This repository was archived by the owner on Nov 2, 2024. It is now read-only.
Frigate Coral TPU Detector Inference Speed very slow #3910
Replies: 1 comment
-
Okay, Now the values of coral went dramatically down. |
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.
-
Hello,
I have a Proxmox system running on Dell Wyse 5070.
The Google Coral is installed as a M.2 Accelerator A+E unit.
I installed firgate with the Proxmox helper Script Frigate LXC.
The driver is installed on the host system, TPU is recognized and the Coral TPU is passed through to frigate.
Frigate recognizes the Coral TPU
Log: frigate.detectors.plugins.edgetpu_tfl TPU found
However I get the warning that Coral TPU is very slow.
Detector Inference Speed: almost constant at 394.24ms
Detector Temperature: 56°C
Detector CPU Usage: constant at 98,8 %
Detector Memory Usage: 0,9 %
Right now I configured 1 camera.
Beta Was this translation helpful? Give feedback.
All reactions