diff --git a/scripts/configs/active_sampler_learning.yaml b/scripts/configs/active_sampler_learning.yaml index 248fde5814..8e4e5c118e 100644 --- a/scripts/configs/active_sampler_learning.yaml +++ b/scripts/configs/active_sampler_learning.yaml @@ -28,15 +28,15 @@ APPROACHES: # FLAGS: # explorer: "active_sampler" # active_sampler_explore_task_strategy: "skill_diversity" - random_score_explore: - NAME: "active_sampler_learning" - FLAGS: - explorer: "active_sampler" - active_sampler_explore_task_strategy: "random" - random_nsrts_explore: - NAME: "active_sampler_learning" - FLAGS: - explorer: "random_nsrts" + # random_score_explore: + # NAME: "active_sampler_learning" + # FLAGS: + # explorer: "active_sampler" + # active_sampler_explore_task_strategy: "random" + # random_nsrts_explore: + # NAME: "active_sampler_learning" + # FLAGS: + # explorer: "random_nsrts" maple_q: NAME: "maple_q" FLAGS: @@ -68,26 +68,9 @@ ENVS: # active_sampler_learning_explore_length_base: 100000 # effectively disable # active_sampler_learning_feature_selection: oracle # max_num_steps_interaction_request: 100 - # spot_sweeping_sim: - # NAME: "spot_main_sweep_env" - # FLAGS: - # active_sampler_learning_exploration_epsilon: 0.5 - # skill_competence_model_optimistic_recency_size: 2 - # skill_competence_model_optimistic_window_size: 2 - # horizon: 10 - # active_sampler_learning_explore_length_base: 100000 # effectively disable - # active_sampler_learning_feature_selection: oracle - # spot_run_dry: True - # approach_wrapper: spot_wrapper - # perceiver: spot_perceiver - # segmenter: spot - # active_sampler_learning_explore_pursue_goal_interval: 5 - # active_sampler_learning_init_cycles_to_pursue_goal: 5 - # max_num_steps_interaction_request: 100 - spot_sweeping_sim_yogurt_only: + spot_sweeping_sim: NAME: "spot_main_sweep_env" FLAGS: - spot_sweep_env_goal_description: "'get the yogurt into the bucket'" active_sampler_learning_exploration_epsilon: 0.5 skill_competence_model_optimistic_recency_size: 2 skill_competence_model_optimistic_window_size: 2 @@ -101,6 +84,23 @@ ENVS: active_sampler_learning_explore_pursue_goal_interval: 5 active_sampler_learning_init_cycles_to_pursue_goal: 5 max_num_steps_interaction_request: 100 + # spot_sweeping_sim_yogurt_only: + # NAME: "spot_main_sweep_env" + # FLAGS: + # spot_sweep_env_goal_description: "'get the yogurt into the bucket'" + # active_sampler_learning_exploration_epsilon: 0.5 + # skill_competence_model_optimistic_recency_size: 2 + # skill_competence_model_optimistic_window_size: 2 + # horizon: 10 + # active_sampler_learning_explore_length_base: 100000 # effectively disable + # active_sampler_learning_feature_selection: oracle + # spot_run_dry: True + # approach_wrapper: spot_wrapper + # perceiver: spot_perceiver + # segmenter: spot + # active_sampler_learning_explore_pursue_goal_interval: 5 + # active_sampler_learning_init_cycles_to_pursue_goal: 5 + # max_num_steps_interaction_request: 100 ARGS: [] FLAGS: