Page 2 of 2

Re: Cobaya: error with DES likelihood

Posted: September 25 2023
by Jesus Torrado
Since `self.acc` is used in different places, it may be the doubling it demands even more extrapolation in k (the `extrap_kmax=3000 * self.acc` line) as a side effect of increased accuracy somewhere else.

I see two possible solutions:

a) we increase the `3000`,
b) or we define a separate the accuracy parameter for `extrap_kmax`.

@Antony Lewis, what do you think?

Re: Cobaya: error with DES likelihood

Posted: September 26 2023
by Antony Lewis
For what specific cosmo/DES parameters are you getting the error, and are they reasonable?