What is model evaluation?
Model evaluation’s primary purpose is to determine the generalizability of the model. Basically, it is used to see how well the model will perform on previously unseen data. Every data scientist seeks to increase the generalization accuracy of the model since it makes the model useless when it performs poorly on new data.
Why is it important?
Model evaluation is vital for any machine learning project since it shows how well the model was trained. It allows us to identify both overfitting and underfitting and helps to determine possible ways to improve the model.