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
Hi,
I am having trouble with the pipeline hanging up on the denoising preprocessing step. I can run template creation fine if I remove the --denoise flag and include --bvalMap, --resample, and --create. I've tried increasing the amount of memory and decreasing the --nproc with no success getting it to finish within 7 days. It has successfully created denoised, bval mapped, and resampled files for all 20 subjects in my reference sample but the furthest it has gotten through my target sample is to make those same files for 4 out of 20 subjects.
I was wondering if you had any insight into what might be causing this or whether I simply need to let it run for longer. I was also wondering if the denoising step was necessary or if I could skip it and still complete the harmonization process. I noticed it wasn't part of the pipeline_test.sh (which did run successfully) but wasn't sure if that was just for the sake of efficiency.
Thank you,
Jenna
The text was updated successfully, but these errors were encountered:
To answer your question, we typically do not use --bvalMap or --resample unless your datasets have different b-values or spatial resolutions. The --denoise flag is also not commonly used.
Please ensure you include the --create flag to generate the harmonization template. Additionally, add the --debug flag to generate mean comparisons before and after harmonization.
You might be running out of memory due to the unnecessary steps.
Again, I apologize for the late response, and you may have already resolved the issues. Please let us know if you still need help, or we can close this GitHub issue.
Hi,
I am having trouble with the pipeline hanging up on the denoising preprocessing step. I can run template creation fine if I remove the --denoise flag and include --bvalMap, --resample, and --create. I've tried increasing the amount of memory and decreasing the --nproc with no success getting it to finish within 7 days. It has successfully created denoised, bval mapped, and resampled files for all 20 subjects in my reference sample but the furthest it has gotten through my target sample is to make those same files for 4 out of 20 subjects.
I was wondering if you had any insight into what might be causing this or whether I simply need to let it run for longer. I was also wondering if the denoising step was necessary or if I could skip it and still complete the harmonization process. I noticed it wasn't part of the pipeline_test.sh (which did run successfully) but wasn't sure if that was just for the sake of efficiency.
Thank you,
Jenna
The text was updated successfully, but these errors were encountered: