IBM / FedMA

Code for Federated Learning with Matched Averaging, ICLR 2020.

Repository from Github https://github.comIBM/FedMA

IBM/FedMA Issues

Running FedMA with simple cnn
Updated a year ago1
UnboundLocalError: local variable 'shape_estimator' referenced before assignment
Updated 2 years ago2
Use Cnn matching as a black box
Updated 2 years ago
I what to run this in heterogeneous data, but there are some errors
Updated 3 years ago
Question on code in language_fedma.py
Updated 3 years ago
unable to install lapsolver
Updated 3 years ago2
i want to change the hyper paramerters language model
Updated 3 years ago
Unable to run lstm_fedma_with_comm.py file
Updated 3 years ago2
Reproduce results from run_dist_skew.sh
Updated 4 years ago
Question about reproducing results in the paper (LSTM on Shakespeare dataset)
Updated 4 years ago
some questions about initialization and retrain
Updated 5 years ago
some qustions about oneshot_matching experiment
Updated 5 years ago1
Running FedMA with large input data shape
Updated 5 years ago
Provide more details about the experiments?
Updated 5 years ago15
Does hyper parameter "retrain_epoch" lead to extra training for FedMA?
Closed 5 years ago5
The problem of "local_retrain" function in the main.py
Closed 5 years ago