Epoch

Friendly Description: An epoch is one full trip through all the training data. Imagine reading an entire textbook from cover to cover, that's one epoch. Most AI models learn by going through their training data many times (many epochs), getting a little better with each pass. Each round helps the model notice patterns it missed before.

Example: Suppose you're training a model on a dataset of 10,000 product photos. After one epoch, the model has looked at every single photo once. After ten epochs, it has seen each photo ten times and is much better at recognizing the products it was being trained to spot.