spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stavros Kontopoulos (JIRA)" <>
Subject [jira] [Commented] (SPARK-26247) SPIP - ML Model Extension for no-Spark MLLib Online Serving
Date Mon, 03 Dec 2018 21:47:00 GMT


Stavros Kontopoulos commented on SPARK-26247:

Some pipelines export models in pmml (I know there is a long history about, license etc) but
would be better to standardize on something? This would minimize a few issues mentioned in
this quote "Recreating and reimplementing model reading outside of the Spark project is costly
and error prone, and over time is hard to keep in-sync with Spark changes over various Spark

> SPIP - ML Model Extension for no-Spark MLLib Online Serving
> -----------------------------------------------------------
>                 Key: SPARK-26247
>                 URL:
>             Project: Spark
>          Issue Type: Improvement
>          Components: MLlib
>    Affects Versions: 2.1.0
>            Reporter: Anne Holler
>            Priority: Major
>              Labels: SPIP
>         Attachments: SPIPMlModelExtensionForOnlineServing.pdf
> This ticket tracks an SPIP to improve model load time and model serving interfaces for
online serving of Spark MLlib models.  The SPIP is here
> []
> The improvement opportunity exists in all versions of spark.  We developed our set of
changes wrt version 2.1.0 and can port them forward to other versions (e.g., we have ported
them forward to 2.3.2).

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message