tf.math.pow

View source on GitHub

Computes the power of one value to another.

View aliases

Main aliases

tf.pow

Compat aliases for migration

See Migration guide for more details.

tf.compat.v1.math.pow, tf.compat.v1.pow

tf.math.pow(
    x, y, name=None
)

Given a tensor x and a tensor y, this operation computes \(x^y\) for corresponding elements in x and y. For example:

x = tf.constant([[2, 2], [3, 3]])
y = tf.constant([[8, 16], [2, 3]])
tf.pow(x, y)  # [[256, 65536], [9, 27]]

Args
x	A Tensor of type float16, float32, float64, int32, int64, complex64, or complex128.
y	A Tensor of type float16, float32, float64, int32, int64, complex64, or complex128.
name	A name for the operation (optional).

Returns
A Tensor.