Got negative prediction value

[Younger330]

Hello, I used your model for training in our Trainer, but the prediction is negative
Here is my code. Am I wrong? Or would you have any suggestions?
`images=images.unsqueeze(4)
out=umamba.forward(images)

    pred=out[0].squeeze(2)
    pred=F.log_softmax(pred,dim=1)
    #the predict is negative

`

[Younger330]

The problem is got by model.apply(InitWeights_He(1e-2)), but I dont know how it influence, maybe only used for radiological images?

Still have this question, although use the loss in nnunet

[Younger330]

Here is my training code:

train_dataloader = DataLoader(train_dataset, batch_size=4, shuffle=True)



umamba=get_umamba_bot_2d_from_plans(num_input_channels=3)
umamba=umamba.cuda()


epochs = 500
umamba.train()
optimizer = torch.optim.Adam(umamba.parameters(), lr=0.001)#,weight_decay=0.005)

dc_cd_loss=DC_and_CE_loss()

for epoch in tqdm(range(epochs),desc='Epoch'):

    epoch_loss=0
    mean_cross=0
    mean_dice=0

    for images, gt in tqdm(train_dataloader):
        images=images.cuda()
        images=images
        gt=gt.cuda()
        images=images.unsqueeze(4)
        out=umamba.forward(images)
        pred=out[0]
        pred=F.log_softmax(pred,dim=1)#negative prediction wheter delete or not
        gt=gt.unsqueeze(4)

        
        loss=dc_cd_loss(pred,gt)


        
        loss.backward()
        optimizer.step()