tensorflow用CNN实现手写数字识别代码

 

下载mnist数据集保存在和MNIST_data文件夹下，和下面程序放在同一目录下

 

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data


x=tf.placeholder(shape=[None,784],dtype=tf.float32)
y=tf.placeholder(shape=[None,10],dtype=tf.float32)
x_image=tf.reshape(x,[-1,28,28,1])

convcore1=tf.Variable(tf.truncated_normal([5,5,1,32],stddev=0.1))
convcore1_bias=tf.Variable(tf.zeros([32])+0.1)
convlayer1=tf.nn.relu(
tf.add(tf.nn.conv2d(x_image,convcore1,strides=[1,1,1,1],padding="SAME"),convcore1_bias)
)
pooling1=tf.nn.max_pool(convlayer1,ksize=[1,2,2,1],strides=[1,2,2,1],padding="SAME")

convcore2=tf.Variable(tf.truncated_normal([3,3,32,64],stddev=0.1))
convcore2_bias=tf.Variable(tf.zeros([64])+0.1)
convlayer2=tf.nn.relu(
tf.nn.conv2d(pooling1,convcore2,strides=[1,1,1,1],padding="SAME")+convcore2_bias
)
pooling2=tf.nn.max_pool(convlayer2,ksize=[1,2,2,1],strides=[1,2,2,1],padding="SAME")

print(pooling2)
pooling2_flat=tf.reshape(pooling2,[-1,7*7*64])

keep_prob=tf.placeholder(tf.float32)
x1=pooling2_flat
w1=tf.Variable(tf.truncated_normal([7*7*64,500],stddev=0.1))
b1=tf.Variable(tf.zeros([500])+0.1)
layer1=tf.nn.relu(
tf.add(tf.matmul(x1,w1),b1)
)
layer1drop=tf.nn.dropout(layer1,keep_prob)

w2=tf.Variable(tf.truncated_normal([500,500],stddev=0.1))
b2=tf.Variable(tf.zeros([500])+0.1)
layer2=tf.nn.relu(
tf.add(tf.matmul(layer1drop,w2),b2)
)
layer2drop=tf.nn.dropout(layer2,keep_prob)

w3=tf.Variable(tf.truncated_normal([500,500],stddev=0.1))
b3=tf.Variable(tf.zeros([500])+0.1)
layer3=tf.nn.relu(
tf.add(tf.matmul(layer2drop,w3),b3)
)
layer3drop=tf.nn.dropout(layer3,keep_prob)


w4=tf.Variable(tf.truncated_normal([500,10],stddev=0.1))
b4=tf.Variable(tf.zeros([10])+0.1)
logit=tf.nn.softmax(
tf.add(tf.matmul(layer3drop,w4),b4)
)

loss=tf.nn.softmax_cross_entropy_with_logits(labels=y,logits=logit)
train_step=tf.train.GradientDescentOptimizer(0.00001).minimize(loss)

precise=tf.reduce_mean(tf.cast(tf.equal(tf.argmax(y,1),tf.argmax(logit,1)),dtype=tf.float32))


mnist=input_data.read_data_sets("MNIST_data",one_hot=True)
batch_size=1000
n_batch_size=mnist.train.num_examples//batch_size

with tf.Session() as sess:
    init=tf.global_variables_initializer()
    sess.run(init)
    for i in range(1000):
        for j in range(n_batch_size):
            batch_x,batch_y=mnist.train.next_batch(batch_size)
            sess.run(train_step,feed_dict={x:batch_x,y:batch_y,keep_prob:1.0})
            loss_p=sess.run(loss,feed_dict={x:batch_x,y:batch_y,keep_prob:1.0})
            precise_p=sess.run(precise,feed_dict={x:batch_x,y:batch_y,keep_prob:1.0})
            # print("loss"+str(loss_p))
            print("precise"+str(precise_p))

 

 

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data

mnist=input_data.read_data_sets("MNIST_data",one_hot=True)
batch_size=1000
n_batch=mnist.train.num_examples//batch_size

def weight_variable(shape):
    initial=tf.truncated_normal(shape,stddev=0.1)
    return tf.Variable(initial)

def bias_variable(shape):
    initial=tf.constant(0.1,shape=shape)
    return tf.Variable(initial)

def conv2d(x,W):
    #strides strides[0]=stride[3]=1 stride[1]x方向步长，y方向步长
    #padding SAME or VALID
    a=tf.nn.conv2d(x,W,strides=[1,1,1,1],padding="SAME")
    return a
def max_pool_2x2(x):
    return tf.nn.max_pool(x,ksize=[1,2,2,1],strides=[1,2,2,1],padding="SAME")

x=tf.placeholder(tf.float32,[None,784])
y=tf.placeholder(tf.float32,[None,10])

#-1表示batch-size的大小
x_image=tf.reshape(x,[-1,28,28,1])
W_conv1=weight_variable([5,5,1,32])
b_conv1=bias_variable([32])

h_conv1=tf.nn.relu(conv2d(x_image,W_conv1)+b_conv1)
h_pool1=max_pool_2x2(h_conv1)

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)

W_fc1=weight_variable([7*7*64,200])
b_fc1=bias_variable([200])

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)

keep_prob=tf.placeholder(tf.float32)
h_fc1_drop=tf.nn.dropout(h_fc1,keep_prob)

W_fc2=weight_variable([200,10])
b_fc2=bias_variable([10])
prediction=tf.nn.softmax(tf.matmul(h_fc1_drop,W_fc2)+b_fc2)

cross_entropy=tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y,logits=prediction))
train_step=tf.train.GradientDescentOptimizer(0.01).minimize(cross_entropy)
correct_predict=tf.equal(tf.argmax(prediction,1),tf.argmax(y,1))
accuracy=tf.reduce_mean(tf.cast(correct_predict,tf.float32))

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    for epoch in range(100):
        for batch in range(n_batch):
            batch_xs,batch_ys=mnist.train.next_batch(batch_size)
            sess.run(train_step,feed_dict={x:batch_xs,y:batch_ys,keep_prob:1.0})
            train_loss=sess.run(cross_entropy,feed_dict={x:batch_xs,y:batch_ys,keep_prob:1.0})
            train_accuracy=sess.run(accuracy,feed_dict={x:batch_xs,y:batch_ys,keep_prob:1.0})
            # print(train_loss)
            print("train_accuracy:"+str(train_accuracy))
        # acc=sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.test.labels,keep_prob:1.0})
        # print("Iter"+str(epoch)+",Testing Accuracy= "+str(acc))