DrChainsaw

I think I have found one instance where the current behaviour is worse: Differences in linestyle become obscured by markers in legends. Seems like one fix would be to scale...

A little late to the party, but here is a related effort w.r.t point 3: https://github.com/DrChainsaw/NaiveGAflux.jl I'm currently just using it to play around with very loosely restricted search spaces...

Thanks @ablaom When I started I was perhaps naively thinking that the search space should not be the million dollar question if you just had a framework which allowed for...

One more shower thought kinda proposal: A PR checklist template (you know, the standard "[] Tests have been added" etc.) with a bullet saying something like "[] Every (undocumented) piece...

Ah, never occured to me to try images as the Array -> RGB -> Array seemed like an unnecssary hassle. Thanks a bunch for pointing it out as it is...

Sorry, forgot to put the versioninfo in the issue. Here is is: ```julia julia> versioninfo() Julia Version 1.7.3 Commit 742b9abb4d (2022-05-06 12:58 UTC) Platform Info: OS: Windows (x86_64-w64-mingw32) CPU: Intel(R)...

> Won’t this not be type stable? The fact that it might return either an Array or a Fill/Ones depending on shape or contents? Yes, I suppose so (ugh for...

Maybe this was not your point, but performance better due to fewer allocations regardless of any additional type instability. code_warntype gives me basically idential output, but maybe I'm not using...

Hi, As for my usecase, without making things too complicated; I'm using FillArrays to help keep memory utilization down, but some parts of my code concatenates arrays. I can easily...

Even this case does not appear to be any more type-unstable than the normal `cat`: ```julia julia> A,B = Fill(1, 1), Fill(2,1) @code_warntype cat(A,B; dims=1) Variables #unused#::Core.Const(Base.var"#cat##kw"()) @_2::NamedTuple{(:dims,), Tuple{Int64}} @_3::Core.Const(cat)...