If your project contains a Jupyter notebook, there are two ways to run it. The first is to use the jupyter nbconvert command in a shell script, rendering it to HTML. The second is to run Jupyter interactively by launching a Jupyter Cloud Workstation environment in a separate tab.

To use Jupyter on Code Ocean:

If your capsule has a pip  or conda installer available, click the Cloud Workstation dropdown and select Jupyter:

This will automatically launch the typical Jupyter notebook interface.

Note: If you don't have pip or conda installers available, click the 'Environment' icon to access your environment configuration screen add python3-pip and python3-setuptools  via apt-get, and then add jupyter  via the pip3 installer that pops up.

General notes:

  1. Leaving a Cloud Workstation running will continuously use quota; click 'Shut Down Cloud Workstation' to close the session.
  2. Every 30 minutes, your interactive session will ask you if you are still there. is to prevent an idly open session from exhausting your quota. 
  3. Changes to files in /code  and /data  in a Cloud Workstation session will be synced back to your capsule. Any changes to /results  or the environment will not be. Please record changes you made to either, and re-implement them in a non-interactive session, if you wish to save them.


Did this answer your question?