def f(params): pi12, pi21 = params LL = 10*np.log(40*60/110**2) + 30*np.log(40*50/110**2) + 20*np.log(20/110*(50/110 + 60/110*pi12)) + \ 50*np.log(50/110*(60/110 + 50/110*pi21)) - 110*np.log(40*60/110**2 + 40*50/110**2 + \ 20/110*(50/110 + 60/110*pi12) + 50/110*(60/110 + 50/110*pi21)) return -LLdef callbackF(Xi): global Nfeval print('pass callback',str(Nfeval)) print(Nfeval, Xi[0], Xi[1], f(Xi)) Nfeval += 1initial_guess = [0, 0]b = (0.0, 1.0)b0 = b1 = bbnb = [b0, b1]res = minimize(f, initial_guess, bounds=bnb, method='bfgs', callback=callbackF, options={'disp':True})print (res)我試圖捕獲目標函數的最佳結果，但它總是顯示 [1, 1]。理想情況下，我應該將兩個參數限制在 0 和 1 之間，以最大化 -LL。我做錯了什么嗎？我想知道是否應該在每次迭代后更新目標函數，但我很困惑如何使其發揮作用。我檢查了幾個帖子但仍然不確定。其他閱讀材料也非常受歡迎。多謝！pass callback 2525 0.6870283538140954 0.7403323855238932 143.98656641020855pass callback 2626 0.7935216169001177 0.7090801503785442 143.93658208323882pass callback 2727 0.8314173041320377 0.7666686643426496 143.84748818067345pass callback 2828 0.9264732632840973 0.8980814706430704 143.7237871814941pass callback 2929 0.9885339111975429 0.9836968132795704 143.69759782341296pass callback 3030 0.999243206123829 0.9988036732413753 143.69694856450647pass callback 3131 1.0000109917713558 0.9999811482899945 143.6969451785149pass callback 3232 1.0000049364520325 0.99999836986115 143.6969451768374pass callback 3333 1.000000303283094 1.000000360615374 143.6969451767528Optimization terminated successfully. Current function value: 143.696945 Iterations: 9 Function evaluations: 44 Gradient evaluations: 11 fun: 143.6969451767528 hess_inv: array([[0.2255719 , 0.08676943], [0.08676943, 0.21320636]]) jac: array([1.90734863e-06, 3.81469727e-06]) message: 'Optimization terminated successfully.' nfev: 44 nit: 9 njev: 11 status: 0 success: True x: array([1.0000003 , 1.00000036])
1 回答
弒天下
TA貢獻1818條經驗 獲得超8個贊
看起來優化器正在 [1,1] 處找到正確的最小值。這是您的函數圖：
這是我用來生成該圖的代碼。
from scipy.optimize import minimize
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import cm
def f(params):
pi12, pi21 = params
LL = 10*np.log(40*60/110**2) + 30*np.log(40*50/110**2) + 20*np.log(20/110*(50/110 + 60/110*pi12)) + \
50*np.log(50/110*(60/110 + 50/110*pi21)) - 110*np.log(40*60/110**2 + 40*50/110**2 + \
20/110*(50/110 + 60/110*pi12) + 50/110*(60/110 + 50/110*pi21))
return -LL
def g(X,Y):
return f([X,Y])
initial_guess = [0, 0]
b = (0.0, 1.0)
b0 = b1 = b
bnb = [b0, b1]
res = minimize(f, initial_guess, bounds=bnb, method='bfgs')
print (res)
X = np.arange(0, 1, 0.05)
Y = np.arange(0, 1, 0.05)
X, Y = np.meshgrid(X, Y)
Z = g(X,Y)
fig = plt.figure()
ax = fig.gca(projection='3d')
# Plot the surface.
surf = ax.plot_surface(X, Y, Z, cmap=cm.coolwarm,
linewidth=0, antialiased=False)
plt.show()
添加回答
舉報
0/150
提交
取消