@yaojieliu I was able to train on your code, but while saving checkpoints to saved_model format signature definition i am unable to figure out.
Here what i understood :
1) According to paper VQ network is the final layer which determines image spoof
2) Out is a mean type of tensor
But i am unable to figure out according to the code how to configure signature definition please do help in this regard have already mailed regarding the same. Thanks in advance.
@yaojieliu I was able to train on your code, but while saving checkpoints to saved_model format signature definition i am unable to figure out. Here what i understood : 1) According to paper VQ network is the final layer which determines image spoof 2) Out is a mean type of tensor
But i am unable to figure out according to the code how to configure signature definition please do help in this regard have already mailed regarding the same. Thanks in advance.