config.py

#Train on simulated data (S, normal random dist.) or external data (E)
Dataset = "S" 

#Path for the dataset (binary file in NumPy .npy format) with shape
#[N_Subjects, N_Nodes, N_Nodes, N_Views]
#ignored if Dataset = S
Path =  "./simulated dataset/example.py"

#Number of simulated subjects (overwriten if Dataset = "E") 
N_Subjects = 100

#Number of nodes for simulated brain networks (overwriten if Dataset = "E") 
N_Nodes = 35

#Number of brain views (overwriten if Dataset = "E")
N_views = 4

#Number of training epochs
N_max_epochs = 100

#Apply early stopping True/False
early_stop =  True

#Random subset size for SNL function  
random_sample_size = 10

#Number of cross validation folds
n_folds = 5

#Learning Rate for Adam optimizer
lr = 0.0005

#Name of the model
model_name = "DGN_TEST"

#dimension of embeddings output by the first GDL layer (for each ROI)
CONV1 = 36
#dimension of embeddings output by the second GDL layer (for each ROI)
CONV2 = 24
#dimension of embeddings output by the third GDL layer (for each ROI)
CONV3 = 5

#----------------------------------------------------------------------------#
#----------------------------------------------------------------------------#
#                 Below is not to be modified manually                       #
#----------------------------------------------------------------------------#
#----------------------------------------------------------------------------#

import numpy as np
import helper 

if Dataset.lower() not in ["e", "E", "s", "S"]:
    raise ValueError("Dataset options are E or S.")
    
if (Dataset.lower() == "e"):
    X = np.load(Path)
    N_Subjects = X.shape[0]
    N_Nodes = X.shape[1]
    N_views = X.shape[3]
else:
    X = helper.create_better_simulated(N_Subjects, N_Nodes) if N_views == 4 else helper.simulate_dataset(N_Subjects, N_Nodes, N_views)


CONFIG = {
        "X": X,
        "N_ROIs":  X.shape[1],
        "N_views":  X.shape[3],
        "N_max_epochs": N_max_epochs,
        "n_folds": n_folds,
        "random_sample_size": random_sample_size,
        "early_stop": early_stop,
        "model_name": model_name
    }

MODEL_PARAMS = {
        "N_ROIs": N_Nodes,
        "learning_rate" : lr,
        "n_attr": X.shape[3],
        "Linear1" : {"in": N_views, "out": CONV1},
        "conv1": {"in" : 1, "out": CONV1},
        
        "Linear2" : {"in": N_views, "out": CONV1*CONV2},
        "conv2": {"in" : CONV1, "out": CONV2},
        
        "Linear3" : {"in": N_views, "out": CONV2*CONV3},
        "conv3": {"in" : CONV2, "out": CONV3} 
    }