diff --git a/bc-experiment.py b/bc-experiment.py index 0d7dc5b..b50489b 100644 --- a/bc-experiment.py +++ b/bc-experiment.py @@ -117,7 +117,7 @@ def training_function(config): pi_opt.step() pi_lr_scheduler.step() - if epoch % 10 == 0: + if epoch % 25 == 0: gen_states, gen_actions, gen_rewards, gen_dones, gen_collisions = rollout(env_fn, policy.cpu(), n_episodes=60, max_steps_per_episode=200) gen_mean_episode_length = (~gen_dones).sum() / gen_states.shape[0] gen_mean_reward_per_episode = gen_rewards[~gen_dones].sum() / gen_states.shape[0] diff --git a/best_configs/bc_expA.json b/best_configs/bc_expA.json index 8581472..42fee47 100644 --- a/best_configs/bc_expA.json +++ b/best_configs/bc_expA.json @@ -10,6 +10,6 @@ "n_hidden_layers": 2, "activation": 0 }, - "train_epochs": 1000, + "train_epochs": 300, "seed": 0 } \ No newline at end of file diff --git a/best_configs/bc_expB.json b/best_configs/bc_expB.json index 7b55224..9d81a2f 100644 --- a/best_configs/bc_expB.json +++ b/best_configs/bc_expB.json @@ -10,6 +10,6 @@ "n_hidden_layers": 2, "activation": 0 }, - "train_epochs": 1000, + "train_epochs": 300, "seed": 0 } \ No newline at end of file diff --git a/gail-experiment.py b/gail-experiment.py index 45084cd..68c9ae9 100644 --- a/gail-experiment.py +++ b/gail-experiment.py @@ -187,8 +187,8 @@ if __name__ == '__main__': }, 'discriminator': { 'learning_rate': 1e-3, - 'weight_decay': 1e-4, # 1e-5 Fixme - 'iterations_per_epoch': 100, # 500 Fixme + 'weight_decay': 1e-4, + 'iterations_per_epoch': 100, 'n_hidden_layers_element': tune.grid_search([3,4]), 'n_hidden_layers_global': tune.grid_search([1,2]), 'hidden_layer_size': 10, diff --git a/gail-trpo-experiment.py b/gail-trpo-experiment.py index ff96929..8a28ccf 100644 --- a/gail-trpo-experiment.py +++ b/gail-trpo-experiment.py @@ -184,8 +184,8 @@ if __name__ == '__main__': }, 'discriminator': { 'learning_rate': 1e-3, - 'weight_decay': 1e-4, # 1e-5 Fixme - 'iterations_per_epoch': 100, # 500 Fixme + 'weight_decay': 1e-4, + 'iterations_per_epoch': 100, 'n_hidden_layers_element': tune.grid_search([3,4]), 'n_hidden_layers_global': tune.grid_search([1,2]), 'hidden_layer_size': 10,