类型错误：'努比·恩达雷'对象不可调用（SciPy优化）

Question

我是python的新用户。目前，我正在编写一个python脚本，该脚本旨在通过SciPy最小化数据协调和参数估计的目标函数。代码如下所示：

import numpy as np
import sys 
import pandas as pd
from scipy.optimize import minimize
import math 
# read from specified excel file
file = 'TP Data.xlsx'
x1 = pd.ExcelFile(file)
df = pd.read_excel(file,sheet_name = 'Sheet1');

# number of trial points
df_nr = len(df.index)
ndata = 1
if ndata > df_nr:
    sys.exit('Error! Input trials exceeds data trials.')

df_lim = df.iloc[0:ndata]

# data initialization
FA6 = df_lim['F(A6)'].values
TA2 = df_lim['T(A2)'].values
TA4 = df_lim['T(A4)'].values
TA5 = df_lim['T(A5)'].values
TA7 = df_lim['T(A7)'].values
TA8 = df_lim['T(A8)'].values
TD1 = df_lim['T(D1)'].values
TC1 = df_lim['T(C1)'].values
TB1 = df_lim['T(B1)'].values
TB2 = df_lim['T(B2)'].values
FA1 = df_lim['F(A1)'].values
FA3 = df_lim['F(A3)'].values
FC1 = df_lim['F(C1)'].values
FD1 = df_lim['F(D1)'].values
FB1 = df_lim['F(B1)'].values
TA1 = df_lim['T(A1)'].values
TB3 = df_lim['T(B3)'].values
TC2 = df_lim['T(C2)'].values
TD2 = df_lim['T(D2)'].values
# unmeasured variables TA3 and TA6, initialized as TA2
TA3 = TA2.copy()
TA6 = TA2.copy()

data_arr = np.concatenate((FA6, TA2, TA4, TA5, TA7, TA8, TD1, TC1, TB1, TB2, FA1, FA3, FC1, FD1, FB1, TA1, TB3, TC2, TD2, TA3, TA6))
data_mat = np.reshape(data_arr,(21,-1))
U = np.array([5.0,4.0,7.0,5.0])
data_arr_full = np.concatenate((data_arr, U))

sflow =  0.5*np.ones((1,ndata))
stemp = 0.01*np.ones((1,ndata))

stdDevAllrow = np.concatenate((sflow, stemp, stemp, stemp, stemp, stemp, stemp, stemp, stemp, stemp, sflow, sflow, sflow, sflow, sflow, stemp, stemp, stemp, stemp, stemp, stemp), axis = 0)
#fudge = len(U) + 2*len(TA3);

def objective(x):
    sumz = 0
    for i in range(0,19):
        sumz = sumz + ((x[i] - data_arr[i])/stdDevAllrow[i])**2
    return sumz

#xlol = np.zeros(len(np.arange(0,19)))
#qlol = objective(xlol)

def constraint_residuals():
    xdata = data_arr_full[0:-4]
    UAint = data_arr_full[-4::]
    FA6=xdata[0]
    TA2=xdata[1]
    TA4=xdata[2]
    TA5=xdata[3]
    TA7=xdata[4]
    TA8=xdata[5]
    TD1=xdata[6]
    TC1=xdata[7]
    TB1=xdata[8]
    TB2=xdata[9]
    FA1=xdata[10]
    FA3=xdata[11]
    FC1=xdata[12]
    FD1=xdata[13]
    FB1=xdata[14]
    TA1=xdata[15]
    TB3=xdata[16]
    TC2=xdata[17]
    TD2=xdata[18]
    TA3=xdata[19]
    TA6=xdata[20]
    dt1_1 = TB2 - TA2
    dt1_2 = TB3 - TA1
    dt2_1 = TB1 - TA4
    dt2_2 = TB2 - TA3
    dt3_1 = TC1 - TA5
    dt3_2 = TC2 - TA4
    dt4_1 = TD1 - TA7
    dt4_2 = TD2 - TA6

    lmtd1 = (dt1_1 - dt1_2)/math.log(dt1_1/dt1_2)
    lmtd2 = (dt2_1 - dt2_2)/math.log(dt2_1/dt2_2)
    lmtd3 = (dt3_1 - dt3_2)/math.log(dt3_1/dt3_2)
    lmtd4 = (dt4_1 - dt4_2)/math.log(dt4_1/dt4_2)

    y = np.zeros(14)

    y[0] = FA1 - FA3 - FA6
    y[1] = FA1*(TA2 - TA1) - FB1*(TB2 - TB3)
    y[2] = FA3*(TA4 - TA3) - FB1*(TB1 - TB2)
    y[3] = FA3*(TA5 - TA4) - FC1*(TC1 - TC2) 
    y[4] = FA6*(TA7 - TA6) - FD1*(TD1 - TD2)
    y[5] = FA1*TA8 - FA3*TA5 - FA6*TA7
    y[6] = UAint[0]*lmtd1 - FB1*(TB2 - TB3)
    y[7] =  UAint[1]*lmtd2 - FB1*(TB1 - TB2)
    y[8] = UAint[2]*lmtd3 - FC1*(TC1 - TC2)
    y[9] = UAint[3]*lmtd4 - FD1*(TD1 - TD2) 
    y[10] = TB2 - TA4
    y[11] = TB1 - TA4
    y[12] = TC1 - TA5
    y[13] = TD1 - TA7

    return y

res_con = constraint_residuals()
complete_con = np.append(res_con,data_arr_full)
#print(complete_con)
consts = {'type':'ineq', 'fun': complete_con}

def constraint_bounds():
    con1_low_lim = np.zeros(len(res_con)) #exactness
    con1_up_lim = np.zeros(len(res_con)) #exactness
    con1_low_lim = np.append(con1_low_lim, [1, 1, 1, 1])
    con1_up_lim = np.append(con1_up_lim, [10, 10, 10, 10])
    con2_low_lim = data_mat - 5*stdDevAllrow
    con2_up_lim = data_mat + 5*stdDevAllrow

    con_low_lim = np.append(con1_low_lim,con2_low_lim)
    con_up_lim = np.append(con1_up_lim,con2_up_lim)
    bound_stack = np.vstack((con_low_lim, con_up_lim))
    bound_stack = np.transpose(bound_stack)
    bounds = tuple(map(tuple, bound_stack))

    return bounds

bnds = constraint_bounds()   
xguess = data_arr_full*1.001  
xguess = xguess.tolist()

sol = minimize(objective, xguess, method ='SLSQP', bounds = bnds, 
constraints = consts)   

当使用最小化时，我得到标题中提到的错误。在

完全回溯错误：

^{pr2}$

我不知道该怎么解决这个问题。感谢任何帮助。在

谢谢你！在