Closed RokIvansek closed 8 years ago
Make sure your implementation handles missing values appropriately. The entropy function should not count NAs as a valid value when calculating probability distribution of an attribute.
Solved using np.isnan which is supposedly the proper way according to: http://docs.orange.biolab.si/3/data-mining-library/tutorial/data.html#missing-values
Make sure your implementation handles missing values appropriately. The entropy function should not count NAs as a valid value when calculating probability distribution of an attribute.