torchrl.policies package

Submodules

torchrl.policies.epsilon_greedy module

torchrl.policies.epsilon_greedy.epsilon_greedy(action_size: int, choices: numpy.array, eps: float = 0.1)[source]

Batched epsilon-greedy :param action_size: Total number of actions :param choices: A list of choices :param eps: Value of epsilon :return:

torchrl.policies.ou_noise module

class torchrl.policies.ou_noise.OUNoise(action_space, mu=0.0, theta=0.15, max_sigma=0.3, min_sigma=0.3, decay_period=100000)[source]

Bases: object

evolve_state()[source]
get_action(action, t=0)[source]
reset()[source]

Module contents