TransUNet

TransUNet copied to clipboard

is there some thing wrong?

4

I am looking at your code and found some problems https://github.com/Beckschen/TransUNet/blob/main/networks/vit_seg_modeling.py#L133 patch_size = (img_size[0] // 16 // grid_size[0], img_size[1] // 16 // grid_size[1]) The training image size is 224 then...

lvzhengyi0204

Thank you for your excellent work. I am working on 2d image segmentation. I have some doubt about how to resize the images and masks. Should I simply use bilinear...

lightersss

Evaluation metric

4

Hi, Thanks a lot for sharing your code. I have a question regarding the computation of the evaluation metrics. Your code is: ``` def calculate_metric_percase(pred, gt): pred[pred > 0] =...

Mordokkai

Thanks for this terrific repository! Your results are impressive! I notice that the Synapse data is not normalized to between 0 and 1 but instead I'm getting the following values...

anoukstein

Have you tried to train the model without pretraining on imagenet?

1

licaizi

Question about how to train with image size with different length and width

2

Very impressive works. I trying to use the transUnet in my dataset. But my dataset shape is (672,448) which length and width not equal. I find this will lead bugs....

xiaoerlaigeid

something about custom dataset

5

Thanks for your excellent work. While I try to make my own dataset, I meet some troubles. My first question is that what the dimensions of "image" and "label" in...

David-zaiwang

The model does not scale on multiple GPU units on a single CPU. It create multiple copies of model (when using distributed data parallel) and add all the load to...

sagarpgs

Question about "patch_size"

2

Thanks for your work. I have some questions about the patch size of patch embedding when using CNN and Transformer as the encoder. In the section 3.2 of the paper,...

benzyz1129

 What kind of file is in the red place? I didn’t understand it. The error is also because of the path problem.

long123524