dfm / dfm/tinygp

Why L1 for stationary kernels by default?

Open
#258 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
349
Forks
35
Avg merge
2d 2h
Merged PRs (30d)
1

Description

Hi Dan,

I really like this library; thank you for designing it.
I'm working on extending it to specialized algebra for spectral kernels.

My question: in stationary.py, why is the default Distance for Stationary kernels L1Distance?

This has the consequence that eg. Matern32 kernel with default instantiation has non-standard behavior whenever the input dimension is larger than 1.
I'm not sure if many users would suspect this behavior, even though it is documented.

Cheers,
Marnix

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

Start with stationary.py and inspect how Stationary selects its default Distance, especially L1Distance, alongside Matern32 behavior for multidimensional inputs. Confirm the design rationale and clarify the documented consequences; done means the default's purpose and non-standard multidimensional behavior are clearly explained.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Documentation
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.