直接上代码:
#-*- coding:utf-8 -*- #author : zhangwei 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 = 100 n_batch = mnist.train.num_examples // batch_size with tf.Graph().as_default(): def variable_summaries(var): with tf.name_scope('summaries'): mean = tf.reduce_mean(var) tf.summary.scalar('mean' , mean) with tf.name_scope('stdeev'): stddev = tf.sqrt(tf.reduce_mean(tf.square(var - mean))) tf.summary.scalar('stddev' , stddev) tf.summary.scalar('max' , tf.reduce_max(var)) tf.summary.scalar('min' , tf.reduce_min(var)) tf.summary.histogram('histogram' , var) #初始化权重 def weight_variable(shape , name): initial = tf.truncated_normal(shape , stddev=0.1) return tf.Variable(initial , name=name) #初始化偏置 def bias_variable(shape , name): initial = tf.constant(0.1 , shape=shape) return tf.Variable(initial , name=name) #卷积层 def conv2d(x , W): return tf.nn.conv2d(x , W , strides=[1 , 1 , 1 , 1] , padding='SAME') #0,3位置都是1,分别代表是x,y方向进行卷积;1 ,2位置代表移动步长,自己设定; #池化层 def max_pool(x): return tf.nn.max_pool(x , ksize=[1 , 2 , 2 , 1] ,strides=[1 , 2 , 2 , 1], padding='SAME') with tf.name_scope('Input'): x = tf.placeholder(tf.float32 , [None , 784] , name='x_input') y = tf.placeholder(tf.float32 , [None , 10] , name='y_input') with tf.name_scope('x_image'): x_image = tf.reshape(x , [-1 , 28 , 28 , 1] , name='x_image') with tf.name_scope('Conv1'): with tf.name_scope('w_conv1'): w_conv1 = weight_variable([3 , 3 , 1 , 64] , name = "w_conv1") with tf.name_scope('b_conv1'): b_conv1 = bias_variable([64] , name = 'b_conv1') with tf.name_scope('conv1'): conv1_1 = conv2d(x_image , w_conv1) + b_conv1 with tf.name_scope('relu'): h_conv1 = tf.nn.relu(conv1_1) with tf.name_scope('pool1'): h_pool1 = max_pool(h_conv1) with tf.name_scope('Conv2'): with tf.name_scope('w_conv2'): w_conv2 = weight_variable([3 , 3 , 64 , 128] , name= 'w_conv2') with tf.name_scope('b_conv2'): b_conv2 = bias_variable([128] , name='b_conv2') with tf.name_scope('conv2'): conv2_1 = conv2d(h_pool1 , w_conv2) + b_conv2 with tf.name_scope('relu'): h_conv2 = tf.nn.relu(conv2_1) with tf.name_scope('pool2'): h_pool2 = max_pool(h_conv2) # with tf.name_scope('Conv3'): # with tf.name_scope('w_conv3'): # w_conv3 = weight_variable([3 , 3 , 64 , 64] , name= 'w_conv3') # with tf.name_scope('b_conv3'): # b_conv3 = bias_variable([64] , name='b_conv3') # with tf.name_scope('conv3'): # conv3_1 = conv2d(h_conv2 , w_conv3) + b_conv3 # with tf.name_scope('relu'): # h_conv3 = tf.nn.relu(conv3_1) # with tf.name_scope('pool3'): # h_pool1 = max_pool(h_conv3) # # with tf.name_scope('Conv4'): # with tf.name_scope('w_conv4'): # w_conv4 = weight_variable([3 , 3 , 64 , 128] , name= 'w_conv4') # with tf.name_scope('b_conv4'): # b_conv4 = bias_variable([128] , name='b_conv4') # with tf.name_scope('conv4'): # conv4_1 = conv2d(h_pool1 , w_conv4) + b_conv4 # with tf.name_scope('relu'): # h_conv4 = tf.nn.relu(conv4_1) # with tf.name_scope('pool2'): # h_pool2 = max_pool(h_conv4) with tf.name_scope('FC1'): with tf.name_scope('w_fc1'): w_fcl = weight_variable([7*7*128 , 1024] , name='w_fc1') with tf.name_scope('b_fc1'): b_fc1 = bias_variable([1024] , name = 'b_fc1') with tf.name_scope('h_pool2_flat'): h_pool2_flat = tf.reshape(h_pool2 , [-1 , 7 * 7 * 128] , name = 'h_pool2_flat')#由于h_pool2的数据是三维数据,需要将其转换到二维; with tf.name_scope('wx_plus_b1'): wx_plus_b1 = tf.matmul(h_pool2_flat , w_fcl) + b_fc1 with tf.name_scope('relu'): h_fc1 = tf.nn.relu(wx_plus_b1) with tf.name_scope('keep_prob'): keep_prob = tf.placeholder(tf.float32 , name = 'keep_prob') with tf.name_scope('h_fc1_drop'): h_fc1_drop = tf.nn.dropout(h_fc1 , keep_prob , name = 'h_fc1_drop') with tf.name_scope('FC2'): with tf.name_scope('w_fc2'): w_fc2 = weight_variable([1024 , 10] , name = 'w_fc2') with tf.name_scope('b_fc2'): b_fc2 = weight_variable([10] , name = 'b_fc2') with tf.name_scope('wx_plus_b2'): wx_plus_b2 = tf.matmul(h_fc1_drop , w_fc2) + b_fc2 with tf.name_scope('sotfmax'): prediction = tf.nn.softmax(wx_plus_b2) # with tf.name_scope('keep_prob'): # keep_prob = tf.placeholder(tf.float32 , name='keep_prob') # with tf.name_scope('h_fc2_drop'): # h_fc2_drop = tf.nn.dropout(h_fc2 , keep_prob , name='h_fc2_prob') # with tf.name_scope('FC3'): # with tf.name_scope('w_fc3'): # w_fc3 = weight_variable([1024 , 10] , name = 'w_fc3') # with tf.name_scope('b_fc3'): # b_fc3 = weight_variable([10] , name = 'b_fc3') # with tf.name_scope('wx_plus_b3'): # wx_plus_b3 = tf.matmul(h_fc2_drop , w_fc3) + b_fc3 # with tf.name_scope('softmax'): # prediction = tf.nn.softmax(wx_plus_b3) with tf.name_scope('cross_entropy'): cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y , logits=prediction) , name='cross_entropy') tf.summary.scalar('cross_entropy' , cross_entropy) with tf.name_scope('train'): train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy) with tf.name_scope('accuracy'): with tf.name_scope('correct_prediction'): correct_prediction = tf.equal(tf.argmax(prediction , 1) , tf.argmax(y , 1)) with tf.name_scope('accuracy'): accuracy = tf.reduce_mean(tf.cast(correct_prediction , tf.float32)) tf.summary.scalar('accuracy' , accuracy) writer = tf.summary.FileWriter(logdir='/home/zhangwei/tensorboard' , graph=tf.get_default_graph()) writer.close() merged = tf.summary.merge_all() with tf.Session() as sess: init = tf.global_variables_initializer() sess.run(init) train_writer = tf.summary.FileWriter('/home/zhangwei/PycharmProjects/tensorboard/train' , sess.graph) test_writer = tf.summary.FileWriter('/home/zhangwei/PycharmProjects/tensorboard/test' , sess.graph) for i in range(500): batch_xs , batch_ys = mnist.train.next_batch(batch_size) trainloss , _ = sess.run([cross_entropy , train_step] , feed_dict={x : batch_xs , y : batch_ys , keep_prob : 0.6}) summary = sess.run(merged , feed_dict={x :batch_xs , y : batch_ys , keep_prob : 0.6}) train_writer.add_summary(summary , i) batch_xs , batch_ys = mnist.test.next_batch(batch_size) summary = sess.run(merged , feed_dict={x :batch_xs , y :batch_ys , keep_prob : 0.6}) test_writer.add_summary(summary , i) if (i+1) % 5 == 0: test_acc = sess.run(accuracy , feed_dict={x : mnist.test.images , y : mnist.test.labels , keep_prob : 1.0}) train_acc = sess.run(accuracy , feed_dict={x : mnist.train.images[:10000] , y : mnist.train.labels[:10000] , keep_prob : 1.0}) print "Iter" + str(i+1) + " , Testing Accuracy = " + str(test_acc) + ' , Training Accuracy = ' + str(train_acc) + ' , trainingloss' + str(trainloss)
最后用tensorboard可视化:
由于本人机器计算能力有限,跑100次,损失在不断降低,继续搬砖!