Hi,
I just noticed that in recurrent nets, e.g. http://pytorch.org/docs/master/nn.html#gru there are two bias vectors. Is there a particular reason for this?
Hi,
I just noticed that in recurrent nets, e.g. http://pytorch.org/docs/master/nn.html#gru there are two bias vectors. Is there a particular reason for this?
Hello,
I think this is to match CUDNN:
Best regards
Thomas
I see. That’s interesting. Don’t understand why CuDNN design it like this.