CPS Class Instructions¶
Important
The following instructions will only work for CPS classes.
These instructions describe the process of opening a CPS JupyterLab environment on the Open OnDemand (OOD) on the cluster web portal and accessing class /courses
folders.
Note
Due to problems with launching OOD on Safari, we recommend using Google Chrome, Mozilla Firefox or Microsoft Edge browsers instead for best experience.
Open JupyterLab For Classes¶
Important
The class instructor needs to fill in the: [Discovery Classroom Use Request] You will only be able to find your class resources if a request was already made.
In a web browser, go to http://ood.discovery.neu.edu. Login with your NU credentials.
Under the Courses menu, select your Class Name (For example: ALY6080 JupyterLab).
Select the default options and click Launch. Wait until the session is successfully created and ready to be launched (turns green).
For more control of the session, modify Time for the session time (in hours), Memory to get more memory in GB, and the Working Directory where JupyterLab will launch.
Note
If Working Directory is left blank, the session will launch in the main class folder (in this example
/courses/ALY6080.202335/data
). Alternatively, start the session directly from your personal working directory by entering:/courses/ALY6080.202335/students/[username]
, where[username]
is your username on Discovery. The instructions below assume the field is left blank.courses/ └── ALY6080.202335 ├── data ├── staff │ ├── doug │ ├── evelyn │ └── frank └── students │── alice │── bab └── cindy
The above file tree represents the folder structure for the
ALY6080.202335
course.Click Connect to Jupyter to open JupyterLab. This will open a JupyterLab interface in another tab.
Caution
Select Cancel when prompted with the Build Recommended option. The package jupyterlab-dash does not require a build, and will not work when build is enabled.
Access Class Directories¶
After you are connected to a CPS JupyterLab session on OOD, you can access any shared class directories and your private class directory.
You can navigate between the class folders using the left menu. The files in the shared class directory are read ony and the students do not have permissions to edit or remove the files. Your instructor may use the shared class directory to share files related to the classwork
Navigate to your personal class directory under:
/courses/[coursename.termcode]/students/<your username>
. Now you can create and edit Jupyter notebook files here.Open a new Python notebook session from the Launcher menu by clicking the Python 3 (ipykernel).
A new file will be created inside your directory called Untitled.ipynb. You can rename it by right-clicking on it and using the rename option.
This Python notebook has ready-to-use Python packages needed for your class.
Note
Permission Denied errors:
Do not attempt to create, edit or write files that are outside your personal student directory. Most “Permission Denied” errors are due to directories or files having read-only access permissions.