RCAC - Knowledge Base: Gilbreth User Guide: Jupyter Notebook - Deep Neural Networks Demo (GPU)

Rosen Center for Advanced Computing

Compute
Bell User Guide
Gilbreth User Guide
- Overview of Gilbreth
- Accounts
- File Storage and Transfer
- Gateway (Open OnDemand)
  - Interactive Apps
    - Compute Node Desktop
    - Windows Desktop
    - Jupyter Notebook
    - Jupyter Notebook - Deep Neural Networks Demo (GPU)
    - MATLAB
    - RStudio Server
  - Files
  - Jobs
  - Cluster Tools
- Software
- Compiling Source Code
- Running Jobs
- Frequently Asked Questions
- Biography of Lillian Moller Gilbreth
- Datasets
Weber User Guide
Scholar User Guide
Hammer User Guide
Negishi User Guide
Geddes User Guide
Anvil User Guide
Gautschi User Guide
Datasets
iGenomes
Software Catalog
Compilers
MPIs
Applications
Utilities
Biocontainers
NVIDIA NGC containers
AMD ROCm containers
FAQs
Storage
Data Depot User Guide
Fortress User Guide
REED Folder User Guide
Box Research Lab Folder User Guide
Scratch User Guide
Home Directory User Guide
Services
High-Performance Computing
Slurm
Depot Object User Guide
Rowdy User Guide
Environment Management with the Module Command
Protected Data Filesystem User Guide
Protected Data Archive User Guide
Purdue GenAI Studio
Environment Management with the Module Command
Environment Management with the Module Command

Expand Topics

Jupyter Notebook - Deep Neural Networks Demo (GPU)

The Notebook app will launch a Notebook session on a compute node and allow you to connect directly to it in a web browser. It can be used to run GPU applications such as Tensorflow and Keras. Below is a demo of this to get you started.

Download the demo notebook to your computer.
Launch a Notebook session from the Gateway Interactive Apps menu:

Open OnDemand launch page for Jupyter Notebooks — "Jupyter Notebook" can be found under "GUIs" in the "Interactive Apps" menu. This takes you to the launch page, with options for selecting the 'Queue', 'Number of hours', and email notifications.

Select the queue to which you wish to submit and enter the number of wallclock hours you require. Your notebook will be terminated after this number of hours elapses.
Click Launch.
Wait for your interactive session to change to Running state. This may take some time depending on how busy the queue and system is.
Click on 'Connect to Jupyter' once the button appears.

Active Jupiter Notebook session in Open OnDemand — When ready, the session will show a "Running state" with details about the session such as "Host", "Created at", "Time Remaining", and "Session ID". The "Connect to Jupyter" button will also become available.

Once in Jupyter, select 'Upload' in the upper right corner. You may wish to create a folder or change into a different directory to put the demo notebook first.

Upload button in a Jupyter Notebook — The 'Upload' button in a Notebook can be found in the upper right corner next to a directory selector and refresh button.

Select the demo notebook file you downloaded earlier. Click the blue Upload button to complete the upload. Then click the dnn.ipynb item from the file list to launch the notebook.
You should now have the notebook loaded and you should be able to re-execute the code cells, or modify them to your needs.

A running Jupyter Notebook — A running Notebook will have a main menu and toolbar buttons across the top with individually marked code and text cells below.

Helpful?