mirror of
https://github.com/harvard-edge/cs249r_book.git
synced 2026-05-07 10:08:50 -05:00
[GH-ISSUE #805] adding labs using arduino nano 33 ble sense #4206
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/cs249r_book#4206
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @MSMRo on GitHub (Apr 11, 2025).
Original GitHub issue: https://github.com/harvard-edge/cs249r_book/issues/805
I’d like to develop a series of hands‑on labs for the Arduino Nano 33 BLE Sense, which may contain:
No-Code ML with Edge Impulse: Introduce students to data collection, model training, and deployment entirely through Edge Impulse’s web interface—no programming required.
Code‑first ML with TensorFlow & TensorFlow Lite: Guide learners through writing Python/TensorFlow code for model design, then demonstrate how to convert and optimize models for on‑device inference using TensorFlow Lite and the X-Cube-DL (XDD) library.
Each lab may include:
An explanation of how TensorFlow Lite and xxd could infer the model into low‑power devices like Nano 33 BLE Sense.
I was doing something similar in a personal web page for the Peruvian community: https://tinymldoc.streamlit.app/.
Perhaps in the future, we can translate the content of the "Machine LEarning system" book to spanish.
@profvjreddi commented on GitHub (Apr 12, 2025):
Hi @MSMRo, thanks for your suggestion. I like the idea of doing it via a two-step process.
Did you happen to check out the labs here—labs?
We used to have the Nano 33 BLE sense, but Arduino has been wanting to step away from it for some reason, so we went down the route of the Nicla Vision.
I am adding @Mjrovai, who is our awesome labs guru-ji 🙏, so that he can chime in, but as long as the Arduino Nano 33 BLE implementation is consistent with the format we are following that'd be OK.
I tried accessing this page but it doesn't render or work. Could you please double-check @MSMRo?
That'd be awesome. I am currently working with a publisher
@MSMRo commented on GitHub (Apr 15, 2025):
Hi @profvjreddi,
Apologies for my delayed response.
Yes, I’ve reviewed the hands-on lab section. Those tutorials are very informative and well-organized. I also believe it would be fantastic to collaborate with Marcelo Rovai, as he is an expert in this area (TinyML).
Regarding my web page, I’m currently using the free-tier server provided by Streamlit, which can cause some delays during startup. That might be why the site appeared as shown in the image attached. To restore access, simply click on "Yes, get this app back up." It usually takes a few minutes to reload (I know, it's a bit unfortunate).
After waiting a few moments, the site should look like this:
In this tutorial, I aim to explain in a simple and accessible way how to understand the workflow of the TensorFlow Lite library in Arduino:
In this tutorial, I aim to explain in a simple and accessible way how to understand the workflow of the TensorFlow Lite library in Arduino:
@profvjreddi commented on GitHub (Apr 22, 2025):
Ah, got it — wonderful.
Not sure if you're aware of this, but here's a link that might be useful:
https://github.com/tinyMLx/courseware/tree/master/edX
Feel free to use the material however you see fit to teach others and share the knowledge!