Evaluation metric issues

[Han2001JL]

Hello, I have recently been conducting experiments on federated unlearning. May I use the model inversion attack evaluation code that you used in your experiments? If possible, I would be very grateful!