Bolt: Faster matrix and vector operations that run on compressed data

Re:Bolt: Faster matrix and vector operations that run on com... by raxxorraxor on Saturday June 18, 2022 @03:21PM (#31792905)

This looks good. Why do the vectors have to be dense? Just because of overhead/speed gain being the lowest? Just asking if you could use it universally for all operations if I don't know the density.

[ Parent | on HN ]

Re:Bolt: Faster matrix and vector operations that run on com... by anderskaseorg on Saturday June 18, 2022 @03:51PM (#31793174)

If your data is represented as sparse vectors, that sparse representation is already compressed. You wouldn’t want to decompress it to a dense representation just to apply a different, less effective, lossy compression algorithm. That would be like taking a screenshot of a paragraph of text so you can post a JPEG of it to Twitter.

Oh, wait.

[ Parent | on HN ]

Re:Bolt: Faster matrix and vector operations that run on com... by ffast-math on Sunday June 19, 2022 @12:09AM (#31796086)

Exactly. You can run it on sparse inputs. It's just that our implementation doesn't exploit the sparsity, so we don't claim that it will work better.

[ Parent | on HN ]