I trained a neural network. A simple one.
I plotted the model's training loss. As you can see, it's too high.
This network is underfitting. It's not learning.
I need to make the model larger.
I increased the capacity of the model. The training loss is now low.
The model is not underfitting anymore, but it might be memorizing the data. I don't know yet.
I need to evaluate this model.
I tested the model and plotted the testing loss.
The training loss is low. The testing loss is high.
This model is overfitting. It's not learning. Instead, it's memorizing the training data.
I need to regularize this model.
