Appendices for the Paper pFL Bench A Comprehensive Benchmark for Personalized Federated Learning

Sec.A: the details of adopted datasets and models ( e.g., tasks, heterogeneous partitions, and FL processes as shown in Figure 4 in the main body of the paper. The CIF AR10 is a popular dataset for 10-class image classification containing 60,000 colored images with resolution of 32x32 pixels. The train/valid/test sets are with a ratio of about 3:1:5. The train/valid/test sets are with ratio about 14:3:3. For all these experimental datasets, we randomly select 20% clients as new clients that do not participate in the FL processes.