Open EmCity opened 7 years ago
Thank you for this update. Now it's wokring with TF0.12.1 But I still have an issue with the eval-function.
Traceback (most recent call last): File "main.py", line 84, in <module> print "Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}) File "/usr/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 575, in eval return _eval_using_default_session(self, feed_dict, self.graph, session) File "/usr/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 3619, in _eval_using_default_session raise ValueError("Cannot evaluate tensor using eval(): No default " 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)
Hi jkronen,
Are you running an interactive session?
Try passing session=sess to the eval function, e.g. print "Accuracy:", accuracy.eval(session=sess, {x: mnist.test.images, y: mnist.test.labels})
Good luck! Sebastian
Hey Sebastian,
thank you for your reply. I am not running an interactive session. I just have the code from here with your updates and run it from the terminal.
Changing the line to your suggestion, produces following output:
print "Accuracy:", accuracy.eval(session=sess,{x: mnist.test.images, y: mnist.test.labels})
SyntaxError: non-keyword arg after keyword arg](url)
I even tried accuracy.eval(sess,{x:
or setting the session as default, but none of them worked
Greetings Jan
Hi Jan,
I'm sorry, I was a little careless, try this line:
print "Accuracy:", accuracy.eval(session=sess, feed_dict={x: mnist.test.images, y: mnist.test.labels})
If this throws an error, please report the whole error message.
Cheers, Sebastian
There you go, it's working now. Thank you very much!
Greeting Jan
Hey Siraj,
first of all I want to say I love your videos and your code! It's really helpful for getting into Machine Learning and Deep Learning in particular.
I updated some of the Tensorflow commands in that code (some warning will be issued when executing your current code with latest the Tensorflow version)
Cheers, Sebastian