Implications of array broadcasting Bradley Lucier (24 Oct 2024 18:55 UTC)
Re: Implications of array broadcasting Alex Shinn (28 Oct 2024 06:17 UTC)
Re: Implications of array broadcasting Bradley Lucier (30 Oct 2024 23:57 UTC)
Re: Implications of array broadcasting Alex Shinn (31 Oct 2024 13:15 UTC)
Re: Implications of array broadcasting Bradley Lucier (02 Nov 2024 04:24 UTC)
Re: Implications of array broadcasting Alex Shinn (03 Nov 2024 22:21 UTC)
Re: Implications of array broadcasting Bradley Lucier (05 Nov 2024 03:08 UTC)
Re: Implications of array broadcasting Alex Shinn (07 Nov 2024 00:27 UTC)
Re: Implications of array broadcasting Bradley Lucier (07 Nov 2024 20:29 UTC)
Re: Implications of array broadcasting Alex Shinn (08 Nov 2024 00:01 UTC)
Re: Implications of array broadcasting Bradley Lucier (08 Nov 2024 20:51 UTC)
Re: Implications of array broadcasting Alex Shinn (08 Nov 2024 22:54 UTC)

Re: Implications of array broadcasting Bradley Lucier 07 Nov 2024 20:29 UTC

On 11/6/24 19:26, Alex Shinn wrote:
> Also, even for mappable operations like `array+`, what I've been doing is
> roughly equivalent to:
>
> (define (array+ a b)
>    (receive (a b) (array-broadcast-both a b)
>      (if (can-use-fast-path? a b)
>        (fast-array+ a b)
>        (array-map + a b))))
>
> where `fast-array+` is currently BLAS but I plan to move to CUDA.

I'm looking at your github repositories to test what you're doing.

I've built and installed the latest chibi and downloaded the latest
alschemist; how do I test the stuff in linalg.*?

Brad