Support Partitioning for Visual ML Clustering Type Models
Currently one of the limitations with partitioning is being able to partition clustering type visual ml models (here). As a work around end users can either export the visual ml model as a python notebook, and/or create their own python script to build their clustering models with partitions.
We actively leverage cluster models with partitions for various use cases: from segmenting field representatives for numerous regions/countries, to identifying patient sub-groups for different populations in a therapy area, and so on.
By supporting partitioning for visual ml clustering models this will help expedite scaling model creation across the enterprise and make available the solution to additional users who aren't as familiar with python code.
Comments
-
Krishna Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Product Ideas Manager Posts: 18 Dataiker
Thanks @kathyqingyuxu
! Adding to the backlog -
I would also like this feature.