[
https://issues.apache.org/jira/browse/SPARK-18755?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16247316#comment-16247316
]
Ganesh Sivalingam commented on SPARK-18755:
-------------------------------------------
[~yuhaoyan] Whilst working at Parkopedia, I have developed a solution for this
and would like to submit it to the community for review.
It is not exactly one of the two options you gave, but is closer to the first.
My approach was to create a `RandomParamGridBuilder` class which generates the
same param map data structure as `ParamGridBuilder`, but does that through
random sampling of provided parameter distributions.
The build method takes an integer as an argument which indicates how many
iterations of random search you would like to perform (and so how many
parameter maps to produce). The method can be run again to generate a new set
of random parameters.
This change needs no changes to the code base other than to add this single
class (exluding tests).
> Add Randomized Grid Search to Spark ML
> --------------------------------------
>
> Key: SPARK-18755
> URL: https://issues.apache.org/jira/browse/SPARK-18755
> Project: Spark
> Issue Type: Improvement
> Components: ML
> Reporter: yuhao yang
>
> Randomized Grid Search implements a randomized search over parameters, where
> each setting is sampled from a distribution over possible parameter values.
> This has two main benefits over an exhaustive search:
> 1. A budget can be chosen independent of the number of parameters and
> possible values.
> 2. Adding parameters that do not influence the performance does not decrease
> efficiency.
> Randomized Grid search usually gives similar result as exhaustive search,
> while the run time for randomized search is drastically lower.
> For more background, please refer to:
> sklearn: http://scikit-learn.org/stable/modules/grid_search.html
> http://blog.kaggle.com/2015/07/16/scikit-learn-video-8-efficiently-searching-for-optimal-tuning-parameters/
> http://www.jmlr.org/papers/volume13/bergstra12a/bergstra12a.pdf
> https://www.r-bloggers.com/hyperparameter-optimization-in-h2o-grid-search-random-search-and-the-future/.
> There're two ways to implement this in Spark as I see:
> 1. Add searchRatio to ParamGridBuilder and conduct sampling directly during
> build. Only 1 new public function is required.
> 2. Add trait RadomizedSearch and create new class RandomizedCrossValidator
> and RandomizedTrainValiationSplit, which can be complicated since we need to
> deal with the models.
> I'd prefer option 1 as it's much simpler and straightforward. We can support
> Randomized grid search via some smallest change.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
