[Detector Support]: ONNX Detector times out and is killed everytime it is turned on. #20867
Unanswered
gg4414
asked this question in
Detector Support
Replies: 1 comment
-
|
This is probably a driver issue, the 5060Ti has a lot of issues, I had my own troubles using mine for other things (Ollama LLM). Especially trying to use it through WSL which is not officially supported and more complicated. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Describe the problem you are having
I have an Nvidia 5060ti which I was hoping to use with Frigate. Found ONNX needs to be used in documentation. Downloaded a few yolo_nas_s.onnx or yolo_nas_l.onnx models from huggingface. After sorting out other problems related to image size and input type, no error shows up.
However, whenever I turn on detection for the stream, the video stream freezes immediately.
Logs show: INFO : Turning on detection for my_camera
Then approx 12-15 sec later: INFO : Detection appears to be stuck. Restarting detection process...
INFO : Waiting for detection process to exit gracefully...
And this loop continues. Video stream is only visible if I turn off detection after this.
In the logs the ffmpeg error was fixed by itself, video was clear. It paused right after detection turned on. Unpaused after "INFO : Turning off detection for my_camera".
Version
0.16.2-4d58206
Frigate config file
docker-compose file or Docker CLI command
Relevant Frigate log output
Install method
Windows WSL2
Object Detector
TensorRT
Screenshots of the Frigate UI's System metrics pages
Any other information that may be helpful
I used to use frigate on my CPU a few months ago, using OpenVino. It worked fine back then. TensorRT has been a nightmare...
Beta Was this translation helpful? Give feedback.
All reactions