cifar_10.py

import cifar10,cifar10_input
import tensorflow as tf
import numpy as np
import time
max_steps= 3000
batch_size =128
data_dir = '/tmp/cifar10_data/cifar-10-batches-bin'
def variable_with_weight_loss(shape,stddev,w1):
    var = tf.Variable(tf.truncated_normal(shape,stddev=stddev))#使用截断的正态分布来初始化权重
    if w1 is not None:
        weight_loss = tf.multiply(tf.nn.l2_loss(var),w1,name='weight_loss')#做l2的正则化，用w1来控制l2 loss的大小
        tf.add_to_collection('losses',weight_loss)#把结果保存在一个collection
    return var
cifar10.maybe_download_and_extract()#使用cifar10类下载数据集，并解压，展开到默认位置
images_train,labels_train = cifar10_input.distorted_inputs(data_dir=data_dir,batch_size=batch_size)
images_test,labels_test = cifar10_input.inputs(eval_data=True,data_dir=data_dir,batch_size=batch_size)#生成测试数据

image_holder = tf.placeholder(tf.float32,[batch_size,24,24,3])
label_holder = tf.placeholder(tf.int32,[batch_size])
    #第一层
weight1 = variable_with_weight_loss(shape=[5,5,3,64],stddev=5e-2,w1=0.0)#第一层卷积核使用5乘5的卷积核输入通道为3输出通道为64 不使用l2正则化因此w1设为0
kernel1 = tf.nn.conv2d(image_holder,weight1,[1,1,1,1],padding='SAME')#卷积
bias1 = tf.Variable(tf.constant(0.0,shape=[64]))#设置偏置
conv1 = tf.nn.relu(tf.nn.bias_add(kernel1,bias1))#激活
pool1 = tf.nn.max_pool(conv1,ksize=[1,3,3,1],strides=[1,2,2,1],padding='SAME')#最大池化
norm1 = tf.nn.lrn(pool1,4,bias=1.0,alpha=0.001/9.0,beta=0.75)#局部响应归一化
    #第二层
weight2 = variable_with_weight_loss(shape=[5,5,64,64],stddev=5e-2,w1=0.0)#输入通道64输出通道64
kernel2 = tf.nn.conv2d(norm1,weight2,[1,1,1,1],padding='SAME')#卷积
bias2 = tf.Variable(tf.constant(0.1,shape=[64]))#设置偏置
conv2 = tf.nn.relu(tf.nn.bias_add(kernel2,bias2))#激活
norm2 = tf.nn.lrn(conv2,4,bias=1.0,alpha=0.001/9.0,beta=0.75)#局部响应归一化
pool2 = tf.nn.max_pool(norm2,ksize=[1,3,3,1],strides=[1,2,2,1],padding='SAME')#池化
    #调换了lrn与池化的顺序
    #设置两层全连接层
reshape = tf.reshape(pool2,[batch_size,-1])#将函数每个样本都变为一维向量
dim = reshape.get_shape()[1].value#获得数据扁平化后的值
weight3 = variable_with_weight_loss(shape=[dim,384],stddev=0.04,w1=0.004)#设置权重
bias3 =tf.Variable(tf.constant(0.1,shape=[384])) #设置偏置
local3 = tf.nn.relu(tf.matmul(reshape,weight3)+bias3)
weight4 = variable_with_weight_loss(shape=[384,192],stddev=0.04,w1=0.004)
bias4 = tf.Variable(tf.constant(0.1,shape=[192]))
local4 = tf.nn.relu(tf.matmul(local3,weight4)+bias4)
    #最后一层
weight5 = variable_with_weight_loss(shape=[192,10],stddev=1/192.0,w1=0.0)
bias5 = tf.Variable(tf.constant(0.0,shape=[10]))
logits = tf.add(tf.matmul(local4,weight5),bias5)
def loss(logits,labels):
    labels = tf.cast(labels,tf.int64)#转换数据类型
    cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
        logits=logits,labels=labels,name='cross_entropy_per_example'
    )#交叉熵加入了softmax计算
    cross_entropy_mean = tf.reduce_mean(cross_entropy,name='cross_entropy')#求平均
    tf.add_to_collection('losses',cross_entropy_mean)#把损失加到collection里
    return tf.add_n(tf.get_collection('losses'),name='total_loss')
loss=loss(logits,label_holder)#得到损失
train_op = tf.train.AdamOptimizer(1e-3).minimize(loss)#定义训练，学习速率为1e-3最小化loss
top_k_op = tf.nn.in_top_k(logits,label_holder,1)#得到概率最高的预测值与标签相同的正确率
sess = tf.InteractiveSession()#创建默认的session
sess.run(tf.initialize_all_variables())#初始化变量
tf.train.start_queue_runners()#启动线程队列
for step in range(max_steps):
    start_time = time.time()
    image_batch,label_batch=sess.run([images_train,labels_train])
    _,loss_value = sess.run([train_op,loss],feed_dict={image_holder:image_batch,label_holder:label_batch})
    duration = time.time()-start_time
    if step % 10 == 0:
        examples_per_sec = batch_size / duration
        sec_per_batch = float(duration)
        format_str=('step %d,loss=%.2f (%.1f example/sec;%.3f sec/batch)')
        print(format_str % (step,loss_value,examples_per_sec,sec_per_batch))
num_examples = 10000
import math
num_iter = int(math.ceil(num_examples/batch_size))#返回大于等于参数的最小整数
true_count = 0
total_sample_count = num_iter * batch_size
step = 0
while step < num_iter:
    image_batch,label_batch = sess.run([images_test,labels_test])
    predictions = sess.run([top_k_op],feed_dict={image_holder:image_batch,label_holder:label_batch})
    true_count += np.sum(predictions)#正确的个数
    step += 1
precision = true_count / total_sample_count#正确率等于正确的个数除总个数
print('precision @ 1=%.3f'% precision)