Open castleKing1997 opened 5 years ago
It looks like the measurements_sigma
might be too small and it could be over fitting. There might be errors that are at the noise level you are trying to image at.
With the last release the measurements_sigma
value has not been using the correct units of Jy, so it is hard to set. This should be fixed in the next release, but I have been held since I just handed in my PhD Thesis. Sorry!
You might get a bit better results if you have relVarianceConvergence: 1e-3
change to relVarianceConvergence: 1e-4
.
Also, maybe you don't need the wide-field imaging algorithms on if you don't have a big w term?
I got better result with a smaller value of
measurements_sigma
. But the result is till not so good. There are some wave structure around point source and some extended noise on the background. I have uploaded the images here. Is there till any parameter needed updating in the configure file?