神经网络模型python代码

以下是一个简单的神经网络模型的Python代码示例：

import numpy as np

# 定义神经网络类
class NeuralNetwork:
    def __init__(self, input_size, hidden_size, output_size):
        self.input_size = input_size
        self.hidden_size = hidden_size
        self.output_size = output_size
        
        # 随机初始化权重和偏置
        self.weights1 = np.random.randn(self.input_size, self.hidden_size)
        self.bias1 = np.random.randn(self.hidden_size)
        self.weights2 = np.random.randn(self.hidden_size, self.output_size)
        self.bias2 = np.random.randn(self.output_size)

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

    # 定义神经网络的前向传播
    def forward(self, x):
        z1 = np.dot(x, self.weights1) + self.bias1
        a1 = self.sigmoid(z1)
        z2 = np.dot(a1, self.weights2) + self.bias2
        y = self.sigmoid(z2)
        return y

# 测试代码
nn = NeuralNetwork(2, 3, 1)
x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([[0], [1], [1], [0]])
for i in range(10000):
    output = nn.forward(x)
    loss = np.mean(np.square(output - y))
    if i % 1000 == 0:
        print("Loss:", loss)
    # 反向传播
    delta2 = (output - y) * output * (1 - output)
    d_weights2 = np.dot(nn.sigmoid(nn.weights1), delta2)
    delta1 = np.dot(delta2, nn.weights2.T) * nn.sigmoid(nn.weights1) * (1 - nn.sigmoid(nn.weights1))
    d_weights1 = np.dot(x.T, delta1)
    nn.weights1 -= 0.1 * d_weights1
    nn.bias1 -= 0.1 * np.mean(delta1, axis=0)
    nn.weights2 -= 0.1 * d_weights2
    nn.bias2 -= 0.1 * np.mean(delta2, axis=0)

该模型使用 numpy 库来进行矩阵运算，实现了一个包含两个输入、三个隐藏单元和一个输出的神经网络。在训练过程中，使用反向传播算法来更新权重和偏置。