[GitHub] incubator-madlib issue #48: SVM: Novelty detection using 1-class SVM

[iyerr3 <gi...@git.apache.org>]

Github user iyerr3 commented on the issue:

    https://github.com/apache/incubator-madlib/pull/48
  
    @mktal 
    Had to make considerable changes for two reasons: 
    1. With cross validation, the transformer was not being updated with the new CV data. That meant having two ways of outputing tables: using transformer or directly using args. 
    2. Cross validation for SVM one class had another issue: the origin in the transformed table was added before the CV, which meant some (most) of the train data did not have the origin. I changed it to do the origin insert within `svm_parsed_params` - that would ensure origin exists in every CV dataset. 


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---