JuliaGaussianProcesses / JuliaGaussianProcesses/KernelFunctions.jl

Domain of kernel input

Open
#16 6 comments 2 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Julia
Stars
275
Forks
41
PR merge metrics
No merged PRs in 30d

Description

Just opening a separate thread following #14, #10 and #3, to talk about how to parametrize the kernels.
The idea would be to parametrize the kernel as Kernel{T,Transform} where T is the type of the input.

Here are some thoughts :

  • It makes sense to define the input type in the kernel parameters as I cannot imagine that the input type would change over time
  • Does it mean that we restrict the kernel to only work on those inputs, as this could cause problems with the AD types ?
  • Should we also do it for Transform as it is the first type to actually "see" the inputs
  • This could be definitely practical for specific kernels as asked in #9, #10 and #14

What do you think @willtebbutt @trthatcher ?

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

No file or test is named in the issue. Start by reading related issues #14, #10, #3, and #9, then examine the existing Kernel{T,Transform} entry point and discussion; done requires an agreed parametrization design and a concrete implementation scope.

Written by the indexing model from the issue text.

Assessment

Tech stack
julia
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.