Softmax with _HARD_SIGMA
ktrojanowski-tcl opened this issue · 2 comments
ktrojanowski-tcl commented
In quantizer.py (line 82), there are two softmax usages over distances. I wonder if there is any special meaning to phi_hard softmax.
I would say that argmax could have been done directly on distances, but I might be missing something here.
fab-jul commented
That's true! I discovered this a bit late and didn't want to take it out due to not being 100% sure that it would not slightly alter the training dynamics :)
ktrojanowski-tcl commented
Thank you for quick answer :)