search

rmcantin / bayesopt

BayesOpt: A toolbox for bayesian optimization, experimental design and stochastic bandits.
GNU Affero General Public License v3.0
396 stars 92 forks source link

Parallelization for matlab wrapper #2

Closed san-bil closed 7 years ago

san-bil commented 9 years ago

Is there any way to parallelize the function evaluations for the matlab implementation of bayesoptcont()?

If not for the actual optimization (due to the sequential dependency), then at least for the initial function evaluations (set by n_init_samples)? If I can have 20 of those initial samples running simultaneously using HTcondor, it's pretty wasteful to do them serially.

rmcantin commented 9 years ago

We are working on a parallelized version of BayesOpt, but it is complicated as Bayesian optimization is inherently sequential by definition.

Meanwhile, the next release will include a save and restore mechanism which can be use to preload initial samples computed separately.

san-bil commented 9 years ago

Yeah of course, I'm not sure how you'd actually parallelize the optimization, just that initial sampling of the parameter space. The preloading feature would be a handy way of doing it.

PiranjaF commented 8 years ago

Any thoughts on when the next version is released? Looking very much forward to this added functionality!