We introduce pyGPs, an object-oriented implementation ofGaussian processes (gps) for machine learning. Thelibrary provides a wide range of functionalities reaching fromsimple gp specification via mean and covariance andgp inference to more complex implementations ofhyperparameter optimization, sparse approximations, and graphbased learning. Using Python we focus on usability for both"users" and "researchers". Our main goal is to offer a user-friendly and flexible implementation of gps formachine learning. color="gray">
展开▼
机译：我们介绍pyGPs，这是面向机器学习的高斯进程（ gp s）的面向对象实现。该库提供了广泛的功能，从简单的 gp 规范到均值和协方差，再到 gp 推断到超参数优化，稀疏近似和基于图的学习等更复杂的实现。使用Python，我们专注于“用户”和“研究人员”的可用性。我们的主要目标是为机器学习提供 gp 的用户友好和灵活实现。 color =“ gray”>
展开▼
