使用 CUDA 相对较新。经过一段看似随机的时间后，我不断收到以下错误： RuntimeError: CUDA error: an invalid memory access was Been应付
我见过人们提出诸如使用cuda.set_device()而不是cuda.device()设置之类的建议torch.backends.cudnn.benchmark = False
但我似乎无法让错误消失。这是我的一些代码：
torch.cuda.set_device(torch.device('cuda:0'))
torch.backends.cudnn.benchmark = False
class LSTM(nn.Module):
def __init__(self, input_dim, hidden_dim, num_layers, output_dim):
super(LSTM, self).__init__()
self.hidden_dim = hidden_dim
self.num_layers = num_layers
self.lstm = nn.LSTM(input_dim, hidden_dim, num_layers, batch_first=True, dropout=0.2)
self.fc = nn.Linear(hidden_dim, output_dim)
def forward(self, x):
h0 = torch.zeros(self.num_layers, x.size(0), self.hidden_dim).requires_grad_().cuda()
c0 = torch.zeros(self.num_layers, x.size(0), self.hidden_dim).requires_grad_().cuda()
out, (hn, cn) = self.lstm(x, (h0.detach(), c0.detach()))
out = self.fc(out[:, -1, :])
return out
def …从零使用开始：
>>> import gc
>>> import GPUtil
>>> import torch
>>> GPUtil.showUtilization()
| ID | GPU | MEM |
------------------
| 0 | 0% | 0% |
| 1 | 0% | 0% |
| 2 | 0% | 0% |
| 3 | 0% | 0% |
然后创建一个足够大的张量并占用内存：
>>> x = torch.rand(10000,300,200).cuda()
>>> GPUtil.showUtilization()
| ID | GPU | MEM |
------------------
| 0 | 0% | 26% |
| 1 | 0% | 0% |
| 2 | …我正在尝试在启用 GPU 的情况下在 Google Colab 上初始化张量。
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
t = torch.tensor([1,2], device=device)
但我收到了这个奇怪的错误。
RuntimeError: CUDA error: device-side assert triggered CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1
即使将该环境变量设置为 1 似乎也没有显示任何进一步的细节。
有人遇到过这个问题吗？