TensorFlow save-cnn

save不用重复编写程序了

x = tf.placeholder(tf.float32, [None, n_input])
y = tf.placeholder(tf.float32, [None, n_output])
keepratio = tf.placeholder(tf.float32)

# FUNCTIONS

_pred = conv_basic(x, weights, biases, keepratio)['out']
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(_pred, y))
optm = tf.train.AdamOptimizer(learning_rate=0.001).minimize(cost)
_corr = tf.equal(tf.argmax(_pred,1), tf.argmax(y,1)) 
accr = tf.reduce_mean(tf.cast(_corr, tf.float32)) 
init = tf.global_variables_initializer()
    
# SAVER
save_step = 1#每一个epoch值都要保存
saver = tf.train.Saver(max_to_keep=3) #进行初始化save，其中的3表示只保留最后三组迭代模型

print ("GRAPH READY")

do_train = 0#
sess = tf.Session()
sess.run(init)

 

training_epochs = 15
batch_size      = 16
display_step    = 1
if do_train == 1:#如果do_train为1，则进行训练神经网络，如果为0就走下面的测试模块
    for epoch in range(training_epochs):
        avg_cost = 0.
        #total_batch = int(mnist.train.num_examples/batch_size)
        total_batch = 10
        # Loop over all batches
        for i in range(total_batch):
            batch_xs, batch_ys = mnist.train.next_batch(batch_size)
            # Fit training using batch data
            sess.run(optm, feed_dict={x: batch_xs, y: batch_ys, keepratio:0.7})
            # Compute average loss
            avg_cost += sess.run(cost, feed_dict={x: batch_xs, y: batch_ys, keepratio:1.})/total_batch

        # Display logs per epoch step
        if epoch % display_step == 0: 
            print ("Epoch: %03d/%03d cost: %.9f" % (epoch, training_epochs, avg_cost))
            train_acc = sess.run(accr, feed_dict={x: batch_xs, y: batch_ys, keepratio:1.})
            print (" Training accuracy: %.3f" % (train_acc))
            #test_acc = sess.run(accr, feed_dict={x: testimg, y: testlabel, keepratio:1.})
            #print (" Test accuracy: %.3f" % (test_acc))
            
        # Save Net
        if epoch % save_step == 0:#条件保存
            saver.save(sess, "save/nets/cnn_mnist_basic.ckpt-" + str(epoch))

    print ("OPTIMIZATION FINISHED")

 

if do_train == 0:#为0时走测试的模块
    epoch = training_epochs-1
    saver.restore(sess, "save/nets/cnn_mnist_basic.ckpt-" + str(epoch))
    
    test_acc = sess.run(accr, feed_dict={x: testimg, y: testlabel, keepratio:1.})
    print (" TEST ACCURACY: %.3f" % (test_acc))