Scientific Reports

Table 7 Performance of Personalized Federated Models (FedPer-B0 vs. FedPer-B2).

From: A personalized communication efficient federated learning framework with low rank adaptation for intelligent leukemia diagnosis

Client	FedPer-B0				FedPer-B2
Client	Accuracy	F1-Score	Kappa	MCC	Accuracy	F1-Score	Kappa	MCC
Client 0	0.7812	0.7498	0.3584	0.3663	0.9861	0.9862	0.5925	0.5944
Client 1	0.7643	0.7223	0.2683	0.2846	0.9877	0.9877	0.6669	0.6781
Client 2	0.7797	0.7561	0.1889	0.1968	0.9892	0.9892	0.8706	0.8739
Client 3	0.7720	0.7385	0.4208	0.4300	0.9892	0.9892	0.6260	0.6488
Client 4	0.7720	0.7344	0.6533	0.6603	0.9861	0.9861	0.7158	0.7297
Client 5	0.7843	0.7517	0.0578	0.0651	0.9846	0.9846	0.3950	0.4048
Client 6	0.8105	0.7914	0.7549	0.7590	0.9877	0.9877	0.8459	0.8466
Client 7	0.7904	0.7664	0.3158	0.3543	0.9846	0.9846	0.7257	0.7354
Client 8	0.7874	0.7591	0.6257	0.6322	0.9861	0.9861	0.7405	0.7425
Client 9	0.8120	0.7908	0.3506	0.3820	0.9861	0.9862	0.7475	0.7496
Average	0.7854	0.7560	0.3997	0.4139	0.9867	0.9868	0.6926	0.7004

Back to article page

Search

Advanced search

Quick links