{ "cells": [ { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "# Hierarchical Forecast" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "This notebook offers a step by step guide to create a hierarchical forecasting pipeline.\n", "\n", "In the pipeline we will use `NeuralForecast` and `HINT` class, to create fit, predict and reconcile forecasts.\n", "\n", "We will use the TourismL dataset that summarizes large Australian national visitor survey.\n", "\n", "Outline
\n", "1. Installing packages
\n", "2. Load hierarchical dataset
\n", "3. Fit and Predict HINT
\n", "4. Forecast Evaluation" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "You can run these experiments using GPU with Google Colab.\n", "\n", "\"Open" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## 1. Installing packages" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "%%capture\n", "!pip install datasetsforecast hierarchicalforecast\n", "!pip install git+https://github.com/Nixtla/neuralforecast.git" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## 2. Load hierarchical dataset" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "This detailed Australian Tourism Dataset comes from the National Visitor Survey, managed by the Tourism Research Australia, it is composed of 555 monthly series from 1998 to 2016, it is organized geographically, and purpose of travel. The natural geographical hierarchy comprises seven states, divided further in 27 zones and 76 regions. The purpose of travel categories are holiday, visiting friends and relatives (VFR), business and other. The MinT (Wickramasuriya et al., 2019), among other hierarchical forecasting studies has used the dataset it in the past. The dataset can be accessed in the [MinT reconciliation webpage](https://robjhyndman.com/publications/mint/), although other sources are available.\n", "\n", "| Geographical Division | Number of series per division | Number of series per purpose | Total |\n", "| --- | --- | --- | --- |\n", "| Australia | 1 | 4 | 5 |\n", "| States | 7 | 28 | 35 |\n", "| Zones | 27 | 108 | 135 |\n", "| Regions | 76 | 304 | 380 |\n", "| Total | 111 | 444 | 555 |\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "\n", "from datasetsforecast.hierarchical import HierarchicalData\n", "from hierarchicalforecast.utils import aggregate, HierarchicalPlot\n", "\n", "from neuralforecast.utils import augment_calendar_df\n", "\n", "def sort_df_hier(Y_df, S):\n", " # NeuralForecast core, sorts unique_id lexicographically\n", " # by default, this method matches S_df and Y_hat_df hierarchical order.\n", " Y_df.unique_id = Y_df.unique_id.astype('category')\n", " Y_df.unique_id = Y_df.unique_id.cat.set_categories(S.index)\n", " Y_df = Y_df.sort_values(by=['unique_id', 'ds'])\n", " return Y_df\n", "\n", "# Load hierarchical dataset\n", "Y_df, S_df, tags = HierarchicalData.load('./data', 'TourismLarge')\n", "Y_df['ds'] = pd.to_datetime(Y_df['ds'])\n", "Y_df = sort_df_hier(Y_df, S_df)\n", "\n", "Y_df, _ = augment_calendar_df(df=Y_df, freq='M')" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "Mathematically a hierarchical multivariate time series can be denoted by the vector $\\mathbf{y}_{[a,b],t}$ defined by the following aggregation constraint:\n", "\n", "$$\n", "\\mathbf{y}_{[a,b],t} = \\mathbf{S}_{[a,b][b]} \\mathbf{y}_{[b],t} \\quad \\Leftrightarrow \\quad \n", "\\begin{bmatrix}\\mathbf{y}_{[a],t}\n", "\\\\ %\\hline\n", "\\mathbf{y}_{[b],t}\\end{bmatrix} \n", "= \\begin{bmatrix}\n", "\\mathbf{A}_{[a][b]}\\\\ %\\hline\n", "\\mathbf{I}_{[b][b]}\n", "\\end{bmatrix}\n", "\\mathbf{y}_{[b],t}\n", "$$\n", "\n", "where $\\mathbf{y}_{[a],t}$ are the aggregate series, $\\mathbf{y}_{[b],t}$ are the bottom level series and $\\mathbf{S}_{[a,b][b]}$ are the hierarchical aggregation constraints." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Here we plot the hierarchical constraints matrix\n", "hplot = HierarchicalPlot(S=S_df, tags=tags)\n", "hplot.plot_summing_matrix()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Here we plot the top most series from the dataset\n", "# that corresponds to the total tourist monthly visits to Australia\n", "plt.figure(figsize=(10,5))\n", "plt.plot(Y_df[Y_df['unique_id']=='TotalAll']['ds'], \n", " Y_df[Y_df['unique_id']=='TotalAll']['y'], label='target')\n", "plt.plot(Y_df[Y_df['unique_id']=='TotalAll']['ds'], \n", " Y_df[Y_df['unique_id']=='TotalAll']['month']*80000, label='month dummy')\n", "plt.xlabel('Date')\n", "plt.ylabel('Tourist Visits')\n", "plt.legend()\n", "plt.grid()\n", "plt.show()\n", "plt.close()" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## 3. Fit and Predict HINT\n", "\n", "The Hierarchical Forecast Network (HINT) combines into an easy to use model three components:
\n", "1. SoTA neural forecast model.
\n", "2. An efficient and flexible multivariate probability distribution.
\n", "3. Builtin reconciliation capabilities.
" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "\n", "from neuralforecast import NeuralForecast\n", "from neuralforecast.models import NBEATSx, NHITS, HINT\n", "from neuralforecast.losses.pytorch import GMM, PMM, DistributionLoss, sCRPS" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# Train test splits\n", "horizon = 12\n", "Y_test_df = Y_df.groupby('unique_id').tail(horizon)\n", "Y_train_df = Y_df.drop(Y_test_df.index)\n", "Y_test_df = Y_test_df.set_index('unique_id')\n", "Y_train_df = Y_train_df.set_index('unique_id')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "INFO:lightning_fabric.utilities.seed:Global seed set to 1\n" ] } ], "source": [ "# Horizon and quantiles\n", "level = np.arange(0, 100, 2)\n", "qs = [[50-lv/2, 50+lv/2] if lv!=0 else [50] for lv in level]\n", "quantiles = np.sort(np.concatenate(qs)/100)\n", "\n", "# HINT := BaseNetwork + Distribution + Reconciliation\n", "nhits = NHITS(h=horizon,\n", " input_size=24,\n", " loss=GMM(n_components=10, quantiles=quantiles),\n", " hist_exog_list=['month'],\n", " max_steps=2000,\n", " early_stop_patience_steps=10,\n", " val_check_steps=50,\n", " scaler_type='robust',\n", " learning_rate=1e-3,\n", " valid_loss=sCRPS(quantiles=quantiles))\n", "\n", "model = HINT(h=horizon, S=S_df.values,\n", " model=nhits, reconciliation='BottomUp')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "%%capture\n", "Y_df['y'] = Y_df['y'] * (Y_df['y'] > 0)\n", "nf = NeuralForecast(models=[model], freq='MS')\n", "# Y_hat_df = nf.cross_validation(df=Y_df, val_size=12, n_windows=1)\n", "nf.fit(df=Y_train_df, val_size=12)\n", "Y_hat_df = nf.predict()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": null, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "unique_id = 'TotalAll'\n", "Y_plot_df = Y_df[Y_df.unique_id==unique_id].tail(12*5)\n", "Y_hat_df = Y_hat_df.reset_index()\n", "plot_df = Y_hat_df[Y_hat_df.unique_id==unique_id]\n", "plot_df = Y_plot_df.merge(plot_df, on=['unique_id', 'ds'], how='left')\n", "\n", "plt.plot(plot_df['ds'], plot_df['y'], c='black', label='True')\n", "plt.plot(plot_df['ds'], plot_df['HINT-median'], c='blue', label='median')\n", "plt.fill_between(x=plot_df['ds'][-12:],\n", " y1=plot_df['HINT-lo-90.0'][-12:].values,\n", " y2=plot_df['HINT-hi-90.0'][-12:].values,\n", " alpha=0.4, label='level 90')\n", "plt.legend()\n", "plt.grid()\n", "plt.plot()" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## 4. Forecast Evaluation" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "To evaluate the coherent probabilistic predictions we use the scaled Continuous Ranked Probability Score (sCRPS), defined as follows:\n", "\n", "$$\n", "\\mathrm{CRPS}(\\hat{F}_{[a,b],\\tau},\\mathbf{y}_{[a,b],\\tau}) = \n", " \\frac{2}{N_{a}+N_{b}} \\sum_{i} \\int^{1}_{0} \\mathrm{QL}(\\hat{F}_{i,\\tau}, y_{i,\\tau})_{q} dq\n", "$$\n", "\n", "$$\n", "\\mathrm{sCRPS}(\\hat{F}_{[a,b\\,],\\tau},\\mathbf{y}_{[a,b\\,],\\tau}) = \n", " \\frac{\\mathrm{CRPS}(\\hat{F}_{[a,b\\,],\\tau},\\mathbf{y}_{[a,b\\,],\\tau})}{\\sum_{i} | y_{i,\\tau} |}\n", "$$\n", "\n", "As you can see the HINT model efficiently achieves state of the art accuracy under minimal tuning." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from hierarchicalforecast.evaluation import scaled_crps \n", " \n", "def _get_hierarchical_scrps(hier_idxs, Y, Yq_hat, quantiles):\n", " # We use the indexes obtained from the aggregation tags\n", " # to compute scaled CRPS across the hierarchy levels \n", " scrps_list = []\n", " for idxs in hier_idxs:\n", " y = Y[idxs, :]\n", " yq_hat = Yq_hat[idxs, :, :]\n", " scrps = scaled_crps(y, yq_hat, quantiles)\n", " scrps_list.append(scrps)\n", " return scrps_list\n", "\n", "hier_idxs = [np.arange(len(S_df))] +\\\n", " [S_df.index.get_indexer(tags[level]) for level in list(tags.keys())]" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "%%capture\n", "n_series = len(S_df)\n", "n_quantiles = len(quantiles)\n", "\n", "# Bootstrap predictions\n", "n_samples = 5\n", "Y_hat_df_list = [nf.predict() for _ in range(n_samples)]\n", "\n", "# Parse y_test and y_rec\n", "# Keep only quantile columns from Y_hat_df\n", "# Removing mean and median default outputs\n", "model_name = type(model).__name__\n", "quantile_columns = [model_name + n for n in nhits.loss.output_names]\n", "quantile_columns.remove(model_name)\n", "Yq_hat = []\n", "for sample_idx in range(n_samples):\n", " Y_hat = Y_hat_df_list[sample_idx][quantile_columns].values\n", " Yq_hat.append(Y_hat.reshape(1, n_series, horizon, n_quantiles))\n", "\n", "Yq_hat = np.concatenate(Yq_hat, axis=0)\n", "Y_test = Y_test_df['y'].values.reshape(n_series, horizon)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Y_test.shape [n_series, horizon] (555, 12)\n", "Yq_hat.shape [n_samples, n_series, horizon, n_quantiles] (5, 555, 12, 99)\n" ] }, { "data": { "text/html": [ "\n", "
\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
LevelsHINTDPMNHierE2EARIMA_MinTrace_BARIMA_BottomUp_BARIMA
0Overall0.1178±0.00020.1249±0.00200.1472±0.00290.1313±0.00090.1375±0.00130.1416
1Country0.0288±0.00070.0431±0.00420.0842±0.00510.0471±0.00180.0622±0.00260.0263
2State0.0593±0.00040.0637±0.00320.1012±0.00290.0723±0.00110.0820±0.00190.0904
3Zone0.1023±0.00030.1084±0.00330.1317±0.00220.1143±0.00070.1207±0.00100.1389
4Region0.1451±0.00040.1554±0.00250.1705±0.00230.1591±0.00060.1646±0.00070.1878
5Country/Purpose0.0752±0.00060.0700±0.00380.0995±0.00610.0723±0.00140.0788±0.00180.0770
6State/Purpose0.1109±0.00010.1070±0.00230.1336±0.00420.1243±0.00140.1268±0.00170.1270
7Zone/Purpose0.1773±0.00030.1887±0.00320.1955±0.00250.1919±0.00080.1949±0.00100.2022
8Region/Purpose0.2438±0.00020.2629±0.00340.2615±0.00160.2694±0.00060.2698±0.00080.2834
\n", "
\n", " \n", " \n", " \n", "\n", " \n", "
\n", "
\n", " " ], "text/plain": [ " Levels HINT DPMN HierE2E \\\n", "0 Overall 0.1178±0.0002 0.1249±0.0020 0.1472±0.0029 \n", "1 Country 0.0288±0.0007 0.0431±0.0042 0.0842±0.0051 \n", "2 State 0.0593±0.0004 0.0637±0.0032 0.1012±0.0029 \n", "3 Zone 0.1023±0.0003 0.1084±0.0033 0.1317±0.0022 \n", "4 Region 0.1451±0.0004 0.1554±0.0025 0.1705±0.0023 \n", "5 Country/Purpose 0.0752±0.0006 0.0700±0.0038 0.0995±0.0061 \n", "6 State/Purpose 0.1109±0.0001 0.1070±0.0023 0.1336±0.0042 \n", "7 Zone/Purpose 0.1773±0.0003 0.1887±0.0032 0.1955±0.0025 \n", "8 Region/Purpose 0.2438±0.0002 0.2629±0.0034 0.2615±0.0016 \n", "\n", " ARIMA_MinTrace_B ARIMA_BottomUp_B ARIMA \n", "0 0.1313±0.0009 0.1375±0.0013 0.1416 \n", "1 0.0471±0.0018 0.0622±0.0026 0.0263 \n", "2 0.0723±0.0011 0.0820±0.0019 0.0904 \n", "3 0.1143±0.0007 0.1207±0.0010 0.1389 \n", "4 0.1591±0.0006 0.1646±0.0007 0.1878 \n", "5 0.0723±0.0014 0.0788±0.0018 0.0770 \n", "6 0.1243±0.0014 0.1268±0.0017 0.1270 \n", "7 0.1919±0.0008 0.1949±0.0010 0.2022 \n", "8 0.2694±0.0006 0.2698±0.0008 0.2834 " ] }, "execution_count": null, "metadata": {}, "output_type": "execute_result" } ], "source": [ "print('Y_test.shape [n_series, horizon]', Y_test.shape)\n", "print('Yq_hat.shape [n_samples, n_series, horizon, n_quantiles]', Yq_hat.shape)\n", "\n", "# Compute bootstraped sCRPS\n", "scrps_hint = [_get_hierarchical_scrps(hier_idxs, Y_test, Yq_hat[sample_idx], quantiles) \\\n", " for sample_idx in range(n_samples)]\n", "crps_mean = np.mean(np.array(scrps_hint), axis=0)\n", "crps_std = np.std(np.array(scrps_hint), axis=0)\n", "scrps_hint = [f'{crps_mean[level_idx]:.4f}±{(1.96 * crps_std[level_idx]):.4f}' \\\n", " for level_idx in range(len(crps_mean))]\n", "\n", "# Add reported baselines' performance\n", "levels = ['Overall', 'Country', 'State', 'Zone', 'Region',\n", " 'Country/Purpose', 'State/Purpose', 'Zone/Purpose', 'Region/Purpose']\n", "scrps_dpmn = [\"0.1249±0.0020\",\"0.0431±0.0042\",\"0.0637±0.0032\",\"0.1084±0.0033\",\n", " \"0.1554±0.0025\",\"0.0700±0.0038\",\"0.1070±0.0023\",\"0.1887±0.0032\",\"0.2629±0.0034\"]\n", "scrps_hiere2e = [\"0.1472±0.0029\",\"0.0842±0.0051\",\"0.1012±0.0029\",\"0.1317±0.0022\",\n", " \"0.1705±0.0023\",\"0.0995±0.0061\",\"0.1336±0.0042\",\"0.1955±0.0025\",\"0.2615±0.0016\"]\n", "scrps_arima_mintrace = [\"0.1313±0.0009\",\"0.0471±0.0018\",\"0.0723±0.0011\",\"0.1143±0.0007\",\n", " \"0.1591±0.0006\",\"0.0723±0.0014\",\"0.1243±0.0014\",\"0.1919±0.0008\",\"0.2694±0.0006\"]\n", "scrps_arima_bu = [\"0.1375±0.0013\",\"0.0622±0.0026\",\"0.0820±0.0019\",\"0.1207±0.0010\",\n", " \"0.1646±0.0007\",\"0.0788±0.0018\",\"0.1268±0.0017\",\"0.1949±0.0010\",\"0.2698±0.0008\"]\n", "scrps_arima = [\"0.1416\",\"0.0263\",\"0.0904\",\"0.1389\",\"0.1878\",\"0.0770\",\"0.1270\",\"0.2022\",\"0.2834\"]\n", "\n", "scrps_results = dict(Levels=levels,\n", " HINT=scrps_hint, \n", " DPMN=scrps_dpmn,\n", " HierE2E=scrps_hiere2e, \n", " ARIMA_MinTrace_B=scrps_arima_mintrace,\n", " ARIMA_BottomUp_B=scrps_arima_bu,\n", " ARIMA=scrps_arima)\n", "scrps_results = pd.DataFrame(scrps_results)\n", "scrps_results" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## References" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "- [Kin G. Olivares, David Luo, Cristian Challu, Stefania La Vattiata, Max Mergenthaler, Artur Dubrawski (2023). \"HINT: Hierarchical Mixture Networks For Coherent Probabilistic Forecasting\". International Conference on Machine Learning (ICML). Workshop on Structured Probabilistic Inference & Generative Modeling. Available at https://arxiv.org/abs/2305.07089.](https://arxiv.org/abs/2305.07089)
\n", "- [Kin G. Olivares, O. Nganba Meetei, Ruijun Ma, Rohan Reddy, Mengfei Cao, Lee Dicker (2023).\"Probabilistic Hierarchical Forecasting with Deep Poisson Mixtures\". International Journal Forecasting, accepted paper. URL https://arxiv.org/pdf/2110.13179.pdf.](https://arxiv.org/pdf/2110.13179.pdf)
\n", "- [Kin G. Olivares, Federico Garza, David Luo, Cristian Challu, Max Mergenthaler, Souhaib Ben Taieb, Shanika Wickramasuriya, and Artur Dubrawski (2023). \"HierarchicalForecast: A reference framework for hierarchical forecasting\". Journal of Machine Learning Research, submitted. URL https://arxiv.org/abs/2207.03517](https://arxiv.org/abs/2207.03517)" ] } ], "metadata": { "kernelspec": { "display_name": "python3", "language": "python", "name": "python3" } }, "nbformat": 4, "nbformat_minor": 0 }