[Biopython-dev] [Bug 2693] LogisticRegression convergence criterion is too lenient

bugzilla-daemon at portal.open-bio.org bugzilla-daemon at portal.open-bio.org
Mon Feb 16 16:05:27 UTC 2009


http://bugzilla.open-bio.org/show_bug.cgi?id=2693





------- Comment #3 from bsouthey at gmail.com  2009-02-16 11:05 EST -------
(In reply to comment #2)
> With this patch, the test_LogisticRegression.py unit test fails.
> Could you check that?

Yes it fails because the test example does not convergence with the defaults
(try the example is R or SAS) and, thus, does not provide a valid check for
logistic regression.

> 
> Also, it is not necessary to pass old_llik to update_fn; if needed, update_fn
> can store the value of llik on each call.

I guess this is all how you define the purpose of the update_fn function.

Bruce


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.



More information about the Biopython-dev mailing list