[
https://issues.apache.org/jira/browse/SPARK-30641?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhengruifeng updated SPARK-30641:
---------------------------------
Description:
We had been refactoring linear models for a long time, and there still are some
works in the future:
# *Blockification (vectorization of vectors)*
** vectors are stacked into matrices, so that high-level BLAS can be used for
better performance. (about ~3x faster on sparse datasets, up to ~15x faster on
dense datasets). Since 3.1.1, LoR/SVC/LiR/AFT supports blockification, and we
need to blockify KMeans in the future.
# *Standardization*
was:
stacking input vectors into blocks will benefit ML algs:
1, less RAM to persist datasets, since the overhead of object header is reduced;
2, optimization potential for impl, since high-level BLAS can be used; Proven
in ALS/MLP;
3, maybe a way to perform efficient mini-batch sampling (To be confirmed)
> Project Matrix: Linear Models revisit and refactor
> --------------------------------------------------
>
> Key: SPARK-30641
> URL: https://issues.apache.org/jira/browse/SPARK-30641
> Project: Spark
> Issue Type: New Feature
> Components: ML, PySpark
> Affects Versions: 3.1.0, 3.2.0
> Reporter: zhengruifeng
> Assignee: zhengruifeng
> Priority: Major
>
> We had been refactoring linear models for a long time, and there still are
> some works in the future:
> # *Blockification (vectorization of vectors)*
> ** vectors are stacked into matrices, so that high-level BLAS can be used
> for better performance. (about ~3x faster on sparse datasets, up to ~15x
> faster on dense datasets). Since 3.1.1, LoR/SVC/LiR/AFT supports
> blockification, and we need to blockify KMeans in the future.
> # *Standardization*
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
