「neural network and deep learning」問題解-ch 02 Networkソースコード分析

http://blog.csdn.net/u011239443/article/details/75008380
完全なコード:https://github.com/xiaoyesoso/neural-networks-and-deep-learning/blob/master/src/network.py
初期化

    # sizes          
    def __init__(self, sizes):
        self.num_layers = len(sizes)
        self.sizes = sizes
        # randn               
        #         biases
        self.biases = [np.random.randn(y, 1) for y in sizes[1:]]
        #          weights
        # y         ，x         
        self.weights = [np.random.randn(y, x)
for x, y in zip(sizes[:-1], sizes[1:])]

トレーニング

    # training_data     
    # epochs     
    # mini_batch_size       
    # test_data     
    def SGD(self, training_data, epochs, mini_batch_size, eta,
            test_data=None):
        if test_data: n_test = len(test_data)
        n = len(training_data)
        for j in xrange(epochs):
            random.shuffle(training_data)
            mini_batches = [
                training_data[k:k+mini_batch_size]
                for k in xrange(0, n, mini_batch_size)]
            for mini_batch in mini_batches:
                #       
                self.update_mini_batch(mini_batch, eta)
            if test_data:
               #   test_data ！= None，
               #      
                print "Epoch {0}: {1} / {2}".format(
                    j, self.evaluate(test_data), n_test)
            else:
print "Epoch {0} complete".format(j)

モデルの更新:

    def update_mini_batch(self, mini_batch, eta):
        nabla_b = [np.zeros(b.shape) for b in self.biases]
        nabla_w = [np.zeros(w.shape) for w in self.weights]
        for x, y in mini_batch:
         #         
            delta_nabla_b, delta_nabla_w = self.backprop(x, y)
            nabla_b = [nb+dnb for nb, dnb in zip(nabla_b, delta_nabla_b)]
            nabla_w = [nw+dnw for nw, dnw in zip(nabla_w, delta_nabla_w)]
         #     
        self.weights = [w-(eta/len(mini_batch))*nw
                        for w, nw in zip(self.weights, nabla_w)]
        self.biases = [b-(eta/len(mini_batch))*nb
for b, nb in zip(self.biases, nabla_b)]

ぎゃくほうこうでんぱ
下方向に伝播する4つの式を先にレビューできます.http://blog.csdn.net/u011239443/article/details/74859614

    def backprop(self, x, y):
        nabla_b = [np.zeros(b.shape) for b in self.biases]
        nabla_w = [np.zeros(w.shape) for w in self.weights]
        # activation         
        #          
        activation = x
        activations = [x] 
        # zs               
        zs = [] 
        for b, w in zip(self.biases, self.weights):
            z = np.dot(w, activation)+b
            zs.append(z)
            activation = sigmoid(z)
            activations.append(activation)
        # cost_derivative         
        # sigmoid_prime   sigmoid    
        #      （BP1）
        delta = self.cost_derivative(activations[-1], y) * \
            sigmoid_prime(zs[-1])
        #      （BP3）
        nabla_b[-1] = delta
        #      （BP4）
        nabla_w[-1] = np.dot(delta, activations[-2].transpose())
        for l in xrange(2, self.num_layers):
            z = zs[-l]
            sp = sigmoid_prime(z)
            #      （BP2）
            delta = np.dot(self.weights[-l+1].transpose(), delta) * sp
            nabla_b[-l] = delta
            nabla_w[-l] = np.dot(delta, activations[-l-1].transpose())
return (nabla_b, nabla_w)

テストSGDのdef evaluateに戻ります.

    def evaluate(self, test_data):
     # np.argmax(self.feedforward(x))        
        test_results = [(np.argmax(self.feedforward(x)), y)
                        for (x, y) in test_data]
return sum(int(x == y) for (x, y) in test_results)

    def feedforward(self, a):
      #              
        for b, w in zip(self.biases, self.weights):
            a = sigmoid(np.dot(w, a)+b)
return a