Osprey¶
Osprey is a tool for practical hyperparameter optimization of machine learning algorithms. It’s designed to provide a practical, easy to use way for application scientists to find parameters that maximize the cross-validation score of a model on their dataset.
Osprey is actively being developed by researchers around the world, with primary application areas in computational protein dynamics and drug design, and distributed under the Apache License (v2.0). All development takes place on GitHub.
Overview¶
osprey is a command line tool. It runs using a simple config file which sets up the problem by describing the estimator, search space, strategy,
dataset, cross validation,
and storage for the results.
Related tools include and spearmint, hyperopt, and GPy. Both hyperopt and GPy can serve as backend search strategies for osprey.
To get started, run osprey skeleton to create an example config file, and
then boot up one or more parallel instances of osprey worker.
If you happen to run into any issues while using Osprey or would like suggest a new feature, please take a moment to read our “Contributing” section.