sissa-data-science/DADApy

range_max argument in return_id_scaling_gride does not indicate the actual max data range

Closed this issue · 1 comments

Subject of the issue

Specify better the behaviour of range_max argument in return ID scaling gride

For some dataset (e.g. helm-mmlu-anatomy?) the range_max argument does behave properly: it computes one scale less than expected

Also, add at the output of the ID methods the "scales" over which the IDs are computed (N, k) rather that the distances over which the IDs are computed