The square root method solves systems of linear equations

import numpy as np


def pingfagenfa(a):
    n = a.shape[0]
    g = np.mat(np.zeros((n, n), dtype=float))
    g[0, 0] = np.sqrt(a[0, 0])
    g[1:, 0] = a[1:, 0]/g[0, 0]
    for j in range(1, n-1):
        g[j, j] = np.sqrt((a[j, j] - np.sum(np.multiply(g[j, 0:j], g[j, 0:j]))))
        for i in range(j+1, n):
            g[i, j] = (a[i, j] - np.sum(np.multiply(g[i, 0:j], g[j, 0:j])))/g[j, j]
    g[n-1, n-1] = np.sqrt((a[n-1, n-1] - np.sum(np.multiply(g[n-1, 0:n-1], g[n-1, 0:n-1]))))
    return g


if __name__ == "__main__":
    a = np.mat([[9, 18, 9, -27],
                [18, 45, 0, -45],
                [9, 0, 126, 9],
                [-27, -45, 9, 135]], dtype=float)
    print('G:')
    G = pingfagenfa(a)
    print(G)

The solution results are shown in the figure below:

column
Python is the direct method of solving linear equations (1) — – gaussian elimination
Python the direct method of solving linear equations (2) — – gaussian column main element elimination
Python the direct method of solving linear equations (3) — – column if primary element gauss – when elimination
Python the direct method of solving linear equations (4) — – LU decomposition of the matrix
Python the direct method of solving linear equations (5) — –
square root method to solve linear equations Python direct method for solving systems of linear equations (6) — chase method for solving tridiagonal equations
Python direct method for solving systems of linear equations (7) — givens transformation based QR decomposition
Python direct method for solving systems of linear equations (8) — haushalde transformation based QR decomposition