Yoni Yang comments

Results 20 comments of


                                            Yoni Yang

feat(ml): ML on Rockchip NPUs

Docker launch command: ``` docker run --security-opt systempaths=unconfined --security-opt apparmor=unconfined --device /dev/dri --device /dev/dma_heap --device /dev/rga --device /dev/mpp_service -v /cache:/cache:ro -p 3004:3003 -v /sys/kernel/debug/:/sys/kernel/debug/:ro --name rknnimmich_name -d rknnimmich ``` and...

feat(ml): ML on Rockchip NPUs

Most issues mentioned in the review should be fixed now.

feat(ml): ML on Rockchip NPUs

Wrote a matrix build script to see what model is not supported, all of them are `nlib` and `XLM`. https://github.com/yoni13/immich_to_rknn2/actions/runs/12892750773

feat(ml): ML on Rockchip NPUs

3000 photos, Everything default, rerun smart search, face detection&recognition RK3566 NPU: - 50mins - 65 c RK3566 CPU: - 4hr 37min - 76c GPU? `/dev/mali0` on RK3566 doesn't even work...

feat(ml): ML on Rockchip NPUs

Oh btw, I am not able to test RKNN's CORE auto mapping because I have a rk3566 which only has one core. Hope their auto mapping is useful, or maybe...

feat(ml): ML on Rockchip NPUs

According to docs, async mode is still underwork (?

feat(ml): ML on Rockchip NPUs

> async_mode=True Another question, if you rerun smart search and facial recognition at the same time, using default configuration (aka without modifying the rknn threads), will these two jobs spitted...

feat(ml): ML on Rockchip NPUs

> I got it working on my RK3588 and made some changes to make things cleaner. The unit tests need to be updated and the model export CLI needs to...

feat(ml): ML on Rockchip NPUs

I was having issues with GH codespaces while trying to fix tests yesterday, can you take a look? 😞

feat(ml): ML on Rockchip NPUs

I was thinking to change its data format before sending it to rknn to quiet the "the data format will be changed" warning But for the "dynamic range" one ,...