[BUG] fixes and improvements to all_objects#69
Conversation
Codecov Report
@@ Coverage Diff @@
## main #69 +/- ##
=======================================
Coverage 54.80% 54.80%
=======================================
Files 16 16
Lines 1352 1352
=======================================
Hits 741 741
Misses 611 611
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. |
|
maybe we should also add a test for this? |
|
@fkiraly there is still a bit of formatting weirdness in the Do you mind taking a look at that and updating the format? I'm going to be opening up a separate PR for the tests. |
Just looked at it, looks ok to me? I also did not change the general format or indentation, I think? Do you mean the "note" boxes? I thought this was ok, but we can change the indentation. |
|
@fkiraly I'm fine leaving the docstring as is for now. I thought I was seeing one part bolded that didn't seem like it should be (the boxes always through me for a loop, but I get its an indentation thing). I think we have an open issue related to the docstring for this, but I'll double-check when I add the tests for lookup functionality. |
RNKuhns
left a comment
There was a problem hiding this comment.
LGTM. Will merge and handle tests and potential tweak to docstring in forthcoming PR.
This fixes some bugs and makes improvements to
all_objectsthat are insktime.Bugs:
issubclass, notisinstance, since we filter classes, not objects. This seems to be correct insktimebut is wrong inskbase. Fixes [BUG]all_objectsargumentobject_typesdoes not work #71Improvements:
modules_to_ignoreparametersall_objectsfixed to make reference to objects, not estimators