How to run a multilabel model on Xavier nx

Hi I’m trying to run a multilabel model but can’t find any support or tutorials for running my imported multilabel model. Any help would be greatly appreciated thanks.

Dear @mcameron.brix,
Are you asking how to get TRT model for your model? Could you share details of your test?

Hi, thank you for your reply.

I’m referring to running a multilabel model. I tried to used my .trt model with imagenet but it errors on label “(9 of 9) of 1” because it’s not set up to run multilabel.

I imagine that the vlm must be running multilabel because it’s attributing multiple labels to the image and stringing them together to form a coherent description of the scene.

If you could point me to any related tutorials, about how multilabel works on jetson-inference, that would be greatly appreciated.

Dear @mcameron.brix,
Could you check jetson-inference/docs/webrtc-recognizer.md at master · dusty-nv/jetson-inference · GitHub if it helps?

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.