Question about ordering of tracers for the sacc likelihood

[ajouellette]

Hi,

I'm trying to compute cmb lensing - cosmic shear power spectra and run forecasts with the sacc module. Everything runs fine when the tracers in the sacc file are ordered as (shear, cmbkappa), but the pipeline is not able to match theory to data for a file with the opposite tracer ordering.
Is this order hard coded or is there some way of setting up the pipeline so that arbitrary tracer ordering works?

From what I can tell this seems to be because shear-cmbkappa is a valid spectrum while the opposite is not in the project_2d module?

[joezuntz]

Hi @ajouellette,

The sacc likelihood is fairly new and is in development. The branch more-sacc adds a new option to it: flip.

You can set flip = cmbGalaxy_convergenceShear_cl_e or whatever data type you are calling it, and it will flip the ordering of that bin. You can put multiple space-separated data types there.

I think that might work for you, but if not then if you're able to share an example file then I can have a go debugging.

Cheers,
Joe

[ajouellette]

Oh cool, that option sounds like what I was looking for. I'll try that out.