Tensorflow实现自编码机

    今天对照Tensorflow的书，实现了一个简单的自编码机。基于MNIST数据集。

    自编码器，简单来说，就是用自身的高阶特征对自身进行编码。自编码器的特征非常明显：1.期望输出与输入一致; 2.希望用高阶的特征来重构自己。也就是说，实现自编码器的目的并不是去关注分类结果，而是去关注训练完成后的网络模型，网络模型的每一层代表了从基础到高阶的特征。无监督的自编码器是一个用处广泛的提取特征的方法。

    自编码器的输入节点和输出节点的数量是一致的，但如果只是单纯地逐个复制输入节点则没有意义，应该加入某些限制。比如给数据加入噪声，形成去噪自编码器（Denoising AutoEncoder)。可以从噪声中学习出数据的特征。

去噪自编码器的实现：

import numpy as np
import sklearn.preprocessing as prep
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data


# 初始化权重，保证深度学习网络中的信号传递
def xavier_init(fan_in, fan_out, constant=1):
    low = -constant * np.sqrt(6.0 / (fan_in + fan_out))
    high = constant * np.sqrt(6.0 / (fan_in + fan_out))
    return tf.random_uniform((fan_in, fan_out),
                             minval=low, maxval=high,
                             dtype=tf.float32)


# 去噪自编码的class
class AdditiveGaussianNoiseAutoencoder(object):
    def __init__(self, n_input, n_hidden, transfer_function=tf.nn.softplus,
                 optimizer=tf.train.AdamOptimizer(), scale=0.1):
        self.n_input = n_input
        self.n_hidden = n_hidden
        # 激活函数
        self.transfer = transfer_function
        # 高斯噪声系数
        self.scale = tf.placeholder(tf.float32)
        self.training_scale = scale
        network_weights = self._initialize_weights()
        self.weights = network_weights

        # 定义网络结构
        self.x = tf.placeholder(tf.float32, [None, self.n_input])
        # 加噪声后对隐藏层进行激活
        self.n_hidden = self.transfer(tf.add(tf.matmul(
            self.x + scale * tf.random_normal((n_input,)),
            self.weights['w1']), self.weights['b1']))
        self.reconstruction = tf.add(tf.matmul(self.n_hidden,
                                               self.weights['w2']), self.weights['b2'])
        # 定义损失函数,采用平方误差
        self.cost = 0.5 * tf.reduce_sum(tf.pow(tf.subtract(
            self.reconstruction, self.x), 2.0
        ))
        # 定义优化器对cost进行优化
        self.optimizer = optimizer.minimize(self.cost)

        # 创建工作域session
        init = tf.global_variables_initializer()
        self.sess = tf.Session()
        self.sess.run(init)

    # 定义weights字典，其中w1用Xavier初始化
    def _initialize_weights(self):
        all_weights = dict()
        all_weights['w1'] = tf.Variable(xavier_init(self.n_input,
                                                    self.n_hidden))
        all_weights['b1'] = tf.Variable(tf.zeros([self.n_hidden],
                                                 dtype=tf.float32))
        all_weights['w2'] = tf.Variable(tf.zeros([self.n_hidden,
                                                  self.n_input], dtype=tf.float32))
        all_weights['b2'] = tf.Variable(tf.zeros([self.n_input],
                                                 dtype=tf.float32))
        return all_weights

    # 用一个batch的数据进行训练
    def partial_fit(self, X):
        cost, opt = self.sess.run((self.cost, self.optimizer),
                                  feed_dict={self.x: X, self.scale: self.training_scale})
        return cost

    # 用来评估测试集
    def calc_total_cost(self, X):
        return self.sess.run(self.cost, feed_dict={self.x: X,
                                                   self.scale: self.training_scale})

    # 用来学习数据的高级特征(返回自编码器隐含层的输出结果)
    def transform(self, X):
        return self.sess.run(self.hidden, feed_dict={self.x: X,
                                                     self.scale: self.training_scale})

    # 重建层
    def generate(self, hidden=None):
        if hidden is None:
            hidden = np.random.normal(size=self.weights["b1"])
        return self.sess.run(self.reconstruction,
                             feed_dict={self.hidden: hidden})

    # 整体过程
    def reconstruct(self, X):
        return self.sess.run(self.reconstruction, feed_dict={self.x: X,
                                                             self.scale: self.training_scale})

    def getWeights(self):
        return self.sess.run(self.weights['w1'])

    def getBiases(self):
        return self.sess.run(self.weights['b1'])


mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)


# 将数据标准化为均值为0,方差为1的数据
def standard_scale(X_train, X_test):
    preprocessor = prep.StandardScaler().fit(X_train)
    X_train = preprocessor.transform(X_train)
    X_test = preprocessor.transform(X_test)
    return X_train, X_test


# 以随机数作为不放回抽样的起点
def get_random_block_from_data(data, batch_size):
    start_index = np.random.randint(0, len(data) - batch_size)
    return data[start_index: (start_index + batch_size)]


X_train, X_test = standard_scale(mnist.train.images, mnist.test.images)

n_samples = int(mnist.train.num_examples)  # 样本总数
training_epoch = 20  # 训练最大轮数量
batch_size = 128  # batch的大小
display_step = 1

# 创建一个自编码器的实例
autoencoder = AdditiveGaussianNoiseAutoencoder(n_input=784,
                                               n_hidden=200,
                                               transfer_function=tf.nn.softplus,
                                               optimizer=tf.train.AdamOptimizer(learning_rate=0.001),
                                               scale=0.01)
# 计算平均损失函数，让损失函数越来越小
for epoch in range(training_epoch):
    avg_cost = 0.
    total_batch = int(n_samples / batch_size)
    for i in range(total_batch):
        batch_xs = get_random_block_from_data(X_train, batch_size)
        cost = autoencoder.partial_fit(batch_xs)
        avg_cost += cost / n_samples * batch_size

    if epoch % display_step == 0:
        print("Epoch:", '%04d' % (epoch + 1), "cost=", "{:.9f}".format(avg_cost))

# 对训练完成的模型进行性能测试
print("Total cost: " + str(autoencoder.calc_total_cost(X_test)))

运行结果如下：