awslabs/gluonts

Introduce model input API

Aberta

#2.650 aberto em 14 de fev. de 2023

 (0 comentário) (0 reação) (0 responsável)Python (753 forks)batch import
enhancementhelp wanted

Métricas do repositório

Stars
 (3.888 estrelas)
Métricas de merge de PR
 (Mesclagem média 26d 17h) (4 fundiu PRs em 30d)

Description

Model networks expect a set of parameters.

For example, DeepAR (mxnet) takes the following values:

https://github.com/awslabs/gluonts/blob/9260639dd28791e48db694d409fe5ef1eeb6806d/src/gluonts/mx/model/deepar/_network.py#L874-L886

It would be useful to make these parameters available on Estimator and Predictor to expose what input values (incl. type, etc) are expected.

We already do this to some extend in torch, where we expose input_shapes and input_types on some models.

Ideally we have single way of retrieving this information from all trainable models, both mxnet and torch.

Guia do colaborador