1
$\begingroup$

If loss is decreasing but val_loss not, what is the problem and how can I fix it?

I get such vague result: enter image description here

| improve this question | | | | |
$\endgroup$
  • $\begingroup$ Are you sure this isn't backwards? It would be odd for validation loss to be consistently lower than train. Not impossible, but atypical. $\endgroup$ – Sean Owen Mar 9 '19 at 22:11
3
$\begingroup$

This indicates that model is not generalizing (it is over-fitting). Few options are :

  1. Get more training data
  2. Reduce complexity of model (Number of LSTM layers, complexity of dense layers)

Andrew NG has a good video on this topic :

https://www.youtube.com/watch?v=OSd30QGMl88

A tutorial specific to LSTM :

https://machinelearningmastery.com/diagnose-overfitting-underfitting-lstm-models/

| improve this answer | | | | |
$\endgroup$

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