An AttributeError when I use Fastsurfer on Docker about fastsurfer HOT 9 CLOSED

Did you create that Docker image yourself (based on what branch?). Maybe try our provided image docker pull deepmi/fastsurfer:cpu-v1.1.1 and run that one.

from fastsurfer.

Did you create that Docker image yourself (based on what branch?). Maybe try our provided image docker pull deepmi/fastsurfer:cpu-v1.1.1 and run that one.

Dear m-reuter,

Thank you for your help! With your help, I have solved this problem. But there is a new problem:

error: not enough inputs supplied to 'subject'
Linux f0c1a84b8035 5.15.0-46-generic #49~20.04.1-Ubuntu SMP Thu Aug 4 19:15:44 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux

recon-all -s subject exited with ERRORS at Wed Sep 7 05:15:32 UTC 2022

Is this problem from fastsurfer or freesurfer? What should I do?

from fastsurfer.

Can you send the exact command that you called? And probably I need to see the recon-surf.log file also.

from fastsurfer.

Can you send the exact command that you called? And probably I need to see the recon-surf.log file also.

The exact command is as follows:

docker run -v /home/zkli/Documents/my_mri_data:/data -v /home/zkli/Documents/ey_fastsurfer_analysis:/output -v /home/zkli/Documents/my_fs_license_dir:/fs_license --rm --user root deepmi/fastsurfer:cpu-v1.1.1 --fs_license /fs_license/license.txt --t1 /data/subject/orig.mgz --no_cuda --sid subject --sd /output --parallel

recon-surf.log is as follows:

cd /output/subject/mri
mris_place_surface --thickness ../surf/rh.white ../surf/rh.pial 20 5 ../surf/rh.thickness
Update not needed
#@# area and vertex vol rh Wed Sep 7 05:15:32 UTC 2022
cd /output/subject/mri
mris_place_surface --thickness ../surf/rh.white ../surf/rh.pial 20 5 ../surf/rh.thickness
Update not needed

mri_brainvol_stats subject

error: not enough inputs supplied to 'subject'
Linux f0c1a84b8035 5.15.0-46-generic #49~20.04.1-Ubuntu SMP Thu Aug 4 19:15:44 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux

recon-all -s subject exited with ERRORS at Wed Sep 7 05:15:32 UTC 2022

For more details, see the log file /output/subject/scripts/recon-all.log
To report a problem, see http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting

Command exited with non-zero status 1
@#@FSTIME 2022:09:07:05:15:30 recon-all N 3 e 1.73 S 0.33 U 1.47 P 104% M 32784 F 1 R 112830 W 0 c 94 w 907 I 8 O 824 L 1.05 1.12 1.72
@#@FSLOADPOST 2022:09:07:05:15:32 recon-all N 3 1.05 1.12 1.72

from fastsurfer.

Thanks, I was able to replicate your problem. I have a few ideas, what it could be and will investigate.

from fastsurfer.

OK, looks like the name of the subject cannot be "subject". recon-all -subject subject -segstats fails otherwise. This is a FreeSurfer bug that we inherit. More specifically mri_brainvol_stats subject fails when called from here: https://github.com/freesurfer/freesurfer/blob/e5ee495b3a2714a55cf0509a519f35b82524095e/scripts/recon-all#L4706
Seems like the argparse is getting confused here:
https://github.com/freesurfer/freesurfer/blob/e5ee495b3a2714a55cf0509a519f35b82524095e/mri_brainvol_stats/mri_brainvol_stats.cpp#L17

Anyway short story: don't ever use "subject" as the subject name :-)

from fastsurfer.

@m-reuter You are correct, freesurfer/utils/argparse.cpp works a bit funny in that way, because it "stores" "subject" in index and therefore recognizes "subject" as a flag (which it clearly is not), instead of the positional, that it actually would be.
In https://github.com/freesurfer/freesurfer/blob/e5ee495b3a2714a55cf0509a519f35b82524095e/utils/argparse.cpp#L225 it then goes into the flag branch instead of the positional branch....

from fastsurfer.

This needs to be fixed on the FreeSurfer side. I submitted a bug report there
freesurfer/freesurfer#1022
where further comments or fixes can go.

from fastsurfer.

was fixed upstream in freesurfer/freesurfer@53b8522 and i added a check for "subject" and error exit (as we use FS7.2) 474a852 . In future releases this can be tested and removed again.
@daozaichen hope that everything works for you now.

from fastsurfer.