Marco Gaido created SPARK-23568:
-----------------------------------
Summary: Silhouette should get number of features from metadata if
available
Key: SPARK-23568
URL: https://issues.apache.org/jira/browse/SPARK-23568
Project: Spark
Issue Type: Improvement
Components: ML
Affects Versions: 2.4.0
Reporter: Marco Gaido
In Silhouette computation we need to know the number of features. This is done
taking the first row and checking the size of the features vector. Despite it
works fine, if the number of attributes is present in the metadata of the
column, we can avoid the additional job which is generated by using `first`.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
