H2OGradientBoostingEstimator train fails on simple example with sample_rate < 1

Description

I'm running h2o 3.26.0.2 on python 3.6.9 and getting an error on a simple example when all I change is setting `sample_rate` < 1. I get the error on both MacOS and Linux. Thanks in advance! Jason

Here is a reprex.

Status

Assignee

Jan Sterba

Fix versions

Reporter

Jason Muhlenkamp

Support ticket URL

None

Labels

None

Affected Spark version

None

Customer Request Type

None

Task progress

None

CustomerVisible

No

Priority

Major
Configure