xuedue

Thank you for your reply, I modified it according to your description, but it brought another error.   If I use AffineAdapterNaive instead of AffineAdapterSigmoid, it works.

> @xuedue Hi, thanks for using MemCNN. Whereas the `memcnn.AdditiveCoupling` expects `Fm` and `Gm` to have a single input x and a single output y of the same shape, `memcnn.AffineCoupling`...

> > Fm and Gm need to have input x and output y of the same shape. If I want to implement a reversible MLP with different input and output...

> > If I change the output to 100 dimensions and only take two of them，It doesn't make sense. > > Why? Could you elaborate? Doesn't this work for your...

> Ok, thanks for clarifying your question. First, I would suggest making layers 1-6 invertible. > This should be simple (the `in_features`/`out_features` ratio is 1:1, which is what memcnn supports...

> > Ok, thanks for clarifying your question. First, I would suggest making layers 1-6 invertible. > > This should be simple (the `in_features`/`out_features` ratio is 1:1, which is what...

我用的代理是clash for windows，我更改了代理地址  但是仍然出现错误  

> Hello author, sorry to bother you again, it seems that more code needs to be changed in the training process, I would like to ask about the order of...

> Hello author, I still have questions about this order. In inference.py I set ganpath to agsmileglasses.pt, set ckpt to pre-trained original e4e without finetune. Then edited with conditional_pti.py, the...

> Sorry for the confusion, you are right. Thank you for your patience. Since StyleCLIP is also one of the methods we compare, I would like to ask you which...