An epoch is the number of times the enitre dataset has passed though the algorith.


An iteration represents the number of batches required to pass the dataset thoutugh the algorithm.


A batch represents the number of samples taken from the dataset and fed into the algorithm before the model is recalibrated.

Learning Curve/Gradient

A graph with x = time and y = accuracy, it's populated with valuation data from each iteration of a model training. It's used to determine if a model is over/under trained.