JuliaGaussianProcesses / KernelFunctions.jl

Julia package for kernel functions for machine learning
https://juliagaussianprocesses.github.io/KernelFunctions.jl/stable/
MIT License
267 stars 32 forks source link

Extend lengthscale tip #476

Closed willtebbutt closed 1 year ago

willtebbutt commented 1 year ago

Summary

It was unclear to a Stheno.jl user how to do ARD stuff -- see this issue. This PR extends our lengthscale tip to multiple dimensions.

Proposed changes

Extend lengthscale tip to cover ARD and factor analysis cases.

What alternatives have you considered?

I considered adding something in Stheno, but I didn't want to dilute the docs. I also considered not adding the factor analyis example, since it is so closely-related to the ARD example, it felt important to include.

Breaking changes

None

codecov[bot] commented 1 year ago

Codecov Report

Base: 40.97% // Head: 47.94% // Increases project coverage by +6.97% :tada:

Coverage data is based on head (39e625a) compared to base (e42d89d). Patch has no changes to coverable lines.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## master #476 +/- ## ========================================== + Coverage 40.97% 47.94% +6.97% ========================================== Files 52 52 Lines 1340 1341 +1 ========================================== + Hits 549 643 +94 + Misses 791 698 -93 ``` | [Impacted Files](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses) | Coverage Δ | | |---|---|---| | [src/mokernels/lmm.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL21va2VybmVscy9sbW0uamw=) | `0.00% <0.00%> (-100.00%)` | :arrow_down: | | [src/mokernels/slfm.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL21va2VybmVscy9zbGZtLmps) | `0.00% <0.00%> (-100.00%)` | :arrow_down: | | [src/distances/delta.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL2Rpc3RhbmNlcy9kZWx0YS5qbA==) | `0.00% <0.00%> (-100.00%)` | :arrow_down: | | [src/mokernels/moinput.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL21va2VybmVscy9tb2lucHV0Lmps) | `0.00% <0.00%> (-100.00%)` | :arrow_down: | | [src/mokernels/mokernel.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL21va2VybmVscy9tb2tlcm5lbC5qbA==) | `0.00% <0.00%> (-100.00%)` | :arrow_down: | | [src/mokernels/independent.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL21va2VybmVscy9pbmRlcGVuZGVudC5qbA==) | `0.00% <0.00%> (-100.00%)` | :arrow_down: | | [src/matrix/kernelkroneckermat.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL21hdHJpeC9rZXJuZWxrcm9uZWNrZXJtYXQuamw=) | `0.00% <0.00%> (-100.00%)` | :arrow_down: | | [src/mokernels/intrinsiccoregion.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL21va2VybmVscy9pbnRyaW5zaWNjb3JlZ2lvbi5qbA==) | `0.00% <0.00%> (-100.00%)` | :arrow_down: | | [src/approximations/nystrom.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL2FwcHJveGltYXRpb25zL255c3Ryb20uamw=) | `0.00% <0.00%> (-92.69%)` | :arrow_down: | | [src/matrix/kernelpdmat.jl](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses#diff-c3JjL21hdHJpeC9rZXJuZWxwZG1hdC5qbA==) | `0.00% <0.00%> (-75.00%)` | :arrow_down: | | ... and [32 more](https://codecov.io/gh/JuliaGaussianProcesses/KernelFunctions.jl/pull/476/diff?src=pr&el=tree-more&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses) | | Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=JuliaGaussianProcesses)

:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.