Behavioral Cloning on N/W reconfiguration

Classes and Functions

phy_train_bc.evaluate_policy(env, model)

Evaluating the learned policy by running experiments in the environment

Parameters
  • env (Gym.Env) – The Open DSS RL environment

  • model (torch.nn.Module) – Trained policy network model

Returns

average episode length, average reward

Return type

float

phy_train_bc.train_and_evaluate(env, bc_train_epoch_lens, exp_trajectory_len)

For different combination of channel bandwidths, router queue limits and expert demonstrations, train and test the policy and saves the results, reward and policy network.

Parameters
  • env (Gym.Env) – The Open DSS RL environment

  • bc_train_epoch_lens (list) – List of the number of epochs the behavioral cloning agent need to be trained

  • exp_tajectory_len (int) – The number of expert demonstrations steps considered for AIRL training

Returns

Nothing

Return type

None