haoheliu / AudioLDM2

Text-to-Audio/Music Generation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

pretrained_model in class FirstStagePostProcessor(nn.Module)?

yeori1 opened this issue · comments

Hello,

I want to use the pre-trained VAE encoder, so I want to use the function 'encode_with_pretained' in provided in the class 'FirstStagePostProcessor', in audioldm2/latent_diffusion/modules/diffusionmodules/model.py.
https://github.com/haoheliu/AudioLDM2/blob/main/audioldm2/latent_diffusion/modules/diffusionmodules/model.py#L1051

However, I don't know how to set the 'pretrained_model' on the class FirstStagePostProcessor.
https://github.com/haoheliu/AudioLDM2/tree/main/audioldm2/latent_diffusion/modules/diffusionmodules/model.py#L1000

Please help me!

Thank you.