Added support for splitting regression sets #112
Conversation
Codecov Report
@@ Coverage Diff @@
## master #112 +/- ##
==========================================
+ Coverage 92.74% 93.45% +0.71%
==========================================
Files 49 50 +1
Lines 1268 1299 +31
==========================================
+ Hits 1176 1214 +38
+ Misses 92 85 -7
Continue to review full report at Codecov.
|
| @@ -43,24 +43,29 @@ def load_data(path, index, label, drop=None, verbose=True, **kwargs): | |||
| return X, y | |||
|
|
|||
|
|
|||
| def split_data(X, y, test_size=.2, random_state=None): | |||
| def split_data(X, y, regression=False, test_size=.2, random_state=None): | |||
There was a problem hiding this comment.
After discussion with Chris, there didn't seem to be a solution to automatically catching the problem type (regression vs. classification) that would satisfy all edges cases. We proposed to set a flag so theres an all-in-one function for users to split data. We saw this as more convenient than creating two separate functions that could be more ambiguous especially when the function names are not indicative of its use case.
Resolves #41