90% of the issues related to rig crashing or not detecting GPU are related to bad USB PCIe x1 to x16 riser failure. The only way to fix it is to change it. First, a user needs to find out which riser is bad: 1. Disconnect ½ of the GPUs and risers, leave the other ½ connected, and start mining. a.
Solved my SMI problem on Windows 7. It seems the old driver thats recommend here does not allow SMI access on Windows 7, I updraft do the windowx7 box to the current driver for win 7 (602) or something and SMI state changing was possible.
I have the code below and I don’t understand why the memory increase twice then stops I searched the forum and can not find answer env: PyTorch 0.4.1, Ubuntu16.04, Python 2.7, CUDA 8.0/9.0 from torchvision.models import vgg16 import torch import pdb net = vgg16().cuda() data1 = torch.rand(16,3,224,224).cuda() for i in range(10): pdb.set_trace ...