Module: tf_agents.bandits.environments.stationary_stochastic_per_arm_py_environment

Stationary Stochastic Python Bandit environment with per-arm features.

Classes

class StationaryStochasticPerArmPyEnvironment: Stationary Stochastic Bandit environment with per-arm features.

GLOBAL_KEY 'global'
NUM_ACTIONS_KEY 'num_actions'
PER_ARM_KEY 'per_arm'