Chocolate Documentation

Chocolate is a completely asynchronous optimisation framework relying solely on a database to share information between workers. Chocolate uses no master process for distributing tasks. Every task is completely independent and only gets its information from the database. Chocolate is thus ideal in controlled computing environments where it is hard to maintain a master process for the duration of the optimisation.

Chocolate has been designed and optimized for hyperparameter optimization where each function evaluation takes very long to complete and is difficult to parallelize.

Chocolate is licensed under the 3-Clause BSD License

• Tutorials
  • Basics (Start here!)
  • Something a bit more realistic
  • Optimizing multiple models at once
  • Optimizing multiple losses at once
  • Let’s go to Tensor Flow
  • Retrieving Results
  • How to choose your algorithm
  • Cross-validating the results
• Installation
• Library Reference
• Release Notes
• About