西瓜书5.2. 编程利用BP算法在西瓜数据集3.0上训练一个单隐层网络。

好的，这是一个深度学习相关的问题。下面是一个简单的Python实现，供你参考：

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

# 读取数据
data = pd.read_csv('watermelon_3.0.csv')
X = data.iloc[:, 1:3].values
y = data.iloc[:, 3].values

# 数据归一化
X = (X - np.mean(X, axis=0)) / np.std(X, axis=0)

# 定义单隐层神经网络
class NeuralNetwork:
    def __init__(self, input_dim, hidden_dim, output_dim):
        self.input_dim = input_dim
        self.hidden_dim = hidden_dim
        self.output_dim = output_dim
        self.W1 = np.random.randn(self.input_dim, self.hidden_dim)
        self.b1 = np.zeros((1, self.hidden_dim))
        self.W2 = np.random.randn(self.hidden_dim, self.output_dim)
        self.b2 = np.zeros((1, self.output_dim))

    # sigmoid激活函数
    def sigmoid(self, x):
        return 1 / (1 + np.exp(-x))

    # 前向传播
    def forward(self, X):
        self.z1 = np.dot(X, self.W1) + self.b1
        self.a1 = self.sigmoid(self.z1)
        self.z2 = np.dot(self.a1, self.W2) + self.b2
        self.y_hat = self.sigmoid(self.z2)
        return self.y_hat

    # sigmoid函数的导数
    def sigmoid_derivative(self, x):
        return x * (1 - x)

    # 反向传播
    def backward(self, X, y, y_hat, learning_rate):
        delta2 = (y_hat - y) * self.sigmoid_derivative(y_hat)
        dW2 = np.dot(self.a1.T, delta2)
        db2 = np.sum(delta2, axis=0, keepdims=True)
        delta1 = np.dot(delta2, self.W2.T) * self.sigmoid_derivative(self.a1)
        dW1 = np.dot(X.T, delta1)
        db1 = np.sum(delta1, axis=0)
        self.W2 -= learning_rate * dW2
        self.b2 -= learning_rate * db2
        self.W1 -= learning_rate * dW1
        self.b1 -= learning_rate * db1

    # 训练模型
    def train(self, X, y, learning_rate=0.1, num_epochs=1000):
        loss_history = []
        for i in range(num_epochs):
            y_hat = self.forward(X)
            loss = np.mean((y_hat - y) ** 2)
            self.backward(X, y, y_hat, learning_rate)
            loss_history.append(loss)
            if i % 100 == 0:
                print(f"Epoch {i}, Loss: {loss}")
        return loss_history

# 定义超参数
input_dim = 2
hidden_dim = 3
output_dim = 1
learning_rate = 0.1
num_epochs = 1000

# 定义并训练模型
model = NeuralNetwork(input_dim, hidden_dim, output_dim)
loss_history = model.train(X, y, learning_rate, num_epochs)

# 绘制损失函数图像
plt.plot(loss_history)
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.show()

# 测试模型
y_hat = model.forward(X)
y_pred = np.round(y_hat)
accuracy = np.mean(y_pred == y)
print(f"Accuracy: {accuracy}")

需要注意的是，这只是一个简单的实现，可能还需要根据实际情况进行更多的调整和优化。希望这个实现能够对你有所帮助。