A generalization curve plots training loss and validation loss over training iterations or epochs. It helps identify learning behavior, including underfitting, overfitting, and the point at which validation performance stops improving.
A plot of both training loss and validation loss as a function of the number of iterations. A generalization curve can help you detect possible overfitting. For example, the following generalization curve suggests overfitting because validation loss ultimately becomes significantly higher than training loss. !A Cartesian graph in which the y-axis is labeled loss and the x-axis is labeled iterations. Two plots appear. One plots shows the training loss and the other shows the validation loss. The two plots start off similarly, but the training loss eventually dips far lower than the validation loss. See Generalization in Machine Learning Crash Course for more information.