GitHub user mengxr opened a pull request:
https://github.com/apache/spark/pull/4144
[SPARK-3424][MLLIB] cache point distances during k-means|| init
This PR ports the following feature implemented in #2634 by @derrickburns:
* During k-means|| initialization, we shouldn't cache costs (squared
distances) previously computed.
It also contains the following optimization:
* aggregate sumCosts directly
* ran multiple (#runs) k-means++ in parallel
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/mengxr/spark SPARK-3424-kmeans-parallel
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/4144.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #4144
----
commit 4341bb8b4916f8e597e803e3007d379d557a3206
Author: Xiangrui Meng <[email protected]>
Date: 2015-01-21T20:31:22Z
do not re-compute point distances during k-means||
----
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
