This PR introduces a function to measure "chance agreement" (see #49 for details). This value can be used as a baseline to compare against the accuracy of a classification model.
This baseline is also included as a new column in the performance table produced by make_performance_table().
Fixes #49.
Description
This PR introduces a function to measure "chance agreement" (see #49 for details). This value can be used as a baseline to compare against the accuracy of a classification model.
This baseline is also included as a new column in the performance table produced by
make_performance_table().To compute the chance agreement we use the new function introduced by this PR: https://github.com/BlueBrain/morphoclass/blob/ed2d4c974dbb5188180b13228b55d2959768c060/src/morphoclass/metrics.py#L25-L26
How to test?
See
tests/.Checklist