#3390: r.learn.ml crossvalidation fails if category map class codes increase more
then +1
--------------------+-------------------------
Reporter: maitl | Owner: grass-dev@…
Type: defect | Status: new
Priority: normal | Milestone: 7.2.2
Component: Addons | Version: unspecified
Keywords: | CPU: All
Platform: All |
--------------------+-------------------------
Crossvalidation fails if class ID values in category maps do not increase
strictly by +1 i.e 1, 2, 3, 4, 5, 6 ... .
E.g. a cass code list 10, 11, 12, 13, 25, 26 is not handled properly.
The bug is here:
if mode == 'classification' and cv > np.histogram(
y, bins=len(np.unique(y)))[0].min():
The solution is to use
if mode == 'classification' and cv > np.histogram(
y, bins=np.unique(y))[0].min():