search

NNPDF / nnpdf

An open-source machine learning framework for global analyses of parton distributions.
https://docs.nnpdf.science/
GNU General Public License v3.0
28 stars 6 forks source link

Effective exponents in reports are plotted always at 1.65 GeV #2030

Open niclaurenti opened 5 months ago

niclaurenti commented 5 months ago

In the effective exponents part of the reports the plots are done taking the value at 1.65 GeV instead of Q0. This is relevant if someone wants to do a fit with Q0 different from 1.65.

For example in the plot in

https://vp.nnpdf.science/QgE-00fgTe2cnIK-3Pwo4g==/figures/exponents_report_current_basisfromfit_plot_alpha_eff_Sigma.pdf

the horizontal lines are plotted for Q=Q0 (i.e. 1.65 for the orange and 1.50 for the green lines), while the bands are plotted at Q=1.65 both for the green and orange one.

scarlehoff commented 5 months ago

The report has a fixed Q all across for plotting (1.65 GeV, which is generally what we use) while the next-iteration-values (vp-nextfitruncard) are taken (correctly now xd) at Q0.

Now... when you do this plot where both the values are relevant I'm not sure what the correct solution would be... Maybe we should just drop this plot when Q =/= Q0 since it would be inconsistent?

(you can always create the plot manually by doing a report where you have two namespaces, one with Q = 1.5 GeV and another with Q = 1.65 GeV)