koriavinash1 / brats2018

Brats segmentation pytorch

Hi there, I'm Avinash 👋

• 🔭 I’m currently working on Interpretability and Causality research.
• 🌱 I love to code and contribute to Open Source.
• 💬 Ask me anything regarding my work, code and research here.
• 📫 How to reach me: [email protected]

Connect with me:

Languages and Tools:

brats2018's Issues

How to train the model?

There are a lot of files in your folder, how should I use it? Is there an order of use?

Multipath3D

checkpoint= '/media/brats/0d4a2225-d6b1-4b80-94fd-7c8ae0b1fa102/varghese/brats_2018/zscore_src/models_zscore_lesion_only_3_classes_103/CE_best_model_loss_based.pth.tar'
Hello, what is this? I did not find the training pipeline in the paper and code.

Mask.nii

Sir,
Where did you get the mask.nii.gz from? It isn't included either in the training nor the validation dataset.

Tiramisu_2D and Tiramisu_3D

Hello, in the model.py in the Tiramisu_2D and Tiramisu_3D folders, I think the model is the same, there is no distinction between 2D and 3D.

Tiramisu_2D

In csv_gen.py in this folder, LGG_patches and HGG_patches are the same as the data in HGG_64_cubes_high_sampling and LGG_64_cubes_high_sampling in Tiramisu_3D?

Data folder problem

What is the structure of the Brats2018 data set?