第一部分:
前一段时间仔细学习了深度学习的CNN,算是对卷积神经网络有了一个初步的了解,想动手实现几个小项目加深理解,第一个自然选中的就是minst数据集做手写识别测试了,选的开源框架是谷歌的tensorflow。
#读入数据集
mnist = input_data.read_data_sets("MNIST_data",one_hot=True)
sess = tf.InteractiveSession()
#x,y为占位符,Wb参数使用变量
x = tf.placeholder("float",[None,784])
W = tf.Variable(tf.zeros([784,10]))
b = tf.Variable(tf.zeros([10]))
y_ = tf.placeholder("float",[None,10])
#更新权重
def weight_variable(shape):
initial = tf.truncated_normal(shape,stddev=0.1)
return tf.Variable(initial)
#更新偏置
def bias_variable(shape):
initial = tf.constant(0.1,shape=shape)
return tf.Variable(initial)
#卷几层计算,滑动窗口1,1,使用0为边缘做padding
def conv2d(x,W):
return tf.nn.conv2d(x,W,strides=[1,1,1,1],padding='SAME')
#使用最大化做池化层
def max_pool_2x2(x):
return tf.nn.max_pool(x,[1,2,2,1],[1,2,2,1],'SAME')
#第一层卷积层,使用relu激活
W_conv1 = weight_variable([5,5,1,32])
b_conv1 = bias_variable([32])
x_image = tf.reshape(x,[-1,28,28,1])
h_conv1 = tf.nn.relu(conv2d(x_image,W_conv1)+b_conv1)
h_pool1 = max_pool_2x2(h_conv1)
#第二层,与上一层配置相同
W_conv2 = weight_variable([5,5,32,64])
b_conv2 = bias_variable([64])
h_conv2 = tf.nn.relu(conv2d(h_pool1,W_conv2)+b_conv2)
h_pool2 = max_pool_2x2(h_conv2)
#全连接层
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.relu(tf.matmul(h_pool2_flat,W_fc1)+b_fc1)
#防止过拟合,使用dropout
keep_prob = tf.placeholder("float")
h_fc1_dp = tf.nn.dropout(h_fc1,keep_prob)
#全连接层输出,使用softmax激活
W_fc2 = weight_variable([1024,10])
b_fc2 = bias_variable([10])
y_conv = tf.nn.softmax(tf.matmul(h_fc1_dp,W_fc2)+b_fc2)
#使用互熵损失
cross_entry = -tf.reduce_sum(y_ * tf.log(y_conv))
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entry)
correct_prediction = tf.equal(tf.arg_max(y_conv,1),tf.arg_max(y_,1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction,"float"))
init = tf.initialize_all_variables()
sess.run(init)
for i in range(1000):
batch_x,batch_y = mnist.train.next_batch(50)
if i %100 == 0:
train_accruacy = accuracy.eval(feed_dict={x:batch_x,y_:batch_y,keep_prob:1.0})
print "step %d,train accuracy %g" %(i,train_accruacy)
train_step.run(feed_dict={x:batch_x,y_:batch_y,keep_prob:0.5})
print "test accuracy %g"%accuracy.eval(feed_dict={
x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0})
由于数据集是28X28,并且是单通道的图片,数据比较容易拟合训练,一个小时不到使用我的air的cpu就可以得到训练准确度和测试准确度达到98%的准确率。
PS:需要mnist数据集的朋友可以在我的分享里下载。