MRG Class weight refactor #1485
Conversation
|
Sorry for pestering but I would really like to merge this soon, as I want to work on #1411, which I think is release critical. |
|
+1 again |
| - :class:`svm.SVC` and :class:`svm.NuSVC` now provide a ``classes_`` | ||
| attribute and support arbitrary dtypes for labels ``y``. | ||
| Also, the dtype returned by ``predict`` now reflects the dtype of | ||
| ``y`` during ``fit`` (used to be ``np.float``). |
There was a problem hiding this comment.
We should probably be consistent in the whole scikit for this (test_common...). Of course, this is not in the scope of this PR.
There was a problem hiding this comment.
There is already a test, that would actually test for strings, but it is commented out still, because something in the SGD didn't like it. I wasn't in the mood to investigate but will after some my current PRs are merged.
|
Apart from my comment, +1. |
|
Moved |
|
\o/ |
|
🍻 |
|
Changes Unknown when pulling 1c1538a on amueller:class_weight_refactor into * on scikit-learn:master*. |
New PR as I messed up #1464.
@ogrisel gave his +1 there.
@erg your comments would still be welcome.
Also more comments by @larsmans and @mblondel ;)