当我运行这段代码时，我得到了错误TypeError: 'Tensor' object is not callable：

torch.manual_seed(50)
epoch = 5
train_loss, train_acc = 0, 0
for epochs in range(epoch):
  for batch, (X, y) in enumerate(train_dataloader):
    model.train()
    y_logits = model(X)
    print(y_logits.shape)
    print(y.shape)
    loss = loss(y_logits, y)
    train_loss += loss
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()

错误发生在我的损失函数上。下面是我创建它的代码：

!pip install torchmetrics
from torchmetrics.classification import MulticlassAccuracy
loss = nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(params=model.parameters(),
                            lr=0.01)

有什么办法解决这个错误吗？