Xavier is announced about 20 times about TX2, but I can not find any model benchmark on Xavier. Could you give some model benchmark, such as time cost of VGG, SSD, Mobilenet and YoloV3 with TensorRT?
Hi fwx5, see the slides from the Jetson AGX Xavier webinar here: https://devtalk.nvidia.com/default/topic/1041822/jetson-agx-xavier/nvidia-webinar-mdash-jetson-agx-xavier-and-the-new-era-of-autonomous-machines/
Is there a tutorial on how we can run these benchmarks below?
I want to verify performance at different temperatures and fan settings. And confirm that all the DLA, Vision accelerators and CPU cores are being utilized.
Hi keithdm, please refer to this post:
The commands to run are found here: https://developer.nvidia.com/embedded/jetson-agx-xavier-dl-inference-benchmarks#trtexec