深度学习之路--Mnist手写数字识别问题分析

 对于每一个初学深度学习的小伙伴, Mnist手写数字识别问题非常经典, tensorflow官网教程首先就介绍了分别使用单隐层网络(准确率约91%)和多层卷积网络(准确率约99.2%)来解决这个问题. luffy也入了深度学习的坑, 刚刚依照官网步骤分别实现了两种方法, 源代码见github, 对于卷积神经网络方法有种似懂非懂的感觉(其实就是不懂(>_<)), 所以写此博客, 一遍分析, 一遍记录心得.
 Mnist问题, 在我看来属于"麻雀虽小五脏俱全", 很有分析的价值.下面进行一一梳理.

基本流程

数据处理

构建模型

损失函数

训练模型

测试模型

附上源代码

'''
构建一个多层卷积网络(输入层+卷积池化x2+全连接层x2)
参照Tensorflow中文社区-MNIST进阶-编写
'''
#encoding=utf-8
import tensorflow as tf 
import input_data
#interactivesession, 允许tensorflow在运行图的时候, 插入一些计算图, 不然只能先构建整个计算图, 然后启动图, 可对比Mnist_Primary的做法
sess = tf.InteractiveSession()

#数据, 如果比赛的话需要自己处理
mnist = input_data.read_data_sets("MNIST_data", one_hot=True)
###################### 构建计算图###############
#占位符, 运行session时填入具体的值
x = tf.placeholder("float", shape = [None, 784])
y_ = tf.placeholder("float", shape = [None, 10])           #真实值, 用于计算交叉熵

#权重初始化
def weight_variable(shape):
    initial = tf.truncated_normal(shape, stddev=0.1)  #用截断的正态分布初始化tensor, 标准差0.1
    return tf.Variable(initial)
def bias_variable(shape):
    initial = tf.constant(0.1, shape=shape)           #使用常数0.1初始化tensor
    return tf.Variable(initial)
#卷积和池化
def conv2d(x, W):
    return tf.nn.conv2d(x, W, strides=[1,1,1,1], padding='SAME')
def max_pool_2x2(x):
    return tf.nn.max_pool(x, ksize=[1,2,2,1], strides=[1,2,2,1], padding='SAME')
#第一层卷积
W_conv1 = weight_variable([5,5,1,32])       #patch大小5x5, 输入通道数1, 输出通道数32
b_conv1 = bias_variable([32])               #每一个输出通道对应一个偏置量
x_image = tf.reshape(x, [-1, 28, 28, 1])    #第二 三维代表图像宽高, 第四维代表通道数, 灰度图为1
h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)    #卷积并整流
h_pool1 = max_pool_2x2(h_conv1)              #池化
#第二层卷积
W_conv2 = weight_variable([5,5,32,64])      #不是很明白第三 四维为什么是32和64
b_conv2 = bias_variable([64])

h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)
h_pool2 = max_pool_2x2(h_conv2)
#密集连接层
W_fc1 = weight_variable([7*7*64, 1024])     #1024个神经元的全连接层
b_fc1 = bias_variable([1024])
h_pool2_flat = tf.reshape(h_pool2, [-1 ,7*7*64])
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)
#Dropout, 防止过拟合
keep_prob = tf.placeholder("float")
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)
#输出层
W_fc2 = weight_variable([1024, 10])
b_fc2 = bias_variable([10])
y_conv = tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2) + b_fc2)
#训练和评估模型
cross_entropy = -tf.reduce_sum(y_ * tf.log(y_conv))                #定义交叉熵
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)  #ADAM优化器来做梯度最速下降
correct_prediction = tf.equal(tf.argmax(y_conv, 1), tf.argmax(y_, 1))  #对比真实值和预测值
accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))    #求准确率
sess.run(tf.global_variables_initializer())
for i in range(20000):
    batch = mnist.train.next_batch(50)        #每次随机抽取50个样本进行训练
    if i%100 == 0:
        train_accuracy = accuracy.eval(feed_dict={x: batch[0], y_:batch[1], keep_prob: 1.0})#训练准确率
        print("step%d, training accuracy %g"%(i, train_accuracy))       
    train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
print("test accuracy %g"%accuracy.eval(feed_dict={x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0}))#测试准确率

input_data源码

# Copyright 2015 Google Inc. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Functions for downloading and reading MNIST data."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import gzip
import os
import numpy
from six.moves import urllib
from six.moves import xrange  # pylint: disable=redefined-builtin
SOURCE_URL = 'http://yann.lecun.com/exdb/mnist/'
def maybe_download(filename, work_directory):
  """Download the data from Yann's website, unless it's already here."""
  if not os.path.exists(work_directory):
    os.mkdir(work_directory)
  filepath = os.path.join(work_directory, filename)
  if not os.path.exists(filepath):
    filepath, _ = urllib.request.urlretrieve(SOURCE_URL + filename, filepath)
    statinfo = os.stat(filepath)
    print('Successfully downloaded', filename, statinfo.st_size, 'bytes.')
  return filepath
 
def _read32(bytestream):
  #采用大尾端存储
  dt = numpy.dtype(numpy.uint32).newbyteorder('>')
  return numpy.frombuffer(bytestream.read(4), dtype=dt)[0]
#提取图片到四维uint8数组
def extract_images(filename):
  """Extract the images into a 4D uint8 numpy array [index, y, x, depth]."""
  print('Extracting', filename)
  with gzip.open(filename) as bytestream:
    magic = _read32(bytestream)
    if magic != 2051:
      raise ValueError(
          'Invalid magic number %d in MNIST image file: %s' %
          (magic, filename))
    num_images = _read32(bytestream)   
    rows = _read32(bytestream)
    cols = _read32(bytestream)
    buf = bytestream.read(rows * cols * num_images) 
    data = numpy.frombuffer(buf, dtype=numpy.uint8)
    data = data.reshape(num_images, rows, cols, 1)
    return data
#将稠密标签向量变成稀疏的标签矩阵
#eg：若原向量的第i行为3，则对应稀疏矩阵的第i行下标为3的值为1，其余为0
def dense_to_one_hot(labels_dense, num_classes=10):
  """Convert class labels from scalars to one-hot vectors."""
  #ndarray.shape展示数组的维度，shape[0]展示行
  num_labels = labels_dense.shape[0]
  #numpy.arange(start,step,stop)创建始于start，终于stop，步长为step的等差数列
  index_offset = numpy.arange(num_labels) * num_classes
  labels_one_hot = numpy.zeros((num_labels, num_classes))
  #numpy.darray.flat将数组变换成一维;numpy.ravel()返回视图
  labels_one_hot.flat[index_offset + labels_dense.ravel()] = 1
  return labels_one_hot
def extract_labels(filename, one_hot=False):
  """Extract the labels into a 1D uint8 numpy array [index]."""
  print('Extracting', filename)
  with gzip.open(filename) as bytestream:
    magic = _read32(bytestream)
    if magic != 2049:
      raise ValueError(
          'Invalid magic number %d in MNIST label file: %s' %
          (magic, filename))
    num_items = _read32(bytestream)
    buf = bytestream.read(num_items)
    labels = numpy.frombuffer(buf, dtype=numpy.uint8)
    if one_hot:
      return dense_to_one_hot(labels)
    return labels
class DataSet(object):
  def __init__(self, images, labels, fake_data=False):
    if fake_data:
      self._num_examples = 10000
    else:
      assert images.shape[0] == labels.shape[0], (
          "images.shape: %s labels.shape: %s" % (images.shape,
                                                 labels.shape))
      self._num_examples = images.shape[0]
      # Convert shape from [num examples, rows, columns, depth]
      # to [num examples, rows*columns] (assuming depth == 1)
      assert images.shape[3] == 1
      images = images.reshape(images.shape[0],
                              images.shape[1] * images.shape[2])
      # Convert from [0, 255] -> [0.0, 1.0].
      images = images.astype(numpy.float32)
      images = numpy.multiply(images, 1.0 / 255.0)
    self._images = images
    self._labels = labels
    self._epochs_completed = 0
    self._index_in_epoch = 0
    
  @property
  def images(self):
    return self._images
  @property
  def labels(self):
    return self._labels
  @property
  def num_examples(self):
    return self._num_examples
  @property
  def epochs_completed(self):
    return self._epochs_completed
  def next_batch(self, batch_size, fake_data=False):
    """Return the next `batch_size` examples from this data set."""
    if fake_data:
      fake_image = [1.0 for _ in xrange(784)]
      fake_label = 0
      return [fake_image for _ in xrange(batch_size)], [
          fake_label for _ in xrange(batch_size)]
    start = self._index_in_epoch
    self._index_in_epoch += batch_size
    if self._index_in_epoch > self._num_examples:
      # Finished epoch
      self._epochs_completed += 1
      # Shuffle the data
      perm = numpy.arange(self._num_examples)
      numpy.random.shuffle(perm)
      self._images = self._images[perm]
      self._labels = self._labels[perm]
      # Start next epoch
      start = 0
      self._index_in_epoch = batch_size
      assert batch_size <= self._num_examples
    end = self._index_in_epoch
    return self._images[start:end], self._labels[start:end]
def read_data_sets(train_dir, fake_data=False, one_hot=False):
  class DataSets(object):
    #python里pass用来占位
    pass
  data_sets = DataSets()
  if fake_data:
    data_sets.train = DataSet([], [], fake_data=True)
    data_sets.validation = DataSet([], [], fake_data=True)
    data_sets.test = DataSet([], [], fake_data=True)
    return data_sets
  TRAIN_IMAGES = 'train-images-idx3-ubyte.gz'
  TRAIN_LABELS = 'train-labels-idx1-ubyte.gz'
  TEST_IMAGES = 't10k-images-idx3-ubyte.gz'
  TEST_LABELS = 't10k-labels-idx1-ubyte.gz'
  VALIDATION_SIZE = 5000
  local_file = maybe_download(TRAIN_IMAGES, train_dir)
  train_images = extract_images(local_file)
  local_file = maybe_download(TRAIN_LABELS, train_dir)
  train_labels = extract_labels(local_file, one_hot=one_hot)
  local_file = maybe_download(TEST_IMAGES, train_dir)
  test_images = extract_images(local_file)
  local_file = maybe_download(TEST_LABELS, train_dir)
  test_labels = extract_labels(local_file, one_hot=one_hot)
  validation_images = train_images[:VALIDATION_SIZE]
  validation_labels = train_labels[:VALIDATION_SIZE]
  train_images = train_images[VALIDATION_SIZE:]
  train_labels = train_labels[VALIDATION_SIZE:]
  data_sets.train = DataSet(train_images, train_labels)
  data_sets.validation = DataSet(validation_images, validation_labels)
  data_sets.test = DataSet(test_images, test_labels)
  return data_sets