Jump to content

    

Ресурсы и публикации по теме NN/ML+μС

предлагаю в теме делиться источниками по теме  NN/ML + микроконтроллеры

Share this post


Link to post
Share on other sites

μTensor

https://github.com/uTensor/uTensor - AI inference library based on mbed and TensorFlow

 

CMSIS-NN

https://www.dlology.com/blog/how-to-run-deep-learning-model-on-microcontroller-with-cmsis-nn/

http://arm-software.github.io/CMSIS_5/NN/html/index.html

 

Examples

http://aqibsaeed.github.io/2016-11-04-human-activity-recognition-cnn/ - Пример реализации CNN для определение паттерна поведения человека

https://community.arm.com/developer/ip-products/processors/b/processors-ip-blog/posts/deploying-convolutional-neural-network-on-cortex-m-with-cmsis-nn?CommentId=e0e42217-6618-4f2f-b4ef-03f56b85c107 - Пример реализации распознавания изображений на ARM + CMSIS-NN

Share this post


Link to post
Share on other sites

ST выкатила продукт под названием X-CUBE-AI

https://www.st.com/en/embedded-software/x-cube-ai.html

 

Quote

 

X-CUBE-AI is an STM32Cube Expansion Package part of the STM32Cube.AI ecosystem and extending STM32CubeMX capabilities with automatic conversion of pre-trained Neural Network and integration of generated optimized library into the user's project. The easiest way to use it is to download it inside the STM32CubeMX tool (version 5.0.1 or newer) as described in user manual Getting started with X-CUBE-AI Expansion Package for Artificial Intelligence (AI) (UM2526).

image.PF267702.en.feature-description-include-personalized-no-cpn-medium.jpg

The X-CUBE-AI Expansion Package offers also several means to validate Neural Network models both on desktop PC and STM32, as well as measure performance on STM32 devices without user handmade ad hoc C code.


 

 

Key Features

  • Generation of an STM32-optimized library from pre-trained Neural Network models
  • Supports various Deep Learning frameworks such as Keras, TensorFlow™ Lite, Caffe, ConvNetJs, and Lasagne
  • Supports 8-bit quantization of Keras networks and TensorFlow™ Lite quantized networks
  • Allows to run larger networks by storing weights in external Flash memory and activation buffers in external RAM
  • Easy portability across different STM32 microcontroller series through STM32Cube integration
  • Free, user-friendly license terms

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now