N3LO matching#83
Conversation
Codecov Report
@@ Coverage Diff @@
## develop #83 +/- ##
==========================================
Coverage 100.00% 100.00%
==========================================
Files 36 65 +29
Lines 2575 3060 +485
==========================================
+ Hits 2575 3060 +485
Flags with carried forward coverage won't be shown. Click here to find out more.
|
felixhekhorn
added
enhancement
This reverts commit 107544a.
|
What are we missing to close this PR? |
|
I'm reviewing right now |
There was a problem hiding this comment.
From now on, we should refrain to address multiple big tasks in a single PR:
- the
harmonicsbranch should have not been merged in here, because it has been reviewed twice in this way - the
numbacache enhancement was really needed, and even related to this PR, but it has been scattered all over, blowing up the number of files, so we should have done in a separate one (and merged immediately on develop, updating afterwards this, or otherwise merge this one before, and next apply thenumbachore from a second PR, we havedevelopbranch exactly for this reason...)
P.S.: of course, I didn't check details and numbers of every single function, I checked tests and relying on them, but MMa numbers I had no way to check, just relying on you
| gqq1p_cfca = -16*g3n + (132 - n*(340 + n*(655 + 51*n*(2 + n))))/(18.*np.power(n,2)*np.power(1 + n,2)) + (-14.666666666666666 + 8/n - 8/(1 + n))*S2 - (4*Sp2p)/(n + np.power(n,2)) + S1*(29.77777777777778 - 16/np.power(n,2) - 16*S2 + 8*Sp2p) + 2*Sp3p + 10*zeta3 + zeta2*(16*S1 - 16*Sp1p + 16*(1/n - np.log(2))) # pylint: disable=line-too-long | ||
| gqq1p_cfcf = 32*g3n - (8 + n*(32 + n*(40 + 3*n*(3 + n)*(3 + np.power(n,2)))))/(2.*np.power(n,3)*np.power(1 + n,3)) + (12 - 8/n + 8/(1 + n))*S2 + S1*(40/np.power(n,2) - 8/np.power(1 + n,2) + 16*S2 - 16*Sp2p) + (8*Sp2p)/(n + np.power(n,2)) - 4*Sp3p - 20*zeta3 + zeta2*(-32*S1 + 32*Sp1p + 32*(-(1/n) + np.log(2))) # pylint: disable=line-too-long | ||
| gqq1p_cfca = -16*g3n + (132 - n*(340 + n*(655 + 51*n*(2 + n))))/(18.*np.power(n,2)*np.power(1 + n,2)) + (-14.666666666666666 + 8/n - 8/(1 + n))*S2 - (4*Sp2p)/(n + np.power(n,2)) + S1*(29.77777777777778 - 16/np.power(n,2) - 16*S2 + 8*Sp2p) + 2*Sp3p + 10*zeta3 + zeta2*(16*S1 - 16*Sp1p + 16*(1/n - log2)) # pylint: disable=line-too-long | ||
| gqq1p_cfcf = 32*g3n - (8 + n*(32 + n*(40 + 3*n*(3 + n)*(3 + np.power(n,2)))))/(2.*np.power(n,3)*np.power(1 + n,3)) + (12 - 8/n + 8/(1 + n))*S2 + S1*(40/np.power(n,2) - 8/np.power(1 + n,2) + 16*S2 - 16*Sp2p) + (8*Sp2p)/(n + np.power(n,2)) - 4*Sp3p - 20*zeta3 + zeta2*(-32*S1 + 32*Sp1p + 32*(-(1/n) + log2)) # pylint: disable=line-too-long |
There was a problem hiding this comment.
Okay, caching even np.log(2) it is rather extreme, but fine of course...
| # TODO: shall we use: 0.4 * 16.0 / (1.0 - logx) ? | ||
| self.r = 0.4 * 16.0 / (-logx) | ||
| # TODO: shall we use: 0.4 * 16.0 / ( - logx) ? | ||
| self.r = 0.4 * 16.0 / (1.0 - logx) |
There was a problem hiding this comment.
I guess this has been tested, are the benefits documented somewhere? It might be useful for the future.
Consider writing something either here in the docstring, or in the Mellin page of the docs.
|
The only relevant point is documenting Mellin path choice #83 (comment), but we can even postpone and just merge, at this point we have to do it as soon as possible. |
|
I suggest to close and instead postpone the documentation to #116 |
|
I also updated dependencies with Poetry, so everything is alright. I leave up to @giacomomagni the task to resolve conflicts with |
3 participants
Integrate the eko relevant features coming from #74 inside develop