This is the class and function reference of the Python API.
You can search for specific modules, classes or functions in the Index.
bolero.controller: Controller¶Controller([config, environment, …]) |
A controller implements the communication between learning components. |
ContextualController([config, environment, …]) |
Controller for contextual problems. |
bolero.environment: Environment¶Environment |
Common interface for environments. |
ContextualEnvironment |
Common interface for (contextual) environments. |
SetContext(contextual_environment, context) |
A contextual environment with a fixed context. |
ObjectiveFunction([name, n_params, random_state]) |
Artificial benchmark function. |
ContextualObjectiveFunction([name, …]) |
Artificial contextual benchmark function. |
OptimumTrajectory([x0, g, execution_time, …]) |
Optimize a trajectory according to some criteria. |
Catapult([segments, catapult_pos, …]) |
Catapult environment, a benchmark for contextual policy search. |
OpenAiGym([env_name, render, log_to_file, …]) |
Wrapper for OpenAI Gym environments. |
bolero.behavior_search: Behavior Search¶BehaviorSearch |
BehaviorSearch (learning algorithm) interface. |
ContextualBehaviorSearch |
Common interface for (contextual) behavior search. |
BlackBoxSearch(behavior, optimizer[, …]) |
Combine a black box optimizer with a black box behavior. |
ContextualBlackBoxSearch(behavior, optimizer) |
Combine a contextual black box optimizer with a black box behavior. |
JustOptimizer(optimizer) |
Wrap only the optimizer. |
JustContextualOptimizer(optimizer) |
Wrap only the contextual optimizer. |
MonteCarloRL(action_space[, gamma, epsilon, …]) |
Tabular Monte Carlo is a model-free reinforcement learning method. |
bolero.optimizer: Optimizer¶Optimizer |
Common interface for (non-contextual) optimizers. |
ContextualOptimizer |
Common interface for (contextual) optimizers. |
NoOptimizer([initial_params]) |
No optimizer. |
RandomOptimizer([initial_params, …]) |
Random optimizer. |
CMAESOptimizer([initial_params, variance, …]) |
Covariance Matrix Adaptation Evolution Strategy. |
RestartCMAESOptimizer([initial_params, …]) |
CMA-ES with restarts. |
IPOPCMAESOptimizer([initial_params, …]) |
Increasing population size CMA-ES. |
BIPOPCMAESOptimizer([initial_params, …]) |
BI-population CMA-ES. |
CCMAESOptimizer([initial_params, variance, …]) |
Contextual Covariance Matrix Adaptation Evolution Strategy. |
REPSOptimizer([initial_params, variance, …]) |
Relative Entropy Policy Search (REPS) as Optimizer. |
CREPSOptimizer([initial_params, variance, …]) |
Contextual Relative Entropy Policy Search. |
SkOptOptimizer(dimensions[, base_estimator, …]) |
Bayesian Optimization from scikit-optimize. |
ACMESOptimizer([initial_params, variance, …]) |
CMA-ES with ranking SVM as surrogate model. |
bolero.representation: Representation¶Behavior |
Behavior interface. |
BlackBoxBehavior |
Can be optimized with black box optimizer. |
ConstantBehavior([outputs]) |
Generates constant outputs. |
DummyBehavior(**kwargs) |
Dummy behavior allows using environments which do not require behaviors. |
RandomBehavior([random_state]) |
Generates random outputs. |
DMPBehavior([execution_time, dt, …]) |
Dynamical Movement Primitive. |
CartesianDMPBehavior([execution_time, dt, …]) |
Cartesian Space Dynamical Movement Primitive. |
DMPSequence([n_dmps, execution_times, dt, …]) |
Sequence of DMPs. |
LinearBehavior |
Linear mapping from inputs to outputs. |
BehaviorTemplate |
Behavior template interface. |
HierarchicalBehaviorTemplate(…[, explore]) |
Behavior template that consists of an upper-level policy and a behavior. |
bolero.datasets: Datasets¶make_minimum_jerk(start, goal[, …]) |
Create a minimum jerk trajectory. |
bolero.wrapper: Wrapper¶CppBLLoader |
Behavior learning loader. |
bolero.utils: Utilities¶from_yaml(filename[, conf_path]) |
Create objects from YAML configuration file. |
from_yaml_string(yaml_str) |
Create objects from YAML string. |
from_dict(config[, name]) |
Create an object of a class that is fully specified by a config dict. |
log.get_logger(obj, log_to_file, log_to_stdout) |
Get logger for given object. |
dependency.compatible_version(…) |
Compares version strings. |
log.HideExtern([stream]) |
Hide one of the standard streams for external components. |
ranking_svm.RankingSVM([n_iter, epsilon, …]) |
Ranking Support Vector Machine. |