0

i have this model that i trained with 100 epochs : Model with 100 Epoch

and then i save the model and train for another 100 epoch (total 200 epoch): Model with additional 100 epoch (200 epoch)

my question is, is my model not overfitting ? is it optimal ?

Willy Wijaya
  • 15
  • 7
  • 1
    How does it perform against your held-out test data? – Andy Jan 26 '21 at 19:28
  • 1
    Overfitting is not an exact term. I'd argue everything beyond the first 30, maybe 50 epochs is overfitting here. – couka Jan 26 '21 at 19:32
  • @Oso i splited my CIFAR-10 dataset to 50000 for training and 10000 for testing, i tested with my 89.55 % val accuracy it get 8955 true prediction out of 10000 test_labels – Willy Wijaya Jan 26 '21 at 19:56
  • @couka so u said that my model is overfit ? in wich model ? the first one or second ? – Willy Wijaya Jan 26 '21 at 19:59
  • @WillyWijaya you need another holdout set that isn't used in validation or training, so that you can test it for things like overfitting – Andy Jan 26 '21 at 20:00
  • i can tell the model is not overfitted if mostly the prediction is true ? @Oso – Willy Wijaya Jan 26 '21 at 20:03

2 Answers2

1

Overfitting is when a model captures patterns that won't recur in the future. This leads to a decrease in prediction accuracy.

You need to test your model on data that has not been seen in training or validation to determine if it is overfitting or not.

BoomBoxBoy
  • 1,770
  • 1
  • 5
  • 23
0

Over fitting is when your model scores very highly on your training set and poorly on a validation test set (or real life post-training predictions).

When you are training your model make sure that you have split your training dataset into two subsets. One for training and one for validation. If you see that your validation accuracy is decreasing as training goes on it means that your CNN has "overfitted" to the training set specifically and should not be generalized.

There are many ways to combat overfitting that should be used while training your model. Seeking more data and using harsh dropout are popular ways to ensure that a model is not overfitting. Check out this article for a good description of your problem and possible solutions.

raceee
  • 477
  • 5
  • 14