HPC friendly Python + Neuroimaging analysis container environment
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
.gitignore
README.md
Singularity

README.md

Cosanlab Singularity Analysis Container

This is a Singularity spec file that can be used to build a container and run on Dartmouth's Discovery HPC cluster. It is a Docker -> Singularity bootstrap of our analysis container

You can either:

  1. Copy the built singularity image from Discovery, located at /ihome/ejolly
  2. Pull the container from Singularity-Hub singularity pull shub://cosanlab/cosanToolsSingularity:master
  3. Build/modify the container from scratch

Building a container from scratch with this repo

You'll need a local machine with sudo privileges and singularity installed. If you're running OSX you can follow the directions here to get a vagrant VM running to do this. Then proceed with:

#on your local machine with sudo privileges
sudo singularity create --size 8000 myContainer.img

#Singularity in the command below is the spec file in this repo, adjust path accordingly
sudo singularity bootstrap myContainer.img Singularity

#You might need to copy the .img out of your vagrant vm first if you're using one; by default /vagrant is shared with your host OS
scp myContainer.img user@discovery.dartmouth.edu:~


#on discovery, from ~
module load singularity
singularity run myContainer.img

#OR
singularity exec ./myContainer.img someCommand

#to mount a folder with data
singularity exec -B /path/to/data:/mnt ./myContainer someCommand