========================================================================== CASA Test Notes 2006/03/01 (JGM) ========================================================================== SUMMARY: If the test focus was to determine whether the necessary tools are available to allow the analysis of single-baseline data ATF data, then I would say that the test was inconclusive. The main problem is that we were given a perfect data set, which is not realistic enough for this test. Also, I found myself asking "why do I need to do this" multiple times as I ran through the examples. Some explanation as to why each step was necessary and why I was doing it would have helped. ========================================================================== Getting the software: - Already installed on rubyps in CV. - Other needed files difficult to get as they are webpage-located. Required two steps to transfer to /users/jmangum. - CASA startup warning message: rubyps<62>% casapy WARNING: Old rc filename ".matplotlibrc" found in working dir and and renamed to new default rc file name "matplotlibrc" (no leading"dot"). ___________________________________________________________ Available tools: ms (MS) mp (MS plot) cb (calibrater) cp (cal plot) tb (table) tp (table plot) im (imager) af (autoflag) me (measures) pl (pylab functions) ___________________________________________________________ Available tasks: data import : vlafiller, fitstoms imaging : clean, feather, invert, mosaic, (wproj) ___________________________________________________________ type 'pdoc tool' or 'pdoc tool.function' for help type 'quickhelp' for the available tool list ___________________________________________________________ ********************************************************************** Welcome to IPython. I will try to create a personal configuration directory where you can customize many aspects of IPython's functionality in: /users/jmangum/.ipython Successful installation! Please read the sections 'Initial Configuration' and 'Quick Tips' in the IPython manual (there are both HTML and PDF versions supplied with the distribution) to make sure that your system environment is properly configured to take advantage of IPython's features. Important note: the configuration system has changed! The old system is still in place, but its setting may be partly overridden by the settings in "~/.ipython/ipy_user_conf.py" config file. Please take a look at the file if some of the new settings bother you. Please press to start IPython. In [1]: *** (1) Why did casa copy my installed .matplotlibrc to matplotlibrc when the installation instructions told me to name it .matplotlibrc? (2) CASA prompt does not correspond to that listed in the Cookbook ("CASA [1]"). ========================================================================== - Command "help" only lists how to use help system. Must type "help()" to enter help system. Same comment for "quickhelp". ========================================================================== - Default logging is no logging, which differs from Cookbook description: In [12]: logstate Logging has not been activated. *** - Cannot change log state after having started logging, even after turning logging off: In [26]: logoff Switching logging OFF In [27]: logstart -o -t ipython.log rotate WARNING: Couldn't start log: Log file is already active: ipython_log.py In [28]: logstart -o -t rotate WARNING: Couldn't start log: Log file is already active: ipython_log.py ========================================================================== - The "casalog" utility does not seem to exist... In [29]: casalog --------------------------------------------------------------------------- exceptions.NameError Traceback (most recent call last) /users/jmangum/ NameError: name 'casalog' is not defined In [30]: casalog? Object `casalog` not found. This was probably my mistake because apparently casalog starts automatically on startup. - Might be useful to be able to start casa with the last-used log state (session save feature)? ========================================================================== - The "autoparen" feature didn't always seem to work... In [1]: quickhelp Out[1]: In [2]: quickhelp() Available tools: ms : MeasurementSet (MS) utilties mp : MS plotting (data (amp/phase) versus other quantities) cb : Calibration utilities cp : Cal solution plotting utilities im : Imaging utilities af : Statistical flagging utilities tb : Table utilities (selection, extraction, etc) tp : Table plotting utilities me : Measures utilities --- pl : pylab functions (e.g., pl.title, etc) --- Available tasks: Data Import ----------- vlafiller : fill VLA data into a MeasurementSet fitstoms : fill UVFITS format into a MeasurementSet Imaging ------- clean : imaging using various clean algorithms invert : dirty beam/map construction mosaic : mosaic imaging feather : Combine images using feathering technique --- In [3]: ========================================================================== - Could not cd after starting casapy... In [4]: !pwd /export/data_1/casa/jmangum In [5]: !cd /users/jmangum In [6]: !pwd /export/data_1/casa/jmangum This means that if you start casa in a certain directory then you find that you need to cd to the directory where the data is, you can't. ========================================================================== -- Estimate of "1 to 3 days" for the time it should take to complete this test is underestimated for a realistic non-sequential test. Took me 4-5 days. ========================================================================== -- Cookbook is a good start. Good basic structure and layout. ========================================================================== -- Test data sets poorly described. Took me quite a while to realize that "G", "F", and "B" sources were "gain", "flux", and "bandpass" calibrators. Also, I only realized at the end that the example plots were posted to the testing web page for comparison. ========================================================================== -- The 3C84 annotated example (section 11.2 in the Cookbook) contains an "os.system" command that is not explained. Why do I need to execute this command to visualize my data? ========================================================================== -- Why does the "mp.plotoptions" command clear the plot display? ========================================================================== -- pdoc should list the defaults for all command options. ========================================================================== -- mp.markflags and mp.flagdata need a lot of work: (1) Appears to be no way to cancel a drawn flag box. (2) Flag region boxes sometimes persist on plot (a good thing), sometimes don't. How is this controlled? (3) Flagging misses some data in boxed regions (see example). (4) It appears that mp.flagdata rescales the plot. This makes it hard to see if data was really flagged. (5) Why werent' we given a data set that had some bad data to be flagged? (6) It is not clear what mp.setspectral is doing. Is "width" the width in input channels of each "nchan" output channel specified? (7) How do you undo flags (i.e. unflag data)? ========================================================================== -- Calibrater: (1) Log messages are difficult to understand and distinguish. (2) How do I know how good the calibrator solution was? (3) When cb.setsolve warns about an existing calibration table, did it really overwrite the file or not? The following error message is unclear. In [23]: cb.setsolve(type='M',t=3,table='B0319.M') Wed Mar 15 16:33:17 2006 WARN calibrater::setsolve(): Calibration table B0319.M exists, and append=false.Therefore, this table will be overwritten.Consider re-running setsolve with append=T. Out[23]: True (4) Are the cb log messages supposed to be informative? Here is what cb.state() says: Wed Mar 15 16:37:09 2006 NORMAL calibrater::state(): The following calibration components will be applied: Wed Mar 15 16:37:09 2006 NORMAL calibrater::state() " None." Wed Mar 15 16:37:09 2006 NORMAL calibrater::state(): The following calibration components will be solved for: Wed Mar 15 16:37:09 2006 NORMAL calibrater::state(): M table=B0319.M t=3 preavg=60 phaseonly=F refant=0 append=F What is this supposed to be telling me? (5) Similarly for the cb.solve messages. What, if anything, does the following mean? And how can I see how the one solution failed? Wed Mar 15 16:39:24 2006 NORMAL calibrater::solve() "Solving:" Wed Mar 15 16:39:24 2006 NORMAL calibrater::state(): The following calibration components will be applied: Wed Mar 15 16:39:24 2006 NORMAL calibrater::state() " None." Wed Mar 15 16:39:24 2006 NORMAL calibrater::state(): The following calibration components will be solved for: Wed Mar 15 16:39:24 2006 NORMAL calibrater::state(): M table=B0319.M t=3 preavg=60 phaseonly=F refant=0 append=F Wed Mar 15 16:39:24 2006 NORMAL Calibrater::ok: Arranging to solve for closure errors (M-matrix) Wed Mar 15 16:39:24 2006 NORMAL Calibrater::ok "Solving for M" Wed Mar 15 16:39:24 2006 NORMAL SolvableVisMueller::solve: For interval of 3 seconds, found 62 slots Wed Mar 15 16:39:24 2006 NORMAL SolvableVisMueller::solve: Found 61 good M Mueller solutions. 1 solution intervals failed, or had insufficient data. Wed Mar 15 16:39:24 2006 NORMAL SolvableVisMueller::solve: Storing M Mueller solutions in table B0319.M (6) Why did I have to solve for the 'M' solution twice (following example 11.2)? (7) Why do I need to re-correct the data after comparing M,MF and K? (8) What *should* I be seeing as I work through the examples? (9) Too much typing to do trivial things (like plot visibilities). ========================================================================== ==========================================================================