This will shrink the x1- and x2-grids such that the minimal and maximal values of each dimension reflect the range of non-zero subgrid values. That should also shrink the size of grids a bit (see https://github.com/NNPDF/pineappl/issues/151), however the main advantage is that one can now read off the x ranges for each subgrid, which can be helpful sometimes. Finally, the generation of EKOs should be faster since the operators have to be generated for fewer points.
This will shrink the
x1
- andx2
-grids such that the minimal and maximal values of each dimension reflect the range of non-zero subgrid values. That should also shrink the size of grids a bit (see https://github.com/NNPDF/pineappl/issues/151), however the main advantage is that one can now read off the x ranges for each subgrid, which can be helpful sometimes. Finally, the generation of EKOs should be faster since the operators have to be generated for fewer points.