espnet2.enh.layers.dcunet.torch_complex_from_reim

source

espnet2.enh.layers.dcunet.torch_complex_from_reim(re, im)

Create a complex tensor from real and imaginary parts.

This function takes two tensors representing the real and imaginary components of a complex number and combines them into a single complex tensor using PyTorch’s built-in complex number support.

• Parameters:
  • re (Tensor) – A tensor representing the real part of the complex number.
  • im (Tensor) – A tensor representing the imaginary part of the complex number.
• Returns: A complex tensor formed by combining the real and imaginary parts.
• Return type: Tensor

Examples

>>> re = torch.tensor([1.0, 2.0])
>>> im = torch.tensor([3.0, 4.0])
>>> complex_tensor = torch_complex_from_reim(re, im)
>>> print(complex_tensor)
tensor([1. + 3.j, 2. + 4.j])

NOTE

The input tensors must have the same shape. If they do not, a runtime error will occur.

• Raises:ValueError – If the input tensors do not have the same shape.