Open
Description
Hello,
I noticed that torch.nn.Hardtanh is used for activation functions in BinaryNet. This is meant to make the model trainable, as introduced in the BNN paper. However, in the inference phase (the validate() function in main_binary.py), shoudn't the activation function be changed to sign function so that the intermediate results are binary?
Thanks!
Metadata
Metadata
Assignees
Labels
No labels