idct#

diffsptk.IDCT#: alias of InverseDiscreteCosineTransform

class diffsptk.InverseDiscreteCosineTransform(dct_length: int, dct_type: int = 2, device: device | None = None, dtype: dtype | None = None)[source]#

See this page for details.

Parameters:

dct_lengthint >= 1: The DCT length, \(L\).
dct_typeint in [1, 4]: The DCT type.
devicetorch.device or None: The device of this module.
dtypetorch.dtype or None: The data type of this module.

forward(y: Tensor) → Tensor[source]#

Apply inverse DCT to the input.

Parameters:

yTensor [shape=(…, L)]: The input.

Returns:

outTensor [shape=(…, L)]: The inverse DCT output.

Examples

>>> x = diffsptk.ramp(3)
>>> dct = diffsptk.DCT(4)
>>> idct = diffsptk.IDCT(4)
>>> x2 = idct(dct(x))
>>> x2
tensor([-4.4703e-08,  1.0000e+00,  2.0000e+00,  3.0000e+00])

diffsptk.functional.idct(y: Tensor, dct_type: int = 2) → Tensor[source]#

Compute inverse DCT.

Parameters:

yTensor [shape=(…, L)]: The input.
dct_typeint in [1, 4]: The DCT type.

Returns:

outTensor [shape=(…, L)]: The inverse DCT output.