2.2 計(jì)算適應(yīng)度

2.3 位置更新

根據(jù)alpha、beta和delta狼的位置來(lái)更新其余灰狼的位置，更新公式如下：

2.3.1?距離向量計(jì)算

2.3.2?計(jì)算新的位置向量

2.3.3 更新灰狼位置

1.2?流程圖

兩圖分別為灰狼優(yōu)化算法流程圖、灰狼的等級(jí)制度(從上到下的優(yōu)勢(shì)遞減)，灰狼優(yōu)化算法通過(guò)模擬灰狼獵食過(guò)程中領(lǐng)袖狼和群體狼的行為，利用領(lǐng)袖狼的位置引導(dǎo)其他灰狼逐步逼近獵物（全局最優(yōu)解），從而實(shí)現(xiàn)全局優(yōu)化，其核心在于通過(guò)位置更新公式，不斷逼近最優(yōu)解

2. 灰狼算法優(yōu)缺點(diǎn)

優(yōu)點(diǎn)：與其他優(yōu)化算法相比，灰狼算法的優(yōu)化過(guò)程更快，因?yàn)樗鼈兿鹊贸龃鸢福侔巡煌鸢高M(jìn)行比較并相應(yīng)地進(jìn)行排序，以此輸出最佳解決方案缺點(diǎn)：灰狼優(yōu)化算法屬于啟發(fā)式優(yōu)化算法，產(chǎn)生的最優(yōu)解僅接近于原始最優(yōu)解，并不是問(wèn)題真正的最優(yōu)解

3. 灰狼算法實(shí)現(xiàn)

3.1?灰狼算法在簡(jiǎn)單函數(shù)上求最值

import numpy as np

import matplotlib.pyplot as plt

plt.rcParams['font.sans-serif'] = 'SimHei'

plt.rcParams['axes.unicode_minus'] = False



# 定義優(yōu)化問(wèn)題的目標(biāo)函數(shù)

def objective_function(x):

    return np.sum(x ** 2 + x)



# 初始化灰狼優(yōu)化算法的參數(shù)

dim = 2  # 解的維度

num_wolves = 300  # 種群大小

max_iter = 100  # 最大迭代次數(shù)



# 初始化灰狼種群的位置

wolves = np.random.uniform(-10, 10, (num_wolves, dim))



# 初始化 alpha、beta、delta 狼的位置

alpha_pos = np.zeros(dim)

beta_pos = np.zeros(dim)

delta_pos = np.zeros(dim)

alpha_score = float("inf")

beta_score = float("inf")

delta_score = float("inf")



# 開(kāi)始迭代

a = 2  # 控制參數(shù)，逐漸減小

convergence_curve = []



for t in range(max_iter):

    for i in range(num_wolves):

        fitness = objective_function(wolves[i])



        if fitness < alpha_score:

            alpha_score = fitness

            alpha_pos = wolves[i].copy()

        elif fitness < beta_score:

            beta_score = fitness

            beta_pos = wolves[i].copy()

        elif fitness < delta_score:

            delta_score = fitness

            delta_pos = wolves[i].copy()



    a = 2 - t * (2 / max_iter)  # 線性減少 a



    for i in range(num_wolves):

        for j in range(dim):

            r1, r2 = np.random.rand(), np.random.rand()

            A1 = 2 * a * r1 - a

            C1 = 2 * r2

            D_alpha = abs(C1 * alpha_pos[j] - wolves[i][j])

            X1 = alpha_pos[j] - A1 * D_alpha



            r1, r2 = np.random.rand(), np.random.rand()

            A2 = 2 * a * r1 - a

            C2 = 2 * r2

            D_beta = abs(C2 * beta_pos[j] - wolves[i][j])

            X2 = beta_pos[j] - A2 * D_beta



            r1, r2 = np.random.rand(), np.random.rand()

            A3 = 2 * a * r1 - a

            C3 = 2 * r2

            D_delta = abs(C3 * delta_pos[j] - wolves[i][j])

            X3 = delta_pos[j] - A3 * D_delta



            wolves[i][j] = (X1 + X2 + X3) / 3



    convergence_curve.append(alpha_score)



print(f"最優(yōu)解: {alpha_pos}")

print(f"最優(yōu)值: {alpha_score}")



# 繪制收斂曲線

plt.figure(figsize=(15,5))

plt.plot(convergence_curve)

plt.title("收斂曲線")

plt.xlabel("迭代次數(shù)")

plt.ylabel("適應(yīng)度值")

plt.show()



# 可視化灰狼種群位置

if dim == 2:

    plt.figure(figsize=(15,5))

    plt.scatter(wolves[:, 0], wolves[:, 1], c='blue', label='Wolves')

    plt.scatter(alpha_pos[0], alpha_pos[1], c='red', label='Alpha', marker='x')

    plt.scatter(beta_pos[0], beta_pos[1], c='green', label='Beta', marker='x')

    plt.scatter(delta_pos[0], delta_pos[1], c='purple', label='Delta', marker='x')

    plt.title("灰狼種群位置")

    plt.legend()

    plt.show()

3.2?灰狼算法在標(biāo)準(zhǔn)測(cè)試函數(shù)Rastrigin上求最值

import numpy as np

import matplotlib.pyplot as plt

from mpl_toolkits.mplot3d import Axes3D



plt.rcParams['font.sans-serif'] = 'SimHei'

plt.rcParams['axes.unicode_minus'] = False



# 定義優(yōu)化問(wèn)題的目標(biāo)函數(shù)（Rastrigin函數(shù)）

def objective_function(x):

    return 10 * len(x) + np.sum(x ** 2 - 10 * np.cos(2 * np.pi * x))



# 初始化灰狼優(yōu)化算法的參數(shù)

dim = 2  # 解的維度

num_wolves = 300  # 種群大小

max_iter = 100  # 最大迭代次數(shù)



# 初始化灰狼種群的位置

wolves = np.random.uniform(-5.12, 5.12, (num_wolves, dim))



# 初始化 alpha、beta、delta 狼的位置

alpha_pos = np.zeros(dim)

beta_pos = np.zeros(dim)

delta_pos = np.zeros(dim)

alpha_score = float("inf")

beta_score = float("inf")

delta_score = float("inf")



# 開(kāi)始迭代

a = 2  # 控制參數(shù)，逐漸減小

convergence_curve = []



for t in range(max_iter):

    for i in range(num_wolves):

        fitness = objective_function(wolves[i])



        if fitness < alpha_score:

            alpha_score = fitness

            alpha_pos = wolves[i].copy()

        elif fitness < beta_score:

            beta_score = fitness

            beta_pos = wolves[i].copy()

        elif fitness < delta_score:

            delta_score = fitness

            delta_pos = wolves[i].copy()



    a = 2 - t * (2 / max_iter)  # 線性減少 a



    for i in range(num_wolves):

        for j in range(dim):

            r1, r2 = np.random.rand(), np.random.rand()

            A1 = 2 * a * r1 - a

            C1 = 2 * r2

            D_alpha = abs(C1 * alpha_pos[j] - wolves[i][j])

            X1 = alpha_pos[j] - A1 * D_alpha



            r1, r2 = np.random.rand(), np.random.rand()

            A2 = 2 * a * r1 - a

            C2 = 2 * r2

            D_beta = abs(C2 * beta_pos[j] - wolves[i][j])

            X2 = beta_pos[j] - A2 * D_beta



            r1, r2 = np.random.rand(), np.random.rand()

            A3 = 2 * a * r1 - a

            C3 = 2 * r2

            D_delta = abs(C3 * delta_pos[j] - wolves[i][j])

            X3 = delta_pos[j] - A3 * D_delta



            wolves[i][j] = (X1 + X2 + X3) / 3



    convergence_curve.append(alpha_score)



print(f"最優(yōu)解: {alpha_pos}")

print(f"最優(yōu)值: {alpha_score}")



plt.figure(figsize=(15,5))

plt.plot(convergence_curve)

plt.title("收斂曲線")

plt.xlabel("迭代次數(shù)")

plt.ylabel("適應(yīng)度值")

plt.show()



if dim == 2:

    x = np.linspace(-5.12, 5.12, 400)

    y = np.linspace(-5.12, 5.12, 400)

    X, Y = np.meshgrid(x, y)

    Z = 10 * 2 + (X ** 2 - 10 * np.cos(2 * np.pi * X)) + (Y ** 2 - 10 * np.cos(2 * np.pi * Y))

    fig = plt.figure(figsize=(8,8))

    ax = fig.add_subplot(111, projection='3d')

    ax.plot_surface(X, Y, Z, cmap='viridis', alpha=0.2)

    ax.scatter(alpha_pos[0], alpha_pos[1], alpha_score, c='red', label='Alpha', marker='x')

    ax.scatter(beta_pos[0], beta_pos[1], beta_score, c='green', label='Beta', marker='x')

    ax.scatter(delta_pos[0], delta_pos[1], delta_score, c='purple', label='Delta', marker='x')

    ax.set_title("Rastrigin求解結(jié)果")

    ax.legend()

    plt.show()

Rastrigin函數(shù)是一種常用于測(cè)試優(yōu)化算法性能的多模態(tài)函數(shù)，其數(shù)學(xué)表達(dá)式為：

4. 灰狼算法在模型上優(yōu)化的運(yùn)用

4.1 數(shù)據(jù)簡(jiǎn)單預(yù)處理

import pandas as pd

import numpy as np

df = pd.read_excel('數(shù)據(jù).xlsx',index_col=0, parse_dates=['數(shù)據(jù)時(shí)間'])



# 數(shù)據(jù)預(yù)處理

df_max = np.max(df['總有功功率（kw）'])

df_min = np.min(df['總有功功率（kw）'])

df_bz = (df['總有功功率（kw）']-df_min)/(df_max-df_min)



def prepare_data(data, win_size):

    X = []  

    y = [] 

    for i in range(len(data) - win_size):

        temp_x = data[i:i + win_size]

        temp_y = data[i + win_size] 

        X.append(temp_x)

        y.append(temp_y)

    X = np.asarray(X)

    y = np.asarray(y)

    X = np.expand_dims(X, axis=-1)



    return X, y

win_size = 12

X, y = prepare_data(df_bz.values, win_size)

train_size = int(len(X) * 0.7)  

X_train, X_test = X[:train_size], X[train_size:]

y_train, y_test = y[:train_size], y[train_size:]

4.2?灰狼算法尋找最優(yōu)參數(shù)

import tensorflow.compat.v1 as tf

from tensorflow.keras.layers import Flatten, Dense

from tensorflow.keras.models import Sequential

import matplotlib.pyplot as plt

from tcn import TCN



# 關(guān)閉TensorFlow 2.x的急切執(zhí)行

tf.disable_eager_execution()



# 定義目標(biāo)函數(shù)

def objective_function(params):

    dense_1, dense_2, filters1 = params

    dense_1, dense_2, filters1 = int(dense_1), int(dense_2), int(filters1)



    model = Sequential()

    model.add(TCN(nb_filters=filters1, kernel_size=6, activation='relu', input_shape=(win_size, 1), dilations=[1, 2, 4, 8, 16]))

    model.add(Flatten())

    model.add(Dense(dense_1, activation='relu'))

    model.add(Dense(dense_2, activation='relu'))

    model.add(Dense(1, activation='sigmoid'))



    model.compile(optimizer='adam', loss='mse')

    history = model.fit(X_train, y_train, epochs=10, batch_size=32, validation_data=(X_test, y_test), verbose=0)

    val_loss = min(history.history['val_loss'])



    return val_loss



# 初始化GWO參數(shù)

dim = 3 # 超參數(shù)的數(shù)量

num_wolves = 5  # 灰狼種群的大小

max_iter = 20  # 最大迭代次數(shù)

lower_bound = [32, 64, 32]  # 超參數(shù)的下界

upper_bound = [128, 256, 128]  # 超參數(shù)的上界



# 初始化灰狼種群的位置

wolves = np.random.uniform(lower_bound, upper_bound, (num_wolves, dim))



# 初始化 alpha、beta、delta 狼的位置

alpha_pos = np.zeros(dim)

beta_pos = np.zeros(dim)

delta_pos = np.zeros(dim)

alpha_score = float("inf")

beta_score = float("inf")

delta_score = float("inf")



# 開(kāi)始迭代

a = 2  # 控制參數(shù)，逐漸減小

convergence_curve = []



for t in range(max_iter):

    for i in range(num_wolves):

        fitness = objective_function(wolves[i])



        if fitness < alpha_score:

            alpha_score = fitness

            alpha_pos = wolves[i].copy()

        elif fitness < beta_score:

            beta_score = fitness

            beta_pos = wolves[i].copy()

        elif fitness < delta_score:

            delta_score = fitness

            delta_pos = wolves[i].copy()



    a = 2 - t * (2 / max_iter)  # 線性減少 a



    for i in range(num_wolves):

        for j in range(dim):

            r1, r2 = np.random.rand(), np.random.rand()

            A1 = 2 * a * r1 - a

            C1 = 2 * r2

            D_alpha = abs(C1 * alpha_pos[j] - wolves[i][j])

            X1 = alpha_pos[j] - A1 * D_alpha



            r1, r2 = np.random.rand(), np.random.rand()

            A2 = 2 * a * r1 - a

            C2 = 2 * r2

            D_beta = abs(C2 * beta_pos[j] - wolves[i][j])

            X2 = beta_pos[j] - A2 * D_beta



            r1, r2 = np.random.rand(), np.random.rand()

            A3 = 2 * a * r1 - a

            C3 = 2 * r2

            D_delta = abs(C3 * delta_pos[j] - wolves[i][j])

            X3 = delta_pos[j] - A3 * D_delta



            wolves[i][j] = np.clip((X1 + X2 + X3) / 3, lower_bound[j], upper_bound[j])



    convergence_curve.append(alpha_score)



print(f"最優(yōu)解: {alpha_pos}")

print(f"最優(yōu)值: {alpha_score}")



plt.plot(convergence_curve)

plt.title("收斂曲線")

plt.xlabel("迭代次數(shù)")

plt.ylabel("驗(yàn)證損失")

plt.show()

在這里目標(biāo)函數(shù) objective_function(params) 負(fù)責(zé)創(chuàng)建和訓(xùn)練模型，并返回驗(yàn)證集上的最小損失值（驗(yàn)證損失），它的輸入 params 是一個(gè)包含超參數(shù)的列表，灰狼算法初始化值如下：

dim = 3  # 超參數(shù)的數(shù)量

num_wolves = 5  # 灰狼種群的大小

max_iter = 20  # 最大迭代次數(shù)

lower_bound = [32, 64, 32]  # 超參數(shù)的下界

upper_bound = [128, 256, 128]  # 超參數(shù)的上界

這里的參數(shù)值都設(shè)置的比較小，實(shí)際上應(yīng)該根據(jù)任務(wù)復(fù)雜度進(jìn)行調(diào)整，通常較大的種群、較大迭代次數(shù)，可以提供更好的解決結(jié)果，當(dāng)然計(jì)算成本也會(huì)增加，這里只是為了演示如何利用灰狼算法進(jìn)行超參數(shù)搜索，通過(guò)算法我們輸出了當(dāng)前搜索范圍，迭代次數(shù)下的最優(yōu)參數(shù)，接下來(lái)使用該參數(shù)進(jìn)行模型訓(xùn)練

4.3?最優(yōu)參數(shù)下的模型訓(xùn)練

# 使用最優(yōu)超參數(shù)重新訓(xùn)練模型

dense_1, dense_2, filters1 = map(int, alpha_pos)



model = Sequential()

model.add(TCN(nb_filters=filters1, kernel_size=6, activation='relu', input_shape=(win_size, 1), dilations=[1, 2, 4, 8, 16]))

model.add(Flatten())

model.add(Dense(dense_1, activation='relu'))

model.add(Dense(dense_2, activation='relu'))

model.add(Dense(1, activation='sigmoid'))



model.compile(optimizer='adam', loss='mse')



history = model.fit(X_train, y_train, epochs=100, batch_size=32, validation_data=(X_test, y_test), verbose=0)



# 可視化訓(xùn)練集和測(cè)試集的損失

plt.plot(history.history['loss'], label='Training Loss')

plt.plot(history.history['val_loss'], label='Validation Loss')

plt.xlabel('Epoch')

plt.ylabel('Loss')

plt.legend()

plt.show()

文章轉(zhuǎn)自微信公眾號(hào)@Python機(jī)器學(xué)習(xí)AI

上一篇：

利用Pytorch框架構(gòu)建lstm時(shí)間序列預(yù)測(cè)模型

下一篇：

SVM多分類(lèi)分析：SHAP解釋各類(lèi)別下各特征對(duì)模型的影響力

我們有何不同？

API服務(wù)商零注冊(cè)

多API并行試用

數(shù)據(jù)驅(qū)動(dòng)選型，提升決策效率

查看全部API→
??

熱門(mén)場(chǎng)景實(shí)測(cè)，選對(duì)API

#AI文本生成大模型API

對(duì)比大模型API的內(nèi)容創(chuàng)意新穎性、情感共鳴力、商業(yè)轉(zhuǎn)化潛力

25個(gè)渠道
一鍵對(duì)比試用API 限時(shí)免費(fèi)

#AI深度推理大模型API

對(duì)比大模型API的邏輯推理準(zhǔn)確性、分析深度、可視化建議合理性

10個(gè)渠道
一鍵對(duì)比試用API 限時(shí)免費(fèi)