PyTorch torch.linalg.matrix_power Function


Pytorch torch 参考手册Pytorch torch Reference Manual

torch.linalg.matrix_powerIt is a function in the PyTorch linear algebra module used to compute matrix powers. It computes the n-th power of a matrix, A^n.

Function Definition

torch.linalg.matrix_power(A, n, out=None)

Parameters:

  • A(Tensor): Input square matrix.
  • n(int): The power, which can be an integer, zero, or negative number.
  • out(Tensor, optional): Output tensor.

Return Value:

  • torch.Tensor: Returns the n-th power of the matrix.

Usage Examples

Example

import torch

# Create a matrix
A = torch.tensor([[1.0, 2.0],
                  [3.0, 4.0]])

# Compute A to the power of 2
A_power_2 = torch.linalg.matrix_power(A, 2)

print("Matrix A:")
print(A)
print("nA^2:")
print(A_power_2)
print("n verification: A @ A =")
print(A @ A)

The output result is:

矩阵 A:
tensor([[1., 2.],
        [3., 4.]])
A^2:
tensor([[ 7., 10.],
        [15., 22.]])
验证: A @ A =
tensor([[ 7., 10.],
        [15., 22.]])

Example - A^0 and A^-1

import torch

A = torch.tensor([[1.0, 2.0],
                  [3.0, 4.0]], dtype=torch.float64)

# A^0 = I (identity matrix)
print("A^0:")
print(torch.linalg.matrix_power(A, 0))

# A^-1 = the inverse matrix of A
print("nA^-1:")
print(torch.linalg.matrix_power(A, -1))

Pytorch torch 参考手册Pytorch torch Reference Manual

Other Extensions