piyanatk
commented
2 years ago Awesome @alphatangojuliett. I am interested to help getting this implemented. I think it might be nice if you can schedule a zoom call with some of us and walk us through the codes. Then, we can brainstorm how to implement it (if @steven-murray haven't thought about it already :)). I will try to look through the codes over the next few week or so too.
@steven-murray , I made a new H4C milestone for this and add several tags. Please feel free to change if they aren't right.
r-pascua
After some discussion, Steven and I feel that the first-order coupling algorithm presented in 2110.10879 can be incorporated into hera_sim.
Basic structure: input into the algorithm non-coupled, simulated visibilities in .uvh5 file format. Output the same visibilities, but with coupling according the the first-order algorithm.
Here you can find the functions associated with the first-order semi-analytic mutual coupling code, which were discussed in the datacon on October 29th 2021. These are the functions I would intend to incorporate into hera_sim.
A demo of how to use that code is in the tutorials section of the repository. The repo also has a link to the semi-analytic coupling paper, which has been submitted to MNRAS. The 'FOC_plotting_funcs' script may be useful for handling data and making plots, but has hera_pspec as a dependency, and so probably doesn't want to live in hera_sim.
The functions script contains both a multi-thread-capable version of the coupling algorithm as well as a non-multi-threaded version. The demo hyperlinked above uses the multi-threaded version of the code; implemented via concurrent.futures