main cluster ditto fedamp fedavg fedrep fedrod knn_per __init__.py config.py utils.py perfedavg pfedgraph_approx pfedgraph_cosine pfedhn pfedme .gitignore README.md attack.py cfl.py cvdataset.py data_partition.py ditto.py fedamp.py fedavg.py fedavg_ft.py fedprox.py fedprox_ft.py ...
python train.py <dataset_name> \ --aggregator_type centralized \ --n_rounds 100 \ --bz 128 \ --lr 0.05 \ --lr_scheduler multi_step \ --log_freq 10 \ --device cuda \ --optimizer sgd \ --seed 1234 \ --logs_dir ./logs \ --chkpts_dir ./chkpts/cifar10_fedavg --verbose 1...
First, one would need to create checkpoint and results directory in ./results directory, with a subdirectory named after the baseline name (e.g., ./results/<dataset_name>/<method_name> to store results of FedAvg baselines which you would run from ./trainers/client_fedavg.py). This would ...