Task modules

Benchmark Tasks definition

class orion.benchmark.task.BenchmarkTask(max_trials: int, **kwargs)[source]

Base class describing what a task can do. A task will define the objective function and search space of it.

Parameters
max_trialsint

Max number of trials the experiment will run against this task.

kwargsdict

Configurable parameters of the task, a particular task implementation can have its own parameters.

Attributes
configuration

Return the configuration of the task.

max_trials

Return the max number of trials to run for the task.

Methods

__call__(*args, **kwargs)

All tasks will be callable by default, and method call() will be executed when a task is called directly.

call()

Define the black box function to optimize, the function will expect hyper-parameters to search and return objective values of trial with the hyper-parameters.

get_search_space()

Return the search space for the task objective function

__call__(*args, **kwargs)[source]

All tasks will be callable by default, and method call() will be executed when a task is called directly.

abstract call() List[Dict][source]

Define the black box function to optimize, the function will expect hyper-parameters to search and return objective values of trial with the hyper-parameters.

This method should be overridden by subclasses. It should receive the hyper-parameters as keyword arguments, with argument names matching the keys of the dictionary returned by get_search_space.

property configuration: Dict[str, Any]

Return the configuration of the task.

abstract get_search_space() Dict[str, str][source]

Return the search space for the task objective function

property max_trials: int

Return the max number of trials to run for the task.

class orion.benchmark.task.Branin(max_trials=20)[source]

Branin function as benchmark task

Methods

call(x)

Evaluate a 2-D branin function.

get_search_space()

Return the search space for the task objective function

call(x)[source]

Evaluate a 2-D branin function.

get_search_space()[source]

Return the search space for the task objective function

class orion.benchmark.task.CarromTable(max_trials=20)[source]

CarromTable function as benchmark task

Methods

call(x)

Evaluate a 2-D CarromTable function.

get_search_space()

Return the search space for the task objective function

call(x)[source]

Evaluate a 2-D CarromTable function.

get_search_space()[source]

Return the search space for the task objective function

class orion.benchmark.task.EggHolder(max_trials=20, dim=2)[source]

EggHolder function as benchmark task

Methods

call(x)

Evaluate a n-D eggholder function.

get_search_space()

Return the search space for the task objective function

call(x)[source]

Evaluate a n-D eggholder function.

get_search_space()[source]

Return the search space for the task objective function

class orion.benchmark.task.Forrester(max_trials: int, alpha: float = 0.5, beta: float = 0.5)[source]

Task based on the Forrester function, as described in https://arxiv.org/abs/1905.12982

\[f(x) = ((lpha x - 2)^2) sin(eta x - 4)\]
Parameters
max_trialsint

Maximum number of trials for this task.

alphafloat, optional

Alpha parameter used in the above equation, by default 0.5

betafloat, optional

Beta parameter used in the above equation, by default 0.5

Methods

call(x)

Define the black box function to optimize, the function will expect hyper-parameters to search and return objective values of trial with the hyper-parameters.

get_search_space()

Return the search space for the task objective function

call(x: float) List[Dict][source]

Define the black box function to optimize, the function will expect hyper-parameters to search and return objective values of trial with the hyper-parameters.

This method should be overridden by subclasses. It should receive the hyper-parameters as keyword arguments, with argument names matching the keys of the dictionary returned by get_search_space.

get_search_space() Dict[str, str][source]

Return the search space for the task objective function

class orion.benchmark.task.RosenBrock(max_trials=20, dim=2)[source]

RosenBrock function as benchmark task

Methods

call(x)

Evaluate a n-D rosenbrock function.

get_search_space()

Return the search space for the task objective function

call(x)[source]

Evaluate a n-D rosenbrock function.

get_search_space()[source]

Return the search space for the task objective function