overfiitting during CNN training

1 visualización (últimos 30 días)
new_user
new_user el 20 de Mzo. de 2022
Comentada: new_user el 23 de Mzo. de 2022
is this the case of overfitting, I used alexnet. After epoch 10 it seems that the validation loss is increasing instead of decreasing. So, is there any specific parameterI should employ to reduce this like augmentaion or just simple regularization.

Iniciar sesión para responder a esta pregunta.

Respuesta aceptada

Prateek Rai
Prateek Rai el 23 de Mzo. de 2022
Hi,
You can try specifying different training options for training the deep learning neural network. Refer to trainingOptions MathWorks Documentation page to find more.

Más respuestas (0)

Categorías

Más información sobre Image Data Workflows en Help Center y File Exchange.

Etiquetas

Productos

Community Treasure Hunt

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

Start Hunting!

Translated by