gcomdensediverge¶
Inputs a 3d Vector of dimensions (-1,gs,param) and uses a simple dense Layer to transform it into (-1,gs,c,paramo) without breaking Graph Permutation Symmetry. So transforms each Feature Vector into a set of c Feature Vectors.
Arguments
- gs: The Number of Nodes of the Graph (Graph Size)
- param: The Number of Features for each node
- paramo: The Number of Features each Outputvector should have
- c=2: The Number of Outputvectors for each Inputfeaturevector
- initializer=glorot_uniform: THe Transformation initializer
- learnabel=True: Shall the Transformation be learnable