[pymvpa] bug reporting

Yaroslav Halchenko debian at onerussian.com
Wed Sep 2 04:21:57 UTC 2009


our little kNN (which I believe started the whole PyMVPA) received some
TLC.  Tara managed to hit the kNN specification we managed not to test
at all (basic majority voting not-weighted) -- so, I guess, it was
always broken ;)

long story short -- everything is fixed, both 'majority' and 'weighted'
are now parts of the unittests battery.

Fixes available within master branch (it is to be released very very
soon as 0.4.3) -- give it a try (since now you can build from the
sources ;))

On Tue, 01 Sep 2009, tg at cs.york.ac.uk wrote:
> k = kNN(k=1, voting='majority') #other values of k cause the same result
> k.train(data)
> k.predict(test)

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





More information about the Pkg-ExpPsy-PyMVPA mailing list