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
bug fix: recompile a more up to date version of sciunit to be included in Flinc
feature request: the ability to select certain sciunit artifacts when running the repeat kernel
Use Case
I'm using FLINC to run and repeat several different notebooks. Audit and repeat work just fine when I execute the first notebook. I shut down the repeat kernel and move on to a totally different second notebook. The audit kernel works just fine once again, but the repeat kernel fails. @sranasir suggested this is because the repeat kernel is defaulting to using the e1 artifact instead of using the e2 artifact.
Feature Request
The ideal scenario would to be able to select the sciunit artifact of my choosing. So I could select e2 when selecting a kernel. For my use case, I wouldn't mind if I only had this option programmatically.
Bug Fix
The ideal workaround would be to use sciunit to delete the e1 kernel in between executing the first notebook with the repeat kernel and executing the second notebook with audit kernel.
sciunit rm e1
@sranasir suggested that this should be possible with the most up to date version sciunit. That there was a bug related the execution counter so that even after e1 is deleted using sciunit rm e1, that the next run in the audit kernel produces an e2 instead of e1 again, which causes some discrepancy.
Current Workaround
Therefore, the actual current workaround is to manually delete the audit kernel using
rm -rf ~/sciunit/audit-kernel/
./install.sh
NOTE: My install.sh (in the chameleon branch of my personal fork) is slightly modified, but I think the behavior should be the same in both
The text was updated successfully, but these errors were encountered:
Thank you for creating the issue as discussed. Meanwhile, I have updated the README so only one notebook per Sciunit project is run with audit and repeat kernels.
tldr
Use Case
I'm using FLINC to run and repeat several different notebooks. Audit and repeat work just fine when I execute the first notebook. I shut down the repeat kernel and move on to a totally different second notebook. The audit kernel works just fine once again, but the repeat kernel fails. @sranasir suggested this is because the repeat kernel is defaulting to using the
e1
artifact instead of using thee2
artifact.Feature Request
The ideal scenario would to be able to select the sciunit artifact of my choosing. So I could select
e2
when selecting a kernel. For my use case, I wouldn't mind if I only had this option programmatically.Bug Fix
The ideal workaround would be to use sciunit to delete the
e1
kernel in between executing the first notebook with the repeat kernel and executing the second notebook with audit kernel.@sranasir suggested that this should be possible with the most up to date version sciunit. That there was a bug related the execution counter so that even after
e1
is deleted usingsciunit rm e1
, that the next run in the audit kernel produces ane2
instead ofe1
again, which causes some discrepancy.Current Workaround
Therefore, the actual current workaround is to manually delete the audit kernel using
NOTE: My install.sh (in the chameleon branch of my personal fork) is slightly modified, but I think the behavior should be the same in both
The text was updated successfully, but these errors were encountered: