Yolov5 model on Drive AGX

Please provide the following info (check/uncheck the boxes after creating this topic):
Software Version
DRIVE OS Linux 5.2.6
DRIVE OS Linux 5.2.6 and DriveWorks 4.0
DRIVE OS Linux 5.2.0
DRIVE OS Linux 5.2.0 and DriveWorks 3.5
[#] NVIDIA DRIVE™ Software 10.0 (Linux)
NVIDIA DRIVE™ Software 9.0 (Linux)
other DRIVE OS version
other

Target Operating System
[#] Linux
QNX
other

Hardware Platform
[#] NVIDIA DRIVE™ AGX Xavier DevKit (E3550)
NVIDIA DRIVE™ AGX Pegasus DevKit (E3550)
other

SDK Manager Version
1.9.1.10844
[#] other

Host Machine Version
[#] native Ubuntu 18.04
other

Target Environment setup:
Software Version = DRIVE OS Linux 5.1.6 and DriveWorks 2.2
Target Operating System = Linux
Hardware Platform = NVIDIA DRIVE™ AGX Xavier DevKit (E3550)
SDK Manager Version = 1.8.1.10392
DRIVE_Software_10.0_Linux_OS_DRIVE_AGX_XAVIER
cuda version = 10.2
python version= 2.7,3.6

We are working on Traffic light detection and classification using Yolov5 model.

The following items are required for compile and run in Host and Target
a CUDA compatible GPU card
NVIDIA driver: 470.82.01
NVIDIA CUDA: 10.2
NVIDIA cuDNN: 8.2.1
NVIDIA TensorRT: 8.2.5.1
OpenCV: 4.5.4
cmake: 3.16 or higher

and exported yolov5s TensorRT engine file

Does Drive AGX supports the above mentioned version of dependencies?

In case I use yolov5 model on Drive AGX, will I have any compatibility issues.

Kindly clarify.

In case of incompatibility, kindly suggest the alternate solution to use Yolo models for detection.

Thanks.

Dear @kagitala.varalakshmi1,
DRIVE SW 10.0 includes cuDNN 7.5.1.14 and TensorRT 5.1.4.2.

You need to check generating TRT model on target using trtexec or TRT c++ API. If any of the layers are not supported natively, you need to implement them as a TRT plugin layer.

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