GitHub user chouqin opened a pull request:
https://github.com/apache/spark/pull/2595
[SPARK-3366][MLLIB]Compute best splits distributively in decision tree
Currently, all best splits are computed on the driver, which makes the
driver a bottleneck for both communication and computation. This PR fix this
problem by computed best splits on executors.
Instead of send all aggregate stats to the driver node, we can send
aggregate stats for a node to a particular executor, using `reduceByKey`
operation, then we can compute best split for this node there.
Implementation details:
Each node now has a nodeStatsAggregator, which save aggregate stats for all
features and bins.
First use mapPartition to compute node aggregate stats for all nodes in
each partition.
Then transform node aggregate stats to (nodeIndex, nodeStatsAggregator)
pairs and use to `reduceByKey` operation to combine nodeStatsAggregator for the
same node.
After all stats have been combined, best splits can be computed for each
node based on the node aggregate stats. Best split result is collected to
driver to construct the decision tree.
CC: mengxr manishamde jkbradley, please help me review this, thanks.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/chouqin/spark dt-dist-agg
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/2595.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 #2595
----
commit 532993a8274e521795de50d348dbae2a4283f79b
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T07:05:37Z
Compute best splits distributively in decision tree
commit f00fc22594a936f1abc22ab53365223bf5071d01
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T07:24:29Z
fix bug
commit 4f564961f699f6a453356de889a80461af8891f8
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T07:29:21Z
fix bug
commit 5f63d6ce2f43c17bc1758bc2d067c71ef2111220
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T08:10:45Z
add multiclassification using One-Vs-All strategy
commit 24eacd8ec2770b3ed9cba5d67ed14bc0d3809170
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T08:23:15Z
fix type-mismatch bug
commit 748bd456809da07b60a984531e83a6b821f71934
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T08:27:39Z
fix type-mismatch bug
commit e76414fd3fe6df19b9352ff6690f12c7d40ee803
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T08:38:19Z
fix testsuite
commit b5b0bc2bafe37510adc79bdbda4d2664456fe738
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T08:45:15Z
fix style
commit a75df27e015b37cd620bc2400c418495ccf9ab39
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T09:40:58Z
fix test suite
commit bdd2a63f2b90058141e16f996984df955c5687a7
Author: qiping.lqp <[email protected]>
Date: 2014-09-30T10:22:08Z
fix test suite
commit 7bbb787c5eaf6652e8300c8f0fb4454b555d6d95
Author: chouqin <[email protected]>
Date: 2014-09-30T16:31:18Z
add comments
----
---
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]
