{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Forecasting with Chronos\n",
    "\n",
    "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/autogluon/autogluon/blob/stable/docs/tutorials/timeseries/forecasting-chronos.ipynb)\n",
    "[![Open In SageMaker Studio Lab](https://studiolab.sagemaker.aws/studiolab.svg)](https://studiolab.sagemaker.aws/import/github/autogluon/autogluon/blob/stable/docs/tutorials/timeseries/forecasting-chronos.ipynb)\n",
    "\n",
    "\n",
    "AutoGluon-TimeSeries (AG-TS) includes [Chronos](https://github.com/amazon-science/chronos-forecasting) family of forecasting models. Chronos models are pretrained on a large collection of real & synthetic time series data, which enables them to make accurate forecasts on new data out of the box.\n",
    "\n",
    "AG-TS provides a robust and easy way to use Chronos through the familiar `TimeSeriesPredictor` API. This tutorial describes how to \n",
    "- Use Chronos models in **zero-shot** mode to make forecasts without any dataset-specific training\n",
    "- **Fine-tune** Chronos models on custom data to improve the accuracy\n",
    "- Handle **covariates & static features** by combining Chronos with a tabular regression model\n",
    "\n",
    ":::{note}\n",
    "\n",
    "**New in v1.2:** AutoGluon now features Chronos-Bolt⚡️ — new, more accurate, and up to 250x faster Chronos models.\n",
    "\n",
    ":::"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": [
     "remove-cell",
     "skip-execution"
    ]
   },
   "outputs": [],
   "source": [
    "# We use uv for faster installation\n",
    "!pip install uv\n",
    "!uv pip install -q autogluon.timeseries --system\n",
    "!uv pip uninstall -q torchaudio torchvision torchtext --system # fix incompatible package versions on Colab"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Getting started with Chronos\n",
    "\n",
    "Being a pretrained model for zero-shot forecasting, Chronos is different from other models available in AG-TS. \n",
    "Specifically, Chronos models do not really `fit` time series data. However, when `predict` is called, they carry out a relatively more expensive computation that scales linearly with the number of time series in the dataset. In this aspect, they behave like local statistical models such as ETS or ARIMA, where all computation happens during inference. \n",
    "\n",
    "AutoGluon supports both the original Chronos models (e.g., [`chronos-t5-large`](https://huggingface.co/autogluon/chronos-t5-large)), as well as the new, more accurate and up to 250x faster Chronos-Bolt⚡ models (e.g., [`chronos-bolt-base`](https://huggingface.co/autogluon/chronos-bolt-base)). \n",
    "\n",
    "The easiest way to get started with Chronos is through the model-specific presets. \n",
    "\n",
    "- **(recommended)** The new, fast Chronos-Bolt️ models can be accessed using the `\"bolt_tiny\"`, `\"bolt_mini\"`, `\"bolt_small\"` and `\"bolt_base\"` presets.\n",
    "- The original Chronos models can be accessed using the `\"chronos_tiny\"`, `\"chronos_mini\"`, `\"chronos_small\"`, `\"chronos_base\"` and `\"chronos_large\"` presets.\n",
    "\n",
    "Note that the original Chronos models of size `small` and above require a GPU to run, while all Chronos-Bolt models can be run both on a CPU and a GPU.\n",
    "\n",
    "Alternatively, Chronos can be combined with other time series models using presets `\"medium_quality\"`, `\"high_quality\"` and `\"best_quality\"`. More details about these presets are available in the documentation for [`TimeSeriesPredictor.fit`](https://auto.gluon.ai/stable/api/autogluon.timeseries.TimeSeriesPredictor.fit.html)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Zero-shot forecasting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's work with a subset of the [Australian Electricity Demand dataset](https://zenodo.org/records/4659727) to see Chronos-Bolt in action.\n",
    "\n",
    "First, we load the dataset as a [TimeSeriesDataFrame](https://auto.gluon.ai/stable/api/autogluon.timeseries.TimeSeriesDataFrame.html)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from autogluon.timeseries import TimeSeriesDataFrame, TimeSeriesPredictor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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></th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>item_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">T000000</th>\n",
       "      <th>2013-03-10 00:00:00</th>\n",
       "      <td>5207.959961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-03-10 00:30:00</th>\n",
       "      <td>5002.275879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-03-10 01:00:00</th>\n",
       "      <td>4747.569824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-03-10 01:30:00</th>\n",
       "      <td>4544.880859</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-03-10 02:00:00</th>\n",
       "      <td>4425.952148</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  target\n",
       "item_id timestamp                       \n",
       "T000000 2013-03-10 00:00:00  5207.959961\n",
       "        2013-03-10 00:30:00  5002.275879\n",
       "        2013-03-10 01:00:00  4747.569824\n",
       "        2013-03-10 01:30:00  4544.880859\n",
       "        2013-03-10 02:00:00  4425.952148"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = TimeSeriesDataFrame.from_path(\n",
    "    \"https://autogluon.s3.amazonaws.com/datasets/timeseries/australian_electricity_subset/test.csv\"\n",
    ")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we create the [TimeSeriesPredictor](https://auto.gluon.ai/stable/api/autogluon.timeseries.TimeSeriesPredictor.html) and select the `\"bolt_small\"` presets to use the Chronos-Bolt (Small, 48M) model in zero-shot mode."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sorting the dataframe index before generating the train/test split.\n",
      "Beginning AutoGluon training...\n",
      "AutoGluon will save models to '/local/home/shchuro/workspace/autogluon/docs/tutorials/timeseries/AutogluonModels/ag-20241126_091557'\n",
      "=================== System Info ===================\n",
      "AutoGluon Version:  1.1.2b20241122\n",
      "Python Version:     3.11.10\n",
      "Operating System:   Linux\n",
      "Platform Machine:   x86_64\n",
      "Platform Version:   #1 SMP Wed Oct 23 01:22:11 UTC 2024\n",
      "CPU Count:          32\n",
      "GPU Count:          4\n",
      "Memory Avail:       230.68 GB / 239.85 GB (96.2%)\n",
      "Disk Space Avail:   563.58 GB / 1968.52 GB (28.6%)\n",
      "===================================================\n",
      "Setting presets to: bolt_small\n",
      "\n",
      "Fitting with arguments:\n",
      "{'enable_ensemble': True,\n",
      " 'eval_metric': WQL,\n",
      " 'hyperparameters': {'Chronos': {'model_path': 'bolt_small'}},\n",
      " 'known_covariates_names': [],\n",
      " 'num_val_windows': 1,\n",
      " 'prediction_length': 48,\n",
      " 'quantile_levels': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],\n",
      " 'random_seed': 123,\n",
      " 'refit_every_n_windows': 1,\n",
      " 'refit_full': False,\n",
      " 'skip_model_selection': True,\n",
      " 'target': 'target',\n",
      " 'verbosity': 2}\n",
      "\n",
      "Inferred time series frequency: '30min'\n",
      "Provided train_data has 172800 rows, 5 time series. Median time series length is 34560 (min=34560, max=34560). \n",
      "\n",
      "Provided data contains following columns:\n",
      "\ttarget: 'target'\n",
      "\n",
      "AutoGluon will gauge predictive performance using evaluation metric: 'WQL'\n",
      "\tThis metric's sign has been flipped to adhere to being higher_is_better. The metric score can be multiplied by -1 to get the metric value.\n",
      "===================================================\n",
      "\n",
      "Starting training. Start time is 2024-11-26 09:16:01\n",
      "Models that will be trained: ['Chronos[bolt_small]']\n",
      "Training timeseries model Chronos[bolt_small]. \n",
      "\t2.01    s     = Training runtime\n",
      "Training complete. Models trained: ['Chronos[bolt_small]']\n",
      "Total runtime: 2.01 s\n",
      "Best model: Chronos[bolt_small]\n"
     ]
    }
   ],
   "source": [
    "prediction_length = 48\n",
    "train_data, test_data = data.train_test_split(prediction_length)\n",
    "\n",
    "predictor = TimeSeriesPredictor(prediction_length=prediction_length).fit(\n",
    "    train_data, presets=\"bolt_small\",\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As promised, Chronos does not take any time to `fit`. The `fit` call merely serves as a proxy for the `TimeSeriesPredictor` to do some of its chores under the hood, such as inferring the frequency of time series and saving the predictor's state to disk. \n",
    "\n",
    "Let's use the `predict` method to generate forecasts, and the `plot` method to visualize them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Model not specified in predict, will default to the model with the best validation score: Chronos[bolt_small]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x350 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictions = predictor.predict(train_data)\n",
    "predictor.plot(\n",
    "    data=data,\n",
    "    predictions=predictions,\n",
    "    item_ids=data.item_ids[:2],\n",
    "    max_history_length=200,\n",
    ");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fine-tuning \n",
    "\n",
    "We have seen above how Chronos models can produce forecasts in zero-shot mode. AutoGluon also makes it easy to fine-tune Chronos models on a specific dataset to maximize the predictive accuracy.\n",
    "\n",
    "The following snippet specifies two settings for the Chronos-Bolt ️(Small) model: zero-shot and fine-tuned. `TimeSeriesPredictor` will perform a lightweight fine-tuning of the pretrained model on the provided training data. We add name suffixes to easily identify the zero-shot and fine-tuned versions of the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Beginning AutoGluon training... Time limit = 60s\n",
      "AutoGluon will save models to '/local/home/shchuro/workspace/autogluon/docs/tutorials/timeseries/AutogluonModels/ag-20241126_091607'\n",
      "=================== System Info ===================\n",
      "AutoGluon Version:  1.1.2b20241122\n",
      "Python Version:     3.11.10\n",
      "Operating System:   Linux\n",
      "Platform Machine:   x86_64\n",
      "Platform Version:   #1 SMP Wed Oct 23 01:22:11 UTC 2024\n",
      "CPU Count:          32\n",
      "GPU Count:          4\n",
      "Memory Avail:       229.82 GB / 239.85 GB (95.8%)\n",
      "Disk Space Avail:   563.58 GB / 1968.52 GB (28.6%)\n",
      "===================================================\n",
      "\n",
      "Fitting with arguments:\n",
      "{'enable_ensemble': False,\n",
      " 'eval_metric': WQL,\n",
      " 'hyperparameters': {'Chronos': [{'ag_args': {'name_suffix': 'ZeroShot'},\n",
      "                                  'model_path': 'bolt_small'},\n",
      "                                 {'ag_args': {'name_suffix': 'FineTuned'},\n",
      "                                  'fine_tune': True,\n",
      "                                  'model_path': 'bolt_small'}]},\n",
      " 'known_covariates_names': [],\n",
      " 'num_val_windows': 1,\n",
      " 'prediction_length': 48,\n",
      " 'quantile_levels': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],\n",
      " 'random_seed': 123,\n",
      " 'refit_every_n_windows': 1,\n",
      " 'refit_full': False,\n",
      " 'skip_model_selection': False,\n",
      " 'target': 'target',\n",
      " 'time_limit': 60,\n",
      " 'verbosity': 2}\n",
      "\n",
      "Inferred time series frequency: '30min'\n",
      "Provided train_data has 172800 rows, 5 time series. Median time series length is 34560 (min=34560, max=34560). \n",
      "\n",
      "Provided data contains following columns:\n",
      "\ttarget: 'target'\n",
      "\n",
      "AutoGluon will gauge predictive performance using evaluation metric: 'WQL'\n",
      "\tThis metric's sign has been flipped to adhere to being higher_is_better. The metric score can be multiplied by -1 to get the metric value.\n",
      "===================================================\n",
      "\n",
      "Starting training. Start time is 2024-11-26 09:16:07\n",
      "Models that will be trained: ['ChronosZeroShot[bolt_small]', 'ChronosFineTuned[bolt_small]']\n",
      "Training timeseries model ChronosZeroShot[bolt_small]. Training for up to 29.9s of the 59.9s of remaining time.\n",
      "\t-0.0417       = Validation score (-WQL)\n",
      "\t0.10    s     = Training runtime\n",
      "\t0.83    s     = Validation (prediction) runtime\n",
      "Training timeseries model ChronosFineTuned[bolt_small]. Training for up to 59.0s of the 59.0s of remaining time.\n",
      "\tSaving fine-tuned model to /local/home/shchuro/workspace/autogluon/docs/tutorials/timeseries/AutogluonModels/ag-20241126_091607/models/ChronosFineTuned[bolt_small]/W0/fine-tuned-ckpt\n",
      "\t-0.0290       = Validation score (-WQL)\n",
      "\t49.36   s     = Training runtime\n",
      "\t0.07    s     = Validation (prediction) runtime\n",
      "Training complete. Models trained: ['ChronosZeroShot[bolt_small]', 'ChronosFineTuned[bolt_small]']\n",
      "Total runtime: 50.38 s\n",
      "Best model: ChronosFineTuned[bolt_small]\n",
      "Best model score: -0.0290\n"
     ]
    }
   ],
   "source": [
    "predictor = TimeSeriesPredictor(prediction_length=prediction_length).fit(\n",
    "    train_data=train_data,\n",
    "    hyperparameters={\n",
    "        \"Chronos\": [\n",
    "            {\"model_path\": \"bolt_small\", \"ag_args\": {\"name_suffix\": \"ZeroShot\"}},\n",
    "            {\"model_path\": \"bolt_small\", \"fine_tune\": True, \"ag_args\": {\"name_suffix\": \"FineTuned\"}},\n",
    "        ]\n",
    "    },\n",
    "    time_limit=60,  # time limit in seconds\n",
    "    enable_ensemble=False,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we used the default fine-tuning configuration for Chronos by only specifying `\"fine_tune\": True`. However, AutoGluon makes it easy to change other parameters for fine-tuning such as the number of steps or learning rate.\n",
    "```python\n",
    "predictor.fit(\n",
    "    ...,\n",
    "    hyperparameters={\"Chronos\": {\"fine_tune\": True, \"fine_tune_lr\": 1e-4, \"fine_tune_steps\": 2000}},\n",
    ")\n",
    "```\n",
    "\n",
    "For the full list of fine-tuning options, see the Chronos documentation in [Forecasting Model Zoo](forecasting-model-zoo.md#autogluon.timeseries.models.ChronosModel).\n",
    "\n",
    "\n",
    "After fitting, we can evaluate the two model variants on the test data and generate a leaderboard."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Additional data provided, testing on additional data. Resulting leaderboard will be sorted according to test score (`score_test`).\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>model</th>\n",
       "      <th>score_test</th>\n",
       "      <th>score_val</th>\n",
       "      <th>pred_time_test</th>\n",
       "      <th>pred_time_val</th>\n",
       "      <th>fit_time_marginal</th>\n",
       "      <th>fit_order</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ChronosFineTuned[bolt_small]</td>\n",
       "      <td>-0.030785</td>\n",
       "      <td>-0.029021</td>\n",
       "      <td>0.541208</td>\n",
       "      <td>0.073925</td>\n",
       "      <td>49.362413</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ChronosZeroShot[bolt_small]</td>\n",
       "      <td>-0.041446</td>\n",
       "      <td>-0.041720</td>\n",
       "      <td>0.859698</td>\n",
       "      <td>0.825092</td>\n",
       "      <td>0.098496</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          model  score_test  score_val  pred_time_test  \\\n",
       "0  ChronosFineTuned[bolt_small]   -0.030785  -0.029021        0.541208   \n",
       "1   ChronosZeroShot[bolt_small]   -0.041446  -0.041720        0.859698   \n",
       "\n",
       "   pred_time_val  fit_time_marginal  fit_order  \n",
       "0       0.073925          49.362413          2  \n",
       "1       0.825092           0.098496          1  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictor.leaderboard(test_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fine-tuning resulted in a more accurate model, as shown by the better `score_test` on the test set.\n",
    "\n",
    "Note that all AutoGluon-TimeSeries models report scores in a \"higher is better\" format, meaning that most forecasting error metrics like WQL are multiplied by -1 when reported."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Incorporating the covariates"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Chronos️ is a univariate model, meaning it relies solely on the historical data of the target time series for making predictions. However, in real-world scenarios, additional exogenous information related to the target series (e.g., holidays, promotions) is often available. Leveraging this information when making predictions can improve forecast accuracy. \n",
    "\n",
    "AG-TS now features covariate regressors that can be combined with univariate models like Chronos-Bolt to incorporate exogenous information. \n",
    "A `covariate_regressor` in AG-TS is a tabular regression model that is fit on the known covariates and static features to predict the target column at the each time step. The predictions of the covariate regressor are subtracted from the target column, and the univariate model then forecasts the residuals."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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></th>\n",
       "      <th>scaled_price</th>\n",
       "      <th>promotion_email</th>\n",
       "      <th>promotion_homepage</th>\n",
       "      <th>unit_sales</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>item_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">1062_101</th>\n",
       "      <th>2018-01-01</th>\n",
       "      <td>0.879130</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>636.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-08</th>\n",
       "      <td>0.994517</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>123.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-15</th>\n",
       "      <td>1.005513</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>391.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-22</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>339.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-29</th>\n",
       "      <td>0.883309</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>661.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     scaled_price  promotion_email  promotion_homepage  \\\n",
       "item_id  timestamp                                                       \n",
       "1062_101 2018-01-01      0.879130              0.0                 0.0   \n",
       "         2018-01-08      0.994517              0.0                 0.0   \n",
       "         2018-01-15      1.005513              0.0                 0.0   \n",
       "         2018-01-22      1.000000              0.0                 0.0   \n",
       "         2018-01-29      0.883309              0.0                 0.0   \n",
       "\n",
       "                     unit_sales  \n",
       "item_id  timestamp               \n",
       "1062_101 2018-01-01       636.0  \n",
       "         2018-01-08       123.0  \n",
       "         2018-01-15       391.0  \n",
       "         2018-01-22       339.0  \n",
       "         2018-01-29       661.0  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = TimeSeriesDataFrame.from_path(\n",
    "    \"https://autogluon.s3.amazonaws.com/datasets/timeseries/grocery_sales/test.csv\",\n",
    ")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We use a grocery sales dataset to demonstrate how Chronos-Bolt can be combined with a covariate regressor. This dataset includes 3 known covariates: `scaled_price`, `promotion_email` and `promotion_homepage` and the task is to forecast the `unit_sales`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction_length = 8\n",
    "train_data, test_data = data.train_test_split(prediction_length=prediction_length)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The following code fits a TimeSeriesPredictor to forecast `unit_sales` for the next 8 weeks. \n",
    "\n",
    "Note that we have specified the target column we are interested in forecasting and the names of known covariates while constructing the TimeSeriesPredictor. \n",
    "\n",
    "We define two configurations for Chronos-Bolt: \n",
    "- zero-shot configuration that uses only the historical values of `unit_sales` without considering the covariates;\n",
    "- a configuration with a CatBoost regression model as the `covariate_regressor`. Note that we recommend to apply a `target_scaler` when using a covariate regressor. Target scaler ensures that all time series have comparable scales, often leading to better accuracy.\n",
    "\n",
    "Like before, we add suffixes to model names to more easily distinguish them in the leaderboard."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Beginning AutoGluon training... Time limit = 60s\n",
      "AutoGluon will save models to '/local/home/shchuro/workspace/autogluon/docs/tutorials/timeseries/AutogluonModels/ag-20241126_091700'\n",
      "=================== System Info ===================\n",
      "AutoGluon Version:  1.1.2b20241122\n",
      "Python Version:     3.11.10\n",
      "Operating System:   Linux\n",
      "Platform Machine:   x86_64\n",
      "Platform Version:   #1 SMP Wed Oct 23 01:22:11 UTC 2024\n",
      "CPU Count:          32\n",
      "GPU Count:          4\n",
      "Memory Avail:       229.61 GB / 239.85 GB (95.7%)\n",
      "Disk Space Avail:   563.40 GB / 1968.52 GB (28.6%)\n",
      "===================================================\n",
      "\n",
      "Fitting with arguments:\n",
      "{'enable_ensemble': False,\n",
      " 'eval_metric': WQL,\n",
      " 'hyperparameters': {'Chronos': [{'ag_args': {'name_suffix': 'ZeroShot'},\n",
      "                                  'model_path': 'bolt_small'},\n",
      "                                 {'ag_args': {'name_suffix': 'WithRegressor'},\n",
      "                                  'covariate_regressor': 'CAT',\n",
      "                                  'model_path': 'bolt_small',\n",
      "                                  'target_scaler': 'standard'}]},\n",
      " 'known_covariates_names': ['scaled_price',\n",
      "                            'promotion_email',\n",
      "                            'promotion_homepage'],\n",
      " 'num_val_windows': 1,\n",
      " 'prediction_length': 8,\n",
      " 'quantile_levels': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],\n",
      " 'random_seed': 123,\n",
      " 'refit_every_n_windows': 1,\n",
      " 'refit_full': False,\n",
      " 'skip_model_selection': False,\n",
      " 'target': 'unit_sales',\n",
      " 'time_limit': 60,\n",
      " 'verbosity': 2}\n",
      "\n",
      "Inferred time series frequency: 'W-MON'\n",
      "Provided train_data has 7337 rows (NaN fraction=6.6%), 319 time series. Median time series length is 23 (min=23, max=23). \n",
      "\n",
      "Provided data contains following columns:\n",
      "\ttarget: 'unit_sales'\n",
      "\tknown_covariates:\n",
      "\t\tcategorical:        []\n",
      "\t\tcontinuous (float): ['scaled_price', 'promotion_email', 'promotion_homepage']\n",
      "\n",
      "To learn how to fix incorrectly inferred types, please see documentation for TimeSeriesPredictor.fit\n",
      "\n",
      "AutoGluon will gauge predictive performance using evaluation metric: 'WQL'\n",
      "\tThis metric's sign has been flipped to adhere to being higher_is_better. The metric score can be multiplied by -1 to get the metric value.\n",
      "===================================================\n",
      "\n",
      "Starting training. Start time is 2024-11-26 09:17:00\n",
      "Models that will be trained: ['ChronosZeroShot[bolt_small]', 'ChronosWithRegressor[bolt_small]']\n",
      "Training timeseries model ChronosZeroShot[bolt_small]. Training for up to 29.9s of the 59.9s of remaining time.\n",
      "\t-0.4523       = Validation score (-WQL)\n",
      "\t0.02    s     = Training runtime\n",
      "\t0.84    s     = Validation (prediction) runtime\n",
      "Training timeseries model ChronosWithRegressor[bolt_small]. Training for up to 59.0s of the 59.0s of remaining time.\n",
      "\t-0.3580       = Validation score (-WQL)\n",
      "\t1.00    s     = Training runtime\n",
      "\t0.92    s     = Validation (prediction) runtime\n",
      "Training complete. Models trained: ['ChronosZeroShot[bolt_small]', 'ChronosWithRegressor[bolt_small]']\n",
      "Total runtime: 2.80 s\n",
      "Best model: ChronosWithRegressor[bolt_small]\n",
      "Best model score: -0.3580\n"
     ]
    }
   ],
   "source": [
    "predictor = TimeSeriesPredictor(\n",
    "    prediction_length=prediction_length,\n",
    "    target=\"unit_sales\",\n",
    "    known_covariates_names=[\"scaled_price\", \"promotion_email\", \"promotion_homepage\"],\n",
    ").fit(\n",
    "    train_data,\n",
    "    hyperparameters={\n",
    "        \"Chronos\": [\n",
    "            # Zero-shot model WITHOUT covariates\n",
    "            {\n",
    "                \"model_path\": \"bolt_small\",\n",
    "                \"ag_args\": {\"name_suffix\": \"ZeroShot\"},\n",
    "            },\n",
    "            # Chronos-Bolt (Small) combined with CatBoost on covariates\n",
    "            {\n",
    "                \"model_path\": \"bolt_small\",\n",
    "                \"covariate_regressor\": \"CAT\",\n",
    "                \"target_scaler\": \"standard\",\n",
    "                \"ag_args\": {\"name_suffix\": \"WithRegressor\"},\n",
    "            },\n",
    "        ],\n",
    "    },\n",
    "    enable_ensemble=False,\n",
    "    time_limit=60,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once the predictor has been fit, we can evaluate it on the test dataset and generate the leaderboard. We see that the model that utilizes the covariates produces a more accurate forecast on the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Additional data provided, testing on additional data. Resulting leaderboard will be sorted according to test score (`score_test`).\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>model</th>\n",
       "      <th>score_test</th>\n",
       "      <th>score_val</th>\n",
       "      <th>pred_time_test</th>\n",
       "      <th>pred_time_val</th>\n",
       "      <th>fit_time_marginal</th>\n",
       "      <th>fit_order</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ChronosWithRegressor[bolt_small]</td>\n",
       "      <td>-0.268969</td>\n",
       "      <td>-0.358048</td>\n",
       "      <td>0.881176</td>\n",
       "      <td>0.916053</td>\n",
       "      <td>1.004376</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ChronosZeroShot[bolt_small]</td>\n",
       "      <td>-0.318562</td>\n",
       "      <td>-0.452296</td>\n",
       "      <td>0.859930</td>\n",
       "      <td>0.844927</td>\n",
       "      <td>0.019435</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              model  score_test  score_val  pred_time_test  \\\n",
       "0  ChronosWithRegressor[bolt_small]   -0.268969  -0.358048        0.881176   \n",
       "1       ChronosZeroShot[bolt_small]   -0.318562  -0.452296        0.859930   \n",
       "\n",
       "   pred_time_val  fit_time_marginal  fit_order  \n",
       "0       0.916053           1.004376          2  \n",
       "1       0.844927           0.019435          1  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictor.leaderboard(test_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that the covariates may not always be useful — for some datasets, the zero-shot model may achieve better accuracy. Therefore, it's always important to try out multiple models and select the one that achieves the best accuracy on held-out data. This is done automatically in AutoGluon's `\"high_quality\"` and `\"best_quality\"` presets."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## FAQ\n",
    "\n",
    "\n",
    "#### How accurate is Chronos?\n",
    "\n",
    "In several independent evaluations we found Chronos to be effective in zero-shot forecasting. \n",
    "The accuracy of Chronos-Bolt (base) often exceeds statistical baseline models, and is often comparable to deep learning \n",
    "models such as `TemporalFusionTransformer` or `PatchTST`.\n",
    "\n",
    "#### What is the recommended hardware for running Chronos models?\n",
    "\n",
    "For fine-tuning and inference with larger Chronos and Chronos-Bolt models, we tested the AWS `g5.2xlarge` and `p3.2xlarge` instances that feature NVIDIA A10G and V100 GPUs, with at least 16GiB of GPU memory and 32GiB of main memory. \n",
    "\n",
    "Chronos-Bolt models can also be used on CPU machines, but this will typically result in a longer runtime.\n",
    "\n",
    "\n",
    "#### Where can I ask specific questions on Chronos?\n",
    "\n",
    "The AutoGluon team are among the core developers of Chronos. So you can ask Chronos-related questions on AutoGluon channels such \n",
    "as the Discord [server](https://discord.gg/wjUmjqAc2N), or [GitHub](https://github.com/autogluon/autogluon). You can also join \n",
    "the discussion on the Chronos GitHub [page](https://github.com/amazon-science/chronos-forecasting/discussions)."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ag",
   "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": "3.11.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}