Model Performance API

Autonomous Machines Jetson & Embedded Systems Jetson Nano
#1

Hello Experts,

CC: @Honey_Patouceul @DaneLLL @amycao @kayccc @WayneWWW @icornejo.a @AastaLLL @dusty_nv

We would like to profile the performance of the model in the GPU core.

Is there any default application, library available as part of CUDA to check the Model’s performance in GPU ?

#2

Hi @techguyz, you can use the trtexec utility that comes with TensorRT (found under /usr/src/tensorrt/bin on your Jetson) to benchmark DNN models on the GPU.

There is also this jetson_benchmarks repo that uses TensorRT to benchmark various networks:

#3