PyTorch on MicroControllers

We are working on deploying PyTorch on Micro Controllers with different sensors.

We are looking for optimization of exported model with quantization, fusion and other memory reduction techniques before exporting the model to onnx.

I’ve searched PyTorch Docs, Tutorials and Mobile Demos without luck.

If you’ve worked, planning to work or have interest, please leave guidance, comment and other helpful resources.

Thank you!

3 Likes

Has there been any update on this?
I’m interested in trying this too.

1 Like

Hello @anandijain, nice to connect with you here. We’re putting effort behind this to get it working. Check out bit.ly/deep-C

1 Like

Now you can try deepC without downloading or installing at http://tinyml.studio/

Here is a brief talk on how to bring ONNX and PyTorch models to your

This video takes you step-by-step through how to develop a deep learning model with PyTorch. Rohit Sharma from AI Tech Systems (AITS), a leading firmware, software and services provider for low power IoT, Endpoint and tinyML devices, demonstrates how PyTorch can be used on Arm Cortex-M based systems. Get an in-depth look at how to bring on-device AI applications to a multitude of verticals, including Industrial IoT, smart space and much more as AITS showcase a live demo using Arduino and STM32F07 boards.

https://link.medium.com/IhbcJzi4jgb

Running a PyTorch or ONNX machine learning model on a microcontroller is as easy as bringing the notebook to cAInvas platform using notebook server and compiling it to MCU board of your choice. Here are simple step by step instructions

https://medium.com/ai-techsystems/bringing-pytorch-models-to-microcontrollers-ec53d81c21d4

Hi, provide another tool for running PyTorch on Micro Controllers: Product- Tiny ONNC – Skymizer