Effacer les filtres
Effacer les filtres

training stops due to NaN loss value

8 vues (au cours des 30 derniers jours)
Shoaib Ali
Shoaib Ali le 22 Août 2022
Commenté : Shoaib Ali le 26 Août 2022
The training process stops due NaN loss... How to avoid this to complete the training ..and what is the possible issue that causses..
  5 commentaires
Shoaib Ali
Shoaib Ali le 23 Août 2022
Weighted cross entropy loss
Chunru
Chunru le 23 Août 2022
try "dbstop error" and then run the program. Check if the network output is 0. There might be a problem if network output is 0 since entropy loss has term of T*log(Y) where T is target and Y is the network output.

Connectez-vous pour commenter.

Réponses (1)

Shoaib Ali
Shoaib Ali le 24 Août 2022
Can you explain, where to use this inside the loss funtion or before the training command??
  2 commentaires
Chunru
Chunru le 25 Août 2022
"dbstop error" can be used in command line before you train the network. Then it should stop when error occurs and then you check out what is wrong at which part of program.
Shoaib Ali
Shoaib Ali le 26 Août 2022
ok Thanks

Connectez-vous pour commenter.

Catégories

En savoir plus sur Statistics and Machine Learning Toolbox dans Help Center et File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by