Improve quadprod with columnar-only implementation #46
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
First, this removes the option to do row-wise quadratic products since they aren't being used within this package anyway. That allows removing the "keyword" argument for choosing which direction to apply. Second, the original implementation was optimized for the fast vector outer product (that I got added to SparseArrays in JuliaLang/julia#24980 and made it into Julia v1.2), but when scaling up to multiple columns the performance was disastrous because the dispatch of a transposed view led to generic matrix multiplication which did the full dense-dense style loops. By not using views, we get the desired sparse matrix multiplication instead.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
First, this removes the option to do row-wise quadratic products since
they aren't being used within this package anyway. That allows removing
the "keyword" argument for choosing which direction to apply.
Second, the original implementation was optimized for the fast vector
outer product (that I got added to SparseArrays in JuliaLang/julia#24980
and made it into Julia v1.2), but when scaling up to multiple columns
the performance was disastrous because the dispatch of a transposed
view led to generic matrix multiplication which did the full
dense-dense style loops. By not using views, we get the desired
sparse matrix multiplication instead.