Page 1 of 1

Cobaya running issue computing convergence extended DE and massive neutrino models

Posted: October 05 2021
by Helena Garcia
The Cobaya code gets stuck when computing the convergence and learn new proposal covmat with extended dark energy and massive neutrino models. There are six chains, and five python processes are ongoing. Debug=true is on, but output isn’t very revealing. Output for recent representative run is below.


[3 : mcmc] Progress @ 2021-10-04 03:14:55 : 50951 steps taken, and 17147 accepted.
[..]
[1 : mcmc] Progress @ 2021-10-04 03:23:46 : 51160 steps taken, and 17169 accepted.
[0 : mcmc] Progress @ 2021-10-04 03:24:00 : 50868 steps taken, and 17023 accepted.
[3 : mcmc] Progress @ 2021-10-04 03:24:11 : 51098 steps taken, and 17195 accepted.
[2 : mcmc] Progress @ 2021-10-04 03:24:25 : 50810 steps taken, and 16959 accepted.
[4 : mcmc] Progress @ 2021-10-04 03:24:27 : 50974 steps taken, and 17011 accepted.
[2 : mcmc] Learn + convergence test @ 16960 samples accepted.
[2 : mcmc] Ready to check convergence and learn a new proposal covmat (waiting for the rest...)
[0 : mcmc] All chains are ready to check convergence and learn a new proposal covmat

What may be a possible solution for this problem? Thank you in advanced!