search

Beckschen / 3D-TransUNet

This is the official repository for the paper "3D TransUNet: Advancing Medical Image Segmentation through Vision Transformers"
Apache License 2.0
182 stars 10 forks source link

Training Data Issue #2

Open gita0326 opened 10 months ago

gita0326 commented 10 months ago

Hi there,

I would like to inquire whether it is sufficient to organize the training data according to the path structure specified by nnUNet's requirements in order to store the training data.

Thank you!

Beckschen commented 10 months ago

Thanks for your interests! Yes it is sufficient to follow the path structure defined by nnUNet-v1.

In the train.sh script, we have export the environmental variables.

export nnUNet_raw_data_base="/data1/data/nnUNet_raw_data_base/" # replace to your database export nnUNet_preprocessed="/data1/data/nnUNet_raw_data_base/nnUNet_preprocessed"

The nnUNet_raw_data_base and nnUNet_preprocessed refer the path from nnUNet data path.

I will get back soon to provide some data examples for your reference.

Best Jieneng