If enough iterations have been run, then you can just use what it has created. To do this, just re-run with the --no-sim option. If you need more iterations, then you can rename the csd files that are there. Currently, they will be numbered 001, 002, ..., 006. Just rename them to something like 101, 102, ..., 106. Then re-run mri_glmfit-sim (with the same csd name) to add the number of iterations you need. It will automatically find all the csd files when it runs surfcluster.
doug
Victoria Williams wrote:
Hi,
I am processing several multiple comparison jobs on launchpad:
'mri_glmfit-sim --glmdir lh.physio2_chol_final.cvram.115.thickness.07092009.doss.30.glmdir --sim mc-z 5000 1.3 mc-z.pos.1.3 --sim-sign pos --bg 6'
and I believe that the job is almost finished running the simulation (it has been running for a few days). However, launchpad, the batch system, is shutting down in a few hours and I am dissappointed that I may lose all the progress it's made. If interrupted, I probably need to create the "glmdir/csd/poll/StopSim" file since I used the --bg flag, correct?
Question: Even though the simulation might not be completely finished, would there be a reason NOT to manually finish the corrections procedure by picking up with the mri_glmfit-sim script at the "now run clustering" section? In other words running the mri_surfcluster command?
Thank You! Tori _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer