Hailo-8L environment setup

General
, ,
1

I’m looking for a step by step that would create a workflow from training to compile for my Hailo-8L (13 TOPS) hat I have on a Raspberry Pi 5. Ideally an end-to-end solution that is tested on Windows 11, using Docker for the environment(s). It seems that this configuration has been abandoned in the community for more modern software and hardware, thus I have not been able to find a working solution. I have two (recent 2025) links that seem to offer hope. 1) GitHub - cyclux/HailoConverter, offers the compile side and I assume will work on Windows 11. 2) Convert ONNX Models to Hailo8L: Step-by-Step Guide Using Hailo SDK, offers a more comprehensive solution, but is based on a Linux platform, not Windows 11. Perhaps I have missed the obvious and someone can provide a link that is exactly what I am looking for.

Any advise is appreciated.

2

Welcome to the Hailo Community!

I highly recommend using the Hailo AI Software Suite Docker on a x86 machine running native Ubuntu 22.04 or 24.04. You can use dual boot to switch between Windows and Ubuntu.

Inside the Docker you can run the built-in tutorials to learn about the model conversion workflow. You can start the Jupyter notebook server with the tutorial notebooks by calling:

hailo tutorial
3

Thanks for the reply. I’ll take a look at it. Ideally I would like a workflow that does not require booting between two OSs. Its Docker based so after looking at it, perhaps I can port it into WIndows Docker Desktop and run it from there. My runtime is all based in Windows (connected to RP5 running the Hailo-8L hat).

Cheers