Advanced Forecasting with Python: With State-of-the-Art-Models Including LSTMs, Facebook’s Prophet, and Amazon’s DeepAR

© The Author(s), under exclusive license to APress Media, LLC, part of Springer Nature 2021

J. KorstanjeAdvanced Forecasting with Pythonhttps://doi.org/10.1007/978-1-4842-7150-6_20

20. The DeepAR Model

Joos Korstanje¹  

(1)

Maisons Alfort, France

 

DeepAR is a model developed by researchers at Amazon. DeepAR provides an interface to building time series models using a deep learning architecture based on RNNs. The advantage of using DeepAR is that it comes with an interface that is easier to use for model building when compared to Keras.

DeepAR can be considered a competitor with Facebook’s Prophet that you have seen in the previous chapter. Both models try to deliver a simple-to-use model interface for models that are very complex under the hood. If ...