tf.keras.ops.diff

Calculate the n-th discrete difference along the given axis.

Main aliases

tf.keras.ops.numpy.diff

 
tf.keras.ops.diff(
    a, n=1, axis=-1
)

The first difference is given by out[i] = a[i+1] - a[i] along the given axis, higher differences are calculated by using diff recursively.

a Input tensor.
n The number of times values are differenced. Defaults to 1.
axis Axis to compute discrete difference(s) along. Defaults to -1.(last axis).

Tensor of diagonals.

Examples:

from keras.src import ops
x = ops.convert_to_tensor([1, 2, 4, 7, 0])
ops.diff(x)
array([ 1,  2,  3, -7])
ops.diff(x, n=2)
array([  1,   1, -10])
 
x = ops.convert_to_tensor([[1, 3, 6, 10], [0, 5, 6, 8]])
ops.diff(x)
array([[2, 3, 4],
       [5, 1, 2]])
ops.diff(x, axis=0)
array([[-1,  2,  0, -2]])