[SOLVED] Should I use torch.clamp after torch.sigmoid?

(Andrei Tenescu) #1

Hello, I’ve tried to write a custom loss function, a Weighted Binary Cross Entropy. As suggested by @miguelvr here:

I’ve tried to use the following function (wrapped inside a class):

def weighted_binary_cross_entropy(output, target, weights=None):
    if weights is not None:
        assert len(weights) == 2
        loss = weights[1] * (target * torch.log(output)) + \
               weights[0] * ((1 - target) * torch.log(1 - output))
        loss = target * torch.log(output) + (1 - target) * torch.log(1 - output)

    return torch.neg(torch.mean(loss))

The problem is that sometimes this outputs nan or -inf.

Tracing it back I reached the conclusion that sometimes my model outputs very small numbers, for example -136. This leads to:

torch.sigmoid(torch.tensor([-136.])) -> tensor([0.]) which leads to -inf in torch.log().

I am on pyTorch 1.0.1.

Is it ok to use torch.clamp(torch.sigmoid(torch.tensor([-136.])), min=1e-8, max=1 - 1e-8) ?

[SOLVED] Class Weighed Binary Crossentropy not working, even with equal weights
(Miguel Varela Ramos) #2

that function had that problem, in the end I used the NLLLoss.

But it is indeed okay to do that to fix the problem.

(Andrei Tenescu) #3

Thank you, I tried to use it like this, but it doesn’t work. The network does not train, even when the weights are (0.5, 0.5). Do you have any idea why?

(K. Frank) #4

Hello Andrei!

The short (narrow) answer – use LogSigmoid.

Please see my reply to your “Weighted Binary Crossentropy”


K. Frank

(Andrei Tenescu) #5

Thank you for your replies, I will mark this as solved.

One thing to note for anyone who might be thinking about implemented custom loss functions in the future: be careful about the numeric stability of your implementation.