前馈神经网络python实现

以下是一个简单的前馈神经网络的Python实现：

import numpy as np

class NeuralNetwork:
    
    def __init__(self, layers):
        self.weights = []
        self.biases = []
        for i in range(1, len(layers)):
            self.weights.append(np.random.randn(layers[i], layers[i-1]))
            self.biases.append(np.random.randn(layers[i], 1))
    
    def sigmoid(self, x):
        return 1 / (1 + np.exp(-x))
    
    def feedforward(self, a):
        for w, b in zip(self.weights, self.biases):
            a = self.sigmoid(np.dot(w, a) + b)
        return a
    
    def train(self, X, y, epochs, lr):
        for i in range(epochs):
            for x, target in zip(X, y):
                x = x.reshape(-1, 1)
                target = target.reshape(-1, 1)
                activations = [x]
                for w, b in zip(self.weights, self.biases):
                    z = np.dot(w, activations[-1]) + b
                    a = self.sigmoid(z)
                    activations.append(a)
                error = activations[-1] - target
                deltas = [error * activations[-1] * (1 - activations[-1])]
                for w, a in zip(reversed(self.weights[1:]), reversed(activations[:-1])):
                    delta = np.dot(w.T, deltas[-1]) * a * (1 - a)
                    deltas.append(delta)
                deltas.reverse()
                for w, b, delta, a in zip(self.weights, self.biases, deltas, activations[:-1] + [x]):
                    w -= lr * np.dot(delta, a.T)
                    b -= lr * delta
    
    def predict(self, X):
        return np.array([np.argmax(self.feedforward(x)) for x in X])

上面的代码实现了一个简单的前馈神经网络，使用sigmoid函数作为激活函数。其中，初始化函数接受一个列表，表示每层神经元的数量；feedforward函数接受一个输入，返回神经网络的输出；train函数接受输入数据、标签、迭代次数和学习率，用于训练神经网络；predict函数接受输入数据，返回预测结果。