Shuai Tan
Shuai Tan
I run the code on a A100 GPU but get error: `module 'fused_dense_lib' has no attribute 'linear_act_forward'`
Any update?
Will you release the code for loss function L_{MI} and L_{ML}, which are the core of this paper, please?

Can someone help me deploy EDTalk on Hugging face with just an A100, but it seems to require a pro user?
您好,请问在计算fid指标出现这个问题 怎么解决呢 
` prompt_embeds = self.text_encoder( text_input_ids.to(device), attention_mask=attention_mask, )` get eroor "RuntimeError: expected scalar type Float but found Half"
Hi, great work! I test the model at huggingface space. The input is https://github.com/user-attachments/assets/310481a8-f688-4345-a385-382d143532d7 and I get the result: https://github.com/user-attachments/assets/1de0db71-8dd7-425c-9630-ed64337c2eb5 Is the result normal?