You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am running SJARACNe through a VM using Oracle VM Virtual Box using a Ubuntu distro (24.04.01). I am running into the following error:
Did not find output with glob pattern : ['TF_run_001.adj']
This appears for every iteration of the bootstrapping. This appears to be because there are p values returning as 0 and p is restricted to (0,1], and thus I assume it is not creating the tmp TF_run_00X.adj file in the first place? I have tried uninstalling and reinstalling as others have suggested, but this does not resolve the issue.
I have also noticed that it throws the following error earlier in the output:
Originally, this CUDA error said it could not find nvidia-smi file, but after reinstalling the nvidia driver, it still throws the non-zero exit status above. However, as it continues to run after this I don't know whether this error results in the downstream issue with the p-values and TF_run_00X.adj files, or whether they are separate issues.
Does anyone have any idea how to overcome this ?
Thanks,
Alistair
The text was updated successfully, but these errors were encountered:
Hi ! I met the same problem just now !!! Does anyone know the solution ? Thanks a lot !
I have found something! The -pb argument sets the bootstrap p value (as opposed to -pc which sets the consensus value) - this should default to 1e-7 but it appears as thought it's defaulting to p=0... So setting -pb 1e-5 has allowed it to run. Hope that helps you too!
Hi Alistair ! Thanks a lot for your kind reply , I just tried as you suggested and this error disappeared .
Hi ! I met the same problem just now !!! Does anyone know the solution ? Thanks a lot !
I have found something! The -pb argument sets the bootstrap p value (as opposed to -pc which sets the consensus value) - this should default to 1e-7 but it appears as thought it's defaulting to p=0... So setting -pb 1e-5 has allowed it to run. Hope that helps you too!
I am running SJARACNe through a VM using Oracle VM Virtual Box using a Ubuntu distro (24.04.01). I am running into the following error:
Did not find output with glob pattern : ['TF_run_001.adj']
This appears for every iteration of the bootstrapping. This appears to be because there are p values returning as 0 and p is restricted to (0,1], and thus I assume it is not creating the tmp TF_run_00X.adj file in the first place? I have tried uninstalling and reinstalling as others have suggested, but this does not resolve the issue.
I have also noticed that it throws the following error earlier in the output:
Originally, this CUDA error said it could not find nvidia-smi file, but after reinstalling the nvidia driver, it still throws the non-zero exit status above. However, as it continues to run after this I don't know whether this error results in the downstream issue with the p-values and TF_run_00X.adj files, or whether they are separate issues.
Does anyone have any idea how to overcome this ?
Thanks,
Alistair
The text was updated successfully, but these errors were encountered: