|View source on GitHub|
Represents the state of the server carried between rounds.
tff.learning.framework.ServerState( model, optimizer_state, delta_aggregate_state, model_broadcast_state )
ModelWeightsstructure, containing Tensors or Variables.
optimizer_state: A list of Tensors or Variables, in the order returned by
delta_aggregate_state: State (possibly empty) of the delta_aggregate_fn.
model_broadcast_state: State (possibly empty) of the model_broadcast_fn.