search

ACEsuit / Polynomials4ML.jl

Polynomials for ML: fast evaluation, batching, differentiation
MIT License
12 stars 5 forks source link

Precompilation #47

Closed cortner closed 1 year ago

cortner commented 1 year ago

It seems P4ML doesn't precompile when used from ACE1x. See also ACEsuit/ACE1pack.jl#130

We need to fix this urgently.

CC @tjjarvinen @CheukHinHoJerry

tjjarvinen commented 1 year ago

The short therm fix is adding __precompile__(false). This has no effect on the actual working of the package, except that it does not try to precompile and thus there will be no failure to precompile. Note that when precompile fails the package works like without precompile.

On fixing the issue itself. My quess is that ObjectPools might have something to do with this. Other than that Julia documentation has section on this topic.

When debugging this you can use Base.compilecache(Base.identify_package("modulename")) to trigger precompile and see does it work.

CheukHinHoJerry commented 1 year ago

I will try to see where is the problem and fix that.

cortner commented 1 year ago

This was not an issue with P4ML but with ACE1x and RepLieGrps