Page 1 of 1

planck_NPIPE_highl_CamSpec.TTTEEE could not be found in COBAYA

Posted: May 02 2023
by Ali Rida Khalife
Hello!
I want to install the Planck NPIPE likelihood. I created a yaml file that includes:

Code: Select all

planck_NPIPE_highl_CamSpec.TTTEEE: null
in its likelihoods section. After running

Code: Select all

cobaya-install CamSpecNPIPE.yaml --packages-path ./Codes/
, which I successfuly used to install CamSpec2018 and CamSpec2021, I get the following error msg:

Code: Select all

[install] *ERROR* 'planck_NPIPE_highl_CamSpec.TTTEEE' could not be found. Tried loading internal and external classes. No component path was given.
[install] *ERROR* Component 'likelihood:planck_NPIPE_highl_CamSpec.TTTEEE' could not be identified. Skipping.
COBAYA has been upgraded to 3.3.1, yet still this is not working.
Could someone please help me with this?
Thanks!

Re: planck_NPIPE_highl_CamSpec.TTTEEE could not be found in COBAYA

Posted: May 05 2023
by Antony Lewis
If packages-path is not your default, you may need to specify it.
What was in CamSpecNPIPE.yaml ?

Re: planck_NPIPE_highl_CamSpec.TTTEEE could not be found in COBAYA

Posted: May 05 2023
by Ali Rida Khalife
Hi Antony,
Thanks for the reply!
I did specify the path to the packages, it's "home/Codes".
CamSpecNPIPE.yaml(attached) is the .yaml file I want to run.
Thanks for the help!

Re: planck_NPIPE_highl_CamSpec.TTTEEE could not be found in COBAYA

Posted: May 06 2023
by Antony Lewis
./Codes/ is only home/codes if you ran the install command in home. Check where the downloaded files are.

Re: planck_NPIPE_highl_CamSpec.TTTEEE could not be found in COBAYA

Posted: May 09 2023
by Ali Rida Khalife
Hi Antony!
The files for CamSpec2018 and 2021 were in the expected directory(home/Codes). However, the problem with NPIPE is that COBAYA is not even finding the likelihood, as was written in the error msg of my first post.
Thanks again for the help!

Re: planck_NPIPE_highl_CamSpec.TTTEEE could not be found in COBAYA

Posted: May 10 2023
by Antony Lewis
I don't know, maybe you have multiple cobaya or local/pip installs. You'd have to track down why it's not finding it.