objectDetector_Yolo crash

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Hi,
The example code path is “/opt/nvidia/deepstream/deepstream-5.0/sources/objectDetector_Yolo”.

I can run this example correctly. However, my computer reboot suddently when I replace its config ,model to my model and run the program. I tried it a lot of times, my compute is still reboot when I ran the program.
My config input size is 416x416. I replace it to 608x608, it can work well. I don’t understand what the problem is.

This is my environment:
hardware:
CPU: i7-9700KF
GPU: GTX1080Ti
Power: 650w

Ubuntu: 18.04
CUDA: 10.2
DeepStream: 5.0
TensorRT: 7.0.0.11

3

Do you mean change input size from 416x416 to 608x608, reboot issue will gone?

4

Yes.

5

Please try docker, it’s hard to analyze the issue without getting any logs