You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In linear_model/omp.py, function _gram_omp, if optimization ends early due to warn/break right after the while statement, then gamma never gets set before the return. This then raises an UnboundLocalError.
The warning displayed is helpful, but it'd be nice to also set some default value of gamma initially, rather than just bailing. I'd fix and submit a pull request, but I have no idea what an appropriate default value for gamma is. :)
Sorry that I don't have a handy test case for this; I encountered it in the middle of a very long test run (> 1 day) over a lot of data.
The text was updated successfully, but these errors were encountered:
On Jan 6, 2012, at 20:50 , Josh Bleecher Snyder wrote:
In linear_model/omp.py, function _gram_omp, if optimization ends early due to warn/break right after the while statement, then gamma never gets set before the return. This then raises an UnboundLocalError.
The warning displayed is helpful, but it'd be nice to also set some default value of gamma initially, rather than just bailing. I'd fix and submit a pull request, but I have no idea what an appropriate default value for gamma is. :)
Sorry that I don't have a handy test case for this; I encountered it in the middle of a very long test run (> 1 day) over a lot of data.
Noted, thank you for submitting. I have actually run into this before and forgot about it because of bigger issues. Now is the time to fix it :)
Reply to this email directly or view it on GitHub: #540
In
linear_model/omp.py
, function_gram_omp
, if optimization ends early due towarn
/break
right after thewhile
statement, thengamma
never gets set before the return. This then raises anUnboundLocalError
.The warning displayed is helpful, but it'd be nice to also set some default value of
gamma
initially, rather than just bailing. I'd fix and submit a pull request, but I have no idea what an appropriate default value forgamma
is. :)Sorry that I don't have a handy test case for this; I encountered it in the middle of a very long test run (> 1 day) over a lot of data.
The text was updated successfully, but these errors were encountered: