[jira] [Commented] (SPARK-10413) Model should support prediction on single instance

["Hussein Hazimeh (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-10413?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15547855#comment-15547855 ] 

Hussein Hazimeh commented on SPARK-10413:
-----------------------------------------

As discussed in [SPARK-16431|https://issues.apache.org/jira/browse/SPARK-16431], I think it would useful to extend this further to support feature transformations on single instances, which enables low-latency feature transformations on single instances and can lead to improved code readability and testing. As a first step, the body of the UDF of each feature transformer can be refactored into a new low-level method (e.g. called "transformInstance") that accepts single instances with raw data types (double, vector, etc). The new method can be utilized later on to add full pipeline support for transforming and predicting single instances. [~josephkb] let me know your thoughts on this.

> Model should support prediction on single instance
> --------------------------------------------------
>
>                 Key: SPARK-10413
>                 URL: https://issues.apache.org/jira/browse/SPARK-10413
>             Project: Spark
>          Issue Type: Umbrella
>          Components: ML
>            Reporter: Xiangrui Meng
>            Priority: Critical
>
> Currently models in the pipeline API only implement transform(DataFrame). It would be quite useful to support prediction on single instance.
> UPDATE: This issue is for making predictions with single models.  We can make methods like {{def predict(features: Vector): Double}} public.
> * This issue is *not* for single-instance prediction for full Pipelines, which would require making predictions on {{Row}}s.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org