Machine-learning-based autotuning and hyperparameter optimization framework

Areas: Development tools, Uses AI

CASS member: RAPIDS

Description

ytopt is a machine learning-based autotuning and hyperparameter optimization software package in python that uses Bayesian Optimization to find the best input parameter/hyperparameter configurations for a given kernel, miniapp, or application with the best system configurations for a given HPC system. ytopt handles both unconstrained and constrained optimization problems, searches and evaluates asynchronously, and can look-ahead on iterations to more effectively adapt to new evaluations and adjust the search towards promising configurations, leading to a more efficient and faster convergence on the best solutions.

Target audience

Application developers and users for autotuning applications in performance and energy efficiency, system developers and administrators for autotuning system configurations, AI/ML developers and users for hyperparameter optimization, and so on.

License: BSD-2-Clause

Additional resources