Jupyter Notebook
Accelerator Server
important
When you use a machine learning environment, please change the current directory to /work_da area and submit the job to CA_001g queue.
Use interactive mode for Jupyter Notebook.
The command is as follows.
$ qsub -I -q CA_001g -v DOCKER_IMAGE=conda3/mlenv:tag $ jupyter notebook --ip=* --no-browser
例)
$ qsub -I -q CA_001g -v DOCKER_IMAGE=conda3/mlenv:cuda10.1-007 qsub: waiting for job 26322.gpu1 to start qsub: job 26322.gpu1 ready Access Port: <proto>://10.1.4.28:6037/ -> container port 8888 bash-4.2$ jupyter notebook --ip=* --no-browser ... To access the notebook, open this file in a browser: file:///home/hitacse/.local/share/jupyter/runtime/nbserver-12-open.html Or copy and paste one of these URLs: http://cgpu28-26322-cgpu28:8888/?token=3a4f5d232de317ef49f51630ba4038e47bdd51f7d191ee2c or http://127.0.0.1:8888/?token=3a4f5d232de317ef49f51630ba4038e47bdd51f7d191ee2c
Check the port corresponding to “container port 8888”. In the example above, it is 6037. (hereinafter referred to as “port A”)
And check the login token. In the example above, it is 3a4f5d232de317ef49f51630ba4038e47bdd51f7d191ee2c.
Port forward any port of your PC to port 22of gpu2 using new terminal. (hereinafter referred to as “port B”)
$ ssh -L port B:gpu2:22 username@cms-ssh.sc.imr.tohoku.ac.jp
(Example)
$ ssh -L 8022:gpu2:22 userA@cms-ssh.sc.imr.tohoku.ac.jp
Port forward any port of your PC to port of Jupyter Notebook using new terminal again. (hereinafter referred to as “port C”)
$ ssh -L port C:10.1.4.28:port A -p port B localhost
(Example)
$ ssh -L 15000:10.1.4.28:6037 -p 8022 localhost
Access the following URL from your PC.
http://localhost:port C/?token=login token
(Example)
http://localhost:15000/?token=3a4f5d232de317ef49f51630ba4038e47bdd51f7d191ee2c