{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from statsforecast import StatsForecast\n",
    "from statsforecast.models import AutoARIMA, AutoETS, AutoCES, AutoTheta, Naive, SeasonalNaive\n",
    "from utilsforecast.plotting import plot_series\n",
    "from utilsforecast.evaluation import evaluate\n",
    "from utilsforecast.losses import mae, rmse, smape, mase\n",
    "\n",
    "from utils import plot_metric_bar_multi, evaluate_cv, get_best_model_forecast"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1) Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load in M4 dataset\n",
    "Y_train_df = pd.read_csv('https://auto-arima-results.s3.amazonaws.com/M4-Hourly.csv')\n",
    "Y_test_df = pd.read_csv('https://auto-arima-results.s3.amazonaws.com/M4-Hourly-test.csv')\n",
    "Y_train_df['ds'] = pd.to_datetime('2024-01-01') + pd.to_timedelta(Y_train_df['ds'], unit='h')\n",
    "Y_test_df['ds'] = pd.to_datetime('2024-01-01') + pd.to_timedelta(Y_test_df['ds'], unit='h')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# randomly select 8 series\n",
    "n_series = 8\n",
    "uids = Y_train_df['unique_id'].drop_duplicates().sample(8, random_state=23).values\n",
    "df_train = Y_train_df.query('unique_id in @uids')\n",
    "df_test = Y_test_df.query('unique_id in @uids')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define Error Metrics\n",
    "from functools import partial\n",
    "hourly_mase = partial(mase, seasonality=24)\n",
    "metrics = [hourly_mase, rmse, smape]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig = plot_series(df_train, df_test.rename(columns={\"y\": \"actual\"}), max_ids=4)\n",
    "fig.savefig('/Users/khuyentran/nixtla_blog/images/statsforecast-automatic-model-selection/selected-series.svg', format='svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2) Baseline: Naive & SeasonalNaive\n",
    "Before diving into more sophisticated models, we begin with two classical and interpretable statistical baselines:\n",
    "\n",
    "**Naive model**: always predicts the last observed value.\n",
    "\n",
    "**SeasonalNaive model**: predicts that each point will repeat the value observed one season ago (e.g.,the same hour yesterday for hourly data)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:384: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:440: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n"
     ]
    }
   ],
   "source": [
    "# Use Naive and SeasonalNaive model as a baseline\n",
    "sf_base = StatsForecast(\n",
    "    models=[Naive(), SeasonalNaive(season_length=24)], \n",
    "    freq='H', \n",
    "    n_jobs=-1\n",
    "    )\n",
    "\n",
    "# Make baseline prediction \n",
    "fcst_base = sf_base.forecast(df = df_train, h = 48)\n",
    "\n",
    "# Evaluate with test dataset\n",
    "eval_base = df_test.merge(fcst_base, on = ['unique_id', 'ds'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plotting the baseline result\n",
    "fig = plot_series(df_train, eval_base, max_ids=4, max_insample_length=5*24)\n",
    "fig.savefig('/Users/khuyentran/nixtla_blog/images/statsforecast-automatic-model-selection/baseline-forecasts.svg', format='svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Naive</th>\n",
       "      <th>SeasonalNaive</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>metric</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mase</th>\n",
       "      <td>8.029174</td>\n",
       "      <td>0.993421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rmse</th>\n",
       "      <td>179.520049</td>\n",
       "      <td>66.529088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>smape</th>\n",
       "      <td>0.252074</td>\n",
       "      <td>0.065754</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Naive  SeasonalNaive\n",
       "metric                           \n",
       "mase      8.029174       0.993421\n",
       "rmse    179.520049      66.529088\n",
       "smape     0.252074       0.065754"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluate performance \n",
    "metrics_base = evaluate(\n",
    "    df=eval_base,\n",
    "    train_df = df_train,\n",
    "    metrics=metrics,\n",
    "    agg_fn='mean',\n",
    ").set_index('metric')\n",
    "metrics_base"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3) StatsForecast Models\n",
    "After establishing the baselines, we now move on to a set of classical statistical forecasting models implemented efficiently in StatsForecast. These models automatically estimate optimal parameters for each series.\n",
    "\n",
    "**AutoARIMA**: Captures autocorrelation patterns and adjusts for both trend and seasonality through differencing and autoregressive components.\n",
    "\n",
    "**AutoETS**: Exponential smoothing model that automatically selects additive or multiplicative trend/seasonal components. Excellent for smooth, structured data.\n",
    "\n",
    "**AutoCES**: Complex Exponential Smoothing, which extends ETS by modeling cyclical components in a more flexible way.\n",
    "\n",
    "**AutoTheta**: A robust, fast method derived from the Theta forecasting model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define models\n",
    "models = [\n",
    "    AutoARIMA(season_length=24),      # With seasonality set as 24 for hourly data\n",
    "    AutoETS(season_length=24),\n",
    "    AutoCES(season_length=24),\n",
    "    AutoTheta(season_length=24)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:384: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:440: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n"
     ]
    }
   ],
   "source": [
    "# Initialize statsforecast \n",
    "sf = StatsForecast(\n",
    "    models=models, \n",
    "    freq='H', \n",
    "    n_jobs=-1\n",
    "    )\n",
    "\n",
    "# Autofit the stats models and make prediction all in one step.\n",
    "fcst_sf_models = sf.forecast(df = df_train, h = 48, level=[90])\n",
    "\n",
    "# Evaluate with test dataset\n",
    "eval_sf_models = df_test.merge(fcst_sf_models, on = ['unique_id', 'ds'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plot the result\n",
    "fig = plot_series(df_train, eval_sf_models, max_ids=4, max_insample_length=5*24)\n",
    "fig.savefig('/Users/khuyentran/nixtla_blog/images/statsforecast-automatic-model-selection/statsforecast-predictions.svg', format='svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AutoARIMA</th>\n",
       "      <th>AutoETS</th>\n",
       "      <th>CES</th>\n",
       "      <th>AutoTheta</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>metric</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mase</th>\n",
       "      <td>0.801971</td>\n",
       "      <td>1.331669</td>\n",
       "      <td>0.729921</td>\n",
       "      <td>1.868366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rmse</th>\n",
       "      <td>71.456459</td>\n",
       "      <td>122.784231</td>\n",
       "      <td>60.979897</td>\n",
       "      <td>65.105242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>smape</th>\n",
       "      <td>0.063301</td>\n",
       "      <td>0.075775</td>\n",
       "      <td>0.079244</td>\n",
       "      <td>0.076261</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        AutoARIMA     AutoETS        CES  AutoTheta\n",
       "metric                                             \n",
       "mase     0.801971    1.331669   0.729921   1.868366\n",
       "rmse    71.456459  122.784231  60.979897  65.105242\n",
       "smape    0.063301    0.075775   0.079244   0.076261"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluation Metrics\n",
    "metrics_sf_models = evaluate(\n",
    "    df= eval_sf_models,\n",
    "    metrics=metrics,\n",
    "    train_df = df_train,\n",
    "    agg_fn='mean',\n",
    ").set_index('metric')\n",
    "metrics_sf_models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plot_metric_bar_multi(dfs = [metrics_sf_models, metrics_base], metric='mase')\n",
    "fig.savefig('/Users/khuyentran/nixtla_blog/images/statsforecast-automatic-model-selection/model-comparison-bar-chart.svg', format='svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4) Cross-Validation with a Rolling Window\n",
    "In this section, we use rolling-origin cross-validation to select the best model for each time series. Nixtla’s implementation of cross-validation ensures the temporal order is respected, avoiding data leakage and producing a more stable and streamlined evaluation process.\n",
    "\n",
    "**How it works:**  \n",
    "1. Start with an initial training window and forecast the next *h* steps.  \n",
    "2. Slide the window forward by *step_size* and repeat.  \n",
    "3. Compute error metrics for each window and model.  \n",
    "4. Select the model with the **lowest average error** for each series.\n",
    "\n",
    "![Rolling-window cross-validation](https://raw.githubusercontent.com/Nixtla/statsforecast/main/nbs/imgs/ChainedWindows.gif)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/fs/__init__.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.\n",
      "  __import__(\"pkg_resources\").declare_namespace(__name__)  # type: ignore\n"
     ]
    }
   ],
   "source": [
    "# Run cross-validation with 2 rolling windows of 24 steps each\n",
    "cv_df = sf.cross_validation(\n",
    "    df=df_train,\n",
    "    h=24,          # forecast horizon\n",
    "    step_size=24,  # roll the window forward by 24 steps each time\n",
    "    n_windows=2    # number of evaluation windows\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>best_statsforecast_model</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AutoARIMA</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AutoETS</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AutoTheta</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CES</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  best_statsforecast_model  count\n",
       "0                AutoARIMA      3\n",
       "1                  AutoETS      2\n",
       "2                AutoTheta      2\n",
       "3                      CES      1"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluate model performance using MSE across cross-validation windows\n",
    "evaluation_df = evaluate_cv(cv_df, mae)\n",
    "\n",
    "# Count how many times each model was selected as best\n",
    "evaluation_df['best_statsforecast_model'].value_counts().to_frame().reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After selecting the best model for each series, we visualize the final forecasts alongside the actual test data.\n",
    "\n",
    "Here, the 90% **prediction interval** (shown as the shaded band) provides an estimate of forecast uncertainty\n",
    "reflecting how much variation we can expect in future observations based on past residuals."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Extract the forecasts from the best-performing model for each series\n",
    "best_fcst_sf = get_best_model_forecast(fcst_sf_models, evaluation_df)\n",
    "eval_best_sf = df_test.merge(best_fcst_sf, on = ['unique_id', 'ds'])\n",
    "\n",
    "# Plot forecasts with 90% interval\n",
    "fig = plot_series(df_train, eval_best_sf, level=[90], max_insample_length=5*24, max_ids=4)\n",
    "fig.savefig('/Users/khuyentran/nixtla_blog/images/statsforecast-automatic-model-selection/best-model-forecasts.svg', format='svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x350 with 1 Axes>"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Extract the forecasts from the best-performing model for each series\n",
    "best_fcst_sf = get_best_model_forecast(fcst_sf_models, evaluation_df)\n",
    "eval_best_sf = df_test.merge(best_fcst_sf, on = ['unique_id', 'ds'])\n",
    "\n",
    "# Filter for H25 only\n",
    "df_train_h25 = df_train[df_train['unique_id'] == 'H25']\n",
    "eval_best_sf_h25 = eval_best_sf[eval_best_sf['unique_id'] == 'H25']\n",
    "\n",
    "# Plot forecasts with 90% interval\n",
    "fig = plot_series(df_train_h25, eval_best_sf_h25, level=[90], max_insample_length=5*24)\n",
    "fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>best_statsforecast_model</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>metric</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mase</th>\n",
       "      <td>0.708592</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rmse</th>\n",
       "      <td>66.165094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>smape</th>\n",
       "      <td>0.057047</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        best_statsforecast_model\n",
       "metric                          \n",
       "mase                    0.708592\n",
       "rmse                   66.165094\n",
       "smape                   0.057047"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics_sf_best = evaluate(\n",
    "    df=eval_best_sf,\n",
    "    train_df = df_train,\n",
    "    metrics=metrics,\n",
    "    agg_fn='mean',\n",
    ").set_index('metric')\n",
    "metrics_sf_best"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plot_metric_bar_multi(dfs = [metrics_sf_models, metrics_base, metrics_sf_best], metric='mase')\n",
    "fig.savefig('/Users/khuyentran/nixtla_blog/images/statsforecast-automatic-model-selection/complete-comparison.svg', format='svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5) Make Forecast with TimeGPT\n",
    "Produce fast, accurate forecasts using TimeGPT, then compare against our statistical baselines and StatsForecast models.\n",
    "\n",
    "**Why TimeGPT?**\n",
    "\n",
    "- Strong out-of-the-box accuracy with minimal tuning\n",
    "\n",
    "- Handles trend/seasonality/holidays automatically\n",
    "\n",
    "- Scales to many series with simple APIs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5 a) Forecast with TimeGPT-1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import necessary packages\n",
    "import os\n",
    "from dotenv import load_dotenv\n",
    "from nixtla import NixtlaClient\n",
    "\n",
    "# Load environment variables from .env\n",
    "load_dotenv(override=True)\n",
    "api_key = os.getenv('NIXTLA_API_KEY')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialize the client\n",
    "nixtla_client = NixtlaClient(api_key=api_key)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:nixtla.nixtla_client:Validating inputs...\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/preprocessing.py:131: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  offset = pd.tseries.frequencies.to_offset(freq)\n",
      "INFO:nixtla.nixtla_client:Preprocessing dataframes...\n",
      "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",
      "INFO:nixtla.nixtla_client:Restricting input...\n",
      "INFO:nixtla.nixtla_client:Calling Forecast Endpoint...\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:384: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:440: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>TimeGPT</th>\n",
       "      <th>TimeGPT-hi-80</th>\n",
       "      <th>TimeGPT-hi-90</th>\n",
       "      <th>TimeGPT-lo-80</th>\n",
       "      <th>TimeGPT-lo-90</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 05:00:00</td>\n",
       "      <td>20.847889</td>\n",
       "      <td>26.581411</td>\n",
       "      <td>27.379568</td>\n",
       "      <td>15.114367</td>\n",
       "      <td>14.316211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 06:00:00</td>\n",
       "      <td>25.407340</td>\n",
       "      <td>34.487167</td>\n",
       "      <td>34.707325</td>\n",
       "      <td>16.327513</td>\n",
       "      <td>16.107355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 07:00:00</td>\n",
       "      <td>49.702620</td>\n",
       "      <td>75.707430</td>\n",
       "      <td>79.375336</td>\n",
       "      <td>23.697813</td>\n",
       "      <td>20.029905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 08:00:00</td>\n",
       "      <td>134.175630</td>\n",
       "      <td>152.434750</td>\n",
       "      <td>153.382050</td>\n",
       "      <td>115.916504</td>\n",
       "      <td>114.969210</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 09:00:00</td>\n",
       "      <td>366.113680</td>\n",
       "      <td>379.054170</td>\n",
       "      <td>381.407230</td>\n",
       "      <td>353.173200</td>\n",
       "      <td>350.820130</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  unique_id                  ds     TimeGPT  TimeGPT-hi-80  TimeGPT-hi-90  \\\n",
       "0      H165 2024-01-30 05:00:00   20.847889      26.581411      27.379568   \n",
       "1      H165 2024-01-30 06:00:00   25.407340      34.487167      34.707325   \n",
       "2      H165 2024-01-30 07:00:00   49.702620      75.707430      79.375336   \n",
       "3      H165 2024-01-30 08:00:00  134.175630     152.434750     153.382050   \n",
       "4      H165 2024-01-30 09:00:00  366.113680     379.054170     381.407230   \n",
       "\n",
       "   TimeGPT-lo-80  TimeGPT-lo-90  \n",
       "0      15.114367      14.316211  \n",
       "1      16.327513      16.107355  \n",
       "2      23.697813      20.029905  \n",
       "3     115.916504     114.969210  \n",
       "4     353.173200     350.820130  "
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Simple zero-shot TimeGPT forecast\n",
    "fcst_timegpt = nixtla_client.forecast(\n",
    "    df=df_train,\n",
    "    h=48,         # forecast horizon (next 48 hours)\n",
    "    freq='H',     # hourly frequency\n",
    "    level = ['80', '90']\n",
    ")\n",
    "fcst_timegpt.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:nixtla.nixtla_client:Validating inputs...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/preprocessing.py:131: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  offset = pd.tseries.frequencies.to_offset(freq)\n",
      "INFO:nixtla.nixtla_client:Preprocessing dataframes...\n",
      "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",
      "INFO:nixtla.nixtla_client:Calling Forecast Endpoint...\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:384: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:440: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>TimeGPT</th>\n",
       "      <th>TimeGPT-hi-80</th>\n",
       "      <th>TimeGPT-hi-90</th>\n",
       "      <th>TimeGPT-lo-80</th>\n",
       "      <th>TimeGPT-lo-90</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 05:00:00</td>\n",
       "      <td>20.847889</td>\n",
       "      <td>26.581411</td>\n",
       "      <td>27.379568</td>\n",
       "      <td>15.114367</td>\n",
       "      <td>14.316211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 06:00:00</td>\n",
       "      <td>25.407340</td>\n",
       "      <td>34.487167</td>\n",
       "      <td>34.707325</td>\n",
       "      <td>16.327513</td>\n",
       "      <td>16.107355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 07:00:00</td>\n",
       "      <td>49.702620</td>\n",
       "      <td>75.707430</td>\n",
       "      <td>79.375336</td>\n",
       "      <td>23.697813</td>\n",
       "      <td>20.029905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 08:00:00</td>\n",
       "      <td>134.175630</td>\n",
       "      <td>152.434750</td>\n",
       "      <td>153.382050</td>\n",
       "      <td>115.916504</td>\n",
       "      <td>114.969210</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>H165</td>\n",
       "      <td>2024-01-30 09:00:00</td>\n",
       "      <td>366.113680</td>\n",
       "      <td>379.054170</td>\n",
       "      <td>381.407230</td>\n",
       "      <td>353.173200</td>\n",
       "      <td>350.820130</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  unique_id                  ds     TimeGPT  TimeGPT-hi-80  TimeGPT-hi-90  \\\n",
       "0      H165 2024-01-30 05:00:00   20.847889      26.581411      27.379568   \n",
       "1      H165 2024-01-30 06:00:00   25.407340      34.487167      34.707325   \n",
       "2      H165 2024-01-30 07:00:00   49.702620      75.707430      79.375336   \n",
       "3      H165 2024-01-30 08:00:00  134.175630     152.434750     153.382050   \n",
       "4      H165 2024-01-30 09:00:00  366.113680     379.054170     381.407230   \n",
       "\n",
       "   TimeGPT-lo-80  TimeGPT-lo-90  \n",
       "0      15.114367      14.316211  \n",
       "1      16.327513      16.107355  \n",
       "2      23.697813      20.029905  \n",
       "3     115.916504     114.969210  \n",
       "4     353.173200     350.820130  "
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Add finetune steps to make it more accurate\n",
    "fcst_timegpt_ft = nixtla_client.forecast(\n",
    "    df=df_train,\n",
    "    h=48,\n",
    "    freq='H',\n",
    "    level = ['80', '90'],\n",
    "    finetune_steps = 10     #Here we use 100 finetune steps. Choose the value base on your usecase\n",
    ")\n",
    "fcst_timegpt.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5 b) Forecast with the newly launched TimeGPT-2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialize nixtla_clienct with TimeGPT-2 credentials\n",
    "nixtla_client = NixtlaClient(api_key=api_key, base_url=\"https://api-preview.nixtla.io\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:nixtla.nixtla_client:Validating inputs...\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/preprocessing.py:131: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  offset = pd.tseries.frequencies.to_offset(freq)\n",
      "INFO:nixtla.nixtla_client:Preprocessing dataframes...\n",
      "INFO:nixtla.nixtla_client:Querying model metadata...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:nixtla.nixtla_client:Restricting input...\n",
      "INFO:nixtla.nixtla_client:Calling Forecast Endpoint...\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:384: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n",
      "/Users/khuyentran/nixtla_blog/.venv/lib/python3.11/site-packages/utilsforecast/processing.py:440: FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead.\n",
      "  freq = pd.tseries.frequencies.to_offset(freq)\n"
     ]
    }
   ],
   "source": [
    "fcst_timegpt_2 = nixtla_client.forecast(\n",
    "    df=df_train,\n",
    "    h=48,\n",
    "    freq='H',\n",
    "    level = ['80', '90'],\n",
    "    model = 'timegpt-2'\n",
    ")\n",
    "eval_tgpt_2 = df_test.merge(fcst_timegpt_2, on = ['unique_id', 'ds'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>TimeGPT_zero_shot</th>\n",
       "      <th>TimeGPT_finetuned</th>\n",
       "      <th>TimeGPT_2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>metric</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>mase</th>\n",
       "      <td>1.118836</td>\n",
       "      <td>0.831705</td>\n",
       "      <td>0.427980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rmse</th>\n",
       "      <td>53.229181</td>\n",
       "      <td>55.354640</td>\n",
       "      <td>27.017126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>smape</th>\n",
       "      <td>0.056424</td>\n",
       "      <td>0.071633</td>\n",
       "      <td>0.035508</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        TimeGPT_zero_shot  TimeGPT_finetuned  TimeGPT_2\n",
       "metric                                                 \n",
       "mase             1.118836           0.831705   0.427980\n",
       "rmse            53.229181          55.354640  27.017126\n",
       "smape            0.056424           0.071633   0.035508"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics_tgpt = evaluate(\n",
    "    df=df_test.merge(fcst_timegpt.rename(columns={'TimeGPT': 'TimeGPT_zero_shot'}), on = ['unique_id', 'ds'])\n",
    "              .merge(fcst_timegpt_ft.rename(columns={'TimeGPT': 'TimeGPT_finetuned'}), on = ['unique_id', 'ds'])\n",
    "              .merge(fcst_timegpt_2.rename(columns={'TimeGPT': 'TimeGPT_2'}), on = ['unique_id', 'ds']),\n",
    "    train_df = df_train,\n",
    "    metrics=metrics,\n",
    "    agg_fn='mean',\n",
    ").set_index('metric')\n",
    "metrics_tgpt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig = nixtla_client.plot(df_train, eval_tgpt_2, level=['80', '90'], max_insample_length=5*24, max_ids=4)\n",
    "fig.savefig('/Users/khuyentran/nixtla_blog/images/statsforecast-automatic-model-selection/timegpt-2-forecasts.svg', format='svg', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plot_metric_bar_multi(dfs = [metrics_base, metrics_sf_best, metrics_tgpt], metric='mase')\n",
    "fig.savefig('/Users/khuyentran/nixtla_blog/images/statsforecast-automatic-model-selection/final-comparison-timegpt.svg', format='svg', bbox_inches='tight')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}