Producer

TODO

orion.core.worker.producer – Produce and register samples to try

class orion.core.worker.producer.Producer(experiment, max_idle_time=None)[source]

Produce suggested sets of problem’s parameter space to try out.

It uses an Experiment object to poll for not yet observed trials which have been already evaluated and to register new suggestions (points of the parameter Space) to be evaluated.

Attributes:
pool_size

Pool-size of the experiment

Methods

backoff() Wait some time and update algorithm.
produce() Create and register new trials.
register_trials(new_point) Register a new set of sampled parameters into the DB guaranteeing their uniqueness
update() Pull all trials to update model with completed ones and naive model with non completed ones.
backoff()[source]

Wait some time and update algorithm.

pool_size

Pool-size of the experiment

produce()[source]

Create and register new trials.

register_trials(new_point)[source]

Register a new set of sampled parameters into the DB guaranteeing their uniqueness

Parameters:
new_point: tuple

tuple of values representing the hyperparameters values

update()[source]

Pull all trials to update model with completed ones and naive model with non completed ones.