如何用Numpy在神经网络中正确计算梯度

问题描述 投票:2回答:1

我正在尝试使用numpy从头开始构建一个简单的神经网络类,并使用XOR问题对其进行测试。但是反向传播功能(backprop)似乎无法正常工作。

[在类中,我通过传递每一层的大小以及在每一层使用的激活函数来构造实例。我假设最终的激活函数是softmax,因此我可以计算出交叉熵损失wrt到最后一层Z的导数。我的课堂上也没有单独的偏倚矩阵集。我只是将它们包含在权重矩阵中,作为结尾处的额外一列。

我知道我的反向传播函数无法正常工作,因为神经网络永远不会收敛于某种程度上正确的输出。在比较两者的结果时,我还创建了一个数值梯度函数。我得到的数字大不相同。

根据我的理解,每层的增量值(L为最后一层,i代表其他任何层)应为:

delta_L = (dE/dZ)

delta_i = WT(i+1)dotdelta(i+1)*sigmoid_prime(zi)

这些层的相应梯度/权重更新应为:

gradient_l = (delta_l)(activations(l-1).T)

W = W - alpha(W)(dE/dW)

[*hardamard product时,a表示某层的激活,而z表示某层的未激活输出。

我用来测试的示例数据在文件的底部。

这是我第一次尝试从头开始实施反向传播算法。所以我对从这里去哪里有些迷茫。

import numpy as np

def sigmoid(n, deriv=False):
    if deriv:
        return np.multiply(n, np.subtract(1, n))
    return 1 / (1 + np.exp(-n))

def softmax(X, deriv=False):
    if not deriv:
        exps = np.exp(X - np.max(X))
        return exps / np.sum(exps)
    else:
        raise Error('Unimplemented')

def cross_entropy(y, p, deriv=False):
    """
    when deriv = True, returns deriv of cost wrt z
    """
    if deriv:
        ret = p - y
        return ret
    else:
        p = np.clip(p, 1e-12, 1. - 1e-12)
        N = p.shape[0]
        return -np.sum(y*np.log(p))/(N)

class NN:
    def __init__(self, layers, activations):
        """random initialization of weights/biases
        NOTE - biases are built into the standard weight matrices by adding an extra column
        and multiplying it by one in every layer"""
        self.activate_fns = activations
        self.weights = [np.random.rand(layers[1], layers[0]+1)]
        for i in range(1, len(layers)):
            if i != len(layers)-1:
                self.weights.append(np.random.rand(layers[i+1], layers[i]+1))

                for j in range(layers[i+1]):
                    for k in range(layers[i]+1):
                        if np.random.rand(1,1)[0,0] > .5:
                            self.weights[-1][j,k] = -self.weights[-1][j,k]

    def ff(self, X, get_activations=False):
         """Feedforward"""
         activations, zs = [], []
         for activate, w in zip(self.activate_fns, self.weights):
             X = np.vstack([X, np.ones((1, 1))]) # adding bias
             z = w.dot(X)
             X = activate(z)
             if get_activations:
                 zs.append(z)
                 activations.append(X)
         return (activations, zs) if get_activations else X

    def grad_descent(self, data, epochs, learning_rate):
        """gradient descent
        data - list of 2 item tuples, the first item being an input, and the second being its label"""
        grad_w = [np.zeros_like(w) for w in self.weights]
        for _ in range(epochs):
            for x, y in data:
                grad_w = [n+o for n, o in zip(self.backprop(x, y), grad_w)]
            self.weights = [w-(learning_rate/len(data))*gw for w, gw in zip(self.weights, grad_w)]

    def backprop(self, X, y):
        """perfoms backprop for one layer of a NN with softmax/cross_entropy output layer"""
        (activations, zs) = self.ff(X, True)
        activations.insert(0, X)

        deltas = [0 for _ in range(len(self.weights))]
        grad_w = [0 for _ in range(len(self.weights))]
        deltas[-1] = cross_entropy(y, activations[-1], True) # assumes output activation is softmax
        grad_w[-1] = np.dot(deltas[-1], np.vstack([activations[-2], np.ones((1, 1))]).transpose())
        for i in range(len(self.weights)-2, -1, -1):
            deltas[i] = np.dot(self.weights[i+1][:, :-1].transpose(), deltas[i+1]) * self.activate_fns[i](activations[i], True)
            grad_w[i] = np.hstack((np.dot(deltas[i], activations[max(0, i-1)].transpose()), deltas[i]))

        # check gradient
        num_gw = self.gradient_check(X, y, i)
        print('numerical:', num_gw, '\nanalytic:', grad_w)

        return grad_w

    def gradient_check(self, x, y, i, epsilon=1e-4):
        """Numerically calculate the gradient in order to check analytical correctness"""
        grad_w = [np.zeros_like(w) for w in self.weights]
        for w, gw in zip(self.weights, grad_w):
            for j in range(w.shape[0]):
                for k in range(w.shape[1]):
                    w[j,k] += epsilon
                    out1 = cross_entropy(self.ff(x), y)
                    w[j,k] -= 2*epsilon
                    out2 = cross_entropy(self.ff(x), y)
                    gw[j,k] = np.float64(out1 - out2) / (2*epsilon)
                    w[j,k] += epsilon # return weight to original value
        return grad_w

##### TESTING #####
X = [np.array([[0],[0]]), np.array([[0],[1]]), np.array([[1],[0]]), np.array([[1],[1]])]
y = [np.array([[1], [0]]), np.array([[0], [1]]), np.array([[0], [1]]), np.array([[1], [0]])]
data = []
for x, t in zip(X, y):
    data.append((x, t))

def nn_test():
    c = NN([2, 2, 2], [sigmoid, sigmoid, softmax])
    c.grad_descent(data, 100, .01)
    for x in X:
        print(c.ff(x))
nn_test()
python numpy machine-learning neural-network backpropagation
1个回答
0
投票
使用交叉熵损失,softmax的导数真的很好(假设您使用的是1热向量,其中“ 1热”本质上是指除单个1之外全为0的数组,即:[0,0,0 ,0,0,0,1,0,0])

对于节点y_n,其最终为y_n-t_n。因此,对于具有输出的softmax:

[0.2,0.2,0.3,0.3]

和所需的输出:

[0,1,0,0]

每个softmax节点处的梯度为:

[0.2,-0.8,0.3,0.3]

好像您要从整个数组中减去1。变量名不是很清楚,所以如果您可以将它们从L重命名为L表示的名称,例如output_layer,我将可以提供更多帮助。

此外,对于其他层来说,只是为了清除问题。当您以a^(L-1)为例时,您是指“ a等于(l-1)的幂”还是“ a异或(l-1)”?因为在python中^表示xor。

© www.soinside.com 2019 - 2024. All rights reserved.