QED anomalous dimensions

[felixhekhorn]

No description provided.

[codecov]

Codecov Report

Merging #111 (246d84e) into develop (acd65aa) will not change coverage.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           develop      #111    +/-   ##
==========================================
  Coverage   100.00%   100.00%            
==========================================
  Files           36        38     +2     
  Lines         2575      2732   +157     
==========================================
+ Hits          2575      2732   +157     

Flag	Coverage Δ
isobench	54.06% <34.81%> (-0.78%)	⬇️
unittests	100.00% <100.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/eko/anomalous_dimensions/__init__.py	100.00% <100.00%> (ø)
src/eko/anomalous_dimensions/aem1.py	100.00% <100.00%> (ø)
src/eko/anomalous_dimensions/aem2.py	100.00% <100.00%> (ø)
src/eko/anomalous_dimensions/as1aem1.py	100.00% <100.00%> (ø)
src/eko/constants.py	100.00% <100.00%> (ø)

[felixhekhorn]

We need a section on "Sum Rules" in the docs, because it's a relevant theory ingredient and we're using all the time in testing.

fine - I guess we can add something under Mellin (as a special example); @niclaurenti can you please take care?

[niclaurenti]

@alecandido @felixhekhorn yes sure. I can take caro of it

[alecandido]

I haven't reviewed the expressions themselves, since this would require almost as much work as it has been done to implement them...
On this respect, I just trusts @niclaurenti

[niclaurenti]

I haven't reviewed the expressions themselves, since this would require almost as much work as it has been done to implement them...
On this respect, I just trusts @niclaurenti

I have reviewed multiple times but I'm trustful from the fact that they satisfy the sum rules.

[alecandido]

I haven't reviewed the expressions themselves, since this would require almost as much work as it has been done to implement them...
On this respect, I just trusts @niclaurenti

I have reviewed multiple times but I'm trustful from the fact that they satisfy the sum rules.

Me too, that's why I reviewed the tests. However, we had some bugs passing sum rules, e.g. normalizations, so I hope you took care enough in the overall implementation :D

[alecandido]

Thank you very much @niclaurenti

[niclaurenti]

@felixhekhorn @niclaurenti

We need a section on "Sum Rules" in the docs, because it's a relevant theory ingredient and we're using all the time in testing.

In particular, I was stupidly wondering "Why do we have multiple tests on momentum sum rules (for quark, gluon, and photon) if the conserved quantity is a single one?". I knew that being anomalous dimensions an operator (PDF in, PDF out) there should have been a clever way to break down on one of the two sides, but it took me quite some time (~5 min) in order to finally work out that I'm free to choose a border condition with only gluon, e.g., and the momentum should be preserved, but I'm ruling out all the other anomalous dimensions not having gluon on the right.

I'm not saying that is a very difficult topic that needs explanation, but explicit is better than implicit.

Do you think I should add this discussion or it's better to just give the sum rules

[alecandido]

Do you think I should add this discussion or it's better to just give the sum rules

If you wish, you can add a brief paragraph about why all these sum rules stem from a single integral relation, i.e. the proton momentum conservation. I leave the choice up to you.

[alecandido]

I'm actually tempted to merge at this point, what do you think @felixhekhorn?

[niclaurenti]

@alecandido @felixhekhorn before merging I wanted to slightly modify the .tex file (for example writing Pqq-P+=Pps) and add a section that explains what has to be changed when passing to O(aem2)

[alecandido]

@niclaurenti you'll merge by yourself, so nothing will happen until you're ready (if you make any significant change to the package itself, ask again for review).

[felixhekhorn]

I think it's fine (still maybe an explicit approval might have been good 😇 )