[ 
https://issues.apache.org/jira/browse/SPARK-61450?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Ruifeng Zheng updated SPARK-61450:
----------------------------------
    Fix Version/s: connect-rust-4.3.0

> Support native DecisionTreeClassifier training and prediction
> -------------------------------------------------------------
>
>                 Key: SPARK-61450
>                 URL: https://issues.apache.org/jira/browse/SPARK-61450
>             Project: Spark
>          Issue Type: New Feature
>          Components: Connect
>    Affects Versions: connect-rust-4.3.0
>            Reporter: Ruifeng Zheng
>            Priority: Major
>             Fix For: connect-rust-4.3.0
>
>
> Add native Rust DecisionTreeClassifier and DecisionTreeClassificationModel 
> using the Connect fit command and session-scoped model cache. Expose all 18 
> official PySpark constructor parameters through Rust setters/getters and 
> native Python bindings. Validate training, predictions, instance weights, 
> class thresholds, raw weighted class counts, scalar leaf-index output and 
> model lifetime through golden fixtures and live Rust/native Python tests. 
> Tree inspection, single-vector prediction and persistence are outside this 
> native API addition.



--
This message was sent by Atlassian Jira
(v10.3.26#10030026)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to