The baseline of our paper, FedDebug, for debugging malicious/faulty clients in Federated Learning is available in the Flower AI framework. Check out the code here.