leeh43

Results 28 comments of leeh43

Hi, thank you for your interest towards our work. It is a great question, as our initial thought does not include converting the model into a Onnx model. I can...

Hi, thank you your interest to our work. The fifth block is in the network_backbone.py and please take a look of the screenshot as follows: ![image](https://github.com/MASILab/3DUX-Net/assets/54121206/ad313b93-dc21-4125-9668-c385f5153778) ![image](https://github.com/MASILab/3DUX-Net/assets/54121206/ee3bfadb-f05d-4a30-a86e-d4bf0be0f41c)

Hi, thank you for your interest towards our work! What is your batch size, cache rate for training? I believe the preprocessing for FLARE dataset have been downsampled to 1.0x1.0x1.2....

Hi, do you mind to print our which sample it is and look into the corresponding sample? Seems like your data contains all pixels/voxels as background.

Are you using the monai random crop function or the one that the current GitHub is using (Label crop)?

Sorry for the late reply. If the label is so small, I believe it still can be cropped. But the above warning is telling us that there is no foreground...

Thank you for your interest towards our work. Sorry for the confusion in the dataset, we only use 361 samples with ground truth label to perform 5-fold cross validation. Please...

Hi, thank you for your interest towards our idea. I am sorry for the inconvenience about the install.md and the finetune.md. Currently, we are organizing the code and the installation...

Both the installation and finetuning tutorials are uploaded and please take a look of the finetuning code main_finetune.py for AMOS 2022. You are also can use to train with AMOS...

For the training/validation/testing split, we have split the training set into 8:1:1, which are 20 cases for validations. We haven't used the official validation set because of lacking of ground-truth...