Issue with auto_selfcal

[JamesAppleton99]

Following a conversation with John Tobin at the 21st NRAO Synthesis Imaging Workshop, I've been trying to use auto_selfcal, initially on a small-scale trial MS (two science target fields plus calibrator fields, all extracted from a larger MS).

I have followed the instructions for installation, preparing the data and running per Github. The procedure runs for a few minutes then stops with an error message.

Versions:

• MS - an extract of science target fields SB1-1-1 and SB1-2-1 plus all calibrator fields from 21A-215.sb39713536.eb40071722.59453.441739444446.ms. The MS was created via the NRAO portal using NRAO pipeline casa-6.6.6-17-pipeline-2025.1.0.32-py3.10.el8 then downloaded to the University of Hertfordshire computing cluster.
• auto_selfcal - v2.0.0
• CASA - casa-6.7.0-31-py3.10.el8

I attach CLI output from the procedure (file run_CLI.txt) and the CASA log file (20260616_094250).

Please don't hesitate to contact me if you'd like any additional information.

Regards,
James Appleton

casa-20260616-094250.log

run_CLI.txt

[jjtobin]

Hi James,

There is an assumption built into the program that the MS you give it will only have science targets. So I thiink that is the issue at present. If you run mstransform on your MS with a command like mstransform(vis='21A-215.sb39713536.eb40071722.59453.441739444446.ms', intent='TARGET, outputvis='21A-215.sb39713536.eb40071722.59453.441739444446_targets.ms') and then retry with the new MS that only has targets, I think that might resolve your issue.

John

[JamesAppleton99]

Hello John, Many thanks for the speedy response! You were right! I re-ran mstransfer so that only science target fields were in the MS for auto_selfcal to work on, and the procedure ran to completion. Unfortunately, the outcome was not a success as the run terminated with a message: Estimated_SNR_too_low_for_solint_EB. The reported SNR was ~7. Interestingly, running hif_selfcal on each of the two fields separately, allowing both phase and amplitude calibration, achieves SNR values of ~33 and ~380, and the two MSes can be jointly imaged into an acceptable mosaic by tclean. (Using the NRAO pipeline casa-6.6.6-17-pipeline-2025.1.0.32-py3.10.el8 for hif_selfcal and calling tclean with specmode=’mvc’ and gridder=’mosaic’.) I’ll continue experimentation with auto_selfcal on some more fields from my data set, and will let you know how I get on. Regards, James From: John J. Tobin ***@***.***> Sent: 16 June 2026 12:09 To: jjtobin/auto_selfcal ***@***.***> Cc: JamesAppleton99 ***@***.***>; Author ***@***.***> Subject: Re: [jjtobin/auto_selfcal] Issue with auto_selfcal (Issue #119) <https://avatars.githubusercontent.com/u/84106857?s=20&v=4> jjtobin left a comment (jjtobin/auto_selfcal#119) <#119 (comment)> Hi James, There is an assumption built into the program that the MS you give it will only have science targets. So I thiink that is the issue at present. If you run mstransform on your MS with a command like mstransform(vis='21A-215.sb39713536.eb40071722.59453.441739444446.ms', intent='TARGET, outputvis='21A-215.sb39713536.eb40071722.59453.441739444446_targets.ms') and then retry with the new MS that only has targets, I think that might resolve your issue. John — Reply to this email directly, view it on GitHub <#119?email_source=notifications&email_token=CGEE5V6LGOS4N4GX6QAPR2D5AETFLA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINZRHAYDQNBSGY3KM4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KYZTPN52GK4S7MNWGSY3L#issuecomment-4718084266> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/CGEE5V4O6O5DZYL6QHUCX6D5AETFLAVCNFSNUABFKJSXA33TNF2G64TZHM2DEOBXGA2DGMRYHNEXG43VMU5TINRXGM2DAOJXGY2KC5QC> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/CGEE5V56EKQZCSZ7OLAHEV35AETFLA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINZRHAYDQNBSGY3KM4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KUZTPN52GK4S7NFXXG> and Android <https://github.com/notifications/mobile/android/CGEE5V4PGPAY7GX5LRCFCIL5AETFLA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINZRHAYDQNBSGY3KM4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2K4ZTPN52GK4S7MFXGI4TPNFSA> . Download it today! You are receiving this because you authored the thread. <https://github.com/notifications/beacon/CGEE5V5UC6WF34GP2WGJU435AETFLBFCNFSM6AAAAAC2LOFRHSWGG33NNVSW45C7OR4XAZNMJFZXG5LFINXW23LFNZ2KUY3PNVWWK3TUL5UWJTYAAAAACGJYICVKM4TFMFZW63VGMF2XI2DPOI.gif> Message ID: ***@***.*** ***@***.***> >