Github user MLnick commented on a diff in the pull request:
https://github.com/apache/spark/pull/11919#discussion_r58854819
--- Diff: mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala
---
@@ -672,7 +673,7 @@ object ALS extends DefaultParamsReadable[ALS] with
Logging {
itemFactors = computeFactors(userFactors, userOutBlocks,
itemInBlocks, rank, regParam,
userLocalIndexEncoder, solver = solver)
if (shouldCheckpoint(iter)) {
- itemFactors.checkpoint()
+ ALS.checkpointAndCleanParents(itemFactors)
itemFactors.count() // checkpoint item factors and cut lineage
deletePreviousCheckpointFile()
--- End diff --
@mengxr I haven't looked at this code in quite a while - can you remind me
why in the explicit case `deletePreviousCheckpointFile` and
`previousCheckpointFile = itemFactors.getCheckpointFile` happens before
`userFactors` is computed, while in the implicit case it happens after?
---
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]
