I am working on an Autoencoder in keras with the following setting: 185-86-32-2-32-86-185. The problem is that its accuracy is fluctuating. Also, it is giving a new accuracy at every run.

If this is due to the random uniform kernel initialization, how can I save the output values of the initializes in order to use them later?

My code:

enter image description here


1 Answer 1


Accuracy makes no sense for an autoencoder, because you're essentially doing regression (from X to X), you should use a different metric instead, like MSE, MAE, RMSE.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.