94 lines
3.4 KiB
JSON
94 lines
3.4 KiB
JSON
|
{
|
||
|
|
"args": {
|
||
|
|
"seed": [1],
|
||
|
|
"agent_rl_algo": ["ppo"],
|
||
|
|
"n_total_updates": [30000],
|
||
|
|
"train_runner": ["paired"],
|
||
|
|
"n_devices": [1],
|
||
|
|
"student_model_name": ["default_student_cnn"],
|
||
|
|
"env_name": ["Maze"],
|
||
|
|
"verbose": [false],
|
||
|
|
"log_dir": ["~/logs/minimax"],
|
||
|
|
"log_interval": [10],
|
||
|
|
"from_last_checkpoint": [true],
|
||
|
|
"checkpoint_interval": [1000],
|
||
|
|
"archive_interval": [0],
|
||
|
|
"archive_init_checkpoint": [false],
|
||
|
|
"test_interval": [100],
|
||
|
|
"n_students": [2],
|
||
|
|
"n_parallel": [32],
|
||
|
|
"n_eval": [1],
|
||
|
|
"n_rollout_steps": [256],
|
||
|
|
"lr": [0.0001],
|
||
|
|
"lr_anneal_steps": [0],
|
||
|
|
"max_grad_norm": [0.5],
|
||
|
|
"adam_eps": [1e-05],
|
||
|
|
"track_env_metrics": [true],
|
||
|
|
"discount": [0.995],
|
||
|
|
"n_unroll_rollout": [10],
|
||
|
|
"render": [false],
|
||
|
|
"ued_score": ["relative_regret"],
|
||
|
|
"student_gae_lambda": [0.98],
|
||
|
|
"teacher_discount": [0.995],
|
||
|
|
"teacher_lr": [0.0001],
|
||
|
|
"teacher_lr_anneal_steps": [0],
|
||
|
|
"teacher_gae_lambda": [0.98],
|
||
|
|
"student_entropy_coef": [0.001],
|
||
|
|
"student_value_loss_coef": [0.5],
|
||
|
|
"student_n_unroll_update": [5],
|
||
|
|
"student_ppo_n_epochs": [5],
|
||
|
|
"student_ppo_n_minibatches": [1],
|
||
|
|
"student_ppo_clip_eps": [0.2],
|
||
|
|
"student_ppo_clip_value_loss": [true],
|
||
|
|
"teacher_entropy_coef": [0.001],
|
||
|
|
"teacher_value_loss_coef": [0.5],
|
||
|
|
"teacher_n_unroll_update": [5],
|
||
|
|
"teacher_ppo_n_epochs": [5],
|
||
|
|
"teacher_ppo_n_minibatches": [1],
|
||
|
|
"teacher_ppo_clip_eps": [0.2],
|
||
|
|
"teacher_ppo_clip_value_loss": [true],
|
||
|
|
"student_recurrent_arch": ["s5"],
|
||
|
|
"student_recurrent_hidden_dim": [256],
|
||
|
|
"student_hidden_dim": [32],
|
||
|
|
"student_n_hidden_layers": [1],
|
||
|
|
"student_n_conv_filters": [16],
|
||
|
|
"student_n_scalar_embeddings": [4],
|
||
|
|
"student_scalar_embed_dim": [5],
|
||
|
|
"student_s5_n_blocks": [2],
|
||
|
|
"student_s5_n_layers": [2],
|
||
|
|
"student_s5_layernorm_pos": ["post"],
|
||
|
|
"student_s5_activation": ["half_glu1"],
|
||
|
|
"teacher_model_name": ["default_teacher_cnn"],
|
||
|
|
"teacher_recurrent_arch": ["s5"],
|
||
|
|
"teacher_recurrent_hidden_dim": [256],
|
||
|
|
"teacher_hidden_dim": [32],
|
||
|
|
"teacher_n_hidden_layers": [1],
|
||
|
|
"teacher_n_conv_filters": [32],
|
||
|
|
"teacher_scalar_embed_dim": [10],
|
||
|
|
"teacher_s5_n_blocks": [2],
|
||
|
|
"teacher_s5_n_layers": [2],
|
||
|
|
"teacher_s5_layernorm_pos": ["post"],
|
||
|
|
"teacher_s5_activation": ["half_glu1"],
|
||
|
|
"maze_height": [13],
|
||
|
|
"maze_width": [13],
|
||
|
|
"maze_n_walls": [60],
|
||
|
|
"maze_replace_wall_pos": [false],
|
||
|
|
"maze_sample_n_walls": [false],
|
||
|
|
"maze_see_agent": [false],
|
||
|
|
"maze_normalize_obs": [true],
|
||
|
|
"maze_obs_agent_pos": [false],
|
||
|
|
"maze_max_episode_steps": [250],
|
||
|
|
"maze_ued_replace_wall_pos": [true],
|
||
|
|
"maze_ued_fixed_n_wall_steps": [true],
|
||
|
|
"maze_ued_first_wall_pos_sets_budget": [false],
|
||
|
|
"maze_ued_noise_dim": [50],
|
||
|
|
"maze_ued_n_walls": [60],
|
||
|
|
"maze_ued_set_agent_dir": [false],
|
||
|
|
"maze_ued_normalize_obs": [true],
|
||
|
|
"test_n_episodes": [10],
|
||
|
|
"test_env_names": ["Maze-SixteenRooms,Maze-Labyrinth,Maze-StandardMaze"],
|
||
|
|
"test_agent_idxs": ["\"*\""],
|
||
|
|
"maze_test_see_agent": [false],
|
||
|
|
"maze_test_normalize_obs": [true]
|
||
|
|
}
|
||
|
|
}
|