[深度学习-1]神经网络-手写数字识别

Date:2018年9月29记

数据及彭亮给出的数据集和源码在下面网址:

https://github.com/mnielsen/neural-networks-and-deep-learning

 

在放上代码之前要先把踩过的坑补一下,不白费国庆前掉发找bug

[bug-1]   

         nabla_w = [nw + dnw for nw, dnw in zip(nabla_w, delta_nabla_w)]
     ValueError: operands could not be broadcast together with shapes (30,784) (30,) 

Traceback (most recent call last):
  File "/home/kongze/PycharmProjects/MNIST_Demo/main.py", line 6, in <module>
    net.SGD(training_data, 1, 10, 3.0, test_data=test_data)
  File "/home/kongze/PycharmProjects/MNIST_Demo/network.py", line 38, in SGD
    self.update_mini_batch(mini_batch, eta)
  File "/home/kongze/PycharmProjects/MNIST_Demo/network.py", line 50, in update_mini_batch
    nabla_w = [nw + dnw for nw, dnw in zip(nabla_w, delta_nabla_w)]
  File "/home/kongze/PycharmProjects/MNIST_Demo/network.py", line 50, in <listcomp>
    nabla_w = [nw + dnw for nw, dnw in zip(nabla_w, delta_nabla_w)]
ValueError: operands could not be broadcast together with shapes (30,784) (30,) 

这是百度,google都解决不了,花了好久才发现,这个bug在def backprop函数中的

        #非输出层
        for l in range(2, self.num_layers):
            z = zs[-l]
            sp = sigmoid_prime(z)
            delta = np.dot(self.weights[-l+1].transpose(), delta) * sp
            nabla_b[-l] = delta
            nabla_w[-l] = np.dot(delta, activation[-l-1].transpose())
        return (nabla_b,nabla_w)

activations,硬生生的写成了activation,绝情的少了一个s,花去了2个小时找这个失误,真是差之毫厘谬以千里

[bug-2] python2 与python3这种爹坑儿子的bug
           TypeError: object of type 'zip' has no len()

Traceback (most recent call last):
training data
<class 'zip'>
  File "/home/kongze/PycharmProjects/MNIST_Demo/demo.py", line 9, in <module>
    print(len(training_data))
TypeError: object of type 'zip' has no len()

完美解决方案就是:在mnist_loader模块中

def load_data_wrapper():
    tr_d, va_d, te_d = load_data()
    #训练集
    training_inputs = [np.reshape(x, (784, 1)) for x in tr_d[0]]
    training_labels = [vectorized_label(y) for y in tr_d[1]]
    training_data = zip(training_inputs, training_labels)

training_data那一行的zip前面加入list,改成 list(zip....).

 

[01-MNIST_loader.py]

import pickle
import gzip
import numpy as np

def load_data():
    file = gzip.open('data/mnist.pkl.gz', 'rb')
    training_data, validation_data, test_data = pickle.load(file, encoding='bytes')
    file.close()
    return (training_data, validation_data, test_data)

def load_data_wrapper():
    tr_d, va_d, te_d = load_data()
    #训练集
    training_inputs = [np.reshape(x, (784, 1)) for x in tr_d[0]]
    training_labels = [vectorized_label(y) for y in tr_d[1]]
    training_data = list(zip(training_inputs, training_labels))

    #验证集
    validation_inputs = [np.reshape(x, (784, 1)) for x in va_d[0]]
    validation_data = list(zip(validation_inputs, va_d[1]))


    #测试集
    test_inputs = [np.reshape(x, (784, 1)) for x in te_d[0]]
    test_data = list(zip(test_inputs, te_d[1]))


    return (training_data, validation_data, test_data)
def vectorized_label(j):

    e = np.zeros((10, 1))
    e[j] = 1.0
    return e

[02-demo.py]

import mnist_loader
import network

training_data, validation_data, test_data = mnist_loader.load_data_wrapper()

print("training data")

print(type(training_data))
print(len(training_data))
print(training_data[0][0].shape)
print(training_data[0][1].shape)

[03-network.py]

import random
import numpy as np




class Network(object):
    #构造初始化神经网络
    def __init__(self, sizes):
        self.num_layers = len(sizes)
        self.sizes = sizes
        self.biases = [np.random.randn(y, 1) for y in sizes[1:]]
        self.weights = [np.random.randn(y, x) for x, y in zip(sizes[:-1], sizes[1:])]

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

    #随机梯度下降(训练数据,迭代次数,小样本数量,学习率,是否有测试集)
    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) #训练数据大小

        #迭代过程
        for j in range(epochs):
            random.shuffle(training_data)
            #mini_batchse是列表中存放切割后的列表
            mini_batches = [training_data[k:k + mini_batch_size] for k in range(0, n, mini_batch_size)]
            #每个mini_batch都更新一次,重复完整个数据集
            for mini_batch in mini_batches:
                #存储C对于各个参数的偏导
                #格式和self.biases和self.weights是一模一样
                self.update_mini_batch(mini_batch, eta)
            if test_data:
                print("Epoch {0}: {1} / {2}".format(j, self.evaluate(test_data), n_test))

            else:
                print("Epoch{0} complete".format(j))
    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)]

    def backprop(self, x, y):

        #存储C对于各个参数的偏导
        #格式和self.biases和self.weight是一摸一样
        nabla_b = [np.zeros(b.shape) for b in self.biases]
        nabla_w = [np.zeros(w.shape) for w in self.weights]

        #前向过程
        activation = x
        activations = [x]
        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)

        #反向过程
        #输出层error
        delta = self.cost_derivative(activations[-1], y) * \
                sigmoid_prime(zs[-1])

        nabla_b[-1] = delta
        nabla_w[-1] = np.dot(delta, activations[-2].transpose())

        #非输出层
        for l in range(2, self.num_layers):
            z = zs[-l]
            sp = sigmoid_prime(z)
            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)

    def evaluate(self, test_data):
        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)

    #输出层cost函数对于a的倒数
    def cost_derivative(self, output_activations, y):
        return (output_activations-y)

def sigmoid(z):
    return 1.0 / (1.0 + np.exp(-z))


def sigmoid_prime(z):
    return sigmoid(z) * (1 - sigmoid(z))

[04-main.py]

import mnist_loader
import network

training_data, validation_data, test_data = mnist_loader.load_data_wrapper()
net = network.Network([784, 30, 10])
net.SGD(training_data, 30, 10, 3.0, test_data=test_data)

另外可以查看,另一博客中代码解释:

https://blog.csdn.net/xierhacker/article/details/53282060/

[05-改进的network]

[network2.py]

import json
import random
import sys
import numpy as np


class QuadraticCost(object):
    @staticmethod
    def fn(a, y):
        return 0.5*np.linalg.norm(a-y)**2

    @staticmethod
    def delta(z, a, y):
        return(a-y)*sigmoid_prime(z)


class CrossEntropyCost(object):
    @staticmethod
    def fn(a, y):
        return np.sum(np.nan_to_num(-y * np.log(a) - (1-y) * np.log(1-a)))

    @staticmethod
    def delta(z, a, y):
        return (a-y)

class Network(object):
    def __init__(self, sizes, cost=CrossEntropyCost):

        self.num_layers = len(sizes)
        self.sizes = sizes
        self.default_weight_initializer()
        self.cost = cost

    def default_weight_initializer(self):
        self.biases = [np.random.rand(y, 1) for y in self.sizes[1:]]
        self.weights = [np.random.randn(y, x) / np.sqrt(x)
                        for x, y in zip(self.sizes[:-1], self.sizes[1:])]
    def large_weight_initializer(self):
        self.biases = [np.random.randn(y, 1) for y in self.sizes[1:]]
        self.weight = [np.random.randn(y, x) for x, y in zip(self.sizes[:-1], self.sizes[1:])]


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

    def SGD(self, training_data, epochs, mini_batch_size, eta,
            lmbda=0.0,
            evaluation_data=None,
            monitor_evaluation_cost=False,
            monitor_evaluation_accuracy=False,
            monitor_training_cost=False,
            monitor_training_accuracy=False):
        if evaluation_data: n_data = len(evaluation_data)
        n = len(training_data)
        evaluation_cost, evaluation_accuracy = [], []
        training_cost, training_accuracy = [], []
        for j in range(epochs):
            random.shuffle(training_data)
            mini_batches = [
                training_data[k:k+mini_batch_size]
                for k in range(0, n, mini_batch_size)]
            for mini_batch in mini_batches:
                self.update_mini_batch(
                    mini_batch, eta, lmbda, len(training_data))
            print("epoch %s training complete" % j)

            if monitor_training_cost:
                cost = self.total_cost(training_data, lmbda)
                training_cost.append(cost)
                print("cost on training data:{}".format(cost))

            if monitor_training_accuracy:
                accuracy = self.accuracy(training_data, convert=True)
                training_accuracy.append(accuracy)
                print("Accuracy on training data:{} / {}".format(accuracy, n))

            if monitor_evaluation_cost:
                cost = self.total_cost(evaluation_data, lmbda, convert=True)
                evaluation_cost.append(cost)
                print("Cost on evaluation data: {}".format(cost))
            if monitor_evaluation_accuracy:
                accuracy = self.accuracy(evaluation_data)
                evaluation_accuracy.append(accuracy)
                print("Accuracy on evaluation data:{} / {}".format(self.accuracy(evaluation_data), n_data))

            print()

        return evaluation_cost, evaluation_accuracy, training_cost, training_accuracy

    def update_mini_batch(self, mini_batch, eta, lmbda, n):
        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 = [(1-eta*(lmbda/n))*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)]
    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 = x
        activations = [x]
        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)
        delta = (self.cost).delta(zs[-1], activations[-1], y)
        nabla_b[-1] = delta
        nabla_w[-1] = np.dot(delta, activations[-2].transpose())

        for l in range(2, self.num_layers):
            z = zs[-l]
            sp = sigmoid_prime(z)
            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)

    def accuracy(self, data, convert=False):
        if convert:
            results = [(np.argmax(self.feedforward(x)), np.argmax(y)) for (x, y) in data]
        else:
            results = [(np.argmax(self.feedforward(x)), y) for (x, y) in data]
        return sum(int(x == y) for (x, y) in results)

    def total_cost(self, data, lmbda, convert=False):
        cost = 0.0
        for x, y in data:
            a = self.feedforward(x)
            if convert: y = vectorized_result(y)
            cost += self.cost.fn(a, y)/len(data)
        cost += 0.5*(lmbda/len(data))*sum(np.linalg.norm(w)**2 for w in self.weights)

        return cost

    def save(self, filename):
        data = {"sizes": self.sizes,
                "weights": [w.tolist() for w in self.weights],
                "biases": [b.tolist() for b in self.biases],
                "cost": str(self.cost.__name__)}
        f = open(filename, "w")
        json.dump(data, f)
        f.close()

def load (filename):
    f = open(filename, "r")
    data = json.load(f)
    f.close()
    cost = getattr(sys.modules[__name__], data["cost"])
    net = Network(data["sizes"], cost=cost)
    net.weights = [np.array(w) for w in data["weights"]]
    net.biases = [np.array(b) for b in data["biases"]]

    return net

def vectorized_result(j):
    e = np.zeros((10, 1))
    e[j] = 1.0
    return e

def sigmoid(z):
    return 1.0/(1.0 + np.exp(-z))

def sigmoid_prime(z):
    return sigmoid(z)*(1-sigmoid(z))

[06-main2.py]

import network2
import mnist_loader

training_data, validation_data, test_data = mnist_loader.load_data_wrapper()
net = network2.Network([784, 30, 10], cost=network2.CrossEntropyCost)
net.SGD(training_data, 30, 10, 0.5, 5.0,
        evaluation_data=validation_data,
        monitor_evaluation_accuracy=True,
        monitor_evaluation_cost=True,
        monitor_training_accuracy=True,
        monitor_training_cost=True)