NVIDIA Developer Forums

Pytorch network -> onnx -> tensorrt performance(run frequency) question

Robotics & Edge Computing Jetson & Embedded Systems Jetson AGX Xavier

AastaLLL December 7, 2023, 6:54am 8

Hi,

Based on the log, the GPU clock doesn’t fix to the maximal.

… GR3D_FREQ 7%@114
… GR3D_FREQ 99%@1109

Have you tried the command shared above?
Thanks.

Topic		Replies	Views	Activity
Inference Speed Jetson Xavier NX pytorch	6	904	April 12, 2023
Onnx -> TensorRT. No speed difference between models TensorRT	1	488	June 24, 2021
Jetson AGX Xavier INT8 Performance Jetson AGX Xavier	4	1781	October 18, 2021
How can i get 1000 FPS by running the inference with TensorRT Tiny-YOLOv3 (Jetson AGX Xavier) Jetson AGX Xavier tensorrt , yolo , onnx , fps	6	2952	October 18, 2021
Performance DECREASE with tensorRT under onnxruntime, pt2 Jetson AGX Xavier tensorrt	5	3069	May 25, 2022
Performance DECREASE with tensorRT under onnxruntime Jetson AGX Xavier tensorrt	2	801	March 8, 2022
Jetson AGX Xavier GPU RAM usage for object detection and instance segmentation inferencing Jetson AGX Xavier tensorrt , jetson-inference , pytorch , onnx	2	902	May 13, 2022
How to check performance of TensorRT/ONNX model running on Jetson NX Jetson Xavier NX tensorrt	4	513	February 2, 2022
TRT inference speed on two AGX Xavier TensorRT	1	313	September 12, 2021
FPS of yolov3 on xavier nx Jetson AGX Xavier yolo	5	1772	April 29, 2022