Jupyter Lab Sessions

In order to start a jupyter lab session within an enroot container, create an sbatch script as demonstrated below. Pick a port number (recommendation: pick a number between 1024 and 9999) and create your script.

#!/bin/bash

#SBATCH --qos=low
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=2
#SBATCH --gpus-per-task=1
#SBATCH --mem-per-cpu=6G
#SBATCH --time=04:00:00
#SBATCH --job-name=jupyter_lab
#SBATCH --no-container-entrypoint
#SBATCH --container-mounts=/data/pathology:/data/pathology
#SBATCH --container-image="doduo1.umcn.nl#rtc/general:interactive"

jupyter lab --ip=0.0.0.0 --port=<port> --no-browser

Make sure to replace <port> with the value you chose, and kick off the script via sbatch <script_name>.sh.

Then, check the output of your job by running tail -f slurm-<job_id>.out. From there, you'll get the jupyter lab session URL when it's up and running. Simply copy & paste it in a new window (should look like http://<nodename>:<port>/lab?token=...).

Once you're done, do not forget to cancel your job via scancel <job_id>.

NB: in case the port you specified is already taken on that machine, you'll automatically get assigned a new port.