gll

gnl + learngraph

Arguments:

• g: a grap object
• m: a constant object (generated for example by getm())
• free: add how many free parameters to each feature vector
• alin=[]: alinearity used, defined in glm
• iterations=1: run each graph update step iteration times (one layer)
• repeat=1: repeat this function repeat times (multiple learnings)
• subrepeat=1: repeat the gnl function this many times (multiple layers)
• usei=False: use inverse graph update layers
• k=4: the k in the topK algorithm