I have these figures of accuracy and loss of training and validation sets of custom CNN models. I have some understanding of underfitting/overfitting however I am having trouble applying those concepts here. Any thoughts would be helpful. I’d like to justify my selection and talk about the others in a report.
[5 layers - accuracy](https://i.sstatic.net/0kLwOpPC.png)
[5 layers - loss](https://i.sstatic.net/B4yKUizu.png)
[7 layers - accuracy](https://i.sstatic.net/gAT3NHIz.png)
[7 layers - loss](https://i.sstatic.net/LR8oEEmd.png)
[9 layers - accuracy](https://i.sstatic.net/DnXw3h4E.png)
[9 layers - loss](https://i.sstatic.net/53XpTiVH.png)
New contributor
Anna TOes is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.