--
bayesian-optimization

"Bayesian Optimization" is a pure Python implementation of bayesian global optimization with gaussian processes, available on GitHub.

SpecsUpdate

Random SearchNo

Quasi Random SearchNo

Pypi PackageYes

Bayesian Optimization SearchYes

Continuous VariablesYes

CMAES SearchNo

MOCMAES SearchNo

Evaluation Cost Sensitive SearchNo

Grid SearchNo

Integer VariablesNo

Categorical VariablesNo

Pros

Pro
Documentation includes code to allow easy visualization

The Jupyter notebook is here.

Note that this is a snapshot from a fixed revision. You can get the latest here.

Update

Comment

Add a Pro

Know any positive aspects of this option?

Cons

Con
Bayesian optimization requires making a smoothness assumption

The Bayesian optimization implementation needs the user to specify a kernel width that determines how smooth the system assumes functions will be with no data. Other implementations do not require this assumption and integrate out the smoothness assumption using Markov Chain Monte Carlo (MCMC) sampling.

The choice here is the width of a Matern kernel which, if you have to have one, is a more flexible kernel than the more commonly found Squared Exponential kernel.

Update

Comment