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

tf.shape_n

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

Returns shape of a list of tensors.

View aliases

Compat aliases for migration

SeeMigration guide for more details.

tf.compat.v1.shape_n

tf.shape_n(
    input,
    out_type=tf.dtypes.int32,
    name=None
)

Given a list of tensors, tf.shape_n is much faster than applying tf.shapeto each tensor individually.

>>> a = tf.ones([1, 2])
>>> b = tf.ones([2, 3])
>>> c = tf.ones([3, 4])
>>> tf.shape_n([a, b, c])
[<tf.Tensor: shape=(2,), dtype=int32, numpy=array([1, 2], dtype=int32)>,
<tf.Tensor: shape=(2,), dtype=int32, numpy=array([2, 3], dtype=int32)>,
<tf.Tensor: shape=(2,), dtype=int32, numpy=array([3, 4], dtype=int32)>]