Open d3v-null opened 7 months ago
Bonus points for calibrating from model column of measurement set to allow for wsclean self-cal
This may be more confusing than helpful, but this code here https://github.com/ADACS-Australia/PyFHD/blob/main/PyFHD/source_modeling/vis_model_transfer.py is how I setup PyFHD
to read WODEN
models and match to real data. The annoying thing being WODEN
doesn't flag tiles, so if you're calibrating a data set which has missing tiles, you have to do a bunch of book keeping to work out which visibilities in the model match what's in the data.
This would allow model visibilities to be generated in Woden.
Would also allow chunks to be done in multiple vis-simulate calls, and combined separately, as an alternative to https://github.com/MWATelescope/mwa_hyperdrive/issues/9