I have downloaded ROOTPWA recently and have installed a machine of mine.
The installation seems well-down and I'd like to understand and progress to the next step.
I found a example file as /rootpwa-code/scripts/pwa_Kpipi_example/run_Kpipi_PWA_analysis.sh.
The shell script seems to download example data automatically from the directory: http://www.staff.uni-mainz.de/jasinsk/demo-data/. However the directory does not exist.
If example data will be available, it will be very helpful.
Thank you very much for any hints or any help...
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
nice to hear of somebody interested in using rootpwa. I removed the
example data recently since analysis went on and the example was
outdated. Scripts are nice to understand programs behind. The gui is now
aimed to lead you through the process. Please try {cmakedir}/bin/***_gui
programs. I do not remember the name anymore since It's been now 2 years
I did not use it anymore. For further support, please stay in contact
with us telling us for example your application. Hopefully there will be
no traps to step in with your channel.
Cheers Promme
On Wed, 2013-06-19 at 10:51 +0000, Tatsuro Matsuda wrote:
Hello,
I have downloaded ROOTPWA recently and have installed a machine of
mine.
The installation seems well-down and I'd like to understand and
progress to the next step.
I found a example file
as /rootpwa-code/scripts/pwa_Kpipi_example/run_Kpipi_PWA_analysis.sh.
The shell script seems to download example data automatically from the
directory: http://www.staff.uni-mainz.de/jasinsk/demo-data/. However the
directory does not exist.
If example data will be available, it will be very helpful.
nice to hear of somebody interested in using rootpwa. I removed the example data recently since analysis went on and the example was outdated. Scripts are nice to understand programs behind. The gui is now aimed to lead you through the process. Please try {cmakedir}/bin/***_gui programs. I do not remember the name anymore since It's been now 2 years I did not use it anymore. For further support, please stay in contact with us telling us for example your application. Hopefully there will be no traps to step in with your channel.
Cheers Promme
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks a lot for introducing me to the nice GUI program which will be /build/bin/rootpwa_gui. When I execute this program, a new window appears and lead me to processes. However I'd still like to run with some concrete files. My object to try to use rootpwa is to analyze meson production channels of B-meson-decay. I think this code is not only applicable to the diffractive reactions but also to the B-decay channels. I had better consult directly...
Tatsuro
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
You will have to copy the example setup file in the gui folder, modify it according to your paths and then to load it with the gui. Setting a new file was not implemented yet. Sorry for that.
More to read about my analysis, if you wish to understand details:
There are now 2 representations of the code and 2 ways to run the analysis. The 2nd alternative one is not fully implemented but gives basically the same results like the 1st one, that is supported by the gui.
Mass dependent fits (last step of analysis) were done by sebastian neubert for a 3 pi and 5 pi analysis. There are corners in the code that need further modification to be applicable for other channels.
We would be happy that improvements to the code would be shared by you.
But first I would recommend to discuss your channel with us whether we do not see any problems that may occur. A diagram might be useful.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I appreciate your great help. Now I am downloading your files. It will still need
hours. I am sorry that I did not introduce myself clearly because this board is
a public one. But I am a colleague of yours at COMPASS. You can find my name at
the collaboration web page. And of course I will keep your instruction. I have
already your thesis which is very nice and beautiful. Your finding of the effect
of RICH on the pid has been wonderful. This time I will try to adapt rootpwa to
the other experiment in Japan.Then I asked my questions in public. And this
project is quit great. If I can take a step forward, I will report.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The rootpwa seems to run fortunately. Thank you very much for your kind help. But when I execute
genpw -n 10000 -M 800 -B 20 -r ~/rootpwa/generators/KpipiDiffractive2008.conf for example, it does not produce correct files and it answers " Warning in <TFile::Init>: no StreamerInfo found" in return. I am not sure the reason but I guess this might be due to the difference of the version of root. If you would remember it, could you tell me the version of root which you used ?
I am so sorry for asking frequently.
Tatsuro
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The last time I know the generator was running, was 1 your ago. But I do not really know who was working with it since then. Funny, as far as I remember the generator is not reading any files but only writing some?!
2 Possibilities: You wait until a collegue fixes the problem, or you find the bug, or you go on.
Or you go on with the step "filter data into bins". Flat phase space events will be restored then from MC acceptance data I have provided you.
But definetly genpw must work otherwise you will be not able to try it with your channel.
Cheers for now, I hope I will find some time to investigate your problem soon.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks a lot again. The genpw has worked well and generated three files when it has been executed without reading a file. It seems to try to extract some parameters (vertex position etc.) from the config file when an argument(a file name) is set following "-r". I will also look for the flat phase space events in the MC file which you provided me.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
OK, this reminds me oft input actally I have implemented. In the config file you can find a path to a root file describing primär vertex properties. For your pwa test you can get rid of it. Please habe a look to the 2004 config file for genpw to See the alternative setup oft primary vertex generator.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
I have restarted the installation of ROOTPWA into my machine. But I encounter the problem of the Building ROOTPWA. When I make "cmake .." to build ROOTPWA, the following message :
-- >>> Setting up ROOT.
CMake Error at cmakeModules/FindROOT.cmake:276 (message):
Unable to find requested ROOT installation: ROOT executable directory ''
does not exist. ROOT include directory '' does not exist. ROOT library
directory '' does not exist. ROOT version .. is lower than requested
version 5.26.
Call Stack (most recent call first):
CMakeLists.txt:224 (find_package)
returns.
The cmake seems not to find the ROOT directory. However ROOT works at my machine. The environment variable of ROOT is set at .bashrc. Is there any reason of this problem?
As you know I am short of the knowledge of UNIX machine. And I am very sorry for such a primitive question.
Thank you in advance.
Tatsuro
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I did "git log -l" (not 1 but l) and I have got a very long output.
At the moment I think I don't have to post the output because the problem may be solved. If you need the output, please notify me. I will do.
I appreciate your comprehensive mind. Danke schoen. Arigato.
Thanks,
Tatsuro
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear Boris,
Thank you very much for helping me. I just tried ccmake and I found one more window appeared and the listed directory of ROOT was wrong. It was an old directory which was I set before. I tried to install many versions of ROOT because I had failed to install and construct ROOT into my machine. I suppose something wrong parameters remained. And I have corrected these parameters by hand and I have probably succeeded in building ROOTPWA. I attach the output of "cmake".
I will go forward. I am very very sorry for disturbing you.
Best regards,
Tatsuro
Great to see that things are working for you. Unfortunately Cmake remembers some paths. So if external libraries change, the safest way is to delete everything in the build directory and build ROOTPWA again.
Cheers,
Boris.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thank you very much for explaining the reason why the old parameters remained.
Now I am running the step7.sh using the "sample data" which was given by Promme and I realize this process needs quite a long time. I started the step yesterday but that is still running. It might not finish during my life ?
Thanks,
Tatsuro
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Sorry Tatsuro, for the late answer. step7.sh submits jobs that calculate the amplitudes. Depending on the amount of data and the wave set size this can may take a while. However, depending on the batch system you are using, you might need to modify the script.
Note that the scripts and keyfiles you are using still use the old PWA2000 amplitude framework. Meanwhile we have switched all active COMPASS analyses that use ROOTPWA to a new amplitude code which is faster and easier to maintain. The old code was removed in this process. Unfortunately there is no real documentation available yet for the new amplitude framework.
Cheers,
Boris.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thank you very much for having informed more useful information to me. And I am sorry
that I did not know that you had posted your new message to this "board" because I
don't know how to let me know an appearance of a new message on this board
automatically. I suppose there is such a method on this system.
By the way I think I have succeeded in running step7.sh on my machines. On one of my
machines it took 3 days and 5 hours, and it took 2 days and 5 hours on the other
machine. Certainly I changed the script slightly. Meanwhile I may enjoy this older
code. Actually I am using the oldest version of ROOTPWA which is _ver1.0.
And may I find the new amplitude code in the latest version of ROOTPWA ?
Danke schoen,
Tatsuro
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
good to hear that the amplitude calculation seems to work for you. Usually this step is performed on a batch system, since the calculation of the amplitudes for the different waves are independent from each other. So if you have a set of 100 waves you could run 100 jobs in parallel.
You may set up email notifications by going to your profile and adding ROOTPWA to your projects. Then you can choose to subscribe to the various boards and email lists by clicking on "Edit Personal Data" and then "Subscriptions".
Using the "_v1.0" tag is perfectly fine. The only potential drawback is that we cannot provide fixes for bugs in the legacy code should they appear.
In order to have look at the new amplitude framework it is recommended that you switch to the "master" branch. This branch is actively maintained, so you should regularly pull the changes. The amplitude code is in the subdirectory "amplitude".
Cheers,
Boris.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
I have downloaded ROOTPWA recently and have installed a machine of mine.
The installation seems well-down and I'd like to understand and progress to the next step.
I found a example file as /rootpwa-code/scripts/pwa_Kpipi_example/run_Kpipi_PWA_analysis.sh.
The shell script seems to download example data automatically from the directory:
http://www.staff.uni-mainz.de/jasinsk/demo-data/. However the directory does not exist.
If example data will be available, it will be very helpful.
Thank you very much for any hints or any help...
Hello,
nice to hear of somebody interested in using rootpwa. I removed the
example data recently since analysis went on and the example was
outdated. Scripts are nice to understand programs behind. The gui is now
aimed to lead you through the process. Please try {cmakedir}/bin/***_gui
programs. I do not remember the name anymore since It's been now 2 years
I did not use it anymore. For further support, please stay in contact
with us telling us for example your application. Hopefully there will be
no traps to step in with your channel.
Cheers Promme
On Wed, 2013-06-19 at 10:51 +0000, Tatsuro Matsuda wrote:
Hello,
nice to hear of somebody interested in using rootpwa. I removed the example data recently since analysis went on and the example was outdated. Scripts are nice to understand programs behind. The gui is now aimed to lead you through the process. Please try {cmakedir}/bin/***_gui programs. I do not remember the name anymore since It's been now 2 years I did not use it anymore. For further support, please stay in contact with us telling us for example your application. Hopefully there will be no traps to step in with your channel.
Cheers Promme
Hi,
Thanks a lot for introducing me to the nice GUI program which will be /build/bin/rootpwa_gui. When I execute this program, a new window appears and lead me to processes. However I'd still like to run with some concrete files. My object to try to use rootpwa is to analyze meson production channels of B-meson-decay. I think this code is not only applicable to the diffractive reactions but also to the B-decay channels. I had better consult directly...
Tatsuro
here is a link to data for my analysis. It is not officially released and therefor you must not show it in public.
http://www.staff.uni-mainz.de/jasinsk/example_data/
You will have to copy the example setup file in the gui folder, modify it according to your paths and then to load it with the gui. Setting a new file was not implemented yet. Sorry for that.
More to read about my analysis, if you wish to understand details:
http://wwwcompass.cern.ch/compass/publications/theses/2012_phd_jasinski.pdf
2 things to mention:
There are now 2 representations of the code and 2 ways to run the analysis. The 2nd alternative one is not fully implemented but gives basically the same results like the 1st one, that is supported by the gui.
Mass dependent fits (last step of analysis) were done by sebastian neubert for a 3 pi and 5 pi analysis. There are corners in the code that need further modification to be applicable for other channels.
We would be happy that improvements to the code would be shared by you.
But first I would recommend to discuss your channel with us whether we do not see any problems that may occur. A diagram might be useful.
Hi,
I appreciate your great help. Now I am downloading your files. It will still need
hours. I am sorry that I did not introduce myself clearly because this board is
a public one. But I am a colleague of yours at COMPASS. You can find my name at
the collaboration web page. And of course I will keep your instruction. I have
already your thesis which is very nice and beautiful. Your finding of the effect
of RICH on the pid has been wonderful. This time I will try to adapt rootpwa to
the other experiment in Japan.Then I asked my questions in public. And this
project is quit great. If I can take a step forward, I will report.
The download of all files finished. Thanks a lot.
Dear Promme,
The rootpwa seems to run fortunately. Thank you very much for your kind help. But when I execute
genpw -n 10000 -M 800 -B 20 -r ~/rootpwa/generators/KpipiDiffractive2008.conf for example, it does not produce correct files and it answers " Warning in <TFile::Init>: no StreamerInfo found" in return. I am not sure the reason but I guess this might be due to the difference of the version of root. If you would remember it, could you tell me the version of root which you used ?
I am so sorry for asking frequently.
Tatsuro
Hi,
The last time I know the generator was running, was 1 your ago. But I do not really know who was working with it since then. Funny, as far as I remember the generator is not reading any files but only writing some?!
2 Possibilities: You wait until a collegue fixes the problem, or you find the bug, or you go on.
Or you go on with the step "filter data into bins". Flat phase space events will be restored then from MC acceptance data I have provided you.
But definetly genpw must work otherwise you will be not able to try it with your channel.
Cheers for now, I hope I will find some time to investigate your problem soon.
Hi,
Thanks a lot again. The genpw has worked well and generated three files when it has been executed without reading a file. It seems to try to extract some parameters (vertex position etc.) from the config file when an argument(a file name) is set following "-r". I will also look for the flat phase space events in the MC file which you provided me.
OK, this reminds me oft input actally I have implemented. In the config file you can find a path to a root file describing primär vertex properties. For your pwa test you can get rid of it. Please habe a look to the 2004 config file for genpw to See the alternative setup oft primary vertex generator.
Sorry for the man autocorrect Terrors by my Herman Android.
Hi,
I have restarted the installation of ROOTPWA into my machine. But I encounter the problem of the Building ROOTPWA. When I make "cmake .." to build ROOTPWA, the following message :
-- >>> Setting up ROOT.
CMake Error at cmakeModules/FindROOT.cmake:276 (message):
Unable to find requested ROOT installation: ROOT executable directory ''
does not exist. ROOT include directory '' does not exist. ROOT library
directory '' does not exist. ROOT version .. is lower than requested
version 5.26.
Call Stack (most recent call first):
CMakeLists.txt:224 (find_package)
returns.
The cmake seems not to find the ROOT directory. However ROOT works at my machine. The environment variable of ROOT is set at .bashrc. Is there any reason of this problem?
As you know I am short of the knowledge of UNIX machine. And I am very sorry for such a primitive question.
Thank you in advance.
Tatsuro
Dear Tatsuro,
it seems that Cmake cannot find the root-config executable or that it gets incorrect information from root-config.
Could you please post the output of the following command?
root-config --version --bindir --incdir --libdir
Thanks,
Boris.
Dear Boris,
Thank you very much for responding to my question.
Here is the output corresponding to your suggestion:
[matsuda@unkai ~]$ root-config --version --bindir --incdir --libdir
5.34/18 /usr/local/hep/root/v5.34.18/bin /usr/local/hep/root/v5.34.18/include /usr/local/hep/root/v5.34.18/lib
I think ROOT was installed successfully and seems to work well...
A lot of thanks,
Tatsuro
Hm, this looks fine. Could you please verify that the directories
/usr/local/hep/root/v5.34.18/bin
/usr/local/hep/root/v5.34.18/include
/usr/local/hep/root/v5.34.18/lib
exist? If they do, it would be great if you could post the full Cmake output as an attachment.
Cheers,
Boris.
I appreciate your kind cooperation.
I think these directories exist. For example,
[matsuda@unkai build]$ ls /usr/local/hep/root/v5.34.18/bin
g2root hadd proofserv.exe root.exe roots.exe thisroot.sh
The rest is ommited.
[matsuda@unkai build]$ ls /usr/local/hep/root/v5.34.18/include |more
Api.h
AuthConst.h
The rest is ommited.
[matsuda@unkai build]$ ls /usr/local/hep/root/v5.34.18/lib
PyCintex.py libHbook.rootmap libRootAuth.so
The rest is ommited.
And I will attach the output file of the procedure.
I did "cmake ..".
Many thanks,
Tatsuro
Hi Tatsuro,
could you please post the output of
git log -1
(executed in your ROOTPWA directory) so that I can see which ROOTPWA version you are using?
Thanks,
Boris.
Dear Boris,
I did "git log -l" (not 1 but l) and I have got a very long output.
At the moment I think I don't have to post the output because the problem may be solved. If you need the output, please notify me. I will do.
I appreciate your comprehensive mind. Danke schoen. Arigato.
Thanks,
Tatsuro
Dear Boris,
Thank you very much for helping me. I just tried ccmake and I found one more window appeared and the listed directory of ROOT was wrong. It was an old directory which was I set before. I tried to install many versions of ROOT because I had failed to install and construct ROOT into my machine. I suppose something wrong parameters remained. And I have corrected these parameters by hand and I have probably succeeded in building ROOTPWA. I attach the output of "cmake".
I will go forward. I am very very sorry for disturbing you.
Best regards,
Tatsuro
Great to see that things are working for you. Unfortunately Cmake remembers some paths. So if external libraries change, the safest way is to delete everything in the build directory and build ROOTPWA again.
Cheers,
Boris.
Thank you very much for explaining the reason why the old parameters remained.
Now I am running the step7.sh using the "sample data" which was given by Promme and I realize this process needs quite a long time. I started the step yesterday but that is still running. It might not finish during my life ?
Thanks,
Tatsuro
Sorry Tatsuro, for the late answer. step7.sh submits jobs that calculate the amplitudes. Depending on the amount of data and the wave set size this can may take a while. However, depending on the batch system you are using, you might need to modify the script.
Note that the scripts and keyfiles you are using still use the old PWA2000 amplitude framework. Meanwhile we have switched all active COMPASS analyses that use ROOTPWA to a new amplitude code which is faster and easier to maintain. The old code was removed in this process. Unfortunately there is no real documentation available yet for the new amplitude framework.
Cheers,
Boris.
Hi Boris,
Thank you very much for having informed more useful information to me. And I am sorry
that I did not know that you had posted your new message to this "board" because I
don't know how to let me know an appearance of a new message on this board
automatically. I suppose there is such a method on this system.
By the way I think I have succeeded in running step7.sh on my machines. On one of my
machines it took 3 days and 5 hours, and it took 2 days and 5 hours on the other
machine. Certainly I changed the script slightly. Meanwhile I may enjoy this older
code. Actually I am using the oldest version of ROOTPWA which is _ver1.0.
And may I find the new amplitude code in the latest version of ROOTPWA ?
Danke schoen,
Tatsuro
Hi Tatsuro,
good to hear that the amplitude calculation seems to work for you. Usually this step is performed on a batch system, since the calculation of the amplitudes for the different waves are independent from each other. So if you have a set of 100 waves you could run 100 jobs in parallel.
You may set up email notifications by going to your profile and adding ROOTPWA to your projects. Then you can choose to subscribe to the various boards and email lists by clicking on "Edit Personal Data" and then "Subscriptions".
Using the "_v1.0" tag is perfectly fine. The only potential drawback is that we cannot provide fixes for bugs in the legacy code should they appear.
In order to have look at the new amplitude framework it is recommended that you switch to the "master" branch. This branch is actively maintained, so you should regularly pull the changes. The amplitude code is in the subdirectory "amplitude".
Cheers,
Boris.