Confusion Matrix for Classification Model Evaluation A confusion matrix is a n x n matrix used to describe the performance of a classification model (where n is the number of labels). In the confusion matrix, each row represents an actual class, while each column represents a predicted class.
To accomplish this, use the model to predict the answer using the evaluation dataset (held out data), and then compare the expected target to the actual answer (ground truth). A variety of measures are used in machine learning to assess a model's predicted performance. The accuracy metric used is determined by the ML task.