PyTorch torch.adjoint Function
PyTorch torch Reference Manual
torch.adjointIt is a function in PyTorch used to compute the conjugate transpose (adjoint) of a matrix. For real matrices, it is equivalent to transpose; for complex matrices, it returns the transpose and takes the conjugate of each element.
Function Definition
torch.adjoint(input)
Parameters:
input(Tensor): Input tensor.
Return Value:
torch.Tensor: Returns the conjugate transpose of the input tensor.
Usage Example
Example - Real Matrix
import torch
# Create a real matrix
A = torch.tensor([[1, 2, 3], [4, 5, 6]])
# Compute the conjugate transpose
result = torch.adjoint(A)
print("Original matrix:")
print(A)
print("Shape:", A.shape)
print("Conjugate transpose:")
print(result)
print("Shape:", result.shape)
# Create a real matrix
A = torch.tensor([[1, 2, 3], [4, 5, 6]])
# Compute the conjugate transpose
result = torch.adjoint(A)
print("Original matrix:")
print(A)
print("Shape:", A.shape)
print("Conjugate transpose:")
print(result)
print("Shape:", result.shape)
The output result is:
原矩阵:
tensor([[1, 2, 3],
[4, 5, 6]])
形状: torch.Size([2, 3])
共轭转置:
tensor([[1, 4],
[2, 5],
[3, 6]])
形状: torch.Size([3, 2])
Example - Complex Matrix
import torch
# Create a complex matrix
A = torch.tensor([[1+1j, 2+2j], [3+3j, 4+4j]])
# Compute the conjugate transpose
result = torch.adjoint(A)
print("Original matrix:")
print(A)
print("Conjugate transpose:")
print(result)
# Create a complex matrix
A = torch.tensor([[1+1j, 2+2j], [3+3j, 4+4j]])
# Compute the conjugate transpose
result = torch.adjoint(A)
print("Original matrix:")
print(A)
print("Conjugate transpose:")
print(result)
Other Extensions