Github user dlwh commented on the pull request:
https://github.com/apache/incubator-spark/pull/575#issuecomment-35218872
@martinjaggi For how it's usually implemented, that's right. But you can
quite likely get better performance doing minibatches with dense vector/CSC
multiply in lieu of a bunch of dot products.
On Sun, Feb 16, 2014 at 2:35 PM, Martin Jaggi
<[email protected]>wrote:
> @fommil <https://github.com/fommil> No matrix operations are performed at
> all so far, only vector addition (of type dense += sparse). See the code
in
> this PR by @mengxr <https://github.com/mengxr> . Vector operations are
> enough for clustering, classification and regression as currently in
MLlib.
> I was referring to the k-Means benchmark posted in the JIRA.
>
> â
> Reply to this email directly or view it on
GitHub<https://github.com/apache/incubator-spark/pull/575#issuecomment-35218573>
> .
>
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. To do so, please top-post your response.
If your project does not have this feature enabled and wishes so, or if the
feature is enabled but not working, please contact infrastructure at
[email protected] or file a JIRA ticket with INFRA.
