Closed 3neutronstar closed 2 years ago
Hello! Please note that
So the problem is that gradient scales between gradients from CE (saliency calculation with clean data) and gradients from BCE (mixup data) is different. We tried to rebalance gradient scales so that we can use an identical hyperparameter (_cleanlam) regardless of the setting.
Hope this helps!
Thank you for the fast answer!
Hi, thank you for the interesting work and its codes.
I have a question about the loss calculation in the code.
In the
imagenet
directory, the loss for obtaining saliency map is calculated as follows:However, in the case of others ('tiny-imagenet' and 'cifar' code),
the loss for obtaining saliency map is calculated as follows:
Is there any special reason to calculate the loss for generating a saliency map in a different way?