LuciferYang commented on pull request #29857:
URL: https://github.com/apache/spark/pull/29857#issuecomment-698111573
cc @srowen The remaining failed case is
```
RandomForestRegressorSuite:
- training with sample weights *** FAILED ***
0.756 was not greater than or equal to 0.78 (MLTestingUtils.scala:285)
```
https://github.com/apache/spark/blob/0bc0e91e4015eb98bd2f4bf17da2ec7135b520a9/mllib/src/test/scala/org/apache/spark/ml/regression/RandomForestRegressorSuite.scala#L171-L200
Input `(50, 10, 0.95, 0.78)` with
```
MLTestingUtils.testOversamplingVsWeighting[RandomForestRegressionModel,
RandomForestRegressor](df.as[LabeledPoint], estimator,
MLTestingUtils.modelPredictionEquals(df, _ ~= _ relTol 0.2, tol),
seed)
```
failed.
I found that the following `RandomForest.runBagged` behave differently for
the same input in Scala 2.12 and Scala 2.13, maybe related to the follow code
block:
https://github.com/apache/spark/blob/0bc0e91e4015eb98bd2f4bf17da2ec7135b520a9/mllib/src/main/scala/org/apache/spark/ml/tree/impl/RandomForest.scala#L191-L215
but I am not familiar with this algorithm and I not find root cause, I think
we need an expert to guide how to fix it
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
