python-3.x - 当模型包含张量操作时，Pytorch DataParallel 不起作用

Question

如果我的模型仅包含 nn.Module 层，例如 nn.Linear，则 nn.DataParallel 可以正常工作。

x = torch.randn(100,10)

class normal_model(torch.nn.Module):
    def __init__(self):
        super(normal_model, self).__init__()
        self.layer = torch.nn.Linear(10,1)

    def forward(self, x):
        return self.layer(x)

model = normal_model()
model = nn.DataParallel(model.to('cuda:0'))
model(x)

但是，当我的模型包含如下张量运算时

class custom_model(torch.nn.Module):
    def __init__(self):
        super(custom_model, self).__init__()
        self.layer = torch.nn.Linear(10,5)
        self.weight = torch.ones(5,1, device='cuda:0')
    def forward(self, x):
        return self.layer(x) @ self.weight

model = custom_model()
model = torch.nn.DataParallel(model.to('cuda:0'))
model(x) 

它给了我以下错误

RuntimeError: Caught RuntimeError in replica 1 on device 1. Original Traceback (most recent call last): File "/opt/conda/lib/python3.6/site-packages/torch/nn/parallel/parallel_apply.py", line 60, in _worker output = module(*input, **kwargs) File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in call result = self.forward(*input, **kwargs) File "", line 7, in forward return self.layer(x) @ self.weight RuntimeError: arguments are located on different GPUs at /pytorch/aten/src/THC/generic/THCTensorMathBlas.cu:277

当我们的模型中有一些张量操作时，如何避免这个错误？

Answer 1

我没有使用 DataParallel 的经验，但我认为这可能是因为您的张量不是模型参数的一部分。你可以这样写:

torch.nn.Parameter(torch.ones(5,1))

请注意，您不必在初始化时将其移动到 gpu，因为现在当您调用 model.to('cuda:0') 时，这是自动完成的。

我可以想象 DataParallel 使用模型参数将它们移动到适当的 gpu。

参见 this answer有关 torch 张量和 torch.nn.Parameter 之间区别的更多信息。

如果你不想在训练过程中通过反向传播更新张量值，你可以添加requires_grad=False。

另一种可行的方法是覆盖 to 方法，并在正向传递中初始化张量:

class custom_model(torch.nn.Module):
    def __init__(self):
        super(custom_model, self).__init__()
        self.layer = torch.nn.Linear(10,5)
    def forward(self, x):
        return self.layer(x) @ torch.ones(5,1, device=self.device)
    def to(self, device: str):
        new_self = super(custom_model, self).to(device)
        new_self.device = device
        return new_self

或者类似这样的东西:

class custom_model(torch.nn.Module):
    def __init__(self, device:str):
        super(custom_model, self).__init__()
        self.layer = torch.nn.Linear(10,5)
        self.weight = torch.ones(5,1, device=device)
    def forward(self, x):
        return self.layer(x) @ self.weight
    def to(self, device: str):
        new_self = super(custom_model, self).to(device)
        new_self.device = device
        new_self.weight = torch.ones(5,1, device=device)
        return new_self