/GB500

Primary LanguageJupyter NotebookMIT LicenseMIT

GB500

How To Run

Data

Get the data from http://medicaldecathlon.com/ and put the files inside one folder called data_heart, the folder structure must be: - data_heart - imagesTr - imagesVal - labelsTr - labelsVal

  • images are the volumes and labels are the groud truths
  • Tr means Training, you should put the nifti volumes for training inside this folder
  • Val means Testing, you shoul put the nifti volumes for testin inside this folder

Files

  • The file directories follow:
    • MainBackbone.ipynb -> Run the Neural Network
    • Load Results.ipynb -> Load the test dataset with trained weights
    • DivergentNets.ipynb -> Load weights from the base folders and make the average results (DivergentNets).
    • Main.ipynb -> Test Script, doesn't have any valuable code.