Licongxu dev

[licongxu]

No description provided.

[xzackli]

@licongxu this looks great; let me know when you think it's ready to merge. Your work has inspired me to develop some improvements to the HEALPix workspace which is now on the main branch. I'll probably test it against your HEALPix code!

[licongxu]

@licongxu this looks great; let me know when you think it's ready to merge. Your work has inspired me to develop some improvements to the HEALPix workspace which is now on the main branch. I'll probably test it against your HEALPix code!

I copied the queryDiscRing function from Healpix.jl from here https://github.com/ziotom78/Healpix.jl/blob/cd12157619442b979e4a075460d322e68416b7e3/src/query.jl#L56C1-L171C4
and it seems that function is working well at the poles.

For that Arnauld 10 pressure profile, I think we don't need a 3D interpolator on a (theta, M, z) grid, because that profile is only a function of theta/theta_500, so a 1D interpolator on theta/theta_500 grid, or a 2D interpolator on (theta, theta_500) would be sufficient. For the current implementation of 3D interpolator, I detected some small bias (but will not ultimately affecting cosmological constraint significantly) on halo simulations using A10 pressure profile at the power spectrum level, I think this is from interpolator or resolution. I am still testing the reason of this bias.