CHI@Edge Tutorial
This artifact includes a Jupyter Notebook that will guide you through the CHI@Edge platform for IoT and edge research.
For more information about using the platform, check out our CHI@Edge documentation and python-chi's container module documentation, the primary interface for orchestrating experiments on CHI@Edge.
What is covered:
- Reserve a CHI@Edge device
- Launch a container on the device
- Interact with the container via python-chi
- Assign a public IP to a container
- Upload and download files to and from the container.
- Orchestrate a full experiment using a popular messaging queue (MQTT)
- New Training a neural network using the GPU (CUDA, PyTorch) on an Nvidia Jetson Nano
- Deprecated accessing camera data from devices w/ attached camera peripherals. Instead, see the newer standalone artifact tutorial showcasing the usage of a Pi Camera Module 3 on one of our devices to capture images and video.
Furthermore, check out our recent tutorial on enabling SSH on CHI@Edge to kickstart your CHI@Edge development with a familiar workflow.
We strongly welcome and encourage collaboration from our users on CHI@Edge. If you are interested in contributing development to the platform, whether in the form of container images with interesting workflows (e.g. SSH, VSCode server, GPU support, or other) or device settings/modifications that enable newer workflows, please contact soufianej@uchicago.edu.
Happy researching!
Launching this artifact will open it within Chameleon’s shared Jupyter experiment environment, which is accessible to all Chameleon users with an active allocation.
Download ArchiveDownload an archive containing the files of this artifact.