- Login to your VM with GPU (interactive for jupyter-notebook)
$ ssh -L 8888:127.0.0.1:8888 $user@bdicdtvm01.bmrc.ox.ac.uk
- Activate your virtual environment
$ ml Anaconda3
$ source activate base
- Copy ipython-Notebook downloaded to your home folder
$ scp -r ~/Downloads/file.ipynb $user@bdicdtvm01.bmrc.ox.ac.uk:/cdthome/$user
- Start Jupyter-notebook where you have your
.ipynb files
- Using tensorboard (make sure you are logged in interactive session)
$ ssh -L 8889:127.0.0.1:8889 $user@bdicdtvm01.bmrc.ox.ac.uk
$ ml Anaconda3
$ source activate base
$ tensorboard --logdir runs/ --port 8889
$ http://127.0.0.1:8889/
- Check GPU usage (e.g. memory, processing etc)
$ ssh $user@bdicdtvm01.bmrc.ox.ac.uk
$ watch nvidia-smi *comment: will continously monitor your session*
- Data for segmentation task (also provided on Canvas)
/cdtshared/Cohort-2021/HDS-M05/segmentation/MontgomerySet