mindspore.ops.HammingWindow

class mindspore.ops.HammingWindow(periodic=True, alpha=0.54, beta=0.46, dtype=mstype.float32)[source]

Computes the hamming window function with input window length.

\[w[n] = \alpha - \beta\ \cos \left( \frac{2 \pi n}{N - 1} \right),\]

where \(N\) is the full window size.

Parameters

  • periodic (bool, optional) –

    a flag determines whether the returned window trims off the last duplicate value from the symmetric window. Default: True.

    • If True, returns a window to be used as periodic function, in above formula, \(N = \text{length} + 1\).

    • If False, return a symmetric window, \(N = \text{length}\).

  • alpha (float, optional) – The coefficient \(\alpha\) in the equation above. Default: 0.54.

  • beta (float, optional) – The coefficient \(\beta\) in the equation above. Default: 0.46.

  • dtype (mindspore.dtype, optional) – An optional data type of mindspore.dtype.float16, mindspore.dtype.float32 and mindspore.dtype.float64. Default: mindspore.dtype.float32.

Inputs:

  • length (Tensor) - a positive integer tensor controlling the returned window size, must be 1D.

Outputs:

Tensor, A 1-D tensor containing the window, whose shape is \(\text{length}\).

Raises

  • TypeError – If length is not a Tensor.

  • TypeError – If dtype of length is not integer data type.

  • TypeError – If periodic is not a bool.

  • TypeError – If alpha is not a float.

  • TypeError – If beta is not a float.

  • TypeError – If dtype is not mindspore.float16, mindspore.float32 or mindspore.float64.

  • ValueError – If dimension of length is not 1.

  • ValueError – If data of length is negative.

Supported Platforms:

GPU CPU

Examples

>>> # case 1: periodic=True.
>>> length = Tensor(np.array([6]).astype(np.int32))
>>> hamming_window = ops.HammingWindow(periodic=True)
>>> y = hamming_window(length)
>>> print(y)
[0.08000001 0.31       0.77000004 1.         0.77000004 0.31      ]
>>> # case 2: periodic=False.
>>> length = Tensor(np.array([7]).astype(np.int32))
>>> hamming_window = ops.HammingWindow(periodic=False)
>>> y = hamming_window(length)
>>> print(y)
[0.08000001 0.31       0.77000004 1.         0.77000004 0.31       0.08000001]