From the course: LinkedIn AI Academy AI-100: 1 Demystifying AI
Join today to access over 23,200 courses taught by industry experts.
Evaluating a linear regression model
From the course: LinkedIn AI Academy AI-100: 1 Demystifying AI
Evaluating a linear regression model
- [Instructor] We know that we can find a sweet spot between bias and variance. What principles can we apply to find that sweet spot in practice? One popular mechanism is to take the data and split that into two parts. The training data and the test data. As the names suggest, training data is the data that we use to train the model. The test data, and we keep that separate and use only for evaluation purposes. This separation is absolutely critical because this allows us to measure the variance of the model independent of the training. One common way in which a high model variance manifests is through the model overfitting to the training data. Overfitting means that the model has not only learned the major patterns in the data but also learned some of the randomness in the data as well. It is not necessarily good because we want to learn the patterns in the data and not the randomness. By keeping the test data…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.