3 ms·
>should rather consider the densities of the vector space and use that as a weight of the errors at different regions. Sounds like the n-dimensional version of
by ur-whale 6y ago
>should rather consider the densities of the vector space and use that as a weight of the errors at different regions.
Sounds like the n-dimensional version of an octree