Tony Yang

Hi! Would you like to release the code for reproducing results in Figure 3 and the Appendix? Especially those using convolutional neural networks Thanks!

Is this project based on Python 3.7?

1

It seems this project is based on Python 3.x It's misleading on README -- “It is tested under Ubuntu Linux 16.04.1 and Python 2.7 environment”

Hello, The dataset used in OOD training has been withdrawed officially and deleted permanently. see [`https://groups.csail.mit.edu/vision/TinyImages/`](https://groups.csail.mit.edu/vision/TinyImages/) Will you provide some alternative datasets for 80 Million Tiny Images ? Thank you!

confidence_pgd_attack.py line 92~102 ` # one_hot_labels = torch.eye(len(outputs[0]))[y].to(CUDA_DEVICE) # other, _ = torch.max((1-one_hot_labels)*outputs, dim=1) # correct = torch.masked_select(outputs, one_hot_labels.byte()) # loss = torch.clamp(other - correct, min=-50.0) # if self.in_distribution: #...

The loss function doesn't match the one mentioned in Sec 3 of paper ?

1

Hello ,I'm very interested in your work and trying to reproduce your results. Q1: I found the loss function in (line 200,main_NL.py) ` ( (loss+loss_neg) / (float((labels>=0).sum())+float((labels_neg[:,0]>=0).sum())) ).backward() ` It...