RuntimeError: only batches of spatial targets supported (3D tensors) but got 4

https://discuss.pytorch.org/t/runtimeerror-only-batches-of-spatial-targets-supported-3d-tensors-but-got-targets-of-dimension-4/82098

pytorch discuss:

解决方案：

target = torch.argmax(target, dim=1)

异常背景：

In my training function:

    #get values from dataloader
    X = normalize_zero_to_one(X) #input
    y = normalize_zero_to_one(y) #target

    images = Variable(torch.from_numpy(X)).to(device) # [batch, channel, H, W]
    masks = Variable(torch.from_numpy(y)).to(device) 
    optim.zero_grad()        
    outputs = model(images) 

    loss = loss_new(outputs, masks) # (preds, target)
    loss.backward()         
    optim.step() # Update weights  

I know the the target (here masks) should be [batch_size, w, h]. However, it is currently [batch_size, channels, w, h].