mlmandude

If you use mmap/munmap directly within your application you could probably get into trouble by hardcoding the page size.

Who exactly are DJI/Hesai copying? They are the market leaders by a mile.

This is what was said about Japan prior to their electronics industry surpassing the rest of the world. Yes, china does copy. However, in many instances those companies move faster and innovate faster than their western counterparts. Look at the lidar industry in china. It's making mass market lidar in the tens of thousands [see hesai]. There is no american or european equivalent at the moment. What about DJI? They massively out innovated western competitors. I wouldn't be so quick to write off that country's capacity for creativity and technological prowess.

It looks like executorch is for edge devices (phones / IoT / etc).

I'm currently doing inference on GPUs with libtorch and have a few concerns: (1) It seems like libtorch/torchscript are on a path to getting deprecated and (2) libtorch/torchscript pull in enormously bloated libraries. Should I be looking at executorch? I currently don't see an nvidia backend / integration with tensor rt in https://github.com/pytorch/executorch/tree/main/backends , but seems like it might be possible. Is this something you are thinking about?