tf.keras.backend.std

Standard deviation of a tensor, alongside the specified axis.

Compat aliases for migration

tf.compat.v1.keras.backend.std, tf.compat.v2.keras.backend.std

tf.keras.backend.std(
    x, axis=None, keepdims=False
)

x: A tensor or variable.
axis: An integer, the axis to compute the standard deviation.
keepdims: A boolean, whether to keep the dimensions or not. If keepdims is False, the rank of the tensor is reduced by 1. If keepdims is True, the reduced dimension is retained with length 1.

A tensor with the standard deviation of elements of x.