AutoGluon Time Series - Forecasting Quick Start#

Open In Colab Open In SageMaker Studio Lab

Via a simple fit() call, AutoGluon can train and tune

  • simple forecasting models (e.g., ARIMA, ETS, Theta),

  • powerful deep learning models (e.g., DeepAR, Temporal Fusion Transformer),

  • tree-based models (e.g., LightGBM),

  • an ensemble that combines predictions of other models

to produce multi-step ahead probabilistic forecasts for univariate time series data.

This tutorial demonstrates how to quickly start using AutoGluon to generate hourly forecasts for the M4 forecasting competition dataset.

Loading time series data as a TimeSeriesDataFrame#

First, we import some required modules

import pandas as pd
from autogluon.timeseries import TimeSeriesDataFrame, TimeSeriesPredictor

To use autogluon.timeseries, we will only need the following two classes:

  • TimeSeriesDataFrame stores a dataset consisting of multiple time series.

  • TimeSeriesPredictor takes care of fitting, tuning and selecting the best forecasting models, as well as generating new forecasts.

We load a subset of the M4 hourly dataset as a pandas.DataFrame

df = pd.read_csv("https://autogluon.s3.amazonaws.com/datasets/timeseries/m4_hourly_subset/train.csv")
df.head()
item_id timestamp target
0 H1 1750-01-01 00:00:00 605.0
1 H1 1750-01-01 01:00:00 586.0
2 H1 1750-01-01 02:00:00 586.0
3 H1 1750-01-01 03:00:00 559.0
4 H1 1750-01-01 04:00:00 511.0

AutoGluon expects time series data in long format. Each row of the data frame contains a single observation (timestep) of a single time series represented by

  • unique ID of the time series ("item_id") as int or str

  • timestamp of the observation ("timestamp") as a pandas.Timestamp or compatible format

  • numeric value of the time series ("target")

The raw dataset should always follow this format with at least three columns for unique ID, timestamp, and target value, but the names of these columns can be arbitrary. It is important, however, that we provide the names of the columns when constructing a TimeSeriesDataFrame that is used by AutoGluon. AutoGluon will raise an exception if the data doesn’t match the expected format.

train_data = TimeSeriesDataFrame.from_data_frame(
    df,
    id_column="item_id",
    timestamp_column="timestamp"
)
train_data.head()
target
item_id timestamp
H1 1750-01-01 00:00:00 605.0
1750-01-01 01:00:00 586.0
1750-01-01 02:00:00 586.0
1750-01-01 03:00:00 559.0
1750-01-01 04:00:00 511.0

We refer to each individual time series stored in a TimeSeriesDataFrame as an item. For example, items might correspond to different products in demand forecasting, or to different stocks in financial datasets. This setting is also referred to as a panel of time series. Note that this is not the same as multivariate forecasting — AutoGluon generates forecasts for each time series individually, without modeling interactions between different items (time series).

TimeSeriesDataFrame inherits from pandas.DataFrame, so all attributes and methods of pandas.DataFrame are available in a TimeSeriesDataFrame. It also provides other utility functions, such as loaders for different data formats (see TimeSeriesDataFrame for details).

Training time series models with TimeSeriesPredictor.fit#

To forecast future values of the time series, we need to create a TimeSeriesPredictor object.

Models in autogluon.timeseries forecast time series multiple steps into the future. We choose the number of these steps — the prediction length (also known as the forecast horizon) — depending on our task. For example, our dataset contains time series measured at hourly frequency, so we set prediction_length = 48 to train models that forecast up to 48 hours into the future.

We instruct AutoGluon to save trained models in the folder ./autogluon-m4-hourly. We also specify that AutoGluon should rank models according to mean absolute scaled error (MASE), and that data that we want to forecast is stored in the column "target" of the TimeSeriesDataFrame.

predictor = TimeSeriesPredictor(
    prediction_length=48,
    path="autogluon-m4-hourly",
    target="target",
    eval_metric="MASE",
)

predictor.fit(
    train_data,
    presets="medium_quality",
    time_limit=600,
)

Beginning AutoGluon training... Time limit = 600s
AutoGluon will save models to 'autogluon-m4-hourly'
=================== System Info ===================
AutoGluon Version:  1.0.1b20240301
Python Version:     3.10.8
Operating System:   Linux
Platform Machine:   x86_64
Platform Version:   #1 SMP Tue Nov 30 00:17:50 UTC 2021
CPU Count:          8
GPU Count:          1
Memory Avail:       28.80 GB / 30.96 GB (93.0%)
Disk Space Avail:   215.42 GB / 255.99 GB (84.2%)
===================================================
Setting presets to: medium_quality

Fitting with arguments:
{'enable_ensemble': True,
 'eval_metric': MASE,
 'hyperparameters': 'light',
 'known_covariates_names': [],
 'num_val_windows': 1,
 'prediction_length': 48,
 'quantile_levels': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],
 'random_seed': 123,
 'refit_every_n_windows': 1,
 'refit_full': False,
 'target': 'target',
 'time_limit': 600,
 'verbosity': 2}

Inferred time series frequency: 'H'
Provided train_data has 148060 rows, 200 time series. Median time series length is 700 (min=700, max=960). 

Provided data contains following columns:
	target: 'target'

AutoGluon will gauge predictive performance using evaluation metric: 'MASE'
	This 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.
===================================================

Starting training. Start time is 2024-03-01 09:37:21
Models that will be trained: ['Naive', 'SeasonalNaive', 'ETS', 'Theta', 'RecursiveTabular', 'DirectTabular', 'TemporalFusionTransformer']
Training timeseries model Naive. Training for up to 74.8s of the 598.4s of remaining time.
	-6.6629       = Validation score (-MASE)
	0.11    s     = Training runtime
	1.94    s     = Validation (prediction) runtime
Training timeseries model SeasonalNaive. Training for up to 85.2s of the 596.3s of remaining time.
	-1.2169       = Validation score (-MASE)
	0.10    s     = Training runtime
	0.21    s     = Validation (prediction) runtime
Training timeseries model ETS. Training for up to 99.3s of the 596.0s of remaining time.
	-1.9684       = Validation score (-MASE)
	0.10    s     = Training runtime
	82.89   s     = Validation (prediction) runtime
Training timeseries model Theta. Training for up to 102.6s of the 513.0s of remaining time.
	-2.1425       = Validation score (-MASE)
	0.10    s     = Training runtime
	18.02   s     = Validation (prediction) runtime
Training timeseries model RecursiveTabular. Training for up to 123.7s of the 494.9s of remaining time.
	-0.9339       = Validation score (-MASE)
	13.19   s     = Training runtime
	2.34    s     = Validation (prediction) runtime
Training timeseries model DirectTabular. Training for up to 159.8s of the 479.3s of remaining time.
	-1.3073       = Validation score (-MASE)
	11.64   s     = Training runtime
	0.69    s     = Validation (prediction) runtime
Training timeseries model TemporalFusionTransformer. Training for up to 233.5s of the 467.0s of remaining time.
	-2.1234       = Validation score (-MASE)
	86.34   s     = Training runtime
	0.28    s     = Validation (prediction) runtime
Fitting simple weighted ensemble.
	Ensemble weights: {'DirectTabular': 0.06, 'ETS': 0.05, 'RecursiveTabular': 0.85, 'SeasonalNaive': 0.03, 'TemporalFusionTransformer': 0.01}
	-0.9140       = Validation score (-MASE)
	3.47    s     = Training runtime
	86.40   s     = Validation (prediction) runtime
Training complete. Models trained: ['Naive', 'SeasonalNaive', 'ETS', 'Theta', 'RecursiveTabular', 'DirectTabular', 'TemporalFusionTransformer', 'WeightedEnsemble']
Total runtime: 221.77 s
Best model: WeightedEnsemble
Best model score: -0.9140

<autogluon.timeseries.predictor.TimeSeriesPredictor at 0x7f37da50ae90>

Here we used the "medium_quality" presets and limited the training time to 10 minutes (600 seconds). The presets define which models AutoGluon will try to fit. For medium_quality presets, these are simple baselines (Naive, SeasonalNaive), statistical models (ETS, Theta), tree-based models based on LightGBM (RecursiveTabular, DirectTabular), a deep learning model TemporalFusionTransformer, and a weighted ensemble combining these. Other available presets for TimeSeriesPredictor are "fast_training", "high_quality" and "best_quality". Higher quality presets will usually produce more accurate forecasts but take longer to train.

Inside fit(), AutoGluon will train as many models as possible within the given time limit. Trained models are then ranked based on their performance on an internal validation set. By default, this validation set is constructed by holding out the last prediction_length timesteps of each time series in train_data.

Generating forecasts with TimeSeriesPredictor.predict#

We can now use the fitted TimeSeriesPredictor to forecast the future time series values. By default, AutoGluon will make forecasts using the model that had the best score on the internal validation set. The forecast always includes predictions for the next prediction_length timesteps, starting from the end of each time series in train_data.

predictions = predictor.predict(train_data)
predictions.head()

Model not specified in predict, will default to the model with the best validation score: WeightedEnsemble
mean 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
item_id timestamp
H1 1750-01-30 04:00:00 627.275082 606.888425 613.932460 618.962773 623.165351 627.275082 631.113552 635.468907 640.433118 647.465578
1750-01-30 05:00:00 557.847274 531.246059 540.381942 547.006266 552.554056 557.847274 562.972849 568.611142 575.161501 584.311445
1750-01-30 06:00:00 516.202320 484.863734 495.625680 503.427965 510.011224 516.202320 522.352083 528.981369 536.701911 547.461087
1750-01-30 07:00:00 485.826997 450.456653 462.620803 471.393327 478.849805 485.826997 492.804439 500.304066 508.998411 521.119145
1750-01-30 08:00:00 460.994987 422.035176 435.455192 445.083017 453.308180 460.994987 468.682854 476.960236 486.541924 499.852745

AutoGluon produces a probabilistic forecast: in addition to predicting the mean (expected value) of the time series in the future, models also provide the quantiles of the forecast distribution. The quantile forecasts give us an idea about the range of possible outcomes. For example, if the "0.1" quantile is equal to 500.0, it means that the model predicts a 10% chance that the target value will be below 500.0.

We will now visualize the forecast and the actually observed values for one of the time series in the dataset. We plot the mean forecast, as well as the 10% and 90% quantiles to show the range of potential outcomes.

import matplotlib.pyplot as plt

# TimeSeriesDataFrame can also be loaded directly from a file
test_data = TimeSeriesDataFrame.from_path("https://autogluon.s3.amazonaws.com/datasets/timeseries/m4_hourly_subset/test.csv")

# Plot 4 randomly chosen time series and the respective forecasts
predictor.plot(test_data, predictions, quantile_levels=[0.1, 0.9], max_history_length=200, max_num_item_ids=4)
../../_images/40751691c245576b605c9991e13740c79d931f8990dfe1a09520d3e786aea7ef.png ../../_images/40751691c245576b605c9991e13740c79d931f8990dfe1a09520d3e786aea7ef.png

Evaluating the performance of different models#

We can view the performance of each model AutoGluon has trained via the leaderboard() method. We provide the test data set to the leaderboard function to see how well our fitted models are doing on the unseen test data. The leaderboard also includes the validation scores computed on the internal validation dataset.

In AutoGluon leaderboards, higher scores always correspond to better predictive performance. Therefore our MASE scores are multiplied by -1, such that higher “negative MASE”s correspond to more accurate forecasts.

# The test score is computed using the last
# prediction_length=48 timesteps of each time series in test_data
predictor.leaderboard(test_data)

Additional data provided, testing on additional data. Resulting leaderboard will be sorted according to test score (`score_test`).
model score_test score_val pred_time_test pred_time_val fit_time_marginal fit_order
0 WeightedEnsemble -0.851040 -0.914033 92.306061 86.401177 3.465552 8
1 RecursiveTabular -0.862797 -0.933874 1.740330 2.339917 13.187542 5
2 SeasonalNaive -1.022854 -1.216909 0.194137 0.206646 0.101999 2
3 DirectTabular -1.445412 -1.307296 0.749681 0.686271 11.638014 6
4 ETS -1.805637 -1.968375 89.219621 82.888903 0.102534 3
5 Theta -1.905365 -2.142531 18.061497 18.024260 0.097715 4
6 TemporalFusionTransformer -1.968007 -2.123390 0.390238 0.279439 86.342039 7
7 Naive -6.696079 -6.662942 0.236997 1.941298 0.109307 1

Summary#

We used autogluon.timeseries to make probabilistic multi-step forecasts on the M4 Hourly dataset. Check out Forecasting Time Series - In Depth to learn about the advanced capabilities of AutoGluon for time series forecasting.