[jira] [Updated] (MADLIB-1517) Do not include intercept as predictor for regression models

["Nikhil Kak (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MADLIB-1517?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Nikhil Kak updated MADLIB-1517:
-------------------------------
    Summary: Do not include intercept as predictor for regression models  (was: PMML for Decision Tree and Logistic regression is not valid)

> Do not include intercept as predictor for regression models
> -----------------------------------------------------------
>
>                 Key: MADLIB-1517
>                 URL: https://issues.apache.org/jira/browse/MADLIB-1517
>             Project: Apache MADlib
>          Issue Type: Bug
>          Components: Module: Utilities
>            Reporter: Ekta Khanna
>            Priority: Major
>             Fix For: v2.2.0
>
>
> Currently PMML export utility ({{madlib.pmml()}}) in madlib supports exporting decision tree, logistic, linear regression model as a pmml file. When trying to read this file by a python code, we see the following:
> 1. For decision tree model : Fails for invalid XML
> 2. Logistic/Linear regression: The model isn't exported correctly as the predictions are off.
> Belos is the code in python to import the logreg pmml generated as shown in [logreg example|https://madlib.apache.org/docs/latest/group__grp__pmml.html]: 
> {code}
> from pypmml import Model
> model = Model.fromFile('logreg.pmml')
> {code}
> As part of this JIRA, need to fix these PMML exports that can be imported by other apps.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)