Tutorials

The best way to learn about GluonTS is to dive right in by following our tutorials. You can download each tutorial as a Jupyter notebook by clicking the download link at the top of each tutorial page.

Forecasting

Quick Start Tutorialforecasting/quick_start_tutorial.html

A quick start guide on how to use the essential features of GluonTS: loading data, training an existing model, evaluating its accuracy.

Extended Forecasting Tutorialforecasting/extended_tutorial.html

An extended tutorial, covering all features of GluonTS more extensively, and explaining how to write your own model using MXNet Gluon.

Write your own GluonTS model with PyTorchforecasting/howto_pytorch_lightning.html

An example of how one can use PyTorch to write a GluonTS forecasting model, and train it with PyTorch Lightning.

Data manipulation

Synthetic Data Generationdata_manipulation/synthetic_data_generation.html

A tutorial on how to use the synthetic data generation toolbox in GluonTS.