search

lucasmansilla / ACRN_Chest_X-ray_IA

Learning Deformable Registration of Medical Images with Anatomical Constraints
MIT License
23 stars 9 forks source link

Which file do we run firstly? #13

Closed tphankr closed 3 years ago

tphankr commented 3 years ago

Thank you for your code, @lucasmansilla. Please guide us, which file do we run firstly? Thank you. image

Thank you.

lucasmansilla commented 3 years ago

Hi @tphankr,

If you want to train AC-RegNet, you first need to train the autoencoder. So, first runtrain_aenet.pyand then train_acregnet.py.

tphankr commented 3 years ago

Thank you @lucasmansilla, Allow me ask, how setting the data folder to run the code, thank you @lucasmansilla. I downloaded the NIH Chest-XRay14 dataset already. And put it in the same folder with the train_aenet.py or train_acregnet.py. image image image

And then, I typed "python train_aenet.py", It is like not run, as following: image

Thank you.

lucasmansilla commented 3 years ago

Hi @tphankr,

What happens is that this code is intended to work with data from JSRT, Montgomery or Shenzhen datasets, at least for now.

tphankr commented 3 years ago

Thank you @lucasmansilla, It means, now I must to download the data from JSRT, Montgomery or Shenzhen datasets, and then run. How about with "the NIH Chest-XRay14 dataset", please guide me some where to I can fix and run with the the NIH Chest-XRay14 dataset. Thank you @lucasmansilla .

lucasmansilla commented 3 years ago

Hi @tphankr,

AC-RegNet requires images and segmentations for training and the Chest-XRay14 dataset does not provide segmentations.