Pols should be XX, XY, YX, YY

[tremou]

Does it work only with linear feeds?
I tried it on VLA -C band data and I get the following error.

Traceback (most recent call last):
  File "/lustre/aoc/sciops/etremou/soft/DynSpecMS/splitcatalog_fixes/DynSpecMS/scripts/ms2dynspec.py", line 237, in <module>
    main(args)
  File "/lustre/aoc/sciops/etremou/soft/DynSpecMS/splitcatalog_fixes/DynSpecMS/scripts/ms2dynspec.py", line 150, in main
    D = ClassDynSpecMS(ListMSName=MSList, 
  File "/users/etremou/.local/lib/python3.9/site-packages/DynSpecMS-0.1.0-py3.9.egg/DynSpecMS/ClassDynSpecMS.py", line 148, in __init__
    self.ReadMSInfos()
  File "/users/etremou/.local/lib/python3.9/site-packages/DynSpecMS-0.1.0-py3.9.egg/DynSpecMS/ClassDynSpecMS.py", line 600, in ReadMSInfos
    raise ValueError("Pols should be XX, XY, YX, YY")
ValueError: Pols should be XX, XY, YX, YY

Thanks,
Lilia

[o-smirnov]

@talon Myburgh ***@***.***> has a meerkat version. Cheers, Oleg

…

-- Sent from my phone, excuse the brevity.

On Tue, 16 Jul 2024, 01:14 Evangelia (Lilia) Tremou, < ***@***.***> wrote: Does it work only with linear feeds? I tried it on VLA -C band data and I get the following error. Traceback (most recent call last): File "/lustre/aoc/sciops/etremou/soft/DynSpecMS/splitcatalog_fixes/DynSpecMS/scripts/ms2dynspec.py", line 237, in <module> main(args) File "/lustre/aoc/sciops/etremou/soft/DynSpecMS/splitcatalog_fixes/DynSpecMS/scripts/ms2dynspec.py", line 150, in main D = ClassDynSpecMS(ListMSName=MSList, File "/users/etremou/.local/lib/python3.9/site-packages/DynSpecMS-0.1.0-py3.9.egg/DynSpecMS/ClassDynSpecMS.py", line 148, in __init__ self.ReadMSInfos() File "/users/etremou/.local/lib/python3.9/site-packages/DynSpecMS-0.1.0-py3.9.egg/DynSpecMS/ClassDynSpecMS.py", line 600, in ReadMSInfos raise ValueError("Pols should be XX, XY, YX, YY") ValueError: Pols should be XX, XY, YX, YY Thanks, Lilia — Reply to this email directly, view it on GitHub <#25>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABRLTPZ3VBJWUJYAGWQUS4TZMRJXBAVCNFSM6AAAAABK5OH5MGVHI2DSMVQWIX3LMV43ASLTON2WKOZSGQYDSNZZG4ZDKNI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[talonmyburgh]

@tremou have a try with my branch SplitCatalog_fixes. I use that for MeerKAT datasets. Can also just pip install it normally (i.e. pip install <path-to-cloned-folder>).

[tremou]

thanks. I do use the SplitCatalog_fixes branch, and I run it without any issue in MeerKAT data since it's XX, YY , etc. the issue is present is with the RR, LL, etc at VLA data.

[talonmyburgh]

Ah I see... apologies, I did not read that correctly. Yes, you are correct it appears from the code that measurement set column "CORR_TYPE" must be [9,10,11,12] which correspond to "XX, XY, YX, YY".
@cyriltasse might be able to comment further on what would be required to support other "CORR_TYPE".

[cyriltasse]

I have this raise condition because the 4 lines here

DynSpecMS/DynSpecMS/ClassDynSpecMS.py

Line 1042 in e3d828f

GOut[..., 0] = 0.5*(Gn[..., 0] + Gn[..., 3]) # I = 0.5(XX + YY)

, ie the mapping from cube slices to IQUV should be different for other corr types. It's quite simple to change but I'm in hollidays right now! @tremou you could add a corr_type attribute to the class and change the mapping and the raise condition accordingly? (if you do so, create a separate branch).