I’m using Arduino Nano 33 BLE Sense, I want to do classification from external sensors (analog, i2c, etc.). I used data forwarder and create my impulse project. I created a impulse model similar to continuous motion documentation but using an external sensor classify signals. I run static buffer example, the impulse model on my Arduino run like a charm, I want to do it stand alone on my Arduino, what I’m trying to do is, can I used the continuous motion classification example as a pattern in my sketches, or it is under limitation of edge impulse? thank you.
Hi @TronixLab, definitely modifying the continuous motion example is the easiest. Just hook in your own sensor, and you should be good to go!
All code that we output, including our SDK and all examples, are open-source and licensed under the Apache 2.0 license - so you can modify, redistribute or deploy them without our permission. Would love to see the final project though!
@janjongboom, Thank you for your response, great! I’m doing a project, a low-cost mechanical ventilator, in response of high demand and collapsing supply of medical equipment such respirator, particularly in low and middle income country like ours. I want to classify the ventilator-patient synchrony, to automatically adjust the ventilator settings and parameters according to patient’s need. Very thanks to Edge Impulse!
For now, what I’m trying to do is to used built-in color sensor of Arduino Nano 33 BLE Sense if it may work. I facing challenges doing this thing. It doesn’t give a right inferencing. I going to post my code later. Then I will share will share my project to Edge Impulse community. Thank you. I’ll come back soon.
Edge Impulse is great! machine learning on microcontroller made easy. I made a color classification to recognized RED, and BLUE color, then modified the accelerometer continues example in Arduino library deployment, and it works!
Thank you! happy Holidays and happy new year! Edge Impulse