Pytorch遇到的坑：为什么模型训练时，L1loss损失无法下降？

最近在用L1loss做一个回归模型的训练，发现模型训练过程中loss及其不稳定，且训练效果很差，终于找到原因了！

原代码如下：

criterion = nn.L1Loss()
def train():
    print('Epoch {}:'.format(epoch + 1))
    model.train()
    # switch to train mode
    for i, sample_batched in enumerate(train_dataloader):
        input, target = sample_batched['geno'], sample_batched['pheno']
        # compute output
        output = model(input.float().cuda())
        loss = criterion(output, target.float().cuda())
        # compute gradient and do SGD step
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

以上代码问题出在：

loss = criterion(output, target.float().cuda())

我输入的batchsize是4，因此output的size是[4,1],也就是一个二维的数据；target的size是[4]。loss输出的结果是一个正确的数值。这也是我没发现问题的原因！我们看一下pytorch库里l1_loss的代码：

def l1_loss(input, target, size_average=None, reduce=None, reduction='mean'):
    # type: (Tensor, Tensor, Optional[bool], Optional[bool], str) -> Tensor
    r"""l1_loss(input, target, size_average=None, reduce=None, reduction='mean') -> Tensor

    Function that takes the mean element-wise absolute value difference.

    See :class:`~torch.nn.L1Loss` for details.
    """
    if not torch.jit.is_scripting():
        tens_ops = (input, target)
        if any([type(t) is not Tensor for t in tens_ops]) and has_torch_function(tens_ops):
            return handle_torch_function(
                l1_loss,