Pdf fkxplot bins

Created by: wilsonmr

Suggestion on how to combine multiple datasets and fktables.

My suggestion is that binning any finer than the export grid is sort of pointless because the PDFs we produce are coarse grained to that when we save them. This has a few caveats but I think it's a reasonable jumping off point. I make the export grid more widely available and then bin the fk table x grids except apply weights which are equal to the differential cross section plotted in the latest iteration of the plot.

If I comment out the part of the code which is currently bugged (and labels the kinematics of each data point which is not used here) I produce: https://vp.nnpdf.science/FW0NdyZYQYu5kih6wusZFA== with 4.0

I still believe that we want to be able to produce the same plot but without elementwise product with one of the PDFs, because I think we want to have an effective DIS FK table (which for DIS observables is just FK table but for hadronic processes is partially convoluted with one of the PDFs) because we want to just see where the FK table points are without looking at the effect of the PDF (as much as possible) but this plot is also quite nice because it gives a slightly better repr. of the data region in terms of the chi2