Closed nguyenpbui closed 1 year ago
Please modify the code in H2Former.py as:
The current code is dealing with 3D data with four modalities
Thank you for the quick response. Just a follow-up question regarding the train-val-test split. I cannot find this information in your paper. Can you clarify this?
We follow the official testing and training split.
Dear authors,
Firstly, thank you for your great work. I am currently re-produce the results on the Skin dataset using your code. However, a problem related to the first patch_embed block occurred as shown in the figure.
Do you know the reason? Thank you in advance.