pytorch: grad can be implicitly created only for scalar outputs

这个错误很早就遇到过但是没看到网上叙述清楚的，这里顺便写一下。
这里贴一下autograd.grad()的注释

grad(outputs, inputs, grad_outputs=None, retain_graph=None, create_graph=False, only_inputs=True, allow_unused=False)
    Computes and returns the sum of gradients of outputs w.r.t. the inputs.
    ``grad_outputs`` should be a sequence of length matching ``output``
    containing the pre-computed gradients w.r.t. each of the outputs. If an
    output doesn't require_grad, then the gradient can be ``None``).
    If ``only_inputs`` is ``True``, the function will only return a list of gradients
    w.r.t the specified inputs. If it's ``False``, then gradient w.r.t. all remaining
    leaves will still be computed, and will be accumulated into their ``.grad``
    attribute.
    
    Arguments:
        outputs (sequence of Tensor): outputs of the differentiated function.
        inputs (sequence of Tensor): Inputs w.r.t. which the gradient will be
            returned (and not accumulated into ``.grad``).
        grad_outputs (sequence of Tensor): Gradients w.r.t. each output.
            None values can be specified for scalar Tensors or ones that don't require
            grad. If a None value would be acceptable for all grad_tensors, then this
            argument is optional. Default: None.
        retain_graph (bool, optional): If ``False``, the graph used to compute the grad
            will be freed. Note that in nearly all cases setting this option to ``True``
            is not needed and often can be worked around in a much more efficient
            way. Defaults to the value of ``create_graph``.
        create_graph (bool, optional): If ``True``, graph of the derivative will
            be constructed, allowing to compute higher order derivative products.
            Default: ``False``.
        allow_unused (bool, optional): If ``False``, specifying inputs that were not
            used when computing outputs (and therefore their grad is always zero)
            is an error. Defaults to ``False``.

如下代码

>>> a=Variable(torch.FloatTensor([1,2,3]),requires_grad=True)
>>> b=3*a    
>>> autograd.grad(outputs=b,inputs=a)  # 这里b为向量
RuntimeError: grad can be implicitly created only for scalar outputs

因为计算梯度时outputs需为标量(未指明grad_outputs或grad_outputs为None时)，所以上面的代码会报错，而如下代码可以正常运行：

>>> a=Variable(torch.FloatTensor([1,2,3]),requires_grad=True)
>>> b=3*a
>>> z=b.sum()    
>>> autograd.grad(outputs=z,inputs=a) # 这里z为标量
(tensor([ 3.,  3.,  3.]),)

也可以通过指定grad_outputs，这时计算梯度就不再需要outputs为标量了，如下

>>> a=Variable(torch.FloatTensor([1,2,3]),requires_grad=True)
>>> b=3*a
>>> autograd.grad(outputs=b,inputs=a,grad_outputs=torch.ones_like(a))
(tensor([ 3.,  3.,  3.]),)

grad_outputs在GPU下时可写作以下形式

    grad_outputs = Variable(torch.Tensor(torch.ones_like(a)),requires_grad=False)