tf.parallel_stack  |  TensorFlow v2.16.1 (original) (raw)

tf.parallel_stack

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

Stacks a list of rank-R tensors into one rank-(R+1) tensor in parallel.

View aliases

Compat aliases for migration

SeeMigration guide for more details.

tf.compat.v1.parallel_stack

tf.parallel_stack(
    values, name='parallel_stack'
)

Requires that the shape of inputs be known at graph construction time.

Packs the list of tensors in values into a tensor with rank one higher than each tensor in values, by packing them along the first dimension. Given a list of length N of tensors of shape (A, B, C); the outputtensor will have the shape (N, A, B, C).

For example:

x = tf.constant([1, 4])
y = tf.constant([2, 5])
z = tf.constant([3, 6])
tf.parallel_stack([x, y, z])  # [[1, 4], [2, 5], [3, 6]]

The difference between stack and parallel_stack is that stack requires all the inputs be computed before the operation will begin but doesn't require that the input shapes be known during graph construction.

parallel_stack will copy pieces of the input into the output as they become available, in some situations this can provide a performance benefit.

Unlike stack, parallel_stack does NOT support backpropagation.

This is the opposite of unstack. The numpy equivalent is

tf.parallel_stack([x, y, z]) = np.asarray([x, y, z])

eager compatibility

parallel_stack is not compatible with eager execution.