o-smirnov
commented
2 years ago @kamva1, once again, full sequence of commands please, from the moment you log in.
Closed kamva1 closed 2 years ago
o-smirnov
commented
2 years ago @kamva1, once again, full sequence of commands please, from the moment you log in.
kamva1
commented
2 years ago On server ike : I go inside my working directory (e.g cd vanqa/Vanqa/Vanqa) -- here I have all the files and scripts I am using to run my main script (sim_time.py). Then I run on the working directory( python sim_time.py) and when running this, I get the above error.
kamva1
commented
2 years ago @o-smirnov Please find this on ike: /home/vanqa/Vanqa/Vanqa/sim_time.py
with all other scripts imported into this one.
bennahugo
commented
2 years ago I've set up a reproducer here based on the input. I had to port the scripts to python3 for this with minor modifications. It looks like the imager module of pyxis no longer works as intended -- don't think this is part of the test as far as I know: The scripts are here: (venv) hugo@jake:~/mtbuild/test-vanqa$
You should activate ~/mtbuild/venv for the python3 build of MT
You can then run his sim_time.py script.
which yields the same error as the 2.7 system packages above:
('### Job result:', None)
### Stopping the meqserver
### All your batch are belong to us. Bye!
11.49user 6.55system 0:16.61elapsed 108%CPU (0avgtext+0avgdata 178664maxresident)k
0inputs+9488outputs (0major+105126minor)pagefaults 0swaps
2022/06/08 15:06:57 PYXIS: meqtree-pipeliner.py succeeded
Visibility column shape: (151200, 1, 4)
Successful readonly open of default-locked table /home/hugo/mtbuild/test-vanqa/MeerKAT_hires.MS/SPECTRAL_WINDOW: 14 columns, 1 rows
Saved plot to Pyplots0.png
2022/06/08 15:06:57 PYXIS: registered module 'im.lwimager'
Traceback (most recent call last):
File "sim_time.py", line 152, in <module>
sim_now()
File "sim_time.py", line 138, in sim_now
weight = "natural")#,basename_image='sim_img_%d.fits'%k, dirty_image = 'sim_img_%d.fits'%k);
File "/home/hugo/mtbuild/venv/lib/python3.6/site-packages/Pyxides/imager.py", line 92, in make_image
call_imager = eval( 'im.%s.make_image'%(imager.lower()) )
File "<string>", line 1, in <module>
NameError: name 'im' is not defined
I suggest porting this issue over to pyxis. It is not a meqtrees problem per say
o-smirnov
commented
2 years ago @kamva1 could you please try this with the issue-108 branch of pyxis.
bennahugo
commented
2 years ago that does not work. Imports are missing. I'm testing a fix now, so far so good I just need to make sure it passes all other tests
bennahugo
commented
2 years ago Ok it runs through under the python3 and all tests pass locally. No guarantees for py2.7 -- I don't think the package will still work with python2.7, there has been lots of changes since then including to the hashbangs at the top which are needed to run in environments without python2.
...
Successful readonly open of default-locked table /home/hugo/mtbuild/test-vanqa/MeerKAT_hires.MS/SPECTRAL_WINDOW: 14 columns, 1 rows
Saved plot to Pyplots5.png
2022/06/08 15:57:05 INFO: im.lwimager.make_image: making dirty image ./plots-spw0/spw0-s1--lwimager.dirty.fits
2022/06/08 15:57:05 PYXIS: executing '/usr/bin/time lwimager stokes=IQUV mode=channel fixed=0 gain=0.1 npix=512 weight=natural cellsize=10.arcsec cachesize=4096 ms=MeerKAT_lores_100s.MS threshold=0 niter=0 wprojplanes=0 robust=0 chanstep=1 img_chanstep=0 nchan=0 phasecenter=j2000,0h0m,-40d0m chanstart=0 data=CORRECTED_DATA img_chanstart=0 img_nchan=1 image=./plots-spw0/spw0-s1--lwimager.dirty.fits.img operation=image spwid=0 field=0':
lwimager: Version 1.4.0
2022-06-08 13:57:06 INFO Imager::open() Opening MeasurementSet /home/hugo/mtbuild/test-vanqa/MeerKAT_lores_100s.MS
2022-06-08 13:57:06 INFO imager::data selection Selecting data
2022-06-08 13:57:06 INFO imager::data selection Selected 1404 out of 1512 visibilities.
2022-06-08 13:57:06 INFO Imager::selectDataChannel() Selecting channel 0 for spw 0
2022-06-08 13:57:06 INFO imager::defineimage() Defining image properties:nx=512 ny=512 cellx='10arcsec' celly='10arcsec' stokes=IQUV' mode=channel nchan=1 start=0 step=0 spwids=[0] fieldid=-1 facets=1 frame=1 distance='0m'
2022-06-08 13:57:06 INFO imager::defineimage() phaseCenter='00:00:00.00, 320.00.00.00, ' mStart='Radialvelocity: 0' qStep='1 km/s'' mFreqStart='Frequency: 0
2022-06-08 13:57:06 INFO imager::weight() Weighting MS: Imaging weights will be changed
2022-06-08 13:57:06 INFO imager::weight() Natural weighting
2022-06-08 13:57:06 INFO imager::setoptions() Setting processing options
2022-06-08 13:57:06 INFO imager::makeimage() Calculating image (without full skyequation)
2022-06-08 13:57:06 INFO imager::makeimage() Making dirty image from corrected data
2022-06-08 13:57:06 INFO imager::makeimage() Image is : ./plots-spw0/spw0-s1--lwimager.dirty.fits.img
2022-06-08 13:57:06 INFO imager::createFTMachine() Performing interferometric gridding...
lwimager normally ended
0.45user 0.25system 0:00.44elapsed 157%CPU (0avgtext+0avgdata 86352maxresident)k
0inputs+24992outputs (0major+12402minor)pagefaults 0swaps
2022/06/08 15:57:06 PYXIS: lwimager succeeded
2022/06/08 15:57:06 ERROR: FITS::./plots-spw0/spw0-s1--lwimager.dirty.fits.img
57.293335
so this happens when I am running the meqtrees simulation script. (e.g python simtime.py). This is on both ike and jake (python 2.7. see the output below:
@o-smirnov