import os
import model
import tensorflow as tf
import input_data

data = input_data.read_data_sets('MNIST_data', one_hot=True)
#model
with tf.variable_scope("convolutional"):
    x = tf.placeholder(tf.float32, [None,784], name='x')
    keep_prob = tf.placeholder(tf.float32)
    y, variables = model.convolutional(x, keep_prob)

#train
y_ = tf.placeholder(tf.float32, [None, 10], name='y')
cross_entropy = -tf.reduce_sum(y_ * tf.log(y))
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

saver = tf.train.Saver(variables)
with tf.Session( ) as sess:
    merged_summary_op = tf.summary.merge_all( )
    summary_writer = tf.summary.FileWriter('/tmp/mnist log/1', sess.graph)
    summary_writer.add_graph(sess.graph)
    sess.run(tf.global_variables_initializer( ))
    for i in range(20000):
        batch = data.train.next_batch(50)
        if i % 100 == 0:
            train_accuracy = accuracy.eval(feed_dict={x: batch[0], y_:batch[1], keep_prob: 1.0})
            print("step %d,training accuracy %g" % (i,train_accuracy))
        sess.run(train_step, feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
   
    print(sess.run(accuracy, feed_dict={x: data.test.images,y_ : data.test.labels, keep_prob: 1.0})

    path = saver.save(
        sess, os.path.join(os.path.dirname(__file__), 'data', 'convolutional.ckpt'),
        write_meta_graph=False, write_state=False)
    print("saved:", path )
    
运行出现  下面错误

C:\ProgramData\Anaconda3\envs\mnist_testdemo\python.exe C:/Users/dbgen/PycharmProjects/mnist_testdemo/mnist/convolutional.py 
  File "C:/Users/dbgen/PycharmProjects/mnist_testdemo/mnist/convolutional.py", line 36    
    path = saver.save(       
^SyntaxError: invalid syntaxProcess finished with exit code 1