Uses cloudpickle to serialize contents (otherwise multiprocessing tries to use pickle)

Action space: - Sleep mode: 0, 1, 2, 3 - Connection mode: 0, 1, 2, 3 - Switch antennas: -16, -4, 0, 4, 16

MAPPO Policy class. Wraps actor and critic networks to compute actions and value function predictions. :param args: (argparse.Namespace) arg

MLP Module to compute actions. :param action_space: (gym.Space) action space. :param inputs_dim: (int) dimension of network input. :p

Actor network class for MAPPO. Outputs actions given observations. :param args: (argparse.Namespace) arguments containing relevant model info

Critic network class for MAPPO. Outputs value function predictions given centralized input (MAPPO) or local observations (IPPO). :param

- The multi-discrete action space consists of a series of discrete action spaces with different parameters - It can be adapted to both a Disc

Buffer to store training data. :param args: (argparse.Namespace) arguments containing relevant model, policy, and env information. :param

Normalize a vector of observations - across the first norm_axes dimensions

Base class for training recurrent policies. :param config: (dict) Config dictionary containing parameters for training.

Trainer class for MAPPO to update policies. :param args: (argparse.Namespace) arguments containing relevant model, policy, and env informatio