M3AE
M3AE copied to clipboard
[MICCAI-2022] This is the official implementation of Multi-Modal Masked Autoencoders for Medical Vision-and-Language Pre-Training.
When I run run_scripts/finetune_m3ae.sh, I get the error "TypeError: compute_vqa() missing 3 required positional arguments: 'outputs', 'loss', and 'labels'", because compute_vqa is defined to take 6 arguments in objectives.py but...
 After the progress bar is loaded, the next step cannot be carried out, and the program ends automatically after a long wait. How can this situation be solved?