I Am newbie to Machine Learning in general. I am currently trying to follow a tutorial on sentiment analysis using BERT and Transformers.

enter image description here

I do not know how i can Read the results to know the overfitting case. Can any one help me to read this and explain to me how to know it afterwards.

  • 1
    $\begingroup$ Welcome to DataScienceSE. I'm not knowledgeable about this type of model but your results show very high overfitting: at every epoch the training loss decreases (good) but the validation loss increases (bad), thus increasing the difference between the two values. The validation loss should be as close as possible to the training loss, and certainly not increase. $\endgroup$
    – Erwan
    Mar 3, 2022 at 12:34
  • $\begingroup$ @Erwan thank u so much. Very helpfull. $\endgroup$
    – amal
    Mar 3, 2022 at 14:07
  • $\begingroup$ @amal If you would like to gain some general understanding of learning curves and how to interpret train/val/test losses this and this question might be worth reading. $\endgroup$
    – Jonathan
    Mar 3, 2022 at 16:38


Your Answer

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