External Email - Use Caution
I don’t believe the workaround will be supported or provided with future releases intended to run on Silicon Macs. If there is still an issue with mri_synthseg running with the current model files in the freesurfer dev build, then please let us know the details. - R.
On Feb 28, 2024, at 20:11, Sorby-Adams, Annabel <ASORBY-ADAMS@mgh.harvard.edu> wrote:Hi Helen,I managed to run mri_synthseg on my M2 Mac by coping over the model file to the 7.3.2 version of freesurfer, and installing the tensorflow workaround previously posted on the forum for apple silicon machines.I have not been able to get it to work on the 7.4 or most recent dev versions unfortunately. The work around for 7.3.2 isn’t ideal, but does the trick. Freesurfer gurus may have more insight though! AnnabelOn 28 Feb 2024, at 20:03, Friedrich, Helen <hfriedrich@bwh.harvard.edu> wrote:Hi R,Thanks for your quick response. We actually did that initially and it did not work unfortunately.Could you provide me with segmentations for a certain scan? Or do you see any other possibility to make it work? Best,Helen______________________ Helen Friedrich, B.Sc.Doctoral Student Netstim Laboratory Center for Brain Circuit TherapeuticsDepartment of Neurology Harvard Medical SchoolBrigham and Women’s Hospital 75 Francis St, Boston MA 02115 brighamandwomens.org <Outlook-rrf3zs4o.png> Von: fsbuild <fsbuild@contbay.com>Gesendet: Mittwoch, 28. Februar 2024 17:20An: freesurfer@nmr.mgh.harvard.edu <freesurfer@nmr.mgh.harvard.edu>Cc: Friedrich, Helen <hfriedrich@bwh.harvard.edu>Betreff: Re: [Freesurfer] Support Request NextBrain External Email - Use Caution The mri_synthseg script and python in the freesurfer disrtibution you are running are not entirely compatible with the arm64 M1, M2, M3 processors in silicon Macs. You could try installing and running the development build of Freesurfer for the silicon Mac via the link below. (After download, right click on the .pkg installer, and select open with default, then follow the prompts).https://secure-web.cisco.com/1lFsgk7vtXlatyZGhw2wfTf_WTyioo4BmLzbKYJKS0chTLA... would be interested to know if you run into such issues with the above development build. Please keep in mind a development build can be unstable and has not been tested like an actual release. Your numerical result may differ with the development build compared to the previous build/release.- R.On Feb 28, 2024, at 14:25, Friedrich, Helen <hfriedrich@bwh.harvard.edu> wrote:Dear Support Team,I'm reaching out for assistance with setting up NextBrain - mri_synthseg. Unfortunately, I could not make it work. There seems to be an issue with the np.float128-section. This error keeps showing up: <image.png>I have also tried to delete np.float128 from the script (as per suggestion from my colleague). However, the same error showed up. I'm working on a macOS M1. I would be very grateful for your assistance. Best, Helen ______________________ Helen Friedrich, B.Sc.Doctoral Student Netstim Laboratory Center for Brain Circuit TherapeuticsDepartment of Neurology Harvard Medical SchoolBrigham and Women’s Hospital 75 Francis St, Boston MA 02115 MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be brighamandwomens.org <Outlook-jaw0jfh4.png> _______________________________________________Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduMailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://secure-web.cisco.com/1rwDcenDHH042rGVnR9ADpR-TvoThnsOLxqL-WqiCsQXlod... mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer%3COutlook-rrf3z... mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr..mgh.harvard.edu/mailman/listinfo/freesurfer