-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
The current N4 bias correction binary is not deterministic across platforms #2
Comments
Hi Christian, Would you mind elaborating on the non-deterministic effect you've observed? I'm in the process of migrating MALPEM (+ N4) from one server to another (both running Ubuntu 10) and I'm seeing very slightly different results in bias field correction across servers. Digging around in the log files, I can see that the N4 convergence estimates differ very slightly across machines, is this similar to what you're seeing? Best, Richard |
Hi Richard, Yes, what you describe sounds like a confirmation of this issue. This is a problem with the N4 binary that ships with MALPEM not MALPEM itself. You can try recompiling a more recent version of ITK N4 and replace the binary. Please let us know if that solves the problem in case you move forward with this. Christian |
Will do, thanks! Richard |
Hi Christian, I ended up fixing the issue by fixing the ITK_GLOBAL_DEFAULT_NUMBER_OF_THREADS variable across servers, as per this post: https://sourceforge.net/p/advants/discussion/840261/thread/354df4c0/?limit=25#116c Hope this is useful! Richard |
Thanks a lot for sharing this information, very useful! Best wishes, |
This is a problem as it hampers reproducibility.
The text was updated successfully, but these errors were encountered: