The equivalent code of multi-layer MLP and One-layer Linear in class Encoder

wyhsirius / LIA

[ICLR 22] Latent Image Animator: Learning to Animate Images via Latent Space Navigation

https://wyhsirius.github.io/LIA-project/

Other

578 stars 64 forks source link

The equivalent code of multi-layer MLP and One-layer Linear in class Encoder #11

Open junleen opened 1 year ago

junleen commented 1 year ago

Hi, I found the activation in this sub-network is set as None. If it was None, the self.fc should equal to the cascaded multiple linear layer without activation, therefore equal to one Linear Layer? Is it right?

TaekyungKi commented 1 year ago

same question

ivan-alles commented 1 year ago

I agree.