Page 2 of 2

checkpoint files not getting created

Posted: April 01 2014
by Jason Dossett
Hi,

burn_in = X sets the number of user defined burn in steps that are done before any samples are recorded in your chains directory. It basically tells the program to throw away X samples before putting any samples in output chains. When running MPI the program still has its own routines that check when true burn in of the chains has completed.

Re: checkpoint files not getting created

Posted: March 10 2021
by Saisandri Saini
Hi,

I am running cosmomc-May2020 on workstation.The chains are stopped after 15 days.I have set convergence stop to 0.01.
When I am rerunninng the chains, I have got invalid check points error.
Here is the output:-

Code: Select all

BK15_dust data param indices:          49          50          51          52          53          54          55          56          57          58          59          60          61          62          63          64
 adding parameters for: plik_rd12_HM_v22b_TTTEEE
 plik_rd12_HM_v22b_TTTEEE data param indices:          65          66          67          68          69          70          71          72          73          74          75          76          77          78          79          80          81          82          83          84          85          86          87          88          89          90          91          92          93          94          95          96          97          98          99         100         101         102         103         104         105         106         107         108         109         110          46
 adding parameters for: DES_1YR_final
 DES_1YR_final data param indices:         111         112         113         114         115         116         117         118         119         120         121         122         123         124         125         126         127         128         129         130         131
 Fast divided into            3  blocks
 Block breaks at:           15          35
 54 parameters ( 7 slow ( 0 semi-slow), 47 fast ( 0 semi-fast))
           3 Reading checkpoint from chains/starobinsky_3.chk
           4 Reading checkpoint from chains/starobinsky_4.chk
 invalid checkpoint files
application called MPI_Abort(MPI_COMM_WORLD, 128) - process 3
I will be grateful if you can kindly let me know how can I solve this problem.

Re: checkpoint files not getting created

Posted: March 12 2021
by Antony Lewis
Are you re-running with the same number of chains and exactly the same parameters?