sktime / sktime/pytorch-forecasting

[ENH] Add v2 interface support for DeepAR

Open
#2,007 5 comments 3 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement module:models
Dominant language
Python
Stars
5k
Forks
912
Avg merge
3d 13h
Merged PRs (30d)
12

Description

Referencing #1992

The DeepAR model currently does not have a v2-compliant estimator/package. Unlike DLinear, SAMformer, and TFT, it does not provide a *_pkg_v2 wrapper inheriting from _BasePtForecasterV2, so it cannot be used through the v2 interface.

Should we add a v2 package for DeepAR? If so, would interfacing with the existing implementation or forking and adapting it be preferred?

CC: @fkiraly

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Compare the v2 wrappers for DLinear, SAMformer, and TFT with the existing DeepAR implementation. First determine whether DeepAR should be interfaced directly or forked and adapted; done means it has a v2-compliant package usable through the v2 interface.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
api, machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.