time_series_forecasting_in_tensorflow.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vlVtweEv7nAx"
      },
      "source": [
        " Time series forecasting in TensorFlow (BitPredict 💰📈)\n",
        "\n",
        "\n",
        "\n",
        "We're going to be building a series of models in an attempt to predict the price of Bitcoin.\n",
        "\n",
        "\n",
        "\n",
        "## What we're going to cover\n",
        "\n",
        "\n",
        "\n",
        "We've got a lot to go through.\n",
        "\n",
        "* Get time series data (the historical price of Bitcoin)\n",
        "  * Load in time series data using pandas/Python's CSV module\n",
        "* Format data for a time series problem\n",
        "  * Creating training and test sets (the wrong way)\n",
        "  * Creating training and test sets (the right way)\n",
        "  * Visualizing time series data\n",
        "  * Turning time series data into a supervised learning problem (windowing)\n",
        "  * Preparing univariate and multivariate (more than one variable) data\n",
        "* Evaluating a time series forecasting model\n",
        "* Setting up a series of deep learning modelling experiments\n",
        "  * Dense (fully-connected) networks\n",
        "  * Sequence models (LSTM and 1D CNN)\n",
        "  * Ensembling (combining multiple models together)\n",
        "  \n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qjEuJ-j7yEdK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "922b9a3a-193d-4323-e2a7-e493cc168fa7"
      },
      "source": [
        "# Check for GPU\n",
        "!nvidia-smi -L"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "GPU 0: Tesla T4 (UUID: GPU-f68fef48-5cd4-cc5d-77e6-571588c1a38a)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mVEjK6ZHDqkQ",
        "outputId": "d8336b0d-ee48-4b1d-82d2-63276d476e4f"
      },
      "source": [
        "# Download Bitcoin historical data from GitHub\n",
        "# Note: you'll need to select \"Raw\" to download the data in the correct format\n",
        "!wget https://raw.githubusercontent.com/mrdbourke/tensorflow-deep-learning/main/extras/BTC_USD_2013-10-01_2021-05-18-CoinDesk.csv"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2024-05-01 01:45:11--  https://raw.githubusercontent.com/mrdbourke/tensorflow-deep-learning/main/extras/BTC_USD_2013-10-01_2021-05-18-CoinDesk.csv\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.109.133, 185.199.110.133, 185.199.108.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.109.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 178509 (174K) [text/plain]\n",
            "Saving to: ‘BTC_USD_2013-10-01_2021-05-18-CoinDesk.csv’\n",
            "\n",
            "BTC_USD_2013-10-01_ 100%[===================>] 174.33K  --.-KB/s    in 0.02s   \n",
            "\n",
            "2024-05-01 01:45:12 (6.85 MB/s) - ‘BTC_USD_2013-10-01_2021-05-18-CoinDesk.csv’ saved [178509/178509]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FRKabu6qB41C"
      },
      "source": [
        "### Importing time series data with pandas\n",
        "Now we've got some data to work with, let's import it using pandas so we can visualize it.\n",
        "\n",
        "Because our data is in **CSV (comma separated values)** format (a very common data format for time series), we'll use the pandas [`read_csv()`](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.read_csv.html) function.\n",
        "\n",
        "And because our data has a date component, we'll tell pandas to parse the dates using the `parse_dates` parameter passing it the name our of the date column (\"Date\")."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "id": "5FOrshVX_dUE",
        "outputId": "8ebd2296-c45b-4d0d-82f0-4aa196ed7eac"
      },
      "source": [
        "# Import with pandas\n",
        "import pandas as pd\n",
        "# Parse dates and set date column to index\n",
        "df = pd.read_csv(\"/content/BTC_USD_2013-10-01_2021-05-18-CoinDesk.csv\",\n",
        "                 parse_dates=[\"Date\"],\n",
        "                 index_col=[\"Date\"]) # parse the date column (tell pandas column 1 is a datetime)\n",
        "df.head()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "           Currency  Closing Price (USD)  24h Open (USD)  24h High (USD)  \\\n",
              "Date                                                                       \n",
              "2013-10-01      BTC            123.65499       124.30466       124.75166   \n",
              "2013-10-02      BTC            125.45500       123.65499       125.75850   \n",
              "2013-10-03      BTC            108.58483       125.45500       125.66566   \n",
              "2013-10-04      BTC            118.67466       108.58483       118.67500   \n",
              "2013-10-05      BTC            121.33866       118.67466       121.93633   \n",
              "\n",
              "            24h Low (USD)  \n",
              "Date                       \n",
              "2013-10-01      122.56349  \n",
              "2013-10-02      123.63383  \n",
              "2013-10-03       83.32833  \n",
              "2013-10-04      107.05816  \n",
              "2013-10-05      118.00566  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e6cd6432-8703-47d2-b5fc-86f11571ae12\" 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>Currency</th>\n",
              "      <th>Closing Price (USD)</th>\n",
              "      <th>24h Open (USD)</th>\n",
              "      <th>24h High (USD)</th>\n",
              "      <th>24h Low (USD)</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",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2013-10-01</th>\n",
              "      <td>BTC</td>\n",
              "      <td>123.65499</td>\n",
              "      <td>124.30466</td>\n",
              "      <td>124.75166</td>\n",
              "      <td>122.56349</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2013-10-02</th>\n",
              "      <td>BTC</td>\n",
              "      <td>125.45500</td>\n",
              "      <td>123.65499</td>\n",
              "      <td>125.75850</td>\n",
              "      <td>123.63383</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2013-10-03</th>\n",
              "      <td>BTC</td>\n",
              "      <td>108.58483</td>\n",
              "      <td>125.45500</td>\n",
              "      <td>125.66566</td>\n",
              "      <td>83.32833</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2013-10-04</th>\n",
              "      <td>BTC</td>\n",
              "      <td>118.67466</td>\n",
              "      <td>108.58483</td>\n",
              "      <td>118.67500</td>\n",
              "      <td>107.05816</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2013-10-05</th>\n",
              "      <td>BTC</td>\n",
              "      <td>121.33866</td>\n",
              "      <td>118.67466</td>\n",
              "      <td>121.93633</td>\n",
              "      <td>118.00566</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e6cd6432-8703-47d2-b5fc-86f11571ae12')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-e6cd6432-8703-47d2-b5fc-86f11571ae12 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-e6cd6432-8703-47d2-b5fc-86f11571ae12');\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",
              "\n",
              "\n",
              "<div id=\"df-47266935-c892-45dc-8df5-d2d29f7a7c3c\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-47266935-c892-45dc-8df5-d2d29f7a7c3c')\"\n",
              "            title=\"Suggest charts\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-47266935-c892-45dc-8df5-d2d29f7a7c3c button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df",
              "summary": "{\n  \"name\": \"df\",\n  \"rows\": 2787,\n  \"fields\": [\n    {\n      \"column\": \"Date\",\n      \"properties\": {\n        \"dtype\": \"date\",\n        \"min\": \"2013-10-01 00:00:00\",\n        \"max\": \"2021-05-18 00:00:00\",\n        \"num_unique_values\": 2787,\n        \"samples\": [\n          \"2019-06-11 00:00:00\",\n          \"2021-05-03 00:00:00\",\n          \"2017-10-05 00:00:00\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Currency\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"BTC\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Closing Price (USD)\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 10887.84744436952,\n        \"min\": 108.58483,\n        \"max\": 63346.7890351052,\n        \"num_unique_values\": 2787,\n        \"samples\": [\n          7954.0062377205\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"24h Open (USD)\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 10863.80077212387,\n        \"min\": 108.58483,\n        \"max\": 63562.6735446382,\n        \"num_unique_values\": 2787,\n        \"samples\": [\n          7600.9225047785\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"24h High (USD)\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 11206.779756254151,\n        \"min\": 118.675,\n        \"max\": 64801.787869525,\n        \"num_unique_values\": 2787,\n        \"samples\": [\n          8064.9001353225\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"24h Low (USD)\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 10475.8112518284,\n        \"min\": 83.32833,\n        \"max\": 62094.6329143272,\n        \"num_unique_values\": 2785,\n        \"samples\": [\n          7714.610306452\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EQiDyqJaCdSS"
      },
      "source": [
        "Looking good! Let's get some more info."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z-YoeINjCzJ_",
        "outputId": "d43d1c4d-79e6-4509-dd0a-c829f259c683"
      },
      "source": [
        "df.info()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "DatetimeIndex: 2787 entries, 2013-10-01 to 2021-05-18\n",
            "Data columns (total 5 columns):\n",
            " #   Column               Non-Null Count  Dtype  \n",
            "---  ------               --------------  -----  \n",
            " 0   Currency             2787 non-null   object \n",
            " 1   Closing Price (USD)  2787 non-null   float64\n",
            " 2   24h Open (USD)       2787 non-null   float64\n",
            " 3   24h High (USD)       2787 non-null   float64\n",
            " 4   24h Low (USD)        2787 non-null   float64\n",
            "dtypes: float64(4), object(1)\n",
            "memory usage: 130.6+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4s00kcEDCrjV"
      },
      "source": [
        "Because we told pandas to parse the date column and set it as the index, its not in the list of columns.\n",
        "\n",
        "You can also see there isn't many samples."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XP3cN0uvlxhi",
        "outputId": "77ebff6f-5918-447f-ef4f-943ab913f386"
      },
      "source": [
        "# How many samples do we have?\n",
        "len(df)"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2787"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "id": "QxoNCv2NCuLY",
        "outputId": "32b09214-3c39-4b39-d49c-958720a6c973"
      },
      "source": [
        "# Only want closing price for each day\n",
        "bitcoin_prices = pd.DataFrame(df[\"Closing Price (USD)\"]).rename(columns={\"Closing Price (USD)\": \"Price\"})\n",
        "bitcoin_prices.head()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                Price\n",
              "Date                 \n",
              "2013-10-01  123.65499\n",
              "2013-10-02  125.45500\n",
              "2013-10-03  108.58483\n",
              "2013-10-04  118.67466\n",
              "2013-10-05  121.33866"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-29860199-836f-4cf3-b2a1-4945c2841b25\" 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>Price</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Date</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2013-10-01</th>\n",
              "      <td>123.65499</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2013-10-02</th>\n",
              "      <td>125.45500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2013-10-03</th>\n",
              "      <td>108.58483</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2013-10-04</th>\n",
              "      <td>118.67466</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2013-10-05</th>\n",
              "      <td>121.33866</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-29860199-836f-4cf3-b2a1-4945c2841b25')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-29860199-836f-4cf3-b2a1-4945c2841b25 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-29860199-836f-4cf3-b2a1-4945c2841b25');\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",
              "\n",
              "\n",
              "<div id=\"df-43ea51f0-0331-4a63-97a8-2bd23c76e370\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-43ea51f0-0331-4a63-97a8-2bd23c76e370')\"\n",
              "            title=\"Suggest charts\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-43ea51f0-0331-4a63-97a8-2bd23c76e370 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "bitcoin_prices",
              "summary": "{\n  \"name\": \"bitcoin_prices\",\n  \"rows\": 2787,\n  \"fields\": [\n    {\n      \"column\": \"Date\",\n      \"properties\": {\n        \"dtype\": \"date\",\n        \"min\": \"2013-10-01 00:00:00\",\n        \"max\": \"2021-05-18 00:00:00\",\n        \"num_unique_values\": 2787,\n        \"samples\": [\n          \"2019-06-11 00:00:00\",\n          \"2021-05-03 00:00:00\",\n          \"2017-10-05 00:00:00\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Price\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 10887.84744436952,\n        \"min\": 108.58483,\n        \"max\": 63346.7890351052,\n        \"num_unique_values\": 2787,\n        \"samples\": [\n          7954.0062377205,\n          56427.0431250152,\n          4186.78374\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "olHSR6poID_w"
      },
      "source": [
        "Much better!\n",
        "\n",
        "But that's only five days worth of Bitcoin prices, let's plot everything we've got."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 643
        },
        "id": "CfkhZhmCEyUe",
        "outputId": "1027afaf-2388-4349-bf2b-f8edeb30965f"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "bitcoin_prices.plot(figsize=(10, 7))\n",
        "plt.ylabel(\"BTC Price\")\n",
        "plt.title(\"Price of Bitcoin from 1 Oct 2013 to 18 May 2021\", fontsize=16)\n",
        "plt.legend(fontsize=14);"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UkTvwP2DCuGK",
        "outputId": "d02849c2-7c4c-446a-fab4-0a49ac13fd44"
      },
      "source": [
        "# Importing and formatting historical Bitcoin data with Python\n",
        "import csv\n",
        "from datetime import datetime\n",
        "\n",
        "timesteps = []\n",
        "btc_price = []\n",
        "with open(\"/content/BTC_USD_2013-10-01_2021-05-18-CoinDesk.csv\", \"r\") as f:\n",
        "  csv_reader = csv.reader(f, delimiter=\",\") # read in the target CSV\n",
        "  next(csv_reader) # skip first line (this gets rid of the column titles)\n",
        "  for line in csv_reader:\n",
        "    timesteps.append(datetime.strptime(line[1], \"%Y-%m-%d\")) # get the dates as dates (not strings), strptime = string parse time\n",
        "    btc_price.append(float(line[2])) # get the closing price as float\n",
        "\n",
        "# View first 10 of each\n",
        "timesteps[:10], btc_price[:10]"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "([datetime.datetime(2013, 10, 1, 0, 0),\n",
              "  datetime.datetime(2013, 10, 2, 0, 0),\n",
              "  datetime.datetime(2013, 10, 3, 0, 0),\n",
              "  datetime.datetime(2013, 10, 4, 0, 0),\n",
              "  datetime.datetime(2013, 10, 5, 0, 0),\n",
              "  datetime.datetime(2013, 10, 6, 0, 0),\n",
              "  datetime.datetime(2013, 10, 7, 0, 0),\n",
              "  datetime.datetime(2013, 10, 8, 0, 0),\n",
              "  datetime.datetime(2013, 10, 9, 0, 0),\n",
              "  datetime.datetime(2013, 10, 10, 0, 0)],\n",
              " [123.65499,\n",
              "  125.455,\n",
              "  108.58483,\n",
              "  118.67466,\n",
              "  121.33866,\n",
              "  120.65533,\n",
              "  121.795,\n",
              "  123.033,\n",
              "  124.049,\n",
              "  125.96116])"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "40o59gpNKtRs"
      },
      "source": [
        "Beautiful! Now, let's see how things look."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 643
        },
        "id": "bTSX7RZ9Cytj",
        "outputId": "cd2195fa-7ac9-41be-c017-557f4f9155a7"
      },
      "source": [
        "# Plot from CSV\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "plt.figure(figsize=(10, 7))\n",
        "plt.plot(timesteps, btc_price)\n",
        "plt.title(\"Price of Bitcoin from 1 Oct 2013 to 18 May 2021\", fontsize=16)\n",
        "plt.xlabel(\"Date\")\n",
        "plt.ylabel(\"BTC Price\");"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2QAAAJyCAYAAACrEAP6AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACoRElEQVR4nOzdd3zT1f7H8Xe6d8tsGWUje88KqAypiANBBScgblAR9wLXVX94ndeBGxdXcXFVFEVUUAGRJXsIKLNQKG1p6c7390ebb5MmaZOutOX1fDz6MP1+T745CU3Nu+ecz7EYhmEIAAAAAFDt/HzdAQAAAAA4VRHIAAAAAMBHCGQAAAAA4CMEMgAAAADwEQIZAAAAAPgIgQwAAAAAfIRABgAAAAA+QiADAAAAAB8hkAEAAACAjxDIAC+0atVKFovF4Ss4OFgtWrTQ+PHj9csvv5TrupMmTZLFYtHcuXMrt8M+tnz5co0cOVL169eXn5+fx8/xrLPOcnqdLRaLwsLC1KlTJ91yyy3au3evy/s+/PDDslgsevjhhyv3yVSi6v73NgxDTz/9tLp27arQ0FDz9ayLli1bpieeeELjxo1zeL/++uuvFb62YRj6+OOPNXbsWMXHxyskJET16tVTz549dffdd7v9mfSFI0eO6L333tPll1+u9u3bKyQkRGFhYerYsaNuvfVW/f3336XePzc3V//3f/+nHj16KDw8XPXq1dNZZ52lTz/91O19tm/frv/85z+aNGmSunXrpoCAAFksFj3++OOlPta3336ra6+9Vn379lWTJk0UHBysyMhI9ezZU/fff7+OHj1anpfAJ9auXat///vfuuyyy3TaaaeZv/c++OCDMu+7d+9eTZs2TR06dFBoaKhCQkLUunVrTZw4UX/++afXfbH9nrFYLOrZs2epbf/44w+H37WV8X6pCuvWrdOTTz6p4cOHKzY2VoGBgapXr56GDBmil19+WXl5eaXe/6+//tKkSZPUvHlzBQcHq3nz5po0aZJ2797tsn1F3kdV+bsIdZABwGMtW7Y0JBmDBg0yJk6caEycONG48MILjWbNmhmSDIvFYjzzzDNeX3fixImGJOOdd96p/E77yIEDB4yYmBjDYrEYQ4YMMa666ipj4sSJxi+//FLmfc8880xDktGjRw/zdb766quNxMREIyYmxpBkREVFGatWrXK676xZswxJxqxZsxyO//TTT4Yk48wzz6ykZ1h+1f3v/fLLLxuSjOjoaOOiiy4yX9O6qEePHoYkpy9Pfu5Kc+DAAaN///7m+7xv377G+PHjjfPPP99o2LChIckIDg42XnrppUp5Hnv27DEkGS1btizX/a+44gpDkuHn52d0797duOSSS4xzzz3XaNSokSHJCA8PN77//nuX983MzDROP/10Q5IRExNjjB071hg5cqQREBBgSDLuuOMOl/e77bbbXL72jz32mEd9bdeunTFixAjjsssuM0aOHGm+1xs3bmxs2rTJq+fv7vdAVbvwwgtdvgbvv/9+qfdbuXKlERkZaUgymjVrZlxwwQXGRRddZLRu3dqQZAQEBBjz58/3qi+23zO2r9WrV7tte8MNN1Tq+6Uq5OXlmf2LiIgwhg4dakyYMMEYPHiw4e/vb0gy+vfvbxw/ftzl/X/99VcjLCzMkGR06dLFGD9+vNGlSxfz/bBixQqn+1TkfVRVv4tQNxHIAC/YAlnJD9JZWVnG1VdfbUgy/P39je3bt3t13YMHDxpbt241UlNTK7G3vvXOO+8YkozLL7/c6/vaApmrD1OpqanGkCFDDElGnz59nM4nJycbW7duNZKTkx2O16RAVt3/3rbX090Hh7rkzjvvNB5++GHjyy+/NPbv32++ZyvyISglJcVo06aNIcno1auXUzjIy8sz/v3vf5sfCl944YWKPo0KB7JbbrnFeOSRR4z9+/c7HD9x4oQxYcIEQ5JRv359IyUlxem+tmDVrVs3h/fR6tWrjYiICEOS8dVXXznd74033jDuvPNO48MPPzS2bt1qXHXVVR4FsnXr1hmHDh1yOn7ixAnjkksuMSQZAwcO9PSpG4bhu0D25JNPGvfff7/x6aefGrt27TLfe2UFsu7duxuSjOuvv97Izc01jxcUFBgPPvigGY6zsrI87ostkPXt29eQZNx8880u2508edKIjo42mjRpYjRv3rzGhoa8vDyjT58+xvz5843s7GyHcxs2bDCaNGliSDImT57sdN/MzEyjadOmhiTjvvvuczh33333GZKM+Ph44+TJkw7nKvI+qorfRai7CGSAF9wFMsMo/AUdHh5uSDKefPLJ6u9cDfPII4+U+wNRaYHMMAxj8eLF5l8bPQ01NSmQVTfbX9n37Nnj665Uu8r4EHT55ZcbkozWrVu7/eu7YRjGSy+9ZEgyAgMDjS1btpT78Qyj4oGsNJmZmeZoTMmgkJKSYgQFBRmSjF9//dXpvo899pjHAckWCMoKZKXZu3ev+V5PS0vz+H6+CmQleRLIjh49aj7HI0eOOJ3Pz883QkNDDUnG2rVrPX5s2+s/a9YsIzY21qhXr57LQPfee+8Zkox77rmnVoeG999/35BkhIaGOoRawyieJXDaaacZBQUFDucKCgqM0047zZBkzJkzx+PHK+195Eptfm1R9VhDBlSSiIgIdejQQZIc5pXbr9d55513lJCQoOjoaFksFrNdWWuK1qxZo4kTJ6p169YKCQlR/fr11aNHD9111136559/nNofPHhQM2bMUKdOnRQWFqbIyEj169dPL730kvLz88v1/D766CMNHz5c9evXV3BwsFq2bKlrrrlGO3bscGg3d+5cWSwWzZo1S5L0yCOPmK9Bq1atyvXYJcXFxZm3Sz4fV2vIzjrrLA0dOlSStHTpUoe1Eq769OOPP+qSSy4x1xk0atRI/fr106xZs3Ts2DGn9t99953OO+88NW7cWEFBQWratKnGjx+v1atXu+y/u39v+74nJydr6tSpio+PV1BQkOLj43XLLbcoNTXVsxdJxWvx9uzZI0lq3bq1+bxtr4/t32vSpElKSUnR9OnT1bZtWwUHB+uss84yr5Wfn685c+bo9NNPV3R0tEJCQtS+fXvdeuutOnDggMvHt//Z/+CDD9S/f39FRESoUaNGuuyyy8w1V4Zh6KWXXlLPnj0VHh6uhg0batKkSTpy5IjHz7Wq7N69Wx999JEk6d///rdiYmLctr355pvVo0cP5eXlafbs2S7bePJenjRpklq3bi1J+ueff5zWUlZUWFiY+btq3759Due++eYb5ebmqkWLFho0aJDTfS+//HJJ0sqVK3Xw4MEK96UsAQEBkiQ/Pz8FBgZ6dB+LxaJHHnlEkuPvH9vPub2UlBTdf//96tKli/m7sk+fPpo9e7aysrIq9bm4Exwc7HHbhg0ben39gIAAXXXVVTp+/Li++OILp/Nvv/22JOmaa65xe43k5GS9+OKLOvfcc9W6dWuFhoYqKipKffv21f/93/8pOzvbof2uXbvk7++vevXq6eTJk26v26VLF1ksFn3zzTdeP6+SevXqJUnKyspyWndoe94TJkyQn5/jR18/Pz+NHz9ekvT55597/HilvY8AbxHIgEqUnp4uyfX/YG+55RZde+21CggI0OjRozVgwACPPlw9/fTT6t+/v9577z0FBQXpwgsv1ODBg5WXl6d///vf+umnnxzaL1u2TF27dtVzzz2n7OxsnX322Ro0aJB27dqlW265RaNHjy5z4bM9wzA0ceJEXXbZZVq2bJl69eqlsWPHKiQkRO+884569eqlRYsWme3btWuniRMnqkePHpKkHj16aOLEiZo4caIuvvhijx+3NKtWrZIkxcbGqkGDBmW2P+ecc5SYmGjex9YfV3269dZbNXz4cH366adq1KiRxo4dq379+iklJUWPPvqoNm7c6ND+oYce0jnnnKNvvvlGp512mi6++GLFxsZq/vz5GjhwoPlhxxv79u1T79699dlnn6l///46++yzdeLECb300ksaOXKkx/9+55xzjiZOnKjw8HBJ0rhx48znXXKR/9GjR9W3b1+999576tq1qy688EI1b95ckpSTk6NRo0bppptu0rp16zRo0CCNGTNGOTk5+s9//qOePXtq7dq1bvtx3333afLkyYqMjNSoUaMUFhamjz76SIMHD9bx48c1YcIE3XXXXWrSpIkSExPl7++vd999V2effbZyc3O9fv0q01dffSWr1aqYmBhdcMEFpba1WCy66qqrzPsZhuFw3tP38uDBgzVu3DhJUnh4uMPP68SJEyv8nPLy8sw/BjVp0sTh3Lp16yRJffv2dXnfNm3aqH79+pKk9evXV7gvpcnJydH9998vSTr77LMVGhrq0f3c/f6ZOHGiBg8ebLbbvXu3evfurSeffFLJyck699xzNWzYMO3cuVP33HOP+fNZ1SIiIjRkyBBJ0oMPPujw/rZarXr44YeVlZWlUaNGKT4+vlyPYQtbJX8f7dq1S0uXLtWgQYN02mmnub3/d999p9tuu00bNmxQy5YtNWbMGPXv31/bt2/Xvffeq2HDhiknJ8ds37ZtW40ePVqpqan68MMPXV7zp59+0pYtW9S2bVuNGjWqXM/L3s6dOyVJQUFB5s+oTVk/17bjtnaeKO19BHjNxyN0QK1S2pTFP//80/Dz8zMkGW+//bZ5XEVTUaKiolwuGjYM90Ue/ve//xmSjJCQEOPjjz92ut/mzZsdpkYdOnTIaNCggWGxWIxXXnnFYWrG0aNHjWHDhhmSjEceecTj5/zqq68akoyGDRsa69atM49brVZzWlBMTIzTVJuKTBlyNWXRarUaSUlJxvvvv280aNDAkGS88sorTvetSFGPF1980ZBkNGjQwPjxxx+dzv/+++/G3r17ze+//fZb89+n5PqsN99805y+VnLNkbt/b1vfJRmTJk1yWCexd+9es3jMvHnz3D4HV2w/t66mLNrW+kkyhg8f7nJa2D333GNIMtq2betwjdzcXGPKlCnmdL6cnByH+9mu26BBA2P9+vXm8ZMnTxqDBw821ym1bdvW+Pvvv83zycnJRrt27QxJxgcffODVc3X33Ms7Tci2Dmro0KEetV+6dKn5vHfv3m0e9/a9XJVTFm3v6dDQUCMpKcnh3NixYw1JxvTp093e37beqawCJt5OWVyzZo1ZwGfUqFFmsZR+/fo5reEpiye/fwYMGGBIMi644AIjIyPDPH7kyBGjd+/e5V4Da8/TNWTbtm0z1yk2a9bMuPDCC42xY8carVu3NoKCgoyrrrrKqymbhuH8+ickJBh+fn7GP//8Y7Z54IEHHP6f5e79smXLFpf//0pJSTFGjhxpSDJmz57tcM42tbxHjx4u+zdu3DhDUrkKYZVktVqNhIQEQ5IxduxYh3Pp6enme9L+95C9tWvXmm3sfxZKU9r7yBWmLKI0BDLAC64CWWpqqrFw4UKjbdu2hiSjadOmDr/Qbb/kH330UbfXdfcBvWfPnl79D8v2wXnatGkuz+/fv98IDAw0GjVqZFitVo+uaXteL774otM5q9Vqfjj717/+5XCuMgKZu6/27dsbX3/9tcv7ljeQ5eXlmZWzPvvsM4/6OXz4cEOSMWPGDJfnzzvvPEOScd111zkcLyuQNW/e3MjMzHS63lNPPWVIMq655hqP+mfjSSALDAw0du3a5XQ+KyvLLOTw5ZdfOp3PzMw0YmNjDUnGhx9+6HDO9u/18ssvO93v888/N88vXLjQ6fwzzzxjSK4X6Hujoh+CzjnnHEOSMWHCBI/ab9u2zXxev//+u3nc2/dyVQWyDRs2mP+eTzzxhNP5s88+25BkPPDAA26vYavA6Or+9rwNZF988YXTe33EiBEufy7LUtbvn19++cWQZISFhbn8ML169WpDKqyut2/fPq8f38bTQGYYhnH48GEz3Nh/de7c2Xjrrbe8fuySr/8bb7xhSDIefvhhwzAK1041b97ciIiIMP+fVZ73y/bt283gXJKtimHJ6+3bt88ICAgwwsLCSl2X6Snbv3dERISxY8cOh3MHDhwwX8udO3e6vP+OHTvMNgcPHizz8cp6H7lCIENpmLIIlMPkyZPNNQkxMTEaPXq0du3apbZt2+qbb74xp4jZ83a6XlJSktavXy8/Pz9NmTLFo/ssXLhQksz58CU1a9ZM7du3V3Jysjm9ozT79+/Xrl27JMnlVCmLxaLJkydLktPUycpQcrrRBRdcoPbt22vnzp2aMWOGOXWxMqxZs0bJyclq2LChLrroojLb5+fn67fffpMkp3UpNrZ/N29fm+HDhyssLMzpeKdOnSTJ7ZqtiujVq5fatGnjdHz16tXKyMhQ/fr1df755zudDwsL04QJEyS5f57nnnuu07H27dtLKlzfMnLkSLfnq2OdUmUySkxTlMr3Xq4K+/fv1/nnn6+MjAxdcMEFuvfee33WF1fGjBkjwzCUn5+vv//+W2+++aa2bt2qrl27lrr/WXn8/PPPkgqn9cbGxjqd79Onj3r06CGr1aqlS5dW6mO78ttvv6lbt27atGmT5s2bp6SkJKWkpOirr75SXl6epkyZUuGfnfHjxys8PFxz586VYRj67rvvtH//fl166aUu/59VUkFBgZYsWaLHHntMN998syZPnqxJkybpX//6l6TCfehKuvXWWyVJL730ksPx1157Tfn5+briiitKXZfpiffee0+PPvqo/Pz89Pbbb5u/O6pKTX8foXYK8HUHgNpo0KBBateunaTC+eqNGzfWwIEDdc4555iL0EvytqCFreBBkyZNFB0d7dF9bJtb2tYjlCY5ObnUNQNS8Qf/Bg0aKCoqymWbtm3bOrStTGPGjHHa4NkwDL366quaOnWqhg4dqq1bt6pFixYVfixbQYUOHTp4tLbv2LFj5kJ2WwGGksr72rh7PrZ/g5IL6CuDu59PW9/dPUep7Ofp6vlERERIKvz5dvWeiYyMlFQ1z9UbtiIKhw8f9qi9fSGSRo0aSSrfe7myJSUlafjw4frnn3+UmJio+fPnu/w5t73umZmZbq+VkZEhSW5/J1SUv7+/WrZsqSlTpmj48OHq0qWLJk+erMGDBzsU9KkIT3+u//zzzyr53WYvNTVVF110kY4ePaoVK1ZowIAB5rnzzjtPnTt3Vrdu3fT222/ryiuvNAsUeSsyMlIXX3yx3n33Xf34448eFfOw2blzpy666CJt3rzZbRvbGmp7V155pe699159/vnnOnTokJo0aaLc3Fy98cYbkqRp06aV67nYfPLJJ2b/33jjDV1yySVObWw/05L7n2vbz7RU+s+1p+8jwFsEMqAcrr32WrejIu54uiC9IqxWq6TC0biy/uLpSTGMmshisejmm2/WW2+9pbVr1+o///mPnn76aV93q1KVrAJWHary57O05+OL5+qNPn366IMPPtDatWuVn5/v9g8uNrZR2wYNGlRaVdGKOnLkiIYNG6YdO3ZoxIgRWrBggdvKfrY+20KkK/v373doW5VatWqloUOHauHChVq8eLFZNKUuWbhwoZKTk9W2bVuHMGbTpk0bDRgwQD/99JN++OGHcgcyqTB8vfvuu3r66af1008/qUOHDi6raZZ08cUXa/PmzTrvvPN09913q3PnzoqKilJgYKByc3Pd/jyFhYXpuuuu0+zZs/X6669r1qxZ+uyzz3T48GENGTJE3bt3L/dz+fzzz3X55ZfLarXqtddecxssIyMjVb9+faWkpGjv3r1mwRd7tiqJDRs2dPv/Tm/eR4C3CGRADWUbVTh06JDS0tI8+st6fHy8WSHMXTUpbzRr1kxS4WhQenq6y78c2kblbG2rS5s2bbR27Vpt3bq1Uq5ne7137NghwzDK/KtngwYNFBwcrJycHO3evdvlBwtfvTaVydZ3W+l8V+rC83Tn/PPP1x133KG0tDT973//M6sfumIYht5//31JhSMbtp+h8ryXK0tycrKGDRumrVu3avjw4fryyy8VEhLitn3v3r0lye2WDbt371ZKSoqk4jLjVc32Abkyt0Gw/azafnZdqa6fa1v4LW1kxvYzY3vty+uMM85Qu3bt9N1330mSOeW8NNu2bdOGDRvUuHFjffHFF05/lChr+vvUqVP1zDPP6PXXX9f9999vTl+syOjYggULNGHCBBUUFOjVV1/VddddV2r73r1764cfftDq1atdTr22/bzbfv5L8vZ9BHirZv9pEjiFxcXFmWsYPC2dbisdPH/+/ErpQ/Pmzc3paK72SDMMwzxekb/alodtbZtt6ltZgoKCJDnvW2bTt29fNWzYUMnJyVqwYEGZ1wsICDBLaLvbP87271bdr01l6tu3ryIiIpSSkqIvv/zS6XxWVpa5T1dtfp7utG3bVpdeeqkk6a677ip1H7hXXnlFGzZsUEBAgO666y7zeHney2X9vHri6NGjGjZsmDZv3qzhw4frq6++KnMk9Nxzz1VQUJD27t1rrpG0N2/ePEnSwIED1bRp03L3zVM5OTn69ddfJanMKdb2ynr9bHvsLVq0yOV01HXr1pnr/s444wwve+0dW+Dbtm2b0tLSnM7n5eWZ20qUNsXSUzfeeKMaNGigxo0b6+qrry6zvS0ENm3a1OUI8QcffFDq/Vu0aKExY8bo4MGDmjlzppYvX66mTZtq7Nix5er/V199pUsvvVT5+fl69dVXdcMNN5R5H9u64I8++sicSWJjtVr18ccfS5LLPpXnfQR4i0AG1GC2zZUfeOABffbZZ07nt2zZ4jBCdNdddykmJkbPPvusnnnmGZd7OO3Zs6fM/4Hau/POOyVJjz32mP7880/zuGEYevzxx7V+/XrFxMSU+RfKymJbQ2bbL+bCCy/06H62PbV27tzpch+vgIAAPfDAA5Kk66+/XsuWLXNq88cff5jTtSTpjjvukCS9+uqrWrJkiUPbuXPn6ssvv1RgYKBuu+02j/pYE4WEhGjq1KmSCp+v/UbkeXl5uu2225SUlKTWrVtX2j5zNc3LL7+sVq1aac+ePeYHM3v5+fl69tlnzX/n//u//1OXLl0c2nj7Xm7UqJGCgoLM4g7eSklJ0fDhw7Vp0yaNGDHC4w+R9erV00033SSpcKNr+43Q165dq//7v/8zn0dlOHLkiF599VWX648OHDigq666SgcPHlSrVq109tlne3xd2/vd3ZqnwYMHa8CAAcrKytINN9zgsHnx0aNHzQ/5EyZMKPfeX54aNWqUwsPDlZWVpeuuu85hPVNubq5uv/127d27V4GBgZXyHrvjjjt09OhRHT582KP9s0477TT5+/tr48aNZjEUm6+++krPPfdcmdewvTeeeuopSdINN9xQ5vRfV7755htdfPHF5kb1noQxqbDwUtOmTbVjxw499NBDDuceeugh7dixQ82bN3cKqOV9HwHeYsoiUINddNFF+te//qUHH3xQF198sTp27KgePXooKytLf/31l7Zs2aJ33nnHrL7XvHlzc1rVnXfeqdmzZ6tr165q0qSJ0tLStHXrVu3atUsDBgzQlVde6VEfbrjhBi1fvlzvv/+++vbtqzPPPFONGzfW2rVrtX37doWGhmrevHlmAYPKtGDBAnPjTalw8fuWLVvMKTJXXXWVWeGvLC1atFDfvn21evVqdevWTX379lVISIgaNmxofki47bbbtH37ds2ZM0dnnnmmevXqpQ4dOig9PV3btm3T7t279dNPP5kf9kaNGqUHH3xQjz/+uLkBd4sWLbRt2zatXbtW/v7+mjNnjtOH89rmkUce0erVq7VkyRJ16tRJQ4cOVWRkpFasWKG9e/eqQYMG+uSTT8xRCV9688039eabb5rfHzp0SFLhz7FtcX+TJk30xRdfeHzN+vXr69dff9WYMWMcfn7atm2rkydPasWKFUpOTlZQUJCeeeYZlwHc2/dyYGCgLrjgAn366afq2bOnBg8ebFbetH9+7lx77bXasGGDLBaL6tevb4asksaMGaMxY8Y4HHviiSe0atUqrVixQu3bt9ewYcOUmZmpJUuWKC8vTzNmzNB5553ndK21a9fq5ptvNr+3jWK/9tpr+vrrr83jX3zxhRkETp48qZtvvlnTp09Xz5491apVKxmGoX379mnt2rXKzc1V06ZNtWDBAq+miCUmJio8PFwLFizQ4MGD1b59e/n7+2vQoEHmNL158+Zp2LBh+t///qfWrVvrjDPOUF5enn766Selp6erd+/eTtUBy7Jw4UI99thj5vdbtmyRJD388MMO11q5cqV5u1GjRpozZ44mT56sTz75RD///LP69eunwMBArV69WgcOHJCfn59efPFFl5VQq1rDhg01bdo0vfDCCxo+fLiGDBmipk2bavv27Vq7dq35O7A0Q4YMUa9evbRu3ToFBgbq+uuv97ofR44c0dixY5Wbm6vmzZtr+fLlWr58ucu2//73v82CPFLhWrb58+dr5MiReuKJJ/Tll1+qa9eu2rRpkzZt2qTw8HB98sknTmGrIu+jqvhdhDrMV/X2gdqotI2h3VHR3ialcbcvlc2KFSuMyy67zGjWrJkRGBho1K9f3+jRo4dx9913O2zyaXP48GHjoYceMnr37m1ERkYaQUFBRvPmzY3TTz/dmDVrlrFhwwaP+28zb94846yzzjJiYmKMwMBAIz4+3pg0aZKxbds2l+2rYh+ywMBAo2nTpsYFF1xgfPHFF14/7j///GNcfvnlRpMmTYyAgAC3+zx9++23xoUXXmjExsaa+7b179/feOSRR4xjx465bH/uuecaDRo0MAICAoy4uDjjkksucdiHyl5Z+5C5e8082dzaFU/2IZs4cWKp18jLyzNeeeUVY+DAgebPVNu2bY1bbrnF7aa9pf3sl7XPVnmfq/3m2u6+yru3V0FBgfHf//7XuPDCC42mTZsaQUFBRlRUlNGtWzfjjjvucPn6luTNe/nYsWPGDTfcYLRo0cIIDAz06HeJTVl7+dm+3P2s5eTkGE8++aTRtWtXIzQ01IiOjjbOOOMMY/78+W4f0/ZvVtaX/euUmZlpPPPMM8aFF15otG3b1oiMjDQCAgKMhg0bGmeccYbx9NNPe70hss2yZcuMESNGGPXq1TP8/Pxc/pwfO3bMuO+++4xOnToZISEhRlhYmNGrVy/jqaeeMk6ePOn1Y9pvtF7alyvr1683Jk2aZLRp08YIDg42goKCjJYtWxpXXHGF298lpfF2HzjDcL9XltVqNd566y2jT58+RkREhBEdHW0MHjzY+OijjwzD8Oz/c7Z9Mi+77DKvn4thFP/O8PZnzN7OnTuNq6++2mjatKn5/5Orr77a+Ouvv1y2r8j7qCp/F6HusRiGi01TAAAAgEpQUFCgtm3b6p9//tHy5cuVkJDg6y4BNQpryAAAAFBlXn/9df3zzz9KSEggjAEusIYMAAAAlWr79u16+umnlZSUpEWLFsnPz0///ve/fd0toEYikAEAAKBSHTp0SG+99ZaCgoLUpUsXPfzwwzr99NN93S2gRmINGQAAAAD4CGvIAAAAAMBHCGQAAAAA4COsIaskVqtVBw8eVGRkpCwWi6+7AwAAAMBHDMPQiRMn1LRpU/n5lT4GRiCrJAcPHlR8fLyvuwEAAACghti3b5+aN29eahsCWSWJjIyUVPiiR0VF+bg3AAAAAHwlPT1d8fHxZkYoDYGsktimKUZFRRHIAAAAAHi0lImiHgAAAADgIwQyAAAAAPARAhkAAAAA+AiBDAAAAAB8hEAGAAAAAD5CIAMAAAAAHyGQAQAAAICPEMgAAAAAwEcIZAAAAADgIwQyAAAAAPARAhkAAAAA+AiBDAAAAAB8hEAGAAAAAD5CIAMAAAAAHyGQAQAAAICPEMgAAAAAwEcIZAAAAADgIwQyAAAAAPARAhkAAAAA+AiBDAAAAAB8hEAGAAAAoEbLzbdq6odr9e7yv33dlUpHIAMAAABQo1373mot3HhIs77c7OuuVDoCGQAAAIAabdmOZF93ocoQyAAAAADARwhkAAAAAOAjBDIAAAAANVrDiCBfd6HKEMgAAAAA1GhRIYHmbcMwfNiTykcgAwAAAFCj+ftZzNv5VgIZAAAAAFQb+0CWV2D1YU8qH4EMAAAAQI3mEMjyGSEDAAAAgGpjKc5jymWEDAAAAACqj9UugzFlEQAAAACqkdWusiKBDAAAAACqEYEMAAAAAHzEvtJ9LkU9AAAAAKD6WK2MkAEAAACATzBlEQAAAAB8pMAukFH2HgAAAACqkWPZe9aQAQAAAEC1cZiymM8IGQAAAABUmn0pJ/WvhVt0MDXL5XnWkAEAAABAFbnuvdV645c9mvzOHy7P22cw1pABAAAAQCXalnRCkrT98AmX5w37oh5MWaxcBw4c0JVXXqkGDRooNDRU3bp10+rVq83zhmFo5syZatKkiUJDQzVixAjt3LnT4RopKSm64oorFBUVpZiYGE2ZMkUZGRkObTZs2KAhQ4YoJCRE8fHxmj17tlNfPvnkE3Xs2FEhISHq1q2bvvnmm6p50gAAAAA8Zl9l0ahbNT18G8iOHz+uQYMGKTAwUN9++622bNmiZ555RvXq1TPbzJ49Wy+++KLmzJmj33//XeHh4UpMTFR2drbZ5oorrtDmzZu1ePFiff3111q2bJmuv/5683x6erpGjhypli1bas2aNXr66af18MMP6/XXXzfbLF++XJdddpmmTJmidevWacyYMRozZow2bdpUPS8GAAAAcIoKCSw9lthvDF1QxxKZxTB894zuvfde/fbbb/rll19cnjcMQ02bNtUdd9yhO++8U5KUlpam2NhYzZ07VxMmTNDWrVvVuXNn/fHHH+rbt68kadGiRTr33HO1f/9+NW3aVK+++qoeeOABJSUlKSgoyHzsBQsWaNu2bZKk8ePHKzMzU19//bX5+AMHDlTPnj01Z86cMp9Lenq6oqOjlZaWpqioqAq9LgAAAMCppM9ji3UsM1eS9PdTo53Od531nTJy8iVJ/7qoq64Y0LJa++ctb7KBT0fIvvzyS/Xt21eXXHKJGjdurF69eumNN94wz+/Zs0dJSUkaMWKEeSw6OloDBgzQihUrJEkrVqxQTEyMGcYkacSIEfLz89Pvv/9utjnjjDPMMCZJiYmJ2r59u44fP262sX8cWxvb45SUk5Oj9PR0hy8AAAAA3vPzs5R63r7KorVuDZD5NpDt3r1br776qtq3b6/vvvtON910k2699Va9++67kqSkpCRJUmxsrMP9YmNjzXNJSUlq3Lixw/mAgADVr1/foY2ra9g/hrs2tvMlPfnkk4qOjja/4uPjvX7+AAAAAMp2MrfAvO3DCX5VwqeBzGq1qnfv3nriiSfUq1cvXX/99bruuus8miLoa/fdd5/S0tLMr3379vm6SwAAAECds/WQ40y0gjo2RObTQNakSRN17tzZ4VinTp20d+9eSVJcXJwk6fDhww5tDh8+bJ6Li4vTkSNHHM7n5+crJSXFoY2ra9g/hrs2tvMlBQcHKyoqyuELAAAAgPdKG/RKSs92+L6O5THfBrJBgwZp+/btDsd27Nihli0LF+m1bt1acXFxWrJkiXk+PT1dv//+uxISEiRJCQkJSk1N1Zo1a8w2P/74o6xWqwYMGGC2WbZsmfLy8sw2ixcvVocOHcyKjgkJCQ6PY2tjexwAAAAA1S89K8/he6YsVqLbb79dK1eu1BNPPKG//vpL8+bN0+uvv66pU6dKkiwWi6ZPn67HH39cX375pTZu3Kirr75aTZs21ZgxYyQVjqidc845uu6667Rq1Sr99ttvmjZtmiZMmKCmTZtKki6//HIFBQVpypQp2rx5sz7++GO98MILmjFjhtmX2267TYsWLdIzzzyjbdu26eGHH9bq1as1bdq0an9dAAAAABRKz853+N5axwJZgC8fvF+/fvriiy9033336dFHH1Xr1q31/PPP64orrjDb3H333crMzNT111+v1NRUDR48WIsWLVJISIjZ5sMPP9S0adM0fPhw+fn5ady4cXrxxRfN89HR0fr+++81depU9enTRw0bNtTMmTMd9io7/fTTNW/ePD344IO6//771b59ey1YsEBdu3atnhcDAAAAOGW5D1knsh1HyAqsVd2X6uXTfcjqEvYhAwAAAMqn7+OLdTTDeR8ywzD03A879eKSneaxuxI7aOrQdtXeR294kw18OkIGAAAAAK78Z8lOzV3+t/q3ru9wvK6NJ/l0DRkAAAAAuPLM4h06lpmrbzc57gtsP2UxJ79AH63aqwOpWdXcu8rDCBkAAAAAnyo56HUsI8dtW/uiHq/+vEvP/7BTkSEB2vhwYlV1r0oxQgYAAACgxmh170L9uO2I2/P2UxZ/3XlUknSiRCXG2oRABgAAAKBGmb96n9tz9htDBwfW/jhT+58BAAAAgFqtZJmOkEB/pzZBAYXRpcBuhCzIv/bHmdr/DAAAAADUKTl5zpuNhRaFNPs1ZLaQVpvV/mcAAAAAoE5Jzcp1OhYRXFiP0L4ASFCA80habUMgAwAAAFCjpJ7MczrWNCZEkmS1W0TmZ6m2LlUZAhkAAAAAnyq52XNqlnMgaxIdKslxDZm1DuwRzT5kAAAAAHzqeIkRsdx85zVkESHFUxbzCqya8u5qLduRXC39q0qMkAEAAADwmTQXo2GuhAQUF/X4YcvhOhHGJAIZAAAAAB/KzPFsU2fbnmNWw9DRjByn89ZaOn+RQAYAAADAZ07mFnjUzrbnWIFVSs92DnEFBoEMAAAAAEq1OzlDT3671RzlyiolkPnblVEM9C+8bRiG8gqc15gV1NIRMop6AAAAAKg2F7z0mzJy8rXt0Am9e01/Zea6n7Lob7GoQIVBy68onFkNQwEu6t3X1kDGCBkAAACAapNRtGZs7d7jkqSTpQQyP7u04mexBTLJ3885xuQTyAAAAADAO65K3Nv4WyxOtwusrkfIKOoBAAAAAB6yRSoXy8FMfnbBKzSosOz9F+sOyJBz+HK1rqw2IJABAAAAqHaWohGvfGspI2R2gSwypLj8xfakDKe2r/y8qxJ7V30IZAAAAACqnW02orWUcvWjusZJkprFhJpryCTXe5e9t+LvSu1fdaHKIgAAAIBq58mUxfaNI/XbvcPUIDxIX/150DweFOA8rhTgXzvHmghkAAAAAHymtGIcRzNy1CwmVJLUMS7KPO5qmqOrQh+1Qe2MkQAAAABqNdsasoKiKYsjOsXqsTFdHdqctNs0ulvzaPN2XoFziPMnkAEAAACAZ4qnLBaGK38/xzL3kmSUWF9mGy1zVVExsJZOWaydvQYAAABQq5Us6uHvZ1HJQa6S42C2UTBXgYwpiwAAAABQiu1JJ+y+K97oWZL8LBaHfcck5wqMAWYgc56y2DM+pvI6Wo0IZAAAAACqRVZe8Zow23TE4imLFofS9pKUXyJ4+ZUyQtaxSZTTsdqAQAYAAACgWgTblavPyS8MVeaURYvzlMWSI2EBpQSykuvNagsCGQAAAIBqEeQQyApHy2zZys/P4lQp8aJezRy+t50/mVOgkkrbYLomI5ABAAAAqHa20S/7ETJLiSmLg9s3dPjeFsh2H810ul4p25nVaAQyAAAAANXC1bRCs6iHn8Wh7H29sECntqXtNcYIGQAAAACUwsXSL+Xb7UNmn7f8/ZyjSsl9yuzV0jxGIAMAAABQPexHsRpGBBUeKwpkAX5+yrVLbDl5zuvESh0hq6VzFglkAAAAAKqFfSDr07KeJKnAKN6HLD0rz2VbmwD/0qYsVlYvqxeBDAAAAEC1sM9YtsEwq92UxQy76oklC3xIctqnzB5ryAAAAACgFPahyVpiY2g/P4tOZOe5vJ9NQClTFmsrAhkAAACAamF1GCErCmR2Ze8v69+i1Pu7KvRRfG1GyAAAAADALVcjZMVTFi2Krx9mnndVIt+/lPRCIAMAAACAUtiHLNtN+6IeZQkodYSsYn3zFQIZAAAAgGphH5oM2daQFX5fWkl7G79S2rgaUasNCGQAAAAAqoX9XmG2/GQ/ZdGeqyqLpRX1sLrYdLo2IJABAAAAqBYFrqosejFlsdSNoRkhAwAAAAD37DOT8whZybYuinq4CG3TR7QvvE7tzGMEMgAAAADVw1rBoh7+/o5tOjeJUliQf9H1amciI5ABAAAAqBauinrku1lD5krJEbJzu8WZQa52xjECGQAAAIBq4mqEzDZlsbSCHTb2oS0yOEA3nNnWLP7BGjIAAAAAKIXhqqhHUSArraS9jX1oG9klToH+frIdYg0ZAAAAAJTCvjS9LT/Zgpmrgh0l2Y+Q2Zr7MUIGAAAAAGWzOoyQFf7X3QhZnoshL/tA5mdx/C9FPQAAAACgFA4Zy9yHrPDbkiNkufnOOz07BrLC2+YaMjaGBgAAAAD37EexzCmL3lRZZMoiAAAAAJRPQSUW9bCNjFHUAwAAAAA88Mz3O8zbJTeGtk1ZjAoJkCQFBzhHFT8Xa8gsZiCrnYmMQAYAAACgWuw5mmnetpbYh8y/KJn89/qBOr1tA310/UCn+zuMkMk2Qla7pywG+LoDAAAAAE49hlnUo2jKYlGw6tI0WvOucw5jkuTvVzyeZMtmtnVlBbV0ziIjZAAAAAB8xpuiHkH+zmvICGQAAAAA4CVryREyTwKZ3boyCyNkAAAAAOC5c7rEmbdtS77yCxyLepQm0N9+ymJh+wACGQAAAACUzepiH7K8gsIdne3Dljv2I2R+JfYhK6ilRT0IZAAAAACqhf0gli2cZecVBrLQIP8y728f2mxryAL8GSEDAAAAgDI5lKYvupmTXyBJCgn0boTMUnKEjEAGAAAAAO7ZBzLb7azcokAWUPYIWZDLNWSFxwhkAAAAAFAK+9BkSJqzdJcyiwKZ11MWi/5r25qMQAYAAAAApbCfsXjgeJae+nab+b0nI2QBdvuQ5eYXrj1jhAwAAAAAPGAfmvJLBKhgD9aQNYoINm8fSM2SJNkGzaiyCAAAAAClsJYSmoI8KHsfXz/MvG27lH/RCJltP7PaxqeB7OGHH5bFYnH46tixo3k+OztbU6dOVYMGDRQREaFx48bp8OHDDtfYu3evRo8erbCwMDVu3Fh33XWX8vPzHdr8/PPP6t27t4KDg9WuXTvNnTvXqS8vv/yyWrVqpZCQEA0YMECrVq2qkucMAAAAnKpKC2R+fmVvDO3qWrYNpUu7dk3m8xGyLl266NChQ+bXr7/+ap67/fbb9dVXX+mTTz7R0qVLdfDgQY0dO9Y8X1BQoNGjRys3N1fLly/Xu+++q7lz52rmzJlmmz179mj06NEaOnSo1q9fr+nTp+vaa6/Vd999Z7b5+OOPNWPGDM2aNUtr165Vjx49lJiYqCNHjlTPiwAAAACcAipzmZftWraiHiWnQNYWPg9kAQEBiouLM78aNmwoSUpLS9Nbb72lZ599VsOGDVOfPn30zjvvaPny5Vq5cqUk6fvvv9eWLVv0wQcfqGfPnho1apQee+wxvfzyy8rNzZUkzZkzR61bt9YzzzyjTp06adq0abr44ov13HPPmX149tlndd1112ny5Mnq3Lmz5syZo7CwML399tvV/4IAAAAAdVRlFt4wikbEbEU9kk/kmMdqE58Hsp07d6pp06Zq06aNrrjiCu3du1eStGbNGuXl5WnEiBFm244dO6pFixZasWKFJGnFihXq1q2bYmNjzTaJiYlKT0/X5s2bzTb217C1sV0jNzdXa9ascWjj5+enESNGmG1cycnJUXp6usMXAAAAAPcqMzCZUxbtEs1vfx2rtOtXF58GsgEDBmju3LlatGiRXn31Ve3Zs0dDhgzRiRMnlJSUpKCgIMXExDjcJzY2VklJSZKkpKQkhzBmO287V1qb9PR0ZWVl6ejRoyooKHDZxnYNV5588klFR0ebX/Hx8eV6DQAAAIBTRWVWQrQNttlPVczOK6i061eXAF8++KhRo8zb3bt314ABA9SyZUvNnz9foaGhPuxZ2e677z7NmDHD/D49PZ1QBgAAAJTCanV93OJdPY/CaxWFu4jg4kgTFRpYnm75lM+nLNqLiYnRaaedpr/++ktxcXHKzc1VamqqQ5vDhw8rLi5OkhQXF+dUddH2fVltoqKiFBoaqoYNG8rf399lG9s1XAkODlZUVJTDFwAAAAD33FVC9CtHIrNdq3m94lL4tXFz6BoVyDIyMrRr1y41adJEffr0UWBgoJYsWWKe3759u/bu3auEhARJUkJCgjZu3OhQDXHx4sWKiopS586dzTb217C1sV0jKChIffr0cWhjtVq1ZMkSsw0AAACAinMXyMoxQOYw2tYxLlISgcxrd955p5YuXaq///5by5cv10UXXSR/f39ddtllio6O1pQpUzRjxgz99NNPWrNmjSZPnqyEhAQNHDhQkjRy5Eh17txZV111lf7880999913evDBBzV16lQFBxfu4n3jjTdq9+7duvvuu7Vt2za98sormj9/vm6//XazHzNmzNAbb7yhd999V1u3btVNN92kzMxMTZ482SevCwAAAFAXuQtM5RkhM1R8Lf+iPcwqc41adfHpGrL9+/frsssu07Fjx9SoUSMNHjxYK1euVKNGjSRJzz33nPz8/DRu3Djl5OQoMTFRr7zyinl/f39/ff3117rpppuUkJCg8PBwTZw4UY8++qjZpnXr1lq4cKFuv/12vfDCC2revLnefPNNJSYmmm3Gjx+v5ORkzZw5U0lJSerZs6cWLVrkVOgDAAAAQPkcTs/WruRMl+dyC9wsLiuF/QhZgC2QuVukVoNZjNpYrL8GSk9PV3R0tNLS0lhPBgAAAJQwdd5aLdxwyO35v58a7dF1Wt27UJLUt2U9fXrT6ZKki175Tev2pur1q/poZBf3dSCqizfZoEatIQMAAABQNx09kVOp17Nfj2YbIXO3Rq0mI5ABAAAAqHKB/pUbPeyXo9nWoOVT1AMAAAAAnNkKb7gyulsTr69nv/IqwN+2hoxABgAAAABOAkoJZI0ig72+nv2Im79f4W0CGQAAAAC4UFpl+0B/z8ve//uSHmpRP0xPjetmHrPdvTZOWfRp2XsAAAAAp4acfPcl6YMCPB8nurhPc13cp7nDMdsImbUWBjJGyAAAAABUudxSAtnFfeIrdG3bdMjaOEJGIAMAAABQ5To1cb8fV+uG4RW6tr8fRT0AAAAAwK25y/+WJI3p2VSTTm9VqdcmkAEAAACAB1buTnEogd8xLrLC1wwgkAEAAACAa/brx0KD/B0C2dUJrSp8fT/WkAEAAACAaymZuebt0EB/+dnVwC9lezKP2UbIrAaBDAAAAAAcZObmm7cLrIbDJtF+pW1Q5iFzhKyAQAYAAAAADuz3B8srsJoBSip9w2hPFa8hc19av6YikAEAAACoUvZru/KsVvlbKneEzKyyyJRFAAAAAHBkX/0wOjRQ/nYpxK8SEokt4FHUAwAAAABKsA9kz4/vKX+7FFYpI2T+RUU9CGQAAAAA4GhbUrokKSokQO0aRzqMkFkqIZAFUPYeAAAAAFy757ONkqT07MJqi2lZeea5yih7b5uyyMbQAAAAAFCGg6nZlXo92xRIAhkAAAAAeKEyQlSAPyNkAAAAAOARw648vbUSStX7UWURAAAAADxjH5vyCyphhMyPKosAAAAA4BH7qYWVMkJGlUUAAAAA8Iz9qFhlhCjbCBlryAAAAACgDJ2bRpm3K2OaoT+BDAAAAAA8c/0ZbczblRGi/JmyCAAAAACeCQn0N29XRogqHiGzVvha1Y1ABgAAAMBnKqOoh7mGrPYNkBHIAAAAAPgOI2QAAAAAUIVio4IlSU+N7eZ0rnFkSIWvX5uLegT4ugMAAAAA6raggMJxoNPiIs1jr1/VRyt2H9OYnk0rfP3aXPaeQAYAAACgShUULe4K9CueoDeyS5xGdomrlOv7WaiyCAAAAAAu2YKSXxWljwD/wkBWGXuaVTcCGQAAAIAqZctJtpGsyuZflPQYIQMAAAAAJ0UjZFUVyCy1dw0ZgQwAAABAlbLlpCrKY7W6yiKBDAAAAECVMoo2f66iPEYgAwAAAAB3bDHJUmVryKiyCAAAAAAu2aofVtWUxdq8DxmBDAAAAECVMkfIquj6TFkEAAAAAHeqvOx9USAzCGQAAAAA4MBqMGXRHQIZAAAAgCpVPGWxahKZn62oR4G1Sq5flQhkAAAAAKqUUcX7kNlGyGrhABmBDAAAAEDVquopi8Vl72vfCFmArzsAAAAAoG6r6n3IggP81aVplIIDat94E4EMAAAAQJUyikbI/KpohKxRZLAW3jqkai5exWpfhAQAAABQq5hryKpsJ7Lai0AGAAAAoErZpixW1QhZbUYgAwAAAFClrMVDZCiBQAYAAACgSjFl0T0CGQAAAIAqYyvoITFl0RUCGQAAAIAqY5fHqqzsfW1GIAMAAABQZezyGBMWXSCQAQAAAKgyjlMWiWQlEcgAAAAAVBkrQ2SlIpABAAAAqDKG3aRFBsicEcgAAAAAVBn7oh5MWXRGIAMAAABQZRyqLPquGzUWgQwAAABAlWHKYukIZAAAAACqDFMWS0cgAwAAAFBlrPaJDE4IZAAAAACqjEPVewbInBDIAAAAAFQZpiyWjkAGAAAAoELeX/mPRjy7VAdSs5zOGXaJjDjmrMYEsqeeekoWi0XTp083j2VnZ2vq1Klq0KCBIiIiNG7cOB0+fNjhfnv37tXo0aMVFhamxo0b66677lJ+fr5Dm59//lm9e/dWcHCw2rVrp7lz5zo9/ssvv6xWrVopJCREAwYM0KpVq6riaQIAAAB1zkMLNumvIxl65MvNTuccyt4zQuakRgSyP/74Q6+99pq6d+/ucPz222/XV199pU8++URLly7VwYMHNXbsWPN8QUGBRo8erdzcXC1fvlzvvvuu5s6dq5kzZ5pt9uzZo9GjR2vo0KFav369pk+frmuvvVbfffed2ebjjz/WjBkzNGvWLK1du1Y9evRQYmKijhw5UvVPHgAAAKgj9qacdDr22dr95m0/8pgTnweyjIwMXXHFFXrjjTdUr14983haWpreeustPfvssxo2bJj69Omjd955R8uXL9fKlSslSd9//722bNmiDz74QD179tSoUaP02GOP6eWXX1Zubq4kac6cOWrdurWeeeYZderUSdOmTdPFF1+s5557znysZ599Vtddd50mT56szp07a86cOQoLC9Pbb79dvS8GAAAAUIvl5Fudjr2+bLd5mxEyZz4PZFOnTtXo0aM1YsQIh+Nr1qxRXl6ew/GOHTuqRYsWWrFihSRpxYoV6tatm2JjY802iYmJSk9P1+bNm802Ja+dmJhoXiM3N1dr1qxxaOPn56cRI0aYbVzJyclRenq6wxcAAABwKgtgCMxrAb588I8++khr167VH3/84XQuKSlJQUFBiomJcTgeGxurpKQks419GLOdt50rrU16erqysrJ0/PhxFRQUuGyzbds2t31/8skn9cgjj3j2RAEAAIBTQIC/z8d7ah2fvWL79u3Tbbfdpg8//FAhISG+6ka53XfffUpLSzO/9u3b5+suAQAAAD4V6M8Imbd8FsjWrFmjI0eOqHfv3goICFBAQICWLl2qF198UQEBAYqNjVVubq5SU1Md7nf48GHFxcVJkuLi4pyqLtq+L6tNVFSUQkND1bBhQ/n7+7tsY7uGK8HBwYqKinL4AgAAAE419mXt2WfMez4LZMOHD9fGjRu1fv1686tv37664oorzNuBgYFasmSJeZ/t27dr7969SkhIkCQlJCRo48aNDtUQFy9erKioKHXu3NlsY38NWxvbNYKCgtSnTx+HNlarVUuWLDHbAAAAAHDNalfW3nBx3tUxFPPZGrLIyEh17drV4Vh4eLgaNGhgHp8yZYpmzJih+vXrKyoqSrfccosSEhI0cOBASdLIkSPVuXNnXXXVVZo9e7aSkpL04IMPaurUqQoODpYk3XjjjXrppZd0991365prrtGPP/6o+fPna+HChebjzpgxQxMnTlTfvn3Vv39/Pf/888rMzNTkyZOr6dUAAAAAaqcCu0RmtRK/vOXToh5lee655+Tn56dx48YpJydHiYmJeuWVV8zz/v7++vrrr3XTTTcpISFB4eHhmjhxoh599FGzTevWrbVw4ULdfvvteuGFF9S8eXO9+eabSkxMNNuMHz9eycnJmjlzppKSktSzZ08tWrTIqdAHAAAAAEdWuymL9rfhGYth8KpVhvT0dEVHRystLY31ZAAAADhlZObkq8us7yRJHeMitWj6GQ7n+/3rByWfyJEk/f3U6Grvny94kw2oSwkAAACg3ArsxndcDfUw/FM6AhkAAACAcrNfN+ZqyiKFF0tHIAMAAABQbvZFPQoYDvMagQwAAABAudmHMFdVFslopSOQAQAAACg3q7X49t/HTuq/q/b6rjO1EIEMAAAAQLmVnKZ43+cbHb4/vW2D6uxOrUMgAwAAAFBuZW0GHRxQGDnuSuxQHd2pdQhkAAAAAMqtoIxAZhtBC/Cj3KIrBDIAAAAA5VZWZUVbYPMnkLlEIAMAAABQbmVNWSSQlY5ABgAAAKDcyhohy8otkCSFBflXR3dqHQIZAAAAgHIraw3Ziex8SVJEcGB1dKfWIZABAAAAKDf7fchsjKJRM6vV0Kq/UyRJESEB1dmtWoNABgAAAKDcXE1ZzCsoPLb5YLp5LMif6OEKrwoAAACAcnM1ZdF2zGJXx6NpTEh1dalWIZABAAAAKDerqxGyonmM9mGtZYPwautTbUIgAwAAAFBurkbI8oumLOYXnWtRP6xa+1SbEMgAAAAAlJurfcjyCxxHyAL82YPMHQIZAAAAgHJzVdRj5Z7Cyoq2YBbAptBuEcgAAAAAlJurKYu3/nedkk/kmFMWA/yIHe7wygAAAAAoN1dFPSQpJTNX+UXFPZiy6B6BDAAAAEC5FbjYGFqS/CzFxT38mbLoFoEMAAAAQLm5mrIoFW4ObTsXyJRFt3hlAAAAAJSbuymLeQVW5VkZISsLgQwAAABAubkbIcu3WlXAGrIyEcgAAAAAlJv7ETJDeQW2KosEMncIZAAAAADKze0Imd0aMn/WkLlVrldm165devDBB3XZZZfpyJEjkqRvv/1WmzdvrtTOAQAAAKjZ3Bf1sJr7kAUyZdEtrwPZ0qVL1a1bN/3+++/6/PPPlZGRIUn6888/NWvWrErvIAAAAICaq7SiHvlFNfEp6uGe14Hs3nvv1eOPP67FixcrKCjIPD5s2DCtXLmyUjsHAAAAoGZztw9ZvrV4yiJryNzzOpBt3LhRF110kdPxxo0b6+jRo5XSKQAAAAC1Q0FpZe9tRT38WUPmjtevTExMjA4dOuR0fN26dWrWrFmldAoAAABA7WAtdWPoorL3jJC55XUgmzBhgu655x4lJSXJYrHIarXqt99+05133qmrr766KvoIAAAAoIZyV9Rj88E0/fv7HZLYh6w0XgeyJ554Qh07dlR8fLwyMjLUuXNnnXHGGTr99NP14IMPVkUfAQAAANRQ7op6vPPb3+btAMreuxXg7R2CgoL0xhtvaObMmdq4caMyMjLUq1cvtW/fvir6BwAAAKAGczdCZo8qi+55Hchs4uPjFR8fX5l9AQAAAFDLuCvqYY8pi+55PXY4btw4/d///Z/T8dmzZ+uSSy6plE4BAAAAqB3cFfWwR1EP97wOZMuWLdO5557rdHzUqFFatmxZpXQKAAAAQO3gbh8ye6whc8/rVyYjI8NhQ2ibwMBApaenV0qnAAAAANQOttL2ktQ0OkTdmkU7tWGEzD2vA1m3bt308ccfOx3/6KOP1Llz50rpFAAAAIDawbaGbNLprbT8vuFq3TDcqY0fgcwtr4t6PPTQQxo7dqx27dqlYcOGSZKWLFmi//73v/rkk08qvYMAAAAAai7blEU/i6Xovz7sTC3kdSA7//zztWDBAj3xxBP69NNPFRoaqu7du+uHH37QmWeeWRV9BAAAAFBD2fYh8y+ae2cLZvYMDyoxnqrKVfZ+9OjRGj16dGX3BQAAAEAtY9uHzL+ocIfFZSCr1i7VKpQ7AQAAAFBuxYGs8HtXUxZDAv2rsUe1i0cjZPXr19eOHTvUsGFD1atXz2XqtUlJSam0zgEAAACo2cwpi0UZYfU/x53ahAQRyNzxKJA999xzioyMlCQ9//zzVdkfAAAAALWIbYTMVknx72OZTm3CGCFzy6NANnHiRElSfn6+LBaLEhMTFRsbW6UdAwAAAFDzffj7XknFI2SBfn7KLbFbdGxUSLX3q7bwag1ZQECAbrzxRmVnZ1dVfwAAAADUEvl2wSs5I0eSFODvuLypYUSwBrVrUK39qk28LurRv39/rVu3rir6AgAAAKAWyc4vDmT5RVMXA0pU9XjovE6l1qA41Xld9v7mm2/WHXfcof3796tPnz4KD3fcibt79+6V1jkAAAAANVdWboF5O7AoiAUFOI75EMZK53UgmzBhgiTp1ltvNY9ZLBYZhiGLxaKCggJ3dwUAAABQh2TnFX/2t201FuDnGMj8CWSl8jqQ7dmzpyr6AQAAAKCWsQ9ktvL3gQGOAcyfnY9L5VUgS09P144dO5Sbm6v+/furUaNGVdUvAAAAADVcll0gs9X3KDlC5scIWak8DmTr16/Xueeeq8OHD8swDEVGRmr+/PlKTEysyv4BAAAAqKHs15AZhuGyjb8fgaw0Hg8g3nPPPWrdurV+/fVXrVmzRsOHD9e0adOqsm8AAAAAarC8guIQZpuyWDKY+RHISuXxCNmaNWv0/fffq3fv3pKkt99+W/Xr11d6erqioqKqrIMAAAAAaqYCu/Blm7JYcpyMoh6l83iELCUlRc2bNze/j4mJUXh4uI4dO1YlHQMAAABQs1mtxfHLMEfIHNswZbF0XhX12LJli5KSkszvDcPQ1q1bdeLECfMY+5ABAAAAp4YCu0BmGy2zlpyyyAhZqbwKZMOHD3eaE3reeeexDxkAAABwCrKfsmjLZiVHyBggK53HgYz9xwAAAADYs5+y2L1ZtMs2rmsvwsbjQNayZcuq7AcAAACAWsZ+hGzSoFaSnKssuqmGjyLsmw0AAACgXGxryBLaNFCgf2G0KJm/Sq4pgyMCGQAAAIBysYUt+0qKJfMXgax0BDIAAAAA5WLbe8x+82dDTFn0BoEMAAAAQLnYinr421VSZITMOx4HsqysLH355ZcOe47ZpKen68svv1ROTk6ldg4AAABAzVXgYspiWJC/QxvyWOk8DmSvv/66XnjhBUVGRjqdi4qK0osvvqg333yzUjsHAAAAoOayFfWw3/z5pct7q03DcPP75vVCq71ftYnHgezDDz/U9OnT3Z6fPn263n33Xa8e/NVXX1X37t0VFRWlqKgoJSQk6NtvvzXPZ2dna+rUqWrQoIEiIiI0btw4HT582OEae/fu1ejRoxUWFqbGjRvrrrvuUn5+vkObn3/+Wb1791ZwcLDatWunuXPnOvXl5ZdfVqtWrRQSEqIBAwZo1apVXj0XAAAA4FTjqqhH12bR+vHOs/T5zadrzpW91T7WeUAHxTwOZDt37lSPHj3cnu/evbt27tzp1YM3b95cTz31lNasWaPVq1dr2LBhuvDCC7V582ZJ0u23366vvvpKn3zyiZYuXaqDBw9q7Nix5v0LCgo0evRo5ebmavny5Xr33Xc1d+5czZw502yzZ88ejR49WkOHDtX69es1ffp0XXvttfruu+/MNh9//LFmzJihWbNmae3aterRo4cSExN15MgRr54PAAAAcCoxR8jsAplN7xb1dE7XJtXdpVrHYpTcuc2NyMhI/fzzz+rTp4/L82vWrNFZZ53lco2ZN+rXr6+nn35aF198sRo1aqR58+bp4osvliRt27ZNnTp10ooVKzRw4EB9++23Ou+883Tw4EHFxsZKkubMmaN77rlHycnJCgoK0j333KOFCxdq06ZN5mNMmDBBqampWrRokSRpwIAB6tevn1566SVJktVqVXx8vG655Rbde++9HvU7PT1d0dHRSktLU1RUVIVeAwAAAKA2eOvXPXrs6y26oEdTvXhZL193p8bwJht4PELWpUsX/fDDD27Pf//99+rSpYvnvSyhoKBAH330kTIzM5WQkKA1a9YoLy9PI0aMMNt07NhRLVq00IoVKyRJK1asULdu3cwwJkmJiYlKT083R9lWrFjhcA1bG9s1cnNztWbNGoc2fn5+GjFihNnGlZycHKWnpzt8AQAAAKcSs8qiixEyeMbjQHbNNdfoscce09dff+107quvvtK//vUvXXPNNV53YOPGjYqIiFBwcLBuvPFGffHFF+rcubOSkpIUFBSkmJgYh/axsbFKSkqSJCUlJTmEMdt527nS2qSnpysrK0tHjx5VQUGByza2a7jy5JNPKjo62vyKj4/3+rkDAAAAtZmtyqJ9UQ94J8DThtdff72WLVumCy64QB07dlSHDh0kFU4j3LFjhy699FJdf/31XnegQ4cOWr9+vdLS0vTpp59q4sSJWrp0qdfXqW733XefZsyYYX6fnp5OKAMAAMAppcAcIfNxR2oxjwPZ3r179d577+mCCy7QvHnztGPHDhmGoQ4dOuiRRx7RpZdeWq4OBAUFqV27dpKkPn366I8//tALL7yg8ePHKzc3V6mpqQ6jZIcPH1ZcXJwkKS4uzqkaoq0Ko32bkpUZDx8+rKioKIWGhsrf31/+/v4u29iu4UpwcLCCg4PL9ZwBAACAuoApixXncZZt3bq1jh49qksvvVQLFizQ5s2btWXLFi1YsKDcYcwVq9WqnJwc9enTR4GBgVqyZIl5bvv27dq7d68SEhIkSQkJCdq4caNDNcTFixcrKipKnTt3NtvYX8PWxnaNoKAg9enTx6GN1WrVkiVLzDYAAAAAnDFlseI8HiHzsBijV+677z6NGjVKLVq00IkTJzRv3jz9/PPP+u677xQdHa0pU6ZoxowZql+/vqKionTLLbcoISFBAwcOlCSNHDlSnTt31lVXXaXZs2crKSlJDz74oKZOnWqOXt1444166aWXdPfdd+uaa67Rjz/+qPnz52vhwoVmP2bMmKGJEyeqb9++6t+/v55//nllZmZq8uTJlf6cAQAAgLqigBGyCvM4kEmSpZKT75EjR3T11Vfr0KFDio6OVvfu3fXdd9/p7LPPliQ999xz8vPz07hx45STk6PExES98sor5v39/f319ddf66abblJCQoLCw8M1ceJEPfroo2ab1q1ba+HChbr99tv1wgsvqHnz5nrzzTeVmJhothk/frySk5M1c+ZMJSUlqWfPnlq0aJFToQ8AAAAAxY5m5EiS6ocH+bgntZfH+5D5+fnp+uuvV1hYWKntnn322UrpWG3DPmQAAAA41Vz99iot25Gs2Rd316V9KXBn40028GqEbOPGjQoKcp9+K3sEDQAAAEDNdSg1S5LUNDrUxz2pvbwKZF988YUaN25cVX0BAAAAUEsYhqGDtkAWE+Lj3tReHldZZPQLAAAAgE1ugVWZuQWSpAbhbAdVXh4HsqqosggAAACgdrJVWJSkwAAGb8rL40D2zjvvKDo6uir7AgAAAKCWyCsoDmQBfh7HCpTg8RqyiRMnSpKOHTumBg0aSJL27dunN954Q1lZWbrgggs0ZMiQquklAAAAgBrFfoQsgH3Iys3jKLtx40a1atVKjRs3VseOHbV+/Xr169dPzz33nF5//XUNHTpUCxYsqMKuAgAAAKgp8guskiQ/i+RHICs3jwPZ3XffrW7dumnZsmU666yzdN5552n06NFKS0vT8ePHdcMNN+ipp56qyr4CAAAAqCHyikbIAvyZrlgRHk9Z/OOPP/Tjjz+qe/fu6tGjh15//XXdfPPN8iuaL3rLLbdo4MCBVdZRAAAAADVHQdEaMqYrVozHcTYlJUVxcXGSpIiICIWHh6tevXrm+Xr16unEiROV30MAAAAANU6etXDKIoGsYrwaXyy5Fxl7kwEAAACnpgKmLFYKj6csStKkSZMUHFy46Vt2drZuvPFGhYeHS5JycnIqv3cAAAAAaqS8AkbIKoPXZe9trrzySqc2V199dcV7BAAAAKDGyy9aQxbICFmFeBzI3nnnnarsBwAAAIBaJL9oyqI/I2QVQpwFAAAA4DXbPmQB/gSyiiCQAQAAAPCabYQs0I9IURG8egAAAAC8lpaVJ0kKD/b3cU9qNwIZAAAAAK8dSsuWJDWJCfVxT2o3AhkAAAAArx1KzZIkNY0O8XFPajcCGQAAAACv/fHPcUlSXDQjZBVBIAMAAADglfwCq/7clypJasIIWYUQyAAAAAB4JSffat5OaNPAhz2p/QhkAAAAALxiK3kvSeHBAT7sSe1HIAMAAADgFdum0JIU4MfG0BVBIAMAAADgFdsImZ9F8iOQVQiBDAAAAIBXbIEswJ84UVG8ggAAAAC8YpuyyHTFiiOQAQAAAPCKOUJGIKswAhkAAAAAr+QXFAayQKYsVhivIAAAAACv5BVNWfRnhKzCCGQAAAAAvFJgZYSssvAKAgAAAPBKvrWoqIc/I2QVRSADAAAA4BXbGjKmLFYcgQwAAACAV2xVFgP9iBMVxSsIAAAAwCu2oh5MWaw4AhkAAAAAr+TkFway4ADiREXxCgIAAADwSnZegSQpOMDfxz2p/QhkAAAAALySk1c4QhYSSJyoKF5BAAAAAB4zDEMncvIlSSGBjJBVVICvOwAAAACg9pj15Wa9t+IfSQSyysAIGQAAAACP2cKYRFGPysArCAAAAKBcGCGrOAIZAAAAgHKJCGYFVEURyAAAAACUCxtDVxyBDAAAAEC5nMjO93UXaj0CGQAAAACPWK2Gw/eXD2jho57UHQQyAAAAAB6x7T8mSR9MGaC2jSJ82Ju6gUAGAAAAwCMb96dJkprFhGpQuwY+7k3dQCADAAAA4JGMnDxJUpPoEFksFPSoDAQyAAAAAB7JKyhcQ0Z1xcpDIAMAAADgkXyrVZIU6E+MqCy8kgAAAAA8Yo6Q+TFCVlkIZAAAAAA8UmC1TVkkRlQWXkkAAAAAHskvsE1ZZISsshDIAAAAAHjENmXR348YUVl4JQEAAAB4xCzqwRqySkMgAwAAAOARyt5XPgIZAAAAAI/kF1DUo7LxSgIAAADwSAFTFisdgQwAAACAR/Ioe1/peCUBAAAAeMRW9p41ZJWHQAYAAADAI2ZRD6YsVhoCGQAAAACP2MreB7APWaXhlQQAAADgEVuVxUCmLFYaAhkAAAAAj+RT1KPS8UoCAAAA8IhZ1IM1ZJXGp4HsySefVL9+/RQZGanGjRtrzJgx2r59u0Ob7OxsTZ06VQ0aNFBERITGjRunw4cPO7TZu3evRo8erbCwMDVu3Fh33XWX8vPzHdr8/PPP6t27t4KDg9WuXTvNnTvXqT8vv/yyWrVqpZCQEA0YMECrVq2q9OcMAAAA1Fa2sveBjJBVGp++kkuXLtXUqVO1cuVKLV68WHl5eRo5cqQyMzPNNrfffru++uorffLJJ1q6dKkOHjyosWPHmucLCgo0evRo5ebmavny5Xr33Xc1d+5czZw502yzZ88ejR49WkOHDtX69es1ffp0XXvttfruu+/MNh9//LFmzJihWbNmae3aterRo4cSExN15MiR6nkxAAAAgBqOsveVz2IYhuHrTtgkJyercePGWrp0qc444wylpaWpUaNGmjdvni6++GJJ0rZt29SpUyetWLFCAwcO1LfffqvzzjtPBw8eVGxsrCRpzpw5uueee5ScnKygoCDdc889WrhwoTZt2mQ+1oQJE5SamqpFixZJkgYMGKB+/frppZdekiRZrVbFx8frlltu0b333ltm39PT0xUdHa20tDRFRUVV9ksDAAAA+NyUuX9oybYj+r9x3TS+Xwtfd6fG8iYb1KixxrS0NElS/fr1JUlr1qxRXl6eRowYYbbp2LGjWrRooRUrVkiSVqxYoW7duplhTJISExOVnp6uzZs3m23sr2FrY7tGbm6u1qxZ49DGz89PI0aMMNuUlJOTo/T0dIcvAAAAoC6zTVmk7H3lqTGvpNVq1fTp0zVo0CB17dpVkpSUlKSgoCDFxMQ4tI2NjVVSUpLZxj6M2c7bzpXWJj09XVlZWTp69KgKCgpctrFdo6Qnn3xS0dHR5ld8fHz5njgAAABQSzBlsfLVmEA2depUbdq0SR999JGvu+KR++67T2lpaebXvn37fN0lAAAAoErlU9Sj0gX4ugOSNG3aNH399ddatmyZmjdvbh6Pi4tTbm6uUlNTHUbJDh8+rLi4OLNNyWqItiqM9m1KVmY8fPiwoqKiFBoaKn9/f/n7+7tsY7tGScHBwQoODi7fEwYAAABqIcreVz6fRlvDMDRt2jR98cUX+vHHH9W6dWuH83369FFgYKCWLFliHtu+fbv27t2rhIQESVJCQoI2btzoUA1x8eLFioqKUufOnc029tewtbFdIygoSH369HFoY7VatWTJErMNAAAAcCp6Y9luTZ23Vjn5BYyQVQGfjpBNnTpV8+bN0//+9z9FRkaa67Wio6MVGhqq6OhoTZkyRTNmzFD9+vUVFRWlW265RQkJCRo4cKAkaeTIkercubOuuuoqzZ49W0lJSXrwwQc1depUcwTrxhtv1EsvvaS7775b11xzjX788UfNnz9fCxcuNPsyY8YMTZw4UX379lX//v31/PPPKzMzU5MnT67+FwYAAACoAV5buktPfrtNkjS8Y2NlZBfu9RscQCCrLD4NZK+++qok6ayzznI4/s4772jSpEmSpOeee05+fn4aN26ccnJylJiYqFdeecVs6+/vr6+//lo33XSTEhISFB4erokTJ+rRRx8127Ru3VoLFy7U7bffrhdeeEHNmzfXm2++qcTERLPN+PHjlZycrJkzZyopKUk9e/bUokWLnAp9AABQm+xLOakVu47pot7N+Is2AK/ZwpgkpWXl6e9jhfsFt20c4asu1Tk1ah+y2ox9yAAANVHXWd8pIydfd5/TQTef1c7X3QFQiyxYd0DTP17vdDwmLFDrHjpbFgvryNyptfuQAQCAypWRUzi9aPai7T7uCYDaZF/KSZdhTJI6xEYSxioRgQwAAACAg+Mnc92e6xAXWY09qfsIZAAAAAAc/HPspNtzjSLY+qkyEcgAAAAAmDJy8nXLf9e5PR8a5F+Nvan7CGQAAAAATAdTs5yOndstzrwdHuzTQu11DoEMAIA6zN+PhfcAvJNZVAzIXu8W9czbIYFEiMrEqwkAQB3G7jYAvJWZU+B0zL6QR+rJvOrsTp1HIAMAoA4jjgHwVoaLEbKQwOJ1Y02iQ6uzO3UeE0ABAKjDGCAD4C1XgaxhRLAWTB2kNf8c18jOsT7oVd1FIAMA4BRhGAabuQIo06YDaU7H4qJC1LphuHrGx1R/h+o4piwCAHCKsDJaBsADO4+ccDpGqfuqQyADAOAUkW+1+roLAGqBfSnOZe9RdQhkAACcIgoYIgPggRPZVFGsTgQyAADqqOOZuQ7f5xPIAHggK8+57D2qDoEMAIA66p+Ukw7fFxQQyACUzmo1lJ3H9ObqRCADAKCOOlmidDUjZADKkpNfHMYaRwb7sCenDgIZAAB1VHa+47SjTQedS1kDgD376YrPXNpDkjQxoaWvunNKIJABAFBHncx1DGR/7kuVJKVl5emxr7doMwENQAm2QBYU4Kch7Rtp1f3DNev8Lj7uVd1GIAMAoI6aNm+dw/exUSGSpHs/26C3ft2jq99a5YtuAajBsor+kBMaWLjvWOOoEPn5saF8VSKQAQBQB1ldrBezlb3/dlOSJOlYiSqMAJCd5xjIUPUIZAAA1EHjX1/hdMxqUNQDQOm+2XhIkhQaRCCrLgQyAADqGKvV0B9/H3c6zsbQAEqTlVugV37eJUnaczTTx705dRDIAACoYwrcjIQRyACUZvfRDF934ZREIAMAoI4pGbwu6tVMElMWAZQu9WSer7twSgrwdQcAAEDlKhm8/IsqpD3xzTZFhwb6oksAaoHjJ4sL/TSLCfVhT04tjJABAFDHlBwh87cUl6y+57ONooI1AFeO242QPTe+p+86coohkAEAUMdYrY7fl9xDyJ9EBqCE1JO5+mT1PknS+L7x6t+6vo97dOpgyiIAAHVMfolE5l/iz68hgf7KK8iXJBmGIYuFgAac6s5/6VftS8mSJMWEM7W5OjFCBgBAHVOyyqJ/icAVFVL8YSs9O79a+gSgZrOFMUmKCGLMpjoRyAAAqGO8mbJ4LCOnOroEoBZp1TDc1104pRDIAACoY8oaIdubctK8nZlTUC19AlB7jO7WxNddOKUQyAAAqGOsJassllLEIyefQAac6kr+zig5qo6qRSADAKCOKVn2vrQPV9l5VrfnAJwa8krOc0a1IpABAFDHlJyyWHKjaHuMkAHIzS8OZKGB/j7syamJQAYAQB1TcoTs241JbtsyQgbAPpAtmj7Ehz05NRHIAACoY0oGshPZeW7bMkIGIK+g8HdGgJ9FLRtQYbG6EcgAAKhjSgay0qRk5lZhTwDUBrYRsqAAooEv8KoDAFDHlFwzVlo8e3zhVi3/62jVdghAjZZbUDhSHuhPNPAFXnUAAOqYkiNkM8/rXGr7rzYcqsruAKjhTuYWBjIKevgGgQwAgDqm5AjZ2N7NS22flZtfld0BUMM9u3iHJKlxVLCPe3JqIpABAFDHFHhZOJFNYIFT10er9urn7cmSpAbhQT7uzamJQAYAQB2TzyavADyw52im7v18o/n9iWxGy32BQAYAQB2TX+B5lUVJOplD6XvgVLRqzzGH7y/o2dRHPTm1EcgAAKhjvB0hy2QNGXBKsi9z37tFjK4c0NKHvTl1EcgAAKhjcvOdR8g6xkW6bZ+ZQyADTkV+luL1o5MHtWY9qY8QyAAAqGNcjZB9fEOCLh/Qwvz+rsQO5u1MpiwCp5wtB9N120frze+HtG/ou86c4ghkAADUMXl2ZRanDW0nSYoODdQZ7RuZx28+q62eGttNkpTBCBlwyrlm7h/m7c5NohQTRoVFXwnwdQcAAEDlyisq6tGtWbTuGHmaeXxYx8Ya0r6hesXHyGKxqE/LepJYQwacipLSs83bY3s382FPQCADAKCOsY2QxUWHyGK3RiQowE/vTxlgfh8S6C9JysmjTD5wqunaLEqbDqQrJNBPUwa39nV3TmlMWQQAoI6xlb0P8i/9f/P+RQv4C6zelcnHqeNEdp6+2XhIWbmsM6yrXr2yj8MfblD9CGQAANQxthGyAP/SP2TZzucWWDXrf5sIZpAk7Us5qeV/HZUk3fbRet384VrN/m6bj3uFkqxWQ4ZR/vdsbn7h74ngMv5wg6rHvwAAAHWMbQ1ZYBkftAL8is+/u+IffbvpUJX2C7XDkNk/6fI3f9fiLYf147YjkqSP/9jn417BXlZugc7698+66YO15b5Gji2QBRIHfI01ZAAA1DG2EbKyApl/iT2HjmXkVlmfUPtc995q83ZMaKAPe4KSVu4+pr0pJ7U35aSsVqNc+4fZRsiC/P0ru3vwEpEYAIA6pviDVhlTFkt8iGNPWLjDGqOaxT6AHT9Zvj+kmL8nAogDvsa/AAAAdYxtX7Hw4NInwpQcIeNDN9ytSXK1vtBqNfSfJTvN9WaoPift9g48nJ6jjJx8zfzfJq3+O8XjaxDIag6mLAIAUMfYAllESOn/m3ceISOQneqy3WyBkJSeLcMwHEL7+yv/0TOLd0iSdv5rVJlTZFExGTn5OnoiRx+v3qdXf95lHv9p+xEln8jReyv+0Xsr/tHfT40u81r5BVZl5RVWzgxhDZnPEcgAAKhjMm2BzMsRspx8Spuf6krbJPznHcka2qGx+f2sLzebt7/ffFijuzep0r6dyo6cyFb/fy1xee7p77Z7da2/jmRow/5U5VsNRQYHKDYypDK6iAogEgMAUMeYUxaDSg9kJacoPvLVFh1Ozy71Pkt3JOvbjVRjrKtK22/s993up8OlZOZURXdQZOGGynnPWa2GRjy7VDPm/ylJat0ovFwFQVC5CGQAANQxR4uqJdYL974y3oAnljisIzIMQ99vTtK+lJMqsBqa+PYq3fThWiWllR7cUDuVNkJW2ozWh/632f1JVJitRH1FlSwAEhfF6FhNQCADAKAOMQxDe45mSJJaN4wo1zW2JZ0wb/+w9Yiuf3+Nhsz+SftSTprHk8oYSUPtdLKUEbKyBlLSs/MquTew2bg/zelY48hgvXhZL6fjpW3w/sW6Aw7fx9cPq3jnUGEEMgAA6pD07HyzMEOT6PL99ftgapZ5e9WeY+btW/67zrx99ART1OqikznuA5mbAowm9rGrGkdOZGuhi2nCM8/vrHO7xjkdP1FKMP7rSIbD92N6Nqt4B1FhFPUAAKAOScks/FAcHuSvkMDybfiallX8gc6+JPbGA8V/pU8p595HqNlKm7KYnWdVXoHVrKZosTiGtNKCAMrv7k83OHx/xmmN9PiFXRVfP1QWi0V/PDBCizYd0qNfb1FegaHUk3ny97Pome93qEl0iK4/o425XtQ2sn3fqI66YmDLMgv/oHowQgYAQB1y4Hjh6Fb9iKByXyPdLpAFB7gOdfZtUHdk5rgPZG//tke9Hl2sfSknZRiGuU1CdGjhWsWMbPf3Rfkt31U8St2nZT29M6mfWjQIM0NWo8hgXZXQSo0igiUV/kHlyW+3ae7yv/Xkt9v0zcYkMyzvPVY47bhbs2jCWA1CIAMAoA5Z/U9hJbx2jcq3fkwqnPZo82zRPlNObQhkddLOElPaSsrIydcrP+/S38cKi7wEBfipRdE6pEe/3qIHvtjodnNplM/gdg3N25/ddLrTdhU20WGFf4RJzcrTvN/3mscfXLBR3R7+XnOW7tLuo5mSpBYNWDtWkxDIAACoQ2wjHO1jIz1qf1diB7VpFK5Qu+mN2Xll70eWRiCrk+wLt9hc1r+Fw/c5eQXanVwY3No3jlC98MIgsC3phD78fa/2ppxU8okcTZn7h5b/dbTqO13HtWkYLkka27v09V7RoYUjXqkncxVpN/p1/GThe/Wpb7eZx5pEh1Z2N1EBBDIAAOqQd1f8I0kK8vfsf/FTh7bTj3ecpVUPDNeUwa0leVZiO8uD0Ibax3XQdhzxyimwmmuRmkSHOHz4lwpH0Z76dpuWbDuiy9/8nfBeQXkFhe/H5jGlhyjb1NFdRzJ0opSpp5LzpvDwLQIZAAB1xInsPOUWhSlvA1NkSKC5piQnv/C+pZXPzq2kfZFQs6SedAxP/n4Wp9LoOXkF2pdSuFaxcVSI01qkQ6nZOmlXHKTHI9/rqz8PVlGPK2bZjmRtPZTu626UKrcokAWW8UeWmNDCkcoXf/zL6VxYUPEI+Oxx3Suxd6gMPg1ky5Yt0/nnn6+mTZvKYrFowYIFDucNw9DMmTPVpEkThYaGasSIEdq5c6dDm5SUFF1xxRWKiopSTEyMpkyZoowMx/nPGzZs0JAhQxQSEqL4+HjNnj3bqS+ffPKJOnbsqJCQEHXr1k3ffPNNpT9fAACq0v7jxeXqSwtT7tiqMtrK5pfcRNae7UMi6hbbaNb4vvEK9Lfozav76ppBrR3a5ORbtWF/qiSpe7No/X0s0+H8te+t1o/bjjgcu+W/63T5GyvN0Z6aYMG6A7r67VUa9cIvOpZR9ds4HEzN0pu/7Pa6GqVtxNq+4qkr0WHuN4K/KqGlFkwdpHUPna1L+8V79fioej4NZJmZmerRo4defvlll+dnz56tF198UXPmzNHvv/+u8PBwJSYmKju7eDPKK664Qps3b9bixYv19ddfa9myZbr++uvN8+np6Ro5cqRatmypNWvW6Omnn9bDDz+s119/3WyzfPlyXXbZZZoyZYrWrVunMWPGaMyYMdq0aVPVPXkAACpZUlrx/x+t5SisEFz0gc/2AdBWQt8VRsjqptSiEH79mW206ZFEDe3Y2Gn7hCPpOTpStA9di/phLkdjXU17Xb7rmCa/80cV9Lp8pn+83rw94IklVR4WL31thR5fuFUPf7nFq/vlFRS+l8sKZIH+jtMQL+nT3LzdtlGEesbHmOv9ULP4NJCNGjVKjz/+uC666CKnc4Zh6Pnnn9eDDz6oCy+8UN27d9d7772ngwcPmiNpW7du1aJFi/Tmm29qwIABGjx4sP7zn//oo48+0sGDhUPjH374oXJzc/X222+rS5cumjBhgm699VY9++yz5mO98MILOuecc3TXXXepU6dOeuyxx9S7d2+99NJL1fI6AABQGezXjZRnhCw4sPBjQUpm4YdtVxv92jab9mSdGWqXAqthVtiMCQ10u+XBgdQsM6zXjwjSrPO7ePwYv/511GXhkOr2++5jDt/nWw39urNqC5DYRrB/2n6kjJaOcoumEJcVyA6lZjt8/+TYbvrsptN158jTNLYXG0DXZDV2DdmePXuUlJSkESNGmMeio6M1YMAArVixQpK0YsUKxcTEqG/fvmabESNGyM/PT7///rvZ5owzzlBQUPFfBBITE7V9+3YdP37cbGP/OLY2tsdxJScnR+np6Q5fAAD4kv1UqPKMkOUX/SX+t78KP6y6GiGrV1Rau7wjZLuSMzyq4ojqZ7+Vga1AhCsZOfnFgSw8SH1a1tPfT43WXYkdnNpO6Bevlg3C9PH1A81jry3bVYm9Lp/V/xx3OjZ57h+VEhbn/rZHV731uw6mZrk8bzUM/W/9Ae0/7tlj2d5rZa0hG9u7eETsvlEdFeDvpz4t62nasPYK8LDID3yjxv7rJCUlSZJiY2MdjsfGxprnkpKS1LhxY4fzAQEBql+/vkMbV9ewfwx3bWznXXnyyScVHR1tfsXHMx8XAOBb9hvzlmeErKXd3kT5BVZzpMxeg6INp8uzhmz+6n0a/sxS3fXpBq/vi6qXUjRdMTI4wOMP8LaALhVW7Nz22DmyL+B3Wf8WWnrXUA1o00BtGhWWb68JG0i7C0Ozvtxc4Ws//NUW/bLzqMa+slxv/rJbOw+fcDifejJPt320XtM/Wu/R9Y4Vhd/SQrIkDWrXQF/fMlif3pig64a0KVff4Rs1NpDVdPfdd5/S0tLMr3379vm6SwCAU5x9efGwoIBSWrrWr1V983ZegWF+ELQvkR0VUvihsDwjZLMXbZekGltx71SXXLQurFFksEfto0ICnEZtQgL9HYJDfbs1S9cXhYT0UgJZgdXQkfRst+crQ4HV0K9u9ker6AiZ/ehvUnq2Hl+4VaNe+EV/H810arv6n+M6kp6ts57+Sf9ZstPpvCSdzM3Xhv1pkqTWRfuRuWOxWNS1WbT6tqovP8ra1yo1NpDFxcVJkg4fPuxw/PDhw+a5uLg4HTniOA83Pz9fKSkpDm1cXcP+Mdy1sZ13JTg4WFFRUQ5fAAD40ha78t23DGvn9f3t16jk5lvNgDcxoZV53PZhPSvXu2mHJ3PzzYIRqJlshToalhLI6oW5Dlv2jtuVzm9er3jvrMiiMJ9Zyh5Z932+Qf2fWKJfdiZ71mkvffzHXrW9/xvtS8lScICfPrKbSik5jviVR8ltA6TC9Wn3fu56VPjMp3/W38dO6pnFO8w/cqzcfcwMpY99XVwApFWD0gMZaq8aG8hat26tuLg4LVmyxDyWnp6u33//XQkJCZKkhIQEpaamas2aNWabH3/8UVarVQMGDDDbLFu2THl5xW+QxYsXq0OHDqpXr57Zxv5xbG1sjwMAQG1wOL3wA/Xcyf3UIMKzUQ57AX4WWYr+sJ5bYDULd0SHBurGM9tqQr94nde9iSRp99FM3TH/T4+u+96Kv9V55nfKt5tGaV8REjXDwg2FI5exUSFu21xrNxWuZ3yMyzYd4yIlSaO7N5HFUjxSU7KKpyvzV++XJF311irN+t8mj9dZeeqezzaat0+LjdTANg30zuR+6teq8DNhSgX/aJCa5fr+K3enuDxuX6HycHq2Vuw6pgmvr9SFL/8mSfqlqNBIs5jQMot6oPby6b9sRkaG1q9fr/Xr10sqLOSxfv167d27VxaLRdOnT9fjjz+uL7/8Uhs3btTVV1+tpk2basyYMZKkTp066ZxzztF1112nVatW6bffftO0adM0YcIENW3aVJJ0+eWXKygoSFOmTNHmzZv18ccf64UXXtCMGTPMftx2221atGiRnnnmGW3btk0PP/ywVq9erWnTplX3SwIAQLnZNuMtuVGvpywWizkFLTuvwJx+FRLop3tHddRT47orxm4E4bO1+8sc9dqdnKGZ/3NelzPwySUuWsNX9qWc1HebC2cLNYsJdTr/5bRBenB0J910Zlud36OpBrSur8fGdHV5rTlX9tF9ozrq2Ut7OBy3VfH0tKjLuyv+0ZiXl3vzNEplLbGu8p+i/dOGdmisp4o2Sz5cwemSrkbISnpncj+Xx5/7YYcue2OlJOlQWrbaP/CNWZlx3nUDKtQv1Gzl+41dSVavXq2hQ4ea39tC0sSJEzV37lzdfffdyszM1PXXX6/U1FQNHjxYixYtUkhI8V9uPvzwQ02bNk3Dhw+Xn5+fxo0bpxdffNE8Hx0dre+//15Tp05Vnz591LBhQ82cOdNhr7LTTz9d8+bN04MPPqj7779f7du314IFC9S1q+tfNAAA1ESZOYUfdMPLGcik4rVhk95ZpY5xhdPxg+3+Mh8Z4njtOUt3695RHd1e759jvi9xjrJtSyouPNEwwnnaXvfmMerePEaS9J/LepV6rVYNw3XDmW2djtvK6Je2/jAowM/h/NGMHG05mK7OTSu+NORAiaqH2XaPY/u5zszJ18b9aZq3aq9mnH2ax+vpbMoKZPOuG6AuTaNdnvt87QGH7237j/lZpPh6Ya7ugjrCp4HsrLPOklFKWV6LxaJHH31Ujz76qNs29evX17x580p9nO7du+uXX34ptc0ll1yiSy65pPQOAwBQg9lGyMLLUdCjpF3JmeYaIfuNgZ0D2S79uO2w3ri6r1q6WOOSnOFcqRE1j/2WCWWVVy+vsqYsFlgNl2Ht3Bd/0d9PjS73457IztPRjFyH6Y/x9UP1wLmdze8jgwvXt1kN6fyXfpUk1Q8P1F2J7v/Y4MrXGxwL1rRuGK49dgU94qJCFB0aqO+mn6HQQH/FhAdq+V9HdeMHa91ec3y/eIp01HE+DWQAAKByWK2GThYV2ggLdr2hr7f++Ltwryb7QBYa6HztHYczNO7VFVr94AinczP/t8nt9Q3DcFhjhOqTejJX0aGB5uu/O7k4NAzt0Njd3SrENmUxJ9/1lEX7UPjx9QM1/vWVFX7MjJx8dXv4e0nS5QNaSJLOPK2R3r2mv0O7kEA/+VkKA5nNsh1HdVei54+1du9xfb3hkMOxMT2bafmuo/p9T+EassZF6/M6FK2zk6ShHRurWUyoOYLXp2U9fXpjgrYfPqEG4cFej9Kh9iGQAQBQB9gXB6iMETJ79sUE3AWooy5GwpLSspWd5zji8e1tQzTqhcJZKz9uO6LhnWKd7oeq9dO2I5o89w81CA/ShT2b6aHzOmnnkcIpi5NOb6UWDapmepxtymJOnusRMltVz9BAfw1o08Dh3Ko9Kerfur6ru5Vq0turzNvzft8rSWoa41y0xGKxKCI4wKEkf0igdyOFN76/xulYTFigrkpoaQYyV+s7gwP89f3tZ8hqGGYlSknmlGHUfZRrAQCgDsgsmq5osXj/QbIsfiVCmG2D35JKjnykZBYX/Ojfqr62PnqOWYFPkqa8u9rr8vmouLd+3SOpcMPht3/bo/X7UrXzSIYkaUQVBmTblMWsvAIZhiHDMPTtxkNa809hWLEFMts+Zo/bFQ2561PPKnqWtPqf407HmkQ7Fy2RpE5NHANQvhebqxuGYf5R4rohrc3jsVHBGt2tiWad31lvXN3X7f3DgwMcwhhOLQQyAADqgJO2gh5BAZU+DXBYR8cpbK9d2cdlu8NpjqNktilobRqFa/6NCQoN8nfq26WvrdCmA2mV2FuUxZBj0DiakavDRdsQNKvnOqxUBtuaxHyrodSTeVq+65hu+nCtJry+Utl5BTpRNDoVFVo4inTlwJZ65IIukgqLw7y2dJfXj9nGxWbKTaJdl/V/a5Jj9UNv/liQbzXM6Y7ThrY3j3dpGi2LxaLJg1rr7M6MBsM1AhkAAHVARtFmu2FBlbN+zF7J/Y/ax0ZqYkJLp3br9jmORtimf0WV8pf/jQfSdPkb5V8r9NtfR80RFnjmRLbjxsx7U04qsyh8uNvsuTKEBPqb13/im61asvWIpMJqgs98v91um4Xin+Gr7X7Onl28w+vHtI0c22vlIqRJhdMJ7x3VUU2LAttJLwKZfSn/4EA//TDjDM2/IUHx9amOiLIRyAAAqANW7j4mSQqtgkDmiqtKfMknHEfIth1Kl1Q8Bc3mhjPbOHyfnu38odkT6dl5uuLN3zXu1RXq9vB3yi9wX079VJdXYNX/1h/Q3Z/+qQ37HUckDxTtdRXgZ1FUSNWWF4grKmrxyZr9evu3PebxN37Zo8cXbpXkuM2CxWLRw+cXVkNs0yjC68c7UvQzedvw4lGr9o3dX+fGM9vqzYmFI2XeBDL7ypHBAX5q1ziyXGvecGoikAEAUMvlF1jND7MV3ffrufE9ym6kwrVqJdmPvPy8/YieKRrRaF1iROK+UZ3K30E7KRnFa9ROZOfroz/2Vcp166KXf/pLt320XvNX73c6983GwsqALRqEVXnVyy1FId0VW3n4kiOy/YqCzb6Ukw6VGEuTV2DVpa+tkG13pSsHttQXN5+uL6cNctjc3BXbKPNJF6Nr7tgCWVCAH5VD4TUCGQAAtZhhGHpt2W7z+0mnt6rQ9UqWPG/spuS2fZl0G9u0SUm67r3V5m1XRUDsp1b6WQr3oPKW/eNJ0l9FhSng7Pkfdjodu6BHU0lSUnrh+rHeLepVeT+mDnXeMLqkvHzHn4VOcVFqGh2ijJx8fbspyaPH+c+Pf2nVnuKprPXCAtWrRT1zc+vSNCjaGPtkboHDNUqTUzRlMTiAj9bwHj81AADUYj9uO6Knv9tufn//uRUbfQooMRVx3nUDXLbLdrGX1Fu/7lF+gVW5+VblFRR/qG7nYqrZt7cNUUDRZrdWQzp+MtepTVlKroWyVelD2WKjghVf37GAR1VtCG3v1uHt9cKEnrorsYN5zP62JOWUmHrq52fRoHYNJUlHisJjWRZvOWzefmdSP6ef69LYVzu88xPPqjvaRsj82cAZ5UAgAwCgFlu6I9m83Swm1Gm6l7cC7D5Q9mtVT+0aR7ps527a4e6jmUpKK/7Q/NiFXZTQtoFTu5YNwrV+1kizyEPyiRy9sWy3LpmzXD9uO+zU3pWSI2Sp5Qh1p4I8F2vr6oUF6egJx9drfL/4Ku9LcIC/LuzZTNcMKiwN7+9n0U1nttWnNyaYbfLynfvbIKJwpHb9vlSNe3W5FpUxUmY/UtWnlfcjf0FFAc7T99O3RdM+XW2cDpSFQAYAQC1mX8Tj7RJlu8vDPpAF+Ln/mNC1WbR5+8qBLczb2XkFOplXGJQahAfpqoRWbtfURAQHmFMiv/rzoP71zVb98fdxXTN3tdbvSy2zryVH1XYlZ8pajqmPdZ19BUCbB0d3dhrl7G73b1rVQoP89ccDI/THAyPk52dR31bFBTDyrc6BrH544ajVD1uPaM0/x3XjB2v09YaDWuNinzGpeMTqol7NSq3y6c6zRWspG3hQdTItK08v/viXJMcN2gFPEcgAAKiFcvILZLUaDh823W3Y7A37KVcB/qVPv/rspgRdNbCl7j6no1m4Iyffquy8wg/DIR6MFjSvV1gW/JWfHfeYWryl9BGQfSkndfenGyRJo7rGKSTQT3tTTuq5H7wvjV7X2f49bKYMbq3B7Rtq+ojTzBLvbRqGy6+ap9s1igx2WWY/v8A5VNcPd17LOG3eOl377h86ciJb76/8Rxk5+crOK9CxjByz4ue1dps0e8P2hwl3axut1sKNrSXpHbtqkZXxRxGceqq2tikAAKh025LSdc7zv0iS+YF6dPcmlbIGyH40K6iM6/VpWV99WhaObNimiL2xbLcSu8RJkkICy+7P5EGt9MNW5ymKtlLs7jzy1Rbzdoe4SEWFBOrj1fv0nx//0ohOseoRH+PQfv7qfbr70w36z2W9dH5RMYtTRckRsviizZ9bNwzX8vuG61hGjsKDff+RsF+revrj7+Pq18q5XLxthKyk4yfzNO7V5dqXkqUXftihrs2i9fP24mm8jSNdbwJdFv+i0eECwzmQbdyfpvNf+lXjejfXg6M76YOVe81zvUr83AGeYIQMAIBa5uYP15q3Dxat12oWE+quebmVNUJmL7hoNOz7LYd1R1EhBE/2RCsZnGx7TmWWsQfU5oPFe2ld3r+Fbh1RvM/UgVTHMJeZk2+Opt3+8foy+1TXfFxiO4CSBS4aRAR7NJpZ1R46r7PuSuygWRd0djrXwMUImc2+lMJ/76MZuQ5hzN0InCdKGyE7/6VfJUmfrd2vXo8t1tGMwtG4B0d3ouQ9ysX3fw4BAABeyXYRVkpuvlwZvBlxc1XuOySg7A/5ESVGZmx7RGWVEchsa4S+vW2IGhdtNnzmaY20dEeyMu2KfZzMzVeXWd+Z37ctx+bCtd1LP/1l3u7Xqp7G9Grmw9641715jNuy9F2aRjl8HxUSUOaG4pf1b1Huqoe2+7maPulOVbwHcWpghAwAgFrm+Enn8u4NI8o3ElCasqYs2nO1B5inge7TGxPUqkGY3p7U19yfLLOMTXltFRYjQ4oDnS3c2Qey1X87Fn3wZtSvLvrkxtOdQnBtEODvZ44CR4YEaNndQ7Xl0USndjefVbzPWYGL4iAeP14pI2TuQl7TKhilxqmBQAYAQC1yMjffrORmX2LbVdGDivImvKRkOpec93Sz576t6uvnu4ZqWMdYhQUVhoV1e1OVV2DV/uMnHdomn8jRsYwc5RaNkNmHi/BgW5grHl2zL8EvSakuwmxJ+QVWl6Xiayvbmrnavmnxf68bqAfO7aQ/HhihmLAghQUFaOGtg83zc67srbvP6agh7Qv3LBvbu3m5H8scISsKdUu2HtayHcma9M4qtz/X7RqfeqOvqBy1708kAACcwo5lFAafoAA/NY0J0a7kTElS83pVsYbM8w/wz43vods/dtxE9/ye3hfPiAkrnvbV6aFFyrcauvucDpp0eiut3H1M0+atk5/dOp1wh0BWeNu2YXR2XoHu/qxw7VhQgJ9y861Oe5V9vzlJN3+4VtcOaSOpcM+uz9fulyFp8e1nqlFk5Qfd6mYb7blj5Gk+7knFtGgQpuvOaONwzFalU5Js9TfemthPqSdzzams5WH7Y0SB1VBaVp6mvLu61PaTTm+l2Ao8Hk5tBDIAAGoR20hUw/Ag2f+h3lZ2vjJ5M2Xxol7NNaZnM730419as/e4LurVTBeUo5qh/Vqh/KInuH5vqu46uEELNxxyaBsS6OcwLbJh0ebBfx3J0L2fbdBHdsUsnriom+785E9l5hYoN99qbvh7/ftrJElzljqW3ZekdXuPa2RRxUjDMPT9lsPqFBelFg3CnNrWZNkuRlTrCvt1W2lZhaOfQQF+FQpjUnGVxXyrYZbQt/fMJT1ULzxQt/53vZ4a103ndT+1KneichHIAACoRY5lFn44rB8RpAy7ogZVUSUvwMuCCBaLRbcMb192wzKuUVJGTr6WbznmdLzkiESToi0ASpbRPy02QmN7NdPdn/4pqyEdTs9WfP2wMjefvv79NQoL8lfvFvWUmpWrTQfSJUl//WuUV6OHvmab4loTKilWhQt7NtWyHclmeK4Mtp99q9VQWpbzdNzzejRRcIC/NswaWe37t6HuIZABAFCL2P5a3zAi2CGQVYUgH605ig4NNEc7JGn5LucwJklxJQKZq/2rpMLqi35+FnVtFq0N+9M0ZPZPGt2tiRZuLB5xaxgRpPaNI7Vit+Njncwt0K9/HXU4Nuj/ftSSO86qNcUxbD8ntvV5dc3z43sqr8Co1J9X27TY5IwcXTJnhdP54KIKooQxVIba8+cdAABOYUdOZOtEdp7mrSqchmcbDaoKUwa3VqPIYF0zuHWVPUZpXrysl8vjDSOCdKfdOihb8Qab+PqupxLefFY7SVK9sOJKlPZh7OI+zbX6wbP16pW9FVi0dijI30+X9Y93eb3D6Tn6rURIq8n2FRVGaVYF6wxrAovFUul/PLCtIcsrMFSyhsedtXwtHmqeuvmnEgAA6pD07Dyd/uSP5poqqfgv9FXhofM6+3ST2zNPa6Stj56jrUnpGvvKcvN4VEigLu4Tr39/v0OSdPmAlqVeJ6FNA825qo+5zuiqgS21dEeyzjitkYZ1aKTUrDy1bhhurv+JCQvSxzckKDTQX52aFK5luyuxo+av3qeRnWPVplGELn9jpZbvOmYWDqnpTubm63B64ahqq1q29s2XStu/LM+LvckATxDIAACo4XYePuEQxiTpgp5NdSgtS38fO+nmXhXjqzBmExrkrx7NY3R62wbmlMUGEUGKiw7Re9f0V1RooOqHO++99tiFXfTQ/zbr6oSWmnleZ4e1XiM6x2rlfcMVGxXs9vn1blHP4fv64UG68cziva2iQgrDXVYZ+6TVFBv2p0kqrF4ZE1b5e9XVVaWtn7RtSg5UFgIZAAA12H9X7dV9n290Ot6ucYSeGttdBdY/dWHPZj7oWdXz97No3nUD9eWfB/XCDzv05NjukqQzTmvk9j5XJbTSVQmt3J6Pq+BUT9vG1Sft9jqrqQ6nZ2vC6yslST2ax/i2M7VMaSNkY3vXzfcbfIdABgBADXU0I8dlGJOkyOAAWSwWvTmxXzX3qvpd0KNpuUroV4XQWhTIPlz5j3l7ZJdYH/ak9okJC5KfRQ7rx9Y+dLbyC6wVLqkPlEQgAwCghkpKy3Z5/LELu/h8SuGpyjZC9sKSnercNEqJlVhqvTIUWA1NnvuHLJKaxhQHh3NqWD9ruojgALVpFKG/jmRIkq4Z1NrlFFmgMhDIAACooWwb+tob1K5BqVPyULWaxRRXKnzrlz01LpBtPZSuZTuSJUmNIgs3yn5ybDc1KNo0G56zD2AbD6T6riOo8whkAADUUNl5xcUDXrmit87pEse+Rz52VUIrPfzVFkmFhTJqgteW7tKT327TDWe00WvLdpvHbXvWtW4Y7quu1Wq7kzPN2+G1ZM851E7sQwYAQA2VVTRC1jM+Rud2a0IYqwH8/SyafXFhcZHcAt9X2/tszX49+e02SXIIYzYNI4KcKkfCM0M7FBePuXdURx/2BHUdgQwAgBrKFshCAvnfdU0SUTRakpnj29L3B1OzdMcnfzodj7MrOjFtaLtK3zT5VDG+X/HG4E2i6+am2qgZGH8FAKCaHU7P1pylu3TtkDYOa5JKyi6q5BcaWHWbQMN74WYg822lxf3Hs5yONYkO0dK7hmrF7mOKiwrRabERPuhZ3RAZUjwlNZIpi6hC/HQBAFDNxr+2Qn8fO6m/jmTo/SkD3La7+7MNkkrfEwnVLyK4MCBn+nhz6MPpxVU4F946WF2aRpvfn1nKXm3wzGmxEZqY0FINI4KZLowqxRg2AADVaFdyhv4+dlKS9MvOo3r4y80yDMOpXYbddLhB7RpWW/9QtrCgmjFl8UR24eO3aRjuEMZQOSwWix65sKtuGd7e111BHUcgAwCgin3150Et2XpYkrT67xSHc3OX/63D6TkOxzJy8vX1nwfN7ydS5r5Gsa0hO5qR69NQdrJohK57c8IYUJsRyAAAqELJJ3J0y3/Xacq7q/XWr3t0z2cbndqs/qc4pBmGoYtfXa57Py9sN6xjY6ZL1TD2JdBt/06+YFvDFsb6JqBWI5ABAFCFjp/MNW8/9vUWl22mzVtnjnY8tWibtiWdMM+N7takajsIr4UHFxdZ+erPgxr/2gr9cyyzlHtUDdsatvAgir4AtRmBDACAKrT8r6NOxwa2qe907GBqtqxWQ68tddxL6sKeTausbyif4ADHAPT7nhSd+fTPyivalywnv0Dz/9inz9bsV9rJPElSfoHV5VpBd17+6S89uGCj030Mw1BKZq6y8wq0aFOSpOI1bQBqJ97BAABUoX9/v8Pp2FsT+2nhxkN6+9c95mjYte/+oUv6xju0e/Pqvgrw52+nNdGi6UN0zvO/OBz760iGOjWJ0pu/7NHT322XJI3p2VRPjeuukc8t06G0LD1yQVddPqBFqdfel3LSvP/VCa10Wmyk/jmWqQ370/To11uUfMJxzWHP+JjKe2IAqp3F8ObPNXArPT1d0dHRSktLU1RUlK+7AwCoIS58+Tf9uS/V4djfT402b1/2+kqt2H3M6X7bHjtHIew/VmMdz8xVr8cWOxw7vW0DzbtuoMt/c5teLWL0xc2D3F73RHaeBjyxRCeL9qCLCglQerb7wiF3JXbQ1KHtvH8CAKqUN9mAETIAAKqQf1E9jqcv7q7FWw7r/B6OUxC7x0c7BbLLB7QgjNVw4S4KaWwvGu2sFxbodM5m84F0fbc5SUlp2bo6oaUsluKCLVarodV/HzfDmCSXYeyuxA5aufuYTm/bUDed1bYiTwNADUAgAwCgCiVnFE4va9UwXK9f3dfpfExokNOxCf3inY6hZgkKcJ5KeqxobVeB1f3ko9wCq254f40kadaXm/XERd10Wf94fbHugO7/YqOy86wu79coMliPXNBFg9o2VHRYIKNiQB1CIAMAoIq89ONO7UvJkiTF1wtz2cZaYuXAud3i1L15TFV3DZVg/g0JmjF/vR4+v4uufW+1JKnjQ4vM8x3jIrUt6YTiokL05S2D1P9fS5yucf8XG/XQ/zY5hbgHzu2k685oo0NpWfppW7Iu7duc9YRAHUUgAwCgChiG4VDQo3FkcJn3+etfo+TPnmO1Rv/W9fXrPcMkOa/1ahwZrAVTB2nroXT1aB4jPz+LFt9+hs5+bpnqhQUqO8+qrLzCqYn2YWxCv3idcVojjeoaJ0lqEh1aZhEQALUbgQwAgCrw+57izZ6vP6ON282dL+0br9eX7dbIzrGMgNRiz0/oqWvmrja/b9kgTCGB/urVop55rH1spENBl6S0bN384Rqt3ZsqSerSNEqPj+nKzwFwiiGQAQBQhr3HTqp+RJAiXBRycOfHbUckSSM7x+r+czu5bdcoMlirHxyhAEbGarVhHWO17bFzNPK5ZdqbclId48quuBwXHaLPiyouFlgN+VnkUOQDwKmBQAYAgBsnc/P15Dfb9P7Kf3R251i94aIohyuF1fIKR8gGt29YZvtARkTqhJBAf/1v6iCt35+qhDYNvLovU1WBUxeBDAAAN27+cK1+3p4sSVq85bDeX/mPFm44qNeu7KtoF6XNZ8xfrx2HT6hDbJQ5DS02KqQ6uwwfqxcepKEdGvu6GwBqEf4kBwCAG7YwZvPQgk1auTtFPR79XkfSs/XUt9u0//hJSdIvO5P1+doD2nQgXZ+t3W/ep3XD8GrtMwCgdmGEDAAAOy/8sFOLNifp1St6l9qu/xOFJcznLN2l6SPaKzjAeSPn0d2a6LTYyCrpJwCgbrAYhuF+90J4LD09XdHR0UpLS1NUVNkLeQEANVOrexc6fN+8XqiGtG+k/67aW+r9mkaH6GBatvn9r/cMVXM3e48BAOo2b7IBUxYBACiSlVvgdOzeUR11dUJLBQX4aUj7hlr70Nnq16qeUztbGDurQyN9e9sQwhgAwCNMWQQAoMjGA2lOx87r3lSS9Ns9wxQZEqCQQH99cuPp2nn4hOqFB+mqt1Zp66F0s/3tI05TpybMlAAAeIYRMgAAiny6Zp/D953tglWjyGCFBBavE2sfG6mGEcH6/KbTHe7TIz6mSvsIAKhbGCEDAEBSSmauFqw/KEn69MYExYQFqUl02SXrQ4P81a1ZtDYeSNNpsRFV3U0AQB1DIAMAnNLyC6yaPPcP/bLzqCSpW7No9WlZTxaL5xv1vjO5n55bvEOX9o2vqm4CAOooAhkA4JT2wpKdZhiTpEmnt/IqjElSw4hg/euibpXdNQDAKYBABgA4JRmGoQcXbNKHvzuWsz+vRxMf9QgAcCoikAEATjnZeQWa/tF6LdqcJElq2yhcd4zsoDNOa+Ryg2cAAKoKgQwAUOtYrYZ2H83Ql+sP6u3f/lZQgJ8Su8RJkm46s61aNCjeAywnv0CLNiUpPStPsli0ZOthLf/rmHILrJKkiQkt9fAFXbyepggAQGUgkMHBn/tSdfOHa5Wenad7R3XUFQNa+rpLAOq47LwC3f/5RvVpVc/t75wCq6H1+1LVqUmkCqyGps5bp2U7kosb5Ej/XVU49XDd3uN6alx3dWkapZn/26wv1x9QposNnyXpnC5xhDEAgE9ZDMMwfN2JuiA9PV3R0dFKS0tTVFTt3RD0old+07q9qeb3sy/u/v/t3Xt0VNXdN/Dvucz9kgu5EwhQG8BbgAgYL1gUhOeRIt5eClrAohaBClrKU6mvYNdCliLWqrS+a1FJrdZKrBe0CygmQKsGH41BCIRwC2AkCQmQe+Z2zn7/GJhlTLiVmTkz5vtZa/7ImTP77P3jJMx39tlncNeIbChy1zcrm/fUI6DpGH95OlSFX2dHRBdP0wU+PdiIX/99F75p6gQAHHr6vyHLEtq9AbR5A7CbFSxbvwd//7LmrO24rCpaPYHzHs9pUeGwKPhhmgsjByTjvmv7o4/TErbxEBERnXEx2YCBLEy+D4HsZLsPI5d/BE3vfkq8M/c6jOifBADYWdOEyS9/Enou2WFGhtsKu1mBT9Nx7aA+cFlUTB/NNztE1J3Hr6GorAb/972KHp8f0T8RFd+0hC4p7IkkAfN+dBl+WpCDdLcVjW1eBDSBuW+U4ctvfagEAL+aMBizrhsAh4UXhRARUXQwkBkg3gPZlr3HcX/h5wCADLcVdS2eLs+7rSoe/tFleGbj3otqd/3863FlVgLk0zNsQgicaPch2W4ObQsnb0CDSZYj0jZRLPl7WQ2WfbAbui6weOIQ3DI0DUdPdsCkyBiS4UKnX0Ntkwdumwl9nGaosgS7OfyB5HirB9uqGlB6MLgmKyvRBrdVxZ7aFpxq98NmDt4gY2dNE8yKjOZOf7fLBwemOFDd2H7WY6Q4LbhzRF/0T7bDrMi4PMuNK7LcPV5m2OYNYPvBE1i5qQoCAq8/MBpprvN/uTMREVE4MZAZIJ4D2ZdHT+HOP3wa+vnl6cPxcskB7K1rPetr7GYFk/Oy4NcEjrd6oMgSOrwaDjS0YeSAJGzaXd/tNf2T7bCZFFTVB9tNdVlgMym4sq8b+TnJGJzuwg0/TOnyGiHEedd2HD3RgQ92HsN75d9g//E2AEC624I+Dgty+tiRmWBDVX0LPq8+BYdFQUaCDWZVhgRAloAEmwkDUhywqAosqgy/puPq7ATcenkGgx3FDCEEjjV78MXhk/jHzlr8c0/337FzkSQgK8GGO4b3hYCA02KC1STD49fR6Qugw6eh3aehvsWDNk8AbpsJgIDFpMCsyOj0afAENLR0+lFzqhO6AAK6jqYO/380HodZwcBUB5b811AU/KAPqupbMff1L3GosR256U7M/dFlGDs4DZoQSLKbuMaLiIjiCgOZAeI5kC0q+gpvlwXXZ8wsyMFTt1+Jr0924NlNVZg2sh+mr/msy/6/+e+hmHFdzjlvDf3OlzV4bN1XF92Xq7MTkO62Ii87AW+X1eDwiQ4AwC1D0jAmNxVWkwxdAO3eAMyqjM+qg29OIykrwQpFkZDptsFlVeG2meCwKBiY4sSQDBeSHWbkpru6rLMLaHqPa+v8mo7Ne+pRMKgPkhzmiPa7t9tZ04Q0lxUZCT3Pjggh8FHlcfx7fwMSbSbIsoRkhxl2swqzKqOpwwevX8fgDBfyc5JQ3+JBdpIdZrXrv2urx49DDe34pqkTZkVGQNeR4rTAagr+fphVGbIkocMXgKaLYJDRdAR0Ab+m41SHD/UtXjS2emEzBz8UaPMGZ5BkCbCZFFhMMjZU1HVZ33khbCYFAV2HX4vsn/nMBCssqozL0pwwKTKEAC5LcyLJEZyZ04XA4AwX7GYVCadn7FwWtceQ5QvoMCkSAxgREcU1BrJLsHr1aqxcuRJ1dXXIy8vDSy+9hFGjRp33dfEYyCq+acYDf/6iy+WJFU9NgPM76yy2VB3HL9d9hUfH5+Kn1174XRf9mo7VWw5gY0VdaLYtM8GKm4ekYXj/JKS7LTApMjZW1OGvnx0953qRCzHp6kzcPqwvhmS48HZZDRravHCYFbR0BiDLEganO5HXLxFNHf7Tb4yDb45rmztR3+LFiTZv8I14p/+iQ16CzQRdD84mqLKEuhZPaG1dZkIwFNhMCkqqjuNQQ/DSrCS7CSZFhsUkwxfQocoy+ifbkXF6f7fVBEUGHBYVdrOCQSlOuG0mDOhjhyRJ0HSBpg7fedfpdfo01DZ34mS7D2ZVRobbijR35C/hEkJA0wVaPQHsqW2BSZGhyIAQwdkas6LArMpIcpiQZA++cb+UN+Eev4bGNi8ON3ag9FAjVm85CCAYiMYNTUOG2wanRUHFsRZ8cqARJkVGm/f8N4L4rjSXBboQ8GsCvoCOTn/Pd++LlCEZLlydnYAhGW78tCAHO2uasf3QCfz46iy4bSqsJgXt3gBcVhPMqgwhBLwBHV993YTVWw/CbVWR4rSgzRtAp1+DVVXgsCiwmRXYTSpSXRZIEqALAVmS4PFr8Gs6bGYVNpMCq0lGitOCBJsJ3oCOvok2pDjNDFBERETfwkD2H3rrrbcwY8YMvPLKKxg9ejReeOEFFBUVoaqqCmlpaed8bSwHsoCmo67Fg9KDJ3Cy3QdFllDd2I6ishr4AsEQdP1lffD/fnpNtzAWTtWN7aEw8V1CCPx7fyMW/K0cP0x34X+rT4b6NX5oOvbWteJQQztsZgXegAaHWYUiS+jjNGPUwGRMGdY3rG8IDza0Yc2/D+HyTDfy+iWiodWLY80eyBLQ5gmgxePHlr0NqKpv7fEmKJH2o8GpqPimBY1tXvwg1QGX1QSbSUFTpx+1zZ1wWVV4/Tp8Ws+XlCmyFHpzbVGDMzBWVUHq6bDhCwRf6wvoECK4Nk+RJZhVGYokQRMCug5YTDLsZgW6Hly709jmhS4EVFnG8VbPRc3MKLKEgSkO2M0K7KfXHXX4NHT4NKiyFLohg6YH++fXdGi6gF/X0eoJ/MeXzgHBMO/xazjV4YfHr8FlVUPrrbbtazjvv3GK04JEuykUKqsb2+CymqBIEnxasK9OS/CclSUJqiJBlSWocjCQprusUBUJ3oAOiyrDYVEhS8GZJY9fQ7tXQ2aiFbOvHxiVME1ERESXhoHsPzR69GiMHDkSL7/8MgBA13X069cPv/jFL/DrX//6nK+NpUD22Lod2Ly7HpCCsxEev4bAOd5QPn3HVZg+un8Ue3h+p9p9UBUJLqvJ6K6c05nZh93HmvHhzlo0tHox/vJ05OckodUTQF2zB7XNHtQ1d6Ldp6Fvog3XDEiCBAk1pzqCQcocDEXegIY9x1rQ4dNQc6oTHT4Nmq6jqdOPytoWnOrwhwL0xTKrMpLsJnT4tAu6PXi4JdhMcNuCIeNM0DgTGE+2+8J2HFWW0L+PHT9IdWJwuguaEDhyoh1pLisUWUKbJ4DMRCucFhWDM4K3Pj9zaeHZNHX4UNvsQXaSDW3e4L+pJEmQJSDRZkay0xzRDzKIiIgo/lxMNuC7iNN8Ph/Kysrw+OOPh7bJsoxx48ahtLS02/5erxderzf0c0tLS1T6eSE0XaD1O5diSRJwRZYbSXYz3DYTshNtGJGThPFD02PyxhXxsr5KkiRYTQryc5KRn5Pc7fmhmWf/BbwqO6Hbtp7a+Lb99a14ZdshJNlNGJrpxg/SnGho9cIb0KDpIjhL1uHHgBQHnBYVJkVCH6ely00RGtu88Gs6vH4dnoAGr1+HN6Cj3RvA8VYPTIoMsyrDrMgwnV7/ZFZOX/qm6dB1ETpnfAEdnT4NkhT8LqhkhwWqLMGvBddROU5/75NZkc86g9nhC8AfEKhv9aCqrhWHG9uRfvoST4sqw20zwa8F+wcg2J/T/VMVGaoiwWlRkeYKXkYX7kvnEu1mJNqD56PLakJmgi2s7RMREVHvxkB2WmNjIzRNQ3p6epft6enp2Lu3+63eV6xYgaeeeipa3bso/zNxCB64YRBkGbCbVVhNMtJd1pgMXnRxfpjuwqr/k3dJbaTE2HfD2c0qYAYS7CbkpruM7g4RERFRVHW/DRxdkMcffxzNzc2hx9dff210l0KyEm24KjsBV2QlYGCKA5kJNoYxIiIiIqIYxBmy01JSUqAoCurru363T319PTIyMrrtb7FYYLHE1kwDERERERHFF86QnWY2m5Gfn4/i4uLQNl3XUVxcjIKCAgN7RkRERERE31ecIfuWxx57DDNnzsQ111yDUaNG4YUXXkB7ezvuv/9+o7tGRERERETfQwxk3zJ16lQ0NDTgySefRF1dHYYNG4aNGzd2u9EHERERERFROPB7yMIklr6HjIiIiIiIjHMx2YBryIiIiIiIiAzCQEZERERERGQQBjIiIiIiIiKDMJAREREREREZhIGMiIiIiIjIIAxkREREREREBmEgIyIiIiIiMggDGRERERERkUEYyIiIiIiIiAzCQEZERERERGQQBjIiIiIiIiKDMJAREREREREZhIGMiIiIiIjIIAxkREREREREBmEgIyIiIiIiMggDGRERERERkUFUozvwfSGEAAC0tLQY3BMiIiIiIjLSmUxwJiOcCwNZmLS2tgIA+vXrZ3BPiIiIiIgoFrS2tiIhIeGc+0jiQmIbnZeu6zh27BhcLhckSTK6OxHV0tKCfv364euvv4bb7Ta6O997rHd0sd7Rx5pHF+sdXax39LHm0cV690wIgdbWVmRlZUGWz71KjDNkYSLLMrKzs43uRlS53W7+4kUR6x1drHf0sebRxXpHF+sdfax5dLHe3Z1vZuwM3tSDiIiIiIjIIAxkREREREREBmEgo4tmsViwdOlSWCwWo7vSK7De0cV6Rx9rHl2sd3Sx3tHHmkcX633peFMPIiIiIiIig3CGjIiIiIiIyCAMZERERERERAZhICMiIiIiIjIIAxkREREREZFBGMh6oRUrVmDkyJFwuVxIS0vDlClTUFVV1WUfj8eDefPmoU+fPnA6nbjrrrtQX1/fZZ9HHnkE+fn5sFgsGDZs2DmPeeDAAbhcLiQmJoZ5NLEvWvU+fPgwJEnq9ti+fXskhxeTonmOCyHw3HPPITc3FxaLBX379sXy5csjNbSYFK16L1u2rMdz3OFwRHJ4MSea5/emTZtw7bXXwuVyITU1FXfddRcOHz4coZHFrmjWfN26dRg2bBjsdjtycnKwcuXKSA0rZoWj3l999RWmTZuGfv36wWazYejQofj973/f7Vhbt27FiBEjYLFYcNlll6GwsDDSw4tJ0ap5bW0tpk+fjtzcXMiyjIULF0ZjeDGPgawX2rZtG+bNm4ft27dj8+bN8Pv9uPXWW9He3h7a59FHH8UHH3yAoqIibNu2DceOHcOdd97Zra2f/exnmDp16jmP5/f7MW3aNNx4441hH0s8iHa9P/roI9TW1oYe+fn5YR9TrItmzRcsWIA1a9bgueeew969e7F+/XqMGjUqIuOKVdGq96JFi7qc27W1tbj88stxzz33RGxssSha9a6ursbtt9+Om2++GTt27MCmTZvQ2NjYYzvfd9Gq+YYNG3Dvvfdizpw5qKiowB/+8Af87ne/w8svvxyxscWicNS7rKwMaWlpeP3117F792785je/weOPP96lltXV1bjtttswduxY7NixAwsXLsQDDzyATZs2RXW8sSBaNfd6vUhNTcUTTzyBvLy8qI4xpgnq9Y4fPy4AiG3btgkhhGhqahImk0kUFRWF9qmsrBQARGlpabfXL126VOTl5Z21/cWLF4v77rtPrF27ViQkJIS7+3EnUvWurq4WAER5eXmkuh63IlXzPXv2CFVVxd69eyPW93gU6b8pZ+zYsUMAEP/617/C1vd4FKl6FxUVCVVVhaZpoW3r168XkiQJn88X/oHEkUjVfNq0aeLuu+/usu3FF18U2dnZQtf18A4ijlxqvc+YO3euGDt2bOjnxYsXiyuuuKLLPlOnThUTJkwI8wjiT6Rq/m033XSTWLBgQVj7Ha84Q0Zobm4GACQnJwMIfsLh9/sxbty40D5DhgxB//79UVpaelFtl5SUoKioCKtXrw5fh+NcJOsNAJMnT0ZaWhpuuOEGrF+/PjydjnORqvkHH3yAQYMG4cMPP8TAgQMxYMAAPPDAAzh58mR4BxBnIn2On7FmzRrk5ub22tn3MyJV7/z8fMiyjLVr10LTNDQ3N+Mvf/kLxo0bB5PJFN5BxJlI1dzr9cJqtXbZZrPZUFNTgyNHjoSh5/EpXPVubm4OtQEApaWlXdoAgAkTJlzS36Xvi0jVnHrGQNbL6bqOhQsX4vrrr8eVV14JAKirq4PZbO623is9PR11dXUX3PaJEycwa9YsFBYWwu12h7PbcSuS9XY6nVi1ahWKiorwj3/8AzfccAOmTJnS60NZJGt+6NAhHDlyBEVFRXjttddQWFiIsrIy3H333eEcQlyJZL2/zePx4I033sDs2bMvtctxLZL1HjhwIP75z39iyZIlsFgsSExMRE1NDdatWxfOIcSdSNZ8woQJeOedd1BcXAxd17Fv3z6sWrUKQHDtTW8Urnp/+umneOutt/DQQw+FttXV1SE9Pb1bGy0tLejs7AzvQOJIJGtOPVON7gAZa968eaioqMDHH38c9rYffPBBTJ8+HWPGjAl72/EqkvVOSUnBY489Fvp55MiROHbsGFauXInJkyeH/XjxIpI113UdXq8Xr732GnJzcwEAf/rTn5Cfn4+qqioMHjw47MeMdZGs97e9++67aG1txcyZMyN6nFgXyXrX1dXhwQcfxMyZMzFt2jS0trbiySefxN13343NmzdDkqSwHzMeRPr/zYMHD2LSpEnw+/1wu91YsGABli1bBlnunZ+hh6PeFRUVuP3227F06VLceuutYezd9xNrHn2987ebAADz58/Hhx9+iC1btiA7Ozu0PSMjAz6fD01NTV32r6+vR0ZGxgW3X1JSgueeew6qqkJVVcyePRvNzc1QVRWvvvpquIYRNyJd756MHj0aBw4cuKQ24lmka56ZmQlVVUNhDACGDh0KADh69OildT4ORfMcX7NmDSZNmtTt0+3eJNL1Xr16NRISEvDss89i+PDhGDNmDF5//XUUFxfjs88+C9cw4kqkay5JEp555hm0tbXhyJEjqKurC90kaNCgQWEZQzwJR7337NmDW265BQ899BCeeOKJLs9lZGR0uxNmfX093G43bDZbeAcTJyJdc+oZA1kvJITA/Pnz8e6776KkpAQDBw7s8nx+fj5MJhOKi4tD26qqqnD06FEUFBRc8HFKS0uxY8eO0OO3v/0tXC4XduzYgTvuuCNs44l10ap3T3bs2IHMzMxLaiMeRavm119/PQKBAA4ePBjatm/fPgBATk7OJY4ifkT7HK+ursaWLVt67eWK0ap3R0dHt1kZRVEABGeHe5Non+OKoqBv374wm8148803UVBQgNTU1EseR7wIV713796NsWPHYubMmT1+HUlBQUGXNgBg8+bNl/x/bzyKVs3pLIy7nwgZ5eGHHxYJCQli69atora2NvTo6OgI7TNnzhzRv39/UVJSIr744gtRUFAgCgoKurSzf/9+UV5eLn7+85+L3NxcUV5eLsrLy4XX6+3xuL31LovRqndhYaH461//KiorK0VlZaVYvny5kGVZvPrqq1EdbyyIVs01TRMjRowQY8aMEV9++aX44osvxOjRo8X48eOjOl6jRftvyhNPPCGysrJEIBCIyvhiTbTqXVxcLCRJEk899ZTYt2+fKCsrExMmTBA5OTldjtUbRKvmDQ0N4o9//KOorKwU5eXl4pFHHhFWq1V89tlnUR2v0cJR7127donU1FRx3333dWnj+PHjoX0OHTok7Ha7+NWvfiUqKyvF6tWrhaIoYuPGjVEdbyyIVs2FEKHzPj8/X0yfPl2Ul5eL3bt3R22ssYiBrBcC0ONj7dq1oX06OzvF3LlzRVJSkrDb7eKOO+4QtbW1Xdq56aabemynurq6x+P21kAWrXoXFhaKoUOHCrvdLtxutxg1alSX29P2JtE8x7/55htx5513CqfTKdLT08WsWbPEiRMnojTS2BDNemuaJrKzs8WSJUuiNLrYE816v/nmm2L48OHC4XCI1NRUMXnyZFFZWRmlkcaOaNW8oaFBXHvttcLhcAi73S5uueUWsX379iiONDaEo95Lly7tsY2cnJwux9qyZYsYNmyYMJvNYtCgQV2O0ZtEs+YXsk9vIwkhxAVNpREREREREVFYcQ0ZERERERGRQRjIiIiIiIiIDMJARkREREREZBAGMiIiIiIiIoMwkBERERERERmEgYyIiIiIiMggDGREREREREQGYSAjIiIiIiIyCAMZERERERGRQRjIiIiIvmPWrFmQJAmSJMFkMiE9PR3jx4/Hq6++Cl3XL7idwsJCJCYmRq6jREQU9xjIiIiIejBx4kTU1tbi8OHD2LBhA8aOHYsFCxZg0qRJCAQCRnePiIi+JxjIiIiIemCxWJCRkYG+fftixIgRWLJkCd5//31s2LABhYWFAIDnn38eV111FRwOB/r164e5c+eira0NALB161bcf//9aG5uDs22LVu2DADg9XqxaNEi9O3bFw6HA6NHj8bWrVuNGSgRERmKgYyIiOgC3XzzzcjLy8M777wDAJBlGS+++CJ2796NP//5zygpKcHixYsBANdddx1eeOEFuN1u1NbWora2FosWLQIAzJ8/H6Wlpfjb3/6GnTt34p577sHEiROxf/9+w8ZGRETGkIQQwuhOEBERxZJZs2ahqakJ7733XrfnfvKTn2Dnzp3Ys2dPt+fefvttzJkzB42NjQCCa8gWLlyIpqam0D5Hjx7FoEGDcPToUWRlZYW2jxs3DqNGjcLTTz8d9vEQEVHsUo3uABERUTwRQkCSJADARx99hBUrVmDv3r1oaWlBIBCAx+NBR0cH7HZ7j6/ftWsXNE1Dbm5ul+1erxd9+vSJeP+JiCi2MJARERFdhMrKSgwcOBCHDx/GpEmT8PDDD2P58uVITk7Gxx9/jNmzZ8Pn8501kLW1tUFRFJSVlUFRlC7POZ3OaAyBiIhiCAMZERHRBSopKcGuXbvw6KOPoqysDLquY9WqVZDl4JLsdevWddnfbDZD07Qu24YPHw5N03D8+HHceOONUes7ERHFJgYyIiKiHni9XtTV1UHTNNTX12Pjxo1YsWIFJk2ahBkzZqCiogJ+vx8vvfQSfvzjH+OTTz7BK6+80qWNAQMGoK2tDcXFxcjLy4Pdbkdubi7uvfdezJgxA6tWrcLw4cPR0NCA4uJiXH311bjtttsMGjERERmBd1kkIiLqwcaNG5GZmYkBAwZg4sSJ2LJlC1588UW8//77UBQFeXl5eP755/HMM8/gyiuvxBtvvIEVK1Z0aeO6667DnDlzMHXqVKSmpuLZZ58FAKxduxYzZszAL3/5SwwePBhTpkzB559/jv79+xsxVCIiMhDvskhERERERGQQzpAREREREREZhIGMiIiIiIjIIAxkREREREREBmEgIyIiIiIiMggDGRERERERkUEYyIiIiIiIiAzCQEZERERERGQQBjIiIiIiIiKDMJAREREREREZhIGMiIiIiIjIIAxkREREREREBvn/TewS8i0vGlwAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "G12TARVyK_gN"
      },
      "source": [
        " Just like the pandas plot. And because we formatted the `timesteps` to be `datetime` objects, `matplotlib` displays a fantastic looking date axis."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "prkQhgiHNOKd"
      },
      "source": [
        "### Create train & test sets for time series (the wrong way)\n",
        "\n",
        "Okay, we've figured out we're dealing with a univariate time series, so we only have to make a split on one variable (for multivariate time series, you will have to split multiple variables).\n",
        "\n",
        "How about we first see the *wrong way* for splitting time series data?\n",
        "\n",
        "Let's turn our DataFrame index and column into NumPy arrays."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wxQbuGVTRFKX",
        "outputId": "d220623a-2be7-4206-990b-af55ddd5ec5c"
      },
      "source": [
        "# Get bitcoin date array\n",
        "timesteps = bitcoin_prices.index.to_numpy()\n",
        "prices = bitcoin_prices[\"Price\"].to_numpy()\n",
        "\n",
        "timesteps[:10], prices[:10]"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array(['2013-10-01T00:00:00.000000000', '2013-10-02T00:00:00.000000000',\n",
              "        '2013-10-03T00:00:00.000000000', '2013-10-04T00:00:00.000000000',\n",
              "        '2013-10-05T00:00:00.000000000', '2013-10-06T00:00:00.000000000',\n",
              "        '2013-10-07T00:00:00.000000000', '2013-10-08T00:00:00.000000000',\n",
              "        '2013-10-09T00:00:00.000000000', '2013-10-10T00:00:00.000000000'],\n",
              "       dtype='datetime64[ns]'),\n",
              " array([123.65499, 125.455  , 108.58483, 118.67466, 121.33866, 120.65533,\n",
              "        121.795  , 123.033  , 124.049  , 125.96116]))"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "F0jPVUNztInE"
      },
      "source": [
        "And now we'll use the ever faithful `train_test_split` from Scikit-Learn to create our train and test sets."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eGY7hTfyNLZ1",
        "outputId": "688052c8-a300-4719-f751-d9f773617d7b"
      },
      "source": [
        "# Wrong way to make train/test sets for time series\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(timesteps, # dates\n",
        "                                                    prices, # prices\n",
        "                                                    test_size=0.2,\n",
        "                                                    random_state=42)\n",
        "X_train.shape, X_test.shape, y_train.shape, y_test.shape"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((2229,), (558,), (2229,), (558,))"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PJbMDz1stTxZ"
      },
      "source": [
        "Looks like the splits worked well, but let's not trust numbers on a page, let's visualize, visualize, visualize!"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 619
        },
        "id": "cvH_0RvSNLWm",
        "outputId": "2a26629a-d6b6-4531-b0a3-8d4150e9f2d6"
      },
      "source": [
        "# Let's plot wrong train and test splits\n",
        "plt.figure(figsize=(10, 7))\n",
        "plt.scatter(X_train, y_train, s=5, label=\"Train data\")\n",
        "plt.scatter(X_test, y_test, s=5, label=\"Test data\")\n",
        "plt.xlabel(\"Date\")\n",
        "plt.ylabel(\"BTC Price\")\n",
        "plt.legend(fontsize=14)\n",
        "plt.show();"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OK9sORtjRh5z"
      },
      "source": [
        "Our test data is scattered all throughout the training data.\n",
        "\n",
        "This kind of random split is okay for datasets without a time component (such as images or passages of text for classification problems) but for time series, we've got to take the time factor into account.\n",
        "\n",
        "To fix this, we've got to split our data in a way that reflects what we're actually trying to do.\n",
        "\n",
        "We need to split our historical Bitcoin data to have a dataset that reflects the past (train set) and a dataset that reflects the future (test set).\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ajoASBK2size"
      },
      "source": [
        "### Create train & test sets for time series (the right way)\n",
        "\n",
        "Of course, there's no way we can actually access data from the future.\n",
        "\n",
        "But we can engineer our test set to be in the future with respect to the training set.\n",
        "\n",
        "To do this, we can create an abitrary point in time to split our data.\n",
        "\n",
        "Everything before the point in time can be considered the training set and everything after the point in time can be considered the test set.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lpEVKyXANLT4",
        "outputId": "e04fca51-9bfc-479b-bdd9-09e29a0545b3"
      },
      "source": [
        "# Create train and test splits the right way for time series data\n",
        "split_size = int(0.8 * len(prices)) # 80% train, 20% test\n",
        "\n",
        "# Create train data splits (everything before the split)\n",
        "X_train, y_train = timesteps[:split_size], prices[:split_size]\n",
        "\n",
        "# Create test data splits (everything after the split)\n",
        "X_test, y_test = timesteps[split_size:], prices[split_size:]\n",
        "\n",
        "len(X_train), len(X_test), len(y_train), len(y_test)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2229, 558, 2229, 558)"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ByN_YCotv3zk"
      },
      "source": [
        "Okay, looks like our custom made splits are the same lengths as the splits we made with `train_test_split`.\n",
        "\n",
        "But again, these are numbers on a page.\n",
        "\n",
        "And you know how the saying goes, trust one eye more than two ears.\n",
        "\n",
        "Let's visualize."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 619
        },
        "id": "L9QImNlhOQuS",
        "outputId": "ab1ac5f6-81a0-40f1-dc5c-4f1820f93c08"
      },
      "source": [
        "# Plot correctly made splits\n",
        "plt.figure(figsize=(10, 7))\n",
        "plt.scatter(X_train, y_train, s=5, label=\"Train data\")\n",
        "plt.scatter(X_test, y_test, s=5, label=\"Test data\")\n",
        "plt.xlabel(\"Date\")\n",
        "plt.ylabel(\"BTC Price\")\n",
        "plt.legend(fontsize=14)\n",
        "plt.show();"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sGUelYVKCybB"
      },
      "source": [
        "### Create a plotting function\n",
        "\n",
        "Rather than retyping `matplotlib` commands to continuously plot data, let's make a plotting function we can reuse later."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "trkvizA0C6vc"
      },
      "source": [
        "# Create a function to plot time series data\n",
        "def plot_time_series(timesteps, values, format='.', start=0, end=None, label=None):\n",
        "\n",
        "  # Plot the series\n",
        "  plt.plot(timesteps[start:end], values[start:end], format, label=label)\n",
        "  plt.xlabel(\"Time\")\n",
        "  plt.ylabel(\"BTC Price\")\n",
        "  if label:\n",
        "    plt.legend(fontsize=14) # make label bigger\n",
        "  plt.grid(True)"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 619
        },
        "id": "6TyFY1cGC_sw",
        "outputId": "a78588b6-925b-4206-8145-5cbed39b4343"
      },
      "source": [
        "# Try out our plotting function\n",
        "plt.figure(figsize=(10, 7))\n",
        "plot_time_series(timesteps=X_train, values=y_train, label=\"Train data\")\n",
        "plot_time_series(timesteps=X_test, values=y_test, label=\"Test data\")"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SDeel1jj8ioU"
      },
      "source": [
        "Looking good!\n",
        "\n",
        "Time for some modelling experiments."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cwSpQQRl_dHb"
      },
      "source": [
        "## Modelling Experiments\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "* **horizon** = number of timesteps to predict into future\n",
        "* **window** = number of timesteps from past used to predict **horizon**\n",
        "\n",
        "For example, if we wanted to predict the price of Bitcoin for tomorrow (1 day in the future) using the previous week's worth of Bitcoin prices (7 days in the past), the horizon would be 1 and the window would be 7.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JO1YAv7w_dE7"
      },
      "source": [
        "## Model 0: Naïve forecast (baseline)\n",
        "\n",
        "As usual, let's start with a baseline.\n",
        "\n",
        "One of the most common baseline models for time series forecasting, the naïve model (also called the [naïve forecast](https://otexts.com/fpp3/simple-methods.html#na%C3%AFve-method)), requires no training at all.\n",
        "\n",
        "That's because all the naïve model does is use the previous timestep value to predict the next timestep value.\n",
        "\n",
        "The formula looks like this:\n",
        "\n",
        "$$\\hat{y}_{t} = y_{t-1}$$\n",
        "\n",
        "\n",
        "\n",
        "In an open system (like a stock market or crypto market), you'll often find beating the naïve forecast with *any* kind of model is quite hard.\n",
        "\n",
        "An **open system** is a system where inputs and outputs can freely flow, such as a market (stock or crypto). Where as, a **closed system** the inputs and outputs are contained within the system (like a poker game with your buddies, you know the buy in and you know how much the winner can get). Time series forecasting in **open systems** is generally quite poor."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nCKGUO4OVos9",
        "outputId": "3bbb6c69-2f71-4d9f-af8f-849f28843f5a"
      },
      "source": [
        "# Create a naïve forecast\n",
        "naive_forecast = y_test[:-1] # Naïve forecast equals every value excluding the last value\n",
        "naive_forecast[:10], naive_forecast[-10:] # View frist 10 and last 10"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([9226.48582088, 8794.35864452, 8798.04205463, 9081.18687849,\n",
              "        8711.53433917, 8760.89271814, 8749.52059102, 8656.97092235,\n",
              "        8500.64355816, 8469.2608989 ]),\n",
              " array([57107.12067189, 58788.20967893, 58102.19142623, 55715.54665129,\n",
              "        56573.5554719 , 52147.82118698, 49764.1320816 , 50032.69313676,\n",
              "        47885.62525472, 45604.61575361]))"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 619
        },
        "id": "a0NLHXDhWDX-",
        "outputId": "c6130ab4-c90c-4486-b477-db6176d81430"
      },
      "source": [
        "# Plot naive forecast\n",
        "plt.figure(figsize=(10, 7))\n",
        "plot_time_series(timesteps=X_train, values=y_train, label=\"Train data\")\n",
        "plot_time_series(timesteps=X_test, values=y_test, label=\"Test data\")\n",
        "plot_time_series(timesteps=X_test[1:], values=naive_forecast, format=\"-\", label=\"Naive forecast\");"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8Zj2p8NAkRLD"
      },
      "source": [
        "The naive forecast looks like it's following the data well.\n",
        "\n",
        "Let's zoom in to take a better look.\n",
        "\n",
        "We can do so by creating an offset value and passing it to the `start` parameter of our `plot_time_series()` function."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 619
        },
        "id": "dx0wUhtOW_VX",
        "outputId": "9420b5c9-4c19-44d2-e52d-2342dbc83429"
      },
      "source": [
        "plt.figure(figsize=(10, 7))\n",
        "offset = 300 # offset the values by 300 timesteps\n",
        "plot_time_series(timesteps=X_test, values=y_test, start=offset, label=\"Test data\")\n",
        "plot_time_series(timesteps=X_test[1:], values=naive_forecast, format=\"-\", start=offset, label=\"Naive forecast\");"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AdPsP9SwD72K"
      },
      "source": [
        "When we zoom in we see the naïve forecast comes slightly after the test data. This makes sense because the naive forecast uses the previous timestep value to predict the next timestep value.\n",
        "\n",
        "Forecast made. Time to evaluate it."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UdK4KoFnA90I"
      },
      "source": [
        "## Evaluating a time series model\n",
        "\n",
        "\n",
        "\n",
        "### Scale-dependent errors\n",
        "\n",
        "These are metrics which can be used to compare time series values and forecasts that are on the same scale.\n",
        "\n",
        "For example, Bitcoin historical prices in USD veresus Bitcoin forecast values in USD.\n",
        "\n",
        "| Metric | Details | Code |\n",
        "| ----- | ----- | ----- |\n",
        "| **MAE** (mean absolute error) | Easy to interpret (a forecast is X amount different from actual amount). Forecast methods which minimises the MAE will lead to forecasts of the median. | [`tf.keras.metrics.mean_absolute_error()`](https://www.tensorflow.org/api_docs/python/tf/keras/metrics/MeanAbsoluteError) |\n",
        "| **RMSE** (root mean square error) | Forecasts which minimise the RMSE lead to forecasts of the mean. | `tf.sqrt(`[`tf.keras.metrics.mean_square_error()`](https://www.tensorflow.org/api_docs/python/tf/keras/metrics/RootMeanSquaredError)`)`|\n",
        "\n",
        "### Percentage errors\n",
        "\n",
        "Percentage errors do not have units, this means they can be used to compare forecasts across different datasets.\n",
        "\n",
        "| **Metric** | **Details** | **Code** |\n",
        "| ----- | ----- | ----- |\n",
        "| **MAPE** (mean absolute percentage error) | Most commonly used percentage error. May explode (not work) if `y=0`. | [`tf.keras.metrics.mean_absolute_percentage_error()`](https://www.tensorflow.org/api_docs/python/tf/keras/losses/MAPE) |\n",
        "| **sMAPE** (symmetric mean absolute percentage error) | Recommended not to be used by [Forecasting: Principles and Practice](https://otexts.com/fpp3/accuracy.html#percentage-errors), though it is used in forecasting competitions. | Custom implementation |\n",
        "\n",
        "### Scaled errors\n",
        "\n",
        "Scaled errors are an alternative to percentage errors when comparing forecast performance across different time series.\n",
        "\n",
        "| **Metric** | **Details** | **Code** |\n",
        "| ----- | ----- | ----- |\n",
        "| **MASE** (mean absolute scaled error). | MASE equals one for the naive forecast (or very close to one). A forecast which performs better than the naïve should get <1 MASE. | See sktime's [`mase_loss()`](https://github.com/alan-turing-institute/sktime/blob/ee7a06843a44f4aaec7582d847e36073a9ab0566/sktime/performance_metrics/forecasting/_functions.py#L16) |\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ljaO69huYCZD"
      },
      "source": [
        "# Let's get TensorFlow!\n",
        "import tensorflow as tf"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1ys3PAVUTo6q"
      },
      "source": [
        "And since TensorFlow doesn't have a ready made version of MASE (mean aboslute scaled error), how about we create our own?\n",
        "\n",
        "We'll take inspiration from [sktime](https://github.com/alan-turing-institute/sktime)'s (Scikit-Learn for time series) [`MeanAbsoluteScaledError`](https://www.sktime.org/en/stable/api_reference/auto_generated/sktime.performance_metrics.forecasting.MeanAbsoluteScaledError.html#sktime.performance_metrics.forecasting.MeanAbsoluteScaledError) class which calculates the MASE."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2jbLBJ_IGjla"
      },
      "source": [
        "# MASE implemented courtesy of sktime - https://github.com/alan-turing-institute/sktime/blob/ee7a06843a44f4aaec7582d847e36073a9ab0566/sktime/performance_metrics/forecasting/_functions.py#L16\n",
        "def mean_absolute_scaled_error(y_true, y_pred):\n",
        "  \"\"\"\n",
        "  Implement MASE (assuming no seasonality of data).\n",
        "  \"\"\"\n",
        "  mae = tf.reduce_mean(tf.abs(y_true - y_pred))\n",
        "\n",
        "  # Find MAE of naive forecast (no seasonality)\n",
        "  mae_naive_no_season = tf.reduce_mean(tf.abs(y_true[1:] - y_true[:-1])) # our seasonality is 1 day (hence the shifting of 1 day)\n",
        "\n",
        "  return mae / mae_naive_no_season"
      ],
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yD4P7-wwUVWj"
      },
      "source": [
        "\n",
        "\n",
        "In practice, if we've created the function correctly, the naïve model should achieve an MASE of 1 (or very close to 1). Any model worse than the naïve forecast will achieve an MASE of >1 and any model better than the naïve forecast will achieve an MASE of <1.\n",
        "\n",
        "Let's put each of our different evaluation metrics together into a function."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MwMeVgRXWaJp"
      },
      "source": [
        "def evaluate_preds(y_true, y_pred):\n",
        "  # Make sure float32 (for metric calculations)\n",
        "  y_true = tf.cast(y_true, dtype=tf.float32)\n",
        "  y_pred = tf.cast(y_pred, dtype=tf.float32)\n",
        "\n",
        "  # Calculate various metrics\n",
        "  mae = tf.keras.metrics.mean_absolute_error(y_true, y_pred)\n",
        "  mse = tf.keras.metrics.mean_squared_error(y_true, y_pred) # puts and emphasis on outliers (all errors get squared)\n",
        "  rmse = tf.sqrt(mse)\n",
        "  mape = tf.keras.metrics.mean_absolute_percentage_error(y_true, y_pred)\n",
        "  mase = mean_absolute_scaled_error(y_true, y_pred)\n",
        "\n",
        "  return {\"mae\": mae.numpy(),\n",
        "          \"mse\": mse.numpy(),\n",
        "          \"rmse\": rmse.numpy(),\n",
        "          \"mape\": mape.numpy(),\n",
        "          \"mase\": mase.numpy()}"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VAKTXEIPVKfR"
      },
      "source": [
        "Looking good! How about we test our function on the naive forecast?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HbJlsvUwYEsf",
        "outputId": "b42d9dd3-00f6-451e-929d-e9b5ee9c519e"
      },
      "source": [
        "naive_results = evaluate_preds(y_true=y_test[1:],\n",
        "                               y_pred=naive_forecast)\n",
        "naive_results"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mae': 567.9802,\n",
              " 'mse': 1147547.0,\n",
              " 'rmse': 1071.2362,\n",
              " 'mape': 2.516525,\n",
              " 'mase': 0.99957}"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0bXUPHswVSjp"
      },
      "source": [
        "Alright, looks like we've got some baselines to beat.\n",
        "\n",
        "Taking a look at the naïve forecast's MAE, it seems on average each forecast is ~$567 different than the actual Bitcoin price.\n",
        "\n",
        "How does this compare to the average price of Bitcoin in the test dataset?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Yjl-X9QLVhp1",
        "outputId": "ed3523a7-d4c7-4ef7-ceb3-2515bbc96cda"
      },
      "source": [
        "# Find average price of Bitcoin in test dataset\n",
        "tf.reduce_mean(y_test).numpy()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "20056.632963737222"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RpEk5iOZWD1o"
      },
      "source": [
        "Okay, looking at these two values is starting to give us an idea of how our model is performing:\n",
        "* The average price of Bitcoin in the test dataset is: $20,056 (note: average may not be the best measure here, since the highest price is over 3x this value and the lowest price is over 4x lower)\n",
        "\n",
        "* Each prediction in naive forecast is on average off by: $567\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e8AzwUvjBkgY"
      },
      "source": [
        "## Format Data Part 2: Windowing dataset\n",
        "\n",
        "Surely we'd be ready to start building models by now?\n",
        "\n",
        "We're so close! Only one more step (really two) to go.\n",
        "\n",
        "We've got to window our time series.\n",
        "\n",
        "Why do we window?\n",
        "\n",
        "Windowing is a method to turn a time series dataset into **supervised learning problem**.\n",
        "\n",
        "In other words, we want to use windows of the past to predict the future.\n",
        "\n",
        "For example for a univariate time series, windowing for one week (`window=7`) to predict the next single value (`horizon=1`) might look like:\n",
        "\n",
        "```\n",
        "Window for one week (univariate time series)\n",
        "\n",
        "[0, 1, 2, 3, 4, 5, 6] -> [7]\n",
        "[1, 2, 3, 4, 5, 6, 7] -> [8]\n",
        "[2, 3, 4, 5, 6, 7, 8] -> [9]\n",
        "```\n",
        "\n",
        "Or for the price of Bitcoin, it'd look like:\n",
        "\n",
        "```\n",
        "Window for one week with the target of predicting the next day (Bitcoin prices)\n",
        "\n",
        "[123.654, 125.455, 108.584, 118.674, 121.338, 120.655, 121.795] -> [123.033]\n",
        "[125.455, 108.584, 118.674, 121.338, 120.655, 121.795, 123.033] -> [124.049]\n",
        "[108.584, 118.674, 121.338, 120.655, 121.795, 123.033, 124.049] -> [125.961]\n",
        "```\n",
        "\n",
        "\n",
        "Let's build some functions which take in a univariate time series and turn it into windows and horizons of specified sizes.\n",
        "\n",
        "We'll start with the default horizon size of 1 and a window size of 7 (these aren't necessarily the best values to use, I've just picked them)."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LibzWQXM2ib-"
      },
      "source": [
        "HORIZON = 1 # predict 1 step at a time\n",
        "WINDOW_SIZE = 7 # use a week worth of timesteps to predict the horizon"
      ],
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1AQ0HWWzfRW_"
      },
      "source": [
        "Now we'll write a function to take in an array and turn it into a window and horizon."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zV8pv5K_x81E"
      },
      "source": [
        "# Create function to label windowed data\n",
        "def get_labelled_windows(x, horizon=1):\n",
        "  \"\"\"\n",
        "  Creates labels for windowed dataset.\n",
        "\n",
        "  E.g. if horizon=1 (default)\n",
        "  Input: [1, 2, 3, 4, 5, 6] -> Output: ([1, 2, 3, 4, 5], [6])\n",
        "  \"\"\"\n",
        "  return x[:, :-horizon], x[:, -horizon:]"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ejJ9vCyKff8F",
        "outputId": "6b08993a-8e18-4a17-a943-d34e8edafd31"
      },
      "source": [
        "# Test out the window labelling function\n",
        "test_window, test_label = get_labelled_windows(tf.expand_dims(tf.range(8)+1, axis=0), horizon=HORIZON)\n",
        "print(f\"Window: {tf.squeeze(test_window).numpy()} -> Label: {tf.squeeze(test_label).numpy()}\")"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Window: [1 2 3 4 5 6 7] -> Label: 8\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qc4U4ExKfx6S"
      },
      "source": [
        "Now we need a way to make windows for an entire time series.\n",
        "\n",
        "We\n",
        "To speed things up, we'll leverage [NumPy's array indexing](https://numpy.org/doc/stable/reference/arrays.indexing.html).\n",
        "\n",
        "Let's write a function which:\n",
        "1. Creates a window step of specific window size, for example: `[[0, 1, 2, 3, 4, 5, 6, 7]]`\n",
        "2. Uses NumPy indexing to create a 2D of multiple window steps, for example:\n",
        "=\n",
        "3. Uses the 2D array of multuple window steps to index on a target series\n",
        "4. Uses the `get_labelled_windows()` function we created above to turn the window steps into windows with a specified horizon\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "apea0wiJyCI6"
      },
      "source": [
        "# Create function to view NumPy arrays as windows\n",
        "def make_windows(x, window_size=7, horizon=1):\n",
        "  \"\"\"\n",
        "  Turns a 1D array into a 2D array of sequential windows of window_size.\n",
        "  \"\"\"\n",
        "  # 1. Create a window of specific window_size (add the horizon on the end for later labelling)\n",
        "  window_step = np.expand_dims(np.arange(window_size+horizon), axis=0)\n",
        "  # print(f\"Window step:\\n {window_step}\")\n",
        "\n",
        "  # 2. Create a 2D array of multiple window steps (minus 1 to account for 0 indexing)\n",
        "  window_indexes = window_step + np.expand_dims(np.arange(len(x)-(window_size+horizon-1)), axis=0).T # create 2D array of windows of size window_size\n",
        "  # print(f\"Window indexes:\\n {window_indexes[:3], window_indexes[-3:], window_indexes.shape}\")\n",
        "\n",
        "  # 3. Index on the target array (time series) with 2D array of multiple window steps\n",
        "  windowed_array = x[window_indexes]\n",
        "\n",
        "  # 4. Get the labelled windows\n",
        "  windows, labels = get_labelled_windows(windowed_array, horizon=horizon)\n",
        "\n",
        "  return windows, labels"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u7Yc3V7fc1UG",
        "outputId": "a2e8445f-6387-416c-d808-03a7e91ddef1"
      },
      "source": [
        "full_windows, full_labels = make_windows(prices, window_size=WINDOW_SIZE, horizon=HORIZON)\n",
        "len(full_windows), len(full_labels)"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2780, 2780)"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xhj521fWlJsy"
      },
      "source": [
        "Of course we have to visualize, visualize, visualize!"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "U1P-MT4CkYta",
        "outputId": "723c2414-d7ed-462e-8ad7-160476c65188"
      },
      "source": [
        "# View the first 3 windows/labels\n",
        "for i in range(3):\n",
        "  print(f\"Window: {full_windows[i]} -> Label: {full_labels[i]}\")"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Window: [123.65499 125.455   108.58483 118.67466 121.33866 120.65533 121.795  ] -> Label: [123.033]\n",
            "Window: [125.455   108.58483 118.67466 121.33866 120.65533 121.795   123.033  ] -> Label: [124.049]\n",
            "Window: [108.58483 118.67466 121.33866 120.65533 121.795   123.033   124.049  ] -> Label: [125.96116]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UCDtm6gskjRH",
        "outputId": "8940a286-21b9-429d-f35b-bfa6ddda7bf3"
      },
      "source": [
        "# View the last 3 windows/labels\n",
        "for i in range(3):\n",
        "  print(f\"Window: {full_windows[i-3]} -> Label: {full_labels[i-3]}\")"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Window: [58788.20967893 58102.19142623 55715.54665129 56573.5554719\n",
            " 52147.82118698 49764.1320816  50032.69313676] -> Label: [47885.62525472]\n",
            "Window: [58102.19142623 55715.54665129 56573.5554719  52147.82118698\n",
            " 49764.1320816  50032.69313676 47885.62525472] -> Label: [45604.61575361]\n",
            "Window: [55715.54665129 56573.5554719  52147.82118698 49764.1320816\n",
            " 50032.69313676 47885.62525472 45604.61575361] -> Label: [43144.47129086]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o_8BRgb8YtP3"
      },
      "source": [
        "# Make the train/test splits\n",
        "def make_train_test_splits(windows, labels, test_split=0.2):\n",
        "  \"\"\"\n",
        "  Splits matching pairs of windows and labels into train and test splits.\n",
        "  \"\"\"\n",
        "  split_size = int(len(windows) * (1-test_split)) # this will default to 80% train/20% test\n",
        "  train_windows = windows[:split_size]\n",
        "  train_labels = labels[:split_size]\n",
        "  test_windows = windows[split_size:]\n",
        "  test_labels = labels[split_size:]\n",
        "  return train_windows, test_windows, train_labels, test_labels"
      ],
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z_xY8T7HEaoU",
        "outputId": "bebad61a-3d8d-4f46-de4f-53cefefeff3a"
      },
      "source": [
        "train_windows, test_windows, train_labels, test_labels = make_train_test_splits(full_windows, full_labels)\n",
        "len(train_windows), len(test_windows), len(train_labels), len(test_labels)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2224, 556, 2224, 556)"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "75_YBwiedI2s"
      },
      "source": [
        "Notice the default split of 80% training data and 20% testing data (this split can be adjusted if needed).\n",
        "\n",
        "How do the first 5 samples of the training windows and labels looks?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DDgyT7Gix1fH",
        "outputId": "037f79fa-1ef6-4e63-830e-b11f2bd195d5"
      },
      "source": [
        "train_windows[:5], train_labels[:5]"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([[123.65499, 125.455  , 108.58483, 118.67466, 121.33866, 120.65533,\n",
              "         121.795  ],\n",
              "        [125.455  , 108.58483, 118.67466, 121.33866, 120.65533, 121.795  ,\n",
              "         123.033  ],\n",
              "        [108.58483, 118.67466, 121.33866, 120.65533, 121.795  , 123.033  ,\n",
              "         124.049  ],\n",
              "        [118.67466, 121.33866, 120.65533, 121.795  , 123.033  , 124.049  ,\n",
              "         125.96116],\n",
              "        [121.33866, 120.65533, 121.795  , 123.033  , 124.049  , 125.96116,\n",
              "         125.27966]]),\n",
              " array([[123.033  ],\n",
              "        [124.049  ],\n",
              "        [125.96116],\n",
              "        [125.27966],\n",
              "        [125.9275 ]]))"
            ]
          },
          "metadata": {},
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TkF0HmvadHBB",
        "outputId": "f8ce9232-6d1d-4d5f-b631-da5c9410c890"
      },
      "source": [
        "# Check to see if same (accounting for horizon and window size)\n",
        "np.array_equal(np.squeeze(train_labels[:-HORIZON-1]), y_train[WINDOW_SIZE:])"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PQX0PF5jV4aq"
      },
      "source": [
        "## Make a modelling checkpoint\n",
        "\n",
        "We're so close to building models. So so so close.\n",
        "\n",
        "Because our model's performance will fluctuate from experiment to experiment, we'll want to make sure we're comparing apples to apples.\n",
        "\n",
        "\n",
        "To take of this, we'll implement a [`ModelCheckpoint`](https://www.tensorflow.org/api_docs/python/tf/keras/callbacks/ModelCheckpoint) callback.\n",
        "\n",
        "The `ModelCheckpoint` callback will monitor our model's performance during training and save the best model to file by setting `save_best_only=True`.\n",
        "\n",
        "That way when evaluating our model we could restore its best performing configuration from file.\n",
        "\n",
        "\n",
        "Because we're going to be running multiple experiments, it makes sense to keep track of them by saving models to file under different names.\n",
        "\n",
        "To do this, we'll write a small function to create a `ModelCheckpoint` callback which saves a model to specified filename."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xh8i1f7rV_P2"
      },
      "source": [
        "import os\n",
        "\n",
        "# Create a function to implement a ModelCheckpoint callback with a specific filename\n",
        "def create_model_checkpoint(model_name, save_path=\"model_experiments\"):\n",
        "  return tf.keras.callbacks.ModelCheckpoint(filepath=os.path.join(save_path, model_name), # create filepath to save model\n",
        "                                            verbose=0, # only output a limited amount of text\n",
        "                                            save_best_only=True) # save only the best model to file"
      ],
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "j5uw1AaVY-J9"
      },
      "source": [
        "## Model 1: Dense model (window = 7, horizon = 1)\n",
        "\n",
        "Finally!\n",
        "\n",
        "Time to build one of our models.\n",
        "\n",
        "\n",
        "\n",
        "And once you've got a good model in place, you'll probably notice far more improvements from manipulating the data (e.g. collecting more, improving the quality) than manipulating the model.\n",
        "\n",
        "We're going to start by keeping it simple, `model_1` will have:\n",
        "* A single dense layer with 128 hidden units and ReLU (rectified linear unit) activation\n",
        "* An output layer with linear activation (or no activation)\n",
        "* Adam optimizer and MAE loss function\n",
        "* Batch size of 128\n",
        "* 100 epochs\n",
        "\n",
        "\n",
        "\n",
        "Let's import TensorFlow and build our first deep learning model for time series."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Slr5hrJHY_Ed",
        "outputId": "390927d7-166c-4197-ad84-26caea179b5d"
      },
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras import layers\n",
        "\n",
        "# Set random seed for as reproducible results as possible\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "# Construct model\n",
        "model_1 = tf.keras.Sequential([\n",
        "  layers.Dense(128, activation=\"relu\"),\n",
        "  layers.Dense(HORIZON, activation=\"linear\") # linear activation is the same as having no activation\n",
        "], name=\"model_1_dense\") # give the model a name so we can save it\n",
        "\n",
        "# Compile model\n",
        "model_1.compile(loss=\"mae\",\n",
        "                optimizer=tf.keras.optimizers.Adam(),\n",
        "                metrics=[\"mae\"]) # we don't necessarily need this when the loss function is already MAE\n",
        "\n",
        "# Fit model\n",
        "model_1.fit(x=train_windows, # train windows of 7 timesteps of Bitcoin prices\n",
        "            y=train_labels, # horizon value of 1 (using the previous 7 timesteps to predict next day)\n",
        "            epochs=100,\n",
        "            verbose=1,\n",
        "            batch_size=128,\n",
        "            validation_data=(test_windows, test_labels),\n",
        "            callbacks=[create_model_checkpoint(model_name=model_1.name)]) # create ModelCheckpoint callback to save best model"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "18/18 [==============================] - 2s 41ms/step - loss: 944.9261 - mae: 944.9261 - val_loss: 2482.0864 - val_mae: 2482.0862\n",
            "Epoch 2/100\n",
            "18/18 [==============================] - 0s 27ms/step - loss: 266.1275 - mae: 266.1275 - val_loss: 986.7337 - val_mae: 986.7337\n",
            "Epoch 3/100\n",
            "18/18 [==============================] - 0s 24ms/step - loss: 183.0647 - mae: 183.0647 - val_loss: 940.2244 - val_mae: 940.2244\n",
            "Epoch 4/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 168.9452 - mae: 168.9452 - val_loss: 900.8352 - val_mae: 900.8352\n",
            "Epoch 5/100\n",
            "18/18 [==============================] - 0s 25ms/step - loss: 162.4606 - mae: 162.4606 - val_loss: 866.3202 - val_mae: 866.3202\n",
            "Epoch 6/100\n",
            "18/18 [==============================] - 1s 49ms/step - loss: 155.6816 - mae: 155.6816 - val_loss: 815.1461 - val_mae: 815.1461\n",
            "Epoch 7/100\n",
            "18/18 [==============================] - 1s 35ms/step - loss: 145.5272 - mae: 145.5272 - val_loss: 784.7266 - val_mae: 784.7266\n",
            "Epoch 8/100\n",
            "18/18 [==============================] - 1s 37ms/step - loss: 139.7503 - mae: 139.7503 - val_loss: 768.2231 - val_mae: 768.2231\n",
            "Epoch 9/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 138.3754 - mae: 138.3754 - val_loss: 739.6499 - val_mae: 739.6498\n",
            "Epoch 10/100\n",
            "18/18 [==============================] - 0s 25ms/step - loss: 135.3170 - mae: 135.3170 - val_loss: 719.6373 - val_mae: 719.6373\n",
            "Epoch 11/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 129.4232 - mae: 129.4232 - val_loss: 716.2942 - val_mae: 716.2942\n",
            "Epoch 12/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 126.1494 - mae: 126.1494 - val_loss: 809.3173 - val_mae: 809.3173\n",
            "Epoch 13/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 132.4140 - mae: 132.4140 - val_loss: 783.6660 - val_mae: 783.6660\n",
            "Epoch 14/100\n",
            "18/18 [==============================] - 0s 25ms/step - loss: 125.9141 - mae: 125.9141 - val_loss: 690.6820 - val_mae: 690.6820\n",
            "Epoch 15/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 123.8690 - mae: 123.8690 - val_loss: 656.3855 - val_mae: 656.3855\n",
            "Epoch 16/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 118.0909 - mae: 118.0909 - val_loss: 672.4993 - val_mae: 672.4993\n",
            "Epoch 17/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 117.2150 - mae: 117.2150 - val_loss: 665.2539 - val_mae: 665.2539\n",
            "Epoch 18/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 116.8341 - mae: 116.8341 - val_loss: 626.4887 - val_mae: 626.4887\n",
            "Epoch 19/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 117.0107 - mae: 117.0107 - val_loss: 673.0685 - val_mae: 673.0685\n",
            "Epoch 20/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 115.7666 - mae: 115.7666 - val_loss: 614.7807 - val_mae: 614.7807\n",
            "Epoch 21/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 114.8556 - mae: 114.8556 - val_loss: 616.6033 - val_mae: 616.6033\n",
            "Epoch 22/100\n",
            "18/18 [==============================] - 0s 25ms/step - loss: 115.6321 - mae: 115.6321 - val_loss: 601.3327 - val_mae: 601.3327\n",
            "Epoch 23/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 114.0743 - mae: 114.0743 - val_loss: 599.8010 - val_mae: 599.8010\n",
            "Epoch 24/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 114.0730 - mae: 114.0730 - val_loss: 601.3453 - val_mae: 601.3453\n",
            "Epoch 25/100\n",
            "18/18 [==============================] - 0s 25ms/step - loss: 112.5613 - mae: 112.5613 - val_loss: 597.8423 - val_mae: 597.8423\n",
            "Epoch 26/100\n",
            "18/18 [==============================] - 0s 25ms/step - loss: 113.1534 - mae: 113.1534 - val_loss: 592.0217 - val_mae: 592.0217\n",
            "Epoch 27/100\n",
            "18/18 [==============================] - 0s 25ms/step - loss: 112.7461 - mae: 112.7461 - val_loss: 589.2709 - val_mae: 589.2709\n",
            "Epoch 28/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 111.1342 - mae: 111.1342 - val_loss: 592.3582 - val_mae: 592.3582\n",
            "Epoch 29/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 110.5647 - mae: 110.5647 - val_loss: 624.1107 - val_mae: 624.1107\n",
            "Epoch 30/100\n",
            "18/18 [==============================] - 0s 24ms/step - loss: 114.2837 - mae: 114.2837 - val_loss: 583.5655 - val_mae: 583.5655\n",
            "Epoch 31/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 113.2720 - mae: 113.2720 - val_loss: 637.2823 - val_mae: 637.2823\n",
            "Epoch 32/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 112.4792 - mae: 112.4792 - val_loss: 583.8621 - val_mae: 583.8621\n",
            "Epoch 33/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 110.6602 - mae: 110.6602 - val_loss: 620.5379 - val_mae: 620.5379\n",
            "Epoch 34/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 111.1385 - mae: 111.1385 - val_loss: 588.8915 - val_mae: 588.8915\n",
            "Epoch 35/100\n",
            "18/18 [==============================] - 0s 24ms/step - loss: 109.7951 - mae: 109.7951 - val_loss: 581.8305 - val_mae: 581.8305\n",
            "Epoch 36/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 118.1861 - mae: 118.1861 - val_loss: 584.6062 - val_mae: 584.6062\n",
            "Epoch 37/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 115.6446 - mae: 115.6446 - val_loss: 602.6964 - val_mae: 602.6964\n",
            "Epoch 38/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 110.5838 - mae: 110.5838 - val_loss: 581.8412 - val_mae: 581.8412\n",
            "Epoch 39/100\n",
            "18/18 [==============================] - 0s 25ms/step - loss: 109.3492 - mae: 109.3492 - val_loss: 575.5474 - val_mae: 575.5474\n",
            "Epoch 40/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 108.6510 - mae: 108.6510 - val_loss: 583.9026 - val_mae: 583.9026\n",
            "Epoch 41/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 108.7181 - mae: 108.7181 - val_loss: 592.8460 - val_mae: 592.8460\n",
            "Epoch 42/100\n",
            "18/18 [==============================] - 0s 6ms/step - loss: 112.1748 - mae: 112.1748 - val_loss: 598.0644 - val_mae: 598.0644\n",
            "Epoch 43/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 110.3657 - mae: 110.3657 - val_loss: 620.9410 - val_mae: 620.9410\n",
            "Epoch 44/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 111.1109 - mae: 111.1109 - val_loss: 581.8159 - val_mae: 581.8159\n",
            "Epoch 45/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 108.4587 - mae: 108.4587 - val_loss: 583.3624 - val_mae: 583.3624\n",
            "Epoch 46/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 109.4934 - mae: 109.4934 - val_loss: 601.7881 - val_mae: 601.7881\n",
            "Epoch 47/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 112.2881 - mae: 112.2881 - val_loss: 581.2003 - val_mae: 581.2003\n",
            "Epoch 48/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 110.3572 - mae: 110.3572 - val_loss: 617.7376 - val_mae: 617.7376\n",
            "Epoch 49/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 110.3848 - mae: 110.3848 - val_loss: 596.1797 - val_mae: 596.1797\n",
            "Epoch 50/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 112.4259 - mae: 112.4259 - val_loss: 635.3444 - val_mae: 635.3444\n",
            "Epoch 51/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 112.2443 - mae: 112.2443 - val_loss: 605.4037 - val_mae: 605.4037\n",
            "Epoch 52/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 113.4817 - mae: 113.4817 - val_loss: 587.6744 - val_mae: 587.6744\n",
            "Epoch 53/100\n",
            "18/18 [==============================] - 0s 25ms/step - loss: 114.6399 - mae: 114.6399 - val_loss: 570.0104 - val_mae: 570.0104\n",
            "Epoch 54/100\n",
            "18/18 [==============================] - 0s 27ms/step - loss: 111.0351 - mae: 111.0351 - val_loss: 569.3605 - val_mae: 569.3605\n",
            "Epoch 55/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 113.8187 - mae: 113.8187 - val_loss: 578.2409 - val_mae: 578.2409\n",
            "Epoch 56/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 112.5481 - mae: 112.5481 - val_loss: 792.0372 - val_mae: 792.0372\n",
            "Epoch 57/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 133.4563 - mae: 133.4563 - val_loss: 678.2915 - val_mae: 678.2915\n",
            "Epoch 58/100\n",
            "18/18 [==============================] - 0s 8ms/step - loss: 118.0069 - mae: 118.0069 - val_loss: 683.3015 - val_mae: 683.3015\n",
            "Epoch 59/100\n",
            "18/18 [==============================] - 0s 8ms/step - loss: 111.9316 - mae: 111.9316 - val_loss: 622.1920 - val_mae: 622.1920\n",
            "Epoch 60/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 114.3446 - mae: 114.3446 - val_loss: 616.9111 - val_mae: 616.9111\n",
            "Epoch 61/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 108.7959 - mae: 108.7959 - val_loss: 590.0445 - val_mae: 590.0445\n",
            "Epoch 62/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 109.2598 - mae: 109.2598 - val_loss: 575.7350 - val_mae: 575.7350\n",
            "Epoch 63/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 108.7557 - mae: 108.7557 - val_loss: 611.9833 - val_mae: 611.9833\n",
            "Epoch 64/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 111.1533 - mae: 111.1533 - val_loss: 617.3491 - val_mae: 617.3491\n",
            "Epoch 65/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 110.8755 - mae: 110.8755 - val_loss: 582.5793 - val_mae: 582.5793\n",
            "Epoch 66/100\n",
            "18/18 [==============================] - 1s 36ms/step - loss: 110.4177 - mae: 110.4177 - val_loss: 566.9335 - val_mae: 566.9335\n",
            "Epoch 67/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 109.5653 - mae: 109.5653 - val_loss: 599.6086 - val_mae: 599.6086\n",
            "Epoch 68/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 115.5478 - mae: 115.5478 - val_loss: 692.3923 - val_mae: 692.3923\n",
            "Epoch 69/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 118.9365 - mae: 118.9365 - val_loss: 570.5753 - val_mae: 570.5753\n",
            "Epoch 70/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 125.2682 - mae: 125.2682 - val_loss: 572.3358 - val_mae: 572.3358\n",
            "Epoch 71/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 111.8857 - mae: 111.8857 - val_loss: 587.0344 - val_mae: 587.0344\n",
            "Epoch 72/100\n",
            "18/18 [==============================] - 0s 6ms/step - loss: 108.6970 - mae: 108.6970 - val_loss: 572.1441 - val_mae: 572.1441\n",
            "Epoch 73/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 107.9422 - mae: 107.9422 - val_loss: 575.5754 - val_mae: 575.5754\n",
            "Epoch 74/100\n",
            "18/18 [==============================] - 0s 6ms/step - loss: 107.8742 - mae: 107.8742 - val_loss: 579.0046 - val_mae: 579.0046\n",
            "Epoch 75/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 107.1083 - mae: 107.1083 - val_loss: 577.9851 - val_mae: 577.9851\n",
            "Epoch 76/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 108.7488 - mae: 108.7488 - val_loss: 566.8684 - val_mae: 566.8684\n",
            "Epoch 77/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 107.1439 - mae: 107.1439 - val_loss: 584.9904 - val_mae: 584.9904\n",
            "Epoch 78/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 107.7804 - mae: 107.7804 - val_loss: 576.2996 - val_mae: 576.2996\n",
            "Epoch 79/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 107.8882 - mae: 107.8882 - val_loss: 582.1379 - val_mae: 582.1379\n",
            "Epoch 80/100\n",
            "18/18 [==============================] - 0s 6ms/step - loss: 107.3478 - mae: 107.3478 - val_loss: 573.7223 - val_mae: 573.7223\n",
            "Epoch 81/100\n",
            "18/18 [==============================] - 0s 6ms/step - loss: 106.6845 - mae: 106.6845 - val_loss: 569.0101 - val_mae: 569.0101\n",
            "Epoch 82/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 114.4537 - mae: 114.4537 - val_loss: 565.1332 - val_mae: 565.1332\n",
            "Epoch 83/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 113.4015 - mae: 113.4015 - val_loss: 617.0717 - val_mae: 617.0717\n",
            "Epoch 84/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 111.4692 - mae: 111.4692 - val_loss: 615.0436 - val_mae: 615.0436\n",
            "Epoch 85/100\n",
            "18/18 [==============================] - 0s 4ms/step - loss: 107.9347 - mae: 107.9347 - val_loss: 601.9797 - val_mae: 601.9797\n",
            "Epoch 86/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 108.3696 - mae: 108.3696 - val_loss: 579.1899 - val_mae: 579.1899\n",
            "Epoch 87/100\n",
            "18/18 [==============================] - 0s 7ms/step - loss: 105.7380 - mae: 105.7380 - val_loss: 600.1626 - val_mae: 600.1626\n",
            "Epoch 88/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 108.2727 - mae: 108.2727 - val_loss: 567.1696 - val_mae: 567.1696\n",
            "Epoch 89/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 107.1574 - mae: 107.1574 - val_loss: 577.3275 - val_mae: 577.3275\n",
            "Epoch 90/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 105.8856 - mae: 105.8856 - val_loss: 620.6524 - val_mae: 620.6524\n",
            "Epoch 91/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 109.5507 - mae: 109.5507 - val_loss: 617.7548 - val_mae: 617.7548\n",
            "Epoch 92/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 107.5707 - mae: 107.5707 - val_loss: 584.6118 - val_mae: 584.6118\n",
            "Epoch 93/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 110.8628 - mae: 110.8628 - val_loss: 567.0992 - val_mae: 567.0992\n",
            "Epoch 94/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 108.6358 - mae: 108.6358 - val_loss: 577.5314 - val_mae: 577.5314\n",
            "Epoch 95/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 105.6491 - mae: 105.6491 - val_loss: 567.0229 - val_mae: 567.0229\n",
            "Epoch 96/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 109.7243 - mae: 109.7243 - val_loss: 653.9220 - val_mae: 653.9220\n",
            "Epoch 97/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 109.7000 - mae: 109.7000 - val_loss: 606.5756 - val_mae: 606.5756\n",
            "Epoch 98/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 107.5372 - mae: 107.5372 - val_loss: 572.6625 - val_mae: 572.6625\n",
            "Epoch 99/100\n",
            "18/18 [==============================] - 0s 26ms/step - loss: 107.5164 - mae: 107.5164 - val_loss: 564.9108 - val_mae: 564.9108\n",
            "Epoch 100/100\n",
            "18/18 [==============================] - 0s 5ms/step - loss: 118.1139 - mae: 118.1139 - val_loss: 705.5209 - val_mae: 705.5209\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.src.callbacks.History at 0x7c0951f46560>"
            ]
          },
          "metadata": {},
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZRrvDINUooJO"
      },
      "source": [
        "Because of the small size of our data (less than 3000 total samples), the model trains very fast.\n",
        "\n",
        "Let's evaluate it."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "N1DbcEZBY_Bt",
        "outputId": "3a60b962-866d-4a03-a1a3-7321dc57afea"
      },
      "source": [
        "# Evaluate model on test data\n",
        "model_1.evaluate(test_windows, test_labels)"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 4ms/step - loss: 705.5209 - mae: 705.5209\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[705.5208740234375, 705.5208740234375]"
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FXey-AX-XZPK",
        "outputId": "a322b3e9-4024-4b5b-d64d-c91cbdfd0dc4"
      },
      "source": [
        "# Load in saved best performing model_1 and evaluate on test data\n",
        "model_1 = tf.keras.models.load_model(\"model_experiments/model_1_dense\")\n",
        "model_1.evaluate(test_windows, test_labels)"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 2ms/step - loss: 564.9108 - mae: 564.9108\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[564.9107666015625, 564.9107666015625]"
            ]
          },
          "metadata": {},
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YvVtmLRdpcyH"
      },
      "source": [
        "Much better! Due to the fluctuating performance of the model during training, loading back in the best performing model see's a sizeable improvement in MAE."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hZrExWzyh3ur"
      },
      "source": [
        "def make_preds(model, input_data):\n",
        "\n",
        "  forecast = model.predict(input_data)\n",
        "  return tf.squeeze(forecast) # return 1D array of predictions"
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vcsoVIrGvrDd"
      },
      "source": [
        "Nice!\n",
        "\n",
        "Now let's use our `make_preds()` and see how it goes.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u1aAKQp6vapw",
        "outputId": "469aebe9-804c-45b2-97bc-23f4bfa5c7b7"
      },
      "source": [
        "# Make predictions using model_1 on the test dataset and view the results\n",
        "model_1_preds = make_preds(model_1, test_windows)\n",
        "len(model_1_preds), model_1_preds[:10]"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 2ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(556,\n",
              " <tf.Tensor: shape=(10,), dtype=float32, numpy=\n",
              " array([8831.117, 8777.55 , 9025.324, 8786.271, 8753.426, 8724.108,\n",
              "        8657.216, 8550.94 , 8438.561, 8521.591], dtype=float32)>)"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "q5Ra13OrwIOf"
      },
      "source": [
        "> 🔑 **Note:** With these outputs, our model isn't *forecasting* yet. It's only making predictions on the test dataset. Forecasting would involve a model making predictions into the future, however, the test dataset is only a pseudofuture.\n",
        "\n",
        "Excellent! Now we've got some prediction values, let's use the `evaluate_preds()` we created before to compare them to the ground truth."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eHak7Jqd9aXv",
        "outputId": "5dabac7b-e4a8-4356-87ad-58fb64ee3148"
      },
      "source": [
        "# Evaluate preds\n",
        "model_1_results = evaluate_preds(y_true=tf.squeeze(test_labels), # reduce to right shape\n",
        "                                 y_pred=model_1_preds)\n",
        "model_1_results"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mae': 564.9107,\n",
              " 'mse': 1157089.6,\n",
              " 'rmse': 1075.681,\n",
              " 'mape': 2.5208726,\n",
              " 'mase': 0.99239165}"
            ]
          },
          "metadata": {},
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hGL9Mj6Rwddw"
      },
      "source": [
        "How did our model go? Did it beat the naïve forecast?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5WePaRbqxTXL",
        "outputId": "ae02b128-f57d-487d-f411-f269fce0899e"
      },
      "source": [
        "naive_results"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mae': 567.9802,\n",
              " 'mse': 1147547.0,\n",
              " 'rmse': 1071.2362,\n",
              " 'mape': 2.516525,\n",
              " 'mase': 0.99957}"
            ]
          },
          "metadata": {},
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "skEwK88PxhDL"
      },
      "source": [
        "It looks like our naïve model beats our first deep model on nearly every metric.\n",
        "\n",
        "That goes to show the power of the naïve model and the reason for having a baseline for any machine learning project.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 619
        },
        "id": "Wgxoqg_tIhFJ",
        "outputId": "0f0f92f2-3f67-4d0f-b21a-0d8a52f10648"
      },
      "source": [
        "offset = 300\n",
        "plt.figure(figsize=(10, 7))\n",
        "# Account for the test_window offset and index into test_labels to ensure correct plotting\n",
        "plot_time_series(timesteps=X_test[-len(test_windows):], values=test_labels[:, 0], start=offset, label=\"Test_data\")\n",
        "plot_time_series(timesteps=X_test[-len(test_windows):], values=model_1_preds, start=offset, format=\"-\", label=\"model_1_preds\")"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fr18ffVq4i6m"
      },
      "source": [
        "## Model 2: Dense (window = 30, horizon = 1)\n",
        "\n",
        "A naïve model is currently beating my deep learning model.\n",
        "\n",
        "We'll keep the previous model architecture but use a window size of 30.\n",
        "\n",
        "In other words, we'll use the previous 30 days of Bitcoin prices to try and predict the next day price.\n",
        "\n",
        "\n",
        "The window size (how many timesteps to use to fuel a forecast) and the horizon (how many timesteps to predict into the future) are **hyperparameters**. This means you can tune them to try and find values will result in better performance.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vz0eMi7F4wpn"
      },
      "source": [
        "HORIZON = 1 # predict one step at a time\n",
        "WINDOW_SIZE = 30 # use 30 timesteps in the past"
      ],
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1HlYaAxr6UoC",
        "outputId": "69aedb11-6684-49e7-96cd-b7ad79e4a206"
      },
      "source": [
        "# Make windowed data with appropriate horizon and window sizes\n",
        "full_windows, full_labels = make_windows(prices, window_size=WINDOW_SIZE, horizon=HORIZON)\n",
        "len(full_windows), len(full_labels)"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2757, 2757)"
            ]
          },
          "metadata": {},
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jWxYZtOuelzO",
        "outputId": "42a7ae88-d5dc-466e-fc29-898422dae9b4"
      },
      "source": [
        "# Make train and testing windows\n",
        "train_windows, test_windows, train_labels, test_labels = make_train_test_splits(windows=full_windows, labels=full_labels)\n",
        "len(train_windows), len(test_windows), len(train_labels), len(test_labels)"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2205, 552, 2205, 552)"
            ]
          },
          "metadata": {},
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2xikvf8N0_dU"
      },
      "source": [
        "Data prepared!\n",
        "\n",
        "Now let's construct `model_2`, a model with the same architecture as `model_1` as well as the same training routine."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "b_oz5aJE6UwN",
        "outputId": "661d65be-42f2-42ef-bbc2-abf3c5ec7a77"
      },
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "# Create model (same model as model 1 but data input will be different)\n",
        "model_2 = tf.keras.Sequential([\n",
        "  layers.Dense(128, activation=\"relu\"),\n",
        "  layers.Dense(HORIZON) # need to predict horizon number of steps into the future\n",
        "], name=\"model_2_dense\")\n",
        "\n",
        "model_2.compile(loss=\"mae\",\n",
        "                optimizer=tf.keras.optimizers.Adam())\n",
        "\n",
        "model_2.fit(train_windows,\n",
        "            train_labels,\n",
        "            epochs=100,\n",
        "            batch_size=128,\n",
        "            verbose=0,\n",
        "            validation_data=(test_windows, test_labels),\n",
        "            callbacks=[create_model_checkpoint(model_name=model_2.name)])"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.src.callbacks.History at 0x7c08e044cfd0>"
            ]
          },
          "metadata": {},
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0oYd2nt51Ppj"
      },
      "source": [
        "Once again, training goes nice and fast.\n",
        "\n",
        "Let's evaluate our model's performance."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "W42Fa-aO8rhg",
        "outputId": "64943802-03b5-4fd4-eb64-6955b839fbd0"
      },
      "source": [
        "# Evaluate model 2 preds\n",
        "model_2.evaluate(test_windows, test_labels)"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 2ms/step - loss: 600.1887\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "600.188720703125"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c_hG89F5YV4i",
        "outputId": "b778b9a9-9a86-40f1-a676-911f4e988c2d"
      },
      "source": [
        "# Load in best performing model\n",
        "model_2 = tf.keras.models.load_model(\"model_experiments/model_2_dense/\")\n",
        "model_2.evaluate(test_windows, test_labels)"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 2ms/step - loss: 600.1887\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "600.188720703125"
            ]
          },
          "metadata": {},
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TPfvPgkUknY5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "aaf3b14c-b1f2-4f36-d373-95ffa1a2df2a"
      },
      "source": [
        "# Get forecast predictions\n",
        "model_2_preds = make_preds(model_2,\n",
        "                           input_data=test_windows)"
      ],
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 2ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dy7XZx7FllHX",
        "outputId": "eea58910-70f3-4a7c-aba6-2f81338a7426"
      },
      "source": [
        "# Evaluate results for model 2 predictions\n",
        "model_2_results = evaluate_preds(y_true=tf.squeeze(test_labels), # remove 1 dimension of test labels\n",
        "                                 y_pred=model_2_preds)\n",
        "model_2_results"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mae': 600.1887,\n",
              " 'mse': 1216403.9,\n",
              " 'rmse': 1102.907,\n",
              " 'mape': 2.7519422,\n",
              " 'mase': 1.0491358}"
            ]
          },
          "metadata": {},
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a8djCXli1wGy"
      },
      "source": [
        "It looks like `model_2` performs worse than the naïve model as well as `model_1`!\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 619
        },
        "id": "jHh_qtXflx7Z",
        "outputId": "26339129-1524-4c82-a4d1-b80dda5c29ab"
      },
      "source": [
        "offset = 300\n",
        "plt.figure(figsize=(10, 7))\n",
        "# Account for the test_window offset\n",
        "plot_time_series(timesteps=X_test[-len(test_windows):], values=test_labels[:, 0], start=offset, label=\"test_data\")\n",
        "plot_time_series(timesteps=X_test[-len(test_windows):], values=model_2_preds, start=offset, format=\"-\", label=\"model_2_preds\")"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Sls7mUaq9lMl"
      },
      "source": [
        "## Model 3: Dense (window = 30, horizon = 7)\n",
        "\n",
        "Let's try and predict 7 days ahead given the previous 30 days.\n",
        "\n",
        "First, we'll update the `HORIZON` and `WINDOW_SIZE` variables and create windowed data.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8PtmayTM98Mb",
        "outputId": "de0c7214-b146-4657-b30f-731dc94650ae"
      },
      "source": [
        "HORIZON = 7\n",
        "WINDOW_SIZE = 30\n",
        "\n",
        "full_windows, full_labels = make_windows(prices, window_size=WINDOW_SIZE, horizon=HORIZON)\n",
        "len(full_windows), len(full_labels)"
      ],
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2751, 2751)"
            ]
          },
          "metadata": {},
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "c7lhMfxWZK83"
      },
      "source": [
        "And we'll split the full dataset windows into training and test sets."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E8xgoKunZ__e",
        "outputId": "3c7040dc-6ba8-4196-d189-b56b8b2396e0"
      },
      "source": [
        "train_windows, test_windows, train_labels, test_labels = make_train_test_splits(windows=full_windows, labels=full_labels, test_split=0.2)\n",
        "len(train_windows), len(test_windows), len(train_labels), len(test_labels)"
      ],
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2200, 551, 2200, 551)"
            ]
          },
          "metadata": {},
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dlwu4mQmZvSg"
      },
      "source": [
        "Now let's build, compile, fit and evaluate a model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hxHh7xS2-CoH",
        "outputId": "cb1b8a22-3a8b-4e64-e65e-677c9f8af5c0"
      },
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "# Create model (same as model_1 except with different data input size)\n",
        "model_3 = tf.keras.Sequential([\n",
        "  layers.Dense(128, activation=\"relu\"),\n",
        "  layers.Dense(HORIZON)\n",
        "], name=\"model_3_dense\")\n",
        "\n",
        "model_3.compile(loss=\"mae\",\n",
        "                optimizer=tf.keras.optimizers.Adam())\n",
        "\n",
        "model_3.fit(train_windows,\n",
        "            train_labels,\n",
        "            batch_size=128,\n",
        "            epochs=100,\n",
        "            verbose=0,\n",
        "            validation_data=(test_windows, test_labels),\n",
        "            callbacks=[create_model_checkpoint(model_name=model_3.name)])"
      ],
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.src.callbacks.History at 0x7c08ec138520>"
            ]
          },
          "metadata": {},
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Pd-8pbfh-DDI",
        "outputId": "e7c0b951-a574-40b5-9078-4f16bf034b36"
      },
      "source": [
        "# How did our model with a larger window size and horizon go?\n",
        "model_3.evaluate(test_windows, test_labels)"
      ],
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 4ms/step - loss: 1431.8333\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1431.833251953125"
            ]
          },
          "metadata": {},
          "execution_count": 60
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XsltcXHYSrkX",
        "outputId": "8979ccde-417f-48c0-e62d-e5d2fd5b5037"
      },
      "source": [
        "# Load in best version of model_3 and evaluate\n",
        "model_3 = tf.keras.models.load_model(\"model_experiments/model_3_dense/\")\n",
        "model_3.evaluate(test_windows, test_labels)"
      ],
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 2ms/step - loss: 1251.1997\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1251.19970703125"
            ]
          },
          "metadata": {},
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "q--5dtgITJbc"
      },
      "source": [
        "In this case, the error will be higher because we're predicting 7 steps at a time.\n",
        "\n",
        "This makes sense though because the further you try and predict, the larger your error will be (think of trying to predict the weather 7 days in advance).\n",
        "\n",
        "Let's make predictions with our model using the `make_preds()` function and evaluate them using the `evaluate_preds()` function."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uKBNVovRpvhd",
        "outputId": "35c9c1b0-3deb-4714-8ac2-0e67b61065fc"
      },
      "source": [
        "# The predictions are going to be 7 steps at a time (this is the HORIZON size)\n",
        "model_3_preds = make_preds(model_3,\n",
        "                           input_data=test_windows)\n",
        "model_3_preds[:5]"
      ],
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 2ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(5, 7), dtype=float32, numpy=\n",
              "array([[9241.596 , 9445.92  , 9445.374 , 9475.937 , 9386.076 , 9626.533 ,\n",
              "        9354.451 ],\n",
              "       [8869.436 , 9069.5625, 9136.187 , 9118.451 , 9310.742 , 9290.351 ,\n",
              "        9064.147 ],\n",
              "       [8883.256 , 9009.449 , 9078.209 , 9149.523 , 9232.456 , 9180.712 ,\n",
              "        8924.966 ],\n",
              "       [8949.668 , 9305.893 , 9207.808 , 9241.16  , 9429.12  , 9189.286 ,\n",
              "        8928.253 ],\n",
              "       [8773.8125, 9088.164 , 9127.774 , 9078.996 , 9335.48  , 8952.311 ,\n",
              "        8825.679 ]], dtype=float32)>"
            ]
          },
          "metadata": {},
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "P9VQqenW-5to",
        "outputId": "67778966-625d-43b6-d1ee-2498b4fe3b5c"
      },
      "source": [
        "# Calculate model_3 results - these are going to be multi-dimensional because\n",
        "# we're trying to predict more than one step at a time.\n",
        "model_3_results = evaluate_preds(y_true=tf.squeeze(test_labels),\n",
        "                                 y_pred=model_3_preds)\n",
        "model_3_results"
      ],
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mae': array([ 631.9117  ,  374.762   ,  417.77484 ,  552.1465  ,  471.68317 ,\n",
              "         558.0112  ,  649.12164 ,  651.3516  ,  601.3091  ,  608.19336 ,\n",
              "         845.6866  ,  914.1292  ,  837.0329  ,  796.544   ,  585.05566 ,\n",
              "         208.33252 ,  226.81844 ,  211.15883 ,  168.79904 ,  289.04205 ,\n",
              "         112.3704  ,  213.24107 ,  349.89328 ,  350.4832  ,  234.348   ,\n",
              "         127.94566 ,  135.39041 ,  122.645645,  126.643   ,  199.4887  ,\n",
              "         312.32672 ,  404.15494 ,  325.37164 ,  391.43658 ,  339.07834 ,\n",
              "         308.5152  ,  353.5249  ,  284.946   ,  233.69447 ,  143.99162 ,\n",
              "         267.52545 ,   67.50907 ,  108.30595 ,  102.694336,  197.79346 ,\n",
              "         184.60463 ,  144.77176 ,  159.63518 ,  109.65541 ,  181.87766 ,\n",
              "         175.45613 ,  208.5807  ,  328.98264 ,  246.23807 ,  239.15666 ,\n",
              "         308.53055 ,  447.09494 ,  501.6855  ,  434.95285 ,  462.8042  ,\n",
              "         328.23834 ,  239.64705 ,  302.5223  ,  323.7947  ,  377.00885 ,\n",
              "         387.143   ,  422.36188 ,  502.7894  ,  114.92955 ,  157.56975 ,\n",
              "         277.51312 ,  408.52246 ,  618.47626 ,  429.2585  ,  379.74274 ,\n",
              "         399.66992 ,  405.54242 ,  407.9255  ,  479.0611  ,  599.63617 ,\n",
              "         675.65497 ,  556.5804  ,  314.29547 ,  205.92271 ,  299.62027 ,\n",
              "         263.6818  ,  254.09096 ,  286.65918 ,  318.68707 ,  499.9859  ,\n",
              "         450.55664 ,  369.60434 ,  315.84808 ,  246.26828 ,  132.79883 ,\n",
              "         249.21024 ,  328.38004 ,  497.2888  ,  721.1109  ,  849.1264  ,\n",
              "         613.4632  ,  444.2602  ,  313.09543 ,  385.73828 ,  389.7652  ,\n",
              "         428.69992 ,  466.93805 ,  738.90265 ,  988.18524 , 1015.9346  ,\n",
              "         923.19446 ,  526.17175 ,  255.74846 ,  227.26604 ,  194.66309 ,\n",
              "         361.8289  ,  288.28174 ,  417.17654 ,  577.0046  , 1121.7731  ,\n",
              "        1750.0212  , 2217.7717  , 2115.0105  , 2187.264   , 2325.7048  ,\n",
              "        2644.1055  , 1316.8909  ,  685.3775  ,  532.57855 ,  496.7271  ,\n",
              "         723.95496 ,  577.7316  ,  736.9791  ,  394.48535 ,  260.89566 ,\n",
              "         210.08963 ,  293.2895  ,  373.81348 ,  324.08295 ,  518.7489  ,\n",
              "         358.0446  ,  348.03647 ,  289.23773 ,  655.0309  ,  626.408   ,\n",
              "         706.2415  ,  735.3947  ,  312.9732  ,  153.83656 ,  118.94008 ,\n",
              "         237.44078 ,  413.29947 ,  543.3284  ,  702.17883 ,  577.537   ,\n",
              "         385.29926 ,  219.58998 ,  211.67111 ,  217.49783 ,  228.21213 ,\n",
              "         330.42432 ,  240.78711 ,  321.55365 ,  284.419   ,  302.7415  ,\n",
              "         291.03473 ,  532.02    ,  554.81866 ,  438.50043 ,  582.46405 ,\n",
              "         748.7928  ,  850.39484 ,  870.74506 , 1050.4812  ,  479.00348 ,\n",
              "         345.41586 ,  423.0886  ,  421.1723  ,  365.5046  ,  443.97867 ,\n",
              "         531.21204 ,  506.3972  ,  638.858   ,  887.8545  ,  783.3376  ,\n",
              "         403.49945 ,  404.601   ,  630.607   ,  441.04144 ,   96.154854,\n",
              "         177.81027 ,  248.29758 ,  489.87555 ,  564.5625  ,  686.7995  ,\n",
              "         619.3276  ,  341.17953 ,  405.1331  ,  370.23422 ,  375.82156 ,\n",
              "         502.7539  ,  738.79364 ,  683.1214  ,  350.4121  ,  299.51215 ,\n",
              "         209.71848 ,  280.86398 ,  510.3736  ,  273.95828 ,  164.40123 ,\n",
              "         282.54437 ,  338.11215 ,  284.8245  ,  357.37543 ,  253.2659  ,\n",
              "         451.38184 ,  566.5243  ,  225.5752  ,  343.77847 ,  196.71986 ,\n",
              "         228.62486 ,  291.8354  ,  323.4025  ,  214.54465 ,  241.52303 ,\n",
              "         179.5865  ,  189.14494 ,  284.34082 ,  354.34906 ,  536.4044  ,\n",
              "         475.4654  ,  388.48438 ,  190.87527 ,  169.49455 ,  149.19783 ,\n",
              "         159.16434 ,  205.86482 ,  230.78236 ,  184.44377 ,  141.67244 ,\n",
              "          95.95982 ,   97.75586 ,  151.08775 ,  255.5325  ,  319.78558 ,\n",
              "         314.82993 ,  288.59598 ,  209.28488 ,  262.35672 ,  193.09863 ,\n",
              "         241.27371 ,  238.9103  ,  227.67285 ,  195.55022 ,  201.51186 ,\n",
              "         205.82869 ,  506.49777 ,  572.1137  ,  708.994   ,  795.5251  ,\n",
              "         995.6642  , 1195.205   , 1206.9401  ,  417.86343 ,  234.14328 ,\n",
              "         225.72488 ,  307.3718  ,  330.62515 ,  378.8379  ,  179.69852 ,\n",
              "         226.63881 ,  159.70076 ,  147.59348 ,  281.54395 ,  362.87625 ,\n",
              "         254.28152 ,  306.2977  ,  361.20438 ,  224.61328 ,  184.44936 ,\n",
              "         152.94727 ,  214.9177  ,  210.96346 ,  332.71875 ,  447.76773 ,\n",
              "         559.8574  ,  416.5103  ,  447.62234 ,  266.5402  ,  182.51424 ,\n",
              "         308.31168 ,  412.07687 ,  280.54184 ,  130.51912 ,  203.69699 ,\n",
              "         355.70465 ,  631.25    ,  870.81555 , 1146.9668  , 1647.4363  ,\n",
              "        1574.2444  , 1080.1898  ,  702.5921  ,  167.19518 ,  169.1907  ,\n",
              "         168.899   ,  290.67203 ,  307.02985 ,  345.1063  ,  343.67047 ,\n",
              "         243.10979 ,  315.66196 ,  185.81041 ,  311.11343 ,  482.0395  ,\n",
              "         527.44324 ,  541.2804  ,  559.6826  ,  462.79184 ,  223.20857 ,\n",
              "         126.21568 ,  310.4407  ,  110.51325 ,  136.50363 ,  177.44043 ,\n",
              "         309.25836 ,  439.07883 ,  465.27008 ,  565.732   ,  309.97906 ,\n",
              "         221.83287 ,  238.0035  ,  335.7147  ,  375.45047 ,  425.024   ,\n",
              "         570.9985  ,  473.23715 ,  303.53418 ,  102.08705 ,  193.67982 ,\n",
              "         371.90402 ,  343.0848  ,  432.08148 ,  573.51984 ,  681.76105 ,\n",
              "         911.14716 , 1008.73016 , 1058.2045  , 1102.1506  ,  351.6282  ,\n",
              "         258.29086 ,  256.0088  ,  228.286   ,  214.82031 ,  223.97656 ,\n",
              "         283.7507  ,  237.41154 ,  417.37067 ,  690.80383 ,  714.7465  ,\n",
              "         792.9399  , 1010.2603  , 1263.1677  , 1100.7892  ,  467.49985 ,\n",
              "         521.6977  ,  390.64273 ,  317.35867 ,  407.69797 ,  703.8407  ,\n",
              "         776.41473 ,  919.273   , 1065.1592  , 1174.6249  , 1868.0723  ,\n",
              "        1455.4668  , 1039.1537  ,  816.7768  ,  344.06473 ,  657.3851  ,\n",
              "         846.18195 , 1070.2986  , 1120.6406  , 1088.3053  , 1018.4897  ,\n",
              "         766.3312  , 1307.0037  , 1129.0731  , 1099.8365  ,  399.53433 ,\n",
              "         249.80078 ,  498.0371  ,  898.15906 ,  806.77344 ,  846.68976 ,\n",
              "         898.07367 ,  629.2235  ,  591.37164 ,  750.7374  , 1330.2478  ,\n",
              "        2140.8186  , 2808.1802  , 3010.4973  , 3166.5747  , 3434.9412  ,\n",
              "        2412.64    ,  761.1928  ,  491.72348 ,  916.95734 , 1211.7196  ,\n",
              "        1234.2852  , 1588.9414  , 2266.49    , 3121.7402  , 3251.287   ,\n",
              "        2736.4846  , 2779.9211  , 2965.1755  , 3683.0261  , 3352.3716  ,\n",
              "        4088.1792  , 5199.644   , 4754.7993  , 4364.7866  , 4814.626   ,\n",
              "        3424.694   , 2706.5652  , 2168.819   , 3228.3086  , 4518.086   ,\n",
              "        4351.883   , 1391.5881  , 1262.8231  ,  681.72156 , 2452.3403  ,\n",
              "        2266.4614  , 2473.4153  , 3118.615   , 3820.8813  , 3897.1096  ,\n",
              "        3632.7358  , 1275.7029  , 1482.466   , 1441.9232  , 1494.4208  ,\n",
              "        1484.3318  , 2399.707   , 4422.2754  , 3300.3765  , 2626.9734  ,\n",
              "        2512.677   , 3052.666   , 4887.4854  , 4493.9033  , 4481.524   ,\n",
              "        5966.6704  , 6033.1807  , 5075.6577  , 7445.9717  , 4757.9688  ,\n",
              "        2037.4375  , 2357.894   , 1712.2377  , 2235.2175  , 3108.7974  ,\n",
              "        3305.1853  , 3731.312   , 4018.1602  , 2330.0508  , 3172.0369  ,\n",
              "        4937.7124  , 7069.418   , 9308.35    , 9178.018   , 5221.249   ,\n",
              "        2634.135   , 2228.1277  , 1728.4531  , 2458.4045  , 4044.3762  ,\n",
              "        2705.9934  , 3095.9275  , 2966.3113  , 3192.3638  , 4846.8657  ,\n",
              "        6848.556   , 5794.37    , 6026.454   , 3813.543   , 2235.3315  ,\n",
              "        1708.2617  , 1480.6903  , 2301.789   , 2912.0847  , 1128.8466  ,\n",
              "        1629.413   , 2759.7925  , 3270.8013  , 4398.929   , 5629.995   ,\n",
              "        5720.9404  , 2921.8884  , 2022.3783  , 1921.8253  , 3263.131   ,\n",
              "        4176.487   , 2393.1497  , 2289.444   , 1063.5642  ,  540.2528  ,\n",
              "        1227.649   , 1608.8075  , 1644.3711  , 1286.0831  , 1738.9012  ,\n",
              "        1499.6278  , 1743.9119  , 2347.6262  , 3379.8186  , 2652.4949  ,\n",
              "        1826.5319  , 1632.058   , 2252.737   , 3929.5208  , 5975.8584  ,\n",
              "        8232.931   , 9651.854   , 9152.968   , 7317.093   , 6203.379   ,\n",
              "        4787.8706  , 3374.4692  , 2014.2031  , 2788.3164  , 3816.399   ,\n",
              "        5681.2935  , 4174.341   , 1774.0262  , 1503.8158  , 2463.8364  ,\n",
              "        1341.8633  , 2097.8203  , 1568.9889  , 1332.5419  , 1042.4598  ,\n",
              "        1344.0251  , 1990.7155  , 3496.5386  , 5303.228   , 7600.757   ,\n",
              "        7593.7354  ], dtype=float32),\n",
              " 'mse': array([4.26170750e+05, 1.92440906e+05, 1.98566500e+05, 3.48336469e+05,\n",
              "        2.82539000e+05, 3.83829062e+05, 4.81249844e+05, 5.06474938e+05,\n",
              "        5.32538562e+05, 5.04135406e+05, 9.07809000e+05, 1.04468488e+06,\n",
              "        7.95180312e+05, 6.99229188e+05, 3.87991938e+05, 6.85639531e+04,\n",
              "        6.09342305e+04, 5.87046172e+04, 4.45722070e+04, 1.10576648e+05,\n",
              "        1.82281270e+04, 5.91971641e+04, 1.59812125e+05, 1.43799188e+05,\n",
              "        7.57895156e+04, 2.92295918e+04, 2.49712363e+04, 1.82474492e+04,\n",
              "        2.48094238e+04, 4.90360430e+04, 1.18415289e+05, 1.75775359e+05,\n",
              "        1.22896133e+05, 1.94613969e+05, 1.75098891e+05, 1.56551641e+05,\n",
              "        1.83522719e+05, 1.67053797e+05, 1.05637102e+05, 4.20609102e+04,\n",
              "        1.02932766e+05, 6.34223340e+03, 2.11344824e+04, 1.47987402e+04,\n",
              "        4.77714297e+04, 4.62672227e+04, 2.96376992e+04, 3.81983828e+04,\n",
              "        1.33777686e+04, 4.56403516e+04, 4.00349102e+04, 6.78793359e+04,\n",
              "        1.41306484e+05, 1.03333375e+05, 1.34690906e+05, 1.88566969e+05,\n",
              "        2.94477281e+05, 3.37676344e+05, 2.46287547e+05, 2.35941953e+05,\n",
              "        1.31971078e+05, 1.15020359e+05, 1.47967047e+05, 1.94727250e+05,\n",
              "        2.07930406e+05, 1.94069672e+05, 2.09866406e+05, 2.67075188e+05,\n",
              "        1.90756055e+04, 4.28093477e+04, 9.35507344e+04, 2.07306250e+05,\n",
              "        4.04841656e+05, 2.29012094e+05, 1.88742391e+05, 1.99175953e+05,\n",
              "        1.78988078e+05, 2.48756094e+05, 3.49966719e+05, 4.46531188e+05,\n",
              "        4.89354562e+05, 3.19591938e+05, 1.15160664e+05, 7.22925469e+04,\n",
              "        1.18905984e+05, 8.96590625e+04, 1.02467703e+05, 1.23269500e+05,\n",
              "        1.35092109e+05, 2.94499469e+05, 2.99278156e+05, 2.19981828e+05,\n",
              "        1.27277945e+05, 7.80100703e+04, 2.53654688e+04, 7.82946016e+04,\n",
              "        1.75920469e+05, 3.45958750e+05, 6.04345000e+05, 8.05120688e+05,\n",
              "        4.36825750e+05, 2.36622141e+05, 1.11523500e+05, 1.68348219e+05,\n",
              "        2.14705266e+05, 3.14288188e+05, 3.50514906e+05, 7.02072562e+05,\n",
              "        1.15233288e+06, 1.14945375e+06, 8.74671562e+05, 2.98269781e+05,\n",
              "        9.48786797e+04, 7.50886406e+04, 4.47696953e+04, 1.62079484e+05,\n",
              "        1.31613672e+05, 2.77330094e+05, 4.25208719e+05, 2.01619212e+06,\n",
              "        4.37694950e+06, 6.39782400e+06, 5.99759450e+06, 6.18300250e+06,\n",
              "        6.34208900e+06, 7.07939300e+06, 1.86921888e+06, 6.70133438e+05,\n",
              "        3.50882094e+05, 3.14276719e+05, 6.57727000e+05, 5.03673000e+05,\n",
              "        6.73013125e+05, 2.06092391e+05, 9.84831094e+04, 6.86804844e+04,\n",
              "        1.15768125e+05, 1.75507969e+05, 1.84051734e+05, 3.77295250e+05,\n",
              "        2.45719672e+05, 2.03656469e+05, 1.32289516e+05, 4.65466000e+05,\n",
              "        5.10814000e+05, 6.33583312e+05, 6.17672000e+05, 1.68164703e+05,\n",
              "        2.63196230e+04, 1.86135781e+04, 8.09317109e+04, 2.42892750e+05,\n",
              "        3.69780656e+05, 5.69369688e+05, 3.79257281e+05, 1.98705750e+05,\n",
              "        6.19138008e+04, 6.38810625e+04, 6.63751172e+04, 6.36967734e+04,\n",
              "        1.50575578e+05, 1.07182578e+05, 1.28152961e+05, 1.08683641e+05,\n",
              "        1.13773711e+05, 1.11256359e+05, 3.14330531e+05, 5.20064469e+05,\n",
              "        4.90185656e+05, 6.93186062e+05, 9.05461562e+05, 9.88075875e+05,\n",
              "        8.79213125e+05, 1.12571925e+06, 2.92733375e+05, 3.07353188e+05,\n",
              "        3.43274062e+05, 2.22079031e+05, 1.89917109e+05, 2.46128266e+05,\n",
              "        2.99238250e+05, 3.18697438e+05, 6.64413562e+05, 1.07791375e+06,\n",
              "        7.55403062e+05, 1.96021328e+05, 2.05619000e+05, 4.17619688e+05,\n",
              "        2.10705797e+05, 1.63497393e+04, 4.27656055e+04, 9.86786953e+04,\n",
              "        3.10046781e+05, 4.02964312e+05, 5.43333188e+05, 4.40688500e+05,\n",
              "        1.91417922e+05, 2.01833547e+05, 2.03978141e+05, 1.62042172e+05,\n",
              "        3.89608906e+05, 6.68600438e+05, 5.40565438e+05, 2.07621828e+05,\n",
              "        1.33938641e+05, 6.42924570e+04, 1.35552766e+05, 2.67330750e+05,\n",
              "        9.09545859e+04, 4.14288711e+04, 1.13355930e+05, 1.60768609e+05,\n",
              "        1.36686516e+05, 1.54084766e+05, 9.83807969e+04, 2.54297969e+05,\n",
              "        3.28341594e+05, 7.96870000e+04, 1.27498141e+05, 5.46136562e+04,\n",
              "        6.91609141e+04, 1.26278836e+05, 1.72127531e+05, 6.75833516e+04,\n",
              "        7.94092031e+04, 4.15311172e+04, 5.08004727e+04, 1.02005883e+05,\n",
              "        1.53947281e+05, 3.12685250e+05, 2.58912344e+05, 1.80977141e+05,\n",
              "        4.08415391e+04, 3.72090234e+04, 2.42331602e+04, 3.05361934e+04,\n",
              "        5.84890703e+04, 6.78819062e+04, 5.99796680e+04, 2.65834727e+04,\n",
              "        1.35539326e+04, 1.73210605e+04, 3.21904961e+04, 7.51757344e+04,\n",
              "        1.16364484e+05, 1.08212148e+05, 9.70553203e+04, 5.36959375e+04,\n",
              "        7.76758906e+04, 4.95578086e+04, 7.45636953e+04, 7.49439141e+04,\n",
              "        6.34736055e+04, 4.98062109e+04, 5.70953711e+04, 7.18335234e+04,\n",
              "        6.16588000e+05, 7.82014375e+05, 9.48799438e+05, 1.04703862e+06,\n",
              "        1.33690200e+06, 1.67704425e+06, 1.52162825e+06, 2.39286578e+05,\n",
              "        7.37736406e+04, 6.82689297e+04, 1.13812055e+05, 1.37461484e+05,\n",
              "        2.08326109e+05, 4.78767812e+04, 6.76475703e+04, 3.92902617e+04,\n",
              "        3.43073242e+04, 1.37126172e+05, 1.82150359e+05, 1.03689539e+05,\n",
              "        1.37470719e+05, 1.49051844e+05, 7.76267578e+04, 4.62961250e+04,\n",
              "        5.27745625e+04, 7.18970078e+04, 6.35186445e+04, 1.30877766e+05,\n",
              "        2.32185328e+05, 3.32530531e+05, 2.20504328e+05, 2.39760859e+05,\n",
              "        1.22527055e+05, 7.93853906e+04, 1.33627203e+05, 2.09899719e+05,\n",
              "        1.03805383e+05, 3.54098164e+04, 8.77053203e+04, 2.79881844e+05,\n",
              "        6.59944000e+05, 1.09669375e+06, 1.67025062e+06, 2.97444850e+06,\n",
              "        2.56344175e+06, 1.28343025e+06, 5.06733812e+05, 4.17506484e+04,\n",
              "        4.45121953e+04, 5.26697852e+04, 1.73961453e+05, 1.94702656e+05,\n",
              "        1.56906875e+05, 1.61626844e+05, 8.42987344e+04, 1.09336266e+05,\n",
              "        4.63877109e+04, 1.09253953e+05, 3.13154812e+05, 3.32500531e+05,\n",
              "        3.73519438e+05, 3.69692750e+05, 2.47333750e+05, 8.39415859e+04,\n",
              "        2.78023809e+04, 1.16402492e+05, 1.77936133e+04, 2.30023926e+04,\n",
              "        4.24837422e+04, 1.16625609e+05, 2.30374016e+05, 2.56472469e+05,\n",
              "        3.52634312e+05, 1.38924453e+05, 6.54949883e+04, 9.01883281e+04,\n",
              "        1.66099094e+05, 2.08969781e+05, 3.06600156e+05, 3.71231656e+05,\n",
              "        2.56479797e+05, 1.05866891e+05, 1.76879434e+04, 5.17518398e+04,\n",
              "        1.65726891e+05, 1.57775281e+05, 3.81507438e+05, 6.14783375e+05,\n",
              "        8.04329062e+05, 1.24629462e+06, 1.28638500e+06, 1.22706425e+06,\n",
              "        1.34073800e+06, 2.26328234e+05, 8.55294844e+04, 8.37505859e+04,\n",
              "        1.01793789e+05, 6.29294766e+04, 7.28699844e+04, 1.01707773e+05,\n",
              "        8.76479219e+04, 5.29326312e+05, 9.91636500e+05, 8.61327188e+05,\n",
              "        1.00046312e+06, 1.34705812e+06, 1.78851588e+06, 1.39109938e+06,\n",
              "        4.35217469e+05, 3.72617844e+05, 2.02885484e+05, 1.25443438e+05,\n",
              "        3.23774156e+05, 9.35752000e+05, 1.12190438e+06, 1.42538288e+06,\n",
              "        1.69313575e+06, 2.01447000e+06, 3.91261825e+06, 2.33501250e+06,\n",
              "        1.26401900e+06, 7.93359625e+05, 1.62204500e+05, 7.53725625e+05,\n",
              "        1.30869975e+06, 1.77725500e+06, 1.81265225e+06, 1.94052025e+06,\n",
              "        1.69492012e+06, 8.63533500e+05, 2.00005375e+06, 1.40975988e+06,\n",
              "        1.26662175e+06, 2.18133875e+05, 1.01444016e+05, 2.89763938e+05,\n",
              "        8.95635062e+05, 8.71341312e+05, 9.98311875e+05, 1.06904112e+06,\n",
              "        5.46639375e+05, 3.88459656e+05, 1.27847788e+06, 4.24055650e+06,\n",
              "        7.77481700e+06, 1.17494180e+07, 1.24452850e+07, 1.21123020e+07,\n",
              "        1.23035180e+07, 5.92084650e+06, 7.06784000e+05, 3.83711562e+05,\n",
              "        1.35111825e+06, 1.99823112e+06, 2.23767350e+06, 3.78779525e+06,\n",
              "        7.62221650e+06, 1.20161430e+07, 1.26618060e+07, 1.18462480e+07,\n",
              "        1.34828690e+07, 1.24954130e+07, 1.78832840e+07, 1.68802740e+07,\n",
              "        2.68376960e+07, 3.85855320e+07, 3.31626140e+07, 2.45360520e+07,\n",
              "        2.83136340e+07, 1.68379340e+07, 8.78316500e+06, 7.27802500e+06,\n",
              "        1.57137020e+07, 2.33372120e+07, 2.06449560e+07, 2.38492200e+06,\n",
              "        2.11368850e+06, 7.71749688e+05, 9.37650700e+06, 7.64486150e+06,\n",
              "        1.12456050e+07, 1.44237330e+07, 1.82337940e+07, 1.79811380e+07,\n",
              "        1.40555730e+07, 1.83803200e+06, 3.34367550e+06, 3.72795325e+06,\n",
              "        3.22646025e+06, 3.35254850e+06, 6.92405050e+06, 2.31628900e+07,\n",
              "        1.52435790e+07, 1.08026770e+07, 1.09424790e+07, 1.24006590e+07,\n",
              "        3.20703560e+07, 3.45687840e+07, 3.55824600e+07, 4.94022320e+07,\n",
              "        4.75712560e+07, 3.10661980e+07, 5.78276720e+07, 2.49233760e+07,\n",
              "        5.52419550e+06, 8.65933500e+06, 6.07182850e+06, 9.54779800e+06,\n",
              "        1.72449040e+07, 1.68591700e+07, 1.83619300e+07, 1.95581860e+07,\n",
              "        7.00478400e+06, 1.14721850e+07, 3.30982220e+07, 5.97597720e+07,\n",
              "        9.55701920e+07, 8.81888000e+07, 3.25257180e+07, 8.96544100e+06,\n",
              "        6.91916450e+06, 4.19955350e+06, 7.52175950e+06, 1.98867420e+07,\n",
              "        1.10678880e+07, 1.61857670e+07, 1.88217340e+07, 2.01637960e+07,\n",
              "        3.39746680e+07, 6.01878040e+07, 4.43897800e+07, 4.06911600e+07,\n",
              "        1.69550020e+07, 7.55745250e+06, 3.61449400e+06, 2.83970875e+06,\n",
              "        7.57117650e+06, 9.25349000e+06, 2.43220800e+06, 4.08131200e+06,\n",
              "        1.27114050e+07, 1.61785620e+07, 2.58089980e+07, 3.59565800e+07,\n",
              "        3.61166040e+07, 1.19999210e+07, 6.43669250e+06, 5.92174250e+06,\n",
              "        1.32501430e+07, 2.02688140e+07, 6.52411900e+06, 5.65700550e+06,\n",
              "        1.60859812e+06, 4.24327219e+05, 2.02933175e+06, 3.82929600e+06,\n",
              "        3.66077725e+06, 2.44241675e+06, 3.38819050e+06, 3.67938800e+06,\n",
              "        4.87388250e+06, 1.02501010e+07, 1.61299280e+07, 9.63716500e+06,\n",
              "        3.91816675e+06, 3.46123675e+06, 7.09518750e+06, 2.36207240e+07,\n",
              "        4.67508360e+07, 7.90942240e+07, 1.05217480e+08, 9.26967200e+07,\n",
              "        6.29534640e+07, 4.73979800e+07, 3.05733120e+07, 1.74079740e+07,\n",
              "        6.15781150e+06, 1.13768640e+07, 1.88408540e+07, 3.48213480e+07,\n",
              "        2.09131840e+07, 4.32005450e+06, 2.53467475e+06, 7.09359200e+06,\n",
              "        2.85866600e+06, 6.51586700e+06, 5.48877550e+06, 2.71571475e+06,\n",
              "        1.56679288e+06, 3.19376750e+06, 7.08283550e+06, 1.86220700e+07,\n",
              "        3.87225160e+07, 7.02177920e+07, 7.46336560e+07], dtype=float32),\n",
              " 'rmse': array([  652.8175 ,   438.68088,   445.608  ,   590.2003 ,   531.544  ,\n",
              "          619.53937,   693.72174,   711.67053,   729.7524 ,   710.0249 ,\n",
              "          952.7901 ,  1022.09827,   891.7288 ,   836.1992 ,   622.89   ,\n",
              "          261.8472 ,   246.8486 ,   242.29036,   211.1213 ,   332.53067,\n",
              "          135.01158,   243.30467,   399.76508,   379.20862,   275.29895,\n",
              "          170.96663,   158.0229 ,   135.08311,   157.51007,   221.44083,\n",
              "          344.11523,   419.2557 ,   350.56543,   441.15073,   418.44818,\n",
              "          395.66608,   428.3955 ,   408.72214,   325.01862,   205.08757,\n",
              "          320.83136,    79.63814,   145.37703,   121.65007,   218.56676,\n",
              "          215.09816,   172.15604,   195.44406,   115.6623 ,   213.63602,\n",
              "          200.08725,   260.53662,   375.90753,   321.45508,   367.0026 ,\n",
              "          434.24298,   542.6576 ,   581.09924,   496.27365,   485.73856,\n",
              "          363.27823,   339.1465 ,   384.66483,   441.2791 ,   455.99387,\n",
              "          440.5334 ,   458.1118 ,   516.79315,   138.11446,   206.90419,\n",
              "          305.86063,   455.30896,   636.27167,   478.55206,   434.44492,\n",
              "          446.29132,   423.06982,   498.75455,   591.57983,   668.22986,\n",
              "          699.5388 ,   565.3246 ,   339.3533 ,   268.87274,   344.82745,\n",
              "          299.4312 ,   320.10577,   351.09756,   367.5488 ,   542.67804,\n",
              "          547.0632 ,   469.0222 ,   356.76035,   279.30283,   159.2654 ,\n",
              "          279.8117 ,   419.42874,   588.18256,   777.3963 ,   897.28516,\n",
              "          660.9279 ,   486.4382 ,   333.95135,   410.30258,   463.363  ,\n",
              "          560.61414,   592.04297,   837.8977 ,  1073.4677 ,  1072.1257 ,\n",
              "          935.2388 ,   546.1408 ,   308.02383,   274.02307,   211.5885 ,\n",
              "          402.59097,   362.78598,   526.6214 ,   652.0803 ,  1419.9268 ,\n",
              "         2092.116  ,  2529.392  ,  2448.9985 ,  2486.5645 ,  2518.3503 ,\n",
              "         2660.713  ,  1367.1937 ,   818.61676,   592.353  ,   560.6039 ,\n",
              "          811.00366,   709.6992 ,   820.3738 ,   453.974  ,   313.8202 ,\n",
              "          262.0696 ,   340.24713,   418.9367 ,   429.0125 ,   614.24365,\n",
              "          495.7012 ,   451.28314,   363.71625,   682.2507 ,   714.7125 ,\n",
              "          795.97943,   785.92114,   410.0789 ,   162.23323,   136.43158,\n",
              "          284.485  ,   492.8415 ,   608.0959 ,   754.5659 ,   615.8387 ,\n",
              "          445.76422,   248.82484,   252.74702,   257.6337 ,   252.38219,\n",
              "          388.04068,   327.3875 ,   357.9846 ,   329.67203,   337.3036 ,\n",
              "          333.55112,   560.65186,   721.15497,   700.13257,   832.57794,\n",
              "          951.55743,   994.0201 ,   937.66364,  1060.9991 ,   541.0484 ,\n",
              "          554.3944 ,   585.89594,   471.25262,   435.7948 ,   496.11316,\n",
              "          547.02673,   564.53296,   815.11566,  1038.2263 ,   869.1392 ,\n",
              "          442.74295,   453.4523 ,   646.23505,   459.027  ,   127.8661 ,\n",
              "          206.79846,   314.13165,   556.8184 ,   634.7947 ,   737.1114 ,\n",
              "          663.84375,   437.51334,   449.25888,   451.63937,   402.54462,\n",
              "          624.1866 ,   817.67993,   735.23157,   455.65536,   365.9763 ,\n",
              "          253.55957,   368.1749 ,   517.04034,   301.58676,   203.54083,\n",
              "          336.68372,   400.9596 ,   369.7114 ,   392.53632,   313.65714,\n",
              "          504.27966,   573.011  ,   282.28885,   357.06882,   233.69565,\n",
              "          262.98462,   355.35733,   414.88254,   259.968  ,   281.7964 ,\n",
              "          203.79184,   225.3896 ,   319.3836 ,   392.36115,   559.1827 ,\n",
              "          508.8343 ,   425.41406,   202.0929 ,   192.89641,   155.67003,\n",
              "          174.74608,   241.84514,   260.54156,   244.90747,   163.04439,\n",
              "          116.42136,   131.6095 ,   179.4171 ,   274.18195,   341.12238,\n",
              "          328.95615,   311.53702,   231.72383,   278.70395,   222.61583,\n",
              "          273.06354,   273.75885,   251.93968,   223.17305,   238.94638,\n",
              "          268.01776,   785.23114,   884.3158 ,   974.06335,  1023.249  ,\n",
              "         1156.2448 ,  1295.0074 ,  1233.543  ,   489.16928,   271.61304,\n",
              "          261.28323,   337.3604 ,   370.75797,   456.42755,   218.80763,\n",
              "          260.09146,   198.21771,   185.22237,   370.3055 ,   426.79077,\n",
              "          322.0086 ,   370.77042,   386.07233,   278.61578,   215.16534,\n",
              "          229.72714,   268.13617,   252.02905,   361.77032,   481.8561 ,\n",
              "          576.6546 ,   469.5789 ,   489.6538 ,   350.03864,   281.75412,\n",
              "          365.55054,   458.14813,   322.18842,   188.17496,   296.15085,\n",
              "          529.0386 ,   812.3694 ,  1047.2314 ,  1292.3818 ,  1724.6589 ,\n",
              "         1601.0752 ,  1132.8857 ,   711.85236,   204.32976,   210.97914,\n",
              "          229.49898,   417.08685,   441.25122,   396.11472,   402.0284 ,\n",
              "          290.34244,   330.66034,   215.37807,   330.53586,   559.60236,\n",
              "          576.6286 ,   611.16235,   608.0236 ,   497.3266 ,   289.72675,\n",
              "          166.74046,   341.1781 ,   133.3927 ,   151.6654 ,   206.11584,\n",
              "          341.5049 ,   479.97293,   506.4311 ,   593.8302 ,   372.7257 ,\n",
              "          255.91989,   300.31372,   407.55255,   457.1321 ,   553.71484,\n",
              "          609.28784,   506.43832,   325.37192,   132.99602,   227.4903 ,\n",
              "          407.09567,   397.20935,   617.66284,   784.08124,   896.84393,\n",
              "         1116.3756 ,  1134.1891 ,  1107.7294 ,  1157.9023 ,   475.73965,\n",
              "          292.45425,   289.39694,   319.0514 ,   250.85748,   269.9444 ,\n",
              "          318.91656,   296.05392,   727.54816,   995.80945,   928.07715,\n",
              "         1000.2315 ,  1160.6283 ,  1337.354  ,  1179.4487 ,   659.71014,\n",
              "          610.4243 ,   450.4281 ,   354.17996,   569.01154,   967.3427 ,\n",
              "         1059.1998 ,  1193.894  ,  1301.2054 ,  1419.3202 ,  1978.0339 ,\n",
              "         1528.0747 ,  1124.286  ,   890.70734,   402.7462 ,   868.1737 ,\n",
              "         1143.9841 ,  1333.1372 ,  1346.3478 ,  1393.0255 ,  1301.891  ,\n",
              "          929.265  ,  1414.2325 ,  1187.333  ,  1125.4429 ,   467.04803,\n",
              "          318.50278,   538.29724,   946.37994,   933.4566 ,   999.1556 ,\n",
              "         1033.9445 ,   739.35065,   623.2653 ,  1130.6979 ,  2059.2612 ,\n",
              "         2788.336  ,  3427.7424 ,  3527.7876 ,  3480.2732 ,  3507.637  ,\n",
              "         2433.2788 ,   840.70447,   619.4446 ,  1162.3761 ,  1413.588  ,\n",
              "         1495.8855 ,  1946.2258 ,  2760.8362 ,  3466.431  ,  3558.343  ,\n",
              "         3441.838  ,  3671.9026 ,  3534.8853 ,  4228.8633 ,  4108.561  ,\n",
              "         5180.511  ,  6211.7256 ,  5758.6987 ,  4953.3877 ,  5321.0557 ,\n",
              "         4103.405  ,  2963.6404 ,  2697.7815 ,  3964.0513 ,  4830.8604 ,\n",
              "         4543.672  ,  1544.3193 ,  1453.8529 ,   878.4928 ,  3062.1082 ,\n",
              "         2764.934  ,  3353.4468 ,  3797.859  ,  4270.1045 ,  4240.417  ,\n",
              "         3749.0762 ,  1355.7404 ,  1828.572  ,  1930.7909 ,  1796.235  ,\n",
              "         1830.9966 ,  2631.3591 ,  4812.7837 ,  3904.3025 ,  3286.7427 ,\n",
              "         3307.9417 ,  3521.4568 ,  5663.0693 ,  5879.5225 ,  5965.1035 ,\n",
              "         7028.672  ,  6897.192  ,  5573.706  ,  7604.4507 ,  4992.3315 ,\n",
              "         2350.3606 ,  2942.6748 ,  2464.108  ,  3089.9512 ,  4152.698  ,\n",
              "         4105.9917 ,  4285.082  ,  4422.464  ,  2646.6553 ,  3387.0615 ,\n",
              "         5753.1055 ,  7730.4443 ,  9776.001  ,  9390.889  ,  5703.1323 ,\n",
              "         2994.2346 ,  2630.4304 ,  2049.2812 ,  2742.5825 ,  4459.4556 ,\n",
              "         3326.8435 ,  4023.1538 ,  4338.4023 ,  4490.4116 ,  5828.7793 ,\n",
              "         7758.08   ,  6662.5654 ,  6378.9624 ,  4117.645  ,  2749.082  ,\n",
              "         1901.1823 ,  1685.1436 ,  2751.5771 ,  3041.955  ,  1559.5537 ,\n",
              "         2020.2257 ,  3565.3057 ,  4022.2583 ,  5080.256  ,  5996.3804 ,\n",
              "         6009.709  ,  3464.09   ,  2537.0637 ,  2433.463  ,  3640.0745 ,\n",
              "         4502.09   ,  2554.2356 ,  2378.446  ,  1268.3052 ,   651.40405,\n",
              "         1424.5461 ,  1956.8588 ,  1913.3158 ,  1562.8234 ,  1840.7037 ,\n",
              "         1918.1731 ,  2207.687  ,  3201.578  ,  4016.208  ,  3104.3784 ,\n",
              "         1979.4359 ,  1860.44   ,  2663.6792 ,  4860.1157 ,  6837.458  ,\n",
              "         8893.493  , 10257.557  ,  9627.913  ,  7934.322  ,  6884.619  ,\n",
              "         5529.314  ,  4172.286  ,  2481.494  ,  3372.9607 ,  4340.6055 ,\n",
              "         5900.9614 ,  4573.0938 ,  2078.474  ,  1592.0662 ,  2663.3796 ,\n",
              "         1690.759  ,  2552.6196 ,  2342.8135 ,  1647.9426 ,  1251.716  ,\n",
              "         1787.1115 ,  2661.3596 ,  4315.3296 ,  6222.7417 ,  8379.605  ,\n",
              "         8639.077  ], dtype=float32),\n",
              " 'mape': array([ 7.2072616 ,  4.322151  ,  4.8316526 ,  6.4204774 ,  5.557499  ,\n",
              "         6.6642904 ,  7.8306303 ,  8.069127  ,  7.7801466 ,  8.015834  ,\n",
              "        11.390957  , 12.477421  , 11.518562  , 10.998036  ,  8.111763  ,\n",
              "         2.8779716 ,  3.0744495 ,  2.812423  ,  2.2368815 ,  3.8488972 ,\n",
              "         1.5013149 ,  2.8811574 ,  4.771615  ,  4.7703547 ,  3.1885545 ,\n",
              "         1.7440077 ,  1.8450122 ,  1.6641762 ,  1.7276596 ,  2.743418  ,\n",
              "         4.332176  ,  5.625597  ,  4.5808125 ,  5.6489806 ,  4.91518   ,\n",
              "         4.488555  ,  5.132786  ,  4.170611  ,  3.4067426 ,  2.0786827 ,\n",
              "         3.669066  ,  0.9396151 ,  1.5042145 ,  1.4215177 ,  2.742714  ,\n",
              "         2.553856  ,  2.0011847 ,  2.2104008 ,  1.513887  ,  2.5482407 ,\n",
              "         2.4498339 ,  2.9220035 ,  4.6008744 ,  3.4046645 ,  3.1380334 ,\n",
              "         3.94095   ,  5.6919327 ,  6.320761  ,  5.434943  ,  5.7637343 ,\n",
              "         4.056909  ,  2.8792968 ,  3.5811462 ,  3.7358613 ,  4.3226724 ,\n",
              "         4.4151826 ,  4.8007975 ,  5.71747   ,  1.2991856 ,  1.816785  ,\n",
              "         3.2198417 ,  4.7754464 ,  7.2600746 ,  5.0655255 ,  4.4784436 ,\n",
              "         4.6843505 ,  4.668666  ,  4.4902267 ,  5.1536717 ,  6.4338593 ,\n",
              "         7.23935   ,  5.950255  ,  3.3560672 ,  2.1725254 ,  3.2021139 ,\n",
              "         2.7964005 ,  2.6821423 ,  2.9608293 ,  3.243988  ,  5.0155253 ,\n",
              "         4.4418926 ,  3.6158004 ,  3.0878317 ,  2.413625  ,  1.3237141 ,\n",
              "         2.4994764 ,  3.315175  ,  5.0558696 ,  7.356037  ,  8.714648  ,\n",
              "         6.3206153 ,  4.5829363 ,  3.206883  ,  4.0087724 ,  4.186044  ,\n",
              "         4.7339926 ,  5.219607  ,  8.318615  , 11.25989   , 11.610146  ,\n",
              "        10.572622  ,  6.0269203 ,  2.9336674 ,  2.568259  ,  2.1920075 ,\n",
              "         4.067093  ,  3.451388  ,  5.092969  ,  7.069678  , 16.003912  ,\n",
              "        26.728895  , 36.149284  , 36.667988  , 40.109024  , 43.15279   ,\n",
              "        49.178665  , 24.528706  , 13.022664  ,  9.453182  ,  8.533849  ,\n",
              "        11.783623  ,  9.013297  , 11.356509  ,  6.09042   ,  3.9851558 ,\n",
              "         3.3517246 ,  4.634303  ,  5.802707  ,  5.2143097 ,  8.268144  ,\n",
              "         5.7680755 ,  5.50771   ,  4.385736  ,  9.745615  ,  9.059866  ,\n",
              "        10.092343  , 10.38244   ,  4.3248487 ,  2.183698  ,  1.6816801 ,\n",
              "         3.2948513 ,  5.945572  ,  7.8348227 , 10.196295  ,  8.394834  ,\n",
              "         5.60371   ,  3.1493993 ,  3.0475817 ,  3.0571597 ,  3.2217197 ,\n",
              "         4.6175814 ,  3.3295016 ,  4.488962  ,  3.973788  ,  4.2298775 ,\n",
              "         3.858734  ,  6.99019   ,  6.856369  ,  5.1074576 ,  6.6878123 ,\n",
              "         8.522768  ,  9.626133  ,  9.836579  , 11.825304  ,  5.2966266 ,\n",
              "         3.6192782 ,  4.384258  ,  4.417011  ,  3.8322747 ,  4.7148895 ,\n",
              "         5.732214  ,  5.54319   ,  7.20215   ,  9.947615  ,  8.734119  ,\n",
              "         4.4352055 ,  4.2294145 ,  6.567152  ,  4.576721  ,  1.0267688 ,\n",
              "         1.8859744 ,  2.6630378 ,  5.3028164 ,  6.190579  ,  7.591355  ,\n",
              "         6.871237  ,  3.7727945 ,  4.466834  ,  4.0630875 ,  4.032505  ,\n",
              "         5.1642785 ,  7.612958  ,  7.0181684 ,  3.5478048 ,  3.0429685 ,\n",
              "         2.134641  ,  2.8434384 ,  5.268584  ,  2.8199975 ,  1.6900891 ,\n",
              "         2.9501138 ,  3.532907  ,  3.00938   ,  3.7699313 ,  2.6998494 ,\n",
              "         4.8041787 ,  6.0124345 ,  2.390323  ,  3.6549249 ,  2.0940554 ,\n",
              "         2.4450886 ,  3.1205416 ,  3.4641185 ,  2.2997572 ,  2.5866392 ,\n",
              "         1.9261304 ,  2.044643  ,  3.0704725 ,  3.8586974 ,  5.8611083 ,\n",
              "         5.204462  ,  4.2623224 ,  2.0900803 ,  1.8579129 ,  1.634719  ,\n",
              "         1.7406033 ,  2.2589397 ,  2.5152197 ,  2.009804  ,  1.5392867 ,\n",
              "         1.0312641 ,  1.0477111 ,  1.6331636 ,  2.7530153 ,  3.4590852 ,\n",
              "         3.410593  ,  3.1331935 ,  2.2766812 ,  2.8553734 ,  2.1055703 ,\n",
              "         2.627226  ,  2.5883064 ,  2.4421525 ,  2.0783617 ,  2.1117213 ,\n",
              "         2.121053  ,  4.844839  ,  5.301431  ,  6.4651327 ,  7.177382  ,\n",
              "         8.928533  , 10.585185  , 10.679875  ,  3.666452  ,  2.0527427 ,\n",
              "         1.9800056 ,  2.6853507 ,  2.9034777 ,  3.343821  ,  1.5648265 ,\n",
              "         1.9355783 ,  1.3613505 ,  1.2715493 ,  2.4416595 ,  3.1333783 ,\n",
              "         2.1974618 ,  2.6394253 ,  3.052805  ,  1.8637369 ,  1.5332236 ,\n",
              "         1.2623246 ,  1.7917633 ,  1.7684687 ,  2.8188958 ,  3.8233254 ,\n",
              "         4.806147  ,  3.6007438 ,  3.8884468 ,  2.3293426 ,  1.604158  ,\n",
              "         2.6992402 ,  3.601592  ,  2.4321508 ,  1.1145098 ,  1.8139882 ,\n",
              "         3.2775528 ,  5.950266  ,  8.319802  , 11.005946  , 15.934302  ,\n",
              "        15.325653  , 10.58087   ,  6.8712735 ,  1.6301589 ,  1.6509948 ,\n",
              "         1.6131692 ,  2.729689  ,  2.8361301 ,  3.2113829 ,  3.1633127 ,\n",
              "         2.2302585 ,  2.8947792 ,  1.7031444 ,  2.8883288 ,  4.54951   ,\n",
              "         4.9846935 ,  5.136431  ,  5.3110805 ,  4.3899536 ,  2.1169522 ,\n",
              "         1.1822591 ,  2.8906436 ,  1.0288837 ,  1.2748137 ,  1.6638577 ,\n",
              "         2.906991  ,  4.1285577 ,  4.3808603 ,  5.3259287 ,  2.917017  ,\n",
              "         2.063061  ,  2.166582  ,  3.0329828 ,  3.3322685 ,  3.7097323 ,\n",
              "         5.015358  ,  4.135603  ,  2.6523542 ,  0.8899571 ,  1.6886237 ,\n",
              "         3.2551374 ,  2.994909  ,  3.543558  ,  4.564973  ,  5.3208103 ,\n",
              "         7.035156  ,  7.774451  ,  8.138774  ,  8.339833  ,  2.6297317 ,\n",
              "         1.9491053 ,  1.9268992 ,  1.6900543 ,  1.578634  ,  1.639938  ,\n",
              "         2.0904183 ,  1.7118409 ,  2.8135333 ,  4.5922933 ,  4.764145  ,\n",
              "         5.210986  ,  6.6182747 ,  8.257227  ,  7.1067715 ,  2.9603074 ,\n",
              "         3.32761   ,  2.4457996 ,  1.9902712 ,  2.4859154 ,  4.1662226 ,\n",
              "         4.491461  ,  5.249271  ,  5.999014  ,  6.473323  , 10.271582  ,\n",
              "         7.9288826 ,  5.5878496 ,  4.371578  ,  1.8873783 ,  3.7343037 ,\n",
              "         4.8430443 ,  6.1034594 ,  6.3588457 ,  6.2163863 ,  5.809095  ,\n",
              "         4.0744376 ,  6.8703046 ,  5.912217  ,  5.736818  ,  2.0849807 ,\n",
              "         1.3063419 ,  2.6369781 ,  4.7851896 ,  4.3496346 ,  4.5820312 ,\n",
              "         4.855376  ,  3.4008224 ,  3.1573913 ,  3.7728515 ,  6.1863728 ,\n",
              "         9.796061  , 12.492434  , 13.125388  , 13.715201  , 14.833133  ,\n",
              "        10.340022  ,  3.2392235 ,  2.0669715 ,  3.7279015 ,  4.872213  ,\n",
              "         4.8083777 ,  6.0394473 ,  8.331229  , 11.307008  , 11.545335  ,\n",
              "         9.183221  ,  8.904576  ,  9.474737  , 11.444388  ,  9.91715   ,\n",
              "        11.372269  , 13.955702  , 12.292148  , 11.266523  , 12.404346  ,\n",
              "         8.709873  ,  7.002915  ,  6.0054326 ,  9.064625  , 12.597691  ,\n",
              "        12.17251   ,  3.8602748 ,  3.4096603 ,  1.8603187 ,  7.2726974 ,\n",
              "         6.796523  ,  7.6618934 ,  9.664016  , 11.847018  , 12.140701  ,\n",
              "        11.449446  ,  3.9774227 ,  4.543256  ,  4.32448   ,  4.491981  ,\n",
              "         4.445234  ,  7.0442257 , 12.566368  ,  9.137582  ,  7.1341715 ,\n",
              "         6.6139708 ,  7.9944806 , 12.184411  , 10.4394045 , 10.065547  ,\n",
              "        13.226329  , 13.067389  , 10.906764  , 15.811827  ,  9.981324  ,\n",
              "         4.226197  ,  4.7441883 ,  3.3690832 ,  4.261094  ,  5.777887  ,\n",
              "         6.047913  ,  6.7989244 ,  7.3649235 ,  4.4619236 ,  6.1108375 ,\n",
              "        10.1569605 , 14.709989  , 19.728037  , 19.498625  , 11.175109  ,\n",
              "         5.630168  ,  4.7729425 ,  3.5658646 ,  5.0107603 ,  8.141646  ,\n",
              "         5.3624334 ,  5.951154  ,  5.464598  ,  5.7247605 ,  8.671238  ,\n",
              "        11.975178  ,  9.893887  , 10.348411  ,  6.5176196 ,  3.773183  ,\n",
              "         2.9161046 ,  2.5375106 ,  4.0105424 ,  5.0534754 ,  1.9917276 ,\n",
              "         2.9043036 ,  5.0534983 ,  6.082178  ,  8.192887  , 10.43093   ,\n",
              "        10.618223  ,  5.433551  ,  3.6908793 ,  3.357258  ,  5.643888  ,\n",
              "         7.1708694 ,  4.106422  ,  3.9210982 ,  1.8111696 ,  0.92831355,\n",
              "         2.1238165 ,  2.7951856 ,  2.8568456 ,  2.225082  ,  2.9866652 ,\n",
              "         2.5919483 ,  2.9400563 ,  3.7939687 ,  5.437543  ,  4.2480855 ,\n",
              "         2.938993  ,  2.6489398 ,  3.791448  ,  6.818137  , 10.536337  ,\n",
              "        14.837869  , 17.854595  , 17.279009  , 14.223083  , 12.143848  ,\n",
              "         9.420596  ,  6.6785088 ,  3.9537616 ,  5.1464567 ,  6.894689  ,\n",
              "        10.181575  ,  7.3769636 ,  3.1313381 ,  2.6931849 ,  4.3560677 ,\n",
              "         2.4053228 ,  3.7655294 ,  2.8344245 ,  2.3897355 ,  1.8089013 ,\n",
              "         2.441088  ,  3.774812  ,  6.7428613 , 10.490687  , 15.431929  ,\n",
              "        16.175167  ], dtype=float32),\n",
              " 'mase': 2.22644}"
            ]
          },
          "metadata": {},
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "t405o940Twum"
      },
      "source": [
        "## Make our evaluation function work for larger horizons\n",
        "\n",
        "The outputs for `model_3_results` are multi-dimensional.\n",
        "\n",
        "This is because the predictions are getting evaluated across the `HORIZON` timesteps (7 predictions at a time).\n",
        "\n",
        "To fix this, let's adjust our `evaluate_preds()` function to work with multiple shapes of data."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ERkl26UFgSbz"
      },
      "source": [
        "def evaluate_preds(y_true, y_pred):\n",
        "  # Make sure float32 (for metric calculations)\n",
        "  y_true = tf.cast(y_true, dtype=tf.float32)\n",
        "  y_pred = tf.cast(y_pred, dtype=tf.float32)\n",
        "\n",
        "  # Calculate various metrics\n",
        "  mae = tf.keras.metrics.mean_absolute_error(y_true, y_pred)\n",
        "  mse = tf.keras.metrics.mean_squared_error(y_true, y_pred)\n",
        "  rmse = tf.sqrt(mse)\n",
        "  mape = tf.keras.metrics.mean_absolute_percentage_error(y_true, y_pred)\n",
        "  mase = mean_absolute_scaled_error(y_true, y_pred)\n",
        "\n",
        "  # Account for different sized metrics (for longer horizons, reduce to single number)\n",
        "  if mae.ndim > 0: # if mae isn't already a scalar, reduce it to one by aggregating tensors to mean\n",
        "    mae = tf.reduce_mean(mae)\n",
        "    mse = tf.reduce_mean(mse)\n",
        "    rmse = tf.reduce_mean(rmse)\n",
        "    mape = tf.reduce_mean(mape)\n",
        "    mase = tf.reduce_mean(mase)\n",
        "\n",
        "  return {\"mae\": mae.numpy(),\n",
        "          \"mse\": mse.numpy(),\n",
        "          \"rmse\": rmse.numpy(),\n",
        "          \"mape\": mape.numpy(),\n",
        "          \"mase\": mase.numpy()}"
      ],
      "execution_count": 64,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_RnoZl28Udx4"
      },
      "source": [
        "Now we've updated `evaluate_preds()` to work with multiple shapes, how does it look?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Kt1Z8yjpUij2",
        "outputId": "61d1a516-6f95-4b05-a510-c021fc4ed0a1"
      },
      "source": [
        "# Get model_3 results aggregated to single values\n",
        "model_3_results = evaluate_preds(y_true=tf.squeeze(test_labels),\n",
        "                                 y_pred=model_3_preds)\n",
        "model_3_results"
      ],
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mae': 1251.1998,\n",
              " 'mse': 5460700.5,\n",
              " 'rmse': 1446.488,\n",
              " 'mape': 5.7115054,\n",
              " 'mase': 2.22644}"
            ]
          },
          "metadata": {},
          "execution_count": 65
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0XNIJfcsTpdA"
      },
      "source": [
        "Time to visualize.\n",
        "\n",
        "If our prediction evaluation metrics were mutli-dimensional, how do you think the predictions will look like if we plot them?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 619
        },
        "id": "gBdg1NcLs3gW",
        "outputId": "d6b29329-fe72-4eba-f974-3ac24d455e0e"
      },
      "source": [
        "offset = 300\n",
        "plt.figure(figsize=(10, 7))\n",
        "plot_time_series(timesteps=X_test[-len(test_windows):], values=test_labels[:, 0], start=offset, label=\"Test_data\")\n",
        "# Checking the shape of model_3_preds results in [n_test_samples, HORIZON] (this will screw up the plot)\n",
        "plot_time_series(timesteps=X_test[-len(test_windows):], values=model_3_preds, start=offset, label=\"model_3_preds\")"
      ],
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 619
        },
        "id": "wYpZb27bU6OO",
        "outputId": "59d1f7f8-c6a8-4967-b48a-862c55af5a54"
      },
      "source": [
        "offset = 300\n",
        "plt.figure(figsize=(10, 7))\n",
        "# Plot model_3_preds by aggregating them (note: this condenses information so the preds will look fruther ahead than the test data)\n",
        "plot_time_series(timesteps=X_test[-len(test_windows):],\n",
        "                 values=test_labels[:, 0],\n",
        "                 start=offset,\n",
        "                 label=\"Test_data\")\n",
        "plot_time_series(timesteps=X_test[-len(test_windows):],\n",
        "                 values=tf.reduce_mean(model_3_preds, axis=1),\n",
        "                 format=\"-\",\n",
        "                 start=offset,\n",
        "                 label=\"model_3_preds\")"
      ],
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ojqon-Y5VpCC"
      },
      "source": [
        "## Which of our models is performing best so far?\n",
        "\n",
        "So far, we've trained 3 models which use the same architecture but use different data inputs.\n",
        "\n",
        "Let's compare them with the naïve model to see which model is performing the best so far."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 616
        },
        "id": "QA4EKseAVrYf",
        "outputId": "2ed26680-a950-4704-8b76-7f0860128fe1"
      },
      "source": [
        "pd.DataFrame({\"naive\": naive_results[\"mae\"],\n",
        "              \"horizon_1_window_7\": model_1_results[\"mae\"],\n",
        "              \"horizon_1_window_30\": model_2_results[\"mae\"],\n",
        "              \"horizon_7_window_30\": model_3_results[\"mae\"]}, index=[\"mae\"]).plot(figsize=(10, 7), kind=\"bar\");"
      ],
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dIRwia9BWeI5"
      },
      "source": [
        "My naïve model is performing best (it's very hard to beat a naïve model in open systems) but the dense model with a horizon of 1 and a window size of 7 looks to be performing cloest.\n",
        "\n",
        "Because of this, let's use `HORIZON=1` and `WINDOW_SIZE=7` for our next series of modelling experiments (in other words, we'll use the previous week of Bitcoin prices to try and predict the next day).\n",
        "\n",
        " One of the reasons could be due the presence of **autocorrelation** in the data. If a time series has **autocorrelation** it means the value at `t+1` (the next timestep) is typically close to the value at `t` (the current timestep). In other words, today's value is probably pretty close to yesterday's value. Of course, this isn't always the case but when it is, a naïve model will often get fairly good results.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CxcL0gi-rWC5"
      },
      "source": [
        "## Model 4: Conv1D\n",
        "\n",
        "Onto the next modelling experiment!\n",
        "\n",
        "This time, we'll be using a Conv1D model. Because as we saw in the sequence modelling notebook, Conv1D models can be used for seq2seq (sequence to sequence) problems.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2_t9hXvXrZPD"
      },
      "source": [
        "HORIZON = 1 # predict next day\n",
        "WINDOW_SIZE = 7 # use previous week worth of data"
      ],
      "execution_count": 69,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CRCy-vXzsEGL",
        "outputId": "a5251f8e-c9a6-4298-d6b9-e8b8ab747d1b"
      },
      "source": [
        "# Create windowed dataset\n",
        "full_windows, full_labels = make_windows(prices, window_size=WINDOW_SIZE, horizon=HORIZON)\n",
        "len(full_windows), len(full_labels)"
      ],
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2780, 2780)"
            ]
          },
          "metadata": {},
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZtsLeB3rsED2",
        "outputId": "a44d04eb-dcbe-4ec2-8906-d1634b3d70c1"
      },
      "source": [
        "# Create train/test splits\n",
        "train_windows, test_windows, train_labels, test_labels = make_train_test_splits(full_windows, full_labels)\n",
        "len(train_windows), len(test_windows), len(train_labels), len(test_labels)"
      ],
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2224, 556, 2224, 556)"
            ]
          },
          "metadata": {},
          "execution_count": 71
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RMtzZKYRXExH"
      },
      "source": [
        "Data windowed!\n",
        "\n",
        "Now, since we're going to be using [Conv1D layers](https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv1D), we need to make sure our input shapes are correct.\n",
        "\n",
        "The Conv1D layer in TensorFlow takes an input of: `(batch_size, timesteps, input_dim)`.\n",
        "\n",
        "In our case, the `batch_size` (by default this is 32 but we can change it) is handled for us but the other values will be:\n",
        "* `timesteps = WINDOW_SIZE` - the `timesteps` is also often referred to as `features`, our features are the previous `WINDOW_SIZE` values of Bitcoin\n",
        "* `input_dim = HORIZON` - our model views `WINDOW_SIZE` (one week) worth of data at a time to predict `HORIZON` (one day)\n",
        "\n",
        "Right now, our data has the `timesteps` dimension ready but we'll have to adjust it to have the `input_dim` dimension."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "To_k7zIAX7Pj",
        "outputId": "b0439885-7ce9-4b5a-afec-b6a12d70c2c7"
      },
      "source": [
        "# Check data sample shapes\n",
        "train_windows[0].shape # returns (WINDOW_SIZE, )"
      ],
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(7,)"
            ]
          },
          "metadata": {},
          "execution_count": 72
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oPIMnb3GYMvP"
      },
      "source": [
        "To fix this, we could adjust the shape of all of our `train_windows` or we could use a [`tf.keras.layers.Lamdba`](https://keras.io/api/layers/core_layers/lambda/) (called a Lambda layer) to do this for us in our model.\n",
        "\n",
        "The Lambda layer wraps a function into a layer which can be used with a model.\n",
        "\n",
        "Let's try it out."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "62oSNy3VvZpv",
        "outputId": "469d382a-3eb3-4cf2-c969-9782d466f57c"
      },
      "source": [
        "# Before we pass our data to the Conv1D layer, we have to reshape it in order to make sure it works\n",
        "x = tf.constant(train_windows[0])\n",
        "expand_dims_layer = layers.Lambda(lambda x: tf.expand_dims(x, axis=1)) # add an extra dimension for timesteps\n",
        "print(f\"Original shape: {x.shape}\") # (WINDOW_SIZE)\n",
        "print(f\"Expanded shape: {expand_dims_layer(x).shape}\") # (WINDOW_SIZE, input_dim)\n",
        "print(f\"Original values with expanded shape:\\n {expand_dims_layer(x)}\")"
      ],
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Original shape: (7,)\n",
            "Expanded shape: (7, 1)\n",
            "Original values with expanded shape:\n",
            " [[123.65499]\n",
            " [125.455  ]\n",
            " [108.58483]\n",
            " [118.67466]\n",
            " [121.33866]\n",
            " [120.65533]\n",
            " [121.795  ]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jODevMWGsEBI",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f8131add-3a3f-4c5b-f596-55df26b456d5"
      },
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "# Create model\n",
        "model_4 = tf.keras.Sequential([\n",
        "  # Create Lambda layer to reshape inputs, without this layer, the model will error\n",
        "  layers.Lambda(lambda x: tf.expand_dims(x, axis=1)), # resize the inputs to adjust for window size / Conv1D 3D input requirements\n",
        "  layers.Conv1D(filters=128, kernel_size=5, padding=\"causal\", activation=\"relu\"),\n",
        "  layers.Dense(HORIZON)\n",
        "], name=\"model_4_conv1D\")\n",
        "\n",
        "# Compile model\n",
        "model_4.compile(loss=\"mae\",\n",
        "                optimizer=tf.keras.optimizers.Adam())\n",
        "\n",
        "# Fit model\n",
        "model_4.fit(train_windows,\n",
        "            train_labels,\n",
        "            batch_size=128,\n",
        "            epochs=100,\n",
        "            verbose=0,\n",
        "            validation_data=(test_windows, test_labels),\n",
        "            callbacks=[create_model_checkpoint(model_name=model_4.name)])"
      ],
      "execution_count": 74,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.src.callbacks.History at 0x7c0903981b70>"
            ]
          },
          "metadata": {},
          "execution_count": 74
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dWqZR5cSjHCf"
      },
      "source": [
        "What does the Lambda layer look like in a summary?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2FcGUALG4RxJ",
        "outputId": "3416edd6-b676-4830-f2ef-4cab87bb0e7f"
      },
      "source": [
        "model_4.summary()"
      ],
      "execution_count": 75,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_4_conv1D\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " lambda_1 (Lambda)           (None, 1, 7)              0         \n",
            "                                                                 \n",
            " conv1d (Conv1D)             (None, 1, 128)            4608      \n",
            "                                                                 \n",
            " dense_6 (Dense)             (None, 1, 1)              129       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 4737 (18.50 KB)\n",
            "Trainable params: 4737 (18.50 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MIZEau4Vjk-M"
      },
      "source": [
        "The Lambda layer appears the same as any other regular layer.\n",
        "\n",
        "Time to evaluate the Conv1D model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KXEL9FuhsD-3",
        "outputId": "7a8451a3-0d0f-4dd2-9e8f-4ae2099e267a"
      },
      "source": [
        "# Load in best performing Conv1D model and evaluate it on the test data\n",
        "model_4 = tf.keras.models.load_model(\"model_experiments/model_4_conv1D\")\n",
        "model_4.evaluate(test_windows, test_labels)"
      ],
      "execution_count": 76,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 4ms/step - loss: 570.5858\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "570.5857543945312"
            ]
          },
          "metadata": {},
          "execution_count": 76
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "j99tbnwSsD8R",
        "outputId": "6445c29e-1ae9-46df-ad88-926c2442d904"
      },
      "source": [
        "# Make predictions\n",
        "model_4_preds = make_preds(model_4, test_windows)\n",
        "model_4_preds[:10]"
      ],
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 2ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(10,), dtype=float32, numpy=\n",
              "array([8870.237 , 8772.3955, 9000.133 , 8782.973 , 8738.429 , 8724.684 ,\n",
              "       8679.324 , 8514.644 , 8458.337 , 8510.8125], dtype=float32)>"
            ]
          },
          "metadata": {},
          "execution_count": 77
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yjfEDWxjrZW2",
        "outputId": "6bae2fe2-1ad7-49cc-efb1-206554abd18e"
      },
      "source": [
        "# Evaluate predictions\n",
        "model_4_results = evaluate_preds(y_true=tf.squeeze(test_labels),\n",
        "                                 y_pred=model_4_preds)\n",
        "model_4_results"
      ],
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mae': 570.58575,\n",
              " 'mse': 1178181.2,\n",
              " 'rmse': 1085.4406,\n",
              " 'mape': 2.5504148,\n",
              " 'mase': 1.0023612}"
            ]
          },
          "metadata": {},
          "execution_count": 78
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pZ2s0Rx0rZjE"
      },
      "source": [
        "## Model 5: RNN (LSTM)\n",
        "\n",
        "\n",
        "Let's reuse the same data we used for the Conv1D model, except this time we'll create an [LSTM-cell](https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTM) powered RNN to model our Bitcoin data.\n",
        "\n",
        "\n",
        "\n",
        "The [`tf.keras.layers.LSTM()`](https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTM) layer takes a tensor with `[batch, timesteps, feature]` dimensions.\n",
        "\n",
        " the `batch` dimension gets taken care of for us but our data is currently only has the `feature` dimension (`WINDOW_SIZE`).\n",
        "\n",
        "To fix this, just like we did with the `Conv1D` model, we can use a `tf.keras.layers.Lambda()` layer to adjust the shape of our input tensors to the LSTM layer.\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "a6YFL_b-rZrt",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b1c5043a-e3b3-48f0-becc-41028706707a"
      },
      "source": [
        "tf.random.set_seed(42)\n",
        "\n",
        "# Let's build an LSTM model with the Functional API\n",
        "inputs = layers.Input(shape=(WINDOW_SIZE))\n",
        "x = layers.Lambda(lambda x: tf.expand_dims(x, axis=1))(inputs) # expand input dimension to be compatible with LSTM\n",
        "# print(x.shape)\n",
        "# x = layers.LSTM(128, activation=\"relu\", return_sequences=True)(x) # this layer will error if the inputs are not the right shape\n",
        "x = layers.LSTM(128, activation=\"relu\")(x) # using the tanh loss function results in a massive error\n",
        "# print(x.shape)\n",
        "# Add another optional dense layer (you could add more of these to see if they improve model performance)\n",
        "# x = layers.Dense(32, activation=\"relu\")(x)\n",
        "output = layers.Dense(HORIZON)(x)\n",
        "model_5 = tf.keras.Model(inputs=inputs, outputs=output, name=\"model_5_lstm\")\n",
        "\n",
        "# Compile model\n",
        "model_5.compile(loss=\"mae\",\n",
        "                optimizer=tf.keras.optimizers.Adam())\n",
        "\n",
        "# Seems when saving the model several warnings are appearing: https://github.com/tensorflow/tensorflow/issues/47554\n",
        "model_5.fit(train_windows,\n",
        "            train_labels,\n",
        "            epochs=100,\n",
        "            verbose=0,\n",
        "            batch_size=128,\n",
        "            validation_data=(test_windows, test_labels),\n",
        "            callbacks=[create_model_checkpoint(model_name=model_5.name)])"
      ],
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:tensorflow:Layer lstm will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.src.callbacks.History at 0x7c08e06b5750>"
            ]
          },
          "metadata": {},
          "execution_count": 79
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cOpHqmEtSsNw",
        "outputId": "19e2953b-050e-427c-e69b-a9a12985241f"
      },
      "source": [
        "# Load in best version of model 5 and evaluate on the test data\n",
        "model_5 = tf.keras.models.load_model(\"model_experiments/model_5_lstm/\")\n",
        "model_5.evaluate(test_windows, test_labels)"
      ],
      "execution_count": 80,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:tensorflow:Layer lstm will not use cuDNN kernels since it doesn't meet the criteria. It will use a generic GPU kernel as fallback when running on GPU.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 1s 2ms/step - loss: 589.7540\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "589.7540283203125"
            ]
          },
          "metadata": {},
          "execution_count": 80
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Pm_ZVNomnWvP"
      },
      "source": [
        "Now we've got the best performing LSTM model loaded in, let's make predictions with it and evaluate them."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yfG11v_51UWa",
        "outputId": "c1afb464-1167-4e24-95bf-e0919d3c5aac"
      },
      "source": [
        "# Make predictions with our LSTM model\n",
        "model_5_preds = make_preds(model_5, test_windows)\n",
        "model_5_preds[:10]"
      ],
      "execution_count": 81,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18/18 [==============================] - 0s 2ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(10,), dtype=float32, numpy=\n",
              "array([8953.819 , 8814.244 , 8969.847 , 8813.682 , 8737.807 , 8734.6875,\n",
              "       8704.712 , 8560.801 , 8466.268 , 8518.116 ], dtype=float32)>"
            ]
          },
          "metadata": {},
          "execution_count": 81
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sVv5Kg5N1axK",
        "outputId": "8c842a3d-909b-492d-f934-a177c46a74f6"
      },
      "source": [
        "# Evaluate model 5 preds\n",
        "model_5_results = evaluate_preds(y_true=tf.squeeze(test_labels),\n",
        "                                 y_pred=model_5_preds)\n",
        "model_5_results"
      ],
      "execution_count": 82,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'mae': 589.754,\n",
              " 'mse': 1252865.6,\n",
              " 'rmse': 1119.3148,\n",
              " 'mape': 2.6630607,\n",
              " 'mase': 1.0360345}"
            ]
          },
          "metadata": {},
          "execution_count": 82
        }
      ]
    }
  ]
}