Add to Chrome
Add to Firefox
Add to EdgeAuptimizer -- an Extensible, Open-Source Framework for Hyperparameter Tuning
Nov 06, 2019
Tuning machine learning models at scale, especially finding the right hyperparameter values, can be difficult and time-consuming. In addition to the computational effort required, this process also requires some ancillary efforts including engineering tasks (e.g., job scheduling) as well as more mundane tasks (e.g., keeping track of the various parameters and associated results). We present Auptimizer, a general Hyperparameter Optimization (HPO) framework to help data scientists speed up model tuning and bookkeeping. With Auptimizer, users can use all available computing resources in distributed settings for model training. The user-friendly system design simplifies creating, controlling, and tracking of a typical machine learning project. The design also allows researchers to integrate new HPO algorithms. To demonstrate its flexibility, we show how Auptimizer integrates a few major HPO techniques (from random search to neural architecture search). The code is available at https://github.com/LGE-ARC-AdvancedAI/auptimizer.
