Github user srowen commented on the issue:
https://github.com/apache/spark/pull/22360
Oh right, the model, not the implementation. CC @mgaido91 too. OK we can
take it out because it was only introduced for 2.4.0.
Hm, so the model still has a distanceMeasure, but it's just not something
that should be settable.
The bisecting k-means model just wraps the older .mllib model. But I don't
see that its getDistanceMeasure would actually return the value inside that
wrapped model? I am missing where the wrapper gets updated to return the actual
distance measure used.
Is that also an issue or did I miss something?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
