Hyukjin Kwon created SPARK-61709:
------------------------------------
Summary: Add PySpark's Params API, param maps and persistence to
the native evaluators
Key: SPARK-61709
URL: https://issues.apache.org/jira/browse/SPARK-61709
Project: Spark
Issue Type: Improvement
Components: Connect
Affects Versions: connect-rust-4.3.0
Reporter: Hyukjin Kwon
Part of completing the native ML classes of pyspark.ml.connect to the full
pyspark.ml 4.3.0 surface, after the PyO3 base (SPARK-61697).
The six native evaluators (BinaryClassificationEvaluator,
MulticlassClassificationEvaluator, MultilabelClassificationEvaluator,
RegressionEvaluator, ClusteringEvaluator, RankingEvaluator) get PySpark's API
of pyspark.ml.evaluation:
- keyword-only constructors, setParams, every getter and setter, and the Params
API;
- evaluate(dataset, params), which evaluates a copy of a non-empty param map,
and isLargerBetter;
- write, save, read and load, which keep the uid and the params, as
RemoteMLWriter and RemoteMLReader save and load them.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]