poli.objective_factory.start

poli.objective_factory.start(name: str, seed: Optional[int] = None, caller_info: Optional[dict] = None, observer: Optional[AbstractObserver] = None, force_register: bool = False, force_isolation: bool = False, **kwargs_for_factory) → AbstractBlackBox

Starts the black-box function.

Works just like create, but it returns only the black-box function and resets the evaluation budget.

One example of running this function:

```python from poli import objective_factory

with objective_factory.start(name=”aloha”) as f:

x = np.array([“F”, “L”, “E”, “A”, “S”]).reshape(1, -1) y = f(x) print(x, y)

```

If the black-box function is created as an isolated process, then it will automatically close when the context manager is closed.

Parameters

• name (str) – The name of the objective function.

• seed (int, optional) – The seed value for random number generation.

• caller_info (dict, optional) – Optional information about the caller that is forwarded to the logger to initialize the run.

• observer (AbstractObserver, optional) – The observer to use.

• force_register (bool, optional) – If True, then the objective function is registered without asking for confirmation, overwriting any previous registration.

• force_isolation (bool, optional) – If True, then the objective function is instantiated as an isolated process.

• **kwargs_for_factory (dict, optional) – Additional keyword arguments for the factory.