PHS: A Toolbox for Parallel Hyperparameter Search
This is an incremental tool for researchers and practitioners in machine learning and related fields to speed up hyperparameter tuning on distributed compute resources.
The authors tackled the problem of hyperparameter optimization for expensive numerical computations by introducing PHS, an open-source Python framework that enables parallel hyperparameter search with minimal modifications to target functions, using Bayesian optimization for sample efficiency.
We introduce an open source python framework named PHS - Parallel Hyperparameter Search to enable hyperparameter optimization on numerous compute instances of any arbitrary python function. This is achieved with minimal modifications inside the target function. Possible applications appear in expensive to evaluate numerical computations which strongly depend on hyperparameters such as machine learning. Bayesian optimization is chosen as a sample efficient method to propose the next query set of parameters.