[
https://issues.apache.org/jira/browse/SPARK-11084?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14954861#comment-14954861
]
Apache Spark commented on SPARK-11084:
--------------------------------------
User 'zero323' has created a pull request for this issue:
https://github.com/apache/spark/pull/9098
> SparseVector.__getitem__ should check if value can be non-zero before
> executing searchsorted
> --------------------------------------------------------------------------------------------
>
> Key: SPARK-11084
> URL: https://issues.apache.org/jira/browse/SPARK-11084
> Project: Spark
> Issue Type: Improvement
> Components: MLlib, PySpark
> Affects Versions: 1.3.0, 1.4.0, 1.5.0, 1.6.0
> Reporter: Maciej Szymkiewicz
> Priority: Minor
>
> At this moment SparseVector.\_\_getitem\_\_ executes np.searchsorted first
> and checks if result is in an expected range after that:
> {code}
> insert_index = np.searchsorted(inds, index)
> if insert_index >= inds.size:
> return 0.
> row_ind = inds[insert_index]
> ...
> {code}
> See: https://issues.apache.org/jira/browse/SPARK-10973
> It is possible to check if index can contain non-zero value before binary
> search:
> {code}
> if (inds.size == 0) or (index > inds.item(-1)):
> return 0.
> insert_index = np.searchsorted(inds, index)
> row_ind = inds[insert_index]
> ...
> {code}
> It is not a huge improvement but should save some work on large vectors.
--
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
