[jira] [Updated] (MADLIB-1351) Add stopping criteria on perplexity to LDA

["Frank McQuillan (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MADLIB-1351?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Frank McQuillan updated MADLIB-1351:
------------------------------------
    Fix Version/s:     (was: v2.0)
                   v1.17

> Add stopping criteria on perplexity to LDA
> ------------------------------------------
>
>                 Key: MADLIB-1351
>                 URL: https://issues.apache.org/jira/browse/MADLIB-1351
>             Project: Apache MADlib
>          Issue Type: New Feature
>          Components: Module: Parallel Latent Dirichlet Allocation
>            Reporter: Frank McQuillan
>            Priority: Major
>             Fix For: v1.17
>
>
> In LDA 
> http://madlib.apache.org/docs/latest/group__grp__lda.html
> make stopping criteria on perplexity rather than just number of iterations.
> Suggested approach is to do what scikit-learn does
> https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.LatentDirichletAllocation.html
> evaluate_every : int, optional (default=0)
> How often to evaluate perplexity. Only used in fit method. set it to 0 or negative number to not evalute perplexity in training at all. Evaluating perplexity can help you check convergence in training process, but it will also increase total training time. Evaluating perplexity in every iteration might increase training time up to two-fold.
> perp_tol : float, optional (default=1e-1)
> Perplexity tolerance in batch learning. Only used when evaluate_every is greater than 0.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)