Strange sigma8 about cobaya HOT 9 CLOSED

I took a look of the source code and sigma8 has inverted redshift

def get_sigma8(self):
r"""
Get :math:\sigma_8 values (must previously have calculated power spectra)

    :return: array of :math:`\sigma_8` values, in order of increasing time (decreasing redshift)
    """
    mtrans = self.get_matter_transfer_data()
    return mtrans.sigma_8[:, 0]

while cobaya just returns the first element (the highest redshift). But then I have another question - how can we be sure that the last element is redshift zero. I couldn't find anywhere the z array that sigma_8 is evaluated.

from cobaya.

If sigma8 is calculated, CAMB always include z=0, so should always be the lowest redshift. The code you quote looks like a pre-v1.0 version of CAMB?

Looks as though get_derived is Cobaya may be returning them the wrong element of the array.

from cobaya.

Hi @vivianmiranda,

Any update on this? Can you confirm that you are using Cobaya>1.1, for CAMB >1.0 support?

from cobaya.

from cobaya.

OK, I'll look into writing a patch for the cobaya 1.0 tree (camb < 1.0) on Monday. Sorry!

from cobaya.

The index is probably also wrong in the 1.0 branch? (should be -1 not 0)

from cobaya.

Taking a look at it right now. For now, I am getting the correct sigma8 for that Cosmology (~0.827) in both Cobaya 1.0 and 1.1. I guess to reproduce your error I need to ask for more than 1 redshift, so that the wrong one is picked? I am going to try that.

from cobaya.

Ok, can confirm. I am going to try to fix it in both 1.0 and 1.1.

from cobaya.

Fixed in both branch 1.0 (deploying as v1.0.5) and master (deploying as v1.1.2)

Thanks a lot for the report!

from cobaya.