Simple GP Regression¶
Here are examples for simple GP regression models. These examples will work for small to medium sized datasets (~2,000 data points). All examples here use exact GP inference (and therefore assume a Gaussian noise observation model).
New to GPyTorch? Check out the GP Regression Tutorial!
- GP Regression Tutorial
- This is the simplest of the notebooks - a GP regression model with an RBF kernel. Start here if you are new to GPyTorch, or new to GPs in general.
- Spectral Mixture GP Regression
- This notebook expands on previous eample with a more complex kernel. The spectral mixture kernel is a great choice if you have a complex extrapolation problem.
- GP Regression (CUDA) with Fast Variances (LOVE)
- This notebook demonstrates LOVE, a technique to rapidly speed up predictive variance computations. Check out this notebook to see how to use LOVE in GPyTorch, and how it compares to standard variance computations.