New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Heavy Ion re-reco software #1564
Conversation
A new Pull Request was created by @yetkinyilmaz for CMSSW_5_3_X. Heavy Ion re-reco software It involves the following packages: RecoHI/HiJetAlgos @nclopezo, @cmsbuild, @thspeer, @slava77 can you please review it and eventually sign? Thanks. |
+tested |
@slava77 started working on this |
|
||
protected: | ||
|
||
double pt_preeq; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why doubles?
Can we use float precision here?
Please change to float if doubles are not justified
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi.
I think that's because the standard reco::Candidate (and math:: LorentzVector) are in double, and this number is to replace the pt of candidates.
I'm not sure if using float would be a problem either. Do you have a strong opinion?
Thanks.
On Nov 29, 2013, at 1:06 PM, slava77 notifications@github.com wrote:
In DataFormats/HeavyIonEvent/interface/VoronoiBackground.h:
+#include
+#include
+
+namespace reco { class VoronoiBackground {
+public:
VoronoiBackground();
VoronoiBackground(double pt0, double pt1, double pt2, double mt0, double mt1, double mt2);
virtual ~VoronoiBackground();
- double pt() const{
return pt_corrected;
- }
+protected:
+
- double pt_preeq;
Why doubles?
Can we use float precision here?
Please change to float if doubles are not justified—
Reply to this email directly or view it on GitHub.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
no strong opinion.
ok
@yetkinyilmaz We had a pretty long discussion related to Centrality, see comments in #189 Please comment on your progress or plans with that. Thanks. |
I think the centrality team was working on this, although a bit slowly. I will ping them again about this. But if you see message: the conclusion was that we must fix these issues for the actual 7_0_X development, and for 5_3_X try to survive with this part as it is. On the other hand, if the fixes are ready early enough (by any chance), it would be very straightforward to apply them in this pull request as well, and in that case we should do it. (A question: Is there a way to ensure to notify the relevant people automatically for the pull request of packages they are responsible of? I'm not sure who are getting these messages) On Nov 29, 2013, at 1:32 PM, slava77 notifications@github.com wrote:
|
* hiEcalClusters | ||
* hiRecoJets | ||
* muonRecoPbPb | ||
* reMuonRecoPbPb |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Please remind me why do we need to run these two versions here now?
At least the naming is pretty odd now (re-reco followed immediately after the regular reco).
Do you have some slides summarizing all the changes in the HI RECO?
Some of it so far looks like a temporary patch-work.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hm, this pull request has been a good exercise - I really need to learn how make the relevant parties participate in this discussion.
I am rather related to the jet developments, but in the new reconstruction we had also updates on tracking and muons, which I only tested without going into the details of the development. I will try to figure these out.
On Nov 29, 2013, at 1:50 PM, slava77 notifications@github.com wrote:
In RecoHI/Configuration/python/Reconstruction_HI_cff.py:
* hiEcalClusters * hiRecoJets * muonRecoPbPb
Please remind me why do we need to run these two versions here now?\* reMuonRecoPbPb
At least the naming is pretty odd now (re-reco followed immediately after the regular reco).Do you have some slides summarizing all the changes in the HI RECO?
Some of it so far looks like a temporary patch-work.—
Reply to this email directly or view it on GitHub.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
yetkin you can mention people in the discussion prefixing with "@" their username e.g. "@yetkinyilmaz" and they'll be automatically notified. You can also ask to watch specific packages and you and your colleagues will receive a notification when a given package is touched. Just create an issue in https://github.com/cms-sw/cmssw/issues/new and specify who should watch which package.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok, so I change the name reMuonRecoPbPb to regionalMuonRecoPbPb, good? @MiheeJo @mandrenguyen ?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
On 12/16/13, 3:17 PM, yetkinyilmaz wrote:
In RecoHI/Configuration/python/Reconstruction_HI_cff.py:
* hiEcalClusters * hiRecoJets * muonRecoPbPb
\* reMuonRecoPbPb
Ok, so I change the name reMuonRecoPbPb to regionalMuonRecoPbPb, good?
Sounds good to me.
--slava
@MiheeJo https://github.com/miheejo @mandrenguyen
https://github.com/mandrenguyen ?—
Reply to this email directly or view it on GitHub
https://github.com/cms-sw/cmssw/pull/1564/files#r8367102.
Vyacheslav (Slava) Krutelyov
TAMU: Physics Dept Texas A&M MS4242, College Station, TX 77843-4242
CERN: 42-R-027
AIM/Skype: siava16 googleTalk: slava77@gmail.com
(630) 291-5128 Cell (US) +41 76 275 7116 Cell (CERN)
I have a few general questions to the new code, use of BPMPD and On the VoronoiAlgorithm (and related files)
On the use of BPMPD
I suggest that you prepare some details on implementation and present in the next RECO meeting. |
Thanks a lot for the comments Slava, On Nov 29, 2013, at 6:13 PM, slava77 notifications@github.com wrote:
|
@@ -4,12 +4,8 @@ | |||
<use name="FWCore/ParameterSet"/> | |||
<use name="RecoHI/HiJetAlgos"/> | |||
<use name="DataFormats/JetReco"/> | |||
<Flags CXXFLAGS="-frounding-math"/> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this required here?
@smuzaffar @ktf
Will this work properly on all architecture that we support?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi, just realized that this point is not resolved, can @yslai please comment on the first question and @smuzaffar and @ktf comment on the second?
Thanks.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It is required due to CGAL. While we do not use rounding-sensitive parts
of CGAL (the Voronoi diagram construction is safe in that regard),
unfortunately CGAL checks during library initialization that
-frounding-math is active. So short of an invasive patching of CGAL this
is the simplest solution.
-frounding-math only affects the optimization, and is portable for all
architectures supported by GCC.
On 29.11.2013 18:35, slava77 wrote:
In RecoHI/HiJetAlgos/plugins/BuildFile.xml:
@@ -4,12 +4,8 @@
+Is this required here?
@smuzaffar https://github.com/smuzaffar @ktf https://github.com/ktf
Will this work properly on all architecture that we support?—
Reply to this email directly or view it on GitHub
https://github.com/cms-sw/cmssw/pull/1564/files#r8005983.
So, I'm told this is the last one ;) |
+1 for #1564 880390e Based mainly on the discussion above (tests done in 5_3_14), and a shorter set of tests in 53X IB. The last iteration of fixes has the same event content, the timing went up a bit The calibration files are read from the external as expected. |
+1 |
This pull request is fully signed and it will be integrated in one of the next CMSSW_5_3_X IBs unless changes or unless it breaks tests. @smuzaffar can you please take care of it? |
+1 |
@yetkinyilmaz
You have DEFINE_FWK_MODULE(AnalyzerWithCentrality); both in test and plugins subdir and I would suggest that please remove the test/AnalyzerWithCentrality.cc and test/BuildFile.xml |
Now that 1564 is merged, this should be done in a new pull-request right? On Feb 27, 2014, at 10:51 AM, Malik Shahzad Muzaffar notifications@github.com wrote:
|
yes
|
Yes. On Feb 27, 2014, at 11:49 AM, yetkinyilmaz notifications@github.com
|
Reconstruction updates for jets and tracking.
Ignore *.C root macro files in test/ directories, which are for user-level analyses.
The RelVal tests will crash because of not having the proper L1 menu in the available global tag compatible with HLT.