Th2QJK0cMZxJvFPqmrTyFuQrgPurqoNwN1tHOByYEP7bAduGsO2JUkLaCEuN20BdrThHcCVI/Xbauhe4LwkqxZg+5KkMZlvSBTwjSR7k2xvtQur6lAbfhq4sA2vBp4aWfZAq0mSlqmz5rn826rqYJK/A+xJ8v3RiVVVSepUVtjCZjvA61//+nm2J0maj3mdSVTVwfb1MPBV4GLgmZnLSO3r4Tb7QWDtyOJrWu3Ydd5cVYOqGkxMTMynPUnSPM05JJK8MsmrZ4aBS4GHgN3AtjbbNuDONrwbuKo95XQJ8NzIZSlJ0jI0n8tNFwJfTTKznj+vqv+Z5AFgV5KrgSeB97b5vw5cAUwBzwPvn8e2JUmLYM4hUVU/BH71OPUfAe88Tr2Aa+a6PUnS4vM3riVJXYaEJKnLkJAkdRkSkqQuQ0KS1GVISJK6DAlJUpchIUnqMiQkSV2GhCSpy5CQJHUZEpKkrjM6JNZd9z+WugVJOq2d0SEhSZofQ0KS1GVISJK6DAlJUteih0SSzUkeTTKV5LrF3r4kafYWNSSSrAA+DVwObATel2TjYvYgSZq9Ob/jeo4uBqba+7FJshPYAjy8UBscfQz2iY/95kJtRpLOSIsdEquBp0bGDwBvGZ0hyXZgexv9WZJH57G9C4C/eWndH5/HmhbHy/o9Tdjzwjvd+gV7Xgwn6vfvjWsjix0SJ1VVNwM3j2NdSSarajCOdS2G061fsOfFcLr1C/a8GBar38W+cX0QWDsyvqbVJEnL0GKHxAPAhiTrk5wDbAV2L3IPkqRZWtTLTVV1NMm1wF3ACuDWqtq/gJscy2WrRXS69Qv2vBhOt37BnhfDovSbqlqM7UiSTkP+xrUkqcuQkCR1nZEhsZR/+iPJ2iT3JHk4yf4kH2j1Dyc5mGRf+1wxssz1rddHk1x2sv1oN/7va/Xb20MA8+37iSTfa71Nttr5SfYkeax9XdnqSfKptv0Hk1w0sp5tbf7Hkmwbqf/Dtv6ptmzm2e/fHzmW+5L8JMkHl9txTnJrksNJHhqpLfhx7W1jjv3+pyTfbz19Ncl5rb4uyf8ZOdafnWtfJ9r3Ofa84N8HSc5t41Nt+rp59nz7SL9PJNm3LI5zVZ1RH4Y3xH8AvAE4B/gusHERt78KuKgNvxr4a4Z/guTDwL89zvwbW4/nAutb7ytOtB/ALmBrG/4s8K/G0PcTwAXH1P4jcF0bvg74eBu+AvhLIMAlwH2tfj7ww/Z1ZRte2abd3+ZNW/byMf83f5rhLxAtq+MM/DpwEfDQYh7X3jbm2O+lwFlt+OMj/a4bne+Y9ZxSX719n0fPC/59APxr4LNteCtw+3x6Pmb6J4B/vxyO85l4JvHSn/6oqv8LzPzpj0VRVYeq6ttt+KfAIwx/07xnC7Czql6oqseBKYb7cNz9aD8pvAO4oy2/A7hyYfaGLW39x25nC3BbDd0LnJdkFXAZsKeqjlTVs8AeYHOb9pqqureG36m3jbnndwI/qKonT7Ivi36cq+pbwJHj9LLQx7W3jVPut6q+UVVH2+i9DH+/qWuOffX2fU49n8A4vw9G9+UO4J0zP8nPp+e2jvcCXzrROhbrOJ+JIXG8P/1xov9JL5h2+vlm4L5Wurad4t06cvrf67dXfx3w45F/tOPavwK+kWRvhn8aBeDCqjrUhp8GLpxjz6vb8LH1cdnKy/9BLefjDItzXHvbmK/fZfiT6Iz1Sb6T5H8l+bVWm0tfC/HvdqG/D15apk1/rs0/X78GPFNVj43Uluw4n4khsSwkeRXwZeCDVfUT4CbgjcAm4BDD08nl5G1VdRHDv9B7TZJfH53YflJZds9Lt+vD7wL+opWW+3F+mcU4ruPaRpIPAUeBL7bSIeD1VfVm4N8Af57kNYvdV8dp9X1wjPfx8h96lvQ4n4khseR/+iPJ2QwD4otV9RWAqnqmql6sqr8FPsfw9PZE/fbqP2J4injWMfV5qaqD7eth4Kutv2dmTkXb18Nz7PkgL79EMc7/JpcD366qZ1r/y/o4N4txXHvbmJMk/xL4LeC32/90aJdsftSG9zK8pv/Lc+xrrP9uF+n74KVl2vTXtvnnrK3nnwO3j+zLkh7nMzEklvRPf7TribcAj1TVJ0fqo9f93g3MPNWwG9janpRYD2xgeDPquPvR/oHeA7ynLb8NuHOePb8yyatnhhneqHyo9TbzJM3odnYDV7UnJS4BnmuntncBlyZZ2U7vLwXuatN+kuSSdnyumm/PI172U9dyPs4jFuO49rZxypJsBv4QeFdVPT9Sn8jwHTEkeQPDY/rDOfbV2/e59rwY3wej+/Ie4JszAToP/xT4flW9dBlpyY/zsXeyz4QPwzv4f80wcT+0yNt+G8NTuweBfe1zBfAF4HutvhtYNbLMh1qvjzLy1E9vPxg+gXE/w5tufwGcO8+e38DwaY7vAvtntsXw+urdwGPAXwHnt3oYvjzqB22fBiPr+t3W1xTw/pH6gOE/1B8Af0b7bf959v1Khj+5vXaktqyOM8MAOwT8P4bXf69ejOPa28Yc+51ieB175vt55omef9G+X/YB3wb+2Vz7OtG+z7HnBf8+AF7Rxqfa9DfMp+dW/zzw+8fMu6TH2T/LIUnqOhMvN0mSxsSQkCR1GRKSpC5DQpLUZUhIkroMCUlSlyEhSer6/63cTINIB95jAAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(14,3))\n",
        "plt.plot(df.iloc[:,0])\n",
        "plt.show()\n",
        "\n",
        "plt.figure(figsize=(14,3))\n",
        "plt.plot(df.iloc[:,4])\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        },
        "id": "M7QQvslWo4qS",
        "outputId": "022ac43b-0795-4fe6-f1e2-7c644d7b2d12"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1008x216 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1008x216 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "raw_data = df.to_numpy()\n",
        "print(raw_data.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FadYdDdcgt47",
        "outputId": "115c9904-1a74-491c-ee5b-a8e719630b9a"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(5181, 5)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(raw_data[:5])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GplAuvN3g30t",
        "outputId": "1d04a5ee-baee-440f-d272-501192530094"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[   4.95     4.95     4.895    4.93  2319.   ]\n",
            " [   4.92     5.045    4.92     5.003 1612.   ]\n",
            " [   5.035    5.035    4.985    5.004 1375.   ]\n",
            " [   4.99     5.005    4.975    4.998  856.   ]\n",
            " [   5.       5.       4.98     4.983  232.   ]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# normalize"
      ],
      "metadata": {
        "id": "ESYLyhWeiRrf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "maxs = raw_data.max(axis=0)\n",
        "print(maxs)\n",
        "mins = raw_data.min(axis=0)\n",
        "print(mins)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NmEhUiEAhxOG",
        "outputId": "4f2ab6b6-e8ca-449b-a4f1-6988d0d234b8"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[4.8459999e+01 4.9520000e+01 4.7540001e+01 4.8584000e+01 1.7029700e+05]\n",
            "[4.026 4.026 4.026 4.026 0.   ]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "normalized = (raw_data - mins) / (maxs - mins)\n",
        "print(normalized[:5])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SwjTIY6aiHQo",
        "outputId": "876eddf4-50d7-49b4-f8ee-99d8736480f1"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[0.02079489 0.02031037 0.01997058 0.02028816 0.01361739]\n",
            " [0.02011973 0.02239856 0.02054511 0.02192648 0.00946582]\n",
            " [0.02270784 0.02217875 0.02203888 0.02194892 0.00807413]\n",
            " [0.0216951  0.02151932 0.02180907 0.02181426 0.00502651]\n",
            " [0.02192015 0.02140942 0.02192398 0.02147762 0.00136233]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "preprocessed = normalized"
      ],
      "metadata": {
        "id": "35hCAbte1YO3"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# by Data Sampling"
      ],
      "metadata": {
        "id": "c0LQqW08jtvW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "numbers = preprocessed"
      ],
      "metadata": {
        "id": "fnzuwFSkkFF2"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "raw_x = []\n",
        "raw_y = []\n",
        "\n",
        "for i in range(len(numbers)):\n",
        "  if i+SEQUENCE_LENGTH+OFFSET >= len(numbers): break\n",
        "  raw_x.append(numbers[i:i+SEQUENCE_LENGTH,:])\n",
        "  raw_y.append(numbers[i+SEQUENCE_LENGTH+OFFSET-1,:])\n",
        "#              0         1              \n",
        "#              01234567890123456789\n",
        "# numbers[0] = 12345678901234567890\n",
        "#                 i = 3\n",
        "#                 <-x->   = [3:8] = 45678\n",
        "#                      y  = [8]   = 9\n",
        "\n"
      ],
      "metadata": {
        "id": "r7NQikGPipTM"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(raw_x[0])\n",
        "print(raw_y[0])\n",
        "print()\n",
        "print(raw_x[1])\n",
        "print(raw_y[1])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2dT4KkJrkRbo",
        "outputId": "07a043f5-a53e-45b5-dd45-97a0fcd34f5f"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[0.02079489 0.02031037 0.01997058 0.02028816 0.01361739]\n",
            " [0.02011973 0.02239856 0.02054511 0.02192648 0.00946582]\n",
            " [0.02270784 0.02217875 0.02203888 0.02194892 0.00807413]\n",
            " [0.0216951  0.02151932 0.02180907 0.02181426 0.00502651]\n",
            " [0.02192015 0.02140942 0.02192398 0.02147762 0.00136233]\n",
            " [0.02147005 0.0209698  0.02077492 0.02040038 0.00176163]\n",
            " [0.02056983 0.02020047 0.0203153  0.02013107 0.00070465]\n",
            " [0.01978215 0.01943113 0.01974077 0.01927824 0.00066355]\n",
            " [0.02045731 0.02009056 0.02008549 0.01981687 0.00123901]\n",
            " [0.02011973 0.01965094 0.01962587 0.01943534 0.00051087]]\n",
            "[0.0195571  0.01943113 0.01962587 0.01934557 0.00060483]\n",
            "\n",
            "[[0.02011973 0.02239856 0.02054511 0.02192648 0.00946582]\n",
            " [0.02270784 0.02217875 0.02203888 0.02194892 0.00807413]\n",
            " [0.0216951  0.02151932 0.02180907 0.02181426 0.00502651]\n",
            " [0.02192015 0.02140942 0.02192398 0.02147762 0.00136233]\n",
            " [0.02147005 0.0209698  0.02077492 0.02040038 0.00176163]\n",
            " [0.02056983 0.02020047 0.0203153  0.02013107 0.00070465]\n",
            " [0.01978215 0.01943113 0.01974077 0.01927824 0.00066355]\n",
            " [0.02045731 0.02009056 0.02008549 0.01981687 0.00123901]\n",
            " [0.02011973 0.01965094 0.01962587 0.01943534 0.00051087]\n",
            " [0.0195571  0.01943113 0.01962587 0.01934557 0.00060483]]\n",
            "[1.89944641e-02 1.87717062e-02 1.90513394e-02 1.88967189e-02\n",
            " 9.39535048e-05]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "x = np.array(raw_x)\n",
        "y = np.array(raw_y)\n",
        "print(x.shape)\n",
        "print(y.shape)\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "train_x, test_x, train_y, test_y = train_test_split(x, y, test_size=0.1, shuffle=True)\n",
        "\n",
        "\n",
        "print(\"Training Data:\")\n",
        "print(train_x.shape)\n",
        "print(train_y.shape)\n",
        "\n",
        "print(\"Test Data:\")\n",
        "print(test_x.shape)\n",
        "print(test_y.shape)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EAJWL5bckiLk",
        "outputId": "f137f12c-7601-4915-cbb6-c6652f5ab208"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(5170, 10, 5)\n",
            "(5170, 5)\n",
            "Training Data:\n",
            "(4653, 10, 5)\n",
            "(4653, 5)\n",
            "Test Data:\n",
            "(517, 10, 5)\n",
            "(517, 5)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"train_x[0]\\n\", train_x[0])\n",
        "print()\n",
        "print(\"train_y[0]\\n\", train_y[0])\n",
        "\n",
        "plt.plot(train_x[0].T[0], '.', color='blue')\n",
        "plt.plot([len(train_x[0])], train_y[0,0], 'x', color='blue')\n",
        "plt.plot(train_x[0].T[1], '.', color='red')\n",
        "plt.plot([len(train_x[0])], train_y[0,1], 'o', color='red')\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 699
        },
        "id": "u_70o6bWknIN",
        "outputId": "72cbc8b6-ffec-4b91-b39f-9f0ccabd6c44"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "train_x[0]\n",
            " [[3.15164093e-01 3.07820834e-01 3.21229942e-01 3.13703488e-01\n",
            "  7.63372226e-05]\n",
            " [3.14803986e-01 3.07469117e-01 3.21459753e-01 3.13927914e-01\n",
            "  1.35058163e-04]\n",
            " [3.15051544e-01 3.07710907e-01 3.10015183e-01 3.02751492e-01\n",
            "  1.17441881e-05]\n",
            " [3.03798900e-01 2.97445839e-01 3.06659941e-01 3.03694084e-01\n",
            "  2.64244232e-04]\n",
            " [3.04136501e-01 2.97599705e-01 2.97467452e-01 2.90497756e-01\n",
            "  2.93604702e-05]\n",
            " [2.88495280e-01 2.82432848e-01 2.92756325e-01 2.86301023e-01\n",
            "  1.82034915e-04]\n",
            " [2.84669381e-01 2.79948982e-01 2.90688025e-01 2.85829683e-01\n",
            "  2.93604702e-05]\n",
            " [2.87482542e-01 2.80894184e-01 2.90688025e-01 2.83922079e-01\n",
            "  1.64418633e-04]\n",
            " [2.86694857e-01 2.82938431e-01 2.92756325e-01 2.88881929e-01\n",
            "  4.11046583e-05]\n",
            " [2.89057958e-01 2.85466215e-01 2.95169387e-01 2.91462813e-01\n",
            "  8.22093167e-05]]\n",
            "\n",
            "train_y[0]\n",
            " [2.91083389e-01 2.84521014e-01 2.93560663e-01 2.89667422e-01\n",
            " 4.11046583e-05]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU\n",
        "\n",
        "input_shape = train_x.shape[1:]\n",
        "output_shape = train_y.shape[-1]\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(LSTM(128, input_shape=input_shape))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(output_shape))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mape\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_x, train_y, epochs=10, verbose=0, validation_split=0.1, batch_size=BATCH_SIZE)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mape = model.evaluate(test_x, test_y)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mape=\", mape)\n",
        "\n",
        "\n",
        "\n",
        "y_ = model.predict(test_x).squeeze()\n",
        "for i in range(output_shape):\n",
        "    plt.scatter(test_y[:,i], y_[:,i])\n",
        "    plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "H_DZZN8oj16K",
        "outputId": "eb93fdc7-ea74-4056-b1d8-bc2fc2e29320"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " lstm (LSTM)                 (None, 128)               68608     \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 128)               0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 5)                 645       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 69,253\n",
            "Trainable params: 69,253\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "17/17 [==============================] - 0s 2ms/step - loss: 5.4675e-04 - mape: 195002.5469\n",
            "loss= 0.0005467520677484572\n",
            "mape= 195002.546875\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(test_y[0], 'b.')\n",
        "plt.plot(y_[0], 'r.')\n",
        "plt.show()\n",
        "\n",
        "plt.plot(test_y[0,:-1], 'b.')\n",
        "plt.plot(y_[0,:-1], 'r.')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 513
        },
        "id": "JglbgKh6mB9R",
        "outputId": "d631daf6-6cc0-40fc-c64f-b37dcca15710"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "denomalized_test_y = test_y * (maxs - mins) + mins\n",
        "denomalized_y_ = y_ * (maxs - mins) + mins"
      ],
      "metadata": {
        "id": "uAK9LDzBoFd4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(denomalized_test_y[0], 'b.')\n",
        "plt.plot(denomalized_y_[0], 'r.')\n",
        "plt.show()\n",
        "\n",
        "plt.plot(denomalized_test_y[0,:-1], 'b.')\n",
        "plt.plot(denomalized_y_[0,:-1], 'r.')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 517
        },
        "id": "53IqHJvtoQjk",
        "outputId": "4559d9fa-35d4-4b87-a06d-a624f6924a15"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAASA0lEQVR4nO3dYYxc13ne8f/j1dKirKJUoy3EiFbpgMmH2EkVZ8B60VZdiZBjFKicIBYgtEjsAoGatEJToIVi94MaqSiIVIULSAkQEG1StSmSCG5dMIxVgWW4CBSsKC1dUjZtp2VqJZbDQrQV2d2qJa3t2w8zbOn17M5d7iyHc/b/AwZ3Zu49V+/RlZ49e+bM3lQVkqR2vWPSBUiStpdBL0mNM+glqXEGvSQ1zqCXpMYZ9JLUuJFBn+TmJC8lOZvkXJLHB+//qyRfTnJm8Lh7nfYfTfJfB4+PjrsDkqSNZdQ6+iQB3lVVK0lmgReAnwV+GjhWVZ/aoO2fAZaBHlDAaeCHq+pPxlS/JGmEkSP66lsZvJwdPLp+y+pHgONV9cYg3I8DH7qmSiVJ1+SmLgclmaE/Gj8A/FJVnUryM8A/SfIYcAL4eFVdWtP0TuArV71+bfDeum6//fbav39/x/IlSQCnT5/+WlXNDdvXKeirahW4O8ke4NNJ3gd8AvjvwC7gCPBzwBPXUmCSh4GHAe666y6Wl5ev5TSStGMl+cP19m1q1U1VvQmcBD5UVRcG0zqXgF8FDg5p8lXg3Ve93jd4b+15j1RVr6p6c3NDfyBJkq5Rl1U3c4ORPEl2A/cDX0qyd/BegB8FPj+k+fPAB5PcluQ24IOD9yRJ10mXqZu9wDODefp3AM9W1bEkv5NkDghwhv4qHJL0gJ+uqp+qqjeS/GPg5cG5nqiqN8bfDUnSekYur7zeer1eOUcvSZuT5HRV9Ybt85uxktQ4g16SGmfQa1stLcHhw/2tpMnotI5euhZLS3DoEFy+DLt2wYkTMD8/6aqknccRvbbN4mI/5FdX+9vFxUlXJO1MBr22zcJCfyQ/M9PfLixMuiJpZ3LqRttmfr4/XbO42A95p22kyTDota3m5w14adKcupGkxhn0ktQ4g16SGmfQS1LjDHpJapxBL0mNM+glqXEGvSQ1zqCXpMYZ9JLUOINekhpn0EtS45oKeu9mJEnfqZm/XundjCRpuGZG9N7NSJKGaybovZuRJA3XzNSNdzOSpOGaCXrwbkaSNEwzUzeSpOEMeklqnEEvSY0z6CWpcSODPsnNSV5KcjbJuSSPr9n/VJKVddrOJnkmyeeSfDHJJ8ZVuCSpmy6rbi4B91XVSpJZ4IUkz1XVi0l6wG0btH0QeGdV/UCSW4AvJPn1qnp166VLkroYOaKvvisj9tnBo5LMAE8Cj27UHHhXkpuA3cBl4JtbK1mStBmd5uiTzCQ5A7wOHK+qU8AjwNGqurBB008B/xO4APwR8M+q6o0h5384yXKS5YsXL266E5Kk9XUK+qparaq7gX3AwST30J+WeXpE04PAKvDdwHuAv5/ke4ac/0hV9aqqNzc3t6kOSJI2tqlVN1X1JnASuBc4AJxP8ipwS5LzQ5r8deA/VtW3qup14PeA3tZKliRtRpdVN3NJ9gye7wbuB05X1R1Vtb+q9gNvVdWBIc3/CLhv0PZdwAeAL42reEnSaF1G9HuBk0leAV6mP0d/bL2DkzyQ5InBy18Cbk1ybtD2V6vqla0WLUnqbuTyykEw/9CIY2696vlR4Ojg+Qr9uXxJ0oT4zVhJapxBL0mNM+glqXEGvSQ1zqCXpMYZ9JLUOINekhpn0EtS4wx6SWqcQS9JjTPoJalxBr0kNc6gl6TGGfSS1DiDXpIaZ9BLUuMMeklqnEEvSY0z6CWpcQa9JDXOoJekxhn0ktQ4g16SGmfQS1LjDHpJapxBL0mNM+glqXEGvSQ1zqCXpMaNDPokNyd5KcnZJOeSPL5m/1NJVjZo/4NJlgZtP5fk5nEULknq5qYOx1wC7quqlSSzwAtJnquqF5P0gNvWa5jkJuDXgJ+oqrNJvgv41lgqlyR1MnJEX31XRuyzg0clmQGeBB7doPkHgVeq6uzgXF+vqtUt1ixJ2oROc/RJZpKcAV4HjlfVKeAR4GhVXdig6ffR/6HwfJLPJhn6QyHJw0mWkyxfvHhxs32QJG2gU9BX1WpV3Q3sAw4muQd4EHh6RNObgL8E/I3B9seSHBpy/iNV1auq3tzc3KY6IEna2KZW3VTVm8BJ4F7gAHA+yavALUnOD2nyGvC7VfW1qnoL+Azw/q2VLEnajC6rbuaS7Bk83w3cD5yuqjuqan9V7QfeqqoDQ5o/D/xAklsGH8z+FeAL4ytfkjRKl1U3e4FnBh++vgN4tqqOrXdwkgeAXlU9VlV/kuSTwMtAAZ+pqt8eR+GSpG5SVZOu4dv0er1aXl6edBmSNFWSnK6q3rB9fjNWkhpn0EtS4wx6SWqcQS9JjTPoJalxBr0kNc6gl6TGGfSS1DiDXpIaZ9BL0g1gaQkOH+5vx63L37qRJG2jpSU4dAguX4Zdu+DECZifH9/5HdFL0oQtLvZDfnW1v11cHO/5DXpJmrCFhf5Ifmamv11YGO/5nbqRpAmbn+9P1ywu9kN+nNM2YNBL0g1hfn78AX+FUzeS1DiDXpIaZ9BLUuMMeklqnEEvSY0z6CWpcQa9JDXOoJekxhn0ktQ4g16SGmfQS1LjDHpJapxBL0mNM+glqXEjgz7JzUleSnI2ybkkj6/Z/1SSlRHnuCvJSpJ/sNWCJUmb0+Xv0V8C7quqlSSzwAtJnquqF5P0gNs6nOOTwHNbKVSSdG1Gjuir78qIfXbwqCQzwJPAoxu1T/KjwJeBc1usVZJ0DTrN0SeZSXIGeB04XlWngEeAo1V1YYN2twI/Bzy+3jGSpO3VKeirarWq7gb2AQeT3AM8CDw9ounPA//8qt8IhkrycJLlJMsXL17sUpKka7S0BIcP97faGTZ1z9iqejPJSeBe4ABwPgnALUnOV9WBNU3+AvCRJP8U2AP8nyT/u6p+cc15jwBHAHq9Xl1bVySNsrQEhw7B5cuwa1f/htTbdZ9S3ThGBn2SOeBbg5DfDdwP/EJV3XHVMStDQp6q+stXHfPzwMrakJd0/Swu9kN+dbW/XVw06HeCLlM3e4GTSV4BXqY/R39svYOTPJDkiXEVKGl8Fhb6I/mZmf52YWHSFel6SNWNNVPS6/VqeXl50mVIzVpa6o/kFxYczbckyemq6g3bt6k5eknTb37egN9p/BMIktQ4g16SGmfQS1LjDHpJapxBL0mNM+glqXEGvSQ1zqCXpMYZ9JLUOINekhpn0EtS4wx6SWqcQS9JjTPoJalxBr0kNc6gl6TGGfSS1DiDXpIaZ9BLUuMMeklqnEEvSY0z6CWpcQa9JDXOoJekxhn0ktQ4g16SGmfQS1LjDHpJapxBL0mNGxn0SW5O8lKSs0nOJXl8zf6nkqys0/b+JKeTfG6wvW9chUuSurmpwzGXgPuqaiXJLPBCkueq6sUkPeC2Ddp+DfhrVfXHSd4HPA/cufWyJUldjRzRV9+VEfvs4FFJZoAngUc3aPufq+qPBy/PAbuTvHOLNUuSNqHTHH2SmSRngNeB41V1CngEOFpVFzr+s34c+GxVXRpy/oeTLCdZvnjxYtfaJUkddAr6qlqtqruBfcDBJPcADwJPd2mf5L3ALwB/a53zH6mqXlX15ubmulUuSepkU6tuqupN4CRwL3AAOJ/kVeCWJOeHtUmyD/g08JNV9QdbK1eStFldVt3MJdkzeL4buB84XVV3VNX+qtoPvFVVB4a03QP8NvDxqvq98ZYuSeqiy4h+L3AyySvAy/Tn6I+td3CSB5I8MXj5CP2R/2NJzgwef3bLVUuSOktVTbqGb9Pr9Wp5eXnSZUjSVElyuqp6w/b5zVhpp1lagsOH+1vtCF2+MCWpFUtLcOgQXL4Mu3bBiRMwPz/pqrTNHNFLO8niYj/kV1f728XFSVek68Cgl3aShYX+SH5mpr9dWJh0RboOnLqRdpL5+f50zeJiP+SdttkRDHppp5mfN+B3GKduJKlxBr0k3Qi2cdmrUzeSNGnbvOzVEb0kTdo2L3s16CVp0rZ52atTN5I0adu87NWgl6QbwTYue3XqRpIaZ9BLUuMMeklqnEEvSY0z6CWpcQa9JDWuraD3FmmS9B3aWUfvLdIkaah2RvTeIk2Shmon6L1FmiQN1c7UjbdIk6Sh2gl68BZpkjREO1M3kqShDHpJapxBL0mNM+glqXEjgz7JzUleSnI2ybkkj6/Z/1SSlQ3afyLJ+SS/n+RHxlG0JKm7LqtuLgH3VdVKklnghSTPVdWLSXrAbes1TPL9wEPAe4HvBv5Tku+rqtVxFC9JGm3kiL76rozYZwePSjIDPAk8ukHzDwO/UVWXqurLwHng4BZrliRtQqc5+iQzSc4ArwPHq+oU8AhwtKoubND0TuArV71+bfCeJOk66fSFqcFUy91J9gCfTnIP8CCwMI4ikjwMPAxw1113jeOUkqSBTa26qao3gZPAvcAB4HySV4Fbkpwf0uSrwLuver1v8N7a8x6pql5V9ebm5jZTkiRphC6rbuYGI3mS7AbuB05X1R1Vtb+q9gNvVdWBIc2PAg8leWeS9wDfC7w0vvIlSaN0mbrZCzwz+PD1HcCzVXVsvYOTPAD0quqxqjqX5FngC8DbwN9xxY0kXV+pqknX8G16vV4tLy9PugxJmipJTldVb9g+vxkrSY0z6CWpcQa9JDXOoJekxhn0ktQ4g16SGmfQS1LjDHpJapxBr+21tASHD/e3kiai01+vlK7J0hIcOgSXL8OuXXDiBMzPT7oqacdxRK/ts7jYD/nV1f52cXHSFUk7kkGv7bOw0B/Jz8z0twsLk65I2pGcutH2mZ/vT9csLvZD3mkbaSIMem2v+XkDXpowp24kqXEGvSQ1zqCXpMYZ9JLUOINekhpn0EtS4264m4MnuQj84RZOcTvwtTGVM0mt9APsy42olX6Afbniz1XV3LAdN1zQb1WS5fXuhD5NWukH2JcbUSv9APvShVM3ktQ4g16SGtdi0B+ZdAFj0ko/wL7ciFrpB9iXkZqbo5ckfbsWR/SSpKtMZdAn+VCS309yPsnHh+x/Z5LfHOw/lWT/9a+ymw59+ViSi0nODB4/NYk6R0nyK0leT/L5dfYnyVODfr6S5P3Xu8auOvRlIck3rromj13vGrtI8u4kJ5N8Icm5JD875JipuC4d+zIt1+XmJC8lOTvoy+NDjhlvhlXVVD2AGeAPgO8BdgFnge9fc8zfBn558Pwh4DcnXfcW+vIx4BcnXWuHvtwDvB/4/Dr7/yrwHBDgA8CpSde8hb4sAMcmXWeHfuwF3j94/qeA/zLkv6+puC4d+zIt1yXArYPns8Ap4ANrjhlrhk3jiP4gcL6q/ltVXQZ+A/jwmmM+DDwzeP4p4FCSXMcau+rSl6lQVb8LvLHBIR8G/nX1vQjsSbL3+lS3OR36MhWq6kJVfXbw/H8AXwTuXHPYVFyXjn2ZCoN/1yuDl7ODx9oPS8eaYdMY9HcCX7nq9Wt85wX/f8dU1dvAN4Dvui7VbU6XvgD8+ODX6k8leff1KW3suvZ1WswPfvV+Lsl7J13MKINf/X+I/ujxalN3XTboC0zJdUkyk+QM8DpwvKrWvS7jyLBpDPqd5reA/VX1g8Bx/v9PeU3OZ+l/3fzPA08D/2HC9Wwoya3AvwP+XlV9c9L1bMWIvkzNdamq1aq6G9gHHEzyvu38501j0H8VuHpUu2/w3tBjktwE/Gng69elus0Z2Zeq+npVXRq8/BfAD1+n2saty3WbClX1zSu/elfVZ4DZJLdPuKyhkszSD8Z/W1X/fsghU3NdRvVlmq7LFVX1JnAS+NCaXWPNsGkM+peB703yniS76H9QcXTNMUeBjw6efwT4nRp8qnGDGdmXNfOlD9Cfm5xGR4GfHKzy+ADwjaq6MOmirkWSO67MlyY5SP//oxtuIDGo8V8CX6yqT65z2FRcly59maLrMpdkz+D5buB+4EtrDhtrhk3dzcGr6u0kjwDP01+18itVdS7JE8ByVR2l/x/Ev0lynv6Hag9NruL1dezL303yAPA2/b58bGIFbyDJr9Nf9XB7kteAf0T/Qyaq6peBz9Bf4XEeeAv4m5OpdLQOffkI8DNJ3gb+F/DQDTqQ+IvATwCfG8wHA/xD4C6YuuvSpS/Tcl32As8kmaH/w+jZqjq2nRnmN2MlqXHTOHUjSdoEg16SGmfQS1LjDHpJapxBL0mNM+glqXEGvSQ1zqCXpMb9X5gRrd4Vx900AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# by Keras Sequence Data Generator"
      ],
      "metadata": {
        "id": "VxPXZ2m106W7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# from tensorflow.keras.utils import Sequence\n",
        "\n",
        "# class DataGenerator(Sequence):\n",
        "\n",
        "#     def __init__(self, numbers, \n",
        "#                  sequence_length=SEQUENCE_LENGTH, \n",
        "#                  offset=OFFSET, \n",
        "#                  batch_size=BATCH_SIZE, \n",
        "#                  shuffle=True,\n",
        "#                  validation_split=0.0,\n",
        "#                  test_split=0.0,\n",
        "#                  indexes=None,\n",
        "#                  max_size=None):\n",
        "#         self.numbers = numbers\n",
        "#         self.sequence_length = sequence_length\n",
        "#         self.offset = offset\n",
        "#         self.batch_size = batch_size\n",
        "#         self.shuffle = shuffle\n",
        "#         self.indexes, self.valid_indexes, self.test_indexes = self._prepare_indexes(validation_split, test_split, indexes, max_size)\n",
        "\n",
        "#     def __len__(self):\n",
        "#         return int(np.floor(len(self.indexes) / self.batch_size))\n",
        "\n",
        "#     def __getitem__(self, index):\n",
        "#         indexes = self.indexes[index*self.batch_size:(index+1)*self.batch_size]\n",
        "#         return self._generate_data(indexes)\n",
        "\n",
        "#     def on_epoch_end(self):\n",
        "#         self._shuffle_index()\n",
        "\n",
        "#     def _prepare_indexes(self, validation_split, test_split, indexes, max_size):\n",
        "#         if indexes is not None:\n",
        "#             return indexes, None, None\n",
        "\n",
        "#         data_count = len(self.numbers) - (self.sequence_length+self.offset) + 1\n",
        "#         all_index = np.arange(data_count)\n",
        "#         np.random.shuffle(all_index)\n",
        "#         if max_size is not None:\n",
        "#             all_index = all_index[:max_size]\n",
        "\n",
        "#         if validation_split+test_split==0.0:\n",
        "#             return all_index, None, None\n",
        "\n",
        "#         valid_i = int(len(all_index)*(1-validation_split-test_split))\n",
        "#         test_i = int(len(all_index)*(1-test_split))\n",
        "\n",
        "#         return all_index[:valid_i], all_index[valid_i:test_i], all_index[test_i:]\n",
        "        \n",
        "#     def _shuffle_index(self):\n",
        "#         if self.shuffle == True:\n",
        "#             np.random.shuffle(self.indexes)\n",
        "\n",
        "#     def _generate_data(self, indexes):\n",
        "#         X = np.empty((len(indexes), self.sequence_length, self.numbers.shape[-1]))\n",
        "#         Y = np.empty((len(indexes), self.numbers.shape[-1]))\n",
        "\n",
        "#         for i, index in enumerate(indexes):\n",
        "#             X[i] = self.numbers[index:index+self.sequence_length]\n",
        "#             Y[i] = self.numbers[index+self.sequence_length+self.offset-1]\n",
        "#         return X, Y\n",
        "\n",
        "#     def get_all_data(self):\n",
        "#         return self._generate_data(self.indexes)        \n",
        "\n",
        "#     def subset(self, type):\n",
        "#         if type==\"train\":\n",
        "#             return DataGenerator(self.numbers, indexes=self.indexes)\n",
        "#         elif type==\"validation\" or type==\"valid\":\n",
        "#             return DataGenerator(self.numbers, indexes=self.valid_indexes)\n",
        "#         elif type==\"test\":\n",
        "#             return DataGenerator(self.numbers, indexes=self.test_indexes)\n",
        "#         else:\n",
        "#             print(\"invalid type {}\".format(type))"
      ],
      "metadata": {
        "id": "BoYqpgU-oTg1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.utils import Sequence\n",
        "\n",
        "class AbstractSequenceDataGenerator(Sequence):\n",
        "\n",
        "    def __init__(self, numbers=None,\n",
        "                 sequence_length=10, \n",
        "                 offset=1, \n",
        "                 batch_size=32, \n",
        "                 shuffle=True,\n",
        "                 validation_split=0.0,\n",
        "                 test_split=0.0,\n",
        "                 indexes=None,\n",
        "                 base_generator=None,\n",
        "                 max_size=None):\n",
        "        if base_generator != None:\n",
        "            numbers = base_generator.numbers\n",
        "            sequence_length = base_generator.sequence_length\n",
        "            offset = base_generator.offset\n",
        "            batch_size = base_generator.batch_size\n",
        "            shuffle = base_generator.shuffle\n",
        "        self.numbers = numbers\n",
        "        self.sequence_length = sequence_length\n",
        "        self.offset = offset\n",
        "        self.batch_size = batch_size\n",
        "        self.shuffle = shuffle\n",
        "        self.indexes, self.valid_indexes, self.test_indexes = self._prepare_indexes(validation_split, test_split, indexes, max_size)\n",
        "\n",
        "    def __len__(self):\n",
        "        return int(np.floor(len(self.indexes) / self.batch_size))\n",
        "\n",
        "    def __getitem__(self, index):\n",
        "        indexes = self.indexes[index*self.batch_size:(index+1)*self.batch_size]\n",
        "        return self._generate_data(indexes)\n",
        "\n",
        "    def on_epoch_end(self):\n",
        "        self._shuffle_index()\n",
        "\n",
        "    def _prepare_indexes(self, validation_split, test_split, indexes, max_size):\n",
        "        if indexes is not None:\n",
        "            return indexes, None, None\n",
        "\n",
        "        data_count = len(self.numbers) - (self.sequence_length+self.offset) + 1\n",
        "        all_index = np.arange(data_count)\n",
        "        np.random.shuffle(all_index)\n",
        "        if max_size is not None:\n",
        "            all_index = all_index[:max_size]\n",
        "\n",
        "        if validation_split+test_split==0.0:\n",
        "            return all_index, None, None\n",
        "\n",
        "        valid_i = int(len(all_index)*(1-validation_split-test_split))\n",
        "        test_i = int(len(all_index)*(1-test_split))\n",
        "\n",
        "        return all_index[:valid_i], all_index[valid_i:test_i], all_index[test_i:]\n",
        "        \n",
        "    def _shuffle_index(self):\n",
        "        if self.shuffle == True:\n",
        "            np.random.shuffle(self.indexes)\n",
        "\n",
        "    def _generate_data(self, indexes):\n",
        "        raise Exception('_generate_data() not implemented.')\n",
        "\n",
        "    def get_all_data(self):\n",
        "        return self._generate_data(self.indexes)        \n",
        "\n",
        "    def subset(self, type):\n",
        "        if type==\"train\":\n",
        "            self.__init__(indexes=self.indexes, base_generator=self)\n",
        "            return self\n",
        "        elif type==\"validation\" or type==\"valid\":\n",
        "            self.__init__(indexes=self.valid_indexes, base_generator=self)\n",
        "            return self\n",
        "        elif type==\"test\":\n",
        "            self.__init__(indexes=self.test_indexes, base_generator=self)\n",
        "            return self\n",
        "        else:\n",
        "            print(\"invalid type {}\".format(type))\n",
        "\n",
        "class SeguenceDataGenerator(AbstractSequenceDataGenerator):\n",
        "\n",
        "    def __init__(self, **kwargs):\n",
        "        super().__init__(**kwargs)\n",
        "\n",
        "    def _generate_data(self, indexes):\n",
        "        X = np.empty((len(indexes), self.sequence_length, self.numbers.shape[-1]))\n",
        "        Y = np.empty((len(indexes), self.numbers.shape[-1]))\n",
        "\n",
        "        for i, index in enumerate(indexes):\n",
        "            X[i] = self.numbers[index:index+self.sequence_length]\n",
        "            Y[i] = self.numbers[index+self.sequence_length+self.offset-1]\n",
        "        return X, Y        "
      ],
      "metadata": {
        "id": "coMz-c9RYCoG"
      },
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data_generator = SeguenceDataGenerator(\n",
        "    numbers=preprocessed, \n",
        "    sequence_length=SEQUENCE_LENGTH,\n",
        "    offset=OFFSET,\n",
        "    batch_size=BATCH_SIZE,\n",
        "    validation_split=0.1, test_split=0.1)\n",
        "train_data_generator = data_generator.subset(\"train\")\n",
        "valid_data_generator = data_generator.subset(\"valid\")\n",
        "test_data_generator = data_generator.subset(\"test\")\n",
        "\n",
        "batch_x, batch_y = train_data_generator.__getitem__(0)\n",
        "print(batch_x.shape)\n",
        "print(batch_y.shape)\n",
        "print(batch_x[0])\n",
        "print(batch_y[0])\n",
        "\n",
        "print(len(preprocessed)-(SEQUENCE_LENGTH+OFFSET)+1)\n",
        "print(len(train_data_generator.indexes))\n",
        "print(len(valid_data_generator.indexes))\n",
        "print(len(test_data_generator.indexes))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "syHJzNvixQ-y",
        "outputId": "802aa496-8602-4143-8733-6a77021bc57d"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(32, 10, 5)\n",
            "(32, 5)\n",
            "[[1.01048749e-02 9.86943333e-03 1.03185179e-02 1.00767539e-02\n",
            "  7.04651286e-04]\n",
            " [9.27217917e-03 9.05613927e-03 9.46821691e-03 9.24637551e-03\n",
            "  0.00000000e+00]\n",
            " [7.53927190e-03 7.36360839e-03 7.69867151e-03 7.51829077e-03\n",
            "  1.76162821e-05]\n",
            " [6.50402859e-03 6.90200906e-03 6.64154050e-03 7.00210961e-03\n",
            "  0.00000000e+00]\n",
            " [5.55880644e-03 5.42928738e-03 5.67633392e-03 5.54333677e-03\n",
            "  4.16918677e-04]\n",
            " [6.16644925e-03 6.19861960e-03 6.29682387e-03 6.32882984e-03\n",
            "  0.00000000e+00]\n",
            " [6.12143868e-03 5.97881039e-03 6.25086165e-03 6.10440325e-03\n",
            "  0.00000000e+00]\n",
            " [6.41400744e-03 6.26456236e-03 6.54961606e-03 6.39615782e-03\n",
            "  0.00000000e+00]\n",
            " [7.47175603e-03 7.29766563e-03 7.62972819e-03 7.45096279e-03\n",
            "  0.00000000e+00]\n",
            " [6.90912380e-03 6.74814261e-03 7.05520046e-03 6.88989631e-03\n",
            "  6.75290815e-04]]\n",
            "[0.00706666 0.00716578 0.00721607 0.00731631 0.        ]\n",
            "5171\n",
            "5171\n",
            "5171\n",
            "5171\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "input_shape = (SEQUENCE_LENGTH, preprocessed.shape[-1])\n",
        "print(input_shape)"
      ],
      "metadata": {
        "id": "rndxXS1yeYUr",
        "outputId": "c639936c-ec45-4cbb-a17c-4d0ef0ae59da",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(10, 5)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Dense, RepeatVector, Flatten\n",
        "from tensorflow.keras.layers import Bidirectional, LSTM, GRU, Input\n",
        "\n",
        "input_shape = (SEQUENCE_LENGTH, preprocessed.shape[-1])\n",
        "output_shape = preprocessed.shape[-1]\n",
        "\n",
        "model = keras.Sequential()\n",
        "model.add(Input(input_shape))\n",
        "model.add(LSTM(128))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(10, activation=\"relu\"))\n",
        "model.add(Dense(output_shape))\n",
        "model.compile(loss=\"mse\", optimizer=\"adam\", metrics=[\"mape\"])\n",
        "model.summary()\n",
        "\n",
        "history = model.fit(train_data_generator, epochs=10, verbose=0, validation_data=valid_data_generator)\n",
        "\n",
        "\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.show()\n",
        "\n",
        "\n",
        "\n",
        "loss, mape = model.evaluate(test_data_generator)\n",
        "print(\"loss=\", loss)\n",
        "print(\"mape=\", mape)\n",
        "\n",
        "\n",
        "batch_x, batch_y = test_data_generator.get_all_data()\n",
        "y_ = model.predict(batch_x).squeeze()\n",
        "for i in range(output_shape):\n",
        "    plt.scatter(batch_y[:,i], y_[:,i])\n",
        "    plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "7LSyMSgzzS3E",
        "outputId": "2739a84f-e0d7-4cb5-84be-eaca7ee7e090"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " lstm (LSTM)                 (None, 128)               68608     \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 128)               0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 10)                1290      \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 5)                 55        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 69,953\n",
            "Trainable params: 69,953\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "161/161 [==============================] - 0s 2ms/step - loss: 5.8190e-04 - mape: 58725.5156\n",
            "loss= 0.0005818966310471296\n",
            "mape= 58725.515625\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "09Qy5zLaDiNz"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}