CODEWITHSUNDEEP
tensorflow
asdlfkjlkj
asdlfkjlkj

Reputation: 2358

Tensorflow: get predicted values

I have downloaded an example code of MLP and trying to understand how it works. I came accross this line where it calculates the accuracy of test datasets. 

accuracy = sess.run(accuracy, feed_dict={X: data_test, y: labels_test, dropout_keep_prob:1.})

Now, I want to get the predicted labels as well. How can I get the predicted labels?

Upvotes: 0

Views: 920

Answers (1)

Sorin
Sorin

Reputation: 11968

You need to get the prediction tensor. If you have the code it's the one you compare to y to compute the accuracy. Say it's called prediction, then you can write:

accuracy, prediction = sess.run([accuracy, prediction], feed_dict={X: data_set, y:labels_test, ...})

Upvotes: 2

Related Questions