tff.backends.mapreduce.get_state_initialization_computation

Validates and transforms a computation to generate state.

tff.backends.mapreduce.get_state_initialization_computation(
    initialize_computation: tff.framework.ConcreteComputation,
    grappler_config: tf.compat.v1.ConfigProto = _GRAPPLER_DEFAULT_CONFIG
) -> tff.Computation

Args
`initialize_computation`	A `computation_impl.ConcreteComputation` that should generate initial state for a computation that is compatible with a federated learning system that implements the contract of a backend defined in the backends/mapreduce directory.
`grappler_config`	An optional instance of `tf.compat.v1.ConfigProto` to configure Grappler graph optimization of the TensorFlow graphs. These options are combined with a set of defaults that aggressively configure Grappler. If the input `grappler_config` has `graph_options.rewrite_options.disable_meta_optimizer=True`, Grappler is bypassed.

Returns
A `computation_base.Computation` that can generate state for a computation.

Raises
`TypeError`	If the arguments are of the wrong types.

tff.backends.mapreduce.get_state_initialization_computation

Args

Returns

Raises