How to use jupyter notebook with NVCaffe

docker container NVCaffe bas been installed on a remote server. Can I use it through jupyter notebook with the NVCaffe? Please let me know how to use jupyter notebook with NVCaffe.

Yes, you can. Run your container like this:

nvidia-docker run -p 8888:8888 ... nvcr.io/nvidia/caffe:19.03-py2 bash

Then inside the container, do this:

jupyter notebook --ip 0.0.0.0 --no-browser --allow-root

(Or you can also use jupyter lab similarly.)

Jupyter will give you back something that looks like this:

# jupyter notebook --ip 0.0.0.0 --no-browser --allow-root
[I 01:16:40.836 NotebookApp] Writing notebook server cookie secret to /root/.local/share/jupyter/runtime/notebook_cookie_secret
[I 01:16:40.993 NotebookApp] JupyterLab extension loaded from /opt/conda/lib/python3.6/site-packages/jupyterlab
[I 01:16:40.993 NotebookApp] JupyterLab application directory is /opt/conda/share/jupyter/lab
[I 01:16:40.994 NotebookApp] Serving notebooks from local directory: /workspace/translation
[I 01:16:40.994 NotebookApp] The Jupyter Notebook is running at:
[I 01:16:40.994 NotebookApp] http://(129873214987 or 127.0.0.1):8888/?token=987e585f8a8e...
[I 01:16:40.994 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[C 01:16:40.997 NotebookApp]

    To access the notebook, open this file in a browser:
        file:///root/.local/share/jupyter/runtime/nbserver-76-open.html
    Or copy and paste one of these URLs:
        http://(129873214987 or 127.0.0.1):8888/?token=987e585f8a8e...

Just replace the (129873214987 or 127.0.0.1) part with the actual hostname/IP address of your system, and you’ll be all set.

Thanks for quick response, Cliff! That works as you guided.

I am using Nvidia Tensorflow container. I run the container successfully and try to open Jupyter Notebook, but the URL that it generates does not work. I also do not have access to the file path that it provides. Do I need to have sudo access to be able to run Jupyter notebook?