lucidrains / DALLE2-pytorch

Implementation of DALL-E 2, OpenAI's updated text-to-image synthesis neural network, in Pytorch
MIT License
11.03k stars 1.07k forks source link

Typo in VQ-VAE #45

Closed CiaoHe closed 2 years ago

CiaoHe commented 2 years ago

Hi Phil, great to see your big contributions. Here I find might be a typo. I thought it should be like last_dec_layer = self.enc_dec.decoders[-1].weight https://github.com/lucidrains/DALLE2-pytorch/blob/c1db2753f5a425220f808da87d3fe208e97cf285/dalle2_pytorch/vqgan_vae.py#L846

Besides, in order to match the name with ResEncDec and ConvNextEncDec, I think it's better to rename ViTEncDec's encoder and decoder as self.encoders and self.decoders

Enjoy!

lucidrains commented 2 years ago

@CiaoHe oh hey, long time! thanks for the code review (as usual :laughing:) and let me know if https://github.com/lucidrains/DALLE2-pytorch/releases/tag/0.0.89 fixes things

CiaoHe commented 2 years ago

great! It should be no problem. Recently I just follow stuffs related to diffusion model and other generation tracks. Hope to participate more on this replicationԅ(¯ㅂ¯ԅ)

lucidrains commented 2 years ago

great! It should be no problem. Recently I just follow stuffs related to diffusion model and other generation tracks. Hope to participate more on this replicationԅ(¯ㅂ¯ԅ)

come on aboard! :D