Clarification on kernels and bias

LSSTDESC / N5K

Non-local No-Nonsense Non-Limber Numerical Knockout

BSD 3-Clause "New" or "Revised" License

7 stars 7 forks source link

Clarification on kernels and bias #7

Closed EiffL closed 3 years ago

EiffL commented 3 years ago

Just to confirm, the bias is NOT included in the kernels right? I was naively expecting it was, and that the problem was literally to solve the integral from the README: but from a quick test, looks like I need to include the bias for the clustering probe if I want to match CCL. Can the organizers confirm?

Because I cheated and looked at what @tilmantroester was up to, it looks like he made the same assumption and is not loading the bias.

damonge commented 3 years ago

Oh, damn. Yes, you're right. Not sure why I didn't see that.

Humm, ok, I think it may be best for us to reload the kernels incorporating the bias so everything is consistent. Let me get back to you.

damonge commented 3 years ago

This was fixed (bias now provided with the kernels)