Format of training data #5
Description
Could you share some information about how the training data is prepared for the model? The main GitHub page suggests the input data is in .vtk format but the train.py script seems to be reading in .mat files for the dataset. I'm trying to use this network architecture but trained on a different dataset for another purpose. I'm assuming the training data is some kind of list input information about the vertex along with a label. Is there a script for converting the vtk data to .mat for training that I'm missing? Thanks