Hello brother zaixi, can you share the inversion attack graph in the black box and the attack code in the label-only scenario?

[huineng1]

The paper 'Model Inversion Attacks against Graph Neural Networks', I saw that this paper proposed zero-order gradient estimation in a black box scenario and a graph model inversion attack based on reinforcement learning, and I would like to make further progress based on the work of zaixi brother After research, I hope I can share the code below these two scenarios. Thank you, Brother Zaixi!