mlos_bench.launcher

A helper class to load the configuration files, parse the command line parameters, and instantiate the main components of mlos_bench system.

It is used in the mlos_bench.run module to run the benchmark/optimizer from the command line.

Classes

Launcher

Command line launcher for mlos_bench and mlos_core.

Module Contents

class mlos_bench.launcher.Launcher(description: str, long_text: str = '', argv: list[str] | None = None)[source]

Command line launcher for mlos_bench and mlos_core.

Parameters:

description (str)
long_text (str)
argv (list[str] | None)

property config_loader: mlos_bench.services.config_persistence.ConfigPersistenceService[source]

Get the config loader service.

Return type:: mlos_bench.services.config_persistence.ConfigPersistenceService

global_config[source]

optimizer[source]

root_env_config[source]

property root_environment: mlos_bench.environments.base_environment.Environment[source]

Gets the root (prototypical) Environment from the first TrialRunner.

Note: All TrialRunners have the same Environment config and are made unique by their use of the unique trial_runner_id assigned to each TrialRunner’s Environment’s global_config.

Notes

This is mostly for convenience and backwards compatibility.

Return type:: mlos_bench.environments.base_environment.Environment

scheduler[source]

property service: mlos_bench.services.base_service.Service[source]

Get the parent service.

Return type:: mlos_bench.services.base_service.Service

storage[source]

teardown: bool[source]

trial_runners = [][source]

tunables[source]