Powering the next generation of IoT with Embedded Machine Learning

День тому

The powerful combination of Edge Impulse #machinelearning at the edge and industry-leading low-power wireless solutions (nRF52 and nRF53 Series SoCs) from Nordic Semiconductor will truly unlock the full potential of the #IoT. Edge Impulse and Nordic are paving the way for every developer to eliminate time-consuming and difficult machine learning software development, deployment, and operations for Nordic silicon, in a singular powerful platform.
Watch and learn how to use Edge Impulse and Nordic’s nRF52840 and nRF5340 developer kits, to rapidly create complex motion detection, sound recognition, and image classification on low-power, memory-constrained, and remote edge devices for free, utilizing Nordic’s ubiquitous #Bluetooth Low Energy solutions.
Content:
00:00 Intro
02:57 Background on the "State of the IoT" and Embedded Machine Learning
08:30 Introduction to Edge Impulse
16:35 Edge Impulse collaboration with Nordic Semiconductor
21:07 Demo
This is a recording of one of the two live webinar sessions held on February 24, 2021. To watch the full webinar complete with Q&A sessions, click here: bit.ly/3dWS8gp
Slides from the presentations can be downloaded here (direct download link): bit.ly/2NQYeUM

КОМЕНТАРІ: 5

@felc95 3 роки тому

Hello. Are there any plans to make these examples work with the NRF52832 ? Or this feature is hardware dependant? Thanks!

@zinkyaw4261 3 роки тому

Hi Felipe, currently no plans but in theory it should work on the nRF52 DK as well but we have not tested it at this point in time. The main dependency is on nRF Connect SDK so I believe that also supports the nRF52832. If you have any further questions, please reach out to us at forum.edgeimpulse.com

@felc95 3 роки тому

@@zinkyaw4261 excellent. Thank you! I will try this soon. Currently working on a project based on reading IMU data. Implementing this would be a major feature.

@evgeni_logic 3 роки тому

Are there any plans to generate model outputs suitable for working in MicroPython or CircuitPython environments ?

@janjongboom7561 3 роки тому

So calling this from MicroPython or CircuitPython should be easy. There's a single entry point (run_classifier) that just takes a very simple structure of data - if you have a way of interacting with native code this should be straightforward. E.g. OpenMV is already doing this (with MicroPython).