用Tensorflow实现卷积神经网络(CNN)
本文系作者原创，转载请注明出处:https://www.cnblogs.com/further-further-further/p/10737065.html
目录
1.踩过的坑(tensorflow)
2.tensorboard
3.代码实现（python3.5）
4.运行结果以及分析
1.踩过的坑(tensorflow)
上一章CNN中各个算法都是纯手工实现的，可能存在一些难以发现的问题，这也是准确率不高的一个原因，这章主要利用tensorflow框架来实现卷积神经网络，数据源还是cifar（具体下载见上一章）
在利用tensorflow框架实现CNN时，需要注意以下几点：
1.输入数据定义时，x只是起到占位符的作用（看不到真实值，只是为了能够运行代码，获取相应的tensor节点，这一点跟我们之前代码流程完全相反， 真正数据流的执行在session会话里）
x：输入数据，y_： 标签数据，keep_prob: 概率因子，防止过拟合。
定义，且是全局变量。
代码语言：javascript复制x = tf.placeholder(tf.float32, [None, 3072], name='x') 后面在session里必须要初始化
代码语言：javascript复制sess.run(tf.global_variables_initializer())在session run时必须要传得到该tensor节点含有参数值（x, y_, keep_prob）
代码语言：javascript复制train_accuracy = accuracy.eval(feed_dict={
x: batch[0], y_: batch[1], keep_prob: 1.0})2.原始数据集标签要向量化；
例如cifar有10个类别，如果类别标签是 6 对应向量[0,0,0,0,0,1,0,0,0,0]
3.知道每一步操作的数据大小的变化，不然，报错的时候很难定位（个人认为这也是tensorflow的弊端，无法实时追踪定位）；
注意padding = 'SAME'和'VALID'的区别
padding = 'SAME' => Height_后 = Height_前/Strides 跟padding无关 向上取整
padding = 'VALID'=> Height_后 = (Height_前 - Filter 1)/Strides 向上取整
4.打印tensorboard流程图，可以直观看到每步操作数据大小的变化；
2. tensorboard
tensorboard就是一个数据结构流程图的可视化工具，通过tensorboard流程图，可以直观看到神经网络的每一步操作以及数据流的变化。
操作步骤：
1. 在session会话里加入如下代码，打印结果会在当前代码文件相同路径的tensorboard文件下，默认是
代码语言：javascript复制tf.summary.FileWriter("tensorboard/", sess.graph)2. 在运行里输入cmd，然后输入（前提是安装好了tensorboard => pip install tensorboard）
代码语言：javascript复制tensorboard --logdir=D:ProjectpythonmyProjectCNNtensorflowcaptchaIdentifytensorboard --host=127.0.0.1'D:ProjectpythonmyProjectCNNtensorflowcaptchaIdentifytensorboard' 是我生成的tensorboard文件的绝对路径，你替换成你自己的就可以了。
正确运行后会显示 ‘Tensorboard at http://127.0.0.1:6006’，说明tensorboard服务已经起来了，在浏览器页面输入
http://127.0.0.1:6006即可显示流程图。
3.代码实现（python3.6）
代码逻辑实现相对比较简单，在一些重要逻辑实现上，我已做了注释，如果大家有什么疑义，可以留言给我，我们一起交流。
因为原始图片数据集太大，不好上传，大家可以直接在http://www.cs.toronto.edu/~kriz/cifar.html下载CIFAR-10 python version，
有163M，放在代码文件同路径下即可。
cifar放置路径
start.py
代码语言：javascript复制 1 # coding=utf-8
2 # Disable linter warnings to maintain consistency with tutorial.
3 # pylint: disable=invalid-name
4 # pylint: disable=g-bad-import-order
5 from __future__ import absolute_import
6 from __future__ import division
7 from __future__ import print_function
8 import argparse
9 import sys
10 import tempfile
11 #from tensorflow.examples.tutorials.mnist import input_data
12 import tensorflow as tf
13 '''
14 卷积神经网络实现10类(airplane, automobile, bird, cat, deer, dog, frog, horse, ship, truck)
15 60000张图片的识别
16 5000次，准确率有 58%；
17 20000次，准确率有 68.89%；
18 相比mnist数字图片识别准确度低，原因有：
19 mnist训练图片是灰度图片，纹理简单，数字的可变性小，而cifar是彩色图片，纹理复杂，动物可变性大；
20 '''
21 try:
22 from . import datesets
23 except Exception:
24 import datesets
25
26 FLAGS = None
27
28 def deepnn(x):
29 with tf.name_scope('reshape'):
30 x_image = tf.reshape(x, [-1, 32, 32, 3])
31 ## 第一层卷积操作 ##
32 with tf.name_scope('conv1'):
33 W_conv1 = weight_variable([5, 5, 3, 32])
34 b_conv1 = bias_variable([32])
35 h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) b_conv1)
36
37 with tf.name_scope('pool1'):
38 h_pool1 = max_pool_2x2(h_conv1)
39
40 # Second convolutional layer -- maps 32 feature maps to 64.
41 ## 第二层卷积操作 ##
42 with tf.name_scope('conv2'):
43 W_conv2 = weight_variable([5, 5, 32, 64])
44 b_conv2 = bias_variable([64])
45 h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) b_conv2)
46
47 with tf.name_scope('pool2'):
48 h_pool2 = max_pool_2x2(h_conv2)
49
50 ## 第三层全连接操作 ##
51 with tf.name_scope('fc1'):
52 W_fc1 = weight_variable([8 * 8 * 64, 1024])
53 b_fc1 = bias_variable([1024])
54 h_pool2_flat = tf.reshape(h_pool2, [-1, 8 * 8 * 64])
55 h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) b_fc1)
56
57 with tf.name_scope('dropout'):
58 keep_prob = tf.placeholder(tf.float32)
59 h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)
60
61 ## 第四层输出操作 ##
62 with tf.name_scope('fc2'):
63 W_fc2 = weight_variable([1024, 10])
64 b_fc2 = bias_variable([10])
65 y_conv = tf.matmul(h_fc1_drop, W_fc2) b_fc2
66 return y_conv, keep_prob
67
68 def conv2d(x, W):
69 return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding='SAME')
70
71 def max_pool_2x2(x):
72 return tf.nn.max_pool(x, ksize=[1, 2, 2, 1],
73 strides=[1, 2, 2, 1], padding='SAME')
74
75 def weight_variable(shape):
76 initial = tf.truncated_normal(shape, stddev=0.1)
77 return tf.Variable(initial)
78
79 def bias_variable(shape):
80 initial = tf.constant(0.1, shape=shape)
81 return tf.Variable(initial)
82
83 def main(_):
84 # Import data
85 mnist = datesets.read_data_sets(train_dir = '.\cifar-10-batches-py\', one_hot=True)
86
87 # Create the model
88 # 声明一个占位符，None表示输入图片的数量不定，28*28图片分辨率
89 x = tf.placeholder(tf.float32, [None, 3072], name='x')
90
91 # 类别是0-9总共10个类别，对应输出分类结果
92 y_ = tf.placeholder(tf.float32, [None, 10], name='y_')
93 y_conv, keep_prob = deepnn(x)
94 # 通过softmax-loss求交叉熵
95 with tf.name_scope('loss'):
96 cross_entropy = tf.nn.softmax_cross_entropy_with_logits(labels=y_, logits=y_conv)
97 # 求均值
98 cross_entropy = tf.reduce_mean(cross_entropy)
99 # 计算梯度，更新参数值
100 with tf.name_scope('adam_optimizer'):
101 train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
102
103 with tf.name_scope('accuracy'):
104 correct_prediction = tf.equal(tf.argmax(y_conv, 1), tf.argmax(y_, 1))
105 correct_prediction = tf.cast(correct_prediction, tf.float32)
106 accuracy = tf.reduce_mean(correct_prediction)
107
108 # graph_location = tempfile.mkdtemp()
109 # print('Saving graph to: %s' % graph_location)
110 # train_writer.add_graph(tf.get_default_graph())
111
112 with tf.Session() as sess:
113 # 打印流程图
114 writer = tf.summary.FileWriter("tensorboard/", sess.graph)
115 sess.run(tf.global_variables_initializer())
116 for i in range(20000):
117 batch = mnist.train.next_batch(50)
118 if i % 1000 == 0:
119 train_accuracy = accuracy.eval(feed_dict={
120 x: batch[0], y_: batch[1], keep_prob: 1.0})
121 print('step %d, training accuracy %g' % (i, train_accuracy))
122 train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
123
124 print('test accuracy %g' % accuracy.eval(feed_dict={
125 x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0}))
126
127 if __name__ == '__main__':
128 parser = argparse.ArgumentParser()
129 parser.add_argument('--data_dir', type=str,
130 default='/tmp/tensorflow/mnist/input_data',
131 help='Directory for storing input data')
132 FLAGS, unparsed = parser.parse_known_args()
133 tf.app.run(main=main, argv=[sys.argv[0]] unparsed)datasets.py
代码语言：javascript复制 1 import numpy
2 from tensorflow.python.framework import dtypes
3 from tensorflow.python.framework import random_seed
4 from six.moves import xrange
5 from tensorflow.contrib.learn.python.learn.datasets import base
6 import pickle
7 import os
8
9 class DataSet(object):
10 """Container class for a dataset (deprecated).
11
12 THIS CLASS IS DEPRECATED. See
13 [contrib/learn/README.md](https://www.tensorflow.org/code/tensorflow/contrib/learn/README.md)
14 for general migration instructions.
15 """
16 def __init__(self,
17 images,
18 labels,
19 fake_data=False,
20 one_hot=False,
21 dtype=dtypes.float32,
22 reshape=True,
23 seed=None):
24 """Construct a DataSet.
25 one_hot arg is used only if fake_data is true. `dtype` can be either
26 `uint8` to leave the input as `[0, 255]`, or `float32` to rescale into
27 `[0, 1]`. Seed arg provides for convenient deterministic testing.
28 """
29 seed1, seed2 = random_seed.get_seed(seed)
30 # If op level seed is not set, use whatever graph level seed is returned
31 numpy.random.seed(seed1 if seed is None else seed2)
32 dtype = dtypes.as_dtype(dtype).base_dtype
33 if dtype not in (dtypes.uint8, dtypes.float32):
34 raise TypeError(
35 'Invalid image dtype %r, expected uint8 or float32' % dtype)
36 if fake_data:
37 self._num_examples = 10000
38 self.one_hot = one_hot
39 else:
40 assert images.shape[0] == labels.shape[0], (
41 'images.shape: %s labels.shape: %s' % (images.shape, labels.shape))
42 self._num_examples = images.shape[0]
43
44 # Convert shape from [num examples, rows, columns, depth]
45 # to [num examples, rows*columns] (assuming depth == 1)
46 if reshape:
47 assert images.shape[3] == 3
48 images = images.reshape(images.shape[0],
49 images.shape[1] * images.shape[2] * images.shape[3])
50 if dtype == dtypes.float32:
51 # Convert from [0, 255] -> [0.0, 1.0].
52 images = images.astype(numpy.float32)
53 images = numpy.multiply(images, 1.0 / 255.0)
54 self._images = images
55 self._labels = labels
56 self._epochs_completed = 0
57 self._index_in_epoch = 0
58
59 @property
60 def images(self):
61 return self._images
62
63 @property
64 def labels(self):
65 return self._labels
66
67 @property
68 def num_examples(self):
69 return self._num_examples
70
71 @property
72 def epochs_completed(self):
73 return self._epochs_completed
74
75 def next_batch(self, batch_size, fake_data=False, shuffle=True):
76 """Return the next `batch_size` examples from this data set."""
77 if fake_data:
78 fake_image = [1] * 784
79 if self.one_hot:
80 fake_label = [1] [0] * 9
81 else:
82 fake_label = 0
83 return [fake_image for _ in xrange(batch_size)], [
84 fake_label for _ in xrange(batch_size)
85 ]
86 start = self._index_in_epoch
87 # Shuffle for the first epoch
88 if self._epochs_completed == 0 and start == 0 and shuffle:
89 perm0 = numpy.arange(self._num_examples)
90 numpy.random.shuffle(perm0)
91 self._images = self.images[perm0]
92 self._labels = self.labels[perm0]
93 # Go to the next epoch
94 if start batch_size > self._num_examples:
95 # Finished epoch
96 self._epochs_completed = 1
97 # Get the rest examples in this epoch
98 rest_num_examples = self._num_examples - start
99 images_rest_part = self._images[start:self._num_examples]
100 labels_rest_part = self._labels[start:self._num_examples]
101 # Shuffle the data
102 if shuffle:
103 perm = numpy.arange(self._num_examples)
104 numpy.random.shuffle(perm)
105 self._images = self.images[perm]
106 self._labels = self.labels[perm]
107 # Start next epoch
108 start = 0
109 self._index_in_epoch = batch_size - rest_num_examples
110 end = self._index_in_epoch
111 images_new_part = self._images[start:end]
112 labels_new_part = self._labels[start:end]
113 return numpy.concatenate(
114 (images_rest_part, images_new_part), axis=0), numpy.concatenate(
115 (labels_rest_part, labels_new_part), axis=0)
116 else:
117 self._index_in_epoch = batch_size
118 end = self._index_in_epoch
119 return self._images[start:end], self._labels[start:end]
120
121 def read_data_sets(train_dir,
122 one_hot=False,
123 dtype=dtypes.float32,
124 reshape=True,
125 validation_size=5000,
126 seed=None):
127
128
129
130
131 train_images,train_labels,test_images,test_labels = load_CIFAR10(train_dir)
132 if not 0 <= validation_size <= len(train_images):
133 raise ValueError('Validation size should be between 0 and {}. Received: {}.'
134 .format(len(train_images), validation_size))
135
136 validation_images = train_images[:validation_size]
137 validation_labels = train_labels[:validation_size]
138 validation_labels = dense_to_one_hot(validation_labels, 10)
139 train_images = train_images[validation_size:]
140 train_labels = train_labels[validation_size:]
141 train_labels = dense_to_one_hot(train_labels, 10)
142
143 test_labels = dense_to_one_hot(test_labels, 10)
144
145 options = dict(dtype=dtype, reshape=reshape, seed=seed)
146 train = DataSet(train_images, train_labels, **options)
147 validation = DataSet(validation_images, validation_labels, **options)
148 test = DataSet(test_images, test_labels, **options)
149
150 return base.Datasets(train=train, validation=validation, test=test)
151
152
153 def load_CIFAR_batch(filename):
154 """ load single batch of cifar """
155 with open(filename, 'rb') as f:
156 datadict = pickle.load(f, encoding='bytes')
157 X = datadict[b'data']
158 Y = datadict[b'labels']
159 X = X.reshape(10000, 3, 32, 32).transpose(0,2,3,1).astype("float")
160 Y = numpy.array(Y)
161 return X, Y
162
163 def load_CIFAR10(ROOT):
164 """ load all of cifar """
165 xs = []
166 ys = []
167 for b in range(1,6):
168 f = os.path.join(ROOT, 'data_batch_%d' % (b, ))
169 X, Y = load_CIFAR_batch(f)
170 xs.append(X)
171 ys.append(Y)
172 Xtr = numpy.concatenate(xs)
173 Ytr = numpy.concatenate(ys)
174 del X, Y
175 Xte, Yte = load_CIFAR_batch(os.path.join(ROOT, 'test_batch'))
176 return Xtr, Ytr, Xte, Yte
177
178 def dense_to_one_hot(labels_dense, num_classes):
179 """Convert class labels from scalars to one-hot vectors."""
180 num_labels = labels_dense.shape[0]
181 index_offset = numpy.arange(num_labels) * num_classes
182 labels_one_hot = numpy.zeros((num_labels, num_classes))
183 labels_one_hot.flat[index_offset labels_dense.ravel()] = 1
184 return labels_one_hot4.运行结果以及分析
这里选取55000张图片作为训练样本，测试样本选取5000张。
tensorboard可视流程图
运行5000次，测试准确率：58%
运行20000次，测试准确率：68.89%
运行40000次，测试准确率71.95%
分析：由最后一张图片可以看出，20000 - 30000次时测试准确率=> 70.27% ->71.44%，30000 - 40000次时=> 71.44% -> 71.95%
而训练准确率已经达到100%，说明测试准确率已经趋于一个稳定值，再增加训练次数，测试准确率提高的可能性不大。
如果想要继续提高测试准确率，就只能增加训练样本。
不要让懒惰占据你的大脑，不要让妥协拖垮了你的人生。青春就是一张票，能不能赶上时代的快车，你的步伐就掌握在你的脚下。
