diff --git a/RLLib_training/experiment_configs/predictions_test/config.gin b/RLLib_training/experiment_configs/predictions_test/config.gin new file mode 100644 index 0000000000000000000000000000000000000000..6d664248302b4f230714237fb3ee15c6d2c73aa8 --- /dev/null +++ b/RLLib_training/experiment_configs/predictions_test/config.gin @@ -0,0 +1,27 @@ +run_experiment.name = "memory_experiment_results" +run_experiment.num_iterations = 2002 +run_experiment.save_every = 50 +run_experiment.hidden_sizes = {"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_with_pred_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 = {"grid_search": [1e-4, 1e-3, 1e-2]} +run_experiment.kl_coeff = 0.2 +run_experiment.lambda_gae = 0.9 +run_experiment.predictor = "dummy_predictor" +run_experiment.step_memory = 1