March 2015
Beginner to intermediate
250 pages
5h 39m
English
The main goal of model building in machine learning is generalization, which is how well the model will perform its intended objective. Generalization means how the model would perform on new unseen data, typically data in production. To estimate the generalization of a model, we need to test its performance based on unseen data. The mechanism to do it varies, depending upon the task at hand.
In this section, we will discuss the common ways to evaluate the efficacy of a classification model.
One of the most common and basic ways of evaluating a model's performance is by creating a confusion matrix and computing various metrics such as accuracy, precision, recall, and so on. We will start with an example ...
Read now
Unlock full access