gimlidc / raven-snippets

Set of scripts and configs for speedup Raven development

Raven snippets

Set of scripts and notes for speedup Raven development

Docker

The most of the code run on Raven should be placed in Docker. Useful templates and command can be found in docker folder.

• Python + PyTorch + NVIDIA example is present

Python

Installed package manager is Poetry. It creates (by default) .venv in your current folder by>

poetry init

JupyterLab

If you would like to run notebooks on Raven create virtualenv with necessary python packages (e.g. through poetry)

poetry new jupyterlab-venv
cd jupyterlab-venv
poetry add jupyterlab

1. log into Raven
2. run screen (to prevent SIGKILL for your jupyterlab)
3. in the screen ./.venv/bin/jupyter-lab --no-browser -p 33333
   • you can leave the screen by CTRL+a, CTRL+d (screen will run in the background)
   • if you want to go back to the screen use screen -r
   • if you want to close the screen press CTRL+d in the screen
4. in another terminal create tunnel for remote work ssh -N -L localhost:33333:localhost:8888 username@raven-ip
5. go to browser [http://localhost:33333/] and enter the token from the screen