Base_agent

class conformer_rl.agents.base_agent.BaseAgent(config: conformer_rl.config.agent_config.Config)

Base interface for building reinforcement learning agents.

Parameters: config (Config) – Configuration object for the agent. See notes for a list of config parameters used by specific pre-built agents.

config

Configuration object used for the agent.

Type: Config

unique_tag

Unique identifier string for the current training session. Used to identify this session in logging.

Type: str

eval_logger

Used for logging environment data when evaluating agent.

Type: EnvLogger

train_logger

Used for logging agent data while training.

Type: TrainLogger

total_steps

Total number of environment interactions/steps taken by the agent.

Type: int

storage

Used to save environment samples, analogous to a replay buffer.

Type: Storage

run_steps() → None

Trains the agent.

Trains the agent until the maximum number of steps (specified by config) is reached. Also periodically saves neural network parameters and performs evaluations on the agent, if specified in the config.

step() → None: Performs one iteration of acquiring samples on the environment and then trains on the acquired samples.

_eval_episode() → dict

Evalutes the agent on a single episode of the evaluation environment.

Returns: Information from the evaluation environment to be logged by the eval_logger.
Return type: dict

evaluate() → None

Evaluates the agent on the evaluation environment.

Information dict returned by the environment’s conformer_rl.environments.conformer_env.ConformerEnv.step() method is logged by the eval_logger and saved.

load(filename: str) → None

Loads the neural network with weights.

Parameters: filename (str) – The path where the neural network weights are saved.

save(filename: str) → None

Saves the neural network weights to a file.

Parameters: filename (str) – The path where the neural network weights are to be saved.