Tensorflow实现自编码器

#encoding: utf-8
'''
AutoEncoder:使用自身的高阶编码器来提取特征，自编码器其实也是一种神经网络，
它的输入和输出是一致的它借助稀疏编码的思想，目标是使用稀疏的一些高阶特征
重新组合来重构自己。
特点：①期望输入/输出一致；②用高阶特征来重构自己，不是复制像素点
'''

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

# 载入MNIST数据
mnist = input_data.read_data_sets('MNIST_data', one_hot=True)


# 标准的均匀分布的Xavier初始化器 fan_in:输入节点的数量 fan_out:输出节点的数量
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)


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 #输入变量数
		self.n_hidden = n_hidden #隐含层结点数
		self.transfer = transfer_function #隐含层激活函数 默认为softplus
		self.scale = tf.placeholder(tf.float32) 
		self.training_scale = scale #高斯噪声系数
		network_weights = self._initialize_weights() #参数初始化
		self.weights = network_weights

		# 定义网络结构
		self.x = tf.placeholder(tf.float32, [None, self.n_input])
		# 将输入x加入噪声
		# 将加了噪声的输入与隐含层的权重w1相乘 再加上隐含层的偏置b1
		# 激活函数处理
		self.hidden = self.transfer(tf.add(tf.matmul(self.x + scale*tf.random_normal((n_input,)),
									self.weights['w1']) , self.weights['b1']))
		# 经过隐含层在输出层进行数据复原、重建
		self.reconstruction = tf.add(tf.matmul(self.hidden, self.weights['w2']), self.weights['b2'])
		
		# 定义自编码函数的损失函数
		self.cost = 0.5*tf.reduce_sum(tf.pow(tf.subtract(self.reconstruction, self.x), 2.0))
		# 定义优化器
		self.optimizer = optimizer.minimize(self.cost)
		init = tf.global_variables_initializer()
		self.sess = tf.Session()
		self.sess.run(init)

	# 参数初始化函数
	def _initialize_weights(self):
		all_weights = {}
		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

	# 定义训练函数
	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

	# 求损失函数的函数 在训练完成后在测试集上你你你你你你你你你对模型性能进行评测
	def calc_total_cost(self, X):
		return self.sess.run(self.cost, feed_dict = {self.x:X, self.scale:self.training_scale})

	# transform函数 返回隐含层的输出结果 学习出的数据中的高阶特征
	def transform(self, X):
		return self.sess.run(self.hidden, feed_dict = {self.x:X, self.scale:self.training_scale})

	# 将隐含层的输出结果作为输入 通过之后重建层将高阶特征复原为原始数据
	def generate(self, hidden=None):
		if hidden==None:
			hidden = np.random.normal(size=self.weights['b1'])

		return self.sess.run(self.reconstruction, feed_dict = {self.hidden:hidden})

	# reconstruct函数 整体运行一遍复原过程 包括提取高阶特征和通过高阶特征复原数据
	def reconstruct(self, X):
		return self.sess.run(self.reconstruction, feed_dict = {self.x:X, self.scale:self.training_scale})

	# 获取权重w1
	def getWeights(self):
		return self.sess.run(self.weights['w1'])
		
	def getBiases(self):
		return self.sess.run(self.weights['b1'])


# 对训练、测试数据进行标准化处理 让数据变成0均值 标准差为1的分布
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

#获取随机block数据 不放回抽样 取一个0--len(data)-batch_size之间的整数作为block的起始位置
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)]

# 数据标准化处理
X_train, X_test = standard_scale(mnist.train.images, mnist.test.images)

n_samples = int(mnist.train.num_examples)
training_epochs = 20
batch_size = 128
display_step = 1

autoencoder = AdditiveGaussianNoiseAutoencoder(n_input = 784, n_hidden=200,
												transfer_function=tf.nn.softplus,
												optimizer = tf.train.AdamOptimizer(learning_rate=0.001),
												scale=0.01)

for epoch in range(training_epochs):
	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))

print('Total cost: ' + str(autoencoder.calc_total_cost(X_test)))

自编码器是一种无监督学习的方法，目的在于提取数据中最有用、最频繁出现的高阶特征，根据这些特征重构数据

内容源自《Tensorflow实战》