HKU-MedAI/MGCA

About the pretrained model of text encoder

o0t1ng0o opened this issue · 2 comments

o0t1ng0o commented

Hi @fuying-wang,

I found that you share the pretrained model of image encoder, such as resnet-50 and vit-base.
Could you please share the pretrained model for text encoder?
Thank you in advance!
image

fuying-wang commented

Hello,

These are the pretrained weights of the whole model. Could you try load it with
model = MGCA.load_from_checkpoint(args.path, strict=False)?
The name denotes the visual encoder it used.

o0t1ng0o commented

Thank you very much!
I load it successfully.