miccunifi / ladi-vton

[ACM MM 2023] - LaDI-VTON: Latent Diffusion Textual-Inversion Enhanced Virtual Try-On
Other
412 stars 56 forks source link

Outdated arguments in function #23

Closed mgp123 closed 1 year ago

mgp123 commented 1 year ago

The Encoder uses the get_down_block here using parameter attn_num_head_channels but get_down_block doesn't have such parameter in newer versions of the diffusers library

ABaldrati commented 1 year ago

Hi @mgp123

As pointed out in the README our code is supposed to work with version 0.14 of diffusers.

Alberto