spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Villu Ruusmann (JIRA)" <>
Subject [jira] [Commented] (SPARK-8546) PMML export for Naive Bayes
Date Thu, 25 Jun 2015 21:54:04 GMT


Villu Ruusmann commented on SPARK-8546:

The NB exporter should be easier to implement than Logistic regression and SVM exporters,
because the PMML representation of NB models does not distinguish between binary and multi-class
classification cases.

I'm currently developing PMML exporters for the Python ML module called Scikit-Learn. Unfortunately,
I haven't made to Scikit-Learn classes GaussianNB and MultinomialNB yet. I shall be much wiser
once I get them done.

> PMML export for Naive Bayes
> ---------------------------
>                 Key: SPARK-8546
>                 URL:
>             Project: Spark
>          Issue Type: New Feature
>          Components: MLlib
>            Reporter: Joseph K. Bradley
>            Priority: Minor
> The naive Bayes section of PMML standard can be found at
We should first figure out how to generate PMML for both binomial and multinomial naive Bayes
models using JPMML (maybe [~vfed] can help).

This message was sent by Atlassian JIRA

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

View raw message