@fenrisulfau commented on GitHub (Feb 3, 2024):
Oh also updated the nvidia drivers
@NickM-27 commented on GitHub (Feb 2, 2024):
@tkffaul can you show the detector and model sections of your config
@Remko76 commented on GitHub (Feb 2, 2024):
For comparison, with the same config this is the CPU load with 0.12.1
:
Maybe try running a different model
@tkffaul commented on GitHub (Feb 2, 2024):
i use a quadro because of the encode/decode limitations on the consumer side.
@NickM-27 commented on GitHub (Feb 2, 2024):
As I mentioned, I have a coral in my own system but experience the same issue when updating to 0.13.1
sure, and maybe they are caused by…
@fenrisulfau commented on GitHub (Feb 3, 2024):
i've got exactly the same issue. It was working perfectly until i updated.
@tkffaul commented on GitHub (Feb 2, 2024):
detectors: tensorrt: type: tensorrt device: 0 model: path: /config/model_cache/tensorrt/8.5.3/yolov7-tiny-416.trt input_tensor:…
@fenrisulfau commented on GitHub (Feb 3, 2024):
hey so for me, it was the changes to the the way tensorrt is handled in v13. ive got mine working now.
one thing ive noticed in your config: in…
@Remko76 commented on GitHub (Feb 2, 2024):
I will give that a try first and will create a seperate issue for my own system with those configs and screenshots. I will update this thread with the…
@shad0wca7 commented on GitHub (Feb 3, 2024):
hey so for me, it was the changes to the the way tensorrt is handled in v13. ive got mine working now.
one thing ive noticed in your config:…
@Remko76 commented on GitHub (Feb 2, 2024):
Definetely a lot better this way.…
@tkffaul commented on GitHub (Feb 2, 2024):
it did do a fresh compile of the model at install, just an FYI
@fenrisulfau commented on GitHub (Feb 3, 2024):
I'm using unraid, so what i ended up doing was deleting my existing image, deleting the model cache folder in the frigate config folder so all that…
@tkffaul commented on GitHub (Feb 2, 2024):
I see there is a newer driver, do you want me to upgrade? 535.154.05... I'm also running this same setup in another spot with a P4000 with the…
@NickM-27 commented on GitHub (Feb 2, 2024):
the detect fps is 32 which is a lot. You probably just need to tune motion detection on his setup https://docs.frigate.video/configuration/motion_detec…
@NickM-27 commented on GitHub (Feb 2, 2024):
I see you have a quadro, I think the versioning might be different for that. @NateMeyer can you confirm?
@NickM-27 commented on GitHub (Feb 2, 2024):
your CPU is not able to keep up with detection, we highly recommend using a Coral or GPU for object detection. You could most likely use your iGPU…
@shad0wca7 commented on GitHub (Feb 3, 2024):
you may want to regenerate your Frigate+ api key since it was included in that dump. I would let it run for 30 minutes and see if it finishes…