[
https://issues.apache.org/jira/browse/SPARK-2426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14302932#comment-14302932
]
Debasish Das commented on SPARK-2426:
-------------------------------------
[~mengxr] [~coderxiang] David is out in Feb and I am not sure if we can cut a
breeze release with the code. I refactored NNLS to breeze.optimize.linear due
to its similarity to CG core. Proximal algorithms and QuadraticMinimizer are
refactored to breeze.optimize.proximal. It will be great if you could also
review the PR https://github.com/scalanlp/breeze/pull/321.
With this solver added to Breeze I am ready to add in ALS modifications to
Spark. The test-cases for default ALS and nnls runs fine with my Spark PR. I
need to add appropriate test-cases for sparse coding and least square loss with
lsa constraints as explained above.
Should I add them to ml.als or mllib.als since we have now two codebases ? My
current PR will merge fine with mllib.als but not with ml.als. I see there is a
CholeskySolver but all those features are supported in
breeze.optimize.proximal.QuadraticMinimizer.
> Quadratic Minimization for MLlib ALS
> ------------------------------------
>
> Key: SPARK-2426
> URL: https://issues.apache.org/jira/browse/SPARK-2426
> Project: Spark
> Issue Type: New Feature
> Components: MLlib
> Affects Versions: 1.3.0
> Reporter: Debasish Das
> Assignee: Debasish Das
> Original Estimate: 504h
> Remaining Estimate: 504h
>
> Current ALS supports least squares and nonnegative least squares.
> I presented ADMM and IPM based Quadratic Minimization solvers to be used for
> the following ALS problems:
> 1. ALS with bounds
> 2. ALS with L1 regularization
> 3. ALS with Equality constraint and bounds
> Initial runtime comparisons are presented at Spark Summit.
> http://spark-summit.org/2014/talk/quadratic-programing-solver-for-non-negative-matrix-factorization-with-spark
> Based on Xiangrui's feedback I am currently comparing the ADMM based
> Quadratic Minimization solvers with IPM based QpSolvers and the default
> ALS/NNLS. I will keep updating the runtime comparison results.
> For integration the detailed plan is as follows:
> 1. Add QuadraticMinimizer and Proximal algorithms in mllib.optimization
> 2. Integrate QuadraticMinimizer in mllib ALS
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
