CrossmodalGroup/DynamicVectorQuantization

infrerence size is not match

Winnie202 opened this issue · 0 comments

Inference with a 512×512 size picture resulted in the following error where I changed the latent_size = 64 because I couldn't reason with a 512×512 size picture when latent_size=32

2023-06-15 19-22-58屏幕截图

What should I do because I need to reconstruct the different sizes picture in the first stage