pytorchrl.agent.actors.noise package

Submodules

pytorchrl.agent.actors.noise.gaussian module

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

Bases: object

Gaussian Noise

reset()[source]

sample()[source]

pytorchrl.agent.actors.noise.nonoise module

class pytorchrl.agent.actors.noise.nonoise.NoNoise(num_outputs)[source]

Bases: object

reset()[source]

sample()[source]

pytorchrl.agent.actors.noise.ou module

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

Bases: object

Ornstein-Uhlenbeck process.

reset()[source]: Reset the internal state (= noise) to mean (mu).

sample()[source]: Update internal state and return it as a noise sample.

Module contents

pytorchrl.agent.actors.noise.get_noise(name)[source]: Returns noise model class from name.