The project is written using Keras with a Tensorflow backend. Therefore, an Cuda enabled Nvidia GPU must be available. Jupyter Notebook is used for the example code, and the top block includes the list of required python 3 libraries.
Training and test data are created in the data_loader.py file. Data is currently not provided due to the size of the files, however interested users can request the data from the authors.
For an example of the training and testing process, please see HoloUNet.ipynb.