tf.raw_ops.PartitionedCall

Stay organized with collections Save and categorize content based on your preferences.

returns f(inputs), where f's body is placed and partitioned.

View aliases

Compat aliases for migration

tf.compat.v1.raw_ops.PartitionedCall

tf.raw_ops.PartitionedCall(
    args,
    Tout,
    f,
    config='',
    config_proto='',
    executor_type='',
    name=None
)

Asynchronously executes a function, potentially across multiple devices but within a single process. The kernel places and partitions a given function's underlying graph, and executes each of the partitioned subgraphs as a function.

Args
args	A list of Tensor objects. A list of input tensors.
Tout	A list of tf.DTypes. A list of output types.
f	A function decorated with @Defun. A function that takes 'args', a list of tensors, and returns 'output', another list of tensors. Input and output types are specified by 'Tin' and 'Tout'. The function body of f will be placed and partitioned across devices, setting this op apart from the regular Call op.
config	An optional string. Defaults to "".
config_proto	An optional string. Defaults to "".
executor_type	An optional string. Defaults to "".
name	A name for the operation (optional).

Returns
A list of Tensor objects of type Tout.

tf.raw_ops.PartitionedCall | TensorFlow v2.16.1 (original) (raw)

View aliases