Training & Learning
Epoch
One complete pass of the training process through the entire training dataset.
An epoch is finished when the model has seen every example in the training set once. Models usually train for multiple epochs so they can keep refining, but there is a balance to strike. Too few epochs and the model underfits, meaning it has not learned enough, while too many can cause overfitting, where it starts memorizing the training data instead of learning general patterns.
Related terms
Overfitting
When a model memorizes its training data too closely and then performs badly on anything new.
Underfitting
When a model is too simple or undertrained to capture the real pattern, so it performs poorly even on its training data.
Training Data
The examples a model learns from, whose size, quality, and bias largely determine how the model turns out.
Hyperparameter
A setting you choose before training that controls how the model learns, as opposed to what it learns.