high_level_policy_main module

high_level_policy_main.evaluate_high_level_policy(nb_episodes_for_test=100, nb_trials=10, trained_agent_file='highlevel_weights.h5f', pretrained=False, visualize=False)
high_level_policy_main.find_good_high_level_policy(nb_steps=25000, load_weights=False, nb_episodes_for_test=100, visualize=False, tensorboard=False, save_path='./highlevel_weights.h5f')
high_level_policy_main.high_level_policy_testing(nb_episodes_for_test=100, trained_agent_file='highlevel_weights.h5f', pretrained=False, visualize=True)
high_level_policy_main.high_level_policy_training(nb_steps=25000, load_weights=False, training=True, testing=True, nb_episodes_for_test=20, max_nb_steps=100, visualize=False, tensorboard=False, save_path='highlevel_weights.h5f')

Do RL of the high-level policy and test it.

Parameters:
  • nb_steps – the number of steps to perform RL
  • load_weights – True if the pre-learned NN weights are loaded (for initializations of NNs)
  • training – True to enable training
  • testing – True to enable testing
  • nb_episodes_for_test – the number of episodes for testing