感知器的三个算法用Python呈现

1. 感知器学习算法

# 定义感知器函数
def perceptron(weights, bias, input):
    dot_product = sum(weight * i for weight, i in zip(weights, input)) + bias
    return 1 if dot_product > 0 else 0

# 定义感知器学习算法函数
def perceptron_learning_algorithm(data, labels, learning_rate):
    # 初始化权重和偏置
    weights = [0] * len(data[0])
    bias = 0

    # 迭代更新权重和偏置
    while True:
        error_count = 0
        for input, label in zip(data, labels):
            prediction = perceptron(weights, bias, input)
            error = label - prediction
            if error != 0:
                error_count += 1
                for i in range(len(weights)):
                    weights[i] += learning_rate * error * input[i]
                bias += learning_rate * error
        if error_count == 0:
            break

    # 返回训练好的模型
    return weights, bias

2. Delta规则算法

# 定义 Delta 规则函数
def delta_rule(weights, bias, input, label, learning_rate):
    y = sum(weight * i for weight, i in zip(weights, input)) + bias
    error = label - y
    for i in range(len(weights)):
        weights[i] += learning_rate * error * input[i]
    bias += learning_rate * error
    return weights, bias

3. Hebbian算法

# 定义 Hebbian 算法函数
def hebbian_learning(data):
    # 初始化权重和偏置
    weights = [0] * len(data[0])
    bias = 0

    # 迭代更新权重和偏置
    for input in data:
        for i in range(len(weights)):
            weights[i] += input[i] * input[i]
        bias += sum(input)

    # 返回训练好的模型
    return weights, bias