search

tommy-qichang / AsynDGAN

AsynDGAN project source code.
BSD 2-Clause "Simplified" License
41 stars 15 forks source link

The data processing #5

Closed shikangouyang closed 3 years ago

shikangouyang commented 3 years ago

Can you give me the data format of BRATS2018 divided into 10 categories? I wanted to run your code, but unfortunately it failed. If I know the format of your processed data, I can test my own data. Looking forward to your reply, thank you.

tommy-qichang commented 3 years ago

Hi @shikangouyang , You could refer to the data/brats_dataset.py for each of the subset, and data/brats_split_dataset.py for combining all(10) subsets together. Each subset has a hdf5 file with format: BraTS18_tumorsize{idx}.h5 The content in each h5 is as followed: data stored in path: "train/{keys}/t2" label stored in path "train/{keys}/seg"

For more details please refer to the enclosed building script.

brats_dataset_build.py.txt

shikangouyang commented 3 years ago

I did it，thank you。I have six subsets. Is netD set to 6? image

tommy-qichang commented 3 years ago

That's right.

shikangouyang commented 3 years ago

Is there anything else that needs to be modified? I see a lot of commented code in asyndgan_model.py.

tommy-qichang commented 3 years ago

Yes, the current version need to manually modify the discriminator number in asyndgan_model.py, and brats_split_dataset.py

shikangouyang commented 3 years ago

I am very interested in your work, thank you for your guidance. I will continue to work hard. thanks

tommy-qichang commented 3 years ago

Glad to hear that. Let me know if you have any other questions. Good luck.

Qi