Alignment LA correction for pixels #2067

dkotlins · 2014-01-17T17:52:20Z

Include alignment LA correction in pixel RecHit, in both CPEs, generic and templates.
Split LA-width from LA-offset in CPEGeneric, they will be adjusted separately.
New DB objects still to come.

cmsbuild · 2014-01-17T17:58:04Z

A new Pull Request was created by @dkotlins for CMSSW_7_1_X.

Danek pre10

It involves the following packages:

EventFilter/SiPixelRawToDigi
RecoLocalTracker/SiPixelClusterizer
RecoLocalTracker/SiPixelRecHits
SimTracker/SiPixelDigitizer

@civanch, @nclopezo, @mdhildreth, @cmsbuild, @anton-a, @thspeer, @slava77, @Degano can you please review it and eventually sign? Thanks.
@gpetruc, @cerati, @GiacomoSguazzoni, @rovere this is something you requested to watch as well.
You can sign-off by replying to this message having '+1' in the first line of your reply.
You can reject by replying to this message having '-1' in the first line of your reply.

nclopezo · 2014-01-20T15:51:33Z

-1

When I ran the RelVals I found an error in the following workflows:

4.53 step2
401.3 step1
5.1 step1
8.0 step2
1306.0 step2
25.0 step2

----- Begin Fatal Exception 20-Jan-2014 14:55:23 CET-----------------------
An exception of category 'Configuration' occurred while
   [0] Processing run: 1
   [1] Calling beginRun for unscheduled module EventSetupRecordDataGetter/'hltGetConditions'
   [2] Using EventSetup component PixelCPETemplateRecoESProducer/'hltESPPixelCPETemplateReco' to make data PixelClusterParameterEstimator/'hltESPPixelCPETemplateReco' in record TkPixelCPERecord
Exception Message:
MissingParameter: Parameter 'DoLorentz' not found.
----- End Fatal Exception -------------------------------------------------

You can see the results of the tests here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-2067/3/summary.html

dkotlins · 2014-01-20T17:31:40Z

Hello,
Sorry for this.
I have added one more parameter to:
RecoLocalTracker/SiPixelRecHits/python/PixelCPETemplateReco_cfi.py

DoLorentz = cms.bool(False),

I forgot that all these parameters are also repeated in HLT config files in
HLTrigger/Configuration/python/*.py .

I suppose there is an automatic way to do it. I have to find out how to do it.

Best regards,
Danek

Danek Bohdan Kotlinski
danek.kotlinski@psi.ch
Phone: +41 563104030 (PSI), 0764875909 or 165909 (CERN)
Paul Scherrer Institut OFLC/004

On 20 Jan 2014, at 16:51, David Mendez notifications@github.com wrote:

-1

When I ran the RelVals I found an error in the following workflows:

4.53 step2
401.3 step1
5.1 step1
8.0 step2
1306.0 step2
25.0 step2

----- Begin Fatal Exception 20-Jan-2014 14:55:23 CET-----------------------
An exception of category 'Configuration' occurred while
[0] Processing run: 1
[1] Calling beginRun for unscheduled module EventSetupRecordDataGetter/'hltGetConditions'
[2] Using EventSetup component PixelCPETemplateRecoESProducer/'hltESPPixelCPETemplateReco' to make data PixelClusterParameterEstimator/'hltESPPixelCPETemplateReco' in record TkPixelCPERecord
Exception Message:
MissingParameter: Parameter 'DoLorentz' not found.
----- End Fatal Exception -------------------------------------------------
You can see the results of the tests here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-2067/3/summary.html

—
Reply to this email directly or view it on GitHub.

dkotlins · 2014-01-20T20:27:07Z

Hello,
I found out from Martin Grunewald that the HLT configurations will be automatically
updated.
Best regards,
Danek

-----Original Message-----
From: David Mendez [mailto:notifications@github.com]
Sent: Mon 1/20/2014 4:51 PM
To: cms-sw/cmssw
Cc: Kotlinski Bohdan
Subject: Re: [cmssw] Danek pre10 (#2067)

-1

When I ran the RelVals I found an error in the following workflows:

4.53 step2
401.3 step1
5.1 step1
8.0 step2
1306.0 step2
25.0 step2

----- Begin Fatal Exception 20-Jan-2014 14:55:23 CET-----------------------
An exception of category 'Configuration' occurred while
   [0] Processing run: 1
   [1] Calling beginRun for unscheduled module EventSetupRecordDataGetter/'hltGetConditions'
   [2] Using EventSetup component PixelCPETemplateRecoESProducer/'hltESPPixelCPETemplateReco' to make data PixelClusterParameterEstimator/'hltESPPixelCPETemplateReco' in record TkPixelCPERecord
Exception Message:
MissingParameter: Parameter 'DoLorentz' not found.
----- End Fatal Exception -------------------------------------------------

You can see the results of the tests here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-2067/3/summary.html

Reply to this email directly or view it on GitHub:
#2067 (comment)

cmsbuild · 2014-01-30T10:21:14Z

+1
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-2067/50/summary.html

mdhildreth · 2014-01-30T17:15:24Z

+1
On Jan 17, 2014, at 12:58 PM, cmsbuild notifications@github.com wrote:

A new Pull Request was created by @dkotlins for CMSSW_7_1_X.

Danek pre10

It involves the following packages:

EventFilter/SiPixelRawToDigi
RecoLocalTracker/SiPixelClusterizer
RecoLocalTracker/SiPixelRecHits
SimTracker/SiPixelDigitizer

@civanch, @nclopezo, @mdhildreth, @cmsbuild, @anton-a, @thspeer, @slava77, @Degano can you please review it and eventually sign? Thanks.
@gpetruc, @cerati, @GiacomoSguazzoni, @rovere this is something you requested to watch as well.
You can sign-off by replying to this message having '+1' in the first line of your reply.
You can reject by replying to this message having '-1' in the first line of your reply.

—
Reply to this email directly or view it on GitHub.

Mike Hildreth e-mail: mikeh@undhep.hep.nd.edu
Department of Physics mikeh@fnal.gov
225 Nieuwland Sciences Hall telephone: 574-631-6458 (office)
University of Notre Dame 574-631-5952 (FAX)
Notre Dame, IN 46556 http://www.hep.nd.edu/MikeHildreth

slava77 · 2014-02-13T14:07:58Z

RecoLocalTracker/SiPixelRecHits/interface/PixelCPEBase.h

@@ -41,6 +41,7 @@

 #include <iostream>

+const bool MYDEBUG = false;


this constant is too generic and should not be in the header file.
Please, use LogDebug in this file.

slava77 · 2014-02-13T14:19:14Z

Hi Danek,

If these changes are expected to introduce changes in physics results,
please add some notes and maybe a few plots on what kind of change is coming.

Thanks

slava77 · 2014-02-13T14:20:56Z

-1

to 34c5307

based on the code review:

couts to be removed or changed to LogDebug or LogInfo
untracked parameters to be changed to tracked.

Please follow up and update in the same PR (my -1 will get reset then).

dkotlins · 2014-02-21T17:48:04Z

Slava,
These changes are part of an bigger update. I just thought it was convenient to do it
in smaller steps.

The main purpose is to:

be able to use the LA (lorentz angle) determined by the alignment group
decouple LA value used for the hit resolution from the one used for hit offset
separate the generic error parametrisation from the template objects.
(1) and (2) is to better deal with radiation effects in the future, (3) is just for speed reasons
and was discussed with Vincenzo.

By itself this does not change any physics results. If the same value is used the results
will be the same. But it gives us more flexibility to handle radiation.
So the validation will be to show that after introducing all these changes the resolution stays the same.

Most of you comments are simple and can be done easily.
I use the untracked parameter for a parameter which
should not be used from the python script but only from the DB. The option from the script is only for testing
purposes, when we test/optimize resolution it is much easier to have in as an input parameter instead of having
to create many DB objects. This usage is protected by a bool variable which is a tracked.
For the couts, I find them very convenient for debugging but will try to delete them in the final code.

I can make the corrections sometime next week and re-commit or simply wait for more modifications to come.
The idea is that it should all be ready before we restart data taking in 2015.

If you see a fundamental reason that the changes I have outline above cannot be implemented let us know
ASAP since several people are working on it now.

Best,
Danek

Danek Bohdan Kotlinski
danek.kotlinski@psi.ch
Phone: +41 563104030 (PSI), 0764875909 or 165909 (CERN)
Paul Scherrer Institut OFLC/004

On 13 Feb 2014, at 15:19, slava77 notifications@github.com wrote:

Hi Danek,

If these changes are expected to introduce changes in physics results,
please add some notes and maybe a few plots on what kind of change is coming.

Thanks

—
Reply to this email directly or view it on GitHub.

slava77 · 2014-02-21T17:55:02Z

Hi Danek,

Thank you for the explanations and the follow up.
For the couts, I can suggest something like
VinInn@27bbeb8#diff-7b9c837596d23ae09d2cd2a322051d5aR7
With a special flag, you should be able to recompile with couts showing up as you need, else they will just be hidden by LogTrace.
If you can live with LogTrace or LogDebug (requires recompilation anyways, but gives you more info on the origin of the printout), I'd suggest to just move to LogTrace.

Dr15Jones · 2014-02-21T18:52:39Z

It is also important to remember that cout is not thread safe so your ouput could be garbled when run under the threaded framework. The MessageLogger is thread safe.

dkotlins · 2014-03-12T18:25:08Z

Salvatore, its done.
But I thought that in test directories we are more free to do whatever is more convenient
for us. The code from test is never used in any official reconstruction/simulation process.
It is just for testing.
Best regards,
Danek

diguida · 2014-03-12T18:28:23Z

@dkotlins
In the tests you can do what you want, provided that you follow the rules of the CMSSW framework, I'd say (e.g., you crete histograms with TFileService, not bare ROOT).

diguida · 2014-03-12T18:28:50Z

+1
@slava77 no changes in the RECO code.

slava77 · 2014-03-12T20:18:26Z

+1

for #2067 8f0eaf3
just compiled as a sanity check (visually code diff ok)

civanch · 2014-03-13T11:54:15Z

+1

cmsbuild · 2014-03-13T11:58:23Z

This pull request is fully signed and it will be integrated in one of the next CMSSW_7_1_X IBs unless changes or unless it breaks tests. @nclopezo, @ktf can you please take care of it?

cmsbuild · 2014-03-13T17:40:08Z

+1
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-2067/473/summary.html

cmsbuild · 2014-03-13T17:43:24Z

This pull request is fully signed and it will be integrated in one of the next CMSSW_7_1_X IBs unless changes (tests are also fine). @nclopezo, @ktf can you please take care of it?

Reco -- Alignment LA correction for pixels

dkotlins added 10 commits January 8, 2014 09:58

small

da304a8

small

494803d

small

ffc365c

small

6c7deef

small

57e7601

test

45c76e1

modify generic CPE, sign, add configurable paramtes, clean

4bfa20c

modify generic CPE, sign, add configurable paramtes, clean

fb30580

add alignment LA to template reco

d65afc3

add alignment LA to template reco

34c5307

smaa changes in test code

35ca660

nclopezo modified the milestones: CMSSW_7_1_0_pre3, CMSSW_7_1_0_pre2 Feb 5, 2014

slava77 reviewed Feb 13, 2014
View reviewed changes

dkotlins added alca-approved labels Mar 12, 2014

dkotlins added alca-approved labels Mar 13, 2014

ktf added a commit that referenced this pull request Mar 13, 2014

Merge pull request #2067 from dkotlins/danek_pre10

4d4633d

Reco -- Alignment LA correction for pixels

ktf merged commit 4d4633d into cms-sw:CMSSW_7_1_X Mar 13, 2014

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Alignment LA correction for pixels #2067

Alignment LA correction for pixels #2067

dkotlins commented Jan 17, 2014

cmsbuild commented Jan 17, 2014

nclopezo commented Jan 20, 2014

dkotlins commented Jan 20, 2014

dkotlins commented Jan 20, 2014

cmsbuild commented Jan 30, 2014

mdhildreth commented Jan 30, 2014

slava77 Feb 13, 2014

slava77 commented Feb 13, 2014

slava77 commented Feb 13, 2014

dkotlins commented Feb 21, 2014

slava77 commented Feb 21, 2014

Dr15Jones commented Feb 21, 2014

dkotlins commented Mar 12, 2014

diguida commented Mar 12, 2014

diguida commented Mar 12, 2014

slava77 commented Mar 12, 2014

civanch commented Mar 13, 2014

cmsbuild commented Mar 13, 2014

cmsbuild commented Mar 13, 2014

cmsbuild commented Mar 13, 2014

		@@ -41,6 +41,7 @@

		#include <iostream>

		const bool MYDEBUG = false;

Alignment LA correction for pixels #2067

Alignment LA correction for pixels #2067

Conversation

dkotlins commented Jan 17, 2014

cmsbuild commented Jan 17, 2014

nclopezo commented Jan 20, 2014

dkotlins commented Jan 20, 2014

dkotlins commented Jan 20, 2014

cmsbuild commented Jan 30, 2014

mdhildreth commented Jan 30, 2014

slava77 Feb 13, 2014

Choose a reason for hiding this comment

slava77 commented Feb 13, 2014

slava77 commented Feb 13, 2014

dkotlins commented Feb 21, 2014

slava77 commented Feb 21, 2014

Dr15Jones commented Feb 21, 2014

dkotlins commented Mar 12, 2014

diguida commented Mar 12, 2014

diguida commented Mar 12, 2014

slava77 commented Mar 12, 2014

civanch commented Mar 13, 2014

cmsbuild commented Mar 13, 2014

cmsbuild commented Mar 13, 2014

cmsbuild commented Mar 13, 2014