Noise

Gaussian Noise

class pytorchrl.agent.actors.noise.gaussian.Normal(action_size, mean=0, std=0.1)

Bases: object

Gaussian Noise

reset()

sample()

Ornstein-Uhlenbeck Noise

class pytorchrl.agent.actors.noise.ou.OUNoise(size, mu=0.0, theta=0.15, sigma=0.2)

Bases: object

Ornstein-Uhlenbeck process.

reset()

Reset the internal state (= noise) to mean (mu).

sample()

Update internal state and return it as a noise sample.