MICA-MNI / micapipe

micapipe from the Multimodal imaging and connectome analysis lab (http://mica-mni.github.io) at the Montreal Neurological Institute. Read The Docs documentation below
http://micapipe.readthedocs.io
GNU General Public License v3.0
80 stars 27 forks source link

building micapipe in docker on MacOS M1 #59

Closed slusna-dominika closed 1 year ago

slusna-dominika commented 1 year ago

Hi all,

when trying to run micapipe on my MacOS Monterey (12.2.1) with M1, I get an error relating to arm-based system of Mac since micapipe requires a linux/amd64 platform. One solution suggested on the net is to build micapipe before using it and so I have run docker build https://github.com/MICA-MNI/micapipe.git --platform linux/amd64 It started to build nicely installing several dependencies, but got stuck on step 6/23 when building FSL python environment:

error message

[ 6/23] RUN bash -c 'bash /opt/fsl-6.0.0/etc/fslconf/fslpython_install.sh -f /opt/fsl-6.0.0':                                                                
#8 8.533 Stage 1                                                                
#8 8.533 By installing this python distribution you agree to the license terms in
#8 8.533 /opt/fsl-6.0.0/fslpython/LICENSE.txt
100%.758 
#8 42.70 Stage 2
100%5.15 
#8 882.4 /opt/fsl-6.0.0/etc/fslconf/fslpython_install.sh: line 203:   759 Killed                  FSLDIR=$fsl_dir "${miniconda_bin_dir}/conda" env create -f "${script_dir}/fslpython_environment.yml" 2>> "${miniconda_install_log}"
#8 882.4        761 Done                    | ${script_dir}/progress.sh 133 ${quiet} >> "${miniconda_install_log}"
#8 882.4 Failed to create FSL Python environment - see /tmp/fslpythonesRS/fslpython_miniconda_installer.log for details
------
executor failed running [/bin/sh -c bash -c 'bash /opt/fsl-6.0.0/etc/fslconf/fslpython_install.sh -f /opt/fsl-6.0.0']: exit code: 4

M1 users, any suggestions on how to make micapipe build/run?

Thank you, Dominika

rcruces commented 1 year ago

We recommend you to download the container from dockerhub instead of trying to build it. micapipe has a complex environment and requires a lot of memory and time to build.

https://hub.docker.com/r/micalab/micapipe