Dear FSL community,
I am running eddy_openmp (FSL 5.0.11) on a DWI data set.
The eddy command is:
eddy_openmp --imain=data.nii.gz --mask=brain_mask.nii.gz --acqp=./acqparams.txt --index=index.txt --bvecs=bvecs.txt --bvals=bvals.txt --topup=my_topup_results --very_verbose --out=data_ECC
Running on my FSL virtualbox with --very_verbose, it returns this error during the fourth "Evaluating prediction maker model" step:
Segmentation violation, Unknown reason, Offending address = (nil)
/lib64/libc.so.6 [0x7f6be582e67f]
eddy_openmp ) [0x6062eb] [
eddy_openmp ) [0x5ff163] [
eddy_openmp ) [0x45fab5] [
eddy_openmp ) [0x4625a5] [
eddy_openmp ) [0x5d9537] [
eddy_openmp ) [0x5d241a] [��k
eddy_openmp ) [0x463827] [��k
Segmentation fault (core dumped)
Running it again with the same files and command, it fails during the "Evaluating prediction maker model" step following the final outlier check.
There is an element of randomness where it will work once, then fail running a second time with the exact same command and data files. It also fails at different points but always within a "Evaluating prediction maker model" step.
There are no NaNs or Infs. I do not think it is the TOPUP files as other posts have suggested since the process runs successfully sometimes. I can provide files as well as the full verbose outputs, if desired. Any suggestions would be greatly appreciated.
Thanks,
Dan
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|