{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "18102cea",
   "metadata": {},
   "outputs": [],
   "source": [
    "#| hide\n",
    "!pip install -Uqq nixtla"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a50470ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "#| hide\n",
    "from nixtla.utils import in_colab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ace35684",
   "metadata": {},
   "outputs": [],
   "source": [
    "#| hide\n",
    "IN_COLAB = in_colab()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b1ea1aa8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#| hide\n",
    "if not IN_COLAB:\n",
    "    from nixtla.utils import colab_badge\n",
    "    from dotenv import load_dotenv"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ecd9d32-9178-4768-bffa-d70c93c98311",
   "metadata": {},
   "source": [
    "## Why incorporate Date/Time Features in your Forecasts\n",
    "\n",
    "Many time series display patterns that repeat based on the calendar like demand \n",
    "increasing on weekends, sales peaking at the end of the month, or traffic \n",
    "varying by hour of the day. Recognizing and capturing these time-based patterns \n",
    "can be a powerful way to improve forecasting accuracy.\n",
    "\n",
    "While you can forecast a time series based solely on its historical values, \n",
    "adding additional date/time related features, such as the day of the \n",
    "week, month, quarter, or hour, can often enhance the model's performance. These \n",
    "features can be especially useful when your dataset lacks exogenous variables, \n",
    "but they can also complement external regressors when available.\n",
    "\n",
    "In this tutorial, we'll walk through how to incorporate these date/time features \n",
    "into TimeGPT to boost the accuracy of your forecasts."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "958454e8",
   "metadata": {},
   "source": [
    "## How to incorporate Date/Time Features in your Forecasts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2c48909a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#| echo: false\n",
    "if not IN_COLAB:\n",
    "    load_dotenv()\n",
    "    colab_badge('docs/tutorials/date_features.ipynb')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03e24b4f-6b8c-4ffa-82c6-f5b889fdd423",
   "metadata": {},
   "source": [
    "### Step 1: Import Packages"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "-SBYXOiTxqyF",
   "metadata": {},
   "source": [
    "Import the necessary libraries and initialize the Nixtla client."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9d984aea-1315-4d4e-8b4d-b23efe947be1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from nixtla import NixtlaClient\n",
    "\n",
    "# For forecast evaluation\n",
    "from utilsforecast.evaluation import evaluate\n",
    "from utilsforecast.losses import mae, rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b73a131-390e-46b9-847b-173f7d3c869a",
   "metadata": {},
   "source": [
    "You can instantiate the `NixtlaClient` class providing your authentication API key."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f1098659-f250-4663-b588-f9e17065cafa",
   "metadata": {},
   "outputs": [],
   "source": [
    "nixtla_client = NixtlaClient(\n",
    "    # defaults to os.environ.get(\"NIXTLA_API_KEY\")\n",
    "    api_key='my_api_key_provided_by_nixtla'  \n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e5b8ea7f-e30e-4001-a7a6-9e935e12180a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#| hide\n",
    "if not IN_COLAB:\n",
    "    nixtla_client = NixtlaClient()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73zXk_wIx8UT",
   "metadata": {},
   "source": [
    "### Step 2: Load Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38IS84oey71C",
   "metadata": {},
   "source": [
    "In this notebook, we use hourly electricity prices as our example dataset, which consists of 5 time series, each with approximately 1700 data points. For demonstration purposes, we focus on the German electricity price series. The time series is split, with the last 240 steps (10 days) set aside as the test set.\n",
    "\n",
    "For simplicity, we will also demonstrate this tutorial without the use of any additional exogenous variables, but you could extend this same technique for datasets that have exogenous variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "D8Ng_1YfqQNF",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('https://raw.githubusercontent.com/Nixtla/transfer-learning-time-series/main/datasets/electricity-short-with-ex-vars.csv')\n",
    "df['ds'] = pd.to_datetime(df['ds'])\n",
    "df_sub = df.query('unique_id == \"DE\"')[['unique_id','ds','y']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "isQ-g8KBqQ0W",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1440, 3), (240, 3))"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train = df_sub.query('ds < \"2017-12-21\"')\n",
    "df_test = df_sub.query('ds >= \"2017-12-21\"')\n",
    "df_train.shape, df_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "n5xWcQOnvSLQ",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x350 with 1 Axes>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nixtla_client.plot(df_train, df_test.rename(columns={'y': 'test'}))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "WMzFOJqG0HBD",
   "metadata": {},
   "source": [
    "### Step 3: Forecasting"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "SGY4Z7qt0L8e",
   "metadata": {},
   "source": [
    "#### Without Datetime Features"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "SuzebyUl0VSM",
   "metadata": {},
   "source": [
    "First, we forecast the univariate time series without the use of datetime features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "Vf-IoRoVqWZq",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:nixtla.nixtla_client:The specified horizon \"h\" exceeds the model horizon, this may lead to less accurate forecasts. Please consider using a smaller horizon.\n"
     ]
    }
   ],
   "source": [
    "fcst_timegpt_no_dt = nixtla_client.forecast(\n",
    "    df = df_train,\n",
    "    h=24*10,\n",
    "    model=\"timegpt-1-long-horizon\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "pQxG1w_n0bjx",
   "metadata": {},
   "source": [
    "We will rename the forecast column for this approach, so that we can distinguish it from forecasts created using other methods later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "egIruxwTtfsx",
   "metadata": {},
   "outputs": [],
   "source": [
    "fcst_timegpt_no_dt.rename(columns={\"TimeGPT\": \"TimeGPT_no_dt\"}, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "jeYxbJ__0Rcx",
   "metadata": {},
   "source": [
    "#### With Inbuilt Datetime Features"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "tNN7Hvie0iZB",
   "metadata": {},
   "source": [
    "Next, let's forecast the same univariate time series with datetime features. This can be done by specifying the `date_features` argument. The data is hourly, so both the hour of the day (`hour`) and the day of the week (`dayofweek`) may impact the usage.\n",
    "\n",
    "For example, the usage may peak in the afternoon and drop off at night. It can also differ between the weekdays and weekends due to working and holiday patterns. Including these features can help the model make better forecasts.\n",
    "\n",
    "> NOTE:\n",
    "1. In order to show how these features are created, we can add the `feature_contribution` agrument. This is just for demonstration purposes in this tutorial and not truly needed to forecast with datetime features.\n",
    "2. If you have a weekly frequency dataset, you can use `date_features = [\"week\", \"month\", \"year\"]` or a subset of these features.\n",
    "3. If you have a monthly frequency dataset, you can use `date_features = [\"month\", \"year\"]` or a subset of these features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "x60kqF3d2uHj",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:nixtla.nixtla_client:The specified horizon \"h\" exceeds the model horizon, this may lead to less accurate forecasts. Please consider using a smaller horizon.\n"
     ]
    }
   ],
   "source": [
    "fcst_timegpt_dt_no_ohe = nixtla_client.forecast(\n",
    "    df = df_train,\n",
    "    h=24*10,\n",
    "    model=\"timegpt-1-long-horizon\",\n",
    "    date_features=['hour', 'dayofweek'],\n",
    "    feature_contributions=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8P9IUb9F2vt6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-121301eb-fb1f-4733-9934-a211d3d7ce1f\" 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>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>TimeGPT</th>\n",
       "      <th>hour</th>\n",
       "      <th>dayofweek</th>\n",
       "      <th>base_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 00:00:00</td>\n",
       "      <td>34.945976</td>\n",
       "      <td>-12.797431</td>\n",
       "      <td>4.236599</td>\n",
       "      <td>43.506810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 01:00:00</td>\n",
       "      <td>33.700954</td>\n",
       "      <td>-14.274811</td>\n",
       "      <td>4.168986</td>\n",
       "      <td>43.806778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 02:00:00</td>\n",
       "      <td>32.120293</td>\n",
       "      <td>-15.785894</td>\n",
       "      <td>4.123096</td>\n",
       "      <td>43.783092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 03:00:00</td>\n",
       "      <td>32.544914</td>\n",
       "      <td>-15.623017</td>\n",
       "      <td>4.542475</td>\n",
       "      <td>43.625454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 04:00:00</td>\n",
       "      <td>33.698105</td>\n",
       "      <td>-14.559433</td>\n",
       "      <td>4.525819</td>\n",
       "      <td>43.731720</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-121301eb-fb1f-4733-9934-a211d3d7ce1f')\"\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-121301eb-fb1f-4733-9934-a211d3d7ce1f 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-121301eb-fb1f-4733-9934-a211d3d7ce1f');\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-aab81b85-10b6-423f-a015-024e806a2d54\">\n",
       "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-aab81b85-10b6-423f-a015-024e806a2d54')\"\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-aab81b85-10b6-423f-a015-024e806a2d54 button');\n",
       "          quickchartButtonEl.style.display =\n",
       "            google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "        })();\n",
       "      </script>\n",
       "    </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "text/plain": [
       "  unique_id                  ds    TimeGPT       hour  dayofweek  base_value\n",
       "0        DE 2017-12-21 00:00:00  34.945976 -12.797431   4.236599   43.506810\n",
       "1        DE 2017-12-21 01:00:00  33.700954 -14.274811   4.168986   43.806778\n",
       "2        DE 2017-12-21 02:00:00  32.120293 -15.785894   4.123096   43.783092\n",
       "3        DE 2017-12-21 03:00:00  32.544914 -15.623017   4.542475   43.625454\n",
       "4        DE 2017-12-21 04:00:00  33.698105 -14.559433   4.525819   43.731720"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "shap_df = nixtla_client.feature_contributions\n",
    "shap_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6hV4WES_3jjK",
   "metadata": {},
   "source": [
    "As we can see, two new exogenous features (`hour` and `dayofweek`) got added to the dataset and the forecast utilized these features."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "LVkK-bzV2wir",
   "metadata": {},
   "source": [
    "However, we need to ensure that the model treats each hour (0, 1, 2, ..., 23) and each day (0, 1, 2, ..., 6) as a categorical variable and not as a numerical variable. If treated numerically, the model may exaggerate differences (e.g., hour 23 might appear 23 times more influential than hour 1), which doesn't reflect real patterns. Electricity usage at hour 23 is typically similar to hour 1, and day 6 usage often resembles day 0. \n",
    "\n",
    "To avoid this distortion, we one-hot encode these variables using the `date_features_to_one_hot` argument. This creates a separate exogenous feature for each hour and each day, allowing the model to capture their effects independently."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "-BmyfxZ0q0iT",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:nixtla.nixtla_client:The specified horizon \"h\" exceeds the model horizon, this may lead to less accurate forecasts. Please consider using a smaller horizon.\n"
     ]
    }
   ],
   "source": [
    "fcst_timegpt_dt = nixtla_client.forecast(\n",
    "    df = df_train,\n",
    "    h=24*10,\n",
    "    model=\"timegpt-1-long-horizon\",\n",
    "    date_features=['hour', 'dayofweek'],\n",
    "    date_features_to_one_hot=['hour', 'dayofweek'],\n",
    "    feature_contributions=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8mAMOlpErcX5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-c3822db1-54e4-407f-8f08-f08f7b6f4d05\" 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>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>TimeGPT</th>\n",
       "      <th>hour_0</th>\n",
       "      <th>hour_1</th>\n",
       "      <th>hour_2</th>\n",
       "      <th>hour_3</th>\n",
       "      <th>hour_4</th>\n",
       "      <th>hour_5</th>\n",
       "      <th>hour_6</th>\n",
       "      <th>...</th>\n",
       "      <th>hour_22</th>\n",
       "      <th>hour_23</th>\n",
       "      <th>dayofweek_0</th>\n",
       "      <th>dayofweek_1</th>\n",
       "      <th>dayofweek_2</th>\n",
       "      <th>dayofweek_3</th>\n",
       "      <th>dayofweek_4</th>\n",
       "      <th>dayofweek_5</th>\n",
       "      <th>dayofweek_6</th>\n",
       "      <th>base_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 00:00:00</td>\n",
       "      <td>35.248108</td>\n",
       "      <td>-13.396377</td>\n",
       "      <td>0.387143</td>\n",
       "      <td>0.423001</td>\n",
       "      <td>0.392672</td>\n",
       "      <td>0.373034</td>\n",
       "      <td>0.333778</td>\n",
       "      <td>0.147671</td>\n",
       "      <td>...</td>\n",
       "      <td>0.271507</td>\n",
       "      <td>0.393282</td>\n",
       "      <td>0.472389</td>\n",
       "      <td>-0.377321</td>\n",
       "      <td>-0.548429</td>\n",
       "      <td>-0.101086</td>\n",
       "      <td>-0.133001</td>\n",
       "      <td>1.455560</td>\n",
       "      <td>2.975230</td>\n",
       "      <td>44.333805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 01:00:00</td>\n",
       "      <td>34.400800</td>\n",
       "      <td>0.358443</td>\n",
       "      <td>-14.488875</td>\n",
       "      <td>0.389985</td>\n",
       "      <td>0.359990</td>\n",
       "      <td>0.341219</td>\n",
       "      <td>0.320964</td>\n",
       "      <td>0.135058</td>\n",
       "      <td>...</td>\n",
       "      <td>0.266497</td>\n",
       "      <td>0.391259</td>\n",
       "      <td>0.445456</td>\n",
       "      <td>-0.306117</td>\n",
       "      <td>-0.436959</td>\n",
       "      <td>-0.172850</td>\n",
       "      <td>-0.151865</td>\n",
       "      <td>1.533456</td>\n",
       "      <td>3.022358</td>\n",
       "      <td>44.539093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 02:00:00</td>\n",
       "      <td>33.175526</td>\n",
       "      <td>0.375983</td>\n",
       "      <td>0.372809</td>\n",
       "      <td>-15.824338</td>\n",
       "      <td>0.348533</td>\n",
       "      <td>0.351379</td>\n",
       "      <td>0.317832</td>\n",
       "      <td>0.123833</td>\n",
       "      <td>...</td>\n",
       "      <td>0.273698</td>\n",
       "      <td>0.410714</td>\n",
       "      <td>0.417348</td>\n",
       "      <td>-0.279551</td>\n",
       "      <td>-0.342991</td>\n",
       "      <td>-0.171547</td>\n",
       "      <td>-0.142890</td>\n",
       "      <td>1.532721</td>\n",
       "      <td>3.042772</td>\n",
       "      <td>44.515614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 03:00:00</td>\n",
       "      <td>33.205390</td>\n",
       "      <td>0.368333</td>\n",
       "      <td>0.366936</td>\n",
       "      <td>0.372584</td>\n",
       "      <td>-15.880591</td>\n",
       "      <td>0.346306</td>\n",
       "      <td>0.319877</td>\n",
       "      <td>0.136488</td>\n",
       "      <td>...</td>\n",
       "      <td>0.276705</td>\n",
       "      <td>0.416273</td>\n",
       "      <td>0.508190</td>\n",
       "      <td>-0.274014</td>\n",
       "      <td>-0.339005</td>\n",
       "      <td>-0.176228</td>\n",
       "      <td>-0.152890</td>\n",
       "      <td>1.588364</td>\n",
       "      <td>3.095226</td>\n",
       "      <td>44.391410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 04:00:00</td>\n",
       "      <td>34.689583</td>\n",
       "      <td>0.363581</td>\n",
       "      <td>0.363459</td>\n",
       "      <td>0.393807</td>\n",
       "      <td>0.362043</td>\n",
       "      <td>-14.755774</td>\n",
       "      <td>0.314718</td>\n",
       "      <td>0.141911</td>\n",
       "      <td>...</td>\n",
       "      <td>0.274819</td>\n",
       "      <td>0.402653</td>\n",
       "      <td>0.531417</td>\n",
       "      <td>-0.277548</td>\n",
       "      <td>-0.360688</td>\n",
       "      <td>-0.159342</td>\n",
       "      <td>-0.169762</td>\n",
       "      <td>1.692538</td>\n",
       "      <td>3.165733</td>\n",
       "      <td>44.505848</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 35 columns</p>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c3822db1-54e4-407f-8f08-f08f7b6f4d05')\"\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-c3822db1-54e4-407f-8f08-f08f7b6f4d05 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-c3822db1-54e4-407f-8f08-f08f7b6f4d05');\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-07f2c431-4d11-4096-974a-daa8e40d1f01\">\n",
       "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-07f2c431-4d11-4096-974a-daa8e40d1f01')\"\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-07f2c431-4d11-4096-974a-daa8e40d1f01 button');\n",
       "          quickchartButtonEl.style.display =\n",
       "            google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "        })();\n",
       "      </script>\n",
       "    </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "text/plain": [
       "  unique_id                  ds    TimeGPT     hour_0     hour_1     hour_2  \\\n",
       "0        DE 2017-12-21 00:00:00  35.248108 -13.396377   0.387143   0.423001   \n",
       "1        DE 2017-12-21 01:00:00  34.400800   0.358443 -14.488875   0.389985   \n",
       "2        DE 2017-12-21 02:00:00  33.175526   0.375983   0.372809 -15.824338   \n",
       "3        DE 2017-12-21 03:00:00  33.205390   0.368333   0.366936   0.372584   \n",
       "4        DE 2017-12-21 04:00:00  34.689583   0.363581   0.363459   0.393807   \n",
       "\n",
       "      hour_3     hour_4    hour_5    hour_6  ...   hour_22   hour_23  \\\n",
       "0   0.392672   0.373034  0.333778  0.147671  ...  0.271507  0.393282   \n",
       "1   0.359990   0.341219  0.320964  0.135058  ...  0.266497  0.391259   \n",
       "2   0.348533   0.351379  0.317832  0.123833  ...  0.273698  0.410714   \n",
       "3 -15.880591   0.346306  0.319877  0.136488  ...  0.276705  0.416273   \n",
       "4   0.362043 -14.755774  0.314718  0.141911  ...  0.274819  0.402653   \n",
       "\n",
       "   dayofweek_0  dayofweek_1  dayofweek_2  dayofweek_3  dayofweek_4  \\\n",
       "0     0.472389    -0.377321    -0.548429    -0.101086    -0.133001   \n",
       "1     0.445456    -0.306117    -0.436959    -0.172850    -0.151865   \n",
       "2     0.417348    -0.279551    -0.342991    -0.171547    -0.142890   \n",
       "3     0.508190    -0.274014    -0.339005    -0.176228    -0.152890   \n",
       "4     0.531417    -0.277548    -0.360688    -0.159342    -0.169762   \n",
       "\n",
       "   dayofweek_5  dayofweek_6  base_value  \n",
       "0     1.455560     2.975230   44.333805  \n",
       "1     1.533456     3.022358   44.539093  \n",
       "2     1.532721     3.042772   44.515614  \n",
       "3     1.588364     3.095226   44.391410  \n",
       "4     1.692538     3.165733   44.505848  \n",
       "\n",
       "[5 rows x 35 columns]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "shap_df = nixtla_client.feature_contributions\n",
    "shap_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "uXcAnvwo363f",
   "metadata": {},
   "source": [
    "As we can see above, this now creates a separate feature for each hour of the day and each day of the week.\n",
    "\n",
    "> NOTE: With one hot encoding, the number of features can increase by a lot. This is especially true if you have weekly frequency data and you are using  `date_feature=[\"week\"]` because this leads to 52 features being created after one hot encoding. Please make sure that your dataset has enough datapoints or else the model will overfit to the data. You can increase the number of datapoints in the dataset by increasing the available history for your time series, or increasing the number of unique time series that share a common pattern in your dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "BsuoC9DnwFSR",
   "metadata": {},
   "outputs": [],
   "source": [
    "fcst_timegpt_dt.rename(columns={\"TimeGPT\": \"fcst_timegpt_dt\"}, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69XJH6Ea6945",
   "metadata": {},
   "source": [
    "#### With Custom Datetime Features"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "nCQ5eZyX7BVn",
   "metadata": {},
   "source": [
    "In the example above, we saw how to incorporate the inbuilt datetime features into the forecast. However, as seen above, in some cases, it may not be feasible to one hot encode the datetime features since it may lead to a large number of features for the dataset size. In that case, we can create a custom datetime feature and use it in the forecast.\n",
    "\n",
    "In this example, we will create a sine/cosine encoder for the week which is a popular technique to encode datetime features due to their circular nature described above (e.g. hour 23 behavior is close to hour 0 behavior, week 52 behavior is very close to week 1 behavior, etc.)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4dt4LWiK93ae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-813cd938-b5bb-4616-9f69-a4a07be3b2d5\" 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>week_sin</th>\n",
       "      <th>week_cos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2023-12-18</th>\n",
       "      <td>-0.3482</td>\n",
       "      <td>0.9374</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-12-25</th>\n",
       "      <td>-0.2349</td>\n",
       "      <td>0.9720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-01-01</th>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-01-08</th>\n",
       "      <td>0.1183</td>\n",
       "      <td>0.9930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-01-15</th>\n",
       "      <td>0.2349</td>\n",
       "      <td>0.9720</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-813cd938-b5bb-4616-9f69-a4a07be3b2d5')\"\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-813cd938-b5bb-4616-9f69-a4a07be3b2d5 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-813cd938-b5bb-4616-9f69-a4a07be3b2d5');\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-2e2edc22-84e1-4437-b932-113b617555fc\">\n",
       "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-2e2edc22-84e1-4437-b932-113b617555fc')\"\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-2e2edc22-84e1-4437-b932-113b617555fc button');\n",
       "          quickchartButtonEl.style.display =\n",
       "            google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "        })();\n",
       "      </script>\n",
       "    </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "text/plain": [
       "            week_sin  week_cos\n",
       "2023-12-18   -0.3482    0.9374\n",
       "2023-12-25   -0.2349    0.9720\n",
       "2024-01-01    0.0000    1.0000\n",
       "2024-01-08    0.1183    0.9930\n",
       "2024-01-15    0.2349    0.9720"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class SinCosWeekOfYear:\n",
    "    \"\"\"\n",
    "    Adds sine and cosine features for each week of the year. This is useful for\n",
    "    models that can benefit from understanding the periodicity of weeks in a year.\n",
    "    \"\"\"\n",
    "    def __call__(self, dates: pd.DatetimeIndex):\n",
    "        df = pd.DataFrame(index=dates)\n",
    "        # Get week of year (1 to 53)\n",
    "        weeks = np.array([date.isocalendar().week for date in dates])\n",
    "\n",
    "        # Calculate sine and cosine features\n",
    "        df[\"week_sin\"] = np.sin((2 * np.pi) * (weeks-1) / 53).round(4)\n",
    "        df[\"week_cos\"] = np.cos((2 * np.pi) * (weeks-1) / 53).round(4)\n",
    "        return df\n",
    "\n",
    "    def __name__(self):\n",
    "        return \"SinCosWeekOfYear\"\n",
    "\n",
    "# Example usage\n",
    "dates = pd.date_range(start='2023-01-01', periods=55, freq='W-MON')\n",
    "sin_cos_week = SinCosWeekOfYear()\n",
    "features = sin_cos_week(dates)\n",
    "features.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "zGh6nAcVLTcN",
   "metadata": {},
   "source": [
    "As we can see above, because of the cyclical encoding of the datetime feature, the encoded values (`week_sin` and `week_cos`) for week 2023-12-25 (week 52) is very close to 2024-01-01 (week 1). This will ensure that the learned features for week 52 will be close to those for week 1. This has also helped us get the feature cardinality down from 53 (in case of one hot encoding) to only 2 features."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "yBA1qOUZLZ0e",
   "metadata": {},
   "source": [
    "In our example, we have the hour feature wich has a relatively high cardinality after one hot encoding. Let's encode this with sine and cosine features and use this instead of the one hot encoding."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "Wq-G6hUBIRIw",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-53b809b9-57de-4dfe-8c7a-8ddc2ae5090b\" 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>hour_sin</th>\n",
       "      <th>hour_cos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2023-01-01 21:00:00</th>\n",
       "      <td>-0.7071</td>\n",
       "      <td>0.7071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-01-01 22:00:00</th>\n",
       "      <td>-0.5000</td>\n",
       "      <td>0.8660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-01-01 23:00:00</th>\n",
       "      <td>-0.2588</td>\n",
       "      <td>0.9659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-01-02 00:00:00</th>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-01-02 01:00:00</th>\n",
       "      <td>0.2588</td>\n",
       "      <td>0.9659</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-53b809b9-57de-4dfe-8c7a-8ddc2ae5090b')\"\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-53b809b9-57de-4dfe-8c7a-8ddc2ae5090b 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-53b809b9-57de-4dfe-8c7a-8ddc2ae5090b');\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-77b223c9-99db-46a4-8a3b-eb74a3779d36\">\n",
       "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-77b223c9-99db-46a4-8a3b-eb74a3779d36')\"\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-77b223c9-99db-46a4-8a3b-eb74a3779d36 button');\n",
       "          quickchartButtonEl.style.display =\n",
       "            google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "        })();\n",
       "      </script>\n",
       "    </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "text/plain": [
       "                     hour_sin  hour_cos\n",
       "2023-01-01 21:00:00   -0.7071    0.7071\n",
       "2023-01-01 22:00:00   -0.5000    0.8660\n",
       "2023-01-01 23:00:00   -0.2588    0.9659\n",
       "2023-01-02 00:00:00    0.0000    1.0000\n",
       "2023-01-02 01:00:00    0.2588    0.9659"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class SinCosHourOfDay:\n",
    "    \"\"\"\n",
    "    Adds sine and cosine features for each hour of the day. This is useful for\n",
    "    models that can benefit from understanding the periodicity of hours in a day.\n",
    "    \"\"\"\n",
    "    def __call__(self, dates: pd.DatetimeIndex):\n",
    "        df = pd.DataFrame(index=dates)\n",
    "        # Get hour of day (0 to 23)\n",
    "        hours = np.array([date.hour for date in dates])\n",
    "\n",
    "        # Calculate sine and cosine features\n",
    "        df[\"hour_sin\"] = np.sin((2 * np.pi) * (hours) / 24).round(4)\n",
    "        df[\"hour_cos\"] = np.cos((2 * np.pi) * (hours) / 24).round(4)\n",
    "        return df\n",
    "\n",
    "    def __name__(self):\n",
    "        return \"SinCosHourOfDay\"\n",
    "\n",
    "# Example usage\n",
    "dates = pd.date_range(start='2023-01-01 00:00', periods=26, freq='h')\n",
    "sin_cos_hour = SinCosHourOfDay()\n",
    "features = sin_cos_hour(dates)\n",
    "features.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "MWxvx8b6Gv7o",
   "metadata": {},
   "source": [
    "In order to use this custom datetime feature, we can simply pass an instance of the class to the `date_features` argument. Since this is alreay encoded, we do not need to include it in the `date_features_to_one_hot` argument."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "TAOFxjO8_Dx4",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:nixtla.nixtla_client:The specified horizon \"h\" exceeds the model horizon, this may lead to less accurate forecasts. Please consider using a smaller horizon.\n"
     ]
    }
   ],
   "source": [
    "fcst_timegpt_dt_custom = nixtla_client.forecast(\n",
    "    df = df_train,\n",
    "    h=24*10,\n",
    "    model=\"timegpt-1-long-horizon\",\n",
    "    date_features=[SinCosHourOfDay(), 'dayofweek'],\n",
    "    date_features_to_one_hot=['dayofweek'],\n",
    "    feature_contributions=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "G8uVPEQ7Hvit",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-3a028d21-3d75-4a61-96d7-4bd024c68059\" 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>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>TimeGPT</th>\n",
       "      <th>hour_sin</th>\n",
       "      <th>hour_cos</th>\n",
       "      <th>dayofweek_0</th>\n",
       "      <th>dayofweek_1</th>\n",
       "      <th>dayofweek_2</th>\n",
       "      <th>dayofweek_3</th>\n",
       "      <th>dayofweek_4</th>\n",
       "      <th>dayofweek_5</th>\n",
       "      <th>dayofweek_6</th>\n",
       "      <th>base_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 00:00:00</td>\n",
       "      <td>35.801600</td>\n",
       "      <td>-3.609636</td>\n",
       "      <td>-9.003666</td>\n",
       "      <td>0.805974</td>\n",
       "      <td>-0.424078</td>\n",
       "      <td>-0.343238</td>\n",
       "      <td>-0.428668</td>\n",
       "      <td>-0.055370</td>\n",
       "      <td>1.462214</td>\n",
       "      <td>3.295479</td>\n",
       "      <td>44.102590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 01:00:00</td>\n",
       "      <td>34.419390</td>\n",
       "      <td>-3.824628</td>\n",
       "      <td>-10.493365</td>\n",
       "      <td>0.714771</td>\n",
       "      <td>-0.400898</td>\n",
       "      <td>-0.282606</td>\n",
       "      <td>-0.331269</td>\n",
       "      <td>-0.115753</td>\n",
       "      <td>1.539153</td>\n",
       "      <td>3.245723</td>\n",
       "      <td>44.368263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 02:00:00</td>\n",
       "      <td>32.892105</td>\n",
       "      <td>-4.959243</td>\n",
       "      <td>-10.772224</td>\n",
       "      <td>0.712402</td>\n",
       "      <td>-0.439891</td>\n",
       "      <td>-0.261654</td>\n",
       "      <td>-0.207954</td>\n",
       "      <td>-0.191223</td>\n",
       "      <td>1.481960</td>\n",
       "      <td>3.206257</td>\n",
       "      <td>44.323673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 03:00:00</td>\n",
       "      <td>32.727295</td>\n",
       "      <td>-5.161374</td>\n",
       "      <td>-10.812295</td>\n",
       "      <td>0.771099</td>\n",
       "      <td>-0.417504</td>\n",
       "      <td>-0.262543</td>\n",
       "      <td>-0.146066</td>\n",
       "      <td>-0.258350</td>\n",
       "      <td>1.578070</td>\n",
       "      <td>3.268950</td>\n",
       "      <td>44.167310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 04:00:00</td>\n",
       "      <td>34.121994</td>\n",
       "      <td>-3.687167</td>\n",
       "      <td>-11.353230</td>\n",
       "      <td>0.846524</td>\n",
       "      <td>-0.387008</td>\n",
       "      <td>-0.278475</td>\n",
       "      <td>-0.169525</td>\n",
       "      <td>-0.255498</td>\n",
       "      <td>1.788180</td>\n",
       "      <td>3.362950</td>\n",
       "      <td>44.255240</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-3a028d21-3d75-4a61-96d7-4bd024c68059')\"\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-3a028d21-3d75-4a61-96d7-4bd024c68059 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-3a028d21-3d75-4a61-96d7-4bd024c68059');\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-13913c95-1306-4ddf-ae50-b4d16d5bf4c3\">\n",
       "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-13913c95-1306-4ddf-ae50-b4d16d5bf4c3')\"\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-13913c95-1306-4ddf-ae50-b4d16d5bf4c3 button');\n",
       "          quickchartButtonEl.style.display =\n",
       "            google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "        })();\n",
       "      </script>\n",
       "    </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "text/plain": [
       "  unique_id                  ds    TimeGPT  hour_sin   hour_cos  dayofweek_0  \\\n",
       "0        DE 2017-12-21 00:00:00  35.801600 -3.609636  -9.003666     0.805974   \n",
       "1        DE 2017-12-21 01:00:00  34.419390 -3.824628 -10.493365     0.714771   \n",
       "2        DE 2017-12-21 02:00:00  32.892105 -4.959243 -10.772224     0.712402   \n",
       "3        DE 2017-12-21 03:00:00  32.727295 -5.161374 -10.812295     0.771099   \n",
       "4        DE 2017-12-21 04:00:00  34.121994 -3.687167 -11.353230     0.846524   \n",
       "\n",
       "   dayofweek_1  dayofweek_2  dayofweek_3  dayofweek_4  dayofweek_5  \\\n",
       "0    -0.424078    -0.343238    -0.428668    -0.055370     1.462214   \n",
       "1    -0.400898    -0.282606    -0.331269    -0.115753     1.539153   \n",
       "2    -0.439891    -0.261654    -0.207954    -0.191223     1.481960   \n",
       "3    -0.417504    -0.262543    -0.146066    -0.258350     1.578070   \n",
       "4    -0.387008    -0.278475    -0.169525    -0.255498     1.788180   \n",
       "\n",
       "   dayofweek_6  base_value  \n",
       "0     3.295479   44.102590  \n",
       "1     3.245723   44.368263  \n",
       "2     3.206257   44.323673  \n",
       "3     3.268950   44.167310  \n",
       "4     3.362950   44.255240  "
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "shap_df = nixtla_client.feature_contributions\n",
    "shap_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "sAHUKfO5L_9W",
   "metadata": {},
   "source": [
    "As we can see above, the hour has now gotten encoded using the sine and cosine features instead of the one hot encoding."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8cVTmHzAHmLS",
   "metadata": {},
   "outputs": [],
   "source": [
    "fcst_timegpt_dt_custom.rename(columns={\"TimeGPT\": \"fcst_timegpt_dt_custom\"}, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "uz3kVkBG56zt",
   "metadata": {},
   "source": [
    "### Step 4: Compare Results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cXJ2A3yD5_zF",
   "metadata": {},
   "source": [
    "#### Visual Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9dffUD9ZMIp-",
   "metadata": {},
   "source": [
    "Let's compare the results visually first. For this, we will merge all the forecasts together. This is why we had renamed the forecast columns above so that we can distinguish the forecasts generated by the different methods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "FeKMfPuLuWrd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-f292f352-7591-4d2d-b3ea-5cbec2122a74\" 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>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>TimeGPT_no_dt</th>\n",
       "      <th>fcst_timegpt_dt</th>\n",
       "      <th>fcst_timegpt_dt_custom</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 00:00:00</td>\n",
       "      <td>34.340740</td>\n",
       "      <td>35.248108</td>\n",
       "      <td>35.801600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 01:00:00</td>\n",
       "      <td>34.376488</td>\n",
       "      <td>34.400800</td>\n",
       "      <td>34.419390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 02:00:00</td>\n",
       "      <td>32.215570</td>\n",
       "      <td>33.175526</td>\n",
       "      <td>32.892105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 03:00:00</td>\n",
       "      <td>34.485695</td>\n",
       "      <td>33.205390</td>\n",
       "      <td>32.727295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 04:00:00</td>\n",
       "      <td>34.359673</td>\n",
       "      <td>34.689583</td>\n",
       "      <td>34.121994</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-f292f352-7591-4d2d-b3ea-5cbec2122a74')\"\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-f292f352-7591-4d2d-b3ea-5cbec2122a74 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-f292f352-7591-4d2d-b3ea-5cbec2122a74');\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-bd0ed602-08e2-4d2e-9ae0-edc01ef3925a\">\n",
       "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-bd0ed602-08e2-4d2e-9ae0-edc01ef3925a')\"\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-bd0ed602-08e2-4d2e-9ae0-edc01ef3925a button');\n",
       "          quickchartButtonEl.style.display =\n",
       "            google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "        })();\n",
       "      </script>\n",
       "    </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "text/plain": [
       "  unique_id                  ds  TimeGPT_no_dt  fcst_timegpt_dt  \\\n",
       "0        DE 2017-12-21 00:00:00      34.340740        35.248108   \n",
       "1        DE 2017-12-21 01:00:00      34.376488        34.400800   \n",
       "2        DE 2017-12-21 02:00:00      32.215570        33.175526   \n",
       "3        DE 2017-12-21 03:00:00      34.485695        33.205390   \n",
       "4        DE 2017-12-21 04:00:00      34.359673        34.689583   \n",
       "\n",
       "   fcst_timegpt_dt_custom  \n",
       "0               35.801600  \n",
       "1               34.419390  \n",
       "2               32.892105  \n",
       "3               32.727295  \n",
       "4               34.121994  "
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_fcst = (\n",
    "    fcst_timegpt_no_dt\n",
    "    .merge(\n",
    "        fcst_timegpt_dt,\n",
    "        on=['unique_id', 'ds'],\n",
    "    )\n",
    "    .merge(\n",
    "        fcst_timegpt_dt_custom,\n",
    "        on=['unique_id', 'ds'],\n",
    "    )\n",
    ")\n",
    "all_fcst.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "Zdqm32ytuHKI",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x350 with 1 Axes>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nixtla_client.plot(df_sub, all_fcst)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "xLAUyJwE6S2V",
   "metadata": {},
   "source": [
    "Visually looking at the results shows that the forecast with the datetime features is closer to the actuals as compared to the forecast without the datetime features."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "yc-K4RnZ6D-N",
   "metadata": {},
   "source": [
    "#### Metric Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "FmYlHLpLMWCe",
   "metadata": {},
   "source": [
    "Next, let's compare the forecast with the actual data quantitatively. We will use two common metrics - `MAE` and `RMSE` for this purpose."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bQetHVkEuOSI",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-d000cb4d-ee1f-4b34-86f1-315cab4ea634\" 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>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>y</th>\n",
       "      <th>TimeGPT_no_dt</th>\n",
       "      <th>fcst_timegpt_dt</th>\n",
       "      <th>fcst_timegpt_dt_custom</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 00:00:00</td>\n",
       "      <td>33.09</td>\n",
       "      <td>34.340740</td>\n",
       "      <td>35.248108</td>\n",
       "      <td>35.801600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 01:00:00</td>\n",
       "      <td>35.26</td>\n",
       "      <td>34.376488</td>\n",
       "      <td>34.400800</td>\n",
       "      <td>34.419390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 02:00:00</td>\n",
       "      <td>31.88</td>\n",
       "      <td>32.215570</td>\n",
       "      <td>33.175526</td>\n",
       "      <td>32.892105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 03:00:00</td>\n",
       "      <td>33.04</td>\n",
       "      <td>34.485695</td>\n",
       "      <td>33.205390</td>\n",
       "      <td>32.727295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>DE</td>\n",
       "      <td>2017-12-21 04:00:00</td>\n",
       "      <td>33.60</td>\n",
       "      <td>34.359673</td>\n",
       "      <td>34.689583</td>\n",
       "      <td>34.121994</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-d000cb4d-ee1f-4b34-86f1-315cab4ea634')\"\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-d000cb4d-ee1f-4b34-86f1-315cab4ea634 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-d000cb4d-ee1f-4b34-86f1-315cab4ea634');\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-6b507356-5d72-403c-8c5e-b01e40d8fc9a\">\n",
       "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-6b507356-5d72-403c-8c5e-b01e40d8fc9a')\"\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-6b507356-5d72-403c-8c5e-b01e40d8fc9a button');\n",
       "          quickchartButtonEl.style.display =\n",
       "            google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "        })();\n",
       "      </script>\n",
       "    </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "text/plain": [
       "  unique_id                  ds      y  TimeGPT_no_dt  fcst_timegpt_dt  \\\n",
       "0        DE 2017-12-21 00:00:00  33.09      34.340740        35.248108   \n",
       "1        DE 2017-12-21 01:00:00  35.26      34.376488        34.400800   \n",
       "2        DE 2017-12-21 02:00:00  31.88      32.215570        33.175526   \n",
       "3        DE 2017-12-21 03:00:00  33.04      34.485695        33.205390   \n",
       "4        DE 2017-12-21 04:00:00  33.60      34.359673        34.689583   \n",
       "\n",
       "   fcst_timegpt_dt_custom  \n",
       "0               35.801600  \n",
       "1               34.419390  \n",
       "2               32.892105  \n",
       "3               32.727295  \n",
       "4               34.121994  "
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_fcst_with_actuals = (\n",
    "    df_test[[\"unique_id\", \"ds\", \"y\"]]\n",
    "    .merge(all_fcst, on=['unique_id', 'ds'])\n",
    ")\n",
    "all_fcst_with_actuals.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8JvO1Ziluw3U",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-fd8fe664-4e0c-4a21-8f2a-37ceae451573\" 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>unique_id</th>\n",
       "      <th>metric</th>\n",
       "      <th>TimeGPT_no_dt</th>\n",
       "      <th>fcst_timegpt_dt</th>\n",
       "      <th>fcst_timegpt_dt_custom</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DE</td>\n",
       "      <td>mae</td>\n",
       "      <td>27.527012</td>\n",
       "      <td>21.644545</td>\n",
       "      <td>21.139603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DE</td>\n",
       "      <td>rmse</td>\n",
       "      <td>33.478168</td>\n",
       "      <td>28.099654</td>\n",
       "      <td>27.616988</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-fd8fe664-4e0c-4a21-8f2a-37ceae451573')\"\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-fd8fe664-4e0c-4a21-8f2a-37ceae451573 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-fd8fe664-4e0c-4a21-8f2a-37ceae451573');\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-bd409960-823d-4751-ba41-007e4a82274c\">\n",
       "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-bd409960-823d-4751-ba41-007e4a82274c')\"\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-bd409960-823d-4751-ba41-007e4a82274c button');\n",
       "          quickchartButtonEl.style.display =\n",
       "            google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "        })();\n",
       "      </script>\n",
       "    </div>\n",
       "\n",
       "  <div id=\"id_cb215978-15ad-4587-a561-01a9cfa7404c\">\n",
       "    <style>\n",
       "      .colab-df-generate {\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-generate:hover {\n",
       "        background-color: #E2EBFA;\n",
       "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "        fill: #174EA6;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate {\n",
       "        background-color: #3B4455;\n",
       "        fill: #D2E3FC;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate: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",
       "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('evaluation')\"\n",
       "            title=\"Generate code using this dataframe.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "    <script>\n",
       "      (() => {\n",
       "      const buttonEl =\n",
       "        document.querySelector('#id_cb215978-15ad-4587-a561-01a9cfa7404c button.colab-df-generate');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      buttonEl.onclick = () => {\n",
       "        google.colab.notebook.generateWithVariable('evaluation');\n",
       "      }\n",
       "      })();\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "text/plain": [
       "  unique_id metric  TimeGPT_no_dt  fcst_timegpt_dt  fcst_timegpt_dt_custom\n",
       "0        DE    mae      27.527012        21.644545               21.139603\n",
       "1        DE   rmse      33.478168        28.099654               27.616988"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics = [mae, rmse]\n",
    "\n",
    "evaluation = evaluate(\n",
    "    all_fcst_with_actuals,\n",
    "    metrics=metrics,\n",
    ")\n",
    "evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "XwNZVPeT6Ize",
   "metadata": {},
   "source": [
    "As we can see, the addition of the datetime features improved the forecasting \n",
    "metrics compared to the baseline model created without these features."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "LxTp9Sxe6fsw",
   "metadata": {},
   "source": [
    "## Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "BgC91itmMmId",
   "metadata": {},
   "source": [
    "As demonstrated in this tutorial\n",
    "\n",
    "1. Providing datetime features to the model during forecasting can improve the metrics substantially.\n",
    "2. However, users must be careful of the cardinality of the features after datetime features have been added. If the feature cardinality is too large for the dataset, it may lead to overfitting.\n",
    "3. In case of high cardinality, users may consider a custom encoding approach as demonstrated."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "python3",
   "language": "python",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}