Github user miro-balaz commented on the issue:
https://github.com/apache/spark/pull/7266
thank you for directions
On Monday, 12 September 2016, Holden Karau <notificati...@github.com> wrote:
> @miro-balaz <https://github.com/miro-balaz> : This
Github user miro-balaz commented on the issue:
https://github.com/apache/spark/pull/7266
For me it would be usefull(when working with trees) if string indexer was
mapping unseen labels to maximum known label+1. The row would not be skipped if
it contained some non-used feature