Currently find_nearest only returns a single index. If the closest value is a duplicated value it would be helpful to return all indices where the value is the same.
find_nearest
