dpctl.tensor.diff

dpctl.tensor.diff(x, /, *, axis=-1, n=1, prepend=None, append=None)[source]

Calculates the n-th discrete forward difference of x along axis.

Parameters:
  • x (usm_ndarray) – input array.

  • axis (int) – axis along which to compute the difference. A valid axis must be on the interval [-N, N), where N is the rank (number of dimensions) of x. Default: -1

  • n (int) – number of times to recursively compute the difference. Default: 1.

  • prepend (Union[usm_ndarray, bool, int, float, complex]) – value or values to prepend to the specified axis before taking the difference. Must have the same shape as x except along axis, which can have any shape. Default: None.

  • append (Union[usm_ndarray, bool, int, float, complex]) – value or values to append to the specified axis before taking the difference. Must have the same shape as x except along axis, which can have any shape. Default: None.

Returns:

an array containing the n-th differences. The array will have the same shape as x, except along axis, which will have shape:

prepend.shape[axis] + x.shape[axis] + append.shape[axis] - n

The data type of the returned array is determined by the Type Promotion Rules.

Return type:

usm_ndarray