Using Conda Environments
Using Widgets
Requesting a GPU
Do Not Run Jupyter on the Login Nodes
Base Conda Environment
Custom Conda Environment
Another Approach
Running as a batch job
#!/bin/bash #PBS -m abe #PBS -M YourEmail@griffithuni.edu.au #PBS -N cembd_jlab #PBS -q dljun #PBS -W group_list=deeplearning -A deeplearning ###Other options group_list=aspen -A aspen ### Number of nodes:Number of CPUs:Number of threads per node. ###If not using gpu,you should not request ngpus #PBS -l select=1:ncpus=16:ngpus=1:mem=100gb,walltime=600:00:00 ###PBS -l select=1:ncpus=32:ngpus=0:mem=100gb,walltime=300:00:00 ### Add current shell environment to job (comment out if not needed) # The job's working directory cd $PBS_O_WORKDIR module load python/3.8.8 module load gcc/4.9.3 source /usr/local/bin/s3proxy.sh unset PYTHONPATH source venv_temp_py38/bin/activate jupyter lab --no-browser --port=5678
After running the above script and the job has started on the compute node, you run the following command on your local machine to start port forwarding.
ssh -CNL 5678:localhost:5678 s123456@n060.rcs.griffith.edu.au