Hello community,
I have a system with a Supermicro X10SRA motherboard with Intel Xeon E7 v4 CPU.
The GPU is TITAN Xp.
Running Ubuntu 20.04 with kernel version: 5.15.0-122-generic #132~20.04.1-Ubuntu SMP
The NVIDIA Driver is installed and seems working fine:
NVIDIA-SMI 535.183.01 Driver Version: 535.183.01 CUDA Version: 12.2
When I use normal graphics programs like CARLA simulator, there is no problem. But when a PyTorch program is started for training the system resets.
Is it possible to run the bug report program under these situations ? Should I try to run in safe mode ? Or downgrade my GPU driver to something more older say 470