[pymvpa] _clf_internals question

[Yaroslav Halchenko]

well... unittests were failing not due to convergence issues, but
because the _R was treated as classifier unless you explicitely provided
'regression=True' to it. That caused unittests to fail while checking
the .confusion in regression tests.

I am yet to explore glmnet on real data ;)

On Wed, 27 May 2009, Per B. Sederberg wrote:

> Indeed, I couldn't prevent non-convergence errors, so some unit tests
> fail, but I have had no issue with either classification or regression
> on actual data.  Have you?

-- 
                                  .-.
=------------------------------   /v\  ----------------------------=
Keep in touch                    // \\     (yoh@|www.)onerussian.com
Yaroslav Halchenko              /(   )\               ICQ#: 60653192
                   Linux User    ^^-^^    [175555]