0
$\begingroup$

I am working on a Keras and TensorFlow in R and I trying to make good predictions for a regression problem. Below you can see how is look like plots two plots one for loss function and second for mean absolute error.

On the first plot validation with green line is above red line and also on second plot green line is above red line.

So can anybody help me how to interpret this result ? Does my neural network fit well ?

enter image description here

Improve this question
$\endgroup$
3
  • 2
    $\begingroup$ I think you forgot to attach the plots that you're referring to. In any event, machine learning and statistical models don't exist in a vacuum. The relevant question isn't "is my loss value good?", but instead "is my model good enough for my particular application?" When the model makes mistakes (which it will, inevitably), you want to know if the costs of those mistakes are outweighed by the value of its correct predictions. $\endgroup$
    – Sycorax ♦
    Commented Jan 4, 2022 at 19:18
  • $\begingroup$ So I think that this opinion is so general and don't give answer about above question $\endgroup$
    – silent_hunter
    Commented Jan 4, 2022 at 19:32
  • $\begingroup$ It's not necessarily the kind of subjective question that is discouraged on Stack Exchange, but it does make the question incomplete. Knowing the answer to "Does my neural network fit well?" requires information that only you have: "Do the predictions from this model solve my problem?" $\endgroup$
    – Sycorax ♦
    Commented Jan 4, 2022 at 19:37

1 Answer 1

Reset to default
1
$\begingroup$

Through each epoch, the model fine tunes its parameters and internal weights, hence why the loss functions of your model decreases with time. The first epoch will typically yield the greatest loss, since the model has seen each sample of the dataset once. Looking at the graphs above, the model converges in accuracy after 15 epochs, and thus running the model through more epochs will run the risk of overfitting in the pursuit of minimal decreases in loss.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Thank you. Much more clear and precise. $\endgroup$
    – silent_hunter
    Commented Jan 4, 2022 at 20:06

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.