You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mahout.apache.org by "Robin Anil (JIRA)" <ji...@apache.org> on 2013/04/13 22:42:16 UTC
[jira] [Updated] (MAHOUT-1190) SequentialAccessSparseVector
function assignment is very slow
[ https://issues.apache.org/jira/browse/MAHOUT-1190?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Robin Anil updated MAHOUT-1190:
-------------------------------
Attachment: MAHOUT-1190.patch
As it turns out the Non Default iterator on the RandomAccessSparseVector was looking up hashmap for values. By iterating and copying a parallel array, I was able to speed up all operations of RandomAccessSparseVector by 25-45% on dot products and distance measures. Its almost twice as fast as SASV on dot product even after applying Dan's Patch
> SequentialAccessSparseVector function assignment is very slow
> -------------------------------------------------------------
>
> Key: MAHOUT-1190
> URL: https://issues.apache.org/jira/browse/MAHOUT-1190
> Project: Mahout
> Issue Type: Bug
> Reporter: Dan Filimon
> Attachments: MAHOUT-1190.patch
>
>
> Currently when calling .assign() on a SASV with another vector and a custom function, it will iterate through it and assign every single entry while also referring it by index.
> This makes the process *hugely* expensive. (on a run of BallKMeans on the 20 newsgroups data set, profiling reveals that 92% of the runtime was spent updating assigning the vectors).
> Here's a prototype patch:
> https://github.com/dfilimon/mahout/commit/63998d82bb750150a6ae09052dadf6c326c62d3d
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira