|View source on GitHub|
Represents a client computation that produces an update to a model.
Returns all the variables of this object.
Note this only includes variables that are part of the state of this object, and not the model variables themselves.
An iterable of
__call__( dataset, initial_weights )
Defines the complete client computation.
Typically implementations should be decorated with
tf.data.Datasetproducing batches than can be fed to
initial_weights: A dictionary of initial values for all trainable and non-trainable model variables, keyed by name. This will be supplied by the server in Federated Averaging.