[WIP] Use Tullio for pairwise distances #385
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
I was checking the docstrings and the one for the `spectral_mixture_kernel` is particularly wrong I think. I double checked with the first reference paper as well as the implementation and changed the formula/description of the arguments
… into tg/correct_doc_spectral_mixture
…ssianProcesses/KernelFunctions.jl into tg/correct_doc_spectral_mixture
Co-authored-by: David Widmann <[email protected]>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: David Widmann <[email protected]>
|
Ah... Ups I think I did not branch from master.... |
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
|
|
||
| function colwise(::DotProduct, x::ColVecs, y::ColVecs=x) | ||
| return @tullio out[i] := x.X[k, i] * y.X[k, i] | ||
| end |
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| end | |
| end |
| return map(d, x, y) | ||
| end | ||
| return @tullio out[i] := d(x[i], y[i]) | ||
| end |
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| end | |
| end |
|
Blessed be |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
We have a long-time problem for binary operations like
DotProductnot satisfying the requirements of theDistances.jlframework (not a proper metric). Additionally,Distances.jlis very incompatible with GPU operations (see JuliaStats/Distances.jl#143 and JuliaStats/Distances.jl#137).Using
Tullio.jlshould solve both these problems. Some quick benchmarks shows that Tullio is both faster and more GPU-able than Distances.jlThere is a longer discussion about this PR in #380
This should also close #98 and replace #194
Proposed changes
pairwiseandcolwiseDistances.jlare their types (we stop usingDistances.pairwise).pairwiseforColVecsandRowVecswhen possible to improve speed (and GPU compatibility)AbstractBinaryOpabstract type for objects likeDotProductandDeltaand combine them with Distances usingBinaryOp = Union{AbstractBinaryOp,Distances.PreMetric}.What alternatives have you considered?
Dropping Distances.jl operations anyway but without Tullio but Tullio shows it's faster.
Breaking changes