hyperopt / hyperopt

Distributed Asynchronous Hyperparameter Optimization in Python
http://hyperopt.github.io/hyperopt
Other
7.25k stars 1.06k forks source link

Warm start and get following combination of parameters #360

Closed nnnet closed 3 weeks ago

nnnet commented 6 years ago

Is it possible(how?) to implement a warm start?

For example the objective function has a single parameter space=hp.uniform('a', -10, 10) We already know 3 of the evaluation of the function [0.25, 0.7, 0.45] for 3 values of 'a', for example, [1.8, -2.7, 9.0]. Loaded them using trials_from_docs() How get the next recommended value of the parameter 'a' (something like method __.next() ) ?

Thanks

github-actions[bot] commented 1 month ago

This issue has been marked as stale because it has been open 120 days with no activity. Remove the stale label or comment or this will be closed in 30 days.