哪里除了问题

print("hello world")
#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import numpy as np
from numpy import matrix as mat
import math

# 导入数据
Label_location = [0.9, 1.2, 2.0]
theta_data = [60.31857894892403, 48.25486315913922, 80.4247719318987, 80.4247719318987]
lambda_data = [0.3125, 0.3125, 0.3125, 0.3125]
xi_data = [0.0, 0.9, 2.5, 0.9]
yi_data = [0.0, 0.0, 0.0, 0.0]
zi_data = [2.0, 2.0, 2.0, 0.4]
# 合并为一个矩阵，然后转置,每一行为一组λ，xi,yi,zi。
Variable_Matrix = mat([lambda_data, xi_data, yi_data, zi_data]).T


def Func(parameter, iput):  # 需要拟合的函数，abc是包含三个参数的一个矩阵[[a],[b],[c]]
   x = parameter[0, 0]
   y = parameter[1, 0]
   z = parameter[2, 0]
   residual = mat((4*np.pi / iput[0, 0])*np.sqrt(np.square(iput[0, 1]-x)+np.square(iput[0, 2]-y)+np.square(iput[0, 3]-z)))
   return residual


def Deriv(parameter, iput):  # 对函数求偏导
   x = parameter[0, 0]
   y = parameter[1, 0]
   z = parameter[2, 0]
   x_deriv = -4*np.pi*(iput[0, 1]-x) / (iput[0, 0] * np.sqrt(np.square(iput[0, 1]-x)+np.square(iput[0, 2]-y) + np.square(iput[0, 3]-z)))
   y_deriv = -4*np.pi*(iput[0, 2]-y) / (iput[0, 0] * np.sqrt(np.square(iput[0, 1]-x)+np.square(iput[0, 2]-y) + np.square(iput[0, 3]-z)))
   z_deriv = -4*np.pi*(iput[0, 3]-z) / (iput[0, 0] * np.sqrt(np.square(iput[0, 1]-x)+np.square(iput[0, 2]-y) + np.square(iput[0, 3]-z)))
   deriv_matrix = mat([x_deriv, y_deriv, z_deriv])
   return deriv_matrix


n = len(theta_data)
J = mat(np.zeros((n, 3)))  # 雅克比矩阵
fx = mat(np.zeros((n, 1)))  # f(x)  3*1  误差
fx_tmp = mat(np.zeros((n, 1)))
initialization_parameters = mat([[10], [400], [30]])  # 参数初始化
lase_mse = 0.0
step = 0.0
u, v = 1.0, 2.0
conve = 100


while conve:
   mse, mse_tmp = 0.0, 0.0
   step += 1
   for i in range(len(theta_data)):
       fx[i] = Func(initialization_parameters, Variable_Matrix[i]) - theta_data[i]  # 注意不能写成  y - Func  ,否则发散
       # print(fx[i])
       mse += fx[i, 0] ** 2
       J[i] = Deriv(initialization_parameters, Variable_Matrix[i])  # 数值求导
   mse = mse/n  # 范围约束
   H = J.T * J + u * np.eye(3)  # 3*3
   dx = -H.I * J.T * fx  # 注意这里有一个负号，和fx = Func - y的符号要对应

   initial_parameters_tmp = initialization_parameters.copy()
   initial_parameters_tmp = initial_parameters_tmp + dx
   for j in range(len(theta_data)):
       fx_tmp[j] = Func(initial_parameters_tmp, Variable_Matrix[j]) - theta_data[j]
       mse_tmp += fx_tmp[j, 0] ** 2
   mse_tmp = mse_tmp/n

   q = (mse - mse_tmp) / ((0.5 * dx.T * (u * dx - J.T * fx))[0, 0])
   print(q)
   if q > 0:
       s = 1.0 / 3.0
       v = 2
       mse = mse_tmp
       initialization_parameters = initial_parameters_tmp
       temp = 1 - pow(2 * q - 1, 3)
       if s > temp:
           u = u * s
       else:
           u = u * temp
   else:
       u = u * v
       v = 2 * v
       mse = mse_tmp
       initialization_parameters = initial_parameters_tmp
   print("step = %d,parameters(mse-lase_mse) = " % step, abs(mse - lase_mse))
   if abs(mse - lase_mse) < math.pow(0.1, 14):
       break
   lase_mse = mse  # 记录上一个 mse 的位置
   conve -= 1
print(lase_mse)
print(initialization_parameters)