lucidrains / meshgpt-pytorch

Implementation of MeshGPT, SOTA Mesh generation using Attention, in Pytorch
MIT License
714 stars 58 forks source link

Typo Error: `dim` Parameter in `MeshAutoencoder` #33

Closed qixuema closed 9 months ago

qixuema commented 9 months ago

Hi Phil,

I've noticed a potential typo error. A few days ago, you removed the dim parameter in the MeshAutoencoder, but I still see it being used here. Is this a typo error?

for _ in range(local_attn_encoder_depth):
    self.encoder_local_attn_blocks.append(nn.ModuleList([
        LocalMHA(dim = dim, **attn_kwargs, **local_attn_kwargs),
        nn.Sequential(RMSNorm(curr_dim), FeedForward(curr_dim, glu = True, dropout = ff_dropout))
    ]))

Best regards, Xueqi

lucidrains commented 9 months ago

@qixuema yes, my bad, thank you