TensorFlow实战(一)是用了只有一个隐藏层神经网络实现的手写数字识别,虽然正确率比较高,但是还没有涉及到最常用的卷积神经网络。这次我们试试使用卷积神经网络做手写数字识别。
'''权重初始化函数'''
def weight_variable(shape):
inital = tf.truncated_normal(shape, stddev=0.1) # 使用truncated_normal进行初始化
return tf.Variable(inital)
'''偏置初始化函数'''
def bias_variable(shape):
inital = tf.constant(0.1,shape=shape) # 偏置定义为常量
return tf.Variable(inital)
'''卷积函数'''
def conv2d(x,W):#x是图片的所有参数,W是此卷积层的权重
return tf.nn.conv2d(x,W,strides=[1,1,1,1],padding='SAME')#strides[0]和strides[3]的两个1是默认值,中间两个1代表padding时在x方向运动1步,y方向运动1步
'''池化函数'''
def max_pool_2x2(x):
return tf.nn.max_pool(x,ksize=[1,2,2,1],
strides=[1,2,2,1], padding='SAME')#池化的核函数大小为2x2,因此ksize=[1,2,2,1],步长为2,因此strides=[1,2,2,1]
mnist = input_data.read_data_sets('MNIST_data', one_hot=True) # 下载数据
xs = tf.placeholder(tf.float32,[None,784]) # 输入图片的大小,28x28=784
ys = tf.placeholder(tf.float32,[None,10]) # 输出0-9共10个数字
keep_prob = tf.placeholder(tf.float32) # 用于接收dropout操作的值,dropout为了防止过拟合
x_image = tf.reshape(xs,[-1,28,28,1]) #-1代表先不考虑输入的图片例子多少这个维度,后面的1是channel的数量,因为我们输入的图片是黑白的,因此channel是1,例如如果是RGB图像,那么channel就是3
'''第一层卷积,池化'''
W_conv1 = weight_variable([5,5,1,32]) # 卷积核定义为5x5,1是输入的通道数目,32是输出的通道数目
b_conv1 = bias_variable([32]) # 每个输出通道对应一个偏置
h_conv1 = tf.nn.relu(conv2d(x_image,W_conv1)+b_conv1) # 卷积运算,并使用ReLu激活函数激活
h_pool1 = max_pool_2x2(h_conv1) # pooling操作
'''第二层卷积,池化'''
W_conv2 = weight_variable([5,5,32,64]) # 卷积核还是5x5,32个输入通道,64个输出通道
b_conv2 = bias_variable([64]) # 与输出通道一致
h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2)+b_conv2)
h_pool2 = max_pool_2x2(h_conv2)
'''全连接层'''
h_pool2_flat = tf.reshape(h_pool2, [-1,7*7*64]) # 将最后操作的数据展开
W_fc1 = weight_variable([7*7*64,1024]) # 下面就是定义一般神经网络的操作了,继续扩大为1024
b_fc1 = bias_variable([1024]) # 对应的偏置
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat,W_fc1)+b_fc1) # 运算、激活(这里不是卷积运算了,就是对应相乘)
'''dropout'''
h_fc1_drop = tf.nn.dropout(h_fc1,keep_prob) # dropout操作
'''最后一层全连接'''
W_fc2 = weight_variable([1024,10]) # 最后一层权重初始化
b_fc2 = bias_variable([10]) # 对应偏置
prediction = tf.nn.softmax(tf.matmul(h_fc1_drop,W_fc2)+b_fc2) # 使用softmax分类器
cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys*tf.log(prediction),reduction_indices=[1])) # 交叉熵损失函数来定义cost function
train_step = tf.train.AdamOptimizer(1e-3).minimize(cross_entropy) # 调用梯度下降
'''下面就是tf的一般操作,定义Session,初始化所有变量,placeholder传入值训练'''
sess = tf.Session()
sess.run(tf.initialize_all_variables())
for i in range(1000):
batch_xs, batch_ys = mnist.train.next_batch(100) # 使用SGD,每次选取100个数据训练
sess.run(train_step, feed_dict={xs: batch_xs, ys: batch_ys, keep_prob: 0.5}) # dropout值定义为0.5
if i % 50 == 0:
print compute_accuracy(xs,ys,mnist.test.images, mnist.test.labels,keep_prob,sess,prediction) # 每50次输出一下准确度
'''计算准确度函数'''
def compute_accuracy(xs,ys,X,y,keep_prob,sess,prediction):
y_pre = sess.run(prediction,feed_dict={xs:X,keep_prob:1.0}) # 预测,这里的keep_prob是dropout时用的,防止过拟合
correct_prediction = tf.equal(tf.argmax(y_pre,1),tf.argmax(y,1)) #tf.argmax 给出某个tensor对象在某一维上的其数据最大值所在的索引值,即为对应的数字,tf.equal 来检测我们的预测是否真实标签匹配
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32)) # 平均值即为准确度
result = sess.run(accuracy,feed_dict={xs:X,ys:y,keep_prob:1.0})
return result
import tensorflow as tf
import numpy as np
from tensorflow.examples.tutorials.mnist import input_data # 导入mnist数据集
'''计算准确度函数'''
def compute_accuracy(xs,ys,X,y,keep_prob,sess,prediction):
y_pre = sess.run(prediction,feed_dict={xs:X,keep_prob:1.0}) # 预测,这里的keep_prob是dropout时用的,防止过拟合
correct_prediction = tf.equal(tf.argmax(y_pre,1),tf.argmax(y,1)) #tf.argmax 给出某个tensor对象在某一维上的其数据最大值所在的索引值,即为对应的数字,tf.equal 来检测我们的预测是否真实标签匹配
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32)) # 平均值即为准确度
result = sess.run(accuracy,feed_dict={xs:X,ys:y,keep_prob:1.0})
return result
'''权重初始化函数'''
def weight_variable(shape):
inital = tf.truncated_normal(shape, stddev=0.1) # 使用truncated_normal进行初始化
return tf.Variable(inital)
'''偏置初始化函数'''
def bias_variable(shape):
inital = tf.constant(0.1,shape=shape) # 偏置定义为常量
return tf.Variable(inital)
'''卷积函数'''
def conv2d(x,W):#x是图片的所有参数,W是此卷积层的权重
return tf.nn.conv2d(x,W,strides=[1,1,1,1],padding='SAME')#strides[0]和strides[3]的两个1是默认值,中间两个1代表padding时在x方向运动1步,y方向运动1步
'''池化函数'''
def max_pool_2x2(x):
return tf.nn.max_pool(x,ksize=[1,2,2,1],
strides=[1,2,2,1],
padding='SAME')#池化的核函数大小为2x2,因此ksize=[1,2,2,1],步长为2,因此strides=[1,2,2,1]
'''运行主函数'''
def cnn():
mnist = input_data.read_data_sets('MNIST_data', one_hot=True) # 下载数据
xs = tf.placeholder(tf.float32,[None,784]) # 输入图片的大小,28x28=784
ys = tf.placeholder(tf.float32,[None,10]) # 输出0-9共10个数字
keep_prob = tf.placeholder(tf.float32) # 用于接收dropout操作的值,dropout为了防止过拟合
x_image = tf.reshape(xs,[-1,28,28,1]) #-1代表先不考虑输入的图片例子多少这个维度,后面的1是channel的数量,因为我们输入的图片是黑白的,因此channel是1,例如如果是RGB图像,那么channel就是3
'''第一层卷积,池化'''
W_conv1 = weight_variable([5,5,1,32]) # 卷积核定义为5x5,1是输入的通道数目,32是输出的通道数目
b_conv1 = bias_variable([32]) # 每个输出通道对应一个偏置
h_conv1 = tf.nn.relu(conv2d(x_image,W_conv1)+b_conv1) # 卷积运算,并使用ReLu激活函数激活
h_pool1 = max_pool_2x2(h_conv1) # pooling操作
'''第二层卷积,池化'''
W_conv2 = weight_variable([5,5,32,64]) # 卷积核还是5x5,32个输入通道,64个输出通道
b_conv2 = bias_variable([64]) # 与输出通道一致
h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2)+b_conv2)
h_pool2 = max_pool_2x2(h_conv2)
'''全连接层'''
h_pool2_flat = tf.reshape(h_pool2, [-1,7*7*64]) # 将最后操作的数据展开
W_fc1 = weight_variable([7*7*64,1024]) # 下面就是定义一般神经网络的操作了,继续扩大为1024
b_fc1 = bias_variable([1024]) # 对应的偏置
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat,W_fc1)+b_fc1) # 运算、激活(这里不是卷积运算了,就是对应相乘)
'''dropout'''
h_fc1_drop = tf.nn.dropout(h_fc1,keep_prob) # dropout操作
'''最后一层全连接'''
W_fc2 = weight_variable([1024,10]) # 最后一层权重初始化
b_fc2 = bias_variable([10]) # 对应偏置
prediction = tf.nn.softmax(tf.matmul(h_fc1_drop,W_fc2)+b_fc2) # 使用softmax分类器
cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys*tf.log(prediction),reduction_indices=[1])) # 交叉熵损失函数来定义cost function
train_step = tf.train.AdamOptimizer(1e-3).minimize(cross_entropy) # 调用梯度下降
'''下面就是tf的一般操作,定义Session,初始化所有变量,placeholder传入值训练'''
sess = tf.Session()
sess.run(tf.initialize_all_variables())
for i in range(1000):
batch_xs, batch_ys = mnist.train.next_batch(100) # 使用SGD,每次选取100个数据训练
sess.run(train_step, feed_dict={xs: batch_xs, ys: batch_ys, keep_prob: 0.5}) # dropout值定义为0.5
if i % 50 == 0:
print(compute_accuracy(xs,ys,mnist.test.images, mnist.test.labels,keep_prob,sess,prediction)) # 每50次输出一下准确度
if __name__ == '__main__':
cnn()
# Copyright 2015 The TensorFlow Authors. 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.
# ==============================================================================
"""A very simple MNIST classifier.
See extensive documentation at
http://tensorflow.org/tutorials/mnist/beginners/index.md
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
# import sys
# sys.path.append("/home/kklv/ML")
from tensorflow.examples.tutorials.mnist import input_data
# from tensorflow.contrib.learn.python.learn.datasets.mnist import read_data_sets
SOURCE_URL = 'http://yann.lecun.com/exdb/mnist/'
import tensorflow as tf
flags = tf.app.flags
FLAGS = flags.FLAGS
flags.DEFINE_string('data_dir', '/tmp/data/', 'Directory for storing data') # 第一次启动会下载文本资料,放在/tmp/data文件夹下
print(FLAGS.data_dir)
mnist = input_data.read_data_sets(FLAGS.data_dir, one_hot=True, source_url=SOURCE_URL)
def weight_variable(shape):
initial = tf.truncated_normal(shape, stddev=0.1) # 变量的初始值为截断正太分布
return tf.Variable(initial)
def bias_variable(shape):
initial = tf.constant(0.1, shape=shape)
return tf.Variable(initial)
def conv2d(x, W):
"""
tf.nn.conv2d功能:给定4维的input和filter,计算出一个2维的卷积结果
前几个参数分别是input, filter, strides, padding, use_cudnn_on_gpu, ...
input 的格式要求为一个张量,[batch, in_height, in_width, in_channels],批次数,图像高度,图像宽度,通道数
filter 的格式为[filter_height, filter_width, in_channels, out_channels],滤波器高度,宽度,输入通道数,输出通道数
strides 一个长为4的list. 表示每次卷积以后在input中滑动的距离
padding 有SAME和VALID两种选项,表示是否要保留不完全卷积的部分。如果是SAME,则保留
use_cudnn_on_gpu 是否使用cudnn加速。默认是True
"""
return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding='SAME')
def max_pool_2x2(x):
"""
tf.nn.max_pool 进行最大值池化操作,而avg_pool 则进行平均值池化操作
几个参数分别是:value, ksize, strides, padding,
value: 一个4D张量,格式为[batch, height, width, channels],与conv2d中input格式一样
ksize: 长为4的list,表示池化窗口的尺寸
strides: 窗口的滑动值,与conv2d中的一样
padding: 与conv2d中用法一样。
"""
return tf.nn.max_pool(x, ksize=[1, 2, 2, 1],
strides=[1, 2, 2, 1], padding='SAME')
def main(_):
sess = tf.InteractiveSession()
x = tf.placeholder(tf.float32, [None, 784])
x_image = tf.reshape(x, [-1,28,28,1]) #将输入按照 conv2d中input的格式来reshape,reshape
"""
# 第一层
# 卷积核(filter)的尺寸是5*5, 通道数为1,输出通道为32,即feature map 数目为32
# 又因为strides=[1,1,1,1] 所以单个通道的输出尺寸应该跟输入图像一样。即总的卷积输出应该为?*28*28*32
# 也就是单个通道输出为28*28,共有32个通道,共有?个批次
# 在池化阶段,ksize=[1,2,2,1] 那么卷积结果经过池化以后的结果,其尺寸应该是?*14*14*32
"""
W_conv1 = weight_variable([5, 5, 1, 32]) # 卷积是在每个5*5的patch中算出32个特征,分别是patch大小,输入通道数目,输出通道数目
b_conv1 = bias_variable([32])
h_conv1 = tf.nn.elu(conv2d(x_image, W_conv1) + b_conv1)
h_pool1 = max_pool_2x2(h_conv1)
"""
# 第二层
# 卷积核5*5,输入通道为32,输出通道为64。
# 卷积前图像的尺寸为 ?*14*14*32, 卷积后为?*14*14*64
# 池化后,输出的图像尺寸为?*7*7*64
"""
W_conv2 = weight_variable([5, 5, 32, 64])
b_conv2 = bias_variable([64])
h_conv2 = tf.nn.elu(conv2d(h_pool1, W_conv2) + b_conv2)
h_pool2 = max_pool_2x2(h_conv2)
# 第三层 是个全连接层,输入维数7*7*64, 输出维数为1024
W_fc1 = weight_variable([7 * 7 * 64, 1024])
b_fc1 = bias_variable([1024])
h_pool2_flat = tf.reshape(h_pool2, [-1, 7*7*64])
h_fc1 = tf.nn.elu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)
keep_prob = tf.placeholder(tf.float32) # 这里使用了drop out,即随机安排一些cell输出值为0,可以防止过拟合
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)
# 第四层,输入1024维,输出10维,也就是具体的0~9分类
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) # 使用softmax作为多分类激活函数
y_ = tf.placeholder(tf.float32, [None, 10])
cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(y_conv), reduction_indices=[1])) # 损失函数,交叉熵
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy) # 使用adam优化
correct_prediction = tf.equal(tf.argmax(y_conv,1), tf.argmax(y_,1)) # 计算准确度
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
sess.run(tf.global_variables_initializer()) # 变量初始化
for i in range(20000):
batch = mnist.train.next_batch(50)
if i%100 == 0:
# print(batch[1].shape)
train_accuracy = accuracy.eval(feed_dict={
x:batch[0], y_: batch[1], keep_prob: 1.0})
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}))
if __name__ == '__main__':
tf.app.run()
https://blog.csdn.net/u013082989/article/details/53695306