[
https://issues.apache.org/jira/browse/SPARK-61229?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ruifeng Zheng updated SPARK-61229:
----------------------------------
Fix Version/s: connect-rust-4.3.0
> Support native KMeans training and prediction in the Rust client
> ----------------------------------------------------------------
>
> Key: SPARK-61229
> URL: https://issues.apache.org/jira/browse/SPARK-61229
> 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 a native KMeans estimator and fitted model using the Spark Connect ML Fit
> command and cached model reference. Expose all 12 constructor parameter
> setters in Rust and the Python binding. Validate held-out cluster
> memberships, separately trained models, sample weights, seed reproducibility,
> Euclidean and cosine distance, row and block solvers, invalid inputs and
> session isolation against Spark 4.3.0 and official PySpark.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
