Kousuke Saruta created SPARK-35530:
--------------------------------------
Summary: Fix rounding error in DifferentiableLossAggregatorSuite
with Java 11
Key: SPARK-35530
URL: https://issues.apache.org/jira/browse/SPARK-35530
Project: Spark
Issue Type: Bug
Components: ML, Tests
Affects Versions: 3.2.0
Reporter: Kousuke Saruta
Assignee: Kousuke Saruta
I'm personally checking whether all the tests pass with Java 11 for the current
master and I found DifferentiableLossAggregatorSuite fails.
https://github.com/sarutak/spark/runs/2661859541?check_suite_focus=true#step:9:13895
The reason seems that the implementation of Blas.daxpy is different between for
Java 8 and Java 11. For Java 11, Math.fma is used.
https://github.com/luhenry/netlib/blob/v2.2.0/blas/src/main/java/dev/ludovic/netlib/blas/Java8BLAS.java#L92
https://github.com/luhenry/netlib/blob/0053ea30b11686336cbdb8c7fceb41d59d268fa2/blas/src/main/java/dev/ludovic/netlib/blas/Java11BLAS.java#L40
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
