Smarter conjugation for inplace BLAS rules

Open

#281 opened on Oct 12, 2020 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement inplace performance

Dominant language: Julia
Stars: 475
Forks: 98
PR merge metrics: No merged PRs in 30d

Description

@sethaxen wrote in https://github.com/JuliaDiff/ChainRules.jl/pull/279#pullrequestreview-506908830

not allocate new conjugated matrices in the BLAS rules? e.g. in

https://github.com/JuliaDiff/ChainRules.jl/blob/c877550430a3cb0f657f43ccbf4e64b13b7177a7/src/rulesets/LinearAlgebra/blas.jl#L134-L137
,
we allocate a new matrix A, which I think can be made faster by allocating a new vector instead and then conjugating a vector in-place:
             ∂x = InplaceableThunk( 
                 @thunk(gemv('N', α', conj(A), ȳ)), 
                 x̄ -> conj!(gemv!('N', α, A, conj(ȳ), one(T), conj!(x̄)))
             ) 

Definately worth benchmarking

Contributor guide

No contributing guide indexed for this repository

First steps

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

Research direction

Start in src/rulesets/LinearAlgebra/blas.jl at lines 134-137 and inspect the existing InplaceableThunk and BLAS calls. Benchmark the current matrix-allocation path against the proposed vector and in-place conjugation approach; done means the chosen implementation is faster without changing rule behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack: julia
Domain: performance
Issue type: Refactor
Difficulty: 4/5
Estimated time: 3-5 days
Activity status: Stale
Clarity: Mostly clear
Newbie friendliness: 35/100

A short digest of beginner-friendly GitHub issues.