In addition to calculating individual numerical metrics for our models, there are a variety of techniques to combine various metrics into a form that gives you a more complete representation of model performance. These include, but are certainly not limited to, confusion matrices and area under the curve (AUC)/Receiver Operating Characteristic (ROC) curves.
Confusion matrices allow us to visualize the various TP, TN, FP, and FN values that we predict in a two-dimensional format. A confusion matrix has rows corresponding to the categories that you were supposed to predict, and columns corresponding to categories that were predicted. Then, the value of each element is the corresponding count:
As you can see, ...
