텐서플로우 코드 에러 - Cannot evaluate tensor using `eval()`

Deep MNIST for Experts 튜토리얼 실행시 발생할 수 있는 텐서플로우 코드 에러

value error 가 발생하는 경우

---

for i in range(20000):
    batch = mnist.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)
    train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
 
-------------------------------------------------------------------------
ValueError                              Traceback (most recent call last)
<ipython-input-21-ea20ade49f79> in <module>()
      2     batch = mnist.train.next_batch(50)
      3     if i%100 == 0:
----> 4         train_accuracy = accuracy.eval(feed_dict={x:batch[0], y_: batch[1], keep_prob: 1.0})
      5         print "step %d, training accuracy %g" % (i, train_accuracy)
      6     train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
 
/root/anaconda/envs/tensorflow/lib/python2.7/site-packages/tensorflow/python/framework/ops.pyc in eval(self, feed_dict, session)
    553 
    554     """
--> 555     return _eval_using_default_session(self, feed_dict, self.graph, session)
    556 
    557 

/root/anaconda/envs/tensorflow/lib/python2.7/site-packages/tensorflow/python/framework/ops.pyc in _eval_using_default_session(tensors, feed_dict, graph, session)
   3482     session = get_default_session()
   3483     if session is None:
-> 3484       raise ValueError("Cannot evaluate tensor using `eval()`: No default "
   3485                        "session is registered. Use `with "
   3486                        "sess.as_default()` or pass an explicit session to "

ValueError: Cannot evaluate tensor using `eval()`: No default session is registered. Use `with sess.as_default()` or pass an explicit session to `eval(session=sess)`

아래와 같이 명시적으로 세션을 넘긴다.

if i%100 == 0:
        train_accuracy = accuracy.eval(feed_dict={x:batch[0], y_: batch[1], keep_prob: 1.0})
 
 
if i%100 == 0:
        train_accuracy = accuracy.eval(session=sess, feed_dict={x:batch[0], y_: batch[1], keep_prob: 1.0})