No output from DWA
Dear all, I have just installed ALPS via conda and faced an unexpected problem with the tutorial dwa-01-bosons. While running the program tutorial1a.py on my laptop after thermalization is finished I get a segmentation fault. I changed the code a bit (see below) to run this tutorial on the bigger machine on which there is ALPS also installed via conda. The segmentation fault does not occur, however the program does not produce any output files and the produced graph is empty. I have also tried to run DWA with these parameters without python - using input file, but this have also ended with a segmentation fault. import pyalps bin_dir = '/net/people/plgprosniak/miniconda2/bin/' parms = [] for t in [0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.1]: parms.append( { 'LATTICE' : "square lattice", 'MODEL' : "boson Hubbard", 'T' : 0.1, 'L' : 4 , 't' : t , 'mu' : 0.5, 'U' : 1.0 , 'Nmax' : 2 , 'THERMALIZATION' : 100000, 'SWEEPS' : 2000000, 'SKIP' : 500, 'MEASURE[Winding Number]': 1 } ) input_file = pyalps.writeInputFiles('parm1a', parms) res = pyalps.runApplication(bin_dir+'dwa', input_file, Tmin=5, writexml=True) # Setting graphical output import matplotlib as mpl mpl.use('agg') # Evaluating the simulation and preparing plots using Python import pyalps import matplotlib.pyplot as plt import pyalps.plot as aplt results = pyalps.getResultFiles(prefix='parm1a') print(results) data = pyalps.loadMeasurements(results,'Stiffness') rhos = pyalps.collectXY(data,x='t',y='Stiffness') fig=plt.figure() aplt.plot(rhos) plt.xlabel('Hopping $t/U$') plt.ylabel('Superfluid density $\\rho _s$') fig.savefig('/net/scratch/people/plgprosniak/test/test_dwa.png') Best regards, Oskar A. Prośniak
Hey Oskar, Greetings, Thank you for reporting the bug. With ALPS/2.3.0-py2 on CentOS/7.4.1708 via Conda/4.3.30, I reproduced the bug via GDB. Program received signal SIGSEGV, Segmentation fault. 0x0000000000452d9e in directed_worm_algorithm::perform_diagonal_measurements() () Missing separate debuginfos, use: debuginfo-install glibc-2.17-196.el7_4.2.x86_64 (gdb) bt #0 0x0000000000452d9e in directed_worm_algorithm::perform_diagonal_measurements() () #1 0x00000000004543a8 in directed_worm_algorithm::docheckpoint() () #2 0x0000000000455fc4 in directed_worm_algorithm::dostep() () #3 0x00007ffff7553dbc in alps::scheduler::MCRun::run() () from /usr/local/lib/anaconda2/anaconda2/bin/../lib/libalps.so.2 #4 0x00007ffff7562fa9 in alps::scheduler::SerialScheduler::run() () from /usr/local/lib/anaconda2/anaconda2/bin/../lib/libalps.so.2 #5 0x00007ffff7561e28 in alps::scheduler::start(int, char**, alps::scheduler::Factory const&) () from /usr/local/lib/anaconda2/anaconda2/bin/../lib/libalps.so.2 #6 0x000000000044c885 in main () Furthermore, via blame, I confirm that there are no recent changes to the DWA code. If you need to use DWA immediately, you need to download an earlier ALPS binary package (I am pretty sure that DWA works perfectly in the 2.2 release branch.) Meanwhile, I need to investigate into what breaks the code internally. Kind regards, Tamama
On 23 Feb 2018, at 1:00 AM, Oskar Amadeusz Prośniak <oskar.krak@gmail.com> wrote:
Dear all, I have just installed ALPS via conda and faced an unexpected problem with the tutorial dwa-01-bosons. While running the program tutorial1a.py on my laptop after thermalization is finished I get a segmentation fault. I changed the code a bit (see below) to run this tutorial on the bigger machine on which there is ALPS also installed via conda. The segmentation fault does not occur, however the program does not produce any output files and the produced graph is empty. I have also tried to run DWA with these parameters without python - using input file, but this have also ended with a segmentation fault.
import pyalps
bin_dir = '/net/people/plgprosniak/miniconda2/bin/'
parms = [] for t in [0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.1]: parms.append( { 'LATTICE' : "square lattice", 'MODEL' : "boson Hubbard", 'T' : 0.1, 'L' : 4 , 't' : t , 'mu' : 0.5, 'U' : 1.0 , 'Nmax' : 2 , 'THERMALIZATION' : 100000, 'SWEEPS' : 2000000, 'SKIP' : 500, 'MEASURE[Winding Number]': 1 } )
input_file = pyalps.writeInputFiles('parm1a', parms) res = pyalps.runApplication(bin_dir+'dwa', input_file, Tmin=5, writexml=True)
# Setting graphical output import matplotlib as mpl mpl.use('agg')
# Evaluating the simulation and preparing plots using Python import pyalps import matplotlib.pyplot as plt import pyalps.plot as aplt
results = pyalps.getResultFiles(prefix='parm1a') print(results)
data = pyalps.loadMeasurements(results,'Stiffness') rhos = pyalps.collectXY(data,x='t',y='Stiffness')
fig=plt.figure() aplt.plot(rhos) plt.xlabel('Hopping $t/U$') plt.ylabel('Superfluid density $\\rho _s$') fig.savefig('/net/scratch/people/plgprosniak/test/test_dwa.png')
Best regards, Oskar A. Prośniak
---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/
List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users
Unsubscribe by writing a mail to comp-phys-alps-users-leave@lists.phys.ethz.ch.
Hi, If it helps I am having this error with revision 7699, but with 7698 the error does not occur ( I did not check systhematically all revisions >=7700 but I did not anyone that worked). r7697 fails at cmake "add_subdirectory given source "imps" which is not an existing directory.” it seems that several revisions below 7696 are working fine. All in all r7699 seems to be place where the error has been introduced. Best, Mateusz Łącki
On 23 Feb 2018, at 02:51, Tama MA - Yotcopi <tamama@yotcopi.com> wrote:
Hey Oskar, Greetings,
Thank you for reporting the bug.
With ALPS/2.3.0-py2 on CentOS/7.4.1708 via Conda/4.3.30, I reproduced the bug via GDB.
Program received signal SIGSEGV, Segmentation fault. 0x0000000000452d9e in directed_worm_algorithm::perform_diagonal_measurements() () Missing separate debuginfos, use: debuginfo-install glibc-2.17-196.el7_4.2.x86_64 (gdb) bt #0 0x0000000000452d9e in directed_worm_algorithm::perform_diagonal_measurements() () #1 0x00000000004543a8 in directed_worm_algorithm::docheckpoint() () #2 0x0000000000455fc4 in directed_worm_algorithm::dostep() () #3 0x00007ffff7553dbc in alps::scheduler::MCRun::run() () from /usr/local/lib/anaconda2/anaconda2/bin/../lib/libalps.so.2 #4 0x00007ffff7562fa9 in alps::scheduler::SerialScheduler::run() () from /usr/local/lib/anaconda2/anaconda2/bin/../lib/libalps.so.2 #5 0x00007ffff7561e28 in alps::scheduler::start(int, char**, alps::scheduler::Factory const&) () from /usr/local/lib/anaconda2/anaconda2/bin/../lib/libalps.so.2 #6 0x000000000044c885 in main ()
Furthermore, via blame, I confirm that there are no recent changes to the DWA code.
If you need to use DWA immediately, you need to download an earlier ALPS binary package (I am pretty sure that DWA works perfectly in the 2.2 release branch.)
Meanwhile, I need to investigate into what breaks the code internally.
Kind regards, Tamama
On 23 Feb 2018, at 1:00 AM, Oskar Amadeusz Prośniak <oskar.krak@gmail.com <mailto:oskar.krak@gmail.com>> wrote:
Dear all, I have just installed ALPS via conda and faced an unexpected problem with the tutorial dwa-01-bosons. While running the program tutorial1a.py on my laptop after thermalization is finished I get a segmentation fault. I changed the code a bit (see below) to run this tutorial on the bigger machine on which there is ALPS also installed via conda. The segmentation fault does not occur, however the program does not produce any output files and the produced graph is empty. I have also tried to run DWA with these parameters without python - using input file, but this have also ended with a segmentation fault.
import pyalps
bin_dir = '/net/people/plgprosniak/miniconda2/bin/'
parms = [] for t in [0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.1]: parms.append( { 'LATTICE' : "square lattice", 'MODEL' : "boson Hubbard", 'T' : 0.1, 'L' : 4 , 't' : t , 'mu' : 0.5, 'U' : 1.0 , 'Nmax' : 2 , 'THERMALIZATION' : 100000, 'SWEEPS' : 2000000, 'SKIP' : 500, 'MEASURE[Winding Number]': 1 } )
input_file = pyalps.writeInputFiles('parm1a', parms) res = pyalps.runApplication(bin_dir+'dwa', input_file, Tmin=5, writexml=True)
# Setting graphical output import matplotlib as mpl mpl.use('agg')
# Evaluating the simulation and preparing plots using Python import pyalps import matplotlib.pyplot as plt import pyalps.plot as aplt
results = pyalps.getResultFiles(prefix='parm1a') print(results)
data = pyalps.loadMeasurements(results,'Stiffness') rhos = pyalps.collectXY(data,x='t',y='Stiffness')
fig=plt.figure() aplt.plot(rhos) plt.xlabel('Hopping $t/U$') plt.ylabel('Superfluid density $\\rho _s$') fig.savefig('/net/scratch/people/plgprosniak/test/test_dwa.png')
Best regards, Oskar A. Prośniak
---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/ <http://alps.comp-phys.org/>
List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users
Unsubscribe by writing a mail to comp-phys-alps-users-leave@lists.phys.ethz.ch.
---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/
List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users
Unsubscribe by writing a mail to comp-phys-alps-users-leave@lists.phys.ethz.ch.
Hey, I am now working with revision 7650 where DWA works fine. However I've found that there is no output from Green's function. Is there any reason why it doesn't work and should it work in more recent versions? Also, is there maybe even earlier version at which computing Green's function works? Best regards, Oskar W dniu 24.02.2018 o 16:03, Mateusz Łącki pisze:
Hi, If it helps I am having this error with revision 7699, but with 7698 the error does not occur ( I did not check systhematically all revisions >=7700 but I did not anyone that worked). r7697 fails at cmake "add_subdirectory given source "imps" which is not an existing directory.” it seems that several revisions below 7696 are working fine.
All in all r7699 seems to be place where the error has been introduced.
Best, Mateusz Łącki
On 23 Feb 2018, at 02:51, Tama MA - Yotcopi <tamama@yotcopi.com <mailto:tamama@yotcopi.com>> wrote:
Hey Oskar, Greetings,
Thank you for reporting the bug.
With ALPS/2.3.0-py2 on CentOS/7.4.1708 via Conda/4.3.30, I reproduced the bug via GDB.
Program received signal SIGSEGV, Segmentation fault. 0x0000000000452d9e in directed_worm_algorithm::perform_diagonal_measurements() () Missing separate debuginfos, use: debuginfo-install glibc-2.17-196.el7_4.2.x86_64 (gdb) bt #0 0x0000000000452d9e in directed_worm_algorithm::perform_diagonal_measurements() () #1 0x00000000004543a8 in directed_worm_algorithm::docheckpoint() () #2 0x0000000000455fc4 in directed_worm_algorithm::dostep() () #3 0x00007ffff7553dbc in alps::scheduler::MCRun::run() () from /usr/local/lib/anaconda2/anaconda2/bin/../lib/libalps.so.2 #4 0x00007ffff7562fa9 in alps::scheduler::SerialScheduler::run() () from /usr/local/lib/anaconda2/anaconda2/bin/../lib/libalps.so.2 #5 0x00007ffff7561e28 in alps::scheduler::start(int, char**, alps::scheduler::Factory const&) () from /usr/local/lib/anaconda2/anaconda2/bin/../lib/libalps.so.2 #6 0x000000000044c885 in main ()
Furthermore, via blame, I confirm that there are no recent changes to the DWA code.
If you need to use DWA immediately, you need to download an earlier ALPS binary package (I am pretty sure that DWA works perfectly in the 2.2 release branch.)
Meanwhile, I need to investigate into what breaks the code internally.
Kind regards, Tamama
On 23 Feb 2018, at 1:00 AM, Oskar Amadeusz Prośniak <oskar.krak@gmail.com <mailto:oskar.krak@gmail.com>> wrote:
Dear all, I have just installed ALPS via conda and faced an unexpected problem with the tutorial dwa-01-bosons. While running the program tutorial1a.py on my laptop after thermalization is finished I get a segmentation fault. I changed the code a bit (see below) to run this tutorial on the bigger machine on which there is ALPS also installed via conda. The segmentation fault does not occur, however the program does not produce any output files and the produced graph is empty. I have also tried to run DWA with these parameters without python - using input file, but this have also ended with a segmentation fault.
import pyalps
bin_dir = '/net/people/plgprosniak/miniconda2/bin/'
parms = [] for t in [0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.1]: parms.append( { 'LATTICE' : "square lattice", 'MODEL' : "boson Hubbard", 'T' : 0.1, 'L' : 4 , 't' : t , 'mu' : 0.5, 'U' : 1.0 , 'Nmax' : 2 , 'THERMALIZATION' : 100000, 'SWEEPS' : 2000000, 'SKIP' : 500, 'MEASURE[Winding Number]': 1 } )
input_file = pyalps.writeInputFiles('parm1a', parms) res = pyalps.runApplication(bin_dir+'dwa', input_file, Tmin=5, writexml=True)
# Setting graphical output import matplotlib as mpl mpl.use('agg')
# Evaluating the simulation and preparing plots using Python import pyalps import matplotlib.pyplot as plt import pyalps.plot as aplt
results = pyalps.getResultFiles(prefix='parm1a') print(results)
data = pyalps.loadMeasurements(results,'Stiffness') rhos = pyalps.collectXY(data,x='t',y='Stiffness')
fig=plt.figure() aplt.plot(rhos) plt.xlabel('Hopping $t/U$') plt.ylabel('Superfluid density $\\rho _s$') fig.savefig('/net/scratch/people/plgprosniak/test/test_dwa.png')
Best regards, Oskar A. Prośniak
---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/
List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users
Unsubscribe by writing a mail to comp-phys-alps-users-leave@lists.phys.ethz.ch <mailto:comp-phys-alps-users-leave@lists.phys.ethz.ch>.
---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/
List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users
Unsubscribe by writing a mail to comp-phys-alps-users-leave@lists.phys.ethz.ch.
---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/
List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users
Unsubscribe by writing a mail to comp-phys-alps-users-leave@lists.phys.ethz.ch.
participants (3)
-
Mateusz Łącki
-
Oskar Amadeusz Prośniak
-
Tama MA - Yotcopi