Epoch
In plain terms
One full pass of the training data through a model.
Going deeper
An epoch is a complete cycle where the model sees every training example once; training usually runs many epochs. More epochs let the model learn more (or overfit).
Under the hood
Too few underfits, too many overfits; early stopping picks the sweet spot.
Examples
Training loops, early stopping, learning curves, neural network training.
Think of it like...
Like reading an entire textbook once, then again to learn more deeply.