Lewis lou

Hi, the F-score calculation on the Lizard dataset takes a long time. Is this normal? The pair numbers are mostly 2000~10000.

 I assum that "x['spatial_query']['rand_shape'].to(self.device) & False)" is to get the reverse map of the input masks. So when the area of the mask is True, it should reture False,...

How to build caption_class_similarity.pth for custom dataset?

Hi, Could you release the registration codes, so that we can try your algorithm correctly on a custom dataset?

Data split for all the few-shot experiments?

6

Hi ,what's your data split for few-shot experiments? For example, in one-shot or two-shot setting, how to split training/validation set?

About the data preprocess codes

1

Hi, Thanks a lot for releasing codes for such a great work. Will you release text generation or text filter codes for convert TCGA captions into question-answer pairs? It would...

How can I segment my own slide and get the segmentation results?

5

Hi, I have tried "runPipeline --datahub breast-tumour-segmentation --file my_slide.svs". But none results are saved. Could you please provide a full command or python codes to segment a testing WSI using...