Closed xwybot closed 10 months ago
Hi, thank you for your interest in our research.
tgt_mask,memory_mask,tgt_key_padding_mask is dummy variable which is 'None' for transformer decoder. So, it doesn't matter if you delete those.
Because images in the batch have different image sizes, we add zero-padding to the images to have the same size. memory_key_padding_mask are generated in dataloader to prevent attending zero-padding area.
Please ask muren.py multiplex_context = self. MURE(output_human,output_obj,output_rel,(memory,tgt_mask,memory_mask,tgt_key_padding_mask,memory_key_padding_mask,pos)), tgt_mask,memory_mask,tgt_key_padding_mask, Where are the memory_key_padding_mask respectively