config.gin

run_experiment.name = "memory_experiment_results"
run_experiment.num_iterations = 2002
run_experiment.save_every = 50
run_experiment.hidden_sizes = [32, 32]#{"grid_search": [[32, 32], [64, 64], [128, 128]]}

run_experiment.map_width = 8
run_experiment.map_height = 8
run_experiment.n_agents = 3
run_experiment.rail_generator = "complex_rail_generator"
run_experiment.nr_extra = 5
run_experiment.policy_folder_name = "ppo_policy_hidden_size_{config[hidden_sizes][0]}_entropy_coeff_{config[entropy_coeff]}_"

run_experiment.horizon = 50
run_experiment.seed = 123

#run_experiment.conv_model = {"grid_search": [True, False]}
run_experiment.conv_model = False

run_experiment.obs_builder = @TreeObsForRailEnv()
TreeObsForRailEnv.max_depth = 2
LocalObsForRailEnv.view_radius = 5

run_experiment.entropy_coeff = 1e-4#{"grid_search": [1e-4, 1e-3, 1e-2]}
run_experiment.kl_coeff = 0.2
run_experiment.lambda_gae = 0.9
run_experiment.predictor = None#@DummyPredictorForRailEnv()
run_experiment.step_memory = 2