I'm looking into the code and find that in def update_cpc() both self.encoder_optimizer.step() and self.cpc_optimizer.step() are called. However the parameters of critic.encoder are carried by both optimizer. Isn't it true that, in def update_cpc(), critic.encoder is updated twice using the same gradient?
I'm looking into the code and find that in
def update_cpc()
bothself.encoder_optimizer.step()
andself.cpc_optimizer.step()
are called. However the parameters ofcritic.encoder
are carried by both optimizer. Isn't it true that, indef update_cpc()
,critic.encoder
is updated twice using the same gradient?