search

Stability-AI / stablediffusion

High-Resolution Image Synthesis with Latent Diffusion Models
MIT License
38.83k stars 5.01k forks source link

Found A Code Typo: Line 318 in ldm/modules/attention.py #237

Open MenghanXia opened 1 year ago

MenghanXia commented 1 year ago

There seems to be a typo at line 318 of attention.py It should be "self.proj_out = zero_module(nn.Linear(inner_dim, in_channels))" instead of "self.proj_out = zero_module(nn.Linear(in_channels, inner_dim))".

Fortunately, it makes no actual problems since "in_channels" is always equal to "inner_dim" in current implementation. But it is a coding error indeed.

ljk628 commented 1 year ago

Exactly, this line needs to be fixed as described above. It happens to work when model_channels can be divided by num_head_channels but fails for other cases, e.g., model_channels can be 320 or 384 but cannot be 416 when num_head_channels is 64.