Hi I would like to know if anyone can tell me what fps counts to expect on the Xavier NX
when using Tensorrt with yolov3-tiny-416 (FP16) if I do everything right.
I was currently only able to have 13fps with 416x416 images in Docker container
regardless of the power mode. The yolo model was created by converting weights to onnx and then converting onnx to .trt . the code calling tensorrt is in python2 and jetpack is 4.4DP