Action Noise

class stable_baselines3.common.noise.ActionNoise

The action noise base class

reset() → None

call end of episode reset for the noise

class stable_baselines3.common.noise.NormalActionNoise(mean: numpy.ndarray, sigma: numpy.ndarray)

A Gaussian action noise

Parameters

• mean – (np.ndarray) the mean value of the noise

• sigma – (np.ndarray) the scale of the noise (std here)

class stable_baselines3.common.noise.OrnsteinUhlenbeckActionNoise(mean: numpy.ndarray, sigma: numpy.ndarray, theta: float = 0.15, dt: float = 0.01, initial_noise: Optional[numpy.ndarray] = None)

An Ornstein Uhlenbeck action noise, this is designed to approximate Brownian motion with friction.

Based on http://math.stackexchange.com/questions/1287634/implementing-ornstein-uhlenbeck-in-matlab

Parameters

• mean – (np.ndarray) the mean of the noise

• sigma – (np.ndarray) the scale of the noise

• theta – (float) the rate of mean reversion

• dt – (float) the timestep for the noise

• initial_noise – (Optional[np.ndarray]) the initial value for the noise output, (if None: 0)

reset() → None

reset the Ornstein Uhlenbeck noise, to the initial position

class stable_baselines3.common.noise.VectorizedActionNoise(base_noise: stable_baselines3.common.noise.ActionNoise, n_envs: int)

A Vectorized action noise for parallel environments.

Parameters

• base_noise – ActionNoise The noise generator to use

• n_envs – (int) The number of parallel environments

reset(indices: Optional[Iterable[int]] = None) → None

Reset all the noise processes, or those listed in indices

Parameters

indices – Optional[Iterable[int]] The indices to reset. Default: None. If the parameter is None, then all processes are reset to their initial position.